14.

1 Methods and Techniques for the Modern Geomorphologist: An

Introduction to the Volume
AD Switzer, Earth Observatory of Singapore, Nanyang Technological University, Singapore
DM Kennedy, The University of Melbourne, Parkville, VIC, Australia
r 2013 Elsevier Inc. All rights reserved.

References 4

Geomorphology is a discipline that covers a wide range of 2007. The group discussed at some length the difficulties of
topics from sediment transport over millennial timescales, to compiling and publishing a ‘recipe book’ of geomorpho-
hydrodynamic flow in urban canals, to remote sensing of logical methods. What we have tried to build in this volume is
other planets and moons in our Solar System. The methods a diverse but by no means exhaustive summary of some of the
employed to understand geomorphic processes are therefore methods available to the modern geomorphologist. This vol-
highly varied. Approaches to geomorphology have also ume is an attempt at producing such a reference. It is divided
changed significantly through time, evolving as the methods into three main themes: the first looks at mapping and land-
available to researchers have advanced. Davisian concepts of form evaluation; the second theme investigates the common
landscape youth and maturity based on visual observation techniques in the field; and the third looks at the primary
dominated academic discussions in the late nineteenth and techniques used in the laboratory. Five years on from the first
early twentieth centuries. With the advent of instrumentation concept meeting, the authors hope that they have produced a
capable of directly measuring geomorphic processes, such as reference work that can be a primary port of call for many in
stream flow, and the standardization of basic methodologies the field of geomorphology.
such as grain size (Folk and Ward, 1957), discussions then The volume begins with a slightly philosophical look at
focused on the frequency and magnitude of processes that modern geomorphological techniques in a paper by Schrott
drove landscape evolution (Wolman and Miller, 1960). The et al. (see Chapter 14.2). The authors open with a classic quote
invention of radiocarbon dating in the mid-twentieth century from Ritter (1986) on the essence of field observation before
(Libby et al., 1949), arguably transformed the discipline providing a commentary on the fundamental principles of
allowing geomorphologists to easily and directly quantify the observation in geomorphology. They then shift focus slightly
age of a landscape for the first time. Most recently, remote highlighting the advantages and disadvantages of recent pro-
sensing of data from satellite and terrestrial platforms has gress in digital mapping and field geophysics.
provided another improvement in our ability to understand The next six papers in the volume focus on some of the
landscapes with the contemporary researcher now faced with methods used in attempts to map, analyze, and manage the
the difficulties of managing large data sets. various landscapes of the Earth. The first paper in this section
The geomorphologist therefore draws on expertise from a is by Minar et al. (see Chapter 14.3) and summarizes the field
wide range of disciplines. In each discipline specialized texts of geomorphometry, a field recently revitalized by the growing
exist; however, the level of details is generally beyond the availability of high-resolution data, much of which can be
needs of an individual researcher, and few are aimed at the remotely sourced from orbiting satellite platforms (see also
geomorphological audience. King (1966) and Goudie (1990) Chapter 3.7). Such data sets have opened up many new ap-
are two texts that have attempted to translate the complexities plications, both theoretical and practical, not only within
of field and laboratory methodologies for geomorphologists, geomorphology but also across many other fields. The authors
but since their publication many new methodologies such as note that the future of the field primarily depends on the
cosmogenic dating or ground penetrating radar (GPR) have integrated development of theoretical, empirical, and tech-
experienced considerable advances. In addition, advances in nological aspects. Minar et al. suggest that in many cases
computing technology in the past two decades have trans- geomorphometric investigation has been led by technological
formed many basic techniques such as surveying. This volume developments (predominantly empirical). Although noting
addresses these issues and is the result of discussions that the importance of technology, they also advocate that geo-
coeditors Switzer and Kennedy had with the then Inter- morphometric theory has considerable potential to influence
national Association of Geomorphologists president, Andrew the content and quality of empirical research. They conclude
Goudie, at the regional International Association of Geo- that high-resolution data have resulted in an exciting phase in
morphologists/Association Internationale des Géomorpholo- geomorphometry, but that endeavors in this field should be
gues regional meeting in Kota Kinabalu, Malaysia, 25–29 May targeted and based on a coherent theoretical–methodological
background.
The follow-on paper discusses the ‘modern’ geomorpho-
Switzer, A.D., Kennedy, D.M., 2013. Methods and techniques for the logical map. Here, Seijmonsbergen (see Chapter 14.4) notes
modern geomorphologist: an introduction to the volume. In: Shroder, J.
(Editor in Chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise on
that geomorphological maps represent a single spatial distri-
Geomorphology. Academic Press, San Diego, CA, vol. 14, Methods in bution of landforms, materials, and of the processes respon-
Geomorphology, pp. 1–5. sible for their formation. The author states that although

Treatise on Geomorphology, Volume 14 doi:10.1016/B978-0-12-374739-6.00368-7 1

2 Methods and Techniques for the Modern Geomorphologist: An Introduction to the Volume

modern maps can contain a wealth of information, there is no effectively synthesizes the analysis of channel form. They point
uniformity among geomorphological legends, as many maps out that the diversity of river channels renders research aims
are developed for use at different scales in a variety of land- and site specificity critical to any investigation, and suggest
scapes. This can cause problems when and where analyses are that there is an inevitable trade-off between resolution, scale,
carried beyond the mapping area or site of investi- and time. This comprehensive paper outlines the investigative
gation from a perspective different to that which created techniques used to assess river channel form in relation to
the original map. To overcome this problem, the author channel planform, cross section, long profile, and three-
describes the preparation, data collection, analysis, storage, dimensional morphology. It provides an analysis of three-
and visualization of the modern geomorphological map not- dimensional forms that provide the most rigorous approach
ing that such work is now primarily a collection of digital that links form to processes, which enables the most accurate
layers of information collectively stored in a geodatabase. predictive modeling of river behavior.
Some common techniques for transforming ‘paper-based’ Catchment development is not, however, entirely the result
maps to digital layers are outlined using conversion legends of fluvial processes, and in the next paper Hyatt and Shul-
(e.g., digital elevation models (DEMs)) as well as the process meister describe and critique the process of mapping glacial
of (semi)automated extraction and classification of geo- geomorphological features (see Chapter 14.8). In this paper,
morphological features from DEMs. they define two types of glacial geomorphological maps and
The next paper is a commentary by Stephen Tooth (see discuss the identification of glacial features before outlining
Chapter 14.5) on the use of Virtual Globes such as Google the basic steps to producing a glacial geomorphological map.
EarthTM in geomorphology where he asks and answers the A series of short case studies conducted in different mapping
query whether it is reenchanting, revolutionizing geomorph- settings reinforce the key concepts before a section on map
ology, or it is just another resource The range of online ‘virtual production and publication is discussed.
globes’ has vastly increased access to digital imagery of the The next section of the volume is focused on the common
surface of Earth (and other planetary bodies) to the public and methods used in the field by geomorphologists, and Switzer
scientific community, which has enabled examination of the and Kennedy (see Chapter 14.9) remind us of the importance
Earth’s surface from various scales and perspectives. This of field studies in an age of computer modeling and remote
commentary concludes that such resources may be providing sensing. The opening paper in this section, by Kennedy (see
opportunities for reenchantment of geomorphology, but Chapter 14.10), delves into the measurement of landscape
when compared to other past and present developments in form (surveying), a fundamental technique for geomorph-
data acquisition and analysis, its use is probably not revo- ologists. He focuses on the common surveying systems (from
lutionizing the discipline. Virtual Globes are just another leveling to terrestrial laser scanning) encountered by re-
resource in the geomorphological ‘toolkit’ but nonetheless a searchers and students and discusses the general regime for
remarkable one, having rapidly become an invaluable part of making an informed choice of which type of instrument to be
many research, education, and outreach projects. used in the field as well as outlining the basic principles of
At a smaller scale than the globe, the paper of Yvonne surveying for a geomorphological audience.
Martin (see Chapter 14.6) explores the numerical modeling Once a landform is surveyed, it is generally time to in-
approach for the study of drainage basin development. Like vestigate the subsurface. To this end, Woodward and Sloss
the preceding chapters, the author notes that considerable open the section on subsurface investigation by introducing
advances in computer technology over recent decades have the principles of coring (see Chapter 14.11). Their paper covers
allowed for the solution of complex numerical models and the objectives for taking a good core and the major factors that
have led to notable advancements in the understanding of should be considered to ensure the collection of non-
drainage basin development. The author outlines a number of disturbed, representative core samples. The paper also pro-
particular challenges in the field and notes that models typi- vides an overview of the main types of coring equipment that
cally incorporate a series of equations for geomorphic and can be used to sample sediments from a variety of settings,
other processes. Martin notes that one of the most significant along with the major advantages and disadvantages of each
contributions of numerical modeling to geomorphology is type of corer.
that it allows for the consideration of a series of ‘what-if’ One disadvantage of a core is it samples a very small spatial
questions that might otherwise be impossible to study given area, which may not provide enough detailed information on
the large scales of drainage basin development. The author subsurface stratigraphy especially for applied geomorpho-
differentiates two stages of drainage basin development, initial logical case studies such as in the field of paleoseismology.
and subsequent, before suggesting that any modeler should A trench can overcome this problem, and McCalpin (see
make clear which of these stages (or both) the focus of Chapter 14.12) gives us a detailed commentary on trench
a particular study is. Martin then provides several examples excavation and putting it to good use. The author notes that
to demonstrate the range of theoretically based research trenching creates artificial exposure of the shallow subsurface
questions about drainage basin development before outlin- from which the common output is a scaled map of the vertical
ing several issues and difficulties surrounding model trench walls, called a trench log.
confirmation. Moving out of the trenches, the authors come back to the
Rivers are generally the drivers of terrestrial drainage basin surface for a summary of methods for working with boulders
development, and at a smaller scale they exhibit an intricate and gravel. In the paper titled ‘Working with gravel and
complexity of form. When researching rivers, Fuller et al. boulders,’ Felton (see Chapter 14.13) outlines two comple-
(see Chapter 14.7) suggest that there is no single approach that mentary descriptive methods, which serve to systematize and
 Methods and Techniques for the Modern Geomorphologist: An Introduction to the Volume 3

integrate the wide range of bed and clast attributes that examples of field instrumentation, their application and op-
characterize complex deposits of coarse gravel. The com- eration, as well as examining the methodological implications
position and fabric method has a material focus and provides of deploying instruments in the field. The authors then discuss
a technique for the detailed description of attributes of the how these measurements contribute to the understanding of
constituent clasts found in gravel deposits. Felton then out- geomorphic environments and describe some of the analytical
lines the method of sedimentary facies model building, which approaches to time series data analysis.
emphasizes the organization of a deposit on a bed-by-bed Field work is only one aspect of geomorphological study,
basis so as to identify facies and infer depositional processes. and it is characteristically focused on gathering samples that
Finally, she concludes by suggesting that the use of a com- need to be analyzed in the laboratory. Laboratory methods are
prehensive gravel fabric and petrography log may ensure the the focus of the next section in this volume (see Chapter
most systematic collection of data. 14.18), and the authors start with particle-size analysis where
Many boulder deposits occur close to the bedrock, which Switzer (see Chapter 14.19) describes one of the fundamental
they were eroded from. Bedrock does not however necessarily tools for interpreting landforms and sediments. The meas-
erode into gravel-sized clasts but can do so on a millimeter urement of the particle-size (grain-size) distribution is a di-
scale. The next paper by Stephenson (see Chapter 14.14) verse and complex field that has undergone significant change
outlines how such erosion on a millimeter scale can be over the last few decades. Switzer points out that although
measured using a microerosion meter. He notes that today the boulder-size material is still characteristically measured by
most widely used application relates to determining direct measurement, a large variety of techniques are used to
the rate at which shore platforms lower before pointing determine grain-size distributions for sand- to clay-dominated
out studies understanding rates of erosion in noncoastal sediments. He outlines the problems faced when combin-
environments are far fewer, although the technique is likely to ations of methods are needed to quantify grain-size popu-
have considerable potential in terrestrial environments. lations; in particular, he provides a commentary on the
After Stephenson has dealt with rocks, it is only fair that assumptions of sphericity and specific gravity before con-
one of the Earth’s primary derivatives, soils, is discussed. (see cluding with a discussion on the various methods used to
also Chapters 4.1 and 4.9). For a synthesis of describing soils present particle-size data.
in a geomorphic context, the authors turn to Morrison and Following the analysis of particle sizing techniques, the
Laurie, who focus on describing soils (sites and profiles) in the authors move to the analysis of shape (see Chapter 14.20),
field, working from a geomorphic perspective (see Chap- where Bullard describes the way geomorphologists look at the
ter 14.15). Topics considered include placing soils in the interrelated variables of sphericity, roundness, and surface
landscape, selection of sites for description, accessing soil texture. She outlines a number of techniques that have been
profiles, describing sites and profiles, field tests that can be developed to assess particle shape and notes the increasing use
done, and soil sampling for laboratory analysis. Aspects like of digital imaging and computer analysis. Bullard concludes
the level of detail required (with examples) and recent that although it is relatively straightforward to assess particle
advances in the field are also considered. Some frequently shape, the many different factors that influence shape mean
encountered problems are highlighted and current approaches making geomorphological interpretations based solely on this
are discussed. variable are difficult.
Imaging the subsurface using noninvasive techniques is Particle size and shape data can allow a geomorphologist
now a key part of modern geomorphic studies and is often to infer the environment of deposition; however, in some
conducted in concert with coring, trenching, and soil analysis. cases more accurate data of transportation can be obtained
In the next paper, Bristow (see Chapter 14.16) covers the de- through the examination of individual grain surfaces. In the
velopment and application of Ground Penetrating Radar next chapter, Bull and Morgan (see Chapter 14.21) provide a
(GPR). This technique provides almost continuous, high- synthesis of the use of the scanning electron microscope and
resolution profiles of the shallow subsurface, which cannot be outline the widespread application of the electron microscope
achieved by any other geophysical technique. GPR is now in geomorphology, including examples from studies of en-
widely used in the investigation of shallow subsurface stra- vironmental reconstruction, soil fabric, and weathering as well
tigraphy due to an increased awareness of the technique as its recent application in geoforensic analysis.
combined with improvements in equipment. Bristow notes The next topic covered in the laboratory section of the
that the development of digital data output, along with par- volume is the determination of organic and carbonate content
allel developments in computing have enabled rapid pro- in sediments. In this paper, Kennedy and Woods (see Chap-
cessing and visualization that have helped to establish GPR in ter 14.22) outline the three main approaches for measuring
the geomorphologist’s tool box. Bristow then goes on to re- carbon in geomorphology: physical, chemical, and relative
view some of the pros and cons of GPR surveys with a number techniques. They note that in many cases, the choice of tech-
of examples to illustrate successful case studies and common nique depends on the accuracy required for the study and that
‘problems.’ many techniques are now automated, allowing for rapid
All the preceding chapters have focused on either the form quantification of both organic and carbonate components.
of the landscape or what it is composed of. In the chapter of In the next paper, the authors continue with the laboratory
Sherman and colleagues (see Chapter 14.17), the authors in- theme with Chague-Goff and Wong (see Chapter 14.23)
vestigate instrumentation systems and the measurement of summarizing a selection of common wet chemical techniques
processes, mainly hydrological and aeolian, which erode and such as pH, conductivity, ion-selective electrodes, colorimetric
transfer material around the landscape. The chapter presents and spectrophotometric analysis, and atomic absorption
4 Methods and Techniques for the Modern Geomorphologist: An Introduction to the Volume

spectrophotometry. They provide basic instructions, tips, and new generation of direct shear devices that genuinely recreate
references to specialized texts for each of these methods. the conditions within, for example, deforming slopes that may
In the following paper, Birch (see Chapter 14.24) focuses allow a proper understanding of process to be established.
on the analysis of sedimentary metals tackling the challenge of The second chapter in this section focuses on hard rocky
identifying and implementing a number of key indicators of surfaces with Goudie (see Chapter 14.29) summarizing the
environmental condition that are inexpensive, easy to use, and field measurement of rock surface hardness using the Schmidt
provide appropriate information to enable managers in the Hammer and Equotip. These techniques are quick, simple,
governance of estuaries. The author notes that sedimentary cheap, and portable. They are commonly used by geo-
metals provide an inexpensive and easily acquired means of morphologists to relate rock hardness to landform develop-
making an initial assessment of estuarine health, and provide ment and are also used to undertake relative dating of
the framework on which more detailed and expensive in- geomorphological phenomena.
vestigations can be based. In many cases, understanding landscape process studies
The focus of the laboratory methods section then switches require considerably more dating than the relative techniques
to the analysis of micropaleontology. Microfossils such as of a Schmidt Hammer. A myriad of qualitative and quantita-
foraminifera and pollen provide key proxies for environ- tive dating techniques are now available to the geomorph-
mental change in landscape studies and form the basis of the ologist, and Sloss et al. (see Chapter 14.30) provide an
long-term (millennial scale) understanding of climate change. introduction to dating techniques in the following paper. This
In the chapter by Horton and colleagues (see Chapter 14.25), chapter provides researchers with a guide to the different
the use of microfossils as indicators of sea-level change, dating techniques that can be used in geomorphological in-
paleoearthquakes, tsunamis, and tropical cyclones is explored. vestigations and comment on several issues that need to be
They note that fine-grained sediments deposited in low-en- addressed when using geochronological data, including
ergy, intertidal settings provide a useful archive of sea-level important issues relating to accuracy and precision.
change, as well as the occurrence of paleoearthquakes, tsuna- Following on from the introductory paper by Sloss et al.
mis, and tropical cyclones. The investigation of microfossils the authors round out the volume with a more case-specific
(pollen, diatoms, and foraminifera, etc.) that accumulate in topic of picking material for 14C dates from organic deposits
salt marsh and other estuarine environments have become (see Chapter 14.31), particularly on samples collected from
increasingly widespread in geomorphology, primarily because marsh environments. In a paper by Kemp et al., their com-
their distribution is closely linked to tidal elevation. The paper mentary focuses on the selection of plant macrofossils for
concludes with a discussion of transfer functions for calcu- radiocarbon dating and application of ages from different
lating quantitative estimates of past environmental conditions types of macrofossils to varied research projects, and they
from microfossil data. make recommendations for selection and preparation of tidal-
Keeping a micropaleontological theme, the authors move marsh samples for dating.
to another realm where Moss (see Chapter 14.26) investigates The mix of case-specific and theoretical methodologies
a number of valuable applications of pollen analysis in the provided here along with specific ‘recipe’ instructions for
geomorphological context. Covering topics including absolute methodological analysis means this volume provides an up-
age control; the production of environmental data in terms of to-date assessment of the common methodological techniques
past climates and depositional environments; and the impli- used in geomorphology. The focus of the volume has been on
cations for investigating impacts of people/human landuse on the general methods of field and laboratory study and their
landforms and geomorphological processes, Moss provides an application to geomorphic research. Each method presented is
overview of these applications, in addition to a detailed supported by its own specific discipline, and the reader is
examination of palynological laboratory techniques, pollen encouraged to investigate each technique further in the rele-
identification/counting, and the production/interpretation of vant specialized literature to expand their understanding of
pollen diagrams. the topic.
The next section of this methods volume investigates the
strength of Earth surface materials and as outlined in the
introduction of Kennedy (see Chapter 14.27), material
strength is a fundamental property for understanding land- References
scape stability and the ability of materials to resist weathering
and erosion. In the first chapter of this section, Petley and Folk, R.L., Ward, W.C., 1957. Brazos River bar: A study in the significance of grain
Clark cover the well-established geotechnical engineering size parameters. Journal of Sedimentary Petrology 27, 3–26.
technique of direct shear testing and note that it has found Goudie, A. (Ed.), 1990. Geomorphological Techniques. Unwin Hyman, New York,
USA.
surprisingly limited application within geomorphology (see
King, C.A.M., 1966. Techniques in Geomorphology. Edward Arnold, London.
Chapter 14.28). They suggest that the reason for this primarily Libby, W.F., Anderson, E.C., Arnold, J.R., 1949. Age determination by radiocarbon
pertains to a lack of insight that the technique provides into content: World-wide assay of natural radiocarbon. Science 109, 227–228.
mechanisms of deformation, beyond simple classification Ritter, D.F., 1986. Process Geomorphology. W.C. Brown, Dubuque, Iowa, USA, 579
(i.e., brittle or ductile). In their commentary, which is focused pp.
Wolman, M.G., Miller, J.P., 1960. Magnitude and frequency of forces in geomorphic
on laboratory analysis, they note that recent advances in
processes. Journal of Geology 68, 54–74.
control systems provides considerable potential to create a
 Methods and Techniques for the Modern Geomorphologist: An Introduction to the Volume 5

Biographical Sketch

Adam Switzer is a sedimentologist and geomorphologist, who holds the positions of Singapore NRF Fellow,
Nanyang Assistant Professor and Principal Investigator at The Earth Observatory of Singapore and Division of
Earth Sciences in SPMS at Nanyang Technological University.
He and his research group are investigating the geological record of coastal hazards (storms and tsunamis) in
Southeast Asia. One of the driving aims of Dr. Switzer’s research group is to make the coastlines of Asia safer
places to live, work, and holiday. Currently, Asia is the focus of his research program and Singapore has the
resources to enable him to effectively study the sedimentary and geomorphic effects of tsunamis and storm surges.
Dr. Switzer’s 10-strong research group currently works in nine different countries throughout the world.
Dr. Switzer obtained a BSc and PhD in Geosciences from the University of Wollongong, Australia, before
completing his postdoctoral studies at the University of Hong Kong.

Dr David M. Kennedy is a geomorphologist who specializes in the response of coastal systems to climatic and
environmental change. He particularly focuses on the Holocene, researching estuarine, coral reef, and rocky coast
landforms. He completed his doctorial study at the University of Wollongong, Australia under the guidance of
Prof. Colin Woodroffe following undergraduate studies at the University of Sydney, Australia. After a postdoctoral
research position at Wollongong he was based at Victoria University of Wellington, New Zealand. He is currently a
senior lecturer in the Department of Resource Management and Geography at The University of Melbourne,
Australia.
 14.2 Fundamental Classic and Modern Field Techniques in Geomorphology:
An Overview
L Schrott, J-C Otto, J Götz, and M Geilhausen, University of Salzburg, Salzburg, Austria
r 2013 Elsevier Inc. All rights reserved.

14.2.1 Introduction 6
14.2.2 Classic Field Techniques in Geomorphology Revisited 7
14.2.2.1 Geomorphological Mapping 7
14.2.2.1.1 Remote sensing and geomorphological mapping 7
14.2.2.2 Shallow Coring and Sampling 8
14.2.2.2.1 Coring 8
14.2.2.2.2 Sampling 8
14.2.3 Modern Field Techniques in Geomorphology 9
14.2.3.1 Principles of GPS and Applications in Geomorphology 9
14.2.3.2 LiDAR in Geomorphology 9
14.2.3.2.1 Principles of LiDAR 10
14.2.3.2.2 Airborne or terrestrial laser scanning 10
14.2.3.2.3 LiDAR – principles of data acquisition and processing 11
14.2.3.3 Geophysical Applications in Geomorphology 11
14.2.3.3.1 Ground-penetrating radar 11
14.2.3.3.1.1 Principle and geomorphic context 11
14.2.3.3.1.2 Advantages and disadvantages of GPR in
geomorphological applications 12
14.2.3.3.2 Geoelectrical resistivity 13
14.2.3.3.2.1 Principle and geomorphic context 13
14.2.3.3.2.2 Advantages and disadvantages in geomorphological applications 15
14.2.3.3.3 Seismic refraction 15
14.2.3.3.3.1 Principle and geomorphic context 15
14.2.3.3.3.2 Advantages and disadvantages in geomorphological studies 18
14.2.4 Conclusions 18
14.2.5 Disclaimer 18
References 18

Glossary reflection is characterized by many angles of outgoing rays

Geomorphometry It is quantitative analysis of land octree approaches.
surface and central theme in theoretical and applied Morhodynamics This refers to dynamic changes in
geomorphology. morphology, landform/landscape changes due to erosion
Laser beam reflection (specular, diffuse) Specular and sedimentation.
reflection is the mirror-like reflection of light and other Morphogenetic This refers mostly to a theoretical area
kinds of waves and describes a reflection of a single reflected devised by geomorphologists to relate climate, geomorphic
incoming into a single outgoing direction, whereas a diffuse processes, and landforms.

Abstract

Over the past three decades field techniques in geomorphology have evolved enormously. The advent of new technologies
influenced even classic techniques such as mapping, because remote sensing, in combination with high-resolution digital
elevation models, has significantly enhanced digital landform mapping and analysis. High-accuracy surveys of surface and
subsurface structures using light detection and ranging (LiDAR), differential global positioning systems (DGPS), and
geophysical techniques offer a wide range of challenges for geomorphological studies. Besides, field geophysics has become
increasingly efficient to capture quantitative subsurface data and provides a better understanding of form–process rela-
tionships. This chapter introduces some classic and modern field techniques in a geomorphological context.

Schrott, L., Otto, J.-C., Götz, J., Geilhausen, M., 2013. Fundamental classic
and modern field techniques in geomorphology: an overview. In: Shroder, J.
(Editor in Chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise on
Geomorphology. Academic Press, San Diego, CA, vol. 14, Methods in
Geomorphology, pp. 6–21.

6 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00369-9

 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview 7

14.2.1 Introduction (Evans, 1990; de Graaff et al., 1987; Gustavsson et al., 2006)
and legends that additionally emphasize the genetic aspects of
landforms (Barsch and Liedtke, 1980; Schoeneich et al., 1998;
ythe real test of geomorphic validity is outdoors, where all the Reynard et al., 2005; Tricart, 1965). For example, specialized
evidence must be pieced together into a lucid picture showing why legend systems exist for mapping high mountain environ-
landforms are the way we find them and why they are located ments (Kneisel et al., 1998) or natural hazards (Bundesamt für
where they are. A prime requisite for a geomorphologist is to be a
Wasser und Geologie, 2002). New approaches also include
careful observer of relevant field relationships.
(Ritter, 1986, p. 3). sediment storages within the geomorphological map. This
extends geomorphological maps into a third dimension and
serves as a basis for sediment budget studies (e.g., Otto et al.,
Since about 1980 field techniques in geomorphology have
2009).
been evolved enormously. This led to new branches in geo-
Traditionally, geomorphological maps are field based. Field
morphology and to new insights commonly connected with
mapping and the accurate delineation of landforms require
the need and use of much larger data sets. As Church (2010)
a lot of user experience. The mapping procedure strongly
pointed out, the bases for these changes are largely techno-
benefits from a detailed preparation involving literature
logical, and this had consequences for the whole discipline of
research, air-photo interpretation, and the consultation
geomorphology. The major changes include: (1) improved
of geological and topographic maps (cf. Chapter 14.4).
and new technological equipment (e.g., high-capacity data
Geomorphological maps are complex thematic maps that re-
loggers, light detection and ranging (LiDAR) techniques, dif-
quire careful application of cartographic design principles to
ferential global positioning system (DGPS), geophysical
preserve legibility and usability. A clear hierarchical organiza-
methods, etc.); (2) new possibilities to gather and store large
tion, thoughtful application of color, contrast, and symbol
data sets in the field using data loggers or field laptops; and (3)
density as well as a balanced arrangement of map items are
the advent of computer facilities that allow the analysis of
preconditions for well-designed maps that deliver the message
large amounts of field data.
intended (Otto et al., 2011).
The advent of new technologies also influenced classic
Field mapping approach can be technically enhanced by
methods in geomorphology such as mapping. This is because
using tablet PCs or hand-held computers and mobile GIS
digital mapping tools using tablet PCs or handheld computers
software. Portable computers provide direct access to digital
can now be used to display digital orthophotos and maps,
maps, orthophotos, and other types of data (e.g., borehole
which can in turn facilitate and potentially improve geo-
information and geophysical data) and when combined with a
morphological mapping in the field.
GPS on-the-fly georegistration of map data is enabled. A pre-
This introductory chapter describes some classic but fun-
requisite for using mobile GIS is the generation of a database
damental methods in geomorphology and provides some case
structure and digital map symbols prior to the field campaign.
studies of modern branches and aims to demonstrate the
This facilitates and improves the mapping procedure in the
application of classic and modern techniques in different
field and directly links geomorphological field data to GIS.
natural environments and to illustrate and assess the advances
Modern geomorphological maps can also incorporate new
and limitations of three different modern branches of field
remote sensing data, high-resolution digital elevation models
techniques in geomorphology (Global Positioning System
(DEMs), and subsurface information derived from geophysics
(GPS), terrestrial laser scanning, geophysical field techniques).
if GIS analysis techniques are used. Enhanced computer
Note that although sophisticated technology may be ap-
graphics and visualization tools as well as new ways of digital
plied to solve a scientific problem, search for suitable field
map dissemination through web mapping are valuable in-
sites still much depends on observation skills and experience.
struments to overcome technical constraints of paper-based
Thus the efficient and appropriate use of a particular techni-
maps leading to a reenchantment of geomorphological
que remains a great challenge for geomorphologists.
maps (see Chapters 3.11 and 14.8). WebGIS applications are
powerful GIS tools to present geomorphological maps as a
scientific result to the general public. The user receives spatially
14.2.2 Classic Field Techniques in Geomorphology
registered geographical information in an interactive web
Revisited
viewer according to the request. GIS processing is performed
on the WebGIS server and transferred to the user’s web brow-
14.2.2.1 Geomorphological Mapping
ser. Standardized data protocols such as Keyhole Markup
Geomorphological maps are classical tools to capture and Language (KML) enable a fast and easy publication of geo-
visualize landform assemblages and process distribution morphological maps within virtual globes like Google Earth
on a single map sheet. Generally, geomorphological maps are (see Chapters 3.11 and 14.5; Tooth, 2009; Mantovani et al.,
complex, multilayer maps covering geomorphometry, proces- 2010).
ses and process domains (morphodynamics), subsurface
material, morphogenetics, and hydrological features. Numer- 14.2.2.1.1 Remote sensing and geomorphological
ous guidelines and legends have evolved in the 20th century mapping
mostly representing country- or region-specific schools of Some high mountain regions, inaccessible or expansive
geomorphology (see Chapter 14.4). Differences bet- on Earth, or planets like Mars cannot be observed from the
ween the legend guidelines can be summarized to those ground. More recently improved satellite and sensor technol-
legends focusing on morphology and current processes ogy, such as High Resolution Stereo Camera (HRSC) or
8 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview

LiDAR technology (see Section 14.2.3.2), offer interesting

possibilities for geomorphological mapping applications
(Otto et al., 2007; Smith and Pain, 2009). These remote
sensing applications, in combination with high-resolution
DEMs, enhance digital landform mapping through improved
visualization possibilities leading to less biased, reproducible
maps (Smith et al., 2006; Smith and Clarke, 2005; Hillier and
Smith, 2008). Furthermore, these new data allow the auto-
matic or semiautomatic detection and classification of
landforms or geomorphic process units to a great extent
(Schneevoigt et al., 2008; Schneevoigt and Schrott, 2006;
Van Asselen and Seijmonsbergen, 2006; Seijmonsbergen et al.,
2011). For example, Advanced Spaceborn Thermal Emission
and Reflection Radiometer (ASTER) and shuttle radar topo-
graphy mission (SRTM) DEMs are now frequently used to
quantify the distribution of glaciers or rock glaciers over
large areas (Bolch and Kamp, 2006; Brenning and Azocar,
2010).

14.2.2.2 Shallow Coring and Sampling

14.2.2.2.1 Coring
Shallow coring (normally between 1 and 20 m) is frequently
applied in geomorphological research, basically for the fol-
lowing reasons: Figure 1 Sediment coring in a high alpine bog. A specially
constructed mounting rack has been developed to facilitate the
1. to analyze stratigraphic sequences of different sediment removal of the vibro-corer and the sediment cores.
and rock properties (e.g., soil depth, bulk density, water/ice
content, pollen, etc.);
2. to measure sediment thicknesses and total depth to
bedrock; surface, to outcrops, or in excavated pits (trenches) and tun-
3. to extract and analyze organic material or specific sedi- nels (see also Chapter 14.12). Apart from clastic sediment,
ment/ice layers (e.g., peat, fossil organic layers, macro- sampling of water and ice is also common in many geo-
fossils, feldspar, or quartz grains) for dating purposes (14C, morphological studies. After a first analysis in the field (e.g.,
Optically Stimulated Luminescence); and Munsell charts and photo documentation) the samples are
4. to instrument boreholes with sensors for measurement and normally taken to the laboratory for further analysis
monitoring purposes (e.g., temperature, inclinometer, etc.). (e.g., grain-size distribution (see Chapter 14.19), bulk density,
Various techniques and equipment for shallow coring are organic carbon content, chemical analysis, etc. (see also
available (see Chapters 14.11 and 14.15). The hand or soil Chapter 14.23).
auger is the most basic of the mechanical drills. Sophisticated Most samples, and specifically the changes in volume and
techniques are generally driven by engines and hydraulic physical/chemical properties of earth material, are used as
components (see Figure 1). Numerous and various systems proxies to interpret variations (temperature, humidity, and
(hammer, weights, etc.), rods, and drill points are used for sedimentation) through time. Many studies have shown that
specific purposes. continental sedimentary deposits (glacial, periglacial, fluvial,
In periglacial geomorphology, for instance, coring and aeolian, and lacustrine) carry to some degree a paleoclimatic
borehole instrumentation is very common since thermal signal (Souch, 2003).
subsurface conditions are important sources of information The number of necessary samples and the design of sample
(e.g., thermal state of permafrost), whereas in fluvial geo- selection can vary considerably from task to task. Bias and
morphology sequences of different sediment layers may dis- measurement errors in sampling are crucial issues and should
play changing sedimentation patterns (see also Chapter 14.7). be minimized through careful sampling strategies and treat-
In landslide research, coring is an in situ testing method to ments. Rice (2003) pointed out that a good sample fulfils two
measure geotechnical properties with the cone penetration test criteria: the sample provides unbiased and precise estimates of
(CPT). In sediment budget studies coring allows, in combin- the parameter of interest. Accuracy of age-depth profiles, for
ation with dating results of organic samples, estimates of example, strongly correlates with the size of sufficient and
sedimentation rates. correctly dated samples (see also Chapter 14.31). In general,
drawing conclusions from only a few samples should be
avoided. A short overview regarding sampling strategy and
14.2.2.2.2 Sampling further literature is given by Rice (2003) and a good practice
Sampling of soils, sediment, or debris is frequently combined guide for sampling in glacial and periglacial environments is
with coring (core sampling) and is also applied directly at the given by Hubbard and Glasser (2005).
 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview 9

14.2.3 Modern Field Techniques in Geomorphology codes, enabling a more accurate positioning of up to a few
millimeters (Abidin, 2002).
This section presents modern techniques that have been most Application of GPS in the field depends on the research
applied and influenced by advances in geomorphological re- question and the accuracy demands. In general, two techni-
search over the past two decades. Given the developments ques of high-accuracy GPS measurement can be distinguished:
(essentially technical), various fields within geomorphology static and kinematic surveying (Rizos, 2002). Both techniques
have changed and improved dramatically in terms of quantity require two dual-frequency receivers for signal correction,
and quality of data, namely: allowing for a differential positioning (referred to as differ-
ential GPS). What distinguishes them is the timing of the
1. surveys of high accuracy in spatial positioning and refer-
signal correction, which is either performed in real time dur-
encing of landforms, objects, or processes by using DGPS;
ing the survey or after the survey as postprocessing. Static
2. spatial coverage and digital high-resolution representation
survey involves postprocessing correction of two receiver
of relief with DEMs using LiIDAR technology;
signals. One receiver needs to be located at a known position
3. subsurface investigation (bedrock, sediment, water, and
(base station), whereas the other must be positioned on the
ice) using field geophysics; and
location measured (rover station). Both receivers record at the
4. surveys combining the above-mentioned techniques to
same rate over a longer period of time (usually minutes
investigate comparative capabilities and to improve the
to hours). Later the signal is corrected by comparing the
accuracy of data.
two recorded data sets. From the distance between the two
The intention is to give an overview to advances in typical receivers (i.e., the baseline) and the known position of the
applications rather than to discuss the technical specification base station, the exact position of the rover can be calculated.
of the different manufacturers (please refer to technical Static survey mode is applied for the exact positioning of
handbooks). single objects or repeated monitoring, for example, in land-
slide monitoring (Squarzoni et al., 2005; Gili et al., 2000).
In kinematic survey mode, correction of the signal is delivered
in real time (real-time kinematic). Thus, the rover can move
14.2.3.1 Principles of GPS and Applications in
independently and it receives both the satellite signal and the
Geomorphology
correction signal from a reference station or base station. Several
Global referencing is fundamental to most mapping and other methods exist to transmit the correction signal, ranging from
applications in geomorphology. Within the past few years GPS short-distance radio transmission of nearby base stations to
receivers have been widely distributed, ranging from car navi- long-distance signals of permanent recording stations. The latter
gation systems to portable (mobile) phones. Precise and more- is commonly provided as a commercial service and inherently
or-less accurate positioning on the Earth’s surface is now comes with extra cost. The accuracy of this signal correction
available almost globally thereby largely benefiting geo- generally declines with increasing distance to the reference
morphological research. Applications of GPS in geomorphology station (baseline). Applications of real-time kinematic GPS
range from digital field mapping to landform monitoring and survey include, for example, landform mapping (Higgitt and
more recently kinematic analysis. Due to the technical prin- Warburton, 1999) or measurement of surface movement from
ciples of GPS, the local position can be accurately determined creeping mountain permafrost (Lambiel and Delaloye, 2004).
from several meters to a few millimeters. Although low-accuracy Even though GPS technology should grant the user their
positioning is performed by low-cost commercial handheld global position, some surface conditions can reduce the
GPS devices (e.g., Garmin or Magellan) and mobile phone re- availability of satellites and data accuracy. GPS accuracy in
ceivers, higher accuracy requires more sophisticated and costly areas close to steep rock faces, for example in high relief areas,
equipment and the use of several receivers or reference data. and under dense vegetation, satellite signals are commonly
Global positions are determined by triangulation using lost or jammed when the elevation angle to the satellite be-
radio signals from 24 orbiting satellites. The radio waves comes too small. Additionally, the received signal may be re-
broadcast different codes consisting of a pseudorandom bin- flected by surrounding walls, leading to multipath effects and
ary sequence that is unique to each satellite, with information delayed travel times. In remote terrain, commercial corrections
on satellite identification and a carrier phase for data trans- signals are generally not available.
mission (Abidin, 2002).
The pseudorandom codes have a fixed frequency and
timing determined by the satellites atomic clock. GPS receivers
14.2.3.2 LiDAR in Geomorphology
generate an identical code, which is correlated to the trans-
mitted satellite code. The time delay between the two signals is LiDAR surveys (terrestrial laser scanning (TLS) or airborne
used to calculate the distance between the two devices. This laser scanning (ALS)) are increasingly being used to generate
delay between the codes represents the travel time of the radio DEMs of high accuracy and resolution. These DEMs can be
waves, which allows for the calculation of the distance range. represented as regular raster information (grids) or as trian-
The code is transmitted at two different frequencies, the gulated irregular networks (TINs) (Hofle and Rutzinger, 2011;
‘Coarse Acquisition Code’ (C/A-code) and the ‘Precision Code’ Liu, 2008). They enable one to answer not only established
(P-Code). Low-accuracy receivers (single-frequency receivers) geomorphological research questions with a higher level of
use only the C/A-code allowing for an accuracy of up to precision, but also formerly unsolved geomorphological
3–5 m, whereas dual-frequency receivers can process both problems can be addressed today using LiDAR technology, as
10 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview

for example, in snow depth modeling (Grünewald et al., 2010) reflected waveform. A single laser beam strikes, for example, a
Current geomorphological studies dealing with LiDAR data leaf of a tree producing a first peak in the reflected waveform
operate on a variety of spatial scales and cover a wide range of (first echo), whereas secondary echoes (of lower, higher, or
topics. For example, LiDAR data enable semiautomated geo- equal intensities) might be received from branches, shrubs or
morphological mapping (van Asselen and Seijmonsbergen, finally from the reflected proportion of the ground surface
2006) and the determination of size, shape, or roughness of underneath. Whereas most popular TLS systems only receive
different land surfaces with a high level of detail (e.g., gravel the last peak exceeding a certain energy threshold (the so-
bar grain sizes (Entwistle and Fuller, 2009) or mass movement called last pulse systems), full range or full wave LiDAR sys-
morphology (Glenn et al., 2006; McKean and Roering, tems record more (up to five) echoes.
2004)). LiDAR data are commonly also used as accurate base Technical specifications of modern laser scanning systems
data for different modeling approaches (e.g., flood modeling are quite remarkable. With an operating distance of up to
(Overton et al., 2009) or for the reconstruction of relict gla- several kilometers, ten or more thousands of points can be
ciation patterns (Kovanen and Slaymaker, 2004)). Further- measured per second with an accuracy of millimeters to
more, repeated laser scan surveying provides a tool for centimeters. Factors controlling the accuracy of LiDAR data are
multiple process monitoring strategies of large areas. As many (Baltsavias, 1999) and can be differentiated in range
probably one of the fastest growing fields in geomorphology, accuracy issues (energy thresholds of returned signals, beam
recently monitoring studies using LiDAR data are focusing on divergence, scan angles, object distance, reflectivity, and
numerous Earth’s surface processes, for example, landslide topography of target) and position accuracy problems using
and mass movement activity (e.g., Prokop and Panholzer, (differential) GPS systems.
2009; Staley et al., 2006), rockfall on steep rock walls and cliffs
(e.g., Oppikofer et al., 2008; Rosser et al., 2005; Rabatel et al.,
2008 Heckmann et al., 2012), glacier or rock glacier creep 14.2.3.2.2 Airborne or terrestrial laser scanning
(e.g., Kennett and Eiken, 1997; Baltsavias et al., 2001; Geist The appropriate mode of data acquisition (i.e., airborne or
et al., 2003; Tamburini et al., 2005; Avian et al., 2009; Haas terrestrial) naturally depends on the geomorphologic research
et al., 2010) and others. Analysis of debris-flow depositional question, but the following aspects are of particular import-
patterns using LiDAR data was also conducted by Staley et al. ance in this regard.
(2006). Geist et al. (2009) gives an overview of principles and Size of area under investigation: Small-scale studies (ap-
laser scanning applications focusing on the significance of proximately o1–5 km2) and especially repetitive small-scale
high-resolution topographic data for natural hazard manage- investigations using identical survey designs will be typically
ment. LiDAR data in hydrological research are used to analyze carried out using TLS systems. If (due to resolution) TLS have
erosion and deposition of river systems (Lane et al., 2003) and to be applied on larger scales (45–10 km2) intensive field
for hydrologic modeling (French, 2003). However, the cap- work incorporating many scan positions is required. There-
ability of the LiDAR system strongly depends, among other fore, larger spatial scales are typically handled using ALS data.
factors, on the wavelength of the laser type applied. This may Data resolution: Subject to the phenomenon to be studied,
cause limitations in the applicability in wetlands or on snow- the desired data resolution should be determined first.
covered surfaces, due to weak reflectivity conditions. Besides Whereas TLS surveys generally achieve 1000 or more points
these limitations and in spite of the obvious wide field of per square meter, ALS typically delivers just 1–10 point m2.
application, it is should be mentioned that LiDAR technology, Final gridded DEM resolution ranges between approximately
especially portable TLS systems, is among the most expensive 1 m (ALS) and 1–50 cm (TSL).
field tools in geomorphology (Rayburg et al., 2009). Topography of the study site: Due to acute angles of impact,
ALS is unfavorable to reproduce vertical rock faces or over-
14.2.3.2.1 Principles of LiDAR hangs. If possible, right angles of impact should be sought and
LiDAR data acquisition is generally based on the two-way the area of interest should be in the near vicinity of the laser
travel time or time-of-flight principle. Emitted laser pulses scanner’s position. Heterogeneous relief conditions generally
propagate through space with the constant velocity of light. complicate the decision of scan positions owing to shadowing
After a certain time (or distance, respectively) laser beams will effects.
strike a surface and get reflected (specular, diffuse, or mixed), Remoteness and accessibility of the study site, availability of
transmitted, or absorbed. Parts of the reflected proportions energy supply, and transport facilities: These critical factors
will be received again by the instrument allowing for exact commonly control the feasibility of a survey design to a large
distance measurements relative to the scanner’s position. part. TLS equipment typically weighs approximately 20 kg and
Combined with the vertical and horizontal angles of each laser can be transported in the field using special carrying devices.
pulse, the typical output of laser scanners are point clouds Energy supply also has to be guaranteed (generator, solar
containing X, Y, Z triples in a scanner’s own coordinate system panel).
commonly supplemented by intensity, timestamp, and add- Weather conditions: Clear and dry conditions are best suited
itional RGB (color) information if a (calibrated) camera is for scanning surveys. Rain, snow, and fog may produce in-
used (Heritage and Large, 2009). creased noise in the point cloud because of reflections at water
Due to their diverging behavior, the footprints of laser droplets. Local wind speed is another crucial factor. Due to
beams grow as distance to target increases. This effect (1) unstable conditions, some laser scanners stop the actual
causes lower data accuracy and resolution with increasing measurement if a certain threshold of wind speed is exceeded.
object distance; and (2) might produce multiple peaks in the Scanning at night is possible but may cause problems in terms
 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview 11

of defining the area to be scanned. Furthermore, RGB infor- questions these ‘raw’ LiDAR data have to be further processed
mation using a calibrated camera cannot be acquired. in order to gain regular gridded elevation data with a homo-
Availability of equipment, financial scope, and available time geneous resolution or TINs, which is usually done using dif-
frame: In some countries, area or nation-wide ALS data ac- ferent filtering and thinning options (e.g., 2.5-D filter, octree
quisition currently takes place through public authorities. approaches). If required, removing vegetation is another im-
Resulting ALS products are frequently available for research portant issue to be managed using specific filters.
institutions at reduced prices. Private companies can also be
commissioned with ALS surveys. If ALS data are not available,
not affordable, or show too low a resolution, TLS should be 14.2.3.3 Geophysical Applications in Geomorphology
considered.
Although geomorphology has always been interested in sub-
surface structures because it is an important component in
14.2.3.2.3 LiDAR – principles of data acquisition and
understanding landforms in space and time, subsurface in-
processing
formation is generally limited to very shallow insights (e.g.,
As ALS data acquisition is generally done by private companies
using a soil auger) reaching not more than 1 m depth.
or public authorities, this section – dealing with practical
Recently, the use of geophysical techniques has become
issues regarding data acquisition and processing – focuses on
increasingly important in many geomorphological studies.
TLS systems. TLS surveys commonly comprise several scan
One reason for the interest in geophysical field methods is
positions. Their locations have to be chosen carefully and (if
certainly related to technical innovations as improved com-
possible) prior to laser scanning measurements. The locations
puter power and the availability of lightweight equipment that
should ensure that shadowing effects are limited to a min-
allows for relatively user-friendly, efficient, and nondestructive
imum, that angles of impact are as large as possible, and that
data gathering. However, correct handling of the geophysical
the whole area of interest is covered homogeneously. This
instruments and subsequent data processing are still difficult
decision demands some experience, but GIS tools, as, for ex-
tasks and the methods generally require advanced math-
ample, viewshed analyses (ArcGIS), might help in this regard.
ematical treatment for interpretation. However, without close
At each scan position, the area to be scanned and the desired
collaboration between geomorphologists and geophysicists
scanning resolution have to be defined. Scanning designs
the accurate and effective use of geophysical techniques and
using reflectors (in many cases fixed in the field with known
their geophysical and geomorphological interpretation are
GPS coordinates) commonly demand separate fine scans of
generally very limited. In addition, correct description and
them with the highest possible resolution. Reflectors are used
interpretation of geomorphological settings and, thus, the
for many purposes. Besides an accurate registration process,
choice of adequate and meaningful field sites are not easy
fixed reflectors in the field can simplify the comparison of
tasks.
repeated scan series (e.g., cut and fill analyses) and – if
Although several potential applications for geophysical
installed on ‘moving objects’ – the assessment of geomor-
methods exist, many of them have not yet been fully inte-
phologic processes (e.g., rates of permafrost creep, solifluction,
grated into geomorphological research. The most common
and landslides). Some systems also capture overlapping digital
geophysical applications are currently focusing on permafrost
photographs covering the area of interest using a calibrated
mapping, sediment thickness determination of talus slopes,
camera. These photographs allow for colorizing point clouds,
block fields, alluvial fans, coastal sedimentary structures
texturing of meshes, or for creating orthorectified photographs
and, increasingly, on the depth and internal structures of
of the study site.
landslides (Hecht, 2000; Tavkhelidse et al., 2000; Hauck,
After data acquisition, each scan position’s point cloud has
2001; Leatherman, 1987; Hoffmann and Schrott, 2002; Hauck
to be merged with each other (often called the registration
and VonderMühll, 2003; Israil and Pachauri, 2003; Kneisel
process). This implies a rotation and translation of the point
and Hauck, 2003; Schrott et al., 2003; Bichler et al., 2004; Sass
clouds to be registered as a whole, but one which would not
et al., 2008). Other landforms such as karst and colluvium are
change the relative positions of points within them. The
comparatively rarely investigated (Hecht, 2003). Currently, the
registration of point clouds can be done in different ways, as
most common geophysical methods in geomorphological re-
for example with (a minimum of three) reflectors visible in
search are ground-penetrating radar, DC resistivity, and seis-
two overlapping scenes. Additional tools, depending on the
mic refraction (Gilbert, 1999). Each geophysical technique is
software package used, enable further (stepwise) registration
based on the interpretation of contrasts in specific physical
processes without using reflectors, as for example the coarse
properties of the subsurface (e.g., dielectric constant, electrical
registration by the manual setting of identical points or
conductivity, and density). The type of physical property to
by using north angles (back sighting orientation) and the
which a particular geophysical method responds determines
subsequent semiautomatic matching of polygon normals.
and limits the range of applications.
The latter is probably the most accurate way and also recom-
mended for point clouds already registered by means of re-
flectors. In contrast to the different locations relative to each 14.2.3.3.1 Ground-penetrating radar
scan position, all points are referred to one common project 14.2.3.3.1.1 Principle and geomorphic context
coordinate system after the registration process. Incorporating Ground-penetrating radar uses high-frequency electro-
GPS data of reflectors or scan positions finally allows for data magnetic waves to acquire information on subsurface com-
conversion in a global coordinate system of choice (e.g., position (see also Chapter 14.16). The electromagnetic pulse is
UTM). To answer numerous geomorphological research emitted from a transmitter antenna and propagates through
12 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview

the subsurface at a velocity determined by the dielectric indicator of potential target depth. As the reflectivity at a layer
properties of the subsurface materials. The pulse is reflected by boundary is determined by the contrast in the dielectric
inhomogeneities and layer boundaries and is received by a properties of the subsurface units, no distinct reflection is
second antenna after a measured travel time. Wide-angle re- found when this contrast is low. Small-scale spatial differences
flection and refraction (WARR) or common midpoint (CMP) in water content and grain-size composition may yield
is measured to calculate depth/resistance/velocity relation- stronger reflections than the target of the investigation (e.g.,
ships. These signal travel time measurements are made with a the bedrock surface or palaeosols). This problem may be
stepwise increase in the distance between the two antennas. overcome by using the radar facies of the sediments for in-
From the distance/travel time diagram, the propagation vel- terpretation. Different sediment units and bedrock yield
ocity of the radar waves in the subsurface can be derived. One typical reflection patterns that can be derived from, for ex-
common mode of GPR data collection is fixed-offset reflection ample, reference profiles. The reconnaissance of these patterns
profiling (Jol and Bristow, 2003). In this step-like procedure, significantly facilitates the interpretation of the radargram.
the antennas are moved along a profile line and the meas- Considering the major restrictions arising from ‘clayey or
urement is repeated at discrete intervals resulting in a 2-D silty subsurface’ and ‘wooded terrain’ it is clear that GPR shows
image of the subsurface. The possible working frequencies can its potential particularly in arctic or alpine areas above the
range from 10 MHz to 1 GHz depending on the aim of in- tree-line and where there is limited soil development. How-
vestigation. Higher frequencies allow higher spatial resolution ever, shallow subsurface investigations are also possible in
of the ground information, but lead to a lower penetration fluvial deposits and even in peat when the electrical con-
depth. ductivity of the groundwater is low.
Correct choice of antenna and measurement strategy has There is a broad range of successful applications of GPR in
important implications in geomorphological applications. geomorphological studies. These include the detection of
Knowledge about the geomorphological context (expected buried structures, assessment of internal sediment structures,
maximum depth and grain-size composition of a sediment and estimation of depth to bedrock (Table 2). Various types of
body) is essential for choosing the appropriate frequencies. A sediments have been investigated for geomorphological pur-
comprehensive ‘good practice guide’ for the application of poses (Bristow et al., 2000). The internal structures of flood-
GPR in sediments is provided by Jol and Bristow (2003). plain deposits and deltaic sediments have been visualized by,
The maximum depth of investigation depends mainly on for example, Leclerc and Hickin (1997); Jol (1996), and Büker
the dielectric constant (e) and the electrical conductivity (s) of et al. (1996). Overgaard and Jakobsen (2001) have investi-
the subsurface. A higher groundwater and/or clay content gated internal deformation structures of push moraines and
(high e, high s) leads to a stronger attenuation and, therefore, Berthling et al. (2000) clearly detected internal structures as
a markedly reduced penetration depth. However, very pure well as the bedrock base of rock glaciers. Sass and Wollny
groundwaters that have a low conductivity (e.g., glacial melt- (2001) and Sass (2006) achieved a penetration depth of up to
water and bogs) are characterized by relatively low levels of 50 m on talus slopes using 25-MHz antennas. Application of
GPR signal loss. Again, geomorphological expertise about GPR on landslides has repeatedly been tested but with limited
possible water and clay layers can help to avoid disappointing success. Bichler et al. (2004) obtained very good results, dis-
measurements. On dry and electrically high-resistive debris, tinguishing seven different facies of loose sediments. GPR
penetration between 30 and 60 m can be achieved (e.g., Smith measurements at landslide sites are only worthwhile where
and Jol, 1995). Sandy sediments are also favorable for GPR comparatively coarse and dry deposits (debris, displaced
measurements at depths of between 15 and 30 m. Due to the blocks) superimpose the silty or clayey material of the slip
strong attenuation in materials that have a high electrical surface. Another possible field of application is the investi-
conductivity the penetration depth in wet, silty, or clayey gation of permafrost features. The active layer thickness has
sediments diminishes rapidly to, for example, less than 5 m in been determined, for example, by Arcone et al. (1998) and
silt (Doolittle and Collins, 1995); in clayey soil the appli- Hinkel et al. (2001). Moorman et al. (2003) provided in-
cation of GPR may be altogether impossible (see Table 1). structive pictures of typical reflection patterns in frozen and
This is, however, only a very rough guide, because the pene- unfrozen ground. Although the presence or absence of
tration depth depends on the device and antenna frequency permafrost is more difficult to establish with GPR than with
used. Vertical resolution of GPR data is a function of frequency electrical resistivity techniques, GPR is superior in detecting
and propagation velocity. With higher velocities, resolution spatially confined structures such as ice wedges (Hinkel et al.,
decreases and vice versa. As a rule of thumb, in the medium- 2001). Thickness and internal structure of glacier ice (unfrozen
velocity range (0.1 m ns1) the resolution is approximately water content and cavities) have also been investigated using
1 m using 25 MHz antennae, 0.25 m using 100 MHz GPR (e.g., Moorman and Michel, 2000). Investigation of quasi
antennae, and 2.5 cm using 1 GHz antennae (Figure 2). point-shaped or linear buried structures in high resolution
is the aim of many studies in the relatively new field of
14.2.3.3.1.2 Advantages and disadvantages of GPR in geoarchaeology that is closely related to geomorphology
geomorphological applications (Baker et al., 1997; Fuchs and Zöller, 2006). Leckebusch
The available antenna frequencies allow for a broad variety of (2003) has provided a detailed description of the GPR method
possible applications. However, the extremely variable pene- for archeological purposes with numerous examples. The
tration depth requires careful assessment of the subsurface working frequencies for these applications are commonly ra-
parameters in the study area to minimize the risk of poor ther high (c. 200 MHz); the target depth is generally between
results. Electric conductivity of the soil provides a rough 1 and 5 m.
 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview 13

Table 1 Comparison of common field geophysical methods in geomorphology. Examples of applications and some technical considerations and
advice

Geophysical Geomorphological application Technical considerations and practical advice

technique

Ground-penetrating • Delineation of the boundaries of massive ice in rock • Small penetration depth in case of conductive near-
radar (GPR) glaciers, moraines and other periglacial phenomena surface layers
• Determining the thickness of a permafrost layer • Less successful in, for example, wooded terrain with
• Active layer thickness many surface reflectors and in electrically conductive
• Glacial ice thickness sediments (e.g., clay-rich)
• Delineation of aquifers • Experience in data processing and interpretation
• Fracture mapping within massive bedrock is needed
• Mapping of internal structures in sediment storage • Correlate radar data with geomorphic/geologic control
types (e.g., talus slopes and rock glaciers) and site observations such as exposures or
borehole logs
DC resistivity • Determining sediment thickness, internal • Obtaining good electrical contact between electrodes
sounding, 2-D DC differentiation and groundwater and ground is essential
resistivity profiling • Depth and extension of landslide bodies • Blocky and dry material is very problematic due to poor
• Detection of massive ice or permafrost in rock glaciers, electrode coupling
moraines and other periglacial phenomena • Experience in data inversion is needed for data
• Ice thickness processing. A priori information influences (improves)
• Determining the altitudinal permafrost limit iterative modeling significantly
• Moisture distribution in rock walls • Differentiation between ice, air, and special rock types
• Seasonal variation in fluid content can sometimes be difficult
• Correlate resistivity data with geomorphic/geologic
control and site observations such as exposures or
borehole logs
Seismic refraction • Sediment thickness (talus, alluvial fans, etc.) • Number of geophones should be at least 12, with shots
(2-D sounding, • Detection of massive ice in rock glaciers, moraines, at every other receiver location
tomography) and other periglacial phenomena • Sledge hammer as source is sufficient for most
• Differentiation between ice, air, and special rock types shallow applications (o30 m)
• Mapping active layer thickness • Experience in data processing and inversion is needed
for data interpretation
• Correlate seismic data with geomorphic/geologic
control and site observations such as exposures or
borehole logs

Source: Reproduced from Schrott, L., Sass, O., 2008. Application of field geophysics in geomorphology: advances and limitations exemplified by case studies. Geomorphology 93,
55–73.

14.2.3.3.2 Geoelectrical resistivity principle of operation is based on a stepwise iteration process

14.2.3.3.2.1 Principle and geomorphic context that tries to minimize the deviation between the measured
Resistivity is measured by applying a constant current into the apparent resistivity and the simulated apparent resistivity
ground through two ‘current electrodes’ and measuring the values calculated from a subsurface model. The 2-D section
resulting voltage differences at two ‘potential electrodes.’ can also be conducted in a Schlumberger array, which
From the current and voltage values, an apparent resistivity provides particularly good resolution for lateral inhomo-
value is calculated. Generally, the two potential electrodes geneities. The Wenner array shows a good signal–noise
remain in position in the middle of the profile, whereas the ratio and is favorable for the detection of horizontal
current electrodes move stepwise to both sides (Schlumberger layers, whereas the Dipole–Dipole array is favorable for the
array). The wider the electrodes are apart, the deeper the delimitation of spatially confined objects in the shallow
electrical field penetrates into the ground. Thus, a depth subsurface. For a comprehensive description of these con-
profile under the approximate middle of the section is cre- figurations, refer to geophysical textbooks (Milsom, 1996;
ated. This array has been widely used in geomorphologic Reynolds, 1997; Kearey et al., 2002).
research (Etzelmüller et al., 2003). Two-dimensional arrays The maximum amount of a priori information on the
represent a further development of the 1-D technique, using geomorphological context should be obtained (e.g., layer
50 or more electrodes at a time. A microcontroller unit thickness, bedrock type, and expected resistivity value) before
automatically switches between numerous electrode con- starting the inversion of the raw data. The a priori estimate
figurations, thus creating a 2-D pseudosection of the sub- (e.g., maximum resistivity value of an expected type of sedi-
surface. Inversion of the gathered data using sophisticated ment) can be set as a fixed parameter and helps to improve the
computer programs produces a 2-D section of the subsurface model. The required information can be derived from test
resistivity. The Res2Dinv-software provided by Loke and profiles of, for example, bedrock or sediment units of known
Barker (1995) is most commonly used in geosciences. The composition (Figure 3).
14 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview

Fixed-offset reflection mode

T1 T2 R1 R2 Tn Rn
Control unit &
antennas

Depth
T R Profile direction Amplitude Layer 1
Layer 2
r1
I WARR mode
T R1 R2 Rn
Layer 1
1 1

Travel time [ns]

Depth
d1 t1
r2 r1
Layer 1
Depth

Layer 2
Layer 2
2 2 CMP mode
d2 t2
Tn T2 T1 R1 R2 Rn
Layer 3 r2
3 3

Depth
Layer 1
Layer 2
(a) (b) Common midpoint

Offset [m]
0 10 20
0
1
100
Travel time [ns]

200 2
300 1 air wave ν = 0.3 m/ns
3
4
400 2 Ground wave ν = 0.12 m/ns 5
500 3 ν = 0.058 m/ns 6
600 4 ν = 0.064 m/ns
5 ν = 0.08 m/ns
Location of CMP measurement
700
6 ν = 0.084 m/ns

Profile distance
0 100 200 300
0 0

1D velocity model]
Travel time [ns]

Depth [m, using

200
10
400
20
600

800 30

0 100 200 300

2080 0
Elevation a.s.l. [m]

2070
Depth [m]

10

2060 20

2050 30
(c)

Figure 2 (a) Principle of a GPR survey (T – transmitter, R – receiver, d1 – depth to reflector 1, d2 – depth to reflector 2, e – permittivity,
s – conductivity, t1 – two-way travel time of the electromagnetic pulse to reflector 1, t2 – two-way travel time of the electromagnetic pulse to
reflector 2). Modified from Blindow, 2005. (b) GPR survey modes. The fixed-offset reflection mode is common for GPR profiling where
transmitter and receiver are moved along a profile line. Pulse radiation and measurement are repeated in a discrete trigger interval resulting in a
2-D radargramm of the subsurface. To estimate the depth of reflections and boundaries, the subsurface propagation velocity of the EM wave is
needed and derived by CMP or WARR measurements. When performing a CMP measurement, the distance between transmitter and receiver is
stepwise increased over a CMP whereas in WARR mode just the receiver is moved whereas the transmitter remains stationary. With both
measurements a distance/travel time diagram is created for the depth calculation. (c) Results of the GPR measurement at a sandur in the
forefield of Pasterze glacier (Austria). Processed radargramm (top) and geomorphological interpretation (bottom) including supposed bedrock
surface (red line) and prominent reflections indicating internal structures (black lines). Migration and time–depth conversion is based on the
propagation velocities of a nonlinear 1-D velocity model derived from a CMP measurement (top left).
 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview 15

Table 2 Common geophysical techniques in geomorphological applications and qualitative assessment in terms of suitable detection of
thickness and/or internal structure. Note: the indication of suitable application may be incorrect under specific or extreme local conditions (e.g.,
wet, dry and blocky)

Physical property Ground-penetrating radar 2-D DC resistivity 2-D seismic refraction

Dielectrical properties Resistivity Elastic moduli, density

Geomorphic context
Talus slopes, debris cones þþa/þþb 7/ þ þþ/7
Block fields þ/þ 7/o þ /o
Alluvial fans, floodplain þ /þþ þ/þ þþ/7
Colluvial þ/þo þ /o 7/ 
Landslides  c/7c þ /þþ 7/7
Karst features 7 o o
Permafrost lenses þ þþ 7
Active layer thickness 7 þþ þþ
Permafrost occurrence (widespread) þ þþ þ
Rock glaciers 7/ þ 7/þþ d
/7
Rock/soil moisture distribution 7 þ 
a
Thickness, depth-to-bedrock.
b
Internal structure/distribution.
c
Only suitable in dry sediments.
d
Unsuitable on active rock glaciers with a permafrost body.
Source: Reproduced from Schrott, L., Sass, O., 2008. Application of field geophysics in geomorphology: advances and limitations exemplified by case studies. Geomorphology 93,
55–73.

14.2.3.3.2.2 Advantages and disadvantages in geomorphological gravel, and bedrock, at different velocities. The denser the
applications material, the faster the waves travel. A prerequisite for the
A great advantage of the method is the high variability of successful application is that each successive underlying layer
electrode spacing and configuration. The distances between of sediment or bedrock must increase in density and therefore,
the electrodes can range from some centimeters to several tens velocity. The seismic velocity of P waves depends on the elastic
or hundreds of meters, allowing a penetration depth from modulus and the density, r, of the material, through which
decimeters to hundreds of meters. The electrode array (e.g., the seismic waves propagate.
Schlumberger, Wenner, and Dipole–Dipole) can be chosen Propagation of seismic waves through layered ground is
according to the aim of the investigation. Almost no re- determined by the reflection and refraction of the waves at the
strictions regarding topography, subsurface features, and interface between different layers. When the wave reaches the
vegetation apply. However, very dry or extremely blocky sub- interface, some energy of the wave is refracted into the deeper
strates are basically unfavorable. Special measures have to be layer, whereas the reflected wave transmits the energy back
taken to improve the electrode-to-ground coupling such as into the overburden layer.
watering of the electrodes or inserting them through wet The critical refracted wave that travels along the interface
sponges. It is also possible to insert electrodes into hard rock produces oscillation stresses, which in turn generate upward
using a hammer drill. However, the best results are generally moving waves, known as head waves. Since the propagation
obtained on loamy and relatively wet subsurface. velocity V2 is faster within the lower layer, at a certain distance
Geoelectrical surveys always integrate the electrical prop- from the shotpoint (crossover distance) these head waves
erties of a certain volume of the subsurface. The extent of this reach the surface (geophones) faster than the direct wave
volume increases in the deeper subsurface. Thus, accurate providing the first arrivals at the geophones. In seismic re-
detection of sharp boundaries is rarely possible especially in fraction studies, it is the first arrivals of the P-waves that are
the deeper parts of a profile. Accurate assessment of depth to utilized.
bedrock is only possible where there is a very sharp resistivity The arrival of a seismic wave is detected by geophones,
contrast between overlying loose sediments and bedrock. which are placed firmly in the ground using a spacing of be-
A further, significant problem is related to significant tween 1 and 5 m. If the expected depths of the interfaces are
overlapping ranges of resistivities for different substrata. Ac- shallower than 30 m the geophone spacing may be reduced
cording to textbook tables, resistivity values of almost every (i.e., between 1 and 3 m) to gain a higher resolution record of
subsurface unit (regardless of loose sediment or bedrock) may the underground. To detect a layer within the time–distance
cover a range of several orders of magnitude, depending on plot the geophone spacing must be smaller than the difference
water content and jointing. Thus, a measured resistivity cannot between two following crossover distances. At the crossover
be directly assigned to a certain substrate. distance, the direct wave is overtaken by a refracted wave.
Beyond this offset distance the first arrival is always a refracted
14.2.3.3.3 Seismic refraction wave. In refraction surveying, recording ranges are chosen
14.2.3.3.3.1 Principle and geomorphic context to be sufficiently large enough to ensure that the crossover
The principle of seismic refraction is based on elastic waves distance is well exceeded to detect a sufficient number of first
traveling through different subsurface materials, such as sand, arrivals of refracted waves. In cases of thin subsurface layers,
16 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview

ΔV

A M N B

Current lines
Potential lines
 Resistivity [Ωm]

(a) (b)

Resistivity [Ωm]
10 Point of intersection
35000
Depth [m]

5 20000
East West
17500
0 Rock glacier Profile length: 96 m 15000
coarse, blocky iterations: 4
–5 RMS error: 5.1%
12500
model reflnment applied 10000
–10 smoothed inversion
7500
6000
10 20 30 40 50 60 70 80
5000
Profile distance [m] 4500
4000
10 Point of intersection 3500
3000
5 Rock glacier 2500
Depth [m]

North South
coarse, blocky 2000
0 1500
Fine grained
Profile length: 96 m 1000
–5
iterations: 4 0
abs. error: 3.7%
–10 robust inversion
Array type: wenner
number of electrodes: 25
10 20 30 40 50 60 70 80 unit electrode spacing: 4 m
Profile distance [m] inversion software: res2dinv
(c) date of data acquisition: 24.07.2009

Figure 3 (a) Configuration of a four-point array. With resistivity measurements a constant current (I) is conducted into the ground by two
current electrodes and the resulting potential difference (DV) is measured by two potential electrodes. For 2-D measurements an array of
electrodes is used at a time and a controller unit switches between the electrode configurations resulting in a linear depth profile or
pseudosection of the resistivity variation with depth and along the profile line. When applying a Wenner configuration (cf. section C) the
minimum electrode spacing is used initially and the array moves along the profile until the last electrode is reached. Electrode spacing is then
stepwise increased and the process is repeated providing a pseudosection. (b) Electrical resistivity measurement at the Gradental
(Schobergruppe, Austria) with the controller unit (red case) and field notebook. (c) Intersecting electrical resistivity tomographies at a high alpine
periglacial test site (approximately 2800 m asl; Glatzbach catchment, Austria; date of acquisition: 24 July 2009). Large resistivity contrasts are
based on different grain sizes. Note also very large resistivity values above 35 kO indicating subsurface permafrost or solid ice.

where the distance between crossover points is small, a narrow by hitting a metal plate placed on the ground. To improve the
geophone spacing (between 1 and 3 m) is necessary. This in signal-to-noise ratio the sledge hammer ‘shot’ is usually re-
turn may result in a lower penetration depth due to a reduced peated several times (typically between 5 and 10 times) at the
total spread. same source spot. The resulting seismograms of each shot are
The most common seismic source in geomorphological stacked (summed) manually or automatically to obtain a
studies is a 5-kg sledge hammer which generates seismic waves single seismogram per shot location.
 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview 17

Generally, a sledge hammer is sufficient to measure the be used for longer surveys with larger penetration depths. To
first arrivals along offset distances of approximately 50 m apply seismic surveys in geomorphic studies it is highly rec-
and seldom more than 90 m. This corresponds to penetration ommended to use instruments with several single light units
depths of 10–30 m and is suitable for many landforms. rather than to use an all-in-one (heavy) seismograph
More powerful sources (e.g., drop weights and explosives) must (Figure 4).

Shot location Geophone

Offset
Surface
Hea
dw
ave
Z
V1 < V2
ic
V1

i Critically refracted wave V2

Xc Profile distance [m]

Travel time [m]

ti

Direct wave Critically refracted wave

(a) slope: 1/v1 slope: 1/v2 (b)

Profile distance [m]

0 10 20 30 40 50 60 70 80 90 v (m/s)
0 p
300

600
Depth [m]

10

900

20 1200

0 10 20 30 40 50 60 70 80 90
0

20
Travel time [ms]

40

60

80
(c)

Figure 4 (a) Principle of seismic wave refraction and reflection and travel time–distance plot (ic – angle of incidence, ti – intercept time,
xc – crossover point, v1 – velocity layer 1, v2 – velocity layer 2). Modified from Brückl et al., 2005. (b) Refraction seismic survey at the Reintal
basin (Bavarian Alps, Germany). (c) Seismic travel times and modeled refractor surfaces at the Reintal basin. Refractors have been modeled
using tomography (top graph, raster image in background) and network-ray tracing (bottom graph) methods.
18 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview

14.2.3.3.3.2 Advantages and disadvantages in geomorphological extend the possibilities of landform detection and process
studies modeling.
In coarse-grained surface conditions (i.e., on talus slopes and There is a need for studies coupling the geophysical back-
debris cones) and hummocky topography (i.e., on rock glaciers ground with geomorphologic approaches in different natural
or rockfall deposits) difficulties may arise from the poor environments and for different landforms. Further investi-
coupling of geophones to the surface. On talus slopes with gations using integrated approaches are required. Only an
larger block sizes directly on the talus surface, it is advisable to appropriate combination of differing methods allows for
improve the coupling of the geophones by removing the upper much more sophisticated data interpretation. For example,
layer of larger boulders and pushing the spike of the geophone attention should be paid, not only to the thickness, but also to
into the fine-grained material underneath. Coupling to larger the internal structure of loose deposits – a task that can only
boulders or to compact bedrock may be established by drilling be achieved in sufficient accuracy with the combination of
into the rock and plunging the spike within the drill hole. several methods. Furthermore, a multimethod approach
In the vicinity of torrents and under strong wind or rainfall, allows for cross-checking of results and determination of the
detection of seismic waves may be impossible due to strong suitable methods for achieving valuable and reliable results in
noise in the seismic record. Special attention should be drawn a particular environment.
to the obtained velocities of materials that are common in However, these new types and amounts of data require a
particular landforms (e.g., talus deposit, till, and rock glacier). complex suite of processing steps and lots of technical skills. It
The large range of observed velocity values, spanning from needs to be verified whether our analysis tools composed for
c. 400 m s1 (loose debris) up to 6500 m s1 (compact rocks), data of lower resolution can still be adequately applied for
are generally conducive to the application of refraction seis- high-resolution data.
mics, as large velocity contrasts between the underlying ma-
terials are necessary. However, ranges of P-wave velocities for
rocks and sediments can overlap significantly. For instance, 14.2.5 Disclaimer
seismic velocities of dolomite and limestone can vary between
2000 and 6500 m s1 depending on the grade of fracturing Section 14.2.3.3 contains parts relating to a previously pub-
and weathering, whereas, for example, glacial sediments lished paper by Schrott, L., Sass, O., 2008. Application of field
compacted by overlying ice possibly range from 1500 to geophysics in geomorphology: advances and limitations ex-
3500 m s1. Consequently, there is no unambiguous rela- emplified by case studies. Geomorphology 93, 55–73.
tionship between certain subsurface units and the linked P-
wave velocities. As a result, it is not always possible to identify
subsurface materials simply based on seismic velocities. To References
differentiate glacial till (without ice) from frozen ground or
solid rock, cross checks with other geophysical methods and Abidin, H., 2002. Fundamentals of GPS signals and data. In: Bossler, J. (Ed.),
evidence from borehole logging are needed. Generally, land- Manual of geospatial science and technology. Taylor & Francis, London,
forms consisting of similar sediments but deposited by dif- New York.
Arcone, S.A., Lawson, D.E., Delaney, A.J., Strasser, J.C., Strasser, J.D., 1998.
ferent geomorphic processes cannot be distinguished Ground-penetrating radar reflection profiling of groundwater and bedrock in an
(Hoffmann and Schrott, 2003). Another very common dis- area of discontinuous permafrost. Geophysics 63, 1573–1584.
advantage in seismic refraction is the ‘hidden layer’ problem, Avian, M., Kellerer-Pirklbauer, A., Bauer, A., 2009. LiDAR for monitoring mass
which may lead to incorrect interpretations (Pullan and movements in permafrost environments at the cirque Hinteres Langtal, Austria,
Hunter, 1990; Hecht, 2001). The hidden layer is caused by a between 2000 and 2008. Natural Hazards and Earth System Sciences 9,
1087–1094.
sandwiched layer with lower velocities between higher velocity
Baker, A., Caseldine, C., Hawkesworth, C., Latham, A., Hatton, J., 1997. A
units of refracted first P-wave arrivals. Cromerian Complex stalagmite from the Mendip Hills, England. Journal of
Quaternary Science 12, 533–537.
Baltsavias, E.P., 1999. Airborne laser scanning: basic relations and formulas.
Journal of Photogrammetry and Remote Sensing 54, 199–214.
Baltsavias, E.P., Favey, E., Bauder, A., Bösch, H., Pateraki, M., 2001. Digital surface
14.2.4 Conclusions
modelling by airborne laser scanning and digital photogrammetry for glacier
monitoring. Photogrammetric Record 17, 243–273.
New techniques such as LiDAR, GPS, or field geophysics have Barsch, D., Liedtke, H., 1980. Principles, scientific value and practical applicability
rendered it possible to obtain new and stimulating solutions of the geomorphological map of the Federal Republic of Germany at the scale of
for geomorphological problems during the past decades. 1:25 000 (GMK 25) and 1:100 000 (GMK 100). Zeitschrift für Geomorphologie.
These techniques have become very common within geo- Berthling, I., Etzelmüller, B., Isaksen, K., Sollid, J.L., 2000. Rock glaciers on Prins
Karls Forland. II: GPR soundings and the development of internal structures.
morphology, but should not marginalize the geomorpho-
Permafrost and Periglacial Processes 11, 357–369.
logical question. Although advanced geomorphological Bichler, A., Bobrowsky, P., Best, M.G., Douma, M., Hunter, J., Calvert, T., Burns, R.,
mapping and stratigraphic analysis will continue to play an 2004. Three-dimensional mapping of a landslide using a multi-geophysical
important role in geomorphology, new high-resolution sur- approach: the Quesnel Forks landslide. Landslides 1, 29–40.
face and subsurface information from DGPS, LiDAR, or field Blindow, N., 2005. Georadar. In: Knödel, K., Krummel, H., Lange, G. (Eds.),
Handbuch zur Erkundung des Untergrundes von Deponien und Altlasten, Band
geophysics can strongly support the geomorphologists’ cap-
3: Geophysik, 2nd ed. Springer, Heidelberg, pp. 389–424.
abilities to assess spatial relationships and landform changes Bolch, T., Kamp, U., 2006. Glacier Mapping in high mountains using DEMs,
in three dimensions. Extremely detailed surface and sub- Landsat and ASTER data. Grazer Schriften für Geographie und Raumforschung
surface data will change our perspective on landforms and 41, 37–48.
 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview 19

Brenning, A., Azocar, G.F., 2010. Statistical Analysis of topographic and Hecht, S., 2003. Investigation of the shallow subsurface with seismic refraction
climatic controls and multispectral signatures of rock glaciers in the dry Andes, methods – application potentials and limitations with examples from
Chile (27 degrees–33 degrees S). Permafrost and Periglacial Processes 21, various field studies. Zeitschrift für Geomorphologie N.F., Supplement Band 132,
54–66. 19–36.
Bristow, C.S., Bailey, S.D., Lancaster, N., 2000. The sedimentary structure of linear Heckmann, T., Bimböse, M., Krautblatter, M., Haas, F., Becht, M., Morche, D., 2012.
sand dunes. Nature 406, 56–59. From geotechnical analysis to quantification and modelling using LIDAR data: a
Brückl, E., Kirchheimer, F., Krummel, K., Liebhardt, G., Reimers, L., Sandmeier, K.J., study on rockfall in the Reintal catchment, Bavarian Alps, Germany. Earth
Utrecht, T., 2005. Refraktionsseismik. In: Knödel, K., Krummel, H., Lange, G. Surface Processes and Landforms 37(1), 119–133.
(Eds.), Handbuch zur Erkundung des Untergrundes von Deponien und Altlasten, Heritage, G.L., Large, A.R.G., 2009. Principles of 3D laser scanning. In: Heritage,
Band 3: Geophysik, 2nd ed. Springer, Heidelberg, pp. 530–619. G.L., Large, A.R.G. (Eds.), Laser Scanning for the Environmental Sciences.
Bundesamt Für, Wasser, Geologie, B., 2002. Naturgefahren Symbolbaukasten zur Blackwell, London.
Kartierung der Phänomene-EDV-Legende für die digitale Karthographie. Bern: Higgitt, D.L., Warburton, J., 1999. Applications of differential GPS in upland fluvial
Bundesamt für Umwelt, Switzerland. geomorphology. Geomorphology 29, 121–134.
Büker, F., Gurtner, M., Horstmeyer, H., Green, A.G., Huggenberger, P., 1996. Three- Hillier, J.K., Smith, M., 2008. Residual relief separation: digital elevation model
dimensional mapping of glaciofluvial and deltaic sediments in Central enhancement for geomorphological mapping. Earth Surface Processes and
Switzerland using ground penetrating radar. Proceedings of the Sixth Landforms 33, 2266–2276.
International Conference on Ground Penetrating Radar, Sendai, pp. 45–50. Hinkel, K.M., Doolittle, J.A., Bockheim, J.G., Nelson, F.E., Paetzold, R., Kimble,
Church, M., 2010. The trajectory of geomorphology. Progress in Physical J.M., Travis, R., 2001. Detection of subsurface permafrost features with ground-
Geography 34, 265–286. penetrating radar, Barrow, Alaska. Permafrost and Periglacial Processes 12,
De Graaff, L.W.S., De Jong, M.G.G., Rupke, J., Verhofstad, J., 1987. A 179–190.
geomorphological mapping system at scale 1:10 000 for mountainous areas Hoffmann, T., Schrott, L., 2002. Modelling sediment thickness and rockwall retreat
(Austria). Zeitschrift fur Geomorphologie 31, 229–242. in an Alpine valley using 2D-seismic refraction (Reintal, Bavarian Alps).
Doolittle, J.A., Collins, M.E., 1995. Use of soil information to determine application Zeitschrift für Geomorphologie, Supplement Band 127, 153–173.
of ground-penetrating radar. Journal of Applied Geophysics 33, 101–108. Hoffmann, T., Schrott, L., 2003. Determining sediment thickness of talus slopes and
Entwistle, N., Fuller, I., 2009. Terrestrial laser scanning to derive facies character of valley fill deposits using seismic refaction – a comparison of 2D interpretation
gravel bars. In: Heritage, G.L., Large, A.R.G. (Eds.), Laser Scanning for the tools. Zeitschrift für Geomorphologie, Supplement Band 132, 71–87.
Environmental Sciences. Blackwell, London. Hofle, B., Rutzinger, M., 2011. Topographic airborne LiDAR in geomorphology: a
Etzelmuller, B., Berthling, I., Sollid, J.L., 2003. Aspects and concepts on the technological perspective. Zeitschrift für Geomorphologie 55, 1–29.
geomorphological significance of Holocene permafrost in southern Norway. Hubbard, B., Glasser, N., 2005. Field Techniques in Glaciology and Glacial
Geomorphology 52, 87–104. Geomorphology. J. Wiley & Son, Chichester.
Evans, I.S., 1990. Cartographic techniques in geomorphology. In: Goudie, A. (Ed.), Israil, M., Pachauri, A.K., 2003. Geophysical characterization of a landslide site
Geomorphological techniques. Unwin Hyman, London. in the Himalayan foothill region. Journal of Asian Earth Sciences 22,
French, J.R., 2003. Airborne LiDAR in support of geomorphological and hydraulic 253–263.
modelling. Earth Surface Processes and Landforms 28, 321–335. Jol, H.M., 1996. Three-dimensional GPR imaging of a fan-foreset delta: an example
Fuchs, M., Zöller, L., 2006. Geoarchäologie aus geomorphologischer Sicht. Eine from Brigham City, Utah, USA. Proceedings of the Sixth International Conference
konzeptionelle Betrachtung. Erdkunde 60, 139–146. on Ground Penetrating Radar (GPR 0 96), Sendai, Japan, pp. 33–37.
Geist, T., Höfele, B., Rutzing, M., Pfeifer, N., Stötter, J., 2009. Laserscanning – a Jol, H.M., Bristow, C.S., 2003. GPR in sediments: advice on data collection, basic
paradigm change in topographic data acquisition for natural hazard processing and interpretation, a good practice guide. In: Bristow, C.S., Jol, H.M.
management. In: Veulliet, E., Stötter, J., Weck-Hannemann, H. (Eds.), (Eds.), Ground-Penetrating Radar in Sediments. Geological Society London,
Sustainable Natural Hazard Management in Alpine Environments. Springer, London.
Heidelberg, Berlin. Kearey, P., Brooks, M., Hill, I., 2002. An Introduction to Geophysical Exploration.
Geist, T., Lutz, E., Stötter, J., 2003. Airborne laser scanning technology and its Blackwell, London.
potential for applications in glaciology. ISPRS Workshop on 3-D Reconstruction Kennett, M., Eiken, T., 1997. Airborne measurement of glacier surface elevation by
from Airborne Laserscanner and INSAR Data, Dresden. scanning laser altimeter. Annals of Glaciology 24, 293–296.
Gilbert, R., 1999. A Handbook of Geophysical Techniques for Geomorphic and Kneisel, C., Hauck, C., 2003. Multi-method geophysical investigation of a sporadic
Environmental Research. Geological Survey of Canada, Ottawa. permafrost occurrence. Zeitschrift für Geomorphologie, Supplementary 132,
Gili, J.A., Corominas, J., Rius, J., 2000. Using global positioning system 145–159.
techniques in landslide monitoring. Engineering Geology 55, 167–192. Kneisel, C., Lehmkuhl, F., Winkler, S., Tressel, E., Schröder, H. (Eds.), 1998.
Glenn, N.F., Streutker, D.R., Chadwick, D.J., Thackray, G.D., Dorsch, S.J., 2006. Legende für geomorphologische Kartierungen im Hochgebirge. Geographische
Analysis of LiDAR-derived topographic information for characterizing and Gesellschaft Trier, Trier.
differentiating landslide morphology and activity. Geomorphology 73, 131–148. Kovanen, D.J., Slaymaker, O., 2004. Relict shorelines and ice flow patterns of the
Grünewald, T., Schirmer, M., Mott, R., Lehning, M., 2010. Spatial and temporal northern Puget Lowland from LiDAR data and digital terrain modelling.
variability of snow depth and ablation rates in a small mountain catchment. Geografiska Annaler, Series A: Physical Geography 86, 385–400.
Cryosphere 4, 215–225. Lambiel, C., Delaloye, R., 2004. Contribution of real-time kinematic GPS in the
Gustavsson, M., Kolstrup, E., Seijmonsbergen, A.C., 2006. A new symbol-and-GIS study of creeping mountain permafrost: examples from the Western Swiss Alps.
based detailed geomorphological mapping system: renewal of a scientific Permafrost and Periglacial Processes 15, 229–241.
discipline for understanding landscape development. Geomorphology 77, Lane, S.N., Westaway, R.M., Hicks, D.M., 2003. Estimation of erosion and
90–111. deposition volumes in a large, gravel-bed, braided river using synoptic remote
Haas, F., Heckmann, T., Klein, T., Becht, M., 2010. Analysing rockfall processes on sensing. Earth Surface Processes and Landforms 28, 249–271.
alpine rockfaces and the corresponding talus cones using Terrestrial Leatherman, S.P., 1987. Coastal geomorphological applications of ground-
Laserscanning. Geophysical Research Abstracts 12, 1607–7962. penetrating radar. Journal of Coastal Research 3(3), 397–399.
Hauck, C., 2001. Geophysical methods for detecting permafrost in high mountains. Leckebusch, J., 2003. Ground-penetrating radar: a modern three dimensional
VAW Mitteilungen 171. Versuchsanstalt für Wasserbau, Hydrologie und prospection method. Archaeological Prospection 10, 213–240.
Glaziologie der Eidgenössischen Technischen Hochschule Zürich. Leclerc, R.F., Hickin, E.J., 1997. The internal structure of scrolled floodplain
Hauck, C., Vonder Mühll, D., 2003. Inversion and interpretation of two-dimensional deposits based on ground-penetrating radar, North Thompson River, British
geoelectrical measurements for detecting permafrost in mountainous regions. Columbia. Geomorphology 21(17–25), 29–38.
Permafrost and Periglacial Processes 14, 305–318. Liu, X., 2008. Airborne LiDAR for DEM generation: some critical issues. Progress
Hecht, S., 2000. Fallbeispiele zur Anwendung refraktionsseismischer Methoden bei in Physical Geography 32(1), 31–49.
der Erkundung des oberflächennahen Untergrundes. Zeitschrift für Loke, M.H., Barker, R.D., 1995. Least-squares deconvolution of apparent resistivity
Geomorphologie N.F., Supplement Band 123, 111–123. pseudosections. Geophysics 60, 1682–1690.
Hecht, S., 2001. Anwendung refraktionsseismischer Methoden zur Erkundung des Mantovani, F., Gracia, F.J., De Cosmo, P.D., Suma, A., 2010. A new approach to
oberflächennahen Untergrundes. Mit acht Fallbeispielen aus Südwestdeutschland. landslide geomorphological mapping using the Open Source software in the
Stuttgarter Geographische Studien, 131. Olvera area (Cadiz, Spain). Landslides 7, 69–74.
20 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview

Mckean, J., Roering, J., 2004. Objective landslide detection and surface Sass, O., Bell, R., Glade, T., 2008. Comparison of georadar, 2D-resistivity and
morphology mapping using high-resolution airborne laser altimetry. traditional techniques for the subsurface exploration of landslides.
Geomorphology 57, 331–351. Geomorphology 93, 89–103.
Milsom, J., 1996. Field Geophysics. Wiley. Sass, O.K.W., 2001. Investigations regarding alpine talus slopes using ground-
Moorman, B.J., Michel, F.A., 2000. Glacial hydrological system characterization penetrating radar (GPR) in the Bavarian Alps, Germany. Earth Surface Processes
using ground-penetrating radar. Hydrological Processes 14, 2645–2667. and Landforms, 26
Moorman, B.J., Robinson, S.D., Burgess, M.M., 2003. Imaging periglacial Schneevoigt, N.J., Schrott, L., 2006. Linking geomorphic systems theory and
conditions with ground-penetrating radar. Permafrost and Periglacial Processes remote sensing. A conceptual approach to Alpine landform detection (Reintal,
14, 319–329. Bavarian Alps, Germany). Geographica Helvetica 61, 181–190.
Oppikofer, T., Jaboyedoff, M., Keusen, H.R., 2008. Collapse at the eastern Eiger Schneevoigt, N.J., vander Linden, S., Thamm, H.P., Schrott, L., 2008. Detecting
flank in the Swiss Alps. Nature Geoscience 1, 531–535. Alpine landforms from remotely sensed imagery. A pilot study in the Bavarian
Otto, J.C., Gustavsson, M., Geilhausen, M., 2011. Cartography: design, Alps. Geomorphology 93, 104–119.
symbolisation and visualisation of geomorphological maps. In: Smith, M.J., Schoeneich, P., Reynard, E., Pierrehumbert, G., 1998. Geomorphological mapping in
Paron, P., Griffiths, J. (Eds.), Geomorphological Mapping: A Handbook of the Swiss Alps and Prealps. In: Kriz, K. (Ed.), Hochgebirgskartographie Silvretta
Techniques and Applications. Elsevier, London. ’98. Wiener Schriften zur Geographie und Kartographie 11, pp. 145–153.
Otto, J.C., Kleinod, K., König, O., et al., 2007. HRSC – A data: a new high- Schrott, L., Hördt, A., Dikau, R. (Eds.), 2003. Geophysical Applications in
resolution data set with multi-purpose applications in physical geography. Geomorphology. Gebr. Bornträger, Berlin, Stuttgart.
Progress in Physical Geography 31, 179–197. Schrott, L., Sass, O., 2008. Application of field geophysics in geomorphology:
Otto, J.C., Schrott, L., Jaboyedoff, M., Dikau, R., 2009. Quantifying sediment advances and limitations exemplified by case studies. Geomorphology 93,
storage in a high alpine valley (Turtmanntal, Switzerland). Earth Surface 55–73.
Processes and Landforms 34, 1726–1742. Seijmonsbergen, A.C., Hengl, T., Anders, N.A., 2011. Semi-automated identi-
Overgaard, T., Jakobsen, P.R., 2001. Mapping of glaciotectonic deformation in an fication and extraction of geomorphological features using digital elevation
ice marginal environment with ground penetrating radar. Journal of Applied data. In: Smith, M.J., Paron, P., Griffiths, J. (Eds.), Geomorphological
Geophysics 47, 191–197. Mapping: A Professional Handbook of Techniques and Applications. Elsevier,
Overton, I.C., Siggins, A., Gallant, J.C., Penton, D., Byrne, G., 2009. Flood London.
modelling and vegetation mapping in large river systems. In: Heritage, G.L., Smith, D.G., Jol, H.M., 1995. Ground penetrating radar – antenna frequencies and
Large, A.R.G. (Eds.), Laser Scanning for the Environmental Sciences. Blackwell, maximum probable depths of penetration in quaternay sediments. Journal of
London. Applied Geophysics 33, 93–100.
Prokop, A., Panholzer, H., 2009. Assessing the capability of terrestrial laser Smith, M.J., Clark, C.D., 2005. Methods for the visualization of digital elevation
scanning for monitoring slow moving landslides. Natural Hazards and Earth models for landform mapping. Earth Surface Processes and Landforms 30,
System Sciences 9, 1921–1928. 885–900.
Pullan, S.E., Hunter, J.A., 1990. Delineation of buried bedrock valleys using the Smith, M.J., Pain, C.F., 2009. Applications of remote sensing in geomorphology.
optimum offset shallow seismic reflection technique. In: Ward, S.H. (Ed.), Progress in Physical Geography 33, 568–582.
Geotechnical and Environmental Geophysiscs. Geotechnical. Society of Smith, M.J., Rose, J., Booth, S., 2006. Geomorphological mapping of
Exploration Geophysicists, Tulsa, Oklahoma. glacial landforms from remotely sensed data: an evaluation of the principal
Rabatel, A., Deline, P., Jaillet, S., Ravanel, L., 2008. Rock falls in high-alpine rock data sources and an assessment of their quality. Geomorphology 76,
walls quantified by terrestrial LiDAR measurements: a case study in the Mont 148–165.
Blanc area. Geophysical Research Letters 35, L10502. doi:10.1029/ Souch, C., 2003. Getting information about the past: palaeo and historical data
2008GL033424. sources. In: Clifford, N., Valentine, G. (Eds.), Key Methods in Geography. Sage,
Rayburg, S., Thoms, M., Neave, M., 2009. A comparison of digital elevation models London.
generated from different data sources. Geomorphology 106(3–4), 261–270. Squarzoni, C., Delacourt, C., Allemand, P., 2005. Differential single-frequency GPS
Reynard, E., Holzmann, C., Lambiel, C., Phillips, M., 2005. Légende monitoring of the La Valette landslide (French Alps). Engineering Geology 79,
géomorphologique de l’IGUL et Guide pratique pour le levé de cartes 215–229.
géomorphologiques. Lausanne, Institut de Géographie, 33 p. Staley, D.M., Wasklewicz, T.A., Blaszczynski, J.S., 2006. Surficial patterns of debris
Reynolds, J.M., 1997. An Introduction to Applied and Environmental Geophysics. flow deposition on alluvial fans in Death Valley, CA using airborne laser swath
John Wiley & Sons, Chichester, UK. mapping data. Geomorphology 74, 152–163.
Rice, S., 2003. Sampling in geography. In: Clifford, N., Valentine, G. (Eds.), Key Tamburini, A., Deline, P., Mortara, G., 2005. Time-space modelling with terrestrial
Methods in Geography. Sage, London. LiDAR: monitoring ice cliff evolution of the Miage Glacier, Italy, with ILRIS-3D.
Ritter, D.F., 1986. Process Geomorphology. Dubuque, Iowa, W.C., Brown. GIM International 19, 31–33.
Rizos, C., 2002. Making sense of the GPS technique. In: Bossler, J. (Ed.), Tavkhelidse, T., Schulte, A., Stumböck, M., 2000. Aufbau und Entwicklung der
Manual of Geospatial Science and Technology. Taylor & Francis, London, Schuttkegel im Finkenbachtal, südlicher Odenwald. Jenaer Geographische
New York. Schriften 9, 95–110.
Rosser, N.J., Petley, D.N., Lim, M., Dunning, S.A., Allison, R.J., 2005. Terrestrial Tooth, S., 2009. Arid geomorphology: emerging research themes and new frontiers.
laser scanning for monitoring the process of hard rock coastal cliff erosion. Progress in Physical Geography 33, 251–287.
Quarterly Journal of Engineering Geology and Hydrogeology 38, 363–375. Tricart, J., 1965. Principes et Methodes de la Geomorphologie. Masson, Paris.
Sass, O., 2006. Determination of the internal structure of alpine talus deposits Van Asselen, S., Seijmonsbergen, A.C., 2006. Expert-driven semi-automated
using different geophysical methods (Lechtaler Alps, Austria). Geomorphology geomorphological mapping for a mountainous area using a laser DTM.
80, 45–58. Geomorphology 78, 309–320.
 Fundamental Classic and Modern Field Techniques in Geomorphology: An Overview 21

Biographical Sketch

Lothar Schrott was born in 1962. He studied geology at the Universities of Tübingen and Geography, Geology and
Physical Education in Heidelberg, Germany. He received his PhD in 1993 from the University of Heidelberg
investigating the role of solar radiation within the geosystem of the semiarid high Andes.
He is a professor for physical geography and head of the research group Geomorphology and Environmental
Systems at the Department of Geography and Geology, University of Salzburg (Austria). His research interests
focus mainly on geomorphic processes in mountain areas (European Alps, Rocky Mountains, Andes, German
upland). Current research projects are related to sediment budgets, high mountain permafrost, and mass
movements. His numerous teaching activities in physical geography focus specifically on theoretical and applied
approaches in geomorphology, global environmental change issues, and natural hazards. Application of quan-
titative methods (lab and field) in geomorphology (e.g., mapping and field geophysics) is of primary concern. He
is associate editor of Geografiska Annaler (Series A, Physical Geography) and member of the editorial board of
Geomorphology. He is also a member of the Executive Committee of the International Association of Geo-
morphologists. Lothar Schrott edited a number of special issues and books on landslides, geophysical appli-
cations in geomorphology and sediment fluxes, and published more than 70 papers mainly in peer-reviewed
journals.

Jan-Christoph Otto was born in 1974. He studied geography, geology, and soil science at the Universities of Bonn
and Grenoble. He received his diploma and PhD in geography from the University of Bonn.
He is currently a postdoctoral researcher in the Department of Geography and Geology, University of Salzburg,
within the Geomorphology and Environmental Systems Research Group. His main research interests include
landscape change in high mountain environments with special focus on sediment budgets and the dynamics and
consequences of permafrost changes in mountain areas. His recent interests include the creation of digital geo-
morphological maps and the combination of high-resolution surface and subsurface data within a Geo-
morphological Information System.

Joachim Götz was born in 1977. He studied geography at the Universities of Augsburg and Bonn, Germany. He
received his diploma thesis from the University of Bonn in 2006 working on sediment budgets in the
German Alps.
He is currently preparing a PhD thesis at the University of Salzburg continuing his research in the field of
sediment budgets. In his thesis he investigates postglacial sediment fluxes and storage in nested Alpine catch-
ments, a project which is embedded into the ESF Topoeurope research framework. Joachim is applying a com-
bined field and modeling approach using geophysical techniques, high-resolution digital terrain data, and GIS
tools for the quantification of sediment storage and fluxes in the Möll catchment, Austrian Alps.

Martin Geilhausen was born in 1979. He studied geography at the University of Bonn and received a diploma
investigating fluvial and glacial deposits using geophysics.
After working for a geoinformation company, where he developed webGIS and web mapping applications, he
enrolled as a PhD student in the Department of Geography and Geology, University of Salzburg (Geomorphology
and Environmental Systems Research Group). His research involves the quantification of recent and postglacial
sediment dynamics in glacier forefields, Austrian Alps. Martin is applying a suite of field techniques ranging from
fluvial erosion measurement and geophysics (ground-penetrating radar, seismic refraction, and resistive tomo-
graphy) to terrestrial laser scanning.
 14.3 Geomorphometry: Quantitative Land-Surface Analysis
J Minár, Comenius University in Bratislava, Bratislava, Slovakia, and University of Ostrava, Ostrava, Czech Republic
IS Evans, Durham University, Durham, UK, and Suceava University, Suceava, Romania
J Krcho, Comenius University in Bratislava, Bratislava, Slovakia
r 2013 Elsevier Inc. All rights reserved.

14.3.1 Introduction 23
14.3.2 Basics: Altitude and Slope Gradient 25
14.3.3 Geomorphometric Field Variables: Local and Regional 26
14.3.4 Linear Objects 29
14.3.5 Areal Objects 29
14.3.6 Scaling and Scale Specificity 30
14.3.7 Conclusions: The Future 31
References 32

Glossary Landform A genetically defined part of the land surface,

Allometry Variation of shape (or other properties) as size generally comprising several adjacent and related
increases. elementary forms.
Digital elevation model (DEM) A numerical discrete Light detection and ranging (LiDAR) A remote-sensing
representation of land surface (ground) altitudes. Often technology based on distance measurement by pulses from
confined to gridded models, but accepted here sensu lato, a laser beam, providing ’first returns’ from building roofs
covering also triangulated irregular models and contour/ and the tree canopy and ’last returns’ from the ground
flowline-based models. Digital terrain model (DEM) and beneath vegetation. Pulses may come from ground-based or
digital ground model (DGM) have also been used. Digital air-based equipment.
surface model (DSM) is used for models including the top Relief Often used (in German and Slavonic languages) as
of vegetation and buildings. synonymous with topography, terrain or the land surface, in
Dupin indicatrix From the intersection of a plane parallel English-language geography ’relief’ more commonly has the
to the plane tangent to the Earth surface, in a differentially narrower sense of difference in altitude between a high
small vicinity: it characterizes local surface shape and is point (variously defined) and a related low point. The latter
related to Gaussian curvature. is known also as relative or available relief: ’Reliefenergie’ in
Elementary form A basic unit (facet and segment) of the German.
land surface, defined by homogeneity in altitude or one or Scale-specific Being limited to a small range of sizes, or
more surface derivatives (local variables) and largely showing a size threshold between different relationships.
bounded by sharp changes in these. Shuttle radar topography mission (SRTM) Using single-
Hypsometry The relation between land surface map area pass interferometry between 11 and 20 February 2000 from
and altitude (elevation above sea level). Commonly, the Space Shuttle Endeavour, an X-band radar scanned
cumulative percentage area is plotted against altitude (either 50-km wide swaths to provide a publicly available near-
absolute or as a percentage of the range). global raster DEM covering the Earth surface from 581S to
Land element A basic unit (facet and segment) of the land 601N. The resolution of the cells is approximatelly 90lm
surface defined in terms of regional (positional) variables as (30lm for USA).
well local ones. Commonly coincident with an elementary
form.

Abstaract

Within geomorphology, geomorphometry applies morphometric analysis to the land surface as a continuous surface
(general geomorphometry), and to objects such as landforms, elementary forms, channels, and drainage basins extracted
from that surface (specific geomorphometry). Special attention is paid here to a comprehensive taxonomy of fundamental

Minár, J., Evans, I.S., Krcho, J., 2013. Geomorphometry: quantitative land
surface analysis. In: Shroder, J. (Editor in chief), Switzer, A.D., Kennedy,
D.M. (Eds.), Treatise on Geomorphology. Academic Press, San Diego, CA,
vol. 14, Methods in Geomorphology, pp. 22–34.

22 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00370-5

 Geomorphometry: Quantitative Land-Surface Analysis 23

geomorphometric variables, both those related to fields and those characterizing objects. The former include field-specific
and field-invariant as well as local and regional variables. Derivatives of the land surface related to the gravity field are
traditionally important in geomorphology, but an extended system of field-invariant curvatures has potential. Solar radi-
ation and wind fields in relation to topography are of both climatic and geomorphological (process) importance. The
morphometry of linear objects has been not only dominated by channels and drainage networks but also includes
structural and tectonic morpholineaments. The problem of elementary form extraction from digital elevation models
(DEMs) is basic to the analysis of areal objects. Scaling and scale-specificity questions are documented with quantile and
allometric plots. Major prospects for the future development of geomorphometry are perceived as resulting not only from
great technological development (new remote-sensing methods providing high-precision DEMs) but also from the effective
integration of the empirical and theoretical streams of geomorphometry.

14.3.1 Introduction nature of the surface while treating it as a continuous

smooth field. Other interpolation and surface-fitting models
Geomorphometry is one type of morphometric analysis, a such as spectral, Fourier, fractal, and spline tend to produce
widespread scientific approach that starts from the assumption overshoots and artificial peaks wherever discontinuities are
that a form or surface contains valuable information about its encountered.
functionality, inner composition, and even genesis. Thus, Precision and accuracy of data are other aspects influencing
morphometry is important, for example, in modern biology, results of geomorphometric investigations. Mixing of data
anthropology, paleontology, sedimentology, and engineering, obtained by various methods in different times (e.g., hypso-
as well as in geomorphology. metry of older topographic maps) can lead to deformations
A simple definition of geomorphometry is ‘quantitative from the point of view of space and also time representation.
land-surface analysis’. It is focused only on the external (outer) But as data have improved from coarse DEMs interpolated
solid surface of the object – Earth, or other planets, or moons. from digitized contours, through photogrammetric and radar-
The object of geomorphometry is topography (in the Anglo- based DEMs, each with their specific problems (Nelson et al.
Saxon tradition) or relief (in Latino-European traditions – cf. in Hengl and Reuter, 2009), to very detailed and accurate light
Shary, 2008), a dynamic surface modeled by geomorpho- detection and ranging (LiDAR) DEMs, our problems have
logical processes and separating the solid from the fluid Earth, shifted away from data error and sampling adequacy to how
that is, the lithosphere or pedosphere from the hydrosphere or the surface is actually defined. This was discussed by Evans
atmosphere. This surface is double sided in that the materials (2012), who suggested that geomorphologists have perceived
on each side differ. the surface essentially at human scale: around 1.5 m. Build-
Surfaces may be analyzed either as continuous fields or as ings and vegetation are filtered out where natural processes are
mosaics. We recognize two complementary approaches: gen- analyzed, but retained for some practical applications such as
eral and specific geomorphometry (Evans, 1972). General flood routing.
geomorphometry analyzes the land surface as a continuous Geodetic measurements (by various instrumental techni-
field and seeks optimal methods for the expression and in- ques using analytical geometry) directly measure the land
terpretation of this field. Specific geomorphometry starts from surface and create a basic geomorphometric database (see
the assumption that parts of the land surface (landforms) are Volume 3, Remote Sensing and GIScience in Geomorpho-
perceived as autonomous entities (objects). By analogy with logy of the Treatise on Geomorphology). But most geomorpho-
the wave and particle concepts of light in physics, there are metric methods focus on the computation of morphometric
situations where either the first or the second approach is characteristics from these primary geodetic data. The methods
preferable. The traditional perception of the land surface as a used by geomorphometry include both mathematical (calcu-
set of geomorphic individuals (landforms or elements) needs lus, algebra, and geometry) (Krcho, 1973) and statistical
to coexist with modern general geomorphometry from digital methods (descriptive, inferential, and graphical) (Chorley,
elevation models (DEMs, see Chapter 3.6), treating the land 1966).
surface as a continuous, commonly smooth field (Schneider, Two streams of development can be identified in geo-
2001). Nevertheless, in principle, objects can be extracted from morphometry. The empirical approach involves statistical
the continuous field (cf. Minár and Evans, 2008) and, in generalization and ad hoc development of morphometric tools
theory, the field can be reconstructed by patching together the (characteristics and methods) for solution of concrete geo-
objects. morphological problems. It involves mainly specific geomor-
Yet, how smooth is the land surface in reality? Shary phometry and also process modeling and land-surface
(2008) demonstrated that it is not truly smooth: there is segmentation methods. The theoretical approach is based on
no limit to surface numerical derivatives as the spacing geomorphometric systems of axioms and theorems, providing
over which they are measured tends to zero. This explains sets of morphometric characteristics (e.g., Krcho, 1973, 2001;
why flows over the surface tend to branch and rejoin. On the Jenčo, 1992; Shary, 1995; Shary et al., 2002).
other hand, the land surface is not as rough as fractal models
imply (Evans and McClean, 1995). In practice, derivatives
such as slope and curvature have to be measured as differ-
entials over finite distances; algorithms such as fitting local
polynomials by least squares cope with the nonsmooth
24 Geomorphometry: Quantitative Land-Surface Analysis

Land surface above reference surface of the Earth (original)

Norm
al (ver
tical
Equipoten ) lin
tial leve Ni es
N t
1750 ls of t i ow
1500 he gr ard
avita N ref
1250 tiona i ere
l sc nc
alar N es
1000 15 field
00 of t
i urf
750 ac
125 he eo
500 0 Ea f th
r th N eE
250
0 1000 i ar
th
750 1250

1500 1750
1250
1000
+2Δ 1500
+x 1250
+z 1250

75
0
1000

+Δ

10 50
+y

00
7 0
50 0
25 0
2000 km2


+3Δ

−Δ +2Δ
ar th
he E
of t
s urface
nce +Δ
Refere
−2Δ

(a) −Δ

Land surface in cartographic projection Cartographic projections of the normal lines Ni

+z
Ni Zi Ni Zi Ni Zi Ni Zi
Cartographic projections
of the equipotential levels

50 1750
17 00 1500
1500 15 0
1250 125 1250
1250 00 1000
10
1000 750
750

+y

0
150 1250
1250
1000 750

Car tographic project

(b) +x ion plane (x,y)

Figure 1 Projection of the orthogonal net of contours (equipotential levels of the gravity field) and stream lines above a reference Earth surface
(ellipsoid or geoid) (a) into an orthogonal net of contours and stream lines in a cartographic coordinate system (b): j, latitude; l, longitude; 7h,
altitude in the line Ni normal to reference surface; x, y, z, Cartesian coordinate system. Without correct cartographic transformation only limited
areas (up to 200 km2 for the most strict geodetic criteria) can be correctly studied in the Cartesian coordinate system, but for most
geomorphological purposes the distortion in an area up to 2000 km2 can be acceptable (top).
 Geomorphometry: Quantitative Land-Surface Analysis 25

2500
2000
Borrowdale
2000
1500

Frequency
Frequency
Helvellyn
1500
1000
1000
500 500

0 0
0 200 400 600 800 1000 0 200 400 600 800 1000
Altitude (m) Altitude (m)

2000 3500
3000 Ambleside
Scafell
1500 2500

Frequency
Frequency

2000
1000
1500

500 1000
500
0 0
0 200 400 600 800 1000 0 200 400 600 800 1000
Altitude (m) Altitude (m)
Figure 2 Altitude–frequency distributions for four adjacent 10  10 km2 (NY 20, 21, 30, 31) in Cumbria: DEM grid mesh 50 m. Class width is
20 m and the horizontal scale is the same throughout, from 0 to 1000 m altitude, but the vertical scale (frequency) varies. Each graph has
39 601 values. Some modes are due to the presence of sizeable lakes; Ambleside includes considerable lowland.

14.3.2 Basics: Altitude and Slope Gradient Limitations to hypsometry and clinometry are those results
which vary with the boundary of the area considered, which is
The history of geomorphometry was summarized by Hengl arbitrary if map sheets, grid squares, or circles are used
et al. in Hengl and Reuter (2009). After an early phase of (Figure 2), and subjective (or not strictly comparable) if
preoccupation with coastal sinuosity, much morphometric topographic regions are used. Many papers focused on relating
work concerned frequency distributions of altitude and of mean values to altitude, with too little use of statistical testing.
slope gradient, and the relation of average slope gradient to However, studies using drainage basins as nonarbitrary areal
altitude. The resulting fields of hypsometry and clinometry, divisions, and relating slopes to channels, produced seminal
respectively, were thoroughly reviewed by Clarke (1966). results such as Strahler (1950, 1952) and Schumm (1956).
Distributions were measured from contour maps, mainly to Ohmori (1993) related denudation rates to uplift, altitude,
establish the altitudes of denudation surface remnants: these and relief (altitude dispersion), with associated changes in
have greater altitude frequencies and lower mean slope gra- hypsometric integral.
dients. Derived summary statistics include the ‘hypsometric Figure 2 shows how variable altitude frequency distri-
integral’ (area under dimensionless hypsometric curve), which butions are even between adjacent areas (map sheets, here).
is the same as the ‘elevation ratio’, and is closely related to the Most are positively skewed, that is, the mode is below the
skewness of altitude (Evans, 1972). mean and there is a longer tail of high values. Figure 3 shows
Altitude is clearly the fundamental and primary variable in the corresponding slope-gradient distributions, which also
geomorphometry. Contour lines represent equipotential levels tend to positively skew especially where some lowland is
of the gravity field, giving physical meaning not only to alti- included.
tude but also to all derived variables. Correct cartographic The availability of large DEMs has revived and extended
expression of altitude is a basic precondition for further an- hypsometric work, for example, in relation to the buzzsaw
alysis (Figure 1); the reference surface and projection are hypothesis (Mitchell and Montgomery, 2006) that glacial
important, especially for large areas. Altitude also has strong erosion limits the height of mountain ranges. This has been
effects on temperature and other climatic variables, and thus applied mainly on regional scales, but Egholm et al. (2009)
controls the magnitude and sometimes the type of geo- have applied it on the near-global-scale of shuttle radar
morphological processes. Slope gradient controls the way topography mission data. The main mode of terrestrial alti-
gravity affects surface runoff of water, and together with relief tude is near sea level, but many secondary modes occur
in the narrow sense of difference in altitude it controls stress between the present snowline and previous (lower) snowlines.
fields and, thus, mass movements.
26 Geomorphometry: Quantitative Land-Surface Analysis

3000 Borrowdale 3000

Helvellyn

Frequency
Frequency
2000 2000

1000 1000

0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Gradient, degrees Gradient, degrees

3000 Scafell 3000

Ambleside
Frequency

Frequency
2000 2000

1000 1000

0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Gradient, degrees Gradient, degrees
Figure 3 Gradient-frequency distributions for the four areas in Figure 2. Class width is 21 and both scales are the same throughout. Dead flat
areas (lakes and floodplains) are excluded, so the number of values varies between 37 957 (Helvellyn) and 39 563 (Scafell). There is some
tendency to positive skew even in mountain areas.

14.3.3 Geomorphometric Field Variables: Local and variables defined by partial derivatives of the altitudinal field.
Regional Both systems are based on changes (first, second, or third
derivatives) of altitude in the direction of the gravitational
Moore et al. (1991), Wilson and Gallant (2000: pp. 7–9), and field (streamlines) and the orthogonal contour line direction
Schmidt and Dikau (1999) listed numerous variables used in (Figure 1). Each contains slope gradient and aspect as well as
geomorphometry. Many further variables are specific to par- plan and profile curvature, and has been extended by these
ticular landforms or elementary forms, as discussed in Hengl and other authors (e.g., Jenčo, 1992; Shary, 1995; Evans, 1998;
and Reuter (2009) and Goudie et al. (1990). In Figure 4, we Krcho, 1992, 2001). Some variables may be defined in slightly
give a classification of the more fundamental variables in both different ways, for example, profile curvature or gradient
general and specific geomorphometry. The primary distinction change in terms of angles (degrees or radians), sines, or tan-
is between field- and object-based variables (Figure 5). Field- gents, but are essentially similar (monotonically related). Note
based variables, defined by the behavior of the land surface in that slope is a vector and its directional (azimuthal) com-
some neighborhood around any point, are either field specific ponent, aspect, should be analyzed by circular (periodic)
or field invariant, and local or regional. Local variables in the statistics and not by conventional linear statistics (Evans and
sense of Shary et al. (2002) include altitude, slope, and Cox, 1999).
curvature. The main local variables are point based, that is, Shary (1995) and Shary et al. (2002, 2005) subdivided
based conceptually on behavior of topography (altitudinal general geomorphometric variables into field invariant (in-
field) over a vanishingly small area around each point, and are variant with respect to surface rotation around any axis) and
specific to the gravity field. A second group of local variables field specific, describing the system land-surface þ vector
are based on arbitrary regular areas (circles or squares) around field. Variables in the Evans and Krcho systems are specific
each point. Regional variables are defined by searching to the gravitational field, as this is relevant to current geo-
upslope or downslope for an indefinite distance determined morphological processes. However, variables describing inter-
by the terrain itself: they include catchment area, depth within action of other fields with the land surface are also important.
a depression, and length of slope and height above thalweg. Point-based examples related to fields of solar radiation and
Object-based variables describe either linear or areal objects wind include incident radiation and shelter, and angle of inci-
(and, less commonly, specific point or volumetric objects); dence of sun or wind.
they are considered in later sections. Field-invariant variables include principal, mean and
Geomorphologists Evans (1972) and Krcho (1973) in- Gaussian curvature independent of any coordinate system. As
dependently introduced systems of the fundamental local discussed by Shary (1995) and Shary et al. (2002), these
 Geomorphometry: Quantitative Land-Surface Analysis 27

Field variables

Variables specific to gravity field

Local: Point-based
Zero order (Primary): Altitude;
First order: Slope gradient and slope aspect;
Second order: Plan, profile and rotor curvatures;
Third order: Change of plan, profile or rotor curvatures.

Local: Area-based
Descriptive statistics of any of the above, within given radius;
Percentile of height, within given radius;
Relief, within given radius;
Drainage density within given radius.

Regional (positional)
Height/depth above/below any regional level (Hill/depression boundary, thalweg, ridge-line);
Distance to stream; Distance to ridge (Flow path);
Total or specific catchment area; Total or Specific dispersal area.

Variables specific to other fields

Angle of incidence of solar radiation;

Angle of incidence of wind flow;
Amount of solar radiation, integrated over a given time period;
Degree of exposure to / shelter from a wind regime.

Field- invariant variables

Local: Point-based
Second order:Principal curvatures: Maximal, Minimal; and Total;
Unsphericity and Mean curvature.

Local: area-based
Descriptive statistics of any point-based variable, within given radius.

Object variables
Areal
Area; Length; Width; Perimeter; Drainage density; Shape; Spatial pattern;
Edge characteristics; Neighbours;
Mean (and other descriptive statistics) of any field variables,
for any specific landform or element.

Linear
Stream order; Stream link length; Stream direction; Flowline length;
Relative height (between stream and ridge);
Morpholineament orientation;
Mean (and other descriptive statistics) of any of the above variables for
any specific line (thalweg, flow line, ridge, morpholineament).

Point
Point-based local variables
Figure 4 Classification of the fundamental geomorphometric variables or characteristics.
28 Geomorphometry: Quantitative Land-Surface Analysis

Area-based local variables computed for moving windows (Value computed from the
whole window is assigned to
the central pixel)

Point-based local
c1 c2  H2 variables. Altitude
d1 dH d1 (H) is distance from
d2
H1 the global sea level
Sea level and others are
derivatives of H:
Slope () = dH/d l
(a)
Regional variables (d = depth of a depression at an arbitrary point, c = specific catchment area
of an arbitrary point) computed from the irregular surrounding of a point

L1 A1

A2
L3
A6

L2 A5
A3

A4

(b)
Areal (A = form area) and linear (L = thalweg length) object variables
Figure 5 Geomorphometric variables derived from (a) the field concept of a land surface (contours represent continuous altitudinal field) and
(b) the object concept of a land surface consisting of a mosaic of individual landforms (areal objects) and thalwegs (linear object).

curvatures have been known since 1827 (but so far they have algorithms to search upslope and downslope, if necessary over
been little used in geomorphology). However, their in- long distances. The resulting set of positional (or regional)
dependence of the coordinate system means that they are variables can be defined at any point on the land surface.
stable with tectonic tilt, whereas normal gravity-defined Runoff cumulates downslope and its quantity is a function of
curvatures (profile, plan, and tangential) change. Krcho (1992, the area drained – the catchment area. On slopes, this is de-
2001) showed the possibilities of investigating relations be- finable (for every point) per unit contour width (specific
tween field-invariant and gravity-normal curvatures using the catchment area), whereas within the channel network absolute
Dupin indicatrix. We can, for example, postulate that both catchment area (km2) is used; this is area-based but relates to
types of curvature are identical only for landforms produced a point on the channel network. Catchment area controls
by gravity-controlled flows. However, when structural con- quantity of water, but distance affects travel times, so further
ditions or tectonics disturb the influence of gravity-controlled positional variables include distance to a stream channel and
flows, principal curvature can be distinctly different from to a ridge or drainage divide; or in a channel, distance to the
gravity-determined main curvatures. Comparison of both outlet or from the farthest headstream. Stream channels are
could, thus, reveal the importance of lithology and tectonics generally defined automatically in terms of a threshold catch-
compared with gravity-controlled flows. ment area per DEM cell, possibly varying with slope gradient.
Point-based local variables require little computing power Although the same algorithm can be applied to an inverted
and were, thus, the first attributes to be calculated from DEMs, DEM, ridges are commonly more diffuse than channels and
in the 1950s, 1960s, and 1970s (Evans, Section 2.3 in Goudie definition of the upslope termination of a slope (drainage) line
et al., 1990). They are important for many gravitational and may require a more arbitrary threshold. Having traced the line
climatic processes, but are incomplete when surface runoff is from ridge to channel, we can also define any point’s position
considered; position is also involved, requiring more complex
 Geomorphometry: Quantitative Land-Surface Analysis 29

in the vertical as a proportion of the drop from ridge to (Woldenberg, 1972). These are geometric laws of networks
channel, as well as the absolute drop in both directions. constrained in a particular way, not only laws of fluvial
All these regional variables are demanding in both com- geomorphology.
putation and data quantity. If drainage lines cross the data More recent syntheses of fluvial geomorphometry include
boundary, values are indeterminate; thus, datasets need to that of Rodrı́guez-Iturbe and Rinaldo (1997), who showed
cover large areas, and even so the outline of areas within that fractal (self-similar or self-affine) models can be applied
which regional variables are definable is irregular. Area-based to many aspects of river systems, although as they noted in
local variables also describe position and are easier to com- (p. 182) ‘‘for real drainage basins strict self-similarity is un-
pute, because they are based on the field distribution within a tenable.’’ Much recent work involves plotting hillslope or
defined radius around any point. The difference between channel gradient against area drained, and shows a maximum
maximum and minimum altitude gives relief, in the narrower at a scale where flow becomes concentrated (channelized)
Anglo-Saxon sense: relief energy in the German sense. The (Dietrich and Montgomery, 1998). Dodds and Rothman
percentile of the point’s altitude within the frequency distri- (2000) showed that real drainage networks cannot be binary
bution of altitudes (within the radius) characterizes its vertical branching trees, nor in general can they be self-similar. Hack’s
position, without the calculation of a drainage line. law (Hack, 1957) states that for medium-sized basins main-
In particular applications, variables calculated from two or stream length increases more rapidly than the square root of
more of these fundamental ones are important. These include catchment area: it increases as area raised to a power around
the Wetness Index and Stream Power, which are compound 0.6. Dodds and Rothman (2000) proposed four scaling re-
regional field variables. gimes for the relation between length and area, defining three
Finally, Shary et al. (2005) distinguished geomorphometric diffuse crossover scales. Dietrich and Montgomery (1998) also
variables that preserve relatively stable statistical characteristics recognized four specific scales in the fluvial landscape. Hence,
with change of the input DEM density (scale-free variables) more than one type of scale specificity occurs even in well-
from others that change without limit as the DEM density integrated fluvial systems.
increases (scale-dependent variables). Altitude is scale free, but Long profiles of rivers reflect system development and
its derivatives (slope gradient and curvature) are scale geological controls; in the steady state, they are most often
dependent (Wood, 1996), and reduce in magnitude as grid concave-up. Also, the gradient of each stream link can be re-
mesh increases. Regional variables such as catchment area and lated to its position within the network, and hillslopes can be
depression area are scale free, but the majority of area-based related to the characteristics of the stream at their base.
local variables are scale dependent. All variables are affected by Richards (Section 2.4 in Goudie et al. (1990)) summarized
random error that increases with the coarseness of sampling work on the geometry of river channels.
(e.g., the spacing between data points in a DEM). Other linear objects mainly morpholineaments, charac-
teristically defined from remote sensing, topographical maps,
or DEM analysis. These are characterized by their orientation
14.3.4 Linear Objects (e.g., east–west, undirected) and length, and have further po-
tential for study of tectonic, structural, and aeolian forms
The analysis of drainage networks (see Chapter 14.6) is of (Minár and Sládek, 2009).
particular importance because of the widespread control of
terrestrial landscapes by fluvial (and related slope) processes.
Surface runoff is directed downslope (approximately at right 14.3.5 Areal Objects
angles to contours) and commonly converges and produces
incised channels. In humid and semi-arid areas, these join Specific geomorphometry is concerned with sets of discrete
together to form integrated drainage networks. The classical objects such as landforms, form elements, and drainage
framework for geomorphometric study of these networks is basins, and especially with their shape and pattern (Jarvis and
Strahler’s ordering system, whereby source streams are first Clifford, Section 2.5 in Goudie et al., 1990). It has several
order, and wherever two streams of the same order join a interconnections with general geomorphometry. Character-
stream of the next higher order is formed. Where a stream istics summarizing the continuous altitudinal fields inside
joins one of the higher order, stream order is unaffected; thus, such objects were exemplified in Evans (1987) (see also Fig-
the concept of links (between two adjacent junctions) and ure 4). However, the most important connection between the
stream magnitude (number of links upstream) was developed field and object approaches concerns methods of object ex-
(for more detailed studies the reader is referred to Gardiner, traction from the altitudinal field. The characteristics of
Section 2.6 in Goudie et al. (1990)). drainage basins, based on segment or link orders within net-
Plots of logarithms of number of streams, of mean works, have received particular attention (Zavoianu, 1985).
stream length and of mean catchment area against stream Deng (2007) (see Figure 4) recognized five types of geo-
order tend to produce linear trends, giving rise to relations morphological objects: bona fide, prototypical, semantic,
known as Horton’s laws. These apply not only to networks multiscale, and landform-class. Many different geomorpho-
developed by fluvial erosion but also to those controlled by metric variables and methods are used for extraction of objects
completely different surfaces, such as drumlin fields. They also from the field defined by a DEM. Among these we may
apply to three-dimensional networks of trees or lungs, and identify several unifying theoretical aspects (Figure 6).
wherever branches of space-filling networks join only in one Objects extracted from the field should be internally
direction, for example, downstream, and do not diverge homogeneous and this homogeneity should change abruptly
30 Geomorphometry: Quantitative Land-Surface Analysis

Analytical approach
B Change of
O Selection Zero Derivation Maximum Discontinuity
Isoline resolution
U isoline line line
N Nearing
D
A Boundary maximizing
R external heterogeneity
Y Statistical
approach
Area maximizing
internal homogeneity
A Degradation
R Nearing
E Selection Selection
Interval Binary Zero Constant
A Derivation
value Interval → 0 value

Analytical approach

Definition relations Possibilities of transformation

Figure 6 Relations among various types of morphometrically defined elementary geomorphic areas (segments) and their boundaries, according
to Minár and Evans, 2008.

at the object boundary (Drăgut¸ and Blaschke, 2006). Recent Minár and Evans, 2008), providing a strong potential tool for
work on landform classification is based mainly on multi- their interpretation (see Chapter 14.4).
variate statistical methods such as cluster analysis (e.g., Irvin It may not always be appropriate to delimit precise
et al., 1997; MacMillan et al., 2000; Iwahashi and Pike, 2007). boundaries for landforms or elementary forms. This has been
Minimization of intraclass and maximization of interclass recognized by a number of papers, such as Irvin et al. (1997)
differences for land elements are provided by cluster analysis and MacMillan et al. (2000), applying fuzzy logic to sub-
but this does not optimize definition of individual objects division or classification of the land surface. Generally, it is
(land elements). Selection of the input variables is another convenient to defuzzify by allocating each pixel to the class in
problem: geomorphological interpretation can be hindered which it has the greatest membership (Arrell et al., 2007).
when a nonsystematic mix of geomorphometric and other Evans (2012) recognized that as well as fuzzy boundaries,
landscape characteristics is used. there may be multiple (alternate) boundaries. Fuzziness may
Alternative (purely geomorphometric) analytical methods also apply to linear objects such as river channels and net-
define inner homogeneity of the segments, most commonly works, especially at different stages of discharge. A final point
by interval values of geomorphometric variables (e.g., Speight, is that we have focused here on linear and areal conceptual-
1968; Krcho, 1973; Dikau, 1989). Isolines of the selected izations of objects, all of which are actually three dimensional
variables are boundaries of the segments in this case. This is in detail. Volumetric analyses are also feasible.
problematic, because isolines do not in general represent any
sudden change of the topography and geomorphic processes
and, therefore, they cannot be genetically interpreted. Another 14.3.6 Scaling and Scale Specificity
analytically based method of topographic subdivision is
focused on the individual character of the boundaries – Once areal and linear objects have been delimited and
various types of discontinuities (structural lines) with clear measured, frequency distributions of their size and shape may
genetic and dynamic interpretation (Waters, 1958; Lastoczkin, be summarized, by descriptive statistics such as moment- or
1987). quantile-based measures. Two fundamental questions are how
An analytically based theoretical synthesis of these meth- constrained or varied are their sizes, and how shape varies
ods is provided by the concept of elementary forms (Minár with size. Evans (2003, 2010) has suggested that most land-
and Evans, 2008), defined morphologically so that they also forms (especially mobile bedforms) are scale specific; they
possess dynamic and genetic unity. They are defined by max- have lower and upper limiting sizes, related to process
imal external heterogeneity on every individual boundary, thresholds and the frameworks within which they develop.
represented in the ideal case by discontinuity of some point- Cirque length and width, and drumlin width, vary over ap-
based local geomorphometric variables, as well as by maximal proximately one metric decimal order of magnitude (e.g.,
internal homogeneity represented in the ideal case by constant from 200 m to 2 km). Their frequency distributions are close
values of some variables (Figure 6). Moreover, ideal elem- to Gaussian (normal) on logarithmic scales: they are com-
entary forms can be defined also by fitted mathematical monly portrayed by histograms, but a more sensitive way of
functions of altitude (cf. Troeh, 1965; Schmidt et al., 2003; comparing them with Gaussian or other models (e.g., gamma)
 Geomorphometry: Quantitative Land-Surface Analysis 31

Sizes of 25 848 british drumlins

Length (m) Length regression
5000 Width (m) Width regression
Amplitude (m) Amplitude regression

Length
1000
2000
Dimensions (m), logarithmic scale

500

Width
1000
100

50

500

10

5 Height

200

100
−4 −2 0 2 4
Standard normal deviate 418 cirques,
Figure 7 Quantile–quantile plots of length, width, and height of 25 Wales and Cumbria
848 British drumlins. Quantiles of the data are plotted against
quantiles of a model probability distribution function; in this case, the
200 500 1000
log-normal (log-Gaussian). The x-axis is labeled in standard deviation
units. Height fits the (straight-line) model well, but length and width Size (m)
show systematic deviations for larger drumlins. Data provided by Figure 8 Allometric plots of cirque length, width, and amplitude in
C. Clark and M. Spagnolo, see Clark et al. (2009). Wales and the English Lake District. The lower gradient for
amplitude shows that larger cirques increase in length and
is from their linearity on a quantile–quantile plot (Figure 7: width more than in vertical dimension. Reproduced from Evans, I.S.,
Wilk and Gnanadesikan, 1968; Cox, 2007). 2012. Geomorphometry and landform mapping: what is a landform?
As mobile bedforms such as sand dunes develop, they Geomorphology 137(1), 94–106 (In ‘Geospatial Technologies and
initially grow, but tend to reach an equilibrium size controlled Geomorphological Mapping’, Proceedings of the 41st Annual
by feedbacks and environmental conditions. This process Binghamton Geomorphology Symposium).
can be observed, but for many erosional forms the timescales
are much greater than human lifetimes. It may still be inferred than they are deeper. The logarithmic regression exponents are
that larger forms are more fully developed than smaller statistically significantly different (Evans, 2010, 2012).
ones, and the variation of size with shape gives clues to
landform development. If shape does not vary with size,
development is said to be isometric, as is required by simple 14.3.7 Conclusions: The Future
fractal models.
Commonly, landform shape does vary with size. This is Geomorphometry has been revitalized by the increasing
clear when height/length or width/length ratios vary between availability of data: globally (terrestrial and bathymetric), for
different size classes. One way of analyzing this more thor- other planetary and lunar surfaces, and in great detail (even
oughly is to plot various size components against an overall 1-m mesh, for quite large areas). This has opened up many
measure of size; logarithmic scales are generally appropriate to new applications, both theoretical and practical, within geo-
give the even scatter of points around a line desirable for re- morphology and in other fields such as soil science, ecology,
gression analysis. Thus, Grosse et al. (2009) demonstrated the and engineering.
allometric growth of Andean volcanoes and showed how this The future of geomorphometry depends on the harmonic
related to processes. Figure 8 shows how vertical amplitude of development of its interacting theoretical, empirical, and tech-
glacial cirques changes with size less than do length and nological aspects. Geomorphometric work has been pre-
width. This is an example of allometry (static rather than dy- dominantly empirical and led by technological developments,
namic, as the data relate to only one point in time) (Church yet geomorphometric theory has potential to influence the
and Mark, 1980): large cirques are wider and longer more content and quality of empirical research. One example of
interaction between geomorphometric theory and empirical
32 Geomorphometry: Quantitative Land-Surface Analysis

investigations came from the positive correlation between Clark, C.D., Hughes, A.L.C., Greenwood, S.L., Spagnolo, M., Ng, F.S.L., 2009. Size
horizontal and vertical curvatures, weak yet consistent over 67 and shape characteristics of drumlins, derived from a large sample, and
varied DEMs (Evans and Cox, 1999). Shary (1995) and Shary associated scaling laws. Quaternary Science Reviews 28, 677–692.
Clarke, J.I., 1966. Morphometry from maps. In: Dury, G.H. (Ed.), Essays in
et al. (2005) introduced a comprehensive system of curvatures Geomorphology. Heinemann, London, pp. 235–274.
and postulated a statistical hypothesis that everyone of 12 types Cox, N.J., 2007. Stata tip 47: quantile–quantile plots without programming. Stata
of forms defined by these curvatures should occupy an ap- Journal 7(2), 275–279.
proximately equal area on the Earth’s surface. This provided a Deng, Y., 2007. New trends in digital terrain analysis: landform definition,
representation and classification. Progress in Physical Geography 31(4),
theoretical explanation for Evans and Cox’s empirical results.
405–419.
Tests are needed for the theoretical postulate that real Dietrich, W.E., Montgomery, D.R., 1998. Hillslopes, channels and landscape scale.
segments of topography have a tendency to approximate ideal In: Sposito, G. (Ed.), Scale Dependence and Scale Invariance in Hydrology.
elementary forms defined by constant values of some geo- Cambridge University Press, Cambridge, pp. 30–60.
morphometric variables (Minár and Evans, 2008). If any Dikau, R., 1989. The application of a digital relief model to landform analysis in
geomorphology. In: Raper, J. (Ed.), Three-dimensional Applications in
variable tends to be locally constant, its derivatives tend to
Geographical Information Systems. Taylor and Francis, London, pp. 51–77.
zero values. Positive skewness of slope gradients (Figure 3) Dodds, P.S., Rothman, D.H., 2000. Scaling, universality, and geomorphology.
supports this for constant altitudes. But testing affinity to Annual Reviews of Earth and Planetary Sciences 28, 571–610.
constant values of curvature requires accurate computation of Drăgut¸, L., Blaschke, T., 2006. Automated classification of landform elements using
third-order geomorphometric variables that alone can define object-based image analysis. Geomorphology 81, 330–344.
Egholm, D.L., Nielsen, S.B., Pedersen, V.K., Lesemann, J.-E., 2009. Glacial effects
some types of elementary forms. In any case, the development
limiting mountain height. Nature 460, 884–888. http://dx.doi.org/10.1038/
of algorithms for third derivative computation (e.g., Florinsky, nature08263.
2009; Jenčo et al., 2009) has acquired a new importance. The Evans, I.S., 1972. General geomorphometry, derivatives of altitude and descriptive
late recognition of the importance of the third derivative statistics. In: Chorley, R.J. (Ed.), Spatial Analysis in Geomorphology. Methuen,
(Minár and Evans, 2008) means that we are reluctant to de- London, pp. 17–90.
Evans, I.S., 1987. The morphometry of specific landforms. In: Gardiner, V. (Ed.),
clare the present set of derivatives to be complete; later
International Geomorphology 1986 Part 2: Proceedings of the First
workers may find higher derivatives to be important. International Conference on Geomorphology, Manchester. Wiley, Chichester,
Geomorphometry is an active field at an exciting phase of pp. 105–124.
development, currently dominated by a huge increase in ac- Evans, I.S., 1998. What do terrain statistics really mean? In: Lane, S.N., Richards,
curate high-resolution data as a result of technological in- K.S., Chandler, J.H. (Eds.), Landform Monitoring, Modelling and Analysis.
Wiley, Chichester, ch. 6, pp. 119–138.
novation in radar and LiDAR sensing of the ground surface. In
Evans, I.S., 2003. Scale-specific landforms and aspects of the land surface.
terms of techniques, the greatest need is for automation of In: Evans, I.S., Dikau, R., Tokunaga, E., Ohmori, H., Hirano, M. (Eds.), Concepts
recognition of elementary forms and landforms (Schmidt and and Modelling in Geomorphology: International Perspectives. Terrapub, Tokyo,
Dikau, 1999). In some cases, this may require relating surface pp. 61–84.
form to the underlying materials; but there is much evidence Evans, I.S., 2010. Allometry, scaling and scale-specificity of cirques, landslides and
other landforms. Transactions of the Japanese Geomorphological Union 31(2),
in surface form alone that has yet to be efficiently related to
133–153.
geomorphological processes. Geomorphometric analysis Evans, I.S., 2012. Geomorphometry and landform mapping: what is a landform?
should, however, be targeted and based on a coherent Geomorphology 137(1), 94–106 (In ‘Geospatial Technologies and
theoretical–methodological background: this is the main Geomorphological Mapping’, Proceedings of the 41st Annual Binghamton
message of this chapter. Theory, for example, on scaling rela- Geomorphology Symposium).
Evans, I.S., Cox, N.J., 1999. Relations between land surface properties: altitude,
tionships of networks in relation to surfaces, requires further
slope and curvature. In: Hergarten, S., Neugebauer, H.J. (Eds.), Process
development. Modelling and Landform Evolution. Springer, Berlin, pp. 13–45.
For much more comprehensive accounts of the concepts, Evans, I.S., McClean, C.J., 1995. The land surface is not unifractal; variograms,
techniques and applications of geomorphometry, the reader is cirque scale and allometry. Zeitschrift für Geomorphologie N.F. Supplement-Band
referred to the texts edited by Hengl and Reuter (2009) and Li 101, 127–147.
Florinsky, I.V., 2009. Computation of the third-order partial derivatives from a digital
et al. (2005), and the collections edited by Pike and Dikau
elevation model. International Journal of Geographical Information Science
(1995); Lane et al. (1998); Wilson and Gallant (2000); 23(2), 213–231.
Peckham and Jordan (2007); Zhou et al. (2008), and Purves Goudie, A., Anderson, M., Burt, T., Lewin, J., Richards, K., Whalley, B., Worsley, P.,
et al. (2009). The field is reviewed also by Pike (2000); Deng et al. (Eds.), 1990. Geomorphological Techniques, Second ed. Unwin Hyman,
(2007), and Evans (2012) as well as Chapters 3.6, 3.7, London, 570 pp.
Grosse, P., van Wyk de Vries, B., Petrinovic, I.A., Euillades, P.A., Alvarado, G.E.,
and 14.2. 2009. Morphometry and evolution of arc volcanoes. Geology 37,
651–654.
Hack, J., 1957. Studies of longitudinal stream profiles in Virginia and Maryland.
U.S. Geological Survey Professional Paper, 294-B.
Hengl, T., Reuter, H.I. (Eds.), 2009. Geomorphometry: Concepts, Software,
References
Applications. Developments in Soil Science. Elsevier, Amsterdam, 765 pp.
Irvin, B.J., Ventura, S.J., Slater, B.K., 1997. Fuzzy and isodata classification of
Arrell, K.E., Fisher, P., Tate, N., Bastin, L., 2007. A fuzzy c-means classification of landform elements from digital terrain data in Pleasant Valley, Wisconsin.
elevation derivatives to extract the morphometric classification of landforms in Geoderma 77, 137–154.
Snowdonia, Wales. Computers and Geosciences 33, 1366–1381. Iwahashi, J., Pike, R.J., 2007. Automated classifications of topography from DEMs
Chorley, R.J., 1966. The application of statistical methods to geomorphology. by an unsupervised nested-means algorithm and a three-part geometric
In: Dury, G.H. (Ed.), Essays in Geomorphology. Heinemann, London, pp. signature. Geomorphology 86, 409–440.
275–387. Jenčo, M., 1992. The morphometric analysis of georelief in terms of a theoretical
Church, M.A., Mark, D.M., 1980. On size and scale in geomorphology. Progress in conception of the complex digital model of georelief. Acta Facultatis Rerum
Physical Geography 4, 342–390. Naturalium Universitatis Comenianae, Geographica 33, 133–154.
 Geomorphometry: Quantitative Land-Surface Analysis 33

Jenčo, M., Pacina, J., Shary, P.A., 2009. Terrain skeleton and local morphometric Natural Environment. E. Schweizbart’sche Verlagsbuchhandlung, Stuttgart, pp.
variables: geosciences and computer vision technique. In: Horák, J., Halounová, 153–173.
L., Kusendová, D., Rapant, P., Voženı́lek, V. (Eds.), Advances in Geoinformation Schmidt, J., Evans, I.S., Brinkmann, J., 2003. Comparison of polynomial models
Technologies. VŠB – Technical University of Ostrava, Ostrava, pp. 57–76. for land surface curvature calculation. International Journal of Geographical
Krcho, J., 1973. Morphometric analysis of relief on the basis of geometric aspect of Information Science 17, 797–814.
field theory. Acta Geographica Universitatis Comenianae, Geographico-physica 1, Schneider, B., 2001. Phenomenon-based specification of the digital representation
11–233. of terrain surfaces. Transactions in GIS 5(1), 39–52.
Krcho, J., 1992. Georelief and its cartographic modelling by complex digital model Schumm, S.A., 1956. Evolution of drainage systems and slopes in badlands at
(CDM) from geographical information systems (GIS) point of view. Acta Perth Amboy. Geological Society of America Bulletin 67, 597–646.
Facultatis Rerum Naturalium Universitatis Comenianae, Geographica 33, Shary, P.A., 1995. Land surface in gravity points classification by a complete
3–132. system of curvatures. Mathematical Geology 27(3), 373–390.
Krcho, J., 2001. Modelling of Georelief and its Geometrical Structure Using DTM: Shary, P.A., 2008. Models of topography. In: Zhou, Q., Lees, B., Tang, G.-A. (Eds.),
Positional and Numerical Accuracy. Q 111. Vydavatel’stvo Q111, Bratislava, 336 Advances in Digital Terrain Modelling. Lecture Notes in Geoinformation and
pp. Cartography. Springer, Berlin, pp. 29–57.
Lane, S.N., Richards, K.S., Chandler, J.H. (Eds.), 1998. Modelling and Analysis. Shary, P.A., Sharaya, L.S., Mitusov, A.V., 2002. Fundamental quantitative methods of
Wiley, Chichester, 453 pp. land surface analysis. Geoderma 107, 1–32.
Lastoczkin, A.N., 1987. Morfodynamicheskiy Analiz. Nedra, Leningrad, 254 pp. Shary, P.A., Sharaya, L.S., Mitusov, A.V., 2005. The problem of scale-specific and
Li, Z., Zhu, Q., Gold, C., 2005. Digital Terrain Modeling: Principles and scale-free approaches in geomorphometry. Geografia Fisica e Dinamica
Methodology. CRC Press, Boca Raton, FL, 319 pp. Quaternaria 28, 81–101.
MacMillan, R.A., Pettapiece, W.W., Nolan, S.C., Goddard, T.W., 2000. A generic Speight, J.G., 1968. Parametric description of land form. In: Stewart, G.A. (Ed.),
procedure for automatically segmenting landforms into landform elements using Land Evaluation. Macmillan, Melbourne, pp. 239–250.
DEMs, heuristic rules and fuzzy logic. Fuzzy Sets and Systems 113, 81–109. Strahler, A.N., 1950. Equilibrium theory of erosional slopes approached by
Minár, J., Evans, I.S., 2008. Elementary forms for land surface segmentation: the frequency distribution analysis. American Journal of Science 248, 673–696 and
theoretical basis of terrain analysis and geomorphological mapping. 800–814.
Geomorphology 95, 236–259. Strahler, A.N., 1952. Hypsometric (area-altitude) analysis of erosional topography.
Minár, J., Sládek, J., 2009. Morphological network as an indicator of a Geological Society of America Bulletin 63, 1117–1142.
morphotectonic field in the central Western Carpathians (Slovakia). Zeitschrift für Troeh, F.R., 1965. Landform equations fitted to contour maps. American Journal of
Geomorphologie, Supplementary 53(2), 23–29. Science 263, 616–627.
Mitchell, S.G., Montgomery, D.R., 2006. Influence of a glacial buzzsaw on the Waters, R.S., 1958. Morphological mapping. Geography 43, 10–17.
height and morphology of the Cascade Range in central Washington State, USA. Wilk, M.B., Gnanadesikan, R., 1968. Probability plotting methods for the analysis of
Quaternary Research 65, 96–107. data. Biometrika 55, 1–17.
Moore, I.D., Grayson, R.B., Ladson, A.R., 1991. Digital terrain modelling: a review Wilson, J.P., Gallant, J.C. (Eds.), 2000. Terrain Analysis: Principles and
of hydrological, geomorphological, and biological applications. Hydrological Applications. Wiley, New York, NY, 479 pp.
Processes 5, 3–30. Woldenberg, M., 1972. The average hexagon in spatial hierarchies. In: Chorley, R.J.
Ohmori, H., 1993. Changes in the hypsometric curve through mountain building (Ed.), Spatial Analysis in Geomorphology. Methuen, London, pp. 323–352.
resulting from concurrent tectonics and denudation. Geomorphology 8, Wood, J., 1996. Scale-based characterization of digital elevation models. In: Parker,
263–277. D. (Ed.), Innovations in GIS 3. Taylor and Francis, London, ch. 13, pp.
Peckham, R.J., Jordan, G. (Eds.), 2007. Digital Terrain Modelling, Development and 163–175.
Applications in a Policy Support Environment, Series: Lecture Notes in Zavoianu, I., 1985. Morphometry of Drainage Basins. Developments in Water
Geoinformation and Cartography. Springer, Berlin, pp. 67–82. Science. Elsevier, Amsterdam, vol. 20, 238 pp.
Pike, R.J., 2000. Geomorphometry – diversity in quantitative surface analysis. Zhou, Q., Lees, B., Tang, G.-A. (Eds.), 2008. Advances in Digital Terrain Analysis.
Progress in Physical Geography 24, 1–20. Series: Lecture Notes in Geoinformation and Cartography. Springer, New York,
Pike, R.J., Dikau, R. (Eds.), 1995. Geomorphometry. Zeitschrift für Geomorphologie, NY, 462 pp.
N.F. Supplement-Band 101.
Purves, R., Gruber, S., Straumann, R., Hengl, T., (Eds.), 2009. Geomorphometry
2009 Conference Proceedings. University of Zurich, Zurich, 263 pp. /http://
www.geo.uzh.ch/~rsm/download/proceedings_geomorphometry2009.pdfS; See Relevant Websites
also http://www.geomorphometry.org/system/files/geomorphometry2009ToC.pdf
Rodrı́guez-Iturbe, I., Rinaldo, A., 1997. Fractal River Basins – Chance and Self- http://www.geomorphometry.org
Organization. Cambridge University Press, Cambridge, 564 pp. Geomorphometry: The science of digital terrain analysis.
Schmidt, J., Dikau, R., 1999. Extracting geomorphometric attributes and objects
from digital elevation models–semantics, methods, future needs. In: Dikau, R.,
Saurer, H. (Eds.), GIS for Earth Surface Systems – Analysis and Modeling of the

Biographical Sketch
Jozef Minár is a professor of physical geography and geoecology at the Comenius University in Bratislava (Slo-
vakia), as well as at the University of Ostrava (Czech Republic). He deals with concepts of geomorphological
information systems as specific tools for detailed geomorphological research and mapping, mainly by develop-
ment of an analytically oriented land-surface segmentation procedure. He also works on the evolutionary geo-
morphology of the Western Carpathians, mainly in the terms of morphostructural analyses (targeted
morphometric analysis and morpholineament analysis). A further area of interest lies in environmental appli-
cations: geoecological mapping and natural hazard assessment.
34 Geomorphometry: Quantitative Land-Surface Analysis

Ian S Evans began working on slope measurement from maps (Wentworth’s method) as an undergraduate at
Cambridge, and in 1967 he started programing computers to achieve this. He formalized a system of surface
derivatives to replace many more arbitrary measures, calculated them from DEMs, and elaborated the distinction
between specific and general geomorphometry. Ian has analyzed the specific geomorphometry of glacial cirques
and drumlins, using both manually measured indices and DEM-based attributes, and testing concepts of fractality,
scale-specificity, and allometry. Other projects included data analysis for ‘People in Britain, a Census Atlas’, and
physical weathering, but the main focus has been on mountain glaciation, including glacier distribution and
change and local asymmetry of glaciation world-wide. Ian worked in the NERC/RCA Experimental Cartography
Unit from 1968 to 1970 and lectured in the Geography Department at Durham University from then until his
retirement in 2008. He is now Emeritus Reader at Durham, associated professor at the University of Suceava, and
Honorary Member of both the Japanese Geomorphological Union and the Quaternary Research Association. He
served as Chairman (1996–97) of the British Geomorphological Research Group, and at the Zurich 2009 Geo-
morphometry Conference he was elected Chairman of the nascent International Society for Geomorphometry.

Jozef Krcho is a professor at the Comenius University in Bratislava (Slovakia). From 1991 to1997 he was dean of the Faculty of Natural Sciences; he founded
the Department of Cartography, Geoinformatics and Remote Sensing in 1993 and was head of the Department until 2004. Between 1964 and 1973, he
elaborated morphometric analysis of relief on the basis of geometrical aspects of field theory. He derived equations of land surface normal curvatures in
relation to Earth’s gravity field (contour lines as intersections of equipotential levels of the gravity scalar field, and slope lines (flow lines) as the shortest
connections of equipotential levels between two contour lines); he defined horizontal (plan) curvature and derived its equation. Using the Dupin indicatrix,
he demonstrated differences between the invariant principal curvatures of Gauss and normal curvatures specific to the gravity field, and derived formulas for
directional angles between them. He defined geometric forms on the basis of these normal curvatures and demonstrated differences between them and
Gaussian forms. He formulated a comprehensive digital elevation model for interdisciplinary use.
 14.4 The Modern Geomorphological Map
AC Seijmonsbergen, University of Amsterdam, Amsterdam, The Netherlands
r 2013 Elsevier Inc. All rights reserved.

14.4.1 Introduction 36
14.4.1.1 Geomorphological Mapping – ‘A World in Motion’ 36
14.4.1.2 New Techniques in Geomorphological Mapping 36
14.4.1.3 Chapter Overview 37
14.4.2 Methods and Geomorphological Maps 37
14.4.2.1 Concepts and Methods Used in Classical Mapping Systems 37
14.4.2.2 Methods Used in Modern Mapping 39
14.4.2.2.1 Digital environments and geomorphological information layers 39
14.4.2.2.2 Digitalization and updating of geomorphological information layers 40
14.4.2.2.3 Geomorphological mapping using digital elevation models 41
14.4.2.2.4 Publishing digital geomorphological maps 43
14.4.2.3 Applications of Geomorphological Maps 43
14.4.3 Modern Geomorphological Mapping and Geoconservation 44
14.4.3.1 Study Area, Classical and Digital Geomorphological Map 44
14.4.3.2 General Steps for Semi-Automated Geomorphological Mapping 45
14.4.3.2.1 Extraction of LSP and segmentation into objects 45
14.4.3.2.2 Rule development, image classification, and accuracy assessment 46
14.4.3.3 Application to Geoconservation of Landscapes 47
14.4.4 Conclusions and Closing Remarks 48
Acknowledgments 50
References 50

Glossary Geospatial portable document format

Differential global positioning system (DGPS) Provides (GeoPDF) Document format enabling some GIS
a network of fixed ground stations used to enhance the functionality within standard PDF files.
accuracy of location data collected by a GPS. Global positioning system (GPS) Satellite-based global
Digital elevation model (DEM) A digital data set navigation satellite system that provides location
consisting of terrain elevations for regularly spaced ground information on/near the Earth’s surface.
locations. Land-surface parameters (LSP) Measures, directly
Geographical information system (GIS) System derived from a DEM.
that integrates hardware, software, and data for Light detection and ranging (LiDAR) Optical remote-
capturing, managing, analyzing, and visualizing spatial sensing technology that measures the distance to an object
information. using light.

Abstract

Classical geomorphological maps are representations of the spatial distribution of landforms, materials and of the processes
responsible for their formation, in a single paper map. They contain a wealth of information that is generally documented
with the aid of symbol and color legends. Uniformity among geomorphological legends does not exist, mainly because the
traditional mapping ‘schools’ independently developed systems for use at different scales in a variety of landscapes.
Technological advances have changed the preparation, data collection, analysis, storage, and visualization of the modern
geomorphological map. Modern geomorphological maps are digital collections of information layers consisting of geor-
eferenced vector, raster and tabular data that are stored in a geodatabase. The data depicted in classical geomorphological
maps can be scanned and converted into digital geomorphological information layers using conversion legends. Both the
content and the precision of digital geomorphological information can be updated by using information from digital

Seijmonsbergen, A.C., 2013. The modern geomorphological map. In:

Shroder, J. (Editor in Chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise
on Geomorphology. Academic Press, San Diego, CA, vol. 14, Methods in
Geomorphology, pp. 35–52.

Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00371-7 35

36 The Modern Geomorphological Map

elevation models (DEMs). A further development is the (semi-)automated extraction and classification of geomorpho-
logical features from DEMs. Extraction and classification of geomorphological features are based on land-surface par-
ameters derived from DEMs. The visualization and publication of digital geomorphological information layers have
drastically changed. Birds-eye views, three-dimensional display and animations, virtual globe visualizations, geospatial
portable document format maps or access to geomorphological maps via web-based services are possible through remote
servers. The end-user interactively controls the information that is on-screen displayed, analyzed, and distributed. A case
study from the Austrian Alps is presented to illustrate how: (1) geomorphological information from classical maps can be
converted to digital information layers; (2) detailed DEMs are used in semi-automated object-based classification of
landforms; and (3) a modern geomorphological map is used for application to geoconservation.

different cartographic map design, symbolization, and visual-

14.4.1 Introduction
ization then classical paper maps (Otto et al., 2011). The
underlying geodatabases in a GIS can be queried on demand
14.4.1.1 Geomorphological Mapping – ‘A World in Motion’
on screen and in the field using, for example, a mobile GIS
The term ‘modern geomorphological map’ implies that there is device and the desired attributes can be displayed as infor-
an ancient geomorphological map. This is the classical paper mation layers in digital layouts. Such a geodatabase may con-
geomorphological map, which has developed over more than tain not only vector maps, for example, contour line maps and
50 years. Pictorial drawings with some geomorphic landscape hydrological features, but also digital elevation data in raster
features and descriptions were already made during early ex- format from which potential stream networks can automatic-
ploratory expeditions, such as the voyage of the Beagle in ally be extracted. For example, point data of existing meas-
1831–36 and the expeditions of John Wesley Powell through urements of process activity, for example, measurements of
the Grand Canyon in 1869 and 1871. The earliest examples of landslide activity over time (Seijmonsbergen and de Graaff,
maps presenting landforms and their genesis are those of H. 2006), and maps of forest change derived from remote-sensing
Gehne from 1912 (Klimaszewski, 1982), and the morpho- (RS) images and elevation data (Dorren et al., 2003) can be
logical atlas of Passarge (1914), which is a collection of eight included and used in analysis. The geomorphological map has,
descriptive map sheets, mainly focused on metrical landscape thus, become a repository of digital geomorphological infor-
elements. It was only in the early 1950s that consistent classical mation layers that comes in a variety of formats and can be
geomorphological maps were developed, without leaving ‘manipulated’ according to the requirements of the end-user.
unmapped areas blank. These classical maps depict the In modern geomorphological mapping, the following
geomorphological diversity by using a variety of symbol- and developments have occurred: (1) conversion of classical geo-
color-based legends. The layout, structure, and content of the morphological maps into digital geomorphological infor-
classical map depend on the ‘school’ that was responsible for its mation layers; (2) updating the content and location precision
development and the physical environment (e.g., mountains of the converted classical geomorphological information layers;
and plains) in which the legend was applied. This means that and (3) automated extraction and classification of geo-
no single approach exists for geomorphological mapping (Lee, morphological features from DEMs. In the last approach, there
2001). Most classical geomorphological maps, however, con- is a trend toward automated and semi-automated detection,
form to a description given by St-Onge (1981: p. 313): ‘‘An ideal isolation and mapping of specific geomorphological features,
geomorphological map should not only describe and explain such as deep-seated landslides (Booth et al., 2009), or erosion
landforms based on the morphogenesis of individual land- assessment of river banks using airborne laser-scanning data
forms, but also, more importantly, the explanations should be (Thoma et al., 2005), and toward mapping of geomorpho-
based on the relations between various landforms affected to logical information layers of the complete landscape. As an
varying degrees by numerous processes.’’ Geomorphologists, example, Dikau et al. (1995) analyzed a 200-m-resolution DEM
educated with fundamental theoretical concepts in geomorph- of New Mexico to classify the terrain into plains, tablelands,
ology, have since produced numerous maps in various geo- hills, and mountain types, which were subdivided into 24
morphological environments, each demanding a particular landform classes. On a global scale, slope gradient, local con-
scale of visualization, specific cartographic symbols, and color vexity, and surface texture were used by Iwahashi and Pike
design. A variety of point symbols, lines, and polygons were (2007) to classify the Earth’s surface into 16 terrain classes
traditionally used to document crisp and gradual, also referred based on combinations of these parameters.
to as diffuse or fuzzy, geomorphological field boundaries. For
example, the boundary between an alluvial plain and bedrock
14.4.1.2 New Techniques in Geomorphological Mapping
slope is in general an abrupt, well-defined crisp boundary. By
contrast, the boundary between a low-angle colluvial footslope New techniques for collecting, storing, analysis, and visual-
and the same alluvial plain may be gradual and, therefore, ization of data are introduced into geomorphological map-
difficult to delineate. ping as the result of technological advances. The most
Technological advances have provided new techniques and important technological advances are increased computer-
ways to map, visualize, and analyze the geomorphology of processing power, storage capacity, and enhanced graphical
landscapes. Paper maps are being replaced by digital platforms displays, along with new techniques to collect and analyze
such as geographical information systems (GISs), in which geomorphological information such as the differential global
geomorphological information is stored and which need positioning systems, (mobile) GIS, a variety of RS imagery of
 The Modern Geomorphological Map 37

high spatial and spectral resolution and DEMs. The wealth of presented of the same area. An accuracy assessment of the
data collected in the past, combined with the development of modern map is presented in a confusion matrix. The potential
an array of new computer-based quantitative techniques for of digital geomorphological information layers for application
characterizing landscapes and landscape change, has enabled in geoconservation using automated application of weighting
a renaissance in geomorphic theory and modeling to occur and ranking of assessment criteria is illustrated. Finally, in
recently (Murray et al., 2009). Modern geomorphologists can Section 14.4.4 conclusions are presented along with closing
analyze a growing set of data with a variety of analysis tools remarks on current and expected trends in geomorphological
that will improve the knowledge of Earth-surface processes. In mapping.
a GIS, geomorphological and thematic information can be
efficiently collected, stored, and analyzed, which enables the
representation and visualization of the land surface. This fa-
14.4.2 Methods and Geomorphological Maps
cilitates the understanding of landforms and other geo-
morphological phenomena, and reveals the spatial, temporal,
14.4.2.1 Concepts and Methods Used in Classical
and thematic relationships among processes, materials, land-
Mapping Systems
forms, and natural and anthropologic-controlling factors
(Remondo and Oguchi, 2009). Various open source and Geomorphological mapping has become an established and
commercial GIS packages are used by geomorphologists. recognized technique for the documentation and re-
Therefore, unique digital mapping and visualization standards construction of landscape development. The developments in
or protocols for displaying, analyzing, and visualization of modern geomorphological mapping will be better understood
layers of geomorphological information do not yet exist. after a short review of some of the most influential classical
Modern software tools and equipment are available that mapping systems and some concepts on which mapping of
speed up and refine geomorphological mapping procedures. geomorphological landform boundaries has been based.
Examples are precision mapping using mobile GIS systems In parallel with the development of classical geomorpho-
with integrated GPS and detailed DEMs from which auto- logical mapping systems, a number of conceptual landscape
mated extraction of terrain information replaces traditional models were developed such as the catena concept of Milne
mapping of the morphometry based on contour lines. Add- (1935), the nine-unit land-surface concept of Dalrymple et al.
itional information can be directly hyperlinked to map co- (1968), and the morphological zoning of Speight (1990).
ordinates in a geodatabase, for example, field descriptions, These concepts all provided a basis for dividing landscapes
terrain photos, lab analyses, and geophysical measurements. into coherent geomorphological zones. Boundaries of geo-
Generating and reading a geomorphological map has become morphological units are traditionally mapped by using a set of
an interactive activity using a computer. predefined legend tools, which comprise point, line, and area
The need for expert knowledge, however, remains; inter- symbols to depict morphometrical transitions in a landscape,
pretation of landscapes by an expert is still crucial for land- hydrological features, the material distribution, and processes
scape reconstruction, validation of current process activity, and to document the genesis of active and fossil landforms. In
and examination of the likely effects in the landscape of some legends, geological information has also been included.
projected climate change. An important step in the change The classical geomorphological map is primarily a scientific
from the classical to the new mapping approach is that expert document used to understand and reconstruct landscape de-
knowledge is translated into expert rules and is implemented velopment and it subsequently became popular in applied
in models for automated extraction of geomorphological studies. A different use requires that the existing classical map
landforms and processes. information is translated into derived thematic themes, which,
for example, depict the activity of slopes in natural hazard or
engineering geomorphological maps. Such applications re-
quire specific classification schemes to be developed for
14.4.1.3 Chapter Overview
mapping fossil or recent landslide activity (see, e.g., Varnes,
In this chapter, an overview is presented of current develop- 1978; Hansen, 1984; Cruden and Varnes, 1996).
ments in modern geomorphological mapping. In Section 14.4.2 The major steps to produce a classical geomorphological
important methods and developments of classical and modern map and subsequent derivative maps are shown in Figure 1.
geomorphological mapping are reviewed. First, a short overview The first step is prefield preparation, which includes the design
of concepts and methods used in classical geomorphological of the legend, selection of mapping scale, and the preparation
maps is presented. Second, examples of new methods of geo- of a base map for mapping, such as a contour line map.
morphological mapping are discussed, with emphasis on the Furthermore, existing information such as literature and rele-
use of DEMs, geodatabases, GIS, and visualization of geo- vant geological information is collected and reviewed. Usually,
morphological information layers. Differences between classic a prefield concept photointerpretation map is made using
applications and modern potential are also presented. panchromatic stereo photos. This results in a first reconnais-
The new approaches in geomorphological mapping are il- sance map, which is updated during the actual fieldwork.
lustrated in Section 14.4.3. First, the steps to convert a classical Delineation of units occurs on the basis of photogeological
map to a geodatabase that stores digital geomorphological in- legends, which are based on combinations of air–photo in-
formation layers and its application to geoconservation in the terpretation elements, such as gray-tone, texture, shape, rela-
Austrian Alps are explained. Second, a semi-automated tive age, morphostructural units, and other classification
method for object-based segmentation and classification is characteristics (Meijerink, 1988; Rivard, 2009).
38 The Modern Geomorphological Map

Contour line maps

geological map Lab. work Report
literature

Field work
Pre-field preparation Concept mapping Cartographic Classical
legend design sampling geomorphological
scale selection map design map
description
measurements

Expert knowledge Expert knowledge

Check
Air-photo / satellite Applied maps
interpretation

Figure 1 Workflow for the preparation of a classical geomorphological map and applied maps.

Table 1 Examples of classical geomorphological mapping systems

Country/system Scale References

Germany 1:25.000/1:100.000 Barsch and Liedtke (1980)

The Netherlands 1:50.000 Maarleveld et al. (1974)
UK 1:25.000 Cooke and Doornkamp (1990); Evans (1990)
Poland 1:50.000 Klimaszewski (1963)
Hungary Detailed Pécsi (1964)
Sweden 1:10.000 Elvhage (1980)
Gustavsson et al. (2006, 2008)
Austria 1:10.000 de Graaff et al. (1987)
World Detailed and medium scale Gilewska and Klimek (1968); Demek and Embleton (1978); Demek et al. (1972)
Italy 1:20.000 CNR (1986)
France 1:50.000, 1:30.000 Tricart (1965)
Canada 1:30.000 St-Onge (1964)
Spain 1:1.000.000 Martin-Serrano (2005)
Portugal 1:500.000 De Brum Ferreira (1981)
Europe 1:250.0000 Demek and Embleton (1978)
Switzerland, IGUL 1:10.000 Reynard et al. (2005), web 5
ITC system All scales Verstappen (1970)

The second step, fieldwork, comprises final delineation of et al., 1972; Demek and Embleton, 1978) with the various
landforms and process units, interpretation and description of mapping schools continuing to use their own specific legends at
key exposures, in situ measurements of material properties, various mapping scales and in different environments.
and additional material sampling for laboratory testing. These Most systems use point, line, and area symbols and colors
tests may include the grain-size distribution, mineral and clay to generate geomorphological maps. Making maps in the
composition, penetrometer testing, rock strength tests, or classical approach is a time-consuming and, thus, expensive
geophysical measurements. approach, especially in areas of high relief, dense vegetation,
In the third step, the final geomorphological map is drawn and inaccessible terrain. A selection of important classical pi-
and accompanied by explanatory notes including laboratory oneer mapping systems is presented in Table 1 (see also Sal-
results, illustrative cross sections, and chronostratigraphic in- omé et al., 1982; Klimaszewski, 1990; Gustavsson et al., 2006;
formation on landscape reconstruction. In the classical ap- Verstappen, 2011). Systematic and nationwide geomorpho-
proach, the experts’ knowledge is translated into hand-drawn logical mapping occurred in the Netherlands at 1:50,000 scales
geomorphological maps using point, line, and area symbols. (Maarleveld et al., 1974). In Germany, the production of maps
These are used in combination with colors to depict landform ceased after successful testing in representative German land-
geometry and boundaries, to present the spatial distribution of scape types (Barsch and Liedtke, 1980). In many countries,
surface materials, and to provide information on landscape however, detailed scale maps are not available, due to a lack of
genesis. A common standard, however, for classical paper geo- finances, absence of mapping expertise, or because there is no
morphological legends never became accepted despite efforts of need for such data in nonpopulated regions. Worth men-
the International Geomorphological Union (IGU) (Demek tioning are the papers of Gustavsson et al. (2006, 2008), who
 The Modern Geomorphological Map 39

recently designed a new geomorphological mapping system for with both analysis and visualization capabilities. A GIS is a
Sweden, based on critical review and comparison of classical computer-based system to aid in the collection, maintenance,
mapping systems. They emphasized the need for using and systematic storage, analysis, output, and distribution of spatial
translating classical geomorphological information into digital data (Bolstad, 2008). GIS was initially used as a mapping
GIS-ready information layers, which can subsequently be up- platform and to perform simple computations, but nowadays
dated. Few classical mapping systems, however, continue to they have become a powerful, integrated technology for sci-
exist in adapted format in the modern mapping era. If they still entific and applied purposes (Remondo and Oguchi, 2009).
exist, their appearance, content, and the precision of the digital Most commercial GIS and RS software (e.g., ArcGIS, eCogni-
geomorphological information layers have drastically changed tion, IDRISI, Intergraph, Mapinfo, ERDAS Imagine, ENVI, and
as the result of technological advances. others) and noncommercial software (e.g., GRASS GIS, SAGA
GIS, ILWIS, Quantum GIS, MapWindow GIS, and others)
allow storage and manipulation of georeferenced raster, vec-
14.4.2.2 Methods Used in Modern Mapping tor, and tabular information. Most GIS’s support three-
dimensional (3D) visualizations as well as computational
Modern geomorphological maps are digital information layers
DEM analyses techniques and provide a database for attribute
that are linked to attributes and stored in GIS geodatabases.
storage and facilitate importing/exporting for data exchange
Information layers can be derived from scanning classical
with specialized database management systems such as Oracle,
paper maps that have been digitized into map features and
PostgreSQL, SQL Server, and DB2. These computer environ-
which can be updated using new data sources and mapping
ments, thus, provide the basic facilities for handling the in-
tools. New information layers that identify specific geo-
formation of modern geomorphological maps.
morphological features are derived from the analysis of DEMs
Features such as rivers and sinkholes and boundaries of
or semi-automatically generated maps. Both approaches use
landforms can be represented in vector format, whereas im-
different data sources and techniques for the generation of
ages and DEMs including derived land-surface parameters
geomorphological information layers. In Table 2, the princi-
(LSPs) can be represented by rasters (e.g., Voženı́lek, 2000). A
pal differences between classical and digital geomorphological
geomorphological geodatabase can include information about
information layers are summarized.
surface and subsurface properties such as stratigraphy and
lithology. The properties can be stored as sets of tables linked
14.4.2.2.1 Digital environments and geomorphological to the geographical positions or as attribute data. The design
information layers of such a GIS geodatabase is shown in Figure 2 along with the
Geomorphological information is mainly collected, generated, transfer process of the classical to the digitized geomorpho-
stored, and distributed through a GIS or modeling software logical map (Gustavsson et al., 2008). The tabular data are

Table 2 Some principal differences between classical and modern maps

Classical geomorphological map Digital geomorphological information layers

Digital classical maps/updated digital maps (Semi-)automated landscape mapping/

identification of specific geomorphological
features

Paper geomorphological map Digital vector/raster layers that carry geomorphological information
Symbol and color-based legend Style sets used for symbols Raster-based map
Content of the paper map based on expert Content based on digitized expert knowledge Automated mapping using DEMs and/or
knowledge and conversion legends imagery, guided by expert-rules
Morphometry/morphography indicated by lines Morphometry/morphography indicated by Morphometry/morphography replaced by
and symbols digital lines and symbols or separate GIS separate layers that carry detailed elevation
layers carrying elevation information information (DEMs, TINs, and/or digital
contour lines)
Materials and processes/genesis well visualized Materials and processes/genesis well visualized Little emphasis on materials, processes and
through symbols and or colors through digital symbols and or colors. genesis
Mapping time-consuming, field expert Preparation of modern map time efficient, less Fast and objective production of DEM-derived
knowledge necessary. Hand-drawn subjective. Digital point and line symbols and LSP raster maps, Updating fast using
visualization of symbols. Updating requires polygons, styles for color coding. Updating predefined rule sets and models
new map print / time consuming fast, time efficient.
Mapping and visualization at fixed scale Map generation and visualization independent Mapping is dependent on cell size of DEM or
of map scale in a GIS quality of triangular irregular network
Combination with other data layers laborious Integration and analysis with other information layers easy
and difficult
Applied maps translated to thematic maps Fast and reproducible applications from geodatabase and GIS analysis tools
Information on reliability/accuracy absent Accuracy increases in case of updated digital Accuracy assessment mostly included in
maps automatically generated maps
40 The Modern Geomorphological Map

(a)
(c)

(a)

(b)

(c)

(d)
Digitized geomorph. map
(e)

(f)

Geomorphological map GIS geodatabase

(b)

Figure 2 Transfer process of a classical geomorphological map (lower left) into a digital geomorphological map of points, lines, and polygons.
Tabular data ((c), top right) are related to vector data and integrated in the geodatabase. In the GIS geodatabase (lower right) the various
information layers are represented. Reproduced from Gustavsson, M., Seijmonsbergen, A.C., Kolstrup E., 2008. Structure and contents of a new
geomorphological GIS database linked to a geomorphological map – with an example from Liden, central Sweden. Geomorphology 95, 335–349.

linked to the vector data and the various information layers Netherlands was one of the first countries in the world to
are represented in the GIS geodatabase. Additional layers and produce and complete a country-wide geomorphological map
attributes can be added to the geodatabase, which may form series (GKN50) at 1:50.000 scales (Ten Cate, 1983).
the basis for applied studies. Gustavsson et al. (2008) also These GKN50 paper maps are static documents that refer to
presented a set of point, line, and polygon features for use in the situation at the time of map preparation before actual
digital geomorphological mapping at various scales and dif- publication. Most maps, globally, are more than 25-years old
ferent environments. The modern geomorphological map is, and likely need revision. It is anticipated that in the near future,
thus, not a single document, but consists of separate infor- most paper maps will be scanned, georeferenced, then con-
mation layers, that can be analyzed or displayed in combin- verted into vector format to be stored in a geomorphological
ation with other information layers. geodatabase. The following steps are generally necessary to
Modern geomorphological mapping requires software that is convert a classical printed geomorphological map into a digital
also specialized in handling (large) digital elevation data and form: (1) scanning, georeferencing, and assigning a coordinate
allows the programming and (semi-)automated extraction of system to the classical map; (2) design of a digital legend and
new LSPs in a flexible way. Examples of popular commercial and geodatabase; (3) digitization of point, line, and polygon in-
free DEM analysis software are TOPAZ, TAPES, Anudem, Land- formation on the scanned map; (4) assigning attribute infor-
Serf and MicroDEM, and the GIS software mentioned earlier. mation to the vector data; and (5) color coding of
In a GIS, the visualization and reading of geomorpho- geomorphological features and design of the cartographic
logical maps depend on the demands of the user. Digital output. Updating of the classical map is then possible, by using
products can be generated on demand, through specific data sources such as detailed DEMs, soil, and geological maps,
querying, selection, and combination of data stored in the high-resolution orthophotos, and hyperspectral imagery.
geodatabase. 3D visualization and modeling techniques and In the Netherlands, a new style of mapping was introduced
web interfaces strongly influence the visualization and distri- by Koomen and Maas (2004) and upgrades the existing clas-
bution of geomorphological information layers to a broader sical maps by using a 5-m cell resolution hillshade map based
audience. on the ‘Algemeen Hoogte Model’ (AHN5), the DEM of the
Netherlands. In combination with the digitized 1:50.000 scale
14.4.2.2.2 Digitalization and updating of soil map of the Netherlands (‘BodemKaart Nederland’ –
geomorphological information layers BKN50), landform boundaries and interpretation of landform
Classical geomorphological maps commonly need to be re- genesis was improved. Part of the digitized classical map of the
vised and/or updated after they have been scanned, georefer- province of Drenthe, along with the updated map, is shown in
enced, and converted into digital information layers. The Figure 3. The DEM, which was used to update the classical
 The Modern Geomorphological Map 41

Figure 3 Fragments of digital geomorphological map (GKN50) of part of the Drenthe Province in the northeast of the Netherlands. Left: digital
geomorphological map converted from the classical paper map. In the center, the 5-m DEM is shown. Right: updated digital geomorphological
map. Area shown approximately 25 km2. Reproduced with permission of WUR – Alterra, The Netherlands.

map, is shown in the center of Figure 3. The improved level of with such analysis, classification rules are developed for the
detail is apparent, which confirms the need to improve identification of units for geomorphological landscape
geomorphological information in a geodatabase. The map mapping.
fragments cover a formerly glaciated area in which primarily Two important developments in geomorphological map-
ice-pushed ridges (reddish colors) on a till-plain (orange/ ping using DEMs are identification of geomorphological fea-
brown colors) are dissected by secondary small streams tures, and (semi-)automated landscape mapping. In both
(greenish/bluish colors) or overlain by younger aeolian inland developments, LSPs derived from DEMs form the basis for the
dunes with accompanying closed depressions (yellowish col- extraction and classification of geomorphological information
ors). Grayish colors refer to built-up areas. In the new map, layers. Geomorphological feature extraction emphasizes the
the detail of the drainage network, depression density (pingo automated identification of single features, such as alluvial
ruins, in light blue), and appearance of newly recognized ice- fans, landslide scars or karst depressions. Automated land-
pushed ridge areas is striking after the update. The update was scape mapping seeks to generate a complete rule-based
also used to redesign the legend descriptions and colors approach for the (semi-)automated extraction of geo-
according to modern standards. New insight in landscape morphological information layers from DEMs. Here, super-
reconstruction and development has accelerated the need for vised classification techniques can be used, which may
landscape preservation in this part of in the Netherlands, incorporate formalized expert rules that are similar to un-
based on this new style of mapping. written expert rules on which classical maps are based.
Another example of updating an existing classical map DEMs are used to extract LSPs (also known as topographic
layer comes from Newell and Clark (2008). The use of a 2-m- derivatives), which carry statistical information for identifi-
resolution light detection and ranging (LiDAR) DEM revealed cation and classification of geomorphological features. This is
new details of surface geomorphology in Maryland (USA) and the home of ‘geomorphometry’, a discipline that specializes
improved knowledge of stratigraphic relations resulting from on quantitative description of the landscape (see Hengl and
the new geomorphic details. Reuter (2008) for an extensive review of techniques and ap-
plications of geomorphometry in soil science and geo-
14.4.2.2.3 Geomorphological mapping using digital morphology). Pike et al. (2008) stated that geomorphometry
elevation models is a discipline that utilizes DEMs for quantitative and quali-
The wide availability of high-resolution elevation information tative description of the landscape. Some examples of LSPs
derived from RS, the increase in computing power, and the include slope angle, plan and profile curvature, aspect, ele-
development of new modeling techniques progresses the vation percentile, and topographic openness (Figure 4 and
automated extraction of geomorphological information layers see, e.g., Hengl and Reuter, 2008). Such LSPs hold potential
from DEMs. Evans et al. (2008) mentioned three applications information on geomorphological landforms and processes.
of DEMs that are useful in automated geomorphological Their statistical analyses may result in precise definition of
mapping: (1) automated recognition and quantification of landform boundaries and also provide morphometrical/mor-
geomorphological properties; (2) automated extraction of phographical information. Many unsupervised and supervised
hydrologic/denudation structures; and (3) automated ex- classification techniques, such as fuzzy k-means techniques
traction of landforms. An example of geomorphometric an- (Burrough et al., 2000), are used to differentiate landforms or
alysis used to automatically generate a geomorphological map identify specific landform elements. Multivariate statistical
of an alpine area is presented in Section 14.4.3. In parallel analysis (Adediran et al., 2004), pixel-based and object-based
42 The Modern Geomorphological Map

Elevation Hillshade

Slope Aspect

Elevation percentile 25 Elevation percentile 250

Topographic openness 25 Topographic openness 250

Figure 4 Selection of land-surface parameters. Extent of the area is the same as in Figure 6.

image segmentation and classification (Drăgut¸ and Blaschke, maximized. Aspect represents a circular variable (0–3601)
2006), unsupervised classification (Bue and Stepinski, 2006; describing the direction of this true slope angle. Elevation
Iwahashi and Pike, 2007), and region-growing classification percentile is a measure of relative position in the landscape,
(Etzelmuller et al., 2007) techniques have been used to classify and is calculated as the percentage of pixels that are located
geomorphological features. In all these studies, a selection or higher relative to a center pixel within a moving window.
combination of various LSPs forms the basis to identify ‘Topographic openness’ is an angular measure of enclosure of
landforms, based on thresholds that exist in gradient, con- an object or pixel in the landscape (Yokoyama et al., 2002).
cavity size, elevation range, and so on. According to Prima et al. (2006), slope in combination with
In Figure 4, some examples of LSPs are shown, which are topographic openness can be used for genetic interpretation of
also used in the case study in Section 14.4.3. The elevation is topography. Therefore, for generation of boundaries of po-
based on processing of a raw laser-altimetry data set that was tential landforms or geomorphological features (objects),
separated in terrain and canopy-reflection points. The re- both fine and coarser LSP values have relevance for identifying
sulting 1-m-resolution DEM forms the basis of a hillshade, geomorphological boundaries. In Figure 4, the elevation
which is a representation of the original DEM created by il- percentile and topographic openness parameters illustrate that
lumination with a virtual light source. Slope is commonly the distance over which LSPs can be measured is flexible and
used in geomorphological feature extraction and represents that these parameters produce different types of information.
the rate of change of altitude in the direction where that rate is For example, elevation percentile calculated within a search
 The Modern Geomorphological Map 43

radius of 25 m gives detailed information on nearby topo- up windows showing geomorphological descriptions, photos,
graphic variation, whereas the same parameter calculated or other desired information. The end-user interactively de-
within a 250-m search radius translates to topographic vari- termines how to visualize the geomorphological information
ation occurring at broader distances. The same differences in and how to consult the underlying geodatabase.
detail occur in the topographic openness maps. The decision
on what search radius is best greatly depends on the purpose
of analysis.
14.4.2.3 Applications of Geomorphological Maps
The extraction of LSPs necessary for (semi-)automated
landscape mapping and/or the identification of specific geo- Classical geomorphological maps are traditionally used in a
morphological features are commonly done using program- wide range of applied disciplines closely related to geo-
ming languages such as R and Matlab. Further examples of morphology. Landforms form the basis for mapping and
popular DEM analysis software are TOPAZ, TAPE’, Anudem, measuring of other environmental issues (Goudie, 2006), such
LandSerf and MicroDEM, and the GIS software mentioned as the rate and causes of erosion and forecasting future erosion
earlier. levels under projected climate-change scenarios. The infor-
mation necessary for applied mapping is commonly ‘hidden’ in
14.4.2.2.4 Publishing digital geomorphological maps the classical geomorphological map, as a consequence of
New visualization techniques such as 3D-visualisations, vir- working with symbol-based legends. Therefore, applied geo-
tual globes such as Google EarthTM (see Chapter 2.10), visu- morphological maps are translated classifications of geo-
alizations of multiple data layers in a GIS or ‘layered maps’ in morphological information, converted by an expert into
GeoPdf format and web publishing are becoming powerful meaningful categories. The process is time consuming, in-
tools for publishing and distribution of digital geomorpho- volving the redrawing into new category boundaries, and the
logical information layers. manual coding of, for example, a hazard zonation or an en-
For use in a GIS, digital legend symbols of classical geo- gineering geomorphological map. The applied maps generally
morphological mapping are in development. Digital con- show a limited number of easily identified and communicated
version legends with predefined mixtures of red, green, and classes, such as highly suitable, suitable, low suitability, and
blue (RGB) color combinations are partly available or under unsuitable. A drawback is also that these maps are difficult to
construction. For the Swiss (Otto and Dikau, 2004) and reproduce, based on unwritten expert rules and without ac-
Swedish geomorphological legend system (Gustavsson et al., curacy assessments. Despite these disadvantages, many appli-
2008), such predefined digital symbol sets exist (further details cations of geomorphological maps were published, for
see Otto et al., 2011) and a scheme for converting the classical example, in highway engineering (Brunsden et al., 1975).
Austrian legend (de Graaff et al., 1987) into a digital legend has Kienholz (1977) used the concept of ‘silent witnesses’ (land-
been published by Seijmonsbergen et al. (2009). For planetary forms and deposits indicative of former process activity) in his
mapping (see Nass et al., 2011), a digital legend for use in geomorphological maps and converted these to so-called
ArcGIS is under development. ‘combined geomorphological hazard maps’ for the Swiss alpine
Many scientific journals publish computer-based modern Grindelwald area. Rupke et al. (1988) published detailed
geomorphological maps as paper copies or in digital format. 1:2.000-scale geomorphological maps that were used in en-
However, the additional attribute information is commonly gineering consultancy for the assessment of temporal landslide
neglected or only partially displayed in tabular format. dynamics in a small alpine catchment in eastern Switzerland.
Promising developments are digital and interactive ‘layered Seijmonsbergen (1992) made an inventory of 125 km2 of al-
maps’ in GeoPdf format, which can be exported from pro- pine geomorphology in Austria at scale 1:10.000 and designed
grams such as ArcGIS. GeoPdf maps carry various information a method for the preparation of natural hazard maps and
layers, where attribute information is displayed upon scrolling geotechnical maps.
over the various map units. Examples of such maps can be Dikau (1989) recognized the possibilities of GIS technol-
found in the Journal of Maps website, which welcomes pub- ogy for geomorphological application, especially for storage,
lication of maps in GeoPdf format, along with additional ex- applications in geo-scientific disciplines, and for simplification
planations on mapping procedures and legend development. of the mapping process by means of modeling and quantifi-
In contrast to geological maps, geomorphological maps cation of landforms. Van Westen et al. (1999) using GIS
have not widely been published through the internet. A compared landslide hazard maps made by different mapping
WebMap Server application for a geomorphological map of experts to generate general rules for the classification of
the Turtmann Valley in Switzerland is a step forward in the landslide intensity into high, medium, and low hazard, with
dissemination of such material. Digital geomorphological the objective of reducing subjectivity in applied geomorpho-
maps of Victoria, Australia, can be inspected at three levels of logical mapping. In a follow-up study, Van Westen et al.
detail. The German geomorphological maps are also access- (2000) tested both traditional ‘symbol-based geomorphological
ible through a web interface (Geilhausen et al., 2007). Web mapping’ and ‘unique condition polygon mapping’ for their
publication helps to develop scalable map symbols that are value in applied landslide mapping in northern Italy. Each
necessary for interactive zooming and integration with other polygon was assigned information that was documented in
thematic layers. Applied information held within the geo- checklists, for example, the genesis of landforms, chronos-
morphological geodatabase can, thus, interactively be com- tratigraphical order of landforms, Quaternary materials, haz-
municated to the general public, for example, using GIS web ard type, and degree of hazard, all based on geomorphological
services. Within digital maps, hyperlinks may open new pop- field indicators. From the geodatabase, five units could be
44 The Modern Geomorphological Map

displayed: main geomorphological units, geomorphological on screen and the results of scenario analyses from geomor-
subunits, process map, surface material map, and hazard map. phometrical models can easily be displayed and communi-
Seijmonsbergen and de Graaff (2006) used GIS to evaluate cated without much difficulty. Digital geomorphological
natural hazards in a small alpine catchment in the Principality information layers serve as valuable input for environmental
of Liechtenstein. They used a conversion key to prepare a impact assessment and in the management of landscapes, for
polygon-based geomorphological map from a classical example, in geoconservation (Gray, 2004).
1:3.000-scale geomorphological map and assigned attributes
to the polygons to spatially display processes and type of
hazards. In addition, data of geophysical measurements and
14.4.3 Modern Geomorphological Mapping and
statistical interpolation of temporal displacement data were
Geoconservation
used in a geodatabase to complete a hazard zonation map.
Mantovani et al. (2010) presented the results of a landslide
Most of the issues discussed in previous sections are illustrated
geomorphological inventory using an open source WebGIS
in the following case study. It shows how a classical geo-
MapServer and PostgreSQL geodatabases from which a geor-
morphological map (Seijmonsbergen, 1992) is converted to
eferenced map layout can interactively be generated.
digital information layers for use in a GIS geodatabase. In
These examples show that digital geomorphological in-
parallel, a semi-automated geomorphological map, accom-
formation layers are far more flexible in their ability to gen-
panied by an accuracy assessment for the same area, is made
erate applied information, due to the underlying geodatabases
based on laser altimetry technology, object-based segmen-
that contain many different types of manually or automatic-
tation, and landform-specific classification rules.
ally assigned attributes.
The information from the two approaches can be used in
Automated extraction of information from high-resolution
combination to produce layers of digital geomorphological
DEMs has initiated new applications in the field of geomor-
information, which form the basis for GIS-based geo-
phometry. Progress in the mathematical derivation of LSPs
conservation application. The application to potential geo-
and statistical parameters such as the terrain roughness, wet-
conservation mapping highlights the possibilities for using
ness index, and solar insolation (Iwahashi and Pike, 2007;
digital geomorphological information as a basis for landscape
Minár and Evans, 2008; Hengl and Reuter, 2008) provide a
conservation (Figure 5). The method for preparing a potential
wealth of newly generated raster data, which may serve as
geoconservation map includes assigning and weighting of as-
input for the modeling of landscape scenarios. Seijmonsber-
sessment criteria and ranking of criteria scores.
gen et al. (2011) provided an overview of developments and
applications of geomorphometry and showed how the use of
LSPs and statistical analyses can be used in the updating of
14.4.3.1 Study Area, Classical and Digital
existing geomorphological maps in the Netherlands. By con-
Geomorphological Map
trast, they illustrated how classical geomorphological maps in
alpine areas can be used for semi-automated mapping using The study area Dünza-Tschengla is located near the village
object-based segmentation and classification. Progress in Bürserberg in the province of Vorarlberg, western Austria.
visualization techniques and the implementation of 3D tools LiDAR data (1-m resolution), false-color orthophotos (25-cm
in common software packages allow visual inspection of resolution), infrastructure, and hydrologic GIS layers are
databases in the form of birds-eye views and fly-troughs. This available. The region suffers from tourist pressure with skiing
means that the underlying spatial database can be consulted areas expanding over recent years. Therefore, there is a need

Assigning and weighting of

assessment criteria

Environmental
Digitized classical vulnerability
geomorphological
map Disturbance Potential
Potential digital Ranking
geomorphosite geoconservation
Automated Frequency
units of occurrence map
geomorphological
map Scientific
relevance

Additional criteria

A: Input B: Analysis C: Output

Figure 5 Workflow for potential geoconservation mapping. Reproduced from Seijmonsbergen, A.C., Sevink, J., Cammeraat, L.H., Recharte, J.,
2010. A potential geoconservation map of the Las Lagunas area, northern Peru. Environmental Conservation 37(2), 107–115, with permission
from Cambridge.
 The Modern Geomorphological Map 45

Figure 6 Left: classical symbol and color geomorphological map fragment of map sheet Dünza-Tschengla (Seijmonsbergen, 1992); center:
polygons resulting from on-screen manual digitizing of the classical geomorphological map; and right: overlay of color-coded vector-based
geomorphological map (50% transparency) on the classical geomorphological map.

for science-based information to be able to objectively pri- In case classical geomorphological maps are not avai-
oritize areas for geoconservation. Selection of potential geo- lable or the content is outdated, geomorphological infor-
conservation sites is mainly based on geomorphological mation can be generated by semi-automated mapping
information. Recently, an evaluation procedure has been techniques.
proposed by Seijmonsbergen et al. (2009) based on layers of
geomorphological information stored in a geodatabase.
Three map fragments of the study area are shown in
14.4.3.2 General Steps for Semi-Automated
Figure 6: the classical geomorphological map (left), the
Geomorphological Mapping
digitized polygon map (center) and an overlay of the Semi-automated geomorphological mapping requires a se-
color-coded vector-based geomorphological map (50% dis- quence of preparatory steps, which are summarized to: (1)
play transparency) on the classical geomorphological deriving LSPs from the LiDAR digital terrain model; (2) gen-
map. The advantage of the latter display is that the original erating segments (or image objects) from the LSPs; (3) de-
symbols and lines, which contain information on the veloping classification rules; (4) classifying image objects; (5)
direction of processes, can still be viewed. The classical selecting and delineation of reference areas; (6) determination
map (Seijmonsbergen, 1992) is based on the legend of classification accuracy; and (7) visualizing the automated
published by de Graaff et al. (1987), whereas the concept map result (see also Figure 7). The individual steps are ex-
of the geodatabase is derived from Gustavsson et al. (2008) plained in more detail below.
(Figure 2). The following six steps are necessary for
converting a classical field-based map to a modern computer- 14.4.3.2.1 Extraction of LSP and segmentation into
based map: objects
In step 1, LSPs are calculated that contain potential value for
1. Scan the paper geomorphological map at high resolution identifying landform boundaries and extracting geomorpho-
(e.g., 300 dpi) in a common file format (e.g., tif) and logical information using Python-scripting language and Arc-
georeference the scanned map to a locally used coordinate GIS. A selection of these LSPs from the case-study area are
system. displayed in Figure 4.
2. Design or use an existing digital geomorphological con- In the second step, eCognition Developer software is
version legend (see Table 3). used to cluster image pixels into objects, based on
3. Digitize the geomorphological map units into a geo-data- heterogeneity criteria of LSP values within objects. First, small
base (see Figure 2). objects are created for extracting small geomorpho-
4. Add GIS-codes to the polygon attribute table (Table 3) and logical features (discussed in step 4, see Section 14.4.3.2.2)
color code the map using a predefined digital legend. and subsequently larger objects extracted for classifying
5. Add selected point and line information (e.g., rivers, faults, larger features. The resulting segments are shown in
springs, and sink holes, not shown in Figure 5). Figure 8. Generally, the automatically segmented objects are
6. Add optional additional attribute columns (see Sec- smaller compared to the manually digitized polygons (Fig-
tion 14.4.3) and link descriptions, sample information or ure 6). These eCognition objects roughly correspond to
digital field photos. landform elements, rather than entire landforms, because the
46 The Modern Geomorphological Map

Table 3 Example of a geomorphological legend for generating digital maps

The legend is used to digitize classical maps into a geodatabase and for semi-automated classification of landforms from LiDAR DEMs. The four additional columns (SR, scientific
relevance; FoO, frequencey of occurrence; V, vulnerability; D, disturbance) contain the scores for use in geoconservation application.

Classification
LiDAR Image Classified Output
accuracy
DTM LSPs objects objects map
2 4 assessment 7
1 6

Classification
rules
3
Expert
knowledge Reference
data points
5

A: Input B: Pre-classification C: Classification D: Post-classification and output

Figure 7 Overview of the steps necessary in the semi-automated geomorphological mapping method.

segmentation algorithm delineates morphologically homo- 14.4.3.2.2 Rule development, image classification, and
geneous areas. In the next steps, the elementary objects are accuracy assessment
combined into larger landforms by means of object The resulting objects represent actual geomorphological fea-
classification. tures, ready for classification. For classification of objects into
 The Modern Geomorphological Map 47

Figure 8 Left: RGB-composite (R ¼ slope, G ¼ openness25, B ¼ openness250), which is used to define exact boundaries of geomorphological
units, which have been used here as training areas for the accuracy assessment. Center: polygons generated after object-based segmentation
process of LSPs calculated from a LiDAR DEM. Right: color-coded semi-automated classified geomorphological map over the classical
geomorphological map. Objects are classified after applying landform specific classification rules.

geomorphological meaningful units, rule sets for fuzzy clas- than the commonly used hillshade map, because it is not
sification are developed in step 3. These rules are sequences of depending on artificial shading, so that every detail is equally
mathematical decisions which are guided by expert know- well displayed. Because multiple-scale openness parameters
ledge, with the objective of identifying statistically homo- are integrated with slope, both the detailed and larger geo-
geneous landforms. For example, in the western Austrian Alps, morphological landform boundaries are accentuated. As a
landforms were separated on the basis of membership func- result, 300 random points were labeled as geomorphological
tions and zonal statistics of the individual objects, for ex- features that served as the reference data set.
ample, alluvial fans (mean slope 7–201, elevation o1200 m) The classification accuracy assessment is evaluated by
and talus slopes (mean slope 30–451) (Van Asselen and Seij- comparing classification results with the reference data set.
monsbergen, 2006). Additional information is required to The results of the classification accuracy assessment are pre-
properly classify fluvial landforms such as a river terrace be- sented as a confusion matrix (Congalton and Green, 1999) in
cause of variations in the slope between its upper surface and Table 4. The table clearly shows that, despite an overall ac-
risers within the terrace or on its edge. Therefore, the upslope curacy of 65%, most classes reach 80% or higher accuracy,
area or the distance to a stream is also used in mapping. This except for denudation landforms (Codes 3111 and 3212). This
means that each landform may require knowledge-based rule means that the overall map accuracy will increase robustly, if
sets (Anders et al., 2010). Application of these rules leads to a the accuracy of classified denudation landforms is improved
final classification of the objects in step 4 (Figure 8). through better segmentation parameters or classification rule
The classification accuracy of the final map (Table 4) was sets.
tested by comparing the classification results with a known
reference data set. Step 5 includes creating such a reference
data set. In the example, the classical geomorphological data
14.4.3.3 Application to Geoconservation of Landscapes
set was used as reference material. However, it appeared that
the landform boundaries on the classical map, especially in The digitized geomorphological map, the semi-automated
steep-sloping terrain and under forest cover, may deviate from geomorphological map, or an updated classical or a combined
exact locations measured in the field with GPS. To overcome geomorphological map forms the basis for geoconservation
this practical problem Anders et al. (2010) suggested using a applications. In this case, a map showing areas of high sig-
RGB composite (Figure 8) in combination with the classical nificance for potential geoconservation is made, based on the
geomorphological map in step 5 to delineate precise bound- digitized classical geomorphological map. The method de-
aries of reference areas. This step guarantees creation of correct signed by Seijmonsbergen et al. (2009) is used. Four criteria
samples for the accuracy assessment. In this RGB composite, which are commonly recognized in literature (Gray, 2004;
the red band is replaced by the slope LSP, and the green and Coratza and Giusti, 2005) are evaluated: scientific relevance,
blue band by the topographic openness LSP, as calculated frequency of occurrence, vulnerability, and disturbance. Using
within a 25- and 250-m search, respectively (Figure 4). The a weighting and ranking system, which was tested in the
RGB composite is a new mapping tool that is more efficient Austrian Alps and in the Peruvian Andes (Seijmonsbergen
48 The Modern Geomorphological Map

Table 4 Confusion matrix showing the number of pixels of the classified data and reference dataset and the producers, user’s and overall
accuracy and KHAT value

Classification data

1111 1221 1222 1223 2111 3111 3112 3211 3212 4211 5111 6111 6213 6214 Total
1111 51 3 0 0 3 0 1 2 3 0 0 1 0 0 64
1221 1 41 0 0 0 0 2 0 3 0 0 0 0 0 47
1222 0 3 9 1 0 0 0 0 0 0 0 0 0 0 13
1223 0 1 0 7 0 0 0 0 0 0 0 0 0 0 8
2111 0 0 0 0 9 0 2 0 3 0 0 0 0 0 14
Reference data

3111 2 2 0 1 1 10 3 1 3 0 0 0 0 0 23
3112 5 2 0 0 5 2 19 5 12 0 0 0 0 0 50
3211 3 0 0 0 1 0 0 13 4 1 0 0 0 0 22
3212 3 3 0 0 0 0 4 4 3 0 0 0 1 0 18
4211 0 0 0 0 0 0 0 0 0 8 0 0 0 0 8
5111 0 0 0 0 0 0 0 0 0 0 5 0 0 0 5
6111 0 1 0 0 0 0 0 0 1 0 0 6 0 0 8
6213 0 1 0 0 0 0 1 0 1 0 0 0 10 1 14
6214 0 0 0 0 0 0 0 0 1 0 0 0 0 5 6
Total 65 57 9 9 19 12 32 25 34 9 5 7 11 6 300

Producers accuracy User’s accuracy Overall accuracy 65.3%

1111 79.7% 1111 78.5% KHAT 0.61

1221 87.2% 1221 71.9%
1222 69.2% 1222 100.0%
1223 87.5% 1223 77.8%
2111 64.3% 2111 47.4%
3111 43.5% 3111 83.3%
3112 38.0% 3112 59.4%
3211 13.6% 3211 52.0%
3212 16.7% 3212 8.8%
4211 100.0% 4211 88.9%
5111 100.0% 5111 100.0%
6111 75.0% 6111 85.7%
6213 71.4% 6213 90.9%
6214 83.3% 6214 83.3%

The codes correspond to the legend units of Table 3.

et al., 2010), the scores of each predefined landform class are areas. In practice, the elevation of a potential geoconservation
assigned in the GIS attribute table (Table 3). Adding of these area to a protected geoconservation area is a decision to be
four scores and subsequent ranking into three categories leads made by the local authorities (Seijmonsbergen et al., 2009).
to a low, medium, or high degree of significance of landforms
for potential geoconservation. The disturbance score is auto-
matically assigned a lower score if polygons are intersected by
roads. The resulting map shows a ranking of potential geo- 14.4.4 Conclusions and Closing Remarks
conservation sites that may serve as science-based information
to support decision making for landscape protection, rural • Modern geomorphological mapping is a consequence of techno-
or urban development, and cultural heritage protection logical advance. Geomorphology has traditionally been a
(Figure 9). discipline with roots in the field of RS and interpretation of
It should be evident that a similar procedure, based on the aerial photography (Walsh et al., 1998), which means it
automatically generated geomorphological map, also leads to is open to new technology, models and data sources. New
a map showing similar patterns of potential geoconservation sensors such as terrestrial laser scanners and hyper-spectral
 The Modern Geomorphological Map 49

N
N

km
High significance Medium significance Low significance
0 0.5 1
Figure 9 Potential geoconservation map of the Tschengla-Dünza area, ranked into three classes: low, medium, and high significance as a
transparent layer over the scanned classical geomorphological map.

sensors will continuously provide new and more detailed between specialized software and databases in which geo-
information of the Earth’s surface, so that global-scale morphological information is stored can be consulted for
issues can be related to regional and local geomorpho- user-specific information via web servers. In this way,
logical problems. In geomorphometry, DEMs are used for digital maps can be assembled and printed on demand as
the automated extraction of specific geomorphological paper or compiled as geoPDF maps.
features, but methods are also developed to classify total • New applications will be developed that use selected digital
landscapes. geomorphological information. Integration of data sources,
• Classical geomorphological maps and modern geomorphological linking of databases in combination with new analyses
information layers coexists in digital environments. Scanned tools opens up a wealth of new applications in which
and georeferenced classical maps serve as base-line infor- modern geomorphological information plays a crucial role.
mation and can be used to train and evaluate modeled The technological advances in geomorphological mapping
results derived from automated approaches to geo- will certainly contribute to solving contemporaneous
morphological mapping. Conversion legends should be questions, for example, to understand the impact of pro-
further developed to digitize the information depicted on jected climate change on landforms and processes, what
classical geomorphological maps. In addition, the classical the consequences are of sea-level change in coastal areas,
digitized geomorphological map can be updated and re- and what the impact is of intensified land-use change on
fined using new data sources such as LiDAR DEMs, which geomorphological processes.
allow precise delineation of local details that were previ- • Geomorphological concepts and expert skills are valuable tools.
ously too laborious or impossible. No doubt that technological advance will continue to offer
• Modern geomorphological information consists of collections of new utilities to derive information on landforms and
digital vector, raster, and attribute data in a geodatabase. In processes from widely available high-resolution spatial,
contrast to classical paper maps, modern geomorphological temporal, and spectral data. It is anticipated that further
information is stored in separate information layers or fea- integration with RS technology will lead to further spe-
tures in a geodatabase. There is no consensus yet about the cializations and applications in the extraction of geo-
design of such a geodatabase. morphological information. Adequate training of
• The end-user creates the digital geomorphological product from geomorphologists in handling new data, the use of new
the geodatabase. The end-users will on demand query and precise observation technologies and field techniques in
display the desired collection of geomorphological and combination with programming skills will further con-
additional information layers in a GIS. Graphical interfaces tribute to developing modern geomorphological mapping
50 The Modern Geomorphological Map

(Murray et al., 2009). This implies that more emphasis will De Brum, Ferreira, 1981. Carte géomorphologique du Portugal. Memórias do
be on development of automated methods and that less Centro de Estudos Geográficos, Lisboa 6, 1–53.
Demek, J., Embleton, C. (Eds.), 1978. Guide to Medium-Scale Geomorphological
time is spent on fieldwork. New generations of geo-
Mapping. IGU Commission on Geomorphological Survey and Mapping. E.
morphologists should be aware that fundamental geo- Schweizerbart’sche Verlagsbuchhandlung, Stuttgart, Germany.
morphological concepts are still valuable and worthwhile Demek, J., Embleton, C., Gellert, J.F., Verstappen, H.T. (Eds.), 1972. Manual of
to integrate in new technology so that they can better ap- Detailed Geomorphological Mapping. International Geographical Union
preciate new findings, which may contribute to further Commission on Geomorphological Survey and Mapping. Academia, Prague.
Dikau, R., 1989. The application of a digital relief model to analysis in
understanding of the functioning of geomorphological geomorphology. In: Raper, J. (Ed.), Three Dimensional Application in GIS. Taylor
systems. & Francis, London, pp. 51–77.
Dikau, R., Brabb, E.E., Mark, R.K., Pike, R.J., 1995. Morphometric landform
analysis of New Mexico. Zeitschrift für Geomorphologie, Supplementband 101,
109–126.
Acknowledgments Dorren, L.K.A., Maier, B., Seijmonsbergen, A.C., 2003. Improved Landsat-based
forest mapping in steep mountainous terrain usding object-based classification.
Forest Ecology and Management 183, 31–46.
The author thanks Niels Anders whose contribution to the
Drăgut¸, L., Blaschke, T., 2006. Automated classification of landform elements using
semi-automated mapping of the case study is much appreci- object-based image analysis. Geomorphology 81, 330–344.
ated. The case presented in Section 14.4.3 was carried out in Elvhage, C., 1980. An experimental series of topogeomorphological maps.
the context of the Virtual Laboratory for e-Science project Geografiska Annaler. Series A. Physical Geography 62, 105–111.
supported by a BSIK grant from the Dutch Ministry of Edu- Etzelmuller, B., Romstad, B., Fjellanger, J., 2007. Automated regional classification
of topography in Norway. Norwegian Journal of Geology 87, 167–180.
cation, Culture and Science (OC&W) and is part of the ICT Evans, I.S., 1990. Cartographic techniques in geomorphology. In: Goudie, A.,
innovation programme of the Ministry of Economic Affairs Anderson, M., Burt, T., Lewin, J., Richards, K., Whalley, B., Worsley, P. (Eds.),
(EZ). The ‘‘Land Vorarlberg’’ (www.Vorarlberg.at) in Austria is Geomorphological Techniques, Second ed. Routledge, London, pp. 97–108.
thanked for the use of the LiDAR data. Evans, I.S., Hengl, T., Gorsevski, P., 2008. Applications in Geomorphology.
In: Hengl, T., Reuter, H.I. (Eds.), Geomorphometry: Concepts, Software,
Applications. Developments in Soil Science. Elsevier, Amsterdam, Volume 33,
pp. 497–525.
References Geilhausen, M., Otto, J.-C., Dikau, R., 2007. GMK Digital – Geomorphologische
Karten im Netz. http://gidimap.giub.uni-bonn.de/gmk.digital/home.htm (accessed
April 2011).
Adediran, A.O., Parcharidid, I., Poscolieri, M., Pavlopoulos, K., 2004. Computer- Gilewska, S., Klimek, M., 1968. Project of the Unified key to the detailed
assisted discrimination of morphological units on north-central Grete (Greece) Geomorphological map of the World. Folia Geographica, Ser. Geographica-
by applying multi-variate statistics to local relief gradients. Geomorphology 58, Physica, Volume II. Polska Akademia Nauk, Kraków.
357–370. Goudie, A.S. (Ed.), 2006. Encyclopedia of Geomorphology. International Association
Anders, N.S., Seijmonsbergen, A.C., Bouten, W., 2010. Stratified object-based of Geomorphology, Routledge, London, 1155 pp.
image analysis of high-res laser altimetry data for semi-automatic de Graaff, L.W.S., De jong, M.G.G., Rupke, J., Verhofstad, J., 1987. A
geomorphological mapping in an alpine area. Geophysical Research Abstracts geomorphological mapping system at scale 1:10.000 for mountainous areas
12, EGU2010-9959-1 (Vienna). (Austria). Zeitschrift für Geomorphologie 31, 229–242.
Barsch, D., Liedtke, H., 1980. Principles, scientific value and practical applicability Gray, M., 2004. Geodiversity: Valuing and Conserving Abiotic Nature. Wiley,
of the geomorphological map of the Federal Republic of Germany at the scale of Chichester, 434 pp.
1:25 000 (GMK 25) and 1:100 000 (GMK 100). Zeitschrift für Geomorphologie, Gustavsson, M., Kolstrup, E., Seijmonsbergen, A.C., 2006. A new symbol-and-GIS
N.F., Supplement Band 36, 296–313. based detailed geomorphological mapping system: renewal of a scientific
Bolstad, P., 2008. GIS Fundamentals – A First Text on Geographical Information discipline for understanding landscape development. Geomorphology 77,
Systems, Third ed. Eider Press, White Bear Lake, Minnesota. 90–111.
Booth, A.M., Roering, J.J., Perron, J.T., 2009. Automated landslide mapping using Gustavsson, M., Seijmonsbergen, A.C., Kolstrup, E., 2008. Structure and contents
spectral 844 analysis and high resolution topographic data: Pudget Sound of a new geomorphological GIS database linked to a geomorphological map –
lowlands, Washington, and Portland Hills, Oregon. Geomorphology 109, with an example from Liden, central Sweden. Geomorphology 95, 335–349.
132–147. Hansen, A., 1984. Landslide hazard analysis. In: Brunsden, D., Prior, D.B. (Eds.),
Brunsden, D., Doornkamp, J.C., Fookes, P.G., Jones, D.K.C., Kelly, J.M.H., 1975. Slope Instability. Wiley, New York, NY, pp. 523–602.
Large-scale geomorphological mapping and highway engineering. Q. JL. Hengl, T., Reuter, H.I. (Eds.), 2008. Geomorphometry: Concepts, Software,
Engineering Geology 8, 227–253. Applications. Developments in Soil Science. Elsevier, Volume 33, 525 pp.
Bue, B.D., Stepinski, T.F., 2006. Automated classification of landforms on Mars. Iwahashi, J., Pike, R.J., 2007. Automated classifications of topography from DEMs
Computers and Geosciences 32, 604–614. by an unsupervised nested-means algorithm and a three-part geometric
Burrough, P.A., van Gaans, P.F.M., MacMillan, R.A., 2000. High resolution signature. Geomorphology 86, 409–440.
landform classification using fuzzy k-means. Fuzzy Sets and Systems 113, Kienholz, H., 1977. Kombinierte Geomorphologische Gefahrenkarte 1:10000 von
37–52. Grindelwald. Catena 3, 265–294.
CNR (Gruppo Nazionale Geografia e Geomorphologia), 1986. Ricerche Klimaszewski, M., 1963. Landforms list and signs used in the detailed
Geomofologia nall’ alta Val di Peio (Gruppo del Cevedale). Geografia Fisica e geomorphological map. Problems of geomorphological mapping. Geographical
Dinamica Quaternaria 9, 137–191. Studies 46, 139–177 (Warszawa).
Congalton, R.G., Green, K., 1999. Assessing the Accuracy of Remotely Sensed Data: Klimaszewski, M., 1982. Detailed geomorphological maps. ITC Journal 3, 265–271.
Principles and Practices. Lewis Publishers, Boca Raton, 137 pp. Klimaszewski, M., 1990. Thirty years of geomorphological mapping. Geographia
Cooke, R.U., Doornkamp, J.C., 1990. Geomorphology in Environmental Polonica 58, 11–18.
Management: A New Introduction. Clarendon Press, Oxford. Koomen, A.J.M, Maas, G.J., 2004. Geomorfologische Kaart Nederland (GKN).
Coratza, P., Giusti, C., 2005. Methodological proposal for the assessment of the Alterra-rapport 1039, Wageningen, 38 pp.
scientific quality of geomorphosites 18. Il Quaternario 18, 305–311. Lee, E.M., 2001. Geomorphological mapping. In: Griffiths, J.S. (Ed.), Land Surface
Cruden, D.M., Varnes, D.J., 1996. Landslide types and processes. In: Turner, A.K., Evaluation for engineering practice. Geological Society Engineering Geology
Schuster, R.L. (Eds.), Landslides, Investigation and Mitigation. Transportation Special Publication No. 18. The Geological Society of London, London, pp.
Research Board, Special report 247, pp. 36–75. 53–56.
Dalrymple, J.B., Blong, R.J., Conacher, A.J., 1968. A hypothetical nine unit land Maarleveld, G.C., Ten Cate, J.A.M., de Lange, G.W., 1974. Die Geomorphologische
surface model. Zeitschrift für Geomorphologie 12, 60–76. Karte der Niederlande. Zeitschrift für Geomorphologie 18, 484–494.
 The Modern Geomorphological Map 51

Mantovani, F., Gracia, F.J., de Cosmo, P.D., Suma, A., 2010. A new approach to (Austria), supported by Geographical Information Systems. Memorie Descrittive
landslide geomorphological mapping using the Open Source Software in the della Carta Geologica d’Italia LXXXVI, 163–172.
Olvera area (Cadiz, Spain). Landslides 7, 69–74. Seijmonsbergen, A.C., Hengl, T., Anders, N.A., 2011. Semi-automated identification
Martin-Serrano, A., 2005. Mapa Geomorphológico de España y del margen and extraction of geomorphological features using digital elevation data.
continental a escala 1:1.000.000. Instituto Geológico y minero de España, In: Smith, M.J., Paron, P., Griffiths, J. (Eds.), Geomorphological Mapping: A
Madrid. Professional Handbook of Techniques and Applications. Elsevier, London.
Meijerink, A.M.J., 1988. Data acquisition and data capture through terrain mapping Seijmonsbergen, A.C., Sevink, J., Cammeraat, L.H., Recharte, J., 2010. A potential
units. ITC Journal 1, 23–44. geoconservation map of the Las Lagunas area, northern Peru. Environmental
Milne, G., 1935. Some suggested units of classification and mapping particularly Conservation 37(2), 107–115.
for East African Soils. Soil Research 4, 183–198. Speight, J.G., 1990. Landform. In: McDonald, R.C., Isbell, R.F., Speight, J.G.,
Minár, J., Evans, I.S., 2008. Elementary forms for land surface segmentation: the Walker, J., Hopkins, M.S. (Eds.), Australian Soil and Land Survey Field
theoretical basis of terrain analysis and geomorphological mapping. Handbook, Second Edition. Inkarta Press, Melbourne, pp. 9–57.
Geomorphology 95, 236–259. St-Onge, D.A., 1964. Geomorphological map legends, their problems and their
Murray, A.B., Lazarus, E., Ashton, A., et al., 2009. Geomorphology, complexity, and value in optimum land utilization. Geographical Bulletin 22, 5–12.
the emerging science of the Earth’s surface. Geomorphology 103, 496–505. St-Onge, D.A., 1981. Theories, paradigms, mapping and geomorphology. Canadian
Nass, A., van Gasselt, S., Jaumann, R., Asche, H., 2011. Implementation of Geographer 4, 307–315.
cartographic symbols for planetary mapping in geographical information Ten Cate, J.A.M., 1983. Detailed systematic geomorphological mapping in the
systems. Planetary and Space Science 59(11–12), 1255–1264, http://dx.doi.org/ Netherlands and its applications. Geologie en Mijnbouw 62, 611–620.
10.1016/j.pss.2010.08.022 (accessed April 2011). Thoma, D.P., Gupta, S.C., Bauer, M.E., Kirchhoff, C.E., 2005. Airborne laser
Newell, W., Clark, I., 2008. Geomorphic map of Worcester County, Maryland, scanning for riverbank erosion assessment. Remote Sensing of Environment 95,
interpreted from a LiDAR-based, digital elevation model. Open-File Report 2008- 493–501.
1005. U.S. Geological Survey, 34 pp. http://pubs.usgs.gov/of/2008/1005/ Tricart, J., 1965. Principes et Methods de la Géomorphologie. Masson & Cie, Paris,
(accessed April 2011). 495 pp.
Otto, J.-C., Dikau, R., 2004. Geomorphic system analysis of a high mountain valley Van Asselen, S., Seijmonsbergen, A.C., 2006. Expert-driven semi-automated
in the Swiss Alps. Zeitschrift für Geomorphologie 48, 323–341.
geomorphological mapping using a laser DTM. Geomorphology 78, 309–320.
Otto, J.-C., Geilhausen, M., Gustavsson, M., 2011. Cartography: design,
Van Westen, C.J., Seijmonsbergen, A.C., Mantovani, F., 1999. Comparing landslide
symbolization and visualization of geomorphological maps. In: Smith, M.J.,
hazard maps. Natural Hazards 20, 137–158.
Paron, P., Griffiths, J. (Eds.), Geomorphological Mapping: A Professional
Van Westen, C.J., Soeters, R., Sijmons, K., 2000. Digital geomorphological
Handbook of Techniques and Applications. Elsevier, London.
landslide hazard mapping of the Alpago area, Italy. International Journal of
Passarge, S., 1914. Morphologischer Atlas. Erläuterungen zu Lief. 1, Morphologie
Applied Earth Observation and Geoinformation 2(1), 51–60.
des Messtischblattes Stadtremba. Mitt. Geogr. Gesell. Hamburg 28.
Varnes, D.J., 1978. Slope movement types and processes. In: Schuster, R.L., Krizek,
Pécsi, M., 1964. Geomorphological mapping in Hungary in the service of theory
R.J. (Eds.), Landslides: Analysis and Control, 234 pp. Transportation Research
and practice. In: Pécsi, M. (Ed.), Applied Geography in Hungary. Akad. Kiadó,
Budapest, 211 pp. Board, Special Report 176, Washington, DC, pp. 11–33.
Verstappen, H., 2011. Old and new trends in geomorphological and landform
Pike, R.J., Evans, I.S., Hengl, T., 2008. Geomorphometry: a brief guide. In: Hengl,
T., Reuter, H.I. (Eds.), Geomorphometry: Concepts, Software, Applications. mapping. In: Smith, M.J., Paron, P., Griffiths, J. (Eds.), Geomorphological
Developments in Soil Science. Elsevier, Volume 33, pp. 3–30. Mapping: A Professional Handbook of Techniques and Applications. Elsevier,
Prima, O.D.A., Echigo, A, Yokoyama, R., Yoshida, T., 2006. Supervised landform London.
classification of Northeast Honshu from DEM-derived thematic maps. Verstappen, H.T., 1970. Introduction to the ITC System of Geomorphological
Geomorphology 78, 373–386. Survey. K.N.A.G. Geografisch Tijdschrift IV, 87–91.
Remondo, J., Oguchi, T., 2009. GIS and SDA applications in geomorphology. Voženı́lek, V., 2000. Spatial database for geomorphological mapping by GPS
Geomorphology 111, 1–3. techniques. Geographica 36, 97–105.
Reynard, E., Holzmann, C., Lambiel, C., Phillips, M., 2005. Légende Walsh, S., Butler, D., Malanson, G., 1998. An overview of scale, pattern, process
geomorphologique de l’IGUL et Guide pratique pour le levé de cartes relationships in geomorphology: a remote sensing and GIS perspective.
géomorphologiques. Lausanne, Institut de Géographie, 33 pp. Geomorphology 21(3–4), 183–204.
Rivard, L.A., 2009. Geo-Hazard Associated Geounits – Atlas and Glossary. Yokoyama, R., Shirasawa, M., Pike, R.J., 2002. Visualizing topography by
Environmental Science and Engineering Series. Springer, Berlin, Heidelberg, openness: a new application of image processing to digital elevation models.
1054 pp. Photogrammetric Engineering and Remote Sensing 68, 257–265.
Rupke, J., Cammeraat, E., Seijmonsbergen, A.C., Van Westen, C.J., 1988.
Engineering geomorphology of the Widentobel catchment, Appenzell and Sankt
Gallen, Switzerland. A geomorphological inventory system applied to
geotechnical appraisal of slope stability. Engineering Geology 26, 33–68.
Relevant Websites
Salomé, A.I., van Dorsser, H.J., Rieff, Ph.L., 1982. A comparison of
geomorphological mapping systems. ITC Journal 3, 272–274. http://www.unil.ch
Seijmonsbergen, A.C., 1992. Geomorphological evolution of an alpine area and its Faculté des geosciences et de 1’environment.
application to geotechnical and natural hazard appraisal in the NW. Rätikon http://www.geomorphometry.org
Mountains and S. Walgau (Vorarlberg, Austria). PhD thesis, Universiteit van Geomorphometry.org.
Amsterdam, 109 pp. http://www.journalofmaps.com
Seijmonsbergen, A.C., de Graaff, L.W.S., 2006. Geomorphological mapping and Journal of Maps.
geophysical profiling for the evaluation of natural hazards in an alpine http://www.geomorphology.at
catchment. Natural Hazards and Earth System Science 6, 185–193. Research Group ‘Geomorphology and Environment Systems’.
Seijmonsbergen, A.C., De Jong, M.G.G., de Graaff, L.W.S., 2009. A method for the http://www.dpi.vic.gov.au
identification and assessment of significance of geomorphosites in Vorarlberg Victoria, Department of Primary Industries.
52 The Modern Geomorphological Map

Biographical Sketch

Harry AC Seijmonsbergen is currently appointed as assistant professor in the Computational Geo-Ecology re-
search group at the Institute for Biodiversity and Ecosystem Dynamics at the University of Amsterdam in the
Netherlands. He received his PhD in environmental sciences in 1992 from the University of Amsterdam, the topic
being ‘Geomorphological Evolution of an Alpine Area and its Application to Geotechnical and Natural Hazard
Appraisal’. Since then, he has been working on geomorphological mapping, geoconservation, and land-use
change projects in which field studies, geographical information systems, remote sensing and digital elevation
information are combined for the development of semi-automated mapping techniques. His teaching activities
comprise field studies in Austria, Peru, Spain, Luxembourg, and the Netherlands, GIS and remote-sensing courses
and landscape analysis practical using both classic stereoscopic air-photo interpretation and 3D ‘Virtual Globe’
applications.
 14.5 Google EarthTM in Geomorphology: Re-Enchanting, Revolutionizing, or
Just another Resource?
S Tooth, Aberystwyth University, Aberystwyth, UK
r 2013 Elsevier Inc. All rights reserved.

14.5.1 Introduction 53
14.5.2 Recent Feature Developments to Google EarthTM 54
14.5.3 Use of Google EarthTM in Geomorphology 55
14.5.3.1 Reach-Scale Fluvial Geomorphology 55
14.5.3.2 Basin-Scale Fluvial Geomorphology 55
14.5.3.3 Planetary Geomorphology 55
14.5.3.4 Geomorphological Education and Outreach 56
14.5.4 Discussion 56
14.5.4.1 Advantages of Google EarthTM in Geomorphology 56
14.5.4.2 Limitations of Google EarthTM in Geomorphology 59
14.5.5 Possible Future Developments in the Use of Google EarthTM in Geomorphology 60
14.5.6 Conclusions 62
References 63

Glossary Google MoonTM A virtual globe that combines

TM topographic representations of the Moon’s surface with
Google Earth A virtual globe that combines topographic
representations of the Earth’s surface with satellite imagery, satellite imagery.
aerial photography, and other information. Google OceanTM A layer within Google EarthTM that
Google MarsTM A virtual globe that combines allows users to zoom below the surface of the ocean and
topographic representations of the Martian surface with view the bathymetry.
imagery obtained from various orbiters and landers.

Abstract

The recent launch of online ’virtual globes’ has vastly increased access to digital imagery of the surfaces of the Earth and other
planetary bodies, enabling their examination from various scales and perspectives. This chapter focuses on Google EarthTM,
one of the most popular virtual globes, and considers whether its growing use in geomorphological research, education, and
outreach provides evidence for disciplinary re-enchantment or revolution, or whether the application is just another resource.
Published case studies and anecdotal accounts highlight many advantages of the use of Google EarthTM in geomorphology
(e.g., generating datasets that provide a firmer basis for hypothesizing about surface processes and landforms), but also some
limitations (e.g., variable image quality, and limited mapping and analytical functions). Although Google EarthTM can provide
an invaluable starting point for geomorphological analyses, its use typically needs to be combined with other datasets and
analyses, including field measurements, laboratory experimentation, more sophisticated geographic information system (GIS)
analyses, geochronology, and computational modeling. Nonetheless, future developments in the use of Google EarthTM in
geomorphology (e.g., development of shared community datasets) will probably continue to increase its appeal and potential.
On balance, Google EarthTM may be providing opportunities for re-enchantment of geomorphology, but when compared to
other past and present developments in data acquisition and analysis, its use is probably not revolutionizing the discipline.
Google EarthTM is just another resource in the geomorphological ’toolkit’ but nonetheless is a remarkable one, having rapidly
become an invaluable part of many research, education, and outreach projects.

14.5.1 Introduction mapmakers and photographers of previous generations.

Iconic images, such as the views from space that reveal the
Over the past few decades, the surfaces of the Earth and many Earth to be a solitary blue sphere shrouded in clouds, have
other planetary bodies in the solar system have been imaged at helped to fundamentally alter our perceptions about our
scales and with perspectives that would have astonished planet and its place in the solar system. Regional or national
aerial photographic surveys with repeat intervals usually dec-
ades apart have been supplemented by regular imaging of
virtually the entire globe at intervals more commonly days to
Tooth, S., 2013. Google EarthTM in geomorphology: re-enchanting,
revolutionizing, or just another resource? In: Shroder, J. (Editor in Chief),
months apart, and in many electromagnetic wavelengths other
Switzer, A.D., Kennedy, D.M. (Eds.), Treatise on Geomorphology. Academic than just the visible. This has led to the recognition of regional
Press, San Diego, CA, vol. 14, Methods in Geomorphology, pp. 53–64. and global Earth phenomena as diverse as ozone depletion,

Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00372-9 53

54 Google EarthTM in Geomorphology: Re-Enchanting, Revolutionizing, or Just another Resource?

ocean phytoplankton blooms, and tropical deforestation these opportunities have been taken up by numerous online,
(National Research Council, 2008). For geomorphology – the interactive communities of virtual globe enthusiasts (e.g.,
science that investigates the spatial distribution and inter- Google EarthTM Blog). Human geographers have commented
relationships of landforms at different scales, and attempts to on the role of Google EarthTM as an ‘alluring digital peep-box’
provide explanatory accounts of landform and landscape de- that provokes an array of frivolous and more serious activities
velopment through time – this unprecedented era of planetary from an online user community reveling in the ability to
surface imaging has been a godsend, and has contributed zoom, tilt, and rotate Earth imagery (e.g., Kingsbury and
greatly to some fundamental empirical and conceptual devel- Jones, 2009).
opments. On Earth, for example, early Landsat images pro- Tooth (2006) suggested that this launch of virtual globes
vided the synoptic view essential for the mapping and analysis might provide an opportunity for the ‘re-enchantment’ of
of global sand seas (e.g., Breed et al., 1979), whereas repeat geomorphology (sensu Baker and Twidale, 1991); for instance,
satellite images of features such as ice sheets and glaciers, by stimulating interest and excitement in landscape form,
rivers, and coastlines have been instrumental in developing process, and change, and provoking the all-important reflect-
our understanding of spatial patterns and rates of landform ive questions such as ‘why?,’ ‘how?,’ and ‘when?.’ In various
and landscape response to recent environmental changes applied contexts (e.g., natural disaster relief efforts, environ-
(Gregory, 2010). Impediments to the wider use of such data- mental protests, and conservationism), virtual globes have
sets in geomorphology, however, have been the difficulty and been claimed to have ‘revolutionized’ ways of thinking and
expense involved in identifying and acquiring the relevant acting (Olsen, 2007), and similar ideas have been mooted
imagery, particularly when faced with the daunting task of regarding the potential revolutionizing effect of virtual globes
grappling with multiple images required to cover large areas. on cognate academic disciplines such as geoarchaeology (e.g.,
To some extent, these impediments have been circumvented see Handwerk, 2006; AHRC ICT Methods Network, no date).
with the recent launch of ‘virtual globes’ (alternatively termed Some academics have been less convinced of such suggestions
‘geobrowsers’) that have vastly increased the online availability and claims, perhaps viewing virtual globes as resources useful
of digital imagery of the Earth’s B510 million km2 surface. As for accessing and viewing imagery of the Earth, but with little
outlined by Tooth (2006), virtual globes such as Google EarthTM, potential for initiating major changes in academic practices or
NASA World Wind, and SkylineGlobe combine topographic reasoning.
representations of the Earth’s surface with satellite imagery or The aim of this chapter is to critically examine these con-
aerial photographs. Using simple controls on user-friendly tentions in the light of recent and possible future develop-
interfaces, one can zoom from any altitude to an identified ments in the use of virtual globes in geomorphology. The
natural or urban area, and using a combination of zoom, pan, chapter focuses on Google EarthTM, which represents one of
and tilt, one can examine the terrain at various scales from the most popular virtual globes, but many points could apply
vertical, oblique, or even three-dimensional (3D) perspectives. to other similar applications. Key questions include:
Images are fully georeferenced, and commonly there are options
1. how widely used is Google EarthTM in geomorphological
to display relevant coordinates, spot elevations, place names,
research and teaching?
ground-level photographs, and other associated information. In
2. what are the advantages and limitations of Google EarthTM
some instances, other georeferenced data (e.g., topographic
for geomorphological analyses?
maps) are also available as layers within the virtual globes. These
3. what is the possible future for Google EarthTM in
facilities have long been available in expensive, more sophisti-
geomorphology?
cated geographic information system (GIS) software packages,
4. is the use of Google EarthTM in geomorphology actually re-
but the basic versions of virtual globes are available free on the
enchanting or revolutionizing the discipline, or is it just
internet, can be downloaded and installed in a few minutes, and
another resource?
will run smoothly on most modern computers with a fast
internet connection. Thus, with minimal technical ability and a
few mouse clicks, virtual globes can bring the Earth’s varied
geomorphology into one’s office or home, and associated ani-
14.5.2 Recent Feature Developments to Google
mations and overlays can be used to visualize landscape-scale
EarthTM
events such as hurricanes, wildfires, floods, and dust storms.
Virtual globes for the Moon and Mars are also available, thus
On its first release, Google EarthTM had one layer of relatively
revealing the geomorphology of other planetary bodies.
recent (past 3 years) satellite imagery stitched together from a
Since their release, virtual globes have captured the im-
variety of commercial sources. Image resolution was variable,
agination of scientists and laypersons alike. In its first week on
with higher-resolution imagery available for some cities and
the web (August 2004), 100 000 users downloaded World
areas of North America and Western Europe. Later versions of
Wind (NASA, 2004), and the subsequent release of Google
Google EarthTM have incorporated a host of developments to
EarthTM (June 2005) triggered a more than 10-fold increase in
its functionality and content, including:
media coverage on virtual globes between 2005 and 2006
(Scharl, 2007) with reports of 100 million installations of 1. some higher resolution and updated imagery (most land
this application in the first year alone (Geography 2.0: Virtual areas are now covered in 415 m per pixel resolution sat-
Globes, no date). Many virtual globes are open-source ellite imagery or by small-scale aerial photography);
programs or enable users to contribute their own data or 2. better 3D representation of natural terrain, buildings, and
modifications within certain predefined parameters, and structures;
 Google EarthTM in Geomorphology: Re-Enchanting, Revolutionizing, or Just another Resource? 55

3. 3601 panoramic ‘street-level’ views in many locations channel width, channel sinuosity, meander length, and oxbow
(‘Google Street View’ feature); length, with a total of 911 lakes being measured among the 30
4. ocean floor 3D bathymetry, imagery, and text information reaches. The assembled global dataset showed that the meander
(‘Google OceanTM’ layer); cutoff process produces a characteristic and predictable size-
5. the ability to experience virtual time travel by comparing frequency distribution of oxbow lakes, and this allows for
images of an area taken on different dates (‘Historical prediction of the temporal rate of cutoff development under a
Imagery’ feature). condition of steady, time-averaged channel sinuosity. These
findings provide a contribution to the improved theoretical
Virtual globes of Mars and the Moon (Google MarsTM and
understanding of the full range of meandering channel be-
Google MoonTM) are now available as modules within Google
havior and demonstrate an additional method of quantifying
EarthTM and also combine topographic representations with
floodplain development (Constantine and Dunne, 2008).
high-resolution imagery and a wide variety of ancillary infor-
mation (e.g., ground-level panoramic images from various
Mars landers, and text information). A Google EarthTM Pro
14.5.3.2 Basin-Scale Fluvial Geomorphology
version offering additional functionality over the basic, free-
to-download version is still available for commercial use. Aggradational depositional systems in subaerial and sub-
aqueous settings (e.g., alluvial fans, deltas, and deep water
fans) are characterized by radial dispersal of sediment from an
14.5.3 Use of Google EarthTM in Geomorphology apex. In subaerial settings, however, the planform character-
istics and sedimentary facies of such aggradational systems
Following their launch, researchers in many of the physical, generally have not been as well documented as adjacent axial
biological, and social sciences have rapidly seized on the po- rivers and tributary rivers located in degradational settings. To
tential of virtual globes to display a variety of spatial data in rectify this discrepancy, Weissmann et al. (2010) systematically
order to enhance basic scientific and applied work (e.g., Butler, scanned the continents using Google EarthTM to identify 724
2006; Nourbakhsh et al., 2006; Gramling, 2007; Conroy et al., modern continental sedimentary basins that are both endoreic
2008; Lefer et al., 2008; Tuttle et al., 2008; Monkkonen, 2008; (internally draining) and exoreic (externally draining) and are
Kamadjeu, 2009). In geomorphology, discussions with col- located in varied climatic and tectonic settings. This provided
leagues, attendance at conferences, and perusal of the recently the basis for assessment of fluvial planform in these basins
published literature reveals varied instances where Google using Landsat and Moderate Resolution Imaging Spectro-
EarthTM has been used to access images of landforms or radiometer (MODIS) imagery and Shuttle Radar Topography
landscapes, display spatial data, or undertake more detailed Mission (SRTM) data. In all cases, Weissmann et al. (2010)
analyses. Google EarthTM is a very young and fast-developing argued that the depositional pattern is dominated by a dis-
application, and the aim of this chapter is simply to provide a tributive fluvial system (DFS), a term used to describe a pat-
snapshot of a moving target by outlining recent examples of tern of channel and floodplain deposits that radiate outward
the use of Google EarthTM in geomorphology, and particularly from an apex located where the river enters the sedimentary
with a view to highlighting its advantages and limitations. The basin. In addition, and in further contrast to many rivers in
emphasis is primarily on the use of Google EarthTM in geo- degradational settings, rivers associated with a DFS typically
morphological research, particularly where the application has decrease in size downstream, exhibit a downstream sediment
formed a fundamental part of the investigative methods, but size decrease, and lack lateral confinement. These character-
the rapidly growing use of Google EarthTM in geomorpho- istics may lead to significant differences in sedimentary pat-
logical education and outreach also merits recognition. terns compared to those patterns in degradational settings, but
Weissmann et al. (2010) noted that this will need to be
clarified by further work.
14.5.3.1 Reach-Scale Fluvial Geomorphology
Growth of meander bends and eventual bend cutoff are key
14.5.3.3 Planetary Geomorphology
processes in the formation of oxbow lakes and the construction
of meandering river floodplains, but few quantitative data are In planetary geomorphology, the use of Earth-based analogs to
available to describe and explain how the meandering process explain surface processes and landform development on other
populates the floodplain with oxbow lakes. To identify the planetary bodies is a well-established method. Planetary geo-
controls on the production of oxbow lakes by meander cutoffs, morphologists thus strive to identify Earth settings with
Constantine and Dunne (2008) used satellite images archived (paleo)climates, tectonics, lithologies, surface processes,
by Google EarthTM, and measured channel and oxbow lake landforms, and landform associations that are known or hy-
characteristics along 30 large meandering rivers. Previous pothesized to be similar to those occurring on other planets
studies of meander cutoff development have tended to focus on (Chapman, 2007; Tooth, 2009a). In the context of the
a restricted set of conditions along one or a few rivers, and so Argentinean Patagonia, Pacifici’s (2009) study exemplified this
the studied rivers were chosen from lowland floodplain en- approach. The Cenozoic geomorphological history of this part
vironments in various geographical regions worldwide. Selected of Patagonia has been dominated by processes similar to those
reaches were chosen from wide valleys, so that meandering that dominated Martian history (e.g., basaltic volcanism, gla-
processes were not affected by interactions with the valley cial, periglacial, fluvial, and aeolian processes). By comparing
margins. For each reach, measurements included bankfull Landsat and QuickBird images of the Earth obtained by
56 Google EarthTM in Geomorphology: Re-Enchanting, Revolutionizing, or Just another Resource?

Google EarthTM with similar resolution, visible wavelength 14.5.4 Discussion

images of Mars taken by the Mars Orbital Camera (MOC),
Thermal Emission Imaging System (THEMIS), and High From the foregoing discussion, several advantages and limi-
Resolution Stereo Camera (HRSC) instruments, possible geo- tations of the use of Google EarthTM in geomorphology are
morphological analogs of the Martian surface were identified apparent, as summarized in Table 1.
from this remote, poorly accessible region. Such features include
volcanic landforms (e.g., lava flows), proglacial and periglacial
landforms (e.g., debris aprons, solifluction deposits, and rock 14.5.4.1 Advantages of Google EarthTM in Geomorphology
glacier-like forms), small circular depressions, aeolian streaks,
gullies, fan deltas, and subaqueously formed dunefields. This One of the most obvious advantages of Google EarthTM is the
may provide opportunities to study in detail, on Earth, the ease, cost effectiveness, and time saving involved in accessing
geomorphological processes that may have contributed to aerial photographs and satellite imagery. Google EarthTM
shaping of the Martian surface (Pacifici, 2009). In the more provides free digital imagery of nearly the entire land surface
specific case of barchan dunes, Bourke and Goudie (2009) and of the Earth and a few other planetary bodies within a simple,
Bourke (2010) adopted a similar approach. By comparing spatially searchable, and interactive interface. Images are dis-
Google EarthTM imagery from the Namib and other Earth des- played at various scales and perspectives within a coordinate
erts with images from the various Mars missions, these authors system (latitude and longitude) and with spot height infor-
were able to describe and classify different varieties of barchan mation, and it is possible to search for features by entering
dunes and dune asymmetries. Potentially, these sorts of studies place coordinates or place names. Without Google EarthTM or
can provide the basis for greater insight into the relationship other virtual globes, aerial photographs or satellite image
between dune shape and past and present environmental con- datasets would have to be assembled piecemeal within an
ditions (e.g., formative wind and sediment regimes) on the appropriate coordinate system, but suitable images are
surface of the Earth and other planetary bodies, including Mars sometimes difficult to source, generally expensive, have fixed
and Titan (Bourke and Goudie, 2009; Bourke, 2010). resolution and perspective, and are physically unwieldy if large
areas are being considered. In particular, by explicitly focusing
attention on issues of landform patterns at different scales,
and by enabling local observation or measurements to be
14.5.3.4 Geomorphological Education and Outreach
placed in broader spatial context, Google EarthTM helps to
In geomorphological (physical geography) education, there formulate and test hypotheses about landforms and processes,
are many references to the use of Google EarthTM (e.g., Pat- which can be beneficial to various activities, including desktop
terson, 2007; Allen, 2007; Dunagan, 2007; Tuttle et al., 2008; research studies, fieldwork planning, manuscript reviewing,
Thorndycraft, et al., 2009; Sinton and Schultz, 2009). Google and teaching.
EarthTM has been used in various ways in school and higher As an example, before the release of Google EarthTM, Leier
education contexts, including: et al. (2005) examined 202 rivers worldwide, combining in-
formation from satellite imagery, monthly discharge and
1. to provide a library of geomorphological imagery suitable
precipitation records, digital elevation models (DEMs), and
for enhancing student essays and presentations;
other sources to look for the conditions that promote the
2. to design geomorphological ‘virtual tours’ capable of sup-
formation of so-called ‘fluvial megafans’ (uncommonly large
porting lectures, practicals, and residential field courses;
fan-shaped bodies of sediment). This represented a monu-
3. to improve the use of spatial thinking and the application
mental task, but within the next five years, Weissmann et al.
of simple quantitative methods (e.g., plotting, measure-
(2010) were able to use Google EarthTM to undertake an even
ments, and manipulation of geomorphological equations).
more comprehensive global survey of 724 continental sedi-
Some textbooks now include online Google EarthTM mentary basins (see Section 14.5.3.2 above). In the process,
worksheets as part of the independent learning exercises (e.g., Weissmann et al. (2010) identified many DFS across varied
Abbot, 2008), whereas various professional societies and or- climatic and tectonic settings, many of which also match Leier
ganizations such as the Royal Geographical Society (with the et al.’s (2005) definition of ‘fluvial megafan.’ As such, Weiss-
Institute of British Geographers), the Geographical Associ- mann et al.’s (2010) database has been used to challenge Leier
ation, and the Association of American Geographers run et al.’s (2005) primary conclusion that modern fluvial mega-
Google EarthTM training days or workshops for instructors. In fans in actively aggrading basins are produced by rivers subject
outreach contexts, Google EarthTM is also being used in vari- to moderate to extreme seasonal discharge fluctuations that
ous ways, with one example being the United States National result from highly seasonal (monsoonal) precipitation pat-
Snow and Ice Data Center (NSIDC) project to enhance the terns (see Hartley et al., 2010). Weissmann and colleagues had
visibility of, and interest in, cryospheric information (Ballagh long suspected that DFS dominate the depositional form in
et al., 2007). Information from key components of the cryo- modern sedimentary basins and that, by inference, the de-
sphere (glacier photographs, permafrost type and extent, posits of DFS may form the bulk of the alluvial sedimentary
sea ice concentration and extent, and snow extent) were fill in ancient basins but Google EarthTM was the technology
converted into a projection and image format compatible with that finally enabled them to start testing these hypotheses
Google EarthTM in order to enhance data availability, under- (G. Weissmann, personal communication).
standing and usability among a diverse, typically nonspecialist Weissmann et al.’s (2010) paper, with a list of authors from
audience. three institutions on two continents, demonstrates another key
 Google EarthTM in Geomorphology: Re-Enchanting, Revolutionizing, or Just another Resource? 57

Table 1 Summary of the principal advantages and limitations of the use of Google EarthTM in geomorphology

Advantages Limitations

Free, instant access to satellite and aerial imagery within a simple, Largely limited to optical satellite and aerial imagery. Other relevant
interactive, and spatially searchable interface. Time and cost-effective, imagery (e.g., radar) may need to be ordered at cost.
particularly when trying to examine large areas.
Near-complete, aerial photograph or satellite image coverage of the land Image quality (e.g., spatial resolution, cloud cover, and shadows)
surface of the Earth and other planetary bodies (e.g., Mars and Earth’s may be variable within and between regions. Local errors in spot
Moon). Terrain can be examined from a variety of scales and height information, 3D rendering, and image stitching may be
perspectives, including vertical, oblique, and 3D perspectives. found.
Images are displayed within a coordinate system (latitude and Map datums and projections cannot be altered (a default Equatorial
longitude). Possible to search for features on the basis of coordinates Cylindrical Projection is used).
or place names.
Can import and overlay images and text from other sources (e.g., Can be time consuming and costly to source appropriate imagery.
SRTM-derived contour maps) using the spatial framework provided by Lack of contours or geological map layers limit possibilities for
Google EarthTM. Geological layers and feature databases suitable for landform analyses without additional data sources and more
importing into Google EarthTM can be shared easily across the internet. sophisticated, GIS-based procedures.
Imagery is updated regularly, so most global regions have imagery Updated imagery does not necessarily result in improved potential
relating to the past few years. In some areas, recent changes can be for geomorphological analyses (e.g., if cloud cover is present).
seen through the ‘historical imagery’ function. ‘Historical imagery’ function is not available for all areas of the
globe, and ‘date stamping’ of images is not always clear.
Basic quantitative measurements can be undertaken using simple tools Accuracy and precision of slope measurements is limited by reliance
(e.g., distances and slopes). on spot height information. Few other quantitative analytical tools
are available in the free-to-download version.
Enables improvements in spatial thinking skills (e.g., comparisons, Temptation for Google EarthTM to be uncritically substituted for other
transitions, and patterns) that are inherent to geomorphological forms of geomorphological analysis (e.g., fieldwork and other
reasoning. forms of image analysis), where these and other spatial thinking
skills may be developed better.
By embedding local observations/measurements in broader spatial Uncritical use of Google EarthTM may lead to superficial reliance on
context, generating regional/global scale datasets, and demonstrating analysis of landforms essentially from planform attributes alone, or
the effects of scale on landform controls and response, Google EarthTM on generating visually appealing images but with little improvement
may provoke more informed hypotheses about landform development. in process understanding or timescales of development.
May encourage researchers, instructors, and students to try other Use of Google EarthTM can only promote deep geomorphological
geomorphological techniques, including more advanced GIS packages. understanding if forming part of an integrated study approach that
may also involve field, laboratory, geochronology, and modeling
techniques.

advantage of Google EarthTM, namely that hypotheses about paper, Weissmann et al. (2010) provide not only a database of
landforms and processes can be discussed more effectively their identified continental sedimentary basins with names
with distant colleagues than perhaps used to be the case. To corresponding to their latitudes and longitudes, but also a
provide a more personal example, a previous global overview KMZ file for importing into Google EarthTM that shows the
of mixed bedrock-alluvial anabranching rivers (Tooth and location of these basins with a series of ‘pins.’ Reviewers and
McCarthy, 2004) was undertaken solely on the basis of a lit- readers of manuscripts based on some types of field-based or
erature search, the limited web-based information available at desktop research will benefit greatly from these sorts of ap-
the time, and text-based email correspondence. The release of proaches, for they can now view study areas and specific lo-
Google EarthTM, however, has enabled identification of other cations, improving their ability to visualize the landforms
mixed bedrock-alluvial anabranching rivers (Figure 1), some being described, and perhaps enabling them to challenge in-
of which lie in remote or politically inaccessible areas, and this terpretations in ways not previously possible.
is facilitating ongoing discussions with colleagues worldwide Other research groups have used Google EarthTM to pro-
regarding their characteristics and controls. Coordinates of vide a framework for importing, displaying, and manipulating
river reaches of interest can be emailed to colleagues so that other forms of imagery besides optical aerial and satellite
they can search for them in their own desktop version of imagery. For example, the Phased Array L-band Synthetic
Google EarthTM, or images can be captured as screen shots and Aperture Radar (PALSAR) carried aboard the Japanese Ad-
emailed. For more advanced data sharing, Keyhole Markup vanced Land Observing Satellite (ALOS) has been used to
Language links (KML/KMZ files) can be compiled and create a radar mosaic of the Sahara and Arabia, which covers a
emailed: the recipient then simply ‘drags and drops’ the file vast area between latitudes 171 N and 371 N and longitudes
into their own Google EarthTM desktop display and is provided 171 W and 601 E (Paillou et al., 2010). Geocoded PALSAR
with additional information layers (e.g., marked localities, images are managed with the help of a web map server that
geological and slope maps, other graphics, and ancillary data) enables one to import and display these data in Google
that can be draped over the Google EarthTM terrain. For in- EarthTM, thereby allowing a very simple and convenient access
stance, as part of the data repository associated with their to the radar data, and facilitating comparison between optical
58 Google EarthTM in Geomorphology: Re-Enchanting, Revolutionizing, or Just another Resource?

(a) (b)

(c) (d)

Figure 1 Illustrations of mixed bedrock-alluvial anabranching rivers using Google EarthTM imagery: (a) Nile River, upstream of Karima, northern
Sudan (flow direction from upper right to lower left) (r 2009 AND, r 2009 Europa Technologies, Image r 2009 DigitalGlobe, Image r 2009
TerraMetrics); (b) Vaal River at Parys, Free State, central South Africa (flow direction from upper right to upper left) (Image r 2010 GeoEye,
r 2010 Cnes/Spot image); (c) Potomac River near Great Falls, eastern USA (flow direction from upper middle to lower right) (Image USDA
Farm Service Agency); (d) Rio Branco, downstream of Boa Vista, northern Brazil (flow direction from upper middle to lower left) (Image r 2010
DigitalGlobe, r 2010 MapLink/Tele Atlas, Image r 2010 TerraMetrics). North is to the top in all images but the scales are variable. Some of
these rivers have long been known or suspected to exhibit the characteristics of mixed bedrock-alluvial anabranching, but with the release of
Google EarthTM, this has been confirmed with imagery, and many other examples have been identified.

and radar images. This aids various ongoing geological and dependence of landform controls can readily be illustrated in
geomorphological investigations, including the detection and Google EarthTM by examining regions or features at different
mapping of features such as buried paleochannels, impact scales. In the Kalahari sand sea, for example, linear dune orien-
craters, and faults (Paillou et al., 2009, 2010). tations at large spatial scales tend to be broadly related to re-
For geomorphological education and outreach, Tooth (2006) gional anticyclonic air circulation (e.g., Thomas and Shaw,
recognized the potential power of 3D visualization and spatial 1991), but at smaller scales, other factors assume greater im-
awareness provided by virtual globes. Indeed, the web-based and portance, with dune orientations commonly adjusting in re-
published examples of the use of Google EarthTM cited (see sponse to wind and sand transport changes resulting from local
Section 14.5.3.4 above) reflect a general consensus that the ap- topographic forcing by river valleys or ranges (e.g., Bullard and
plication is a highly cost-effective and academically beneficial Nash, 1998) (Figure 2). Furthermore, imagery available for other
way to integrate technology into the classroom, lecture theater, or planetary bodies can be used to enhance the understanding of
laboratory. In particular, unfamiliar landscapes can be brought Earth’s processes and landforms (e.g., Baker, 1993). Using virtual
into the lecture theater in ways that are simply not possible with globes to view heavily cratered regions of the Moon or Mars, for
static slides or photographs taken from fixed perspectives, and instance, immediately begs the question as to why craters are less
familiar landscapes can be seen from new perspectives. With numerous and less well preserved on Earth, which leads on to
carefully designed exercises, the use of Google EarthTM can not consideration of the role of plate tectonics in recycling large parts
only provoke more informed contemplation about landform of the Earth’s crust, and the influence of subaerial processes on
processes and controls but can also help promote improvements landscape denudation. Innovative use of such approaches in
in generic spatial thinking skills (e.g., making comparisons, education and outreach can only enhance appreciation of geo-
understanding transitions, and recognizing patterns). Scale morphology, and indeed the educational literature contains an
 Google EarthTM in Geomorphology: Re-Enchanting, Revolutionizing, or Just another Resource? 59

Figure 2 Oblique southwards view of the Orange River and aeolian Figure 3 Example of an error in the Google EarthTM 3D terrain
dunes near the Boegoeberg Range, Northern Cape, western South model along the Senqu River, Lesotho (Image r 2010 DigitalGlobe).
Africa using Google EarthTM imagery (r 2010 Cnes/Spot Image). The river flows from left to right but appears to rise over a hump in
Numerous linear dunes in the southern part of the Kalahari sand sea the valley floor (middle right of image), as confirmed by spot height
tend to be oriented broadly southeast (middle and upper right side of information (not shown) that incorrectly indicates that the river water
image) but are deflected northeast and east by the 200–250 m high surface rises B70 m before once again falling toward the far upper
range, and coalesce into fewer linear dunes that abut the Orange right of the image.
River floodplain. By enabling examination of terrain from different
perspectives, Google EarthTM can help to illustrate visually how
landform controls and responses vary with scale.

increasing number of case studies that provide evidence for such

claims based, for example, on documented student responses to
the use of Google EarthTM in classroom, lecture, or laboratory
settings (e.g., Thorndycraft et al., 2009).

14.5.4.2 Limitations of Google EarthTM in Geomorphology

The use of Google EarthTM in geomorphology is not without
limitations. With the present generation of Google EarthTM,
only optical (aerial photograph or satellite) imagery is avail-
able, and although other forms of imagery (e.g., radar image
mosaics) can be imported and displayed in Google EarthTM
(e.g., Paillou et al., 2010), these data are not always freely Figure 4 Part of Patagonia, southern Argentina, illustrating how the
distributed. In terms of the free imagery available in Google variable image quality in Google EarthTM can be far from optimal for
EarthTM, one of the biggest limitations is that image quality geomorphological analyses (Image r 2010 DigitalGlobe, r 2010
also varies from area to area; spatial resolution in many areas Cnes/Spot Image).
(particularly remote areas) is only sufficient for examination
of macroscale geomorphological features, and although base, which is suitable for global views but inevitably intro-
higher-resolution imagery is being provided for increasing duces errors at smaller scales and in polar regions, yet ad-
number of regions worldwide, images may not have been vanced mapping (e.g., manipulation of a map projection or
taken at optimum times for geomorphological analyses (e.g., datum) is not possible. Spot elevation data and the 3D terrain
when cloud cover, shadows, or high water levels obscure some model underlying Google EarthTM landscapes is derived from
landforms). This can restrict the analyses being undertaken; SRTM data with fill in from DEMs derived from other datasets,
for instance, owing to the variable resolution of satellite im- but local errors in spot height information, 3D rendering, and
ages on Google EarthTM, Constantine and Dunne (2008) were edge matching (stitching) of image sets occur locally (e.g.,
forced to restrict their analysis to large meandering rivers Figure 3; see also examples highlighted in Wikipedia, 2010).
(minimum width of 100 m) so that the spatial attributes could These errors are generally fixed with the release of newer ver-
be measured confidently. A further limitation is that it is not sions of the application, although image updates can intro-
always possible to tell the exact date of the images displayed, duce new errors and the updated imagery is not necessarily an
although regular updates mean that imagery is generally from improvement in terms of its potential for geomorphological
the past few years, and the provision of the ‘Historical Im- analyses, particularly where stitched images vary in contrast,
agery’ feature has improved date stamping for some areas. hue, or levels of cloud cover (Figure 4). The lack of options
The mapping and analytical tools of Google EarthTM are to display contours is also frustrating for some forms of
also limited. Google EarthTM uses an Equatorial Cylindrical geomorphological analysis, although Hoffman and Winde
Equidistant (simple cylindrical) projection for its imagery (2010) have demonstrated how it is possible to use more
60 Google EarthTM in Geomorphology: Re-Enchanting, Revolutionizing, or Just another Resource?

sophisticated GIS procedures to retrieve high-resolution data

at 1 m intervals from the satellite imagery in Google EarthTM.
These data can then be used to produce high-resolution
contour maps and DEMs (e.g., for small wetlands) that com-
pare favorably to those produced using standard topographic
maps (1:50 000 scale) and aerial photography (Hoffman and
Winde, 2010). Similarly, other forms of geomorphological
analysis are not possible without supporting geological
map layers, although one can compile such information in a
more sophisticated GIS package, and then use KML/KMZ files
to import and overlay such information in Google
EarthTM (e.g., see Conroy et al., 2008 in a paleontological
context).
The free-to-download version of Google EarthTM enables
basic length/distance measurements to be made using the
‘ruler’ tool, and simple slope estimates can be made Figure 5 Oblique view of amphitheater-headed valleys near
from distance and spot height measurements, whereas the Pro Grootberg, northwestern Namibia, using Google EarthTM imagery
(r 2010 Cnes/Spot Image, r 2010 Google). Google EarthTM can
version enables basic measurements of polygons and
help in establishing the regional to global distribution of such
circles (e.g., perimeter, area, radius, and circumference). Owing features, but without supporting geological information and field
to the mapping and visualization limitations and errors of investigations, the formative processes remain unclear (in this
Google EarthTM, however, inevitably some questions have particular example, quartz-latite igneous strata overlie weathered
been raised over the accuracy of some of these measurements. basalts, but the relative roles of groundwater and surface runoff in
Nonetheless, the Wikipedia (2010) article on Google EarthTM weathering and erosion are not known).
reports that measurements of the equatorial and meridional
circumferences using the ‘ruler’ tool in Google EarthTM com-
pares favorably to independently determined measurements,
and shorter measured distances in Google EarthTM compare (e.g., Lamb et al., 2006, 2008) (Figure 5). Google EarthTM can
well to those measured from published maps. help to improve understanding of the distribution and char-
Given these limitations, many researchers have found that acteristics of these kinds of landforms, but can provide little or
although Google EarthTM provides an invaluable starting nothing in the way of fundamental process insights or
point for analysis, the application by itself is insufficient and understanding of timescales of development. Clearly, along-
needs to be combined with other datasets and forms of an- side the use of Google EarthTM, there is still a need for other
alysis. For example, in Weissmann et al.’s (2010) investigation, investigative approaches that may include field measure-
Google EarthTM enabled ‘cursory identification’ of the con- ments, laboratory experimentation and analyses, geochron-
tinental sedimentary basins principally using a ‘low relief’ ology, and computational modeling.
criterion (see their data repository item 2010006), but Landsat
imagery and SRTM data still had to be compiled to verify that
the low relief regions were properly identified. In cases where 14.5.5 Possible Future Developments in the Use of
advanced mapping and quantification of key parameters (e.g., Google EarthTM in Geomorphology
areas and slopes) is needed, Google EarthTM is certainly not a
replacement for more sophisticated GIS-based analyses and Given the rapid pace of technological developments, pre-
other digital topographic datasets (e.g., DEMs derived from dicting the trajectory of future developments in the use of
airborne or terrestrial platforms). There is also the risk, Google EarthTM and other virtual globes in geomorphology is
common to existing remote sensing imagery and GIS software, a daunting and probably hazardous task. The technology is
that virtual globes will simply be used to generate visually young and still developing, and its full potential in geo-
appealing imagery without any corresponding advancement morphology perhaps has still to be exploited. Further techni-
in the understanding of geomorphological process and form. cal developments (e.g., higher image resolution, more
In particular, the dangers of superficial interpretations of frequent image updates, and other types of image datasets),
processes based on planform or other form attributes alone and incorporation of innovative visualization of environ-
will be familiar to many geomorphologists acquainted with mental phenomena relevant to geomorphological processes
the concept of equifinality, whereby different processes or (e.g., 3D fly-bys of weather systems – Butler, 2006), will un-
different process combinations converge to produce the same doubtedly increase the appeal and potential of virtual globes
(or similar) landforms. The history of geomorphology is rid- in geomorphology.
dled with examples where process inferences from even ap- A specific example of a possible future development in the
parently simple landforms have later been shown to be use of Google EarthTM in geomorphology and the geosciences
erroneous, or where the process inferences continue to gen- more generally might be the development of more shared
erate considerable debate; examples include ongoing debates community datasets. By and large, geomorphologists tend to
over the processes of pediment formation in arid regions be an individualistic breed generally resistant to the mega-
(e.g., Strudley and Murray, 2007), and the role of ground- team projects favored by other geoscientists such as climat-
water sapping in amphitheater-headed valley formation ologists and oceanographers (Dorn, 1994), but there are
 Google EarthTM in Geomorphology: Re-Enchanting, Revolutionizing, or Just another Resource? 61

(a)

(b)

Figure 6 Example screen shots from community datasets that utilize the Google EarthTM interface and capabilities: (a) Weissmann et al.’s
(2010) database of continental sedimentary basins (each basin is marked by a green pin) (r 2010 Tele Atlas, r 2010 Google, US Department
of State Geographer, r 2010 Europa Technologies); (b) the GLIMS glacier database, showing the footprint of relevant, low-resolution ASTER
imagery that is available for this part of Europe (Image NASA, Data SIO, NOAA, US Navy, NGA, GEBCO).
62 Google EarthTM in Geomorphology: Re-Enchanting, Revolutionizing, or Just another Resource?

growing signs of community-wide projects that attempt to restricted context of ‘DFS’). To facilitate such a task, the
exploit the powers of Google EarthTM. For instance, Weiss- powers of virtual globes perhaps could be combined with
mann et al.’s (2010) KMZ file (part of their data repository ongoing advances in automated feature detection, classifi-
item), provides a database of their identified continental sedi- cation, and mapping (Vickers et al., 2009). There might also
mentary basins that can be imported into Google EarthTM be an opportunity for innovative outreach projects along
(Figure 6(a)). Weissmann et al. (2010) called this database lines similar to the Moon Zoo mapping project (http://
version 1.0 and specifically invited collaboration with others www.moonzoo.org/) that invites public participation in
with a view to updating and improving the database through studying the lunar surface in unprecedented detail. The aim of
identification of additional basins and possible removal of the project is to provide detailed crater counts for the Moon’s
others. A larger initiative is NSIDC’s Global Land Ice surface using images from NASA’s Lunar Reconnaissance
Measurements from Space (GLIMS) Glacier Database, which Orbiter (LRO), which is a daunting task because the sheer
has been developed as an archive of glacier outlines extracted volume of images supplied by the LRO threatens to overwhelm
from ASTER images and other sources, and is intended scientists.
to document and monitor changes in the spatial extent
of glacier systems through time (Raup et al., 2007). Owing
to the enormous volume of ASTER imagery, the GLIMS 14.5.6 Conclusions
Project has developed a unique subset of ASTER imagery
that covers glacierized terrain, making it easier to find, view, The case studies highlighted in the foregoing sections illustrate
and order ASTER images that have been captured over areas the growing use of Google EarthTM in many geomorphological
where glaciers may exist and that may be suitable for per- subfields, but is this leading to re-enchantment of the discip-
forming glacier analyses. The GLIMS ASTER Browse data can line (Tooth, 2006)? The answer is possibly ‘yes,’ in the limited
be viewed in two ways: from within the GLIMS Glacier Viewer sense that the vastly increased availability of images of the
or within Google EarthTM by downloading a KMZ file surfaces of the Earth and other planetary bodies have generated
(Figure 6(b)). new types of discussions with students and colleagues, en-
These examples of the generation of global datasets that livened many conference presentations, helped to richly illus-
may be shared freely amongst the geomorphological and trate a number of publications, and promoted new kinds of
geosciences community – and perhaps among the interested regional and global datasets to be assembled. Even outside the
lay community more widely – could be applied to investi- academic and applied geomorphological community, the
gations of many other Earth surface features such as submarine popularity of virtual globes illustrates that, among many other
topography, aeolian dunes, and river patterns. For example, the topics, public interest in the landscapes of Earth and other
Irish government funded program entitled Integrated Mapping planetary bodies is high; for instance, some of the online dis-
For the Sustainable Development of Ireland’s Marine Resource cussion threads in Google EarthTM Blog and Google EarthTM
(INFOMAR) is providing detailed insight into seabed topo- Community fora revolve around questions of landform genesis
graphy in its submarine territory, including the spatial distri- and development. Although some of these discussions are of
bution and frequency of drowned drumlin fields, canyons, and dubious accuracy and clarity, and typically are far removed
cold water reefs (Mulqueen, 2010). All the surveys are available from the more informed debates in the academic literature, the
for free on the INFOMAR website (http://www.infomar.ie/), interest and even fascination in surface processes and land-
and are provided in a format that allows users to combine forms is nonetheless clearly evident. This provides as-yet-
them with Google OceanTM. Regional and even global surveys untapped opportunities for promoting the poorly understood
of aeolian sand seas have been undertaken and have provided term ‘geomorphology’ and its role as a worthwhile, valid, and
insight into the controls on the spatial distribution and fre- distinctive scientific endeavor (Tooth, 2006, 2009b).
quency of occurrence of different dune types (e.g., Breed et al., Is the growing use of Google EarthTM revolutionizing the
1979; Smith, 1982; Wang et al., 2005; Wolfe et al., 2009), but discipline? The answer here is probably ‘no.’ For many (pos-
as yet there is no complete, geographically accurate global map sibly most) geomorphologists, particularly those concerned
of aeolian sand seas and dune types. At present, however, de- with the details of process-form interactions and landform
velopment of a digital global atlas of Quaternary sand sea ac- change at relatively small (e.g., plot or hillslope) scales, the
tivity is underway (http://inquadunesatlas.dri.edu/) and there potential research advantages of Google EarthTM may appear
are possible plans to make the database available online in a trivial, and the application has probably had little impact on
format suitable for importing to Google EarthTM. By contrast, their working practices. Even for those geomorphologists that
except in a few limited areas, similar levels of systematic an- have made use of Google EarthTM imagery in larger scale
alysis have not yet been undertaken for river patterns. Two landform/landscape investigations, this has not (as yet) led to
decades ago, Newson (1992, p. 20) suggested that ‘‘We can and radical new discoveries or insights. Google EarthTM un-
should make use of remote sensing and image processing doubtedly has been invaluable for helping to generate re-
technology to attempt national or continental, even global gional- or global-scale datasets that provide a firmer basis for
fluvial ‘auditing’ to discover new patterns and quantify the hypothesizing about processes and forms, and thus has helped
occurrence rate of the old.’’ Although the methods for under- to make valuable additions to the corpus of knowledge about
taking this kind of fluvial ‘auditing’ and structuring the re- the Earth’s landforms (e.g., Weissmann et al., 2010). These
sulting databases still need to be determined, we are now in a outcomes, however, are in no way comparable to those in-
situation where we can at least perceive what might be required sights that emerged, say, when maps of sea floor topography
(see Hartley et al., 2010 for an example within the more first became available from the late 1940s onwards, and the
 Google EarthTM in Geomorphology: Re-Enchanting, Revolutionizing, or Just another Resource? 63

incontrovertible evidence for sea floor spreading and contin- Ballagh, L.M., Parsons, M.A., Swick, R., 2007. Visualising cryospheric images in a
ental drift contributed to a true revolution in the geosciences. virtual environment: present challenges and future implications. Polar Record 43,
305–310.
Nor are these outcomes comparable to the fundamental
Bourke, M.C., 2010. Barchan dune asymmetry: observations from Earth and Mars.
transformation in understanding of the dynamics and vari- Icarus 205, 183–197.
ability of Earth surface processes that were provided when Bourke, M.C., Goudie, A.S., 2009. Varieties of barchan form in the Namib Desert
satellite imagery first started to become widely available from and on Mars. Aeolian Research 1, 45–54.
the 1970s onwards, as exemplified recently by the recognition Breed, C.S., Fryberger, S.J., Andrews, S., McCauley, C., Lennartz, F., Gebel, D.,
Horstman, K., 1979. Regional studies of sand seas using Landsat (ERTS)
and growing concern over the possible accelerated collapse of imagery. In: McKee, E.D. (Ed.), A Study of Global Sand Seas. United States
Antarctica’s and Greenland’s ice sheets (National Research Geological Survey Professional Paper. United States Geological Survey, Reston,
Council, 2008). Indeed, for field- and desk-based investi- VA, 1052, pp. 305–397.
gations of surface processes and landform development, it is Bullard, J.E., Nash, D.J., 1998. Linear dune pattern variability in the vicinity of dry
valleys in the southwest Kalahari. Geomorphology 23, 35–54.
arguable whether other developments in spatial data acqui-
Butler, D., 2006. The web-wide world. Nature 439, 776–778.
sition and image processing (e.g., generation of DEMs from Chapman, M.G. (Ed.), 2007. The Geology of Mars: Evidence from Earth-Based
global topographic datasets such as ASTER, airborne laser Analogs. Cambridge Planetary Science Series 5. Cambridge University Press,
altimetry, or terrestrial laser scanning) or in geochronology Cambridge.
(e.g., luminescence and cosmogenic isotope analysis) are Conroy, G.C., Anemone, R.L., van Regenmorter, J., Addison, A., 2008. Google
Earth, GIS, and the Great Divide: a new and simple method for sharing
more revolutionary (e.g., French, 2003; Heimsath and Ehlers,
paleontological data. Journal of Human Evolution 55, 751–755.
2005; Evans and Lindsay, 2010; Milan et al., 2010; Hugenholtz Constantine, J.A., Dunne, T., 2008. Meander cutoff and the controls on the
and Barchyn, 2010). production of oxbow lakes. Geology 36, 23–26.
Is Google EarthTM just another resource? The answer here is Dorn, R.I., 1994. The role of climatic change in alluvial fan development.
both ‘yes’ and ‘no’. Google EarthTM is just another resource in In: Abrahams, A.D., Parsons, A.J. (Eds.), Geomorphology of Desert
Environments. Chapman and Hall, London, pp. 593–615.
the sense that it is only one part of the geomorphological Dunagan, S., 2007. Integrating Google Earth with geomorphology lab activities and
‘toolkit’ and is most effective when used as part of an inte- student presentations. Geological Society of America Abstracts with Programs
grated study approach involving fieldwork, laboratory work, 39, 153.
other forms of image and statistical analyses, geochronology, Evans, M., Lindsay, J., 2010. High resolution quantification of gully erosion in
upland peatlands at the landscape scale. Earth Surface Processes and Landforms
and perhaps computational modeling. Google EarthTM was
35, 876–886.
never intended to be a replacement for these activities or even French, J.R., 2003. Airborne LiDAR in support of geomorphological and hydraulic
for more sophisticated GIS packages, yet it may well encourage modelling. Earth Surface Processes and Landforms 28, 321–335.
researchers and instructors to explore these avenues (Butler, Geography 2.0: Virtual Globes, no date. Virtual Globes: GIS for the masses? Call
2006). But it is not just another resource: despite its limi- for abstracts, Session ID: 3232, Association of American Geographers Annual
Meeting, San Francisco, April 2007. Available at: http://
tations, Google EarthTM remains a remarkable resource. geography2.blogspot.com/ (accessed 20 July 2010].
A decade ago, few geomorphologists could have envisioned Gramling, C., 2007. Google Planet: with virtual globes, Earth scientists see a new
having access to aerial photograph or satellite imagery of vir- world. Geotimes, February 2007. Available at: www.geotimes.org/feb07/
tually all surfaces of the Earth and some other planetary trends.html (accessed 20 July 2010).
Gregory, K.J., 2010. The Earth’s Land Surface: Landforms and Processes in
bodies in one free, easy-to-use application that runs on a
Geomorphology. Sage, Los Angeles.
desktop computer machine. The many advantages of Google Handwerk, B., 2006. Google Earth, satellite maps boost armchair archaeology.
EarthTM surely outweigh the limitations, and so the appli- National Geographic News, 7 November 2006. Available at: http://
cation will continue to help generate more hypotheses re- news.nationalgeographic.com/news/2006/11/061107-archaeology.html (accessed
garding landform genesis that could be tested by field, 20 July 2010).
Hartley, A.J., Weissmann, G.S., Nichols, G.J., Warwick, G.L., 2010. Large
laboratory, geochronological, or modeling approaches. Goo- distributive fluvial systems: characteristics, distribution, and controls on
gle EarthTM and other virtual globes can thus only help pro- development. Journal of Sedimentary Research 80, 167–183.
mote the recent heightened level of research interest in meso Heimsath, A.M., Ehlers, T.A. (Eds.) 2005. Quantifying rates and timescales of
and macroscale landform development (e.g., Summerfield, geomorphic processes: Part 1. Earth Surface Processes and Landforms 30,
917–1067.
2005a, b), as well as contributing to geomorphological edu-
Hoffman, E., Winde, F., 2010. Generating high-resolution digital elevation models
cation and outreach more generally. for wetland research using Google Earth (TM) imagery – an example from
South Africa. Water SA 36, 53–68.
Hugenholtz, C.H., Barchyn, T.E., 2010. Spatial analysis of sand dunes with a new
References global topographic dataset: new approaches and opportunities. Earth Surface
Processes and Landforms 35, 986–992.
Kamadjeu, R., 2009. Tracking the polio virus down the Congo River: a case study on
Abbot, P.L., 2008. Natural Disasters, 6th edn. McGraw-Hill Higher Education, New the use of Google EarthTM in public health planning and mapping. International
York, 524 pp. (Google Earth worksheets available at: http://highered.mcgraw- Journal of Health Geographics 8, 4. http://dx.doi.org/10.1186/1476-072X-8-4.
hill.com/sites/0073050342/student_view0/google_earth_worksheets.html Kingsbury, P., Jones, III J.P., 2009. Walter Benjamin’s Dionysian adventures on
(accessed 20 July 2010). Google Earth. Geoforum 40, 502–513.
AHRC ICT Methods Network, no date. Digital Tools for Archaeology: A Methods Lamb, M.P., Howard, A.D., Johnson, J., Whipple, K.X., Dietrich, W.E., Perron, J.T.,
Network Working Paper. Available at: http://www.arts-humanities.net/ 2006. Can springs cut canyons into rock? Journal of Geophysical Research –
briefingpaper/digital_tools_archaeology (accessed 20 July 2010). Planets 111, E07002.
Allen, T.R., 2007. Digital terrain visualization and virtual globes for teaching Lamb, M.P., Dietrich, W.E., Aciego, S.M., DePaolo, D.J., Manga, M., 2008.
geomorphology. Journal of Geography 106, 253–266. Formation of Box Canyon, Idaho, by megaflood: implications for seepage
Baker, V.R., Twidale, C.R., 1991. The reenchantment of geomorphology. erosion on Earth and Mars. Science 320, 1067–1070.
Geomorphology 4, 73–100. Lefer, T.B., Anderson, M.R., Fornari, A., Lambert, A., Fletcher, J., Baquero, M.,
Baker, V.R., 1993. Extraterrestrial geomorphology: science and philosophy of 2008. Using Google Earth as an innovative tool for community mapping. Public
Earthlike planetary landscapes. Geomorphology 7, 9–35. Health Reports 123, 474–480.
64 Google EarthTM in Geomorphology: Re-Enchanting, Revolutionizing, or Just another Resource?

Leier, A.L., DeCelles, P.G., Pelletier, J.D., 2005. Mountains, monsoons, and Summerfield, M.A., 2005a. The changing landscape of geomorphology. Earth
megafans. Geology 33, 289–292. Surface Processes and Landforms 30, 779–781.
Milan, D.J., Heritage, G.L., Large, A.R.G., Entwistle, N.S., 2010. Mapping hydraulic Summerfield, M.A., 2005b. A tale of two scales, or the two geomorphologies.
biotopes using terrestrial laser scan data of water surface properties. Earth Transactions of the Institute of British Geographers NS 30, 402–415.
Surface Processes and Landforms 35, 918–931. Thomas, D.S.G., Shaw, P.A., 1991. The Kalahari Environment. Cambridge University
Monkkonen, P., 2008. Using online satellite imagery as a research tool: mapping Press, Cambridge.
changing patterns of urbanization in Mexico. Journal of Planning Education and Thorndycraft, V.R., Thompson, D., Tomlinson, E., 2009. Google Earth, virtual
Research 28, 225–236. fieldwork and quantitative methods in Physical Geography. Planet 22,
Mulqueen, É., 2010. Seeing beneath Ireland’s seas. Geographical 82, 48–51. 48–51.
NASA, 2004. NASA computer program is hot download on the internet. Available at: Tooth, S., 2006. Virtual globes: a catalyst for the re-enchantment of
www.nasa.gov/vision/earth/environment/World_Wind.html (accessed 20 July 2010). geomorphology? Earth Surface Processes and Landforms 31, 1192–1194.
National Research Council, 2008. Earth Observations from Space: The Tooth, S., 2009a. Arid geomorphology: emerging research themes and new
First 50 Years of Scientific Achievements. Committee on Scientific frontiers. Progress in Physical Geography 33, 251–287.
Accomplishments of Earth Observations from Space, The National Academies Tooth, S., 2009b. Invisible geomorphology? Earth Surface Processes and Landforms
Press, Washington, DC. 34, 752–754.
Newson, M.D., 1992. Geomorphic thresholds in gravel-bed rivers – refinement for Tooth, S., McCarthy, T.S., 2004. Anabranching in mixed bedrock-alluvial rivers: the
an era of environmental change. In: Billi, P., Hey, R.D., Thorne, C.R., Tacconi, P. example of the Orange River above Augrabies Falls, Northern Cape Province,
(Eds.), Dynamics of Gravel-Bed Rivers. John Wiley and Sons, Chichester, pp. South Africa. Geomorphology 57, 235–262.
3–20. Tuttle, B.T., Anderson, S., Huff, R., 2008. Virtual globes: an overview of their
Nourbakhsh, I., Sargent, R., Wright, A., Cramer, K., McClendon, B., Jones, M., history, uses, and future challenges. Geography Compass 2, 1478–1505.
2006. Mapping disaster zones. Nature 439, 787–788. Vickers, M., Gibbens, M., Cook, A., 2009. Automated texture classification and
Olsen, S., 2007. Do-gooders doing Google Earth. CNET News, 7 June 2007. landform taxonomy for populating a GIS. Proceedings of the Remote Sensing
Available at: http://news.cnet.com/Do-gooders-doing-Google-Earth/2100-1038_3- and Photogrammetry Annual Conference, 8–11 September 2009, Leicester, UK,
6189464.html (accessed 20 July 2010). pp. 604–610.
Pacifici, A., 2009. The Argentinean Patagonia and the Martian landscape. Planetary Wang, T., Xue, X., Chen, G., 2005. Map of the Desert and Aeolian Desertification in
and Space Science 57, 571–578. China. SinoMaps Press, Beijing.
Paillou, P., Lopez, S., Farr, T., Rosenqvist, A., 2010. Mapping subsurface geology in Weissmann, G.S., Hartley, A.J., Nichols, G.J., Scuderi, L.A., Olson, M., Buehler, H.,
Sahara using L-band SAR: first results from the ALOS/PALSAR imaging radar. Banteah, R., 2010. Fluvial form in modern continental sedimentary basins:
Journal of Selected Topics in Earth Observations and Remote Sensing 3(4), Part: 2. distributive fluvial systems. Geology 38, 39–42 (includes data repository item
Paillou, P., Schuster, M., Tooth, S., Farr, T., Rosenqvist, A., Lopez, S., Malezieux, 2010006).
J.-M., 2009. Mapping of a major paleodrainage system in eastern Libya using Wikipedia, 2010. Google Earth. Available at: http://en.wikipedia.org/wiki/
orbital imaging radar: the Kufrah River. Earth and Planetary Science Letters 277, Google_Earth (accessed 20 July 2010).
327–333. Wolfe, S.A., Gillis, A., Robertson, L., 2009. Late Quaternary Eolian Deposits of
Patterson, T.C., 2007. Google Earth as a (not just) Geography education tool. Northern North America: Age and Extent. Geological Survey of Canada, Open
Journal of Geography 106, 145–152. File 6006, CD-ROM. Available at: http://geopub.nrcan.gc.ca/
Raup, B., Racoviteanu, A., Khalsa, S.J.S., Helm, C., Armstrong, R., Arnaud, Y., moreinfo_e.phpid=226434 (accessed 20 July 2010).
2007. The GLIMS geospatial glacier database: a new tool for studying glacier
change. Global and Planetary Change 56, 101–110.
Scharl, A., 2007. Towards the geospatial web: media platforms for managing
geotagged knowledge repositories. In: Scharl, A., Tochtermann, K. (Eds.), The Relevant Websites
Geospatial Web – How Geo-Browsers, Social Software and the Web 2.0 are
Shaping the Network Society. Springer, London, pp. 3–14. http://www.infomar.ie/
Sinton, D.S., Schultz, R.B., 2009. GIS and mapping technologies: applications for INFOMAR (Integrated Mapping for the Sustainable Development of Ireland’s
reasoning and critical thinking. In: Solem, M., Foote, K. (Eds.), Teaching Marine Resources) is a joint venture between the Geological Survey of Ireland
College Geography: A Practical Guide for Graduate Students and Early Career and the Marine Institute.
Faculty. Pearson, Upper Saddle River, NJ, pp. 69–80. http://inquadunesatlas.dri.edu/
Smith, R.S.U., 1982. Sand dunes in the North American deserts. In: Bender, G. ‘Sand Seas and dune fields of the world: a digital Quaternary atlas’ is a project
(Ed.), Reference Handbook of the Deserts of North America. Greenwood Press, funded by the International Union for Quaternary Research (INQUA) (Project
Westport, CT, pp. 481–554. 0704) and the Desert Research Institute (DRI).
Strudley, M.W., Murray, A.B., 2007. Sensitivity analysis of pediment development http://www.moonzoo.org/
through numerical simulation and selected geospatial query. Geomorphology 88, The Moon Zoo public participation program is organized by the Citizen Science
329–351. Alliance http://www.citizensciencealliance.org/

Biographical Sketch

Dr. Stephen Tooth is a Reader in the Institute of Geography and Earth Sciences (IGES) at Aberystwyth University,
Wales, UK, and convenor of the River Basin Dynamics and Hydrology Research Group. He graduated with a
Geography BSc degree from the University of Southampton, England, and completed a PhD at the University of
Wollongong, Australia. Following a short period as a Temporary Lecturer at the University of Nottingham,
England, he undertook postdoctoral work at the University of the Witwatersrand, South Africa, and then joined
IGES in April 2000. His main research interests focus on geomorphology and sedimentology, especially in the
drylands of Australia and southern Africa. Particular research themes include: anabranching rivers; floodplains
and floodouts; wetlands in drylands; channel–vegetation interactions; bedrock-influenced rivers; controls on gully
erosion; long-term fluvial landscape development; paleoenvironmental change; and the use of drylands on Earth
as analogues for Martian surface environments.
 14.6 Methods in Geomorphology: Numerical Modeling of Drainage Basin
Development
YE Martin, University of Calgary, Calgary, AB, Canada
r 2013 Elsevier Inc. All rights reserved.

14.6.1 Background 66
14.6.2 Defining the Numerical Modeling Exercise 67
14.6.3 Geomorphic Process Equations 68
14.6.3.1 Introduction 68
14.6.3.2 Hillslope Processes 68
14.6.3.3 Channel Network Processes 68
14.6.3.4 Other Processes 69
14.6.3.5 Changes in Climate and Hydrology over Time 69
14.6.4 Constructing and Running the Model 69
14.6.5 Model Confirmation 70
14.6.6 Final Comments 70
References 71

Glossary
Channel network The network of channels toward which Landscape morphology Form and spatial structure of the
water flows from the interfluves, collects, and is output from landscape.
the drainage basin. Model confirmation Examination of the ability of a
Computer programming This involves the writing of a model to match prediction (model outcome) with
computer programming code that allows the computer to observation.
simulate the desired phenomenon. Numerical models Mathematical models that solve
Drainage basin development The study of how drainage equations using analytical or numerical methods to
basin morphology changes over time. simulate the behavior of a phenomenon.
Hillslopes These make up most of the nonchannelized Process operation An approach toward investigating
portions of the landscape. They may have notable slope phenomena that consider the various mechanical/chemical
gradients, and sediment and/or rock may be moved and other processes in operation versus the adoption of a
downslope due to a combination of weathering and more empirical approach that does not explicitly consider
sediment-transport processes. such processes.
Historical contingencies Drainage basin development is Sediment-transport equations Mathematical equations
constrained by historical contingencies. The latter term that simulate sediment transport; such equations have been
refers to boundary conditions or specific sequences of developed for hillslope, fluvial, and other systems.
events; these historical contingencies may be unknown, and
may also be unpredictable/random.

Abstract

The adoption of a numerical-modeling approach for the study of drainage-basin development is explored in this chapter.
Advances in computer technology in recent decades have allowed for the solution of very complex numerical models,
leading to notable advancements in our understanding of drainage-basin development. The study of drainage-basin de-
velopment faces particular challenges, as both process operation and historical contingencies must be taken into account.
Numerical models of drainage-basin change typically incorporate a series of equations for geomorphic and other processes
included in the model (e.g., weathering, tectonics, and hydrology). An important contribution of numerical modeling is
that it allows for the consideration of a series of ‘what-if’ questions that might otherwise be impossible to study given the
large scales of drainage-basin development. Two stages of drainage-basin development, initial drainage initiation and
subsequent drainage-basin development, are differentiated. The modeler should make clear which of these stages is the

Martin, Y.E., 2013. Methods in geomorphology: numerical modeling of

drainage basin development. In: Shroder, J. (Editor in Chief), Switzer, A.D.,
Kennedy, D.M. (Eds.), Treatise on Geomorphology. Academic Press, San
Diego, CA, vol. 14, Methods in Geomorphology, pp. 65–72.

Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00373-0 65

66 Methods in Geomorphology: Numerical Modeling of Drainage Basin Development

focus of a particular study, or if both will be considered. Examples are given to demonstrate the range of theoretically-based
research questions about drainage-basin development that can be addressed using a numerical-modeling approach. Next,
the general forms of sediment-transport equations for hillslope and channel-network processes that are typically employed
in models of drainage-basin development are reviewed. Practical issues related to the actual construction and running of
numerical models of drainage-basin development are discussed. Finally, issues and difficulties surrounding model con-
firmation are outlined.

14.6.1 Background must be defined suitably for larger spatial and temporal scales,
generally involving some simplification or abstraction of the
This chapter outlines issues that should be considered when process (Martin and Church, 2004). The model landscape
utilizing a numerical modeling methodology for the study of generally consists of a surface with either regular or irregular
drainage-basin development. Consideration of longer-term dimensions of grid cells; in many cases, a digital elevation
geomorphological change (i.e., timescales ranging from B103 model (DEM) is used as the initial surface.
to 106 þ years) is challenging, as it must take into account both Numerical models provide a powerful tool to assess
process operation and (generally unknown) historical contin- changes in the drainage basin over timescales that cannot be
gencies, the latter referring to boundary conditions and/or observed directly. However, given the large timescales in-
specific sequences of events at a location (Kennedy, 2006). volved, researchers must make decisions regarding the ‘flaws’
Unfortunately, information about historical contingencies (or limitations) within the model that must be accepted to
affecting landscape development is generally unavailable or not maintain computational tractability of numerical solutions
available in sufficient detail to be useful for modeling purposes. (see Martin and Church, 2004). For example, a full rendering
Researchers rely on a number of methodologies to study of landslide mechanics is not possible (nor may it be prefer-
drainage-basin change over larger timescales. For example, able) when considering landsliding processes over 103 þ -year
analyses of depositional and erosional records provide key timescales, in part, due to the impossibility of obtaining and
tools for making inferences about landscape history. These incorporating accurate and/or precise information about
types of analyses are now supported by new quantitative- controlling variables in such detail over these timescales.
dating techniques (e.g., cosmogenic isotopes, thermo- Therefore, generalization of process specification commonly
luminescence, and fission track), allowing for improved becomes necessary (see, for example, Section 14.6.3.2). It is
estimation of rates associated with longer-term landscape also essential that the model preserve what are deemed to be
change (Repka et al., 1997; Willett et al., 2003; Reinhardt et al., essential characteristics of landscape change, despite the ne-
2007). Although extrapolation of short- and medium-term cessary simplification in model construction (Martin and
records of process rates (e.g., measurements of suspended load Church, 2004). An important function of numerical modeling
and landslide inventories) to longer timescales is commonly is that researchers may address a series of ‘what-if’ questions
necessary, the representativeness of such information applied that otherwise would be impossible to explore, given the large
to longer timescales remains questionable. Physical/scale spatial and temporal scales associated with drainage-basin
modeling, including studies in flumes and model basins such development (Oreskes et al., 1994).
as those undertaken at the St. Anthony Falls Laboratory, Uni- Although there has been notable progress in the modeling
versity of Minnesota, is an important methodology to study of drainage basin development in recent decades as indicated
landscape change (e.g., Hasbargen and Paola, 2000). A major by the increasingly sophisticated questions now being con-
advantage of this type of approach is the capacity for experi- sidered in such models (see Tucker and Hancock (2010) for a
mental-type control that is not generally possible in the field, recent review of the state of landscape evolution modeling),
and especially not for the study of long-term landscape history. many key issues still confront researchers. Concerns remain
Numerical modeling of development of drainage basins is about how to best represent geomorphological processes over
the methodology that is the focus of this review. Increased larger spatial and temporal scales. For example, development,
computer-processing power in recent decades has made the calibration, and testing of transport relations used in such
solution of complex numerical models feasible, thus leading models remain problematic because of difficulties in acquiring
to significant advances in the study and understanding of how suitable field data over the representative timescales needed to
drainage basins change over time. A variety of numerical achieve these goals. That being said, the tools and techniques
models to study drainage-basin development have been put now available, including DEM/remote-sensing coverage, im-
forth in the past two decades, and they have been used to proved surveying instruments, and state-of-the-art dating
explore a wide range of research questions (e.g., Rosenbloom techniques, have improved significantly our ability to tackle
and Anderson, 1994; Tucker and Bras, 1998; Martin, 2005; these issues. In addition, feedbacks and interactions among
also see review paper of Tucker and Hancock (2010)). Nu- climate, hydrology, ecology, and geomorphology are increas-
merical modeling of drainage-basin development involves the ingly a focus of geomorphological research, and attempts are
definition of a series of appropriately defined equations rep- being made to incorporate such feedbacks into longer-term
resenting the relevant processes (e.g., hillslope, and fluvial), models of development of drainage basins (e.g., Willett, 1999;
which are then linked into mass-continuity equations and Istanbulluoglu and Bras, 2005). Testing of models of drainage
applied across the model landscape. The process equations basin development remains a topic that requires much further
 Methods in Geomorphology: Numerical Modeling of Drainage Basin Development 67

investigation (Martin and Church, 2004; Tucker and Hancock, ranging from an individual hillslope or drainage basin and up
2010). The historical nature of development of drainage basins to a tectonic unit (see Martin and Church (2004) for an ex-
contributes to difficulties in defining effective tests of model tended discussion of this topic). Temporal and spatial scales
results. Because of historical contingencies (be they geologic, should be chosen in relation to one another, with temporal
climatic, or others) associated with the development of spe- scales starting as low as 102 years for individual hillslopes and
cific landscapes, it is generally not possible to replicate the going up to 107 þ years for tectonic units. Models should
development of any particular landscape. Therefore, the test- include relevant geomorphic equations for processes oper-
ing of such results needs to find means of identifying ating in the region and for the spatial and temporal framework
acceptable performance of models, while at the same time being studied. Most models include various sediment-trans-
relaxing any expectation that specific historical sequences can port equations for hillslope and fluvial processes (the reader is
be replicated exactly in model runs. referred to the list of numerical models provided in Tucker
and Hancock (2010)). The inclusion of glacial processes
within a model is particularly challenging, given the more
14.6.2 Defining the Numerical Modeling Exercise limited knowledge about glacial transport compared to geo-
morphic transport processes for hillslope and fluvial systems.
Two stages of development of a drainage basin can be differ- Finally, tectonic processes may or may not be included in the
entiated: (1) drainage initiation and (2) subsequent drainage- model, depending on their relevance to the problem being
basin development (see Martin and Church (2004) for studied. Tectonics can be included in a model of drainage-
additional discussion of this topic). In either case, a model basin development as a simple constant uplift rate across the
of drainage-basin development requires an initial surface landscape surface (Tucker and Bras, 1998) or, alternatively,
upon which the chosen processes operate. The definition of more detailed formulations of geodynamic processes may be
this initial surface results in studies that may address very included (e.g., Willett, 1999). A discussion of these geody-
different questions and stages in development of the drainage namic formulations is outside the scope of this review, and the
basin. A central concern in many studies is the determination reader is referred to relevant references in the literature for
of where and when channels begin to emerge on the land- further discussion of this matter (e.g., Willett, 1999; Braun and
scape surface. In such cases, a key goal is to develop landscapes Sambridge, 1997; Miller and Slingerland, 2006).
with realistic channel-network properties, such as drainage Models of development of drainage basins can be utilized
density (i.e., Ahnert, 1976; Willgoose et al., 1991a, 1991b; to address a seemingly endless array of research questions; it is
Tucker and Bras, 1998). Such models require the adoption of not possible within this methodological review to provide an
mechanistic equations that include critical thresholds for in- exhaustive discussion about the many theoretical applications
cision by water flow and the processes involved in the devel- that have taken place or that could take place. Instead, a few
opment of the emerging channel network. These equations are examples are given to illustrate the types of questions related
applied to some initial landscape, which is commonly repre- to drainage-basin change that have been addressed using a
sented by a planar or fractal surface (Willgoose et al., 1991a, numerical-modeling methodology. Benda and Dunne (1997)
1991b; Tucker and Bras, 1998). Drainage initiation in its early used their stochastically driven numerical model of drainage-
stages involves rill development. Because rill development basin development to assess how time series of sediment
operates at relatively small scales, the defined grid-cell size supply, transport, and storage are affected by basin size as well
must be appropriate for simulating the operation of the rele- as climatic, topographic, and/or geotechnical controls. Tucker
vant processes for rill development. A relatively detailed and his collaborators have published extensively using their
knowledge of hydrological variables and substrate properties model of drainage-basin development, CHILD (e.g., Tucker
may be necessary for the solution of such equations. and Bras, 1998; Tucker et al., 2001; Bogaart et al., 2003; Attal
As the channel system enlarges, drainage-basin develop- et al., 2008). Tucker and Bras (1998), for example, used
ment may enter another stage, that of ‘subsequent drainage- CHILD to examine how different combinations of geo-
basin development’ (see Martin and Church (2004) for further morphic processes result in particular drainage densities and
discussion of this concept; see Martin (2005) for an example landscape appearances. Istanbulluoglu and Bras (2005) ex-
of a modeling exercise with an established drainage system as tended the application of CHILD to examine the effects of
the initial surface). This term refers to changes in landscape vegetation on geomorphic activity and resultant topographic
morphology that occur after the fluvial system has developed characteristics, including drainage density. Martin (2005) used
into an integrated (and perhaps more stable) drainage net- LandMod to examine how wildfire-induced changes to hill-
work. Instead of focusing on rill incision and initial devel- slope transport affect long-term sediment routing regimes in
opment of the channel network, attention is directed at small drainage basins. Finally, geomorphic processes have
examining how the landscape continues to develop after the been combined with tectonic processes in a number of long-
drainage network has completed its initial period of incision term models of landscape development to examine various
and extension. Modeling studies may consider the stages of questions pertaining to their interactions. For example, Tucker
drainage initiation and/or subsequent development of the and Slingerland (1994) examined the geomorphic conditions
drainage basin. necessary to drive prolonged retreat of rift-generated escarp-
The researcher must define clearly the temporal and spatial ments. Willett (1999) used a coupled model of tectonics and
scales of interest for the particular questions being asked in a geomorphology to examine the impacts of regional climate,
modeling study. A significant range of possible spatial scales dominant wind directions, and orographically enhanced pre-
exist for the investigation of drainage-basin development, cipitation on the asymmetry of mountain ranges.
68 Methods in Geomorphology: Numerical Modeling of Drainage Basin Development

14.6.3 Geomorphic Process Equations coefficient may also be adjusted accordingly to account for
nonlinearity of the creep process. In steeper terrain, it may be
14.6.3.1 Introduction the case that creep rates fall within the margin of error of other
processes that operate at much greater rates, such as land-
Geomorphic equations are a core component of drainage-
sliding; therefore, it is not always necessary to explicitly
basin development models. Such equations may include
include this process for certain landscapes (Martin, 2000).
sediment-transport formulae for a range of possible processes,
Soil erosion by overland flow may be an important process
and may also include weathering processes. The significance of
in certain environments, such as semi-arid landscapes or some
weathering is that it governs the rate of bedrock conversion to
post-wildfire settings. Overland flow and rill erosion are
regolith, which provides sediment supply for the model runs.
typically modeled using some form of an equation containing
Depending on the specification of the geomorphic equations,
a critical threshold for transport (Tucker and Hancock, 2010):
hydrological information may be required in the model;
hydrological variables, in turn, may require climatic inputs. qs ¼ ks ðkt qm Sn  fÞb ½3
Climatic and hydrological information may be input directly
into the model, in the form of precipitation or discharge
where qs is the volumetric sediment-transport rate (L3 T1 L1),
values. Alternatively, the numerical model may include vari-
q the specific surface water discharge (L3 L2 T1 or L T1), S
ous equations to obtain suitable values for the hydrological
the slope gradient (tangent slope), f an erosion threshold
variables.
(depending on the particular application, this may have units
of shear stress or stream power), and, finally, ks, kt, m, n, and b
14.6.3.2 Hillslope Processes constants with values defined for a particular study. Note that
this is not a dimensionally balanced formula and contains
Weathering equations are utilized in models to simulate the
several empirical coefficients and exponents. Some researchers
conversion of bedrock into regolith. Models that consider
may substitute specific discharge with contributing area (see
sediment supply generally monitor the underlying bedrock
Howard, 1997), while others may relate surface-water dis-
surface height for each grid cell, in addition to the height of
charge to rainfall conditions (Tucker and Bras, 1998).
the land surface (note that these two values will be equal if
Rapid, episodic mass movements are an important con-
there is no regolith cover). Dating by cosmogenic isotopes
tributor to landscape change in mountainous environments.
now allows for estimation of bedrock-weathering rates over
Models constructed at the hillslope scale may simulate debris
long temporal scales (e.g., Heimsath et al., 1997, 2000, 2009).
slides using relatively detailed, mechanistically-based equa-
Although the rates reported in such studies are not represen-
tions (see Martin and Church (2004) for a review of such
tative of all locations, they do allow order-of-magnitude
equations). Detailed process equations may not always be
weathering rates to be implemented into numerical models
suitable at larger scales, in large part because a correspond-
of drainage-basin development. These types of bedrock-
ingly detailed knowledge of controlling variables is not feas-
weathering equations have the general form:
ible. Therefore, models constructed at the larger scale of
q hb drainage basins characteristically incorporate some degree of
¼ ro exp½dðhs  hb Þ ½1
qt generalization in equations for debris slides. Debris slides can
be modeled as a diffusive-type process in such cases, based on
where hb is height of the bedrock/regolith interface (L), hs the the assumption that these kinds of mass movements will affect
height of the land surface (L), t the time (T), ro the rate of nearly every location on the hillslope above some threshold
weathering of exposed surface rock (L T1), and d the decay gradient given enough time. Suitable formulations of the dif-
constant (L1) (note that throughout this chapter, length units fusion equation to simulate debris slides may incorporate a
are denoted by L, T is a time unit and M is a mass unit). If threshold gradient for failure and include nonlinear func-
sediment supply is depleted, then despite the potential for a tional dependencies for the diffusion coefficient, k (e.g.,
certain amount of sediment to be transported (i.e., the Martin and Church, 1997; Martin, 2005). Deep-seated, bed-
amount that would be transported, if enough sediment was rock landsliding has been approached in different ways. As an
available), the sediment-transport rates must be adjusted example, Dadson and Church (2005) utilized a stochastic
accordingly. term for the initiation of large landslides in their model (an
Some processes, such as soil creep, may be represented by extension of LandMod, see Martin (1998)). The stated reason
either linear or nonlinear diffusion (e.g., Martin and Church, for their stochastic approach was that the precise location of
1997; Roering et al., 1999, 2001). The linear diffusion equa- large landslide events cannot be determined from variables
tion is a slope-dependent transport equation that incorporates that are resolvable in their model. Certain locations in the
a coefficient, k, which encapsulates other factors, such as cli- landscape were assigned a greater probability for failure based
mate and soil properties: on slope, relief, and elevation.
 2 
qh q h q 2h
¼k þ ½2 14.6.3.3 Channel Network Processes
qt q x2 q y 2
Herein, the channel network is defined as including the main
where h is the height of the land surface (L), t the time (T), k fluvial system (low- and high-gradient streams) and low-order,
the diffusion coefficient (L2 T1), and x and y (L) the spatial debris-flow reaches. A well-developed body of theory exists for
coordinates for the landscape surface grid cells. The diffusion bed-load transport in lower-gradient streams/rivers (generally
 Methods in Geomorphology: Numerical Modeling of Drainage Basin Development 69

defined as less than B1.51). That being said, a significant error In this latter model, certain parameters based on channel
term often exists for predictions based on bed-load transport gradients and angles of tributary junctions dictate the lo-
formulae (Gomez and Church, 1989), and sediment-transport cations where deposition of debris flows occurs.
prediction remains fraught with difficulties despite the
notable research effort that has been focused on this topic.
These types of equations require information about discharge 14.6.3.4 Other Processes
values, so some hydrological inputs or subroutines are re- Processes in addition to those listed above can be incorpor-
quired. A large number of bed-load transport formulae are ated into models of drainage-basin development. For example,
available to choose from (see Gomez and Church (1989) for a biologically-driven processes (for example, animal activity or
listing of the most widely used equations). Although re- tree throw) can be included in models (e.g., Gabet et al., 2003;
searchers may have particular preferences, most of these Gallaway et al., 2009). These processes might be modeled
equations adopt a similar general form. In particular, some using some type of gradient-driven diffusion equation, al-
defined value of critical shear stress or critical stream power though calibration for representation of these particular pro-
must be exceeded for transport to be initiated. A typical form cesses would be necessary. Alternatively, more detailed process
of such equations is specification might be suitable in particular circumstances.
ib ¼ Iðt  tc Þm ½4
14.6.3.5 Changes in Climate and Hydrology over Time
where ib is the unit bed-load transport rate (M L1 T1), t the
shear stress (M L1 T2), tc the critical shear stress for en- It may be the case that the boundary conditions (climatic and/
trainment that depends on factors such as grain size and hy- or hydrological regimes, etc.) may need to change throughout
draulics (M L1 T2), and m is an exponent (see Gomez and the course of a model run. For example, climatic and hydro-
Church (1989) for the detailed form of numerous bed-load logical regime shifts may occur, resulting in wetter or drier
transport equations). Although form resistance (due to various periods. If precipitation and hydrology are incorporated ex-
bed structures and other forms) may also play a significant plicitly into model, then they will need to be adjusted as ap-
role in bed-load transport processes, such resistance terms propriate to reflect changes in these variables. Alternatively,
have not typically been included in the classic bed-load the strength of coefficients/exponents that encapsulate such
transport formulae generally utilized in drainage-basin mod- terms (e.g., the diffusion coefficient) may require modification
els. Although suspended-load transport may show some cor- to adequately reflect climatic and hydrologic shifts.
relation with discharge values of a river, it is in large part often
a supply-dependent phenomenon. The latter attribute ac-
counts for the wide margin of error in relying on hydraulic 14.6.4 Constructing and Running the Model
information, such as discharge values, for its estimation. The
finer components of suspended load are characteristically Once the modeling objectives have been defined, and the
transported directly through the channel system, so they do appropriate equations have been identified for inclusion in
not always play a major role in morphological change of the the model, then it is a question of constructing (or revising)
river (see discussion in Martin and Church (1995)). the numerical code to solve the equations. Although there are
Recent decades have seen a significant increase in under- a number of programming languages available, the choice of
standing of high-gradient fluvial processes. Such streams have programming language (e.g., C, Fortran and others) to a large
very different morphological characteristics than their lower- extent depends on the background and preferences of the
gradient counterparts; in particular, form resistance may have programmer and/or user. Some researchers undertake their
a particularly significant role in the sediment-transport theory own programming, whereas others may hire computer pro-
for such streams (e.g., Lee and Ferguson, 2002; Wilcox and grammers to write the code and/or maintain and upgrade the
Wohl, 2006). Martin (1998) utilized an adapted bed-load code as necessary. In addition to familiarity with the pro-
transport algorithm for these high-gradient streams in her gramming language of choice, a solid grounding in the so-
numerical model of drainage-basin development, although phisticated mathematical techniques necessary for numerical
recent advances in research have set the stage for improved solution of differential equations may be required. The model
incorporation of high-gradient stream processes into models surface may consist of regularly spaced grid cells, whereas
of drainage-basin development (e.g., Zimmerman and other models make use of triangulated irregular network
Church, 2001; Lenzi et al., 2006). Nevertheless, many of the (TINs). The latter allows spacing of grid cells to be higher or
existing models do not include a specific set of rules for these lower in different parts of the model domain (Clevis et al.,
higher-gradient streams to distinguish them from their lower- 2006). TINs can be an effective grid surface if some locations
gradient counterparts. require a higher grid-cell density relative to other locations in
Several researchers have included debris-flow processes the model landscape. To ensure numerical stability in the
within the channel component of their models. For example, solution of differential equations contained within the model,
Benda and Dunne (1997) used stochastic probabilities to criteria must be met regarding the relative sizes of grid-cell size
initiate movement by debris flow processes of material col- (space steps) and time steps. Depending on the complexity of
lected in upstream hollows Martin (1998) used a probabilistic a model and the spatial and temporal scales being investi-
formulation for initiation of debris flows, based on field- gated, computing power needs may vary enormously. In many
derived data for the Queen Charlotte Islands, British Columbia. cases, the programming code can be run on PCs or through
70 Methods in Geomorphology: Numerical Modeling of Drainage Basin Development

typical mainframe systems available at many research insti- questions about development of drainage basins. As an ex-
tutions and government agencies. In other cases, the modeler ample, Tucker and Bras (1998) used their model of drainage
may have to utilize supercomputing facilities on shared development primarily as an exploratory tool to examine the
networks. joint operation of different combinations of hillslope
Numerical models of development of drainage basins are processes.
associated with a series of input files and output files, which Whether specific landscapes are being modeled, or if the
are the primary means for user interaction with the program. model is used for exploratory purposes, it is still necessary to
The items in input files include fundamental information, know if the model is operating in a way that is at least rea-
such as initial landscape surface (typically a DEM), dimen- sonable, and is also producing landscapes that are reasonable.
sions of the model domain (number of grid cells in each The definition of what constitutes a ‘reasonable’ performance
direction), grid-cell size, time step, and total run time of the remains a critical research question (Tucker and Hancock,
model. In addition, various adjustable variables and co- 2010). Methodical study of this issue has been limited, and no
efficients in the numerical model are generally included in consensus has yet been reached about what constitutes an
input files; examples include precipitation rates, uplift rates, adequate test of model performance. Tucker and Hancock
choices of particular processes/transport equations to include, (2010) discussed the possible types of landscapes and data
coefficients, and exponents for various geomorphic process that can be compared against numerical model results: rapidly
equations. Output files may be individualized according to the evolving landscapes (e.g., postmining landscapes in the real
particular needs of the model user; these may consist of the world); long-term sediment fluxes; scaled experimental land-
new landscape surface (typically in a DEM format), height of scape simulations; and what can be termed ‘natural experi-
the regolith/bedrock boundary, and amounts of erosion/ ments’ in the field (cases where initial conditions and forcing
transport/deposition. Any or all of the above may be output at conditions are relatively known). Qualitative comparisons can
stated times (including the final run time for the model) and be made between the observed values and the modeled
for specific locations or across the entire grid. The modeler landscape, and it is possible that qualitative consistency may
may include options for interactive visualization during a be achieved (e.g., Howard, 1997). Some studies have com-
model run and/or results may be imported into visualization pared statistical properties, such as hypsometry, between
and statistical packages for further analysis. simulated and real landscapes (e.g., Anderson, 1994).
Willgoose and his collaborators have made perhaps the most
notable advances in this area of research, using landscape
14.6.5 Model Confirmation statistics and experimental simulators to test model results
(e.g., Willgoose, 1994; Willgoose and Hancock, 1998;
Model confirmation is established by examining the ability of Hancock and Willgoose, 2001). Tucker and Hancock (2010)
a model to match prediction (model outcome) with obser- highlighted that a critical remaining issue in numerical mod-
vation in some formally defined manner. There are several eling of landscape change is to define statistical tests that are
difficulties associated with models of drainage-basin devel- capable of adequately discriminating between observed and
opment that make this a far from trivial exercise: (1) To modeled landscapes.
achieve model outcomes in which we have some degree of
confidence, it is desirable to have some confidence in the
chosen initial conditions and other contingencies (climatic, 14.6.6 Final Comments
tectonic, and geological) incorporated into the model run.
Given the large temporal scales of drainage-basin develop- Numerical modeling is now a well-established methodology
ment, knowledge of the initial conditions and historical for the study of development of drainage basins over medium
contingencies is most often relatively unknown and at the very to large spatial and temporal scales. A particular challenge that
best fraught with significant uncertainty. Therefore, a relatively confronts modelers of longer-term drainage-basin change is
wide margin of ‘error’ generally has to be accepted in model that process operation and historical contingencies both affect
results when attempting to test model performance. (2) Set- the course of landscape change, with each presenting its own
ting aside the uncertainties surrounding point (1) above, no set of unique issues. A number of possible sediment-transport
widely accepted standard exists for what might constitute a equations exist to represent geomorphic processes in numer-
robust comparison between a modeled landscape and its ical-modeling studies; basic forms of such equations are out-
observable counterpart. lined in this chapter. The advent of advanced and modern
It is difficult to disentangle the reasons for differences be- techniques of dating has allowed for better estimation of
tween a model outcome and its observable prototype. For longer-term process rates to aid in the calibration of these
example, the model equations and/or their calibration could geomorphic equations. Despite advances in understanding of
themselves be at fault; inappropriate initial conditions may geomorphic process operation at medium to large scales,
have been chosen; and/or hydroclimatic and other forcings further research is required to further optimize the specifi-
may be inadequate. Given the multitude of possible un- cation and calibration of the relevant equations. Historical
certainties surrounding the modeling of specific landscapes, contingencies associated with particular landscapes are gen-
some researchers instead focus on investigating general prin- erally difficult to incorporate adequately into model runs, thus
ciples of development of drainage basins (but which may be representing a very real limitation of landscape modeling.
based loosely on particular, real-world landscapes). In such Nonetheless, a numerical-modeling approach for the study of
cases, the models may be used to explore a series of ‘what-if’ development of drainage basins represents a powerful
 Methods in Geomorphology: Numerical Modeling of Drainage Basin Development 71

methodology, allowing for improved understanding of general Martin, Y., 1998. Modelling geomorphology in landscape evolution. Ph.D. Thesis,
principles associated with landscape change and for investi- University of British Columbia, Vancouver, BC, Canada, Unpublished.
gating connections among controlling factors, process oper- Martin, Y., 2000. Modelling hillslope evolution: linear and nonlinear transport
relations. Geomorphology 34, 1–21.
ation, and changes in landscape morphology. Martin, Y., 2005. Wildfire disturbance and shallow landsliding in coastal British
Columbia over millennial time scales: A numerical modelling study. Catena 69,
206–219.
References Martin, Y., Church, M., 1995. Bed-material transport estimated from channel
surveys: Vedder River, British Columbia. Earth Surface Processes and Landforms
20, 347–361.
Ahnert, F., 1976. Brief description of a comprehensive three-dimensional process-
Martin, Y., Church, M., 1997. Diffusion in landscape development models: on the
response model of landform development. Zeitschrift für Geomorphologie 25,
29–49. nature of basic transport relations. Earth Surface Processes and Landforms 22,
Anderson, R.S., 1994. Evolution of the Santa Cruz Mountains, California, through 273–279.
tectonic growth and geomorphic decay. Journal of Geophysical Research 99, Martin, Y., Church, M., 2004. Numerical modelling of landscape evolution:
20161–20179. geomorphological perspectives. Progress in Physical Geography 28, 317–339. h
Attal, M., Tucker, G.E., Whittaker, A.C., Cowie, P.A., Roberts, G.P., 2008. Modeling ttp://dx.doi.org/10.1191/0309133304pp412ra.
fluvial incision and transient landscape evolution: influence of dynamic channel Miller, S.R., Slingerland, R.L., 2006. Topographic advection on fault-bend folds:
adjustment. Journal of Geophysical Research – Earth Surface 113, F03013. http: inheritance of valley positions and formation of wind gaps. Geology 34,
//dx.doi.org/10.1029/2007JF000893. 769–772. http://dx.doi.org/10.1130/G22658.1.
Benda, L., Dunne, T., 1997. Stochastic forcing of sediment supply to channel Oreskes, N., Shrader-Frechette, K., Belitz, K., 1994. Verification, validation, and
networks from landsliding and debris flow. Water Resources Research 33, confirmation of numerical models in the earth sciences. Science 263(5147),
2849–2863. 641–646. http://dx.doi.org/10.1126/science.263.5147.641.
Bogaart, P.W., Tucker, G.E., de Vries, J.J., 2003. Channel network morphology and Reinhardt, L.J., Bishop, P., Hoey, T.B., Dempster, T.J., Sanderson, D.C.W., 2007.
sediment dynamics under alternating periglacial and temperate regimes: a Quantification of the transient response to base-level fall in a small mountain
numerical simulation study. Geomorphology 54, 257–277. catchment: Sierra Nevada, southern Spain. Journal of Geophysical Research 112,
Braun, J., Sambridge, M., 1997. Modelling landscape evolution on geologic time F03S05 (20 pp.). http://dx.doi.org/10.1029/2006JF000524
scales: a new method based on irregular spatial discretization. Basin Research Repka, J., Anderson, R.S., Finkel, R.C., 1997. Cosmogenic dating of fluvial terraces,
9, 27–52. Fremont River, Utah. Earth and Planetary Science Letters 152, 59–73. http://dx.d
Clevis, Q., Tucker, G.E., Lock, G., Lancaster, S.T., Gasparini, N.M., Desitter, A., oi.org/10.1016/S0012-821X(97)00149-0.
2006. A simple algorithm for the mapping of TIN data onto a static grid: Roering, J., Kirchner, J., Dietrich, W., 1999. Evidence for nonlinear, diffusive
applied to the stratigraphic simulation of river meander deposits. Computers and sediment transport on hillslopes and implications for landscape morphology.
Geosciences 32, 749–766. Water Resources Research 25, 853–870.
Dadson, S.J., Church, M.A., 2005. Post-glacial topographic evolution of glaciated Roering, J., Kirchner, J., Dietrich, W., 2001. Experimental hillslope evolution by
valleys: a stochastic landscape evolution model. Earth Surface Processes and
nonlinear creep and landsliding. Geology 29, 143–146.
Landforms 30, 1387–1403.
Rosenbloom, N.A., Anderson, R.S., 1994. Evolution of the marine terraced
Gabet, E.J., Reichman, O.J., Seabloom, E.W., 2003. The effects of bioturbation on
landscape, Santa Cruz, California. Journal of Geophysical Research 99,
soil processes and sediment transport. Annual Review of Earth and Planetary
14013–14030.
Sciences 31, 249–273.
Tucker, G.E., Bras, R.L., 1998. Hillslope processes, drainage density, and landscape
Gallaway, J., Martin, Y., Johnson, E., 2009. Sediment transport due to tree root
throw: integrating tree population dynamics, wildfire and geomorphic response. morphology. Water Resources Research 34, 2751–2764.
Earth Surface Processes and Landforms 34, 1255–1269. http://dx.doi.org/ Tucker, G.E., Hancock, G.R., 2010. Modelling landscape evolution. Earth Surface
10.1002/esp.1813. Processes and Landforms 35, 28–50.
Gomez, B., Church, M., 1989. An assessment of bed load sediment transport Tucker, G., Slingerland, R., 1994. Erosional dynamics, flexural isostasy, and long-
formulae for gravel bed rivers. Water Resources Research 25, 1161–1186. lived escarpments. Journal of Geophysical Research 99, 12229–12243.
Hancock, G., Willgoose, G., 2001. Use of a landscape simulator in the validation of Tucker, G.E., Lancaster, S.T., Gasparini, N.M., Bras, R.L., 2001. The Channel-
the SIBERIA catchment evolution model: declining equilibrium landforms. Water Hillslope Integrated Landscape Development (CHILD) model. In: Harmon, R.S.,
Resources Research 37, 1981–1992. Doe, W.W. (Eds.), Landscape Erosion and Evolution Modeling. Kluwer/Plenum,
Hasbargen, L.E., Paola, C., 2000. Landscape instability in an experimental drainage New York, NY, pp. 349–388.
basin. Geology 28, 1067–1070. Wilcox, A., Wohl, E., 2006. Flow resistance dynamics in step–pool channels: 1.
Heimsath, A.M., Chappell, J.C., Dietrich, W.E., Nishiizumi, K., Finkel, R.C., 2000. Large woody debris and controls on total resistance. Water Resources Research
Soil production on a retreating escarpment in southeastern Australia. Geology 42, W05418.
28(9), 787–790. Willett, S.D., 1999. Orogeny and orography: the effects of erosion on the structure
Heimsath, A.M., Dietrich, W.E., Nishiizumi, K., Finkel, R.C., 1997. The soil of mountain belts. Journal of Geophysical Research 104(B12), 28957–28981.
production function and landscape equilibrium. Nature 388, 358–361. Willett, S.D., Fisher, D., Fuller, C., En-Chao, Y., Chia-Yu, L., 2003. Erosion rates
Heimsath, A.M., Hancock, G.R., Fink, D., 2009. The ‘humped’ soil production and orogenic-wedge kinematics in Taiwan inferred from fission-track
function: eroding Arnhem Land, Australia. Earth Surface Processes and thermochronometry. Geology 31(11), 945–948. http://dx.doi.org/10.1130/
Landforms 34, 1674–1684. G19702.1.
Howard, A.D., 1997. Badland morphology and evolution: interpretation using a Willgoose, G., 1994. A statistic for testing the elevation characteristics of landscape
simulation model. Earth Surface Processes and Landforms 22, 211–227. simulation models. Journal of Geophysical Research 99, 13987–13996.
Istanbulluoglu, E., Bras, R.L., 2005. Vegetation-modulated landscape evolution: Willgoose, G., Bras, R.L., Rodriguez-Iturbe, I., 1991a. A coupled channel network
effects of vegetation on landscape processes, drainage density, and topography. growth and hillslope evolution model: 1. Theory. Water Resources Research 27,
Journal of Geophysical Research 110, F02012. http://dx.doi.org/10.1029/ 1671–1684.
2004JF000249. Willgoose, G., Bras, R.L., Rodriguez-Iturbe, I., 1991b. A coupled channel network
Kennedy, B.A., 2006. Inventing the Earth: Ideas on Landscape Development Since
growth and hillslope evolution model: 2. Nondimensionalization and
1740. Blackwell, Oxford, 176 pp.
applications. Water Resources Research 27, 1685–1696.
Lee, A.J., Ferguson, R.I., 2002. Velocity and flow resistance in step–pool streams.
Willgoose, G., Hancock, G., 1998. Hypsometric curves as an indicator of catchment
Geomorphology 46, 59–71.
form: revisited. Earth Surface Processes and Landforms 23, 611–623.
Lenzi, M.A., Mao, L., Comiti, F., 2006. Effective discharge for sediment transport in
Zimmerman, A., Church, M., 2001. Channel morphology, gradient profiles and bed
a mountain river: computational approaches and geomorphic effectiveness.
Journal of Hydrology 326, 257–276. stresses during flood in a step–pool channel. Geomorphology 40, 311–327.
72 Methods in Geomorphology: Numerical Modeling of Drainage Basin Development

Biographical Sketch

Yvonne Martin is an associate professor in the Department of Geography, and an adjunct professor in the
Department of Geoscience, University of Calgary. She has been appointed as an Institute Fellow of the Bio-
geosciences Institute at the University of Calgary. Dr. Martin completed her MSc and PhD degrees at the University
of British Columbia. Her research focuses on the numerical modeling of drainage-basin development and the
interactions between geomorphic and ecological processes.
 14.7 Methods in Geomorphology: Investigating River Channel Form
IC Fuller, Massey University, Palmerston North, New Zealand
HE Reid and GJ Brierley, University of Auckland, Auckland, New Zealand
r 2013 Elsevier Inc. All rights reserved.

14.7.1 Introduction 73
14.7.1.1 Importance of Context 74
14.7.2 History/Background 76
14.7.3 Methods 78
14.7.3.1 Planform 78
14.7.3.2 Cross Section 80
14.7.3.3 Long Profile 82
14.7.3.4 3D Form 83
14.7.4 Case Studies 84
14.7.4.1 Case Study 1: River Styles 84
14.7.4.2 Case Study 2: Waipaoa 84
14.7.5 Future Work and Direction 87
14.7.6 Conclusions 88
References 88

Glossary global positioning system (GPS) base station and Rover unit
Digital elevation model (DEM) A digitally constructed to map topography on the ground using positions derived
surface of topography generated by interpolation of a digital from satellites. This setup permits rapid data acquisition
surface between data points derived from ground or aerial using 1-s occupation time per observation and real-time
survey. coordinate calculation using on-the-fly algorithms.
Geographical information system (GIS) A geographical Terrestrial laser scanning (TLS) Ground-based laser
information system such as ArcGISs is used to analyze and mapping that provides a rapid means of acquiring
interpret aerial photos, maps, and DEMs derived from topographic data in ground surface surveys at a small spatial
spatial data. scale, very high resolution, typically hundreds of points per
Light detection and ranging (LiDAR) An airborne laser- square meter and high degree of accuracy (mm).
mapping system, providing a rapid means of acquiring Total station A combined theodolite and electromagnetic
topographic data from an aircraft, covering a large spatial distance measurement (EDM) device used in conventional
scale with point densities typically at a point per square meter. ground survey measuring topography using angles and
Positional accuracy of points is in the order of c. 10–20 cm. distances. Very high precision (mm), used in conjunction
Photogrammetry Quantification of landsurface with GPS survey where satellite coverage, is limited
parameters using rectified (dimensionally accurate and (e.g., narrow valleys or beneath vegetation canopy).
precise) aerial photography. Positional accuracy tends to be Unmanned aerial vehicle (UAV) Unmanned aerial
measured in meters, but submeter accuracy is now feasible. vehicle to which a camera or sensor may be mounted to
Real-time kinematic differential global positioning acquire airborne imagery without the need for a
system (RTK-dGPS) survey A precise survey tool using a conventional aircraft.

Abstract

No single approach effectively synthesizes analysis of river channel form. The diversity of river channels renders research
aims and site specificity critical to any investigation, framed within an open-ended approach to enquiry. This inevitably
entails a trade-off between resolution, scale, and time: information collected must reflect the specific needs of the task at
hand at a given site. In this chapter, investigative techniques used to assess river channel form are appraised in relation to
channel planform, cross section, long profile, and three-dimensional morphology. Analysis of three-dimensional forms
provides the most rigorous approach to link form adjustment to channel processes, enabling predictive modeling of river
behavior.

Fuller, I.C., Reid, H.E., Brierley, G.J., 2013. Methods in geomorphology:

investigating river channel form. In: Shroder, J. (Editor in Chief), Switzer,
A.D., Kennedy, D.M. (Eds.), Treatise on Geomorphology. Academic Press,
San Diego, CA, vol. 14, Methods in Geomorphology, pp. 73–91.

Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00374-2 73

74 Methods in Geomorphology: Investigating River Channel Form

14.7.1 Introduction distribution of geomorphic units, and types of lateral

adjustment.
2. Channel cross section. A cross-cutting 2D slice of the chan-
In a two-dimensional (2D) planimetric dimension, river channel
nel, used to calculate energy, confinement, and understand
form, or morphology, is manifest as a continuum of channel
controls on shape.
patterns, or planform, which reflects the combined response to
3. Channel long profile. Longitudinal changes down the thal-
multiple variables in any given reach and catchment. Planform is
weg, assessing how slope, energy, and geomorphic units
commonly expressed through differences in sinuosity and
change downstream.
braiding index, although interpretation of differing forms of
4. 3D perspective. Combine planform, cross section, and lon-
lateral adjustment and mobility provides an important context
gitudinal investigations of river form through terrain
to processes that created that morphology (Figure 1). Flow–
modeling procedures.
sediment interactions generate distinctive channel planform
types within particular environmental settings. At the catchment These perspectives should not be disconnected from
scale, this is governed by independent variables, such as relief, broader environmental considerations. Indeed, all investi-
rock type(s), valley slope and morphology (width and shape), gations of channel form should consider the adjacent
climate, and rainfall regime (and associated vegetation inter- floodplain and hillslope environments and processes, and
actions). These factors, in turn, condition the dependent vari- catchment-specific considerations. This chapter does not
ables of sediment and water discharges that modulate or amplify seek to explain variability in channel form; rather, it con-
runoff characteristics and catchment erosion. siders approaches and techniques that can be used to in-
Assessment of channel form can also be performed in two vestigate and record river form. Approaches to analysis,
dimensions across channels (the cross section) and along however, must reflect the specific purpose for which infor-
channels (the long profile). Figure 2 provides a representation mation gathered is to be used.
of these views. 3D analysis using terrain modeling can be
applied to pull together planform, cross section, and long
profile perspectives for a given channel reach. Changes to 14.7.1.1 Importance of Context
channel form over time provide a fourth (temporal) dimen-
Valley setting presents a conceptually simple way to categorize
sion to this topic. In each dimension, channel form represents,
the degree to which imposed boundaries influence a river
or is characterized by, a dynamic assemblage of morphological
(Figure 3; Brierley and Fryirs, 2005). Confined valleys abut the
(or geomorphic) units. These features include an array of
channel margin on both sides for most of the length of a
erosional and depositional forms, the assemblage of which
reach. Partly confined valleys have discontinuous floodplain
varies markedly from bedrock-confined situations where im-
pockets, enabling a degree of lateral movement. Laterally un-
posed (forced) river morphologies are created by erosional
confined valleys have continuous floodplains, where channels
processes through to fully alluvial (self-adjusting) morpholo-
self-adjust within alluvial deposits. Confinement may be a
gies created by a mix of depositional and erosional processes
product of a variety of antecedent controls such as bedrock,
(Brierley and Fryirs, 2005).
river terrace or fan materials, or human-induced attributes
The precise assemblage of morphological units, which ex-
(e.g., channelization within concrete), which determine the
presses channel form in a range of aspects/dimensions in a
extent to which the river can adjust on the valley floor (Fryirs
given reach, responds to and reflects the flow regime, sediment
and Brierley, 2010). In these instances, the channel retains a
supply, and boundary conditions within a catchment. Chan-
memory of conditions set at some time in the past (Brierley,
nel form and patterns are the result of many overlapping
2010), whereby contemporary channel form is partly, or
controls. Hence, understanding of governing controls is often
wholly, forced (i.e., the channel has limited capacity to adjust
difficult and complex. A wide range of controls creates huge
its boundaries). Valley confinement influences the capacity of
diversity of fluvial environments across the globe. The influ-
a channel to adjust its form, concentrating or dissipating en-
ence of differing controls varies with scale, ranging from
ergy in response to energy and sediment loads (Nanson and
broad-scale factors, such as slope and valley confinement,
Huang, 2008). Hence, this provides an important foundation
through to more transient, local-scale factors, such as con-
for choosing the method to assess river channel form.
nectivity to sediment sources, vegetation cover, and the load-
Each study needs to consider the specific circumstances
ing of wood in the channel. Analysis of underlying controls
and the data output required prior to determining what to
upon river channel form and function is a critical consider-
measure and the procedures best suited for this purpose (in
ation in management and rehabilitation planning, aiding ef-
terms of precision, site selection, etc.). Field visits provide
forts to work with prevailing processes to improve river
insight into formative processes, enabling assessment of bed
condition (Brierley et al., 2008).
and bank morphology, geomorphic units, and habitat po-
Investigating river channel form entails research into the
tential. The size and type of landform under investigation
physical dimensions of river channels, that is, size, shape and
dictate the type and resolution of data that can be acquired.
pattern, and the surrounding land they rework (Hart, 1986).
For example, light detection and ranging (LiDAR) imagery is
No single method can be used to capture this extremely broad
good for capturing the morphology of larger-scale features
topic. For the sake of brevity, this chapter discusses four dis-
and landscapes (such as braid plains) relatively quickly.
crete perspectives:
However, it has limited ability to record surface topography
1. Channel planform. Analysis of channel shape from a bird’s- under deep water, or more local-scale morphology. Where
eye perspective, assessing sinuosity, number of channels, smaller scales and higher resolution of data are necessary,
 Methods in Geomorphology: Investigating River Channel Form 75

Step−pool Low sinuosity with Anabranching

Straight
cascade meandering thalweg rivers Flow
direction

Bed-load-
dominated
channels

Boulders,
cobbles

Sin- 1 Sin- 1.19 Sin- 1.57 Sin- 1 Sin- 1.02

BI- 0 BI- 0.31 BI- 2.18 BI- 0 BI- 2.22
Laterally Laterally Laterally Laterally Lateraly
inactive inactive inactive inactive inactive

Straight Wandering channels Braided

Gravel

Sin- 1.11 Sin- 1.16 Sin- 1.12 Sin- 1.09 Sin- 1.25
BI- 0.74 BI- 0.92 BI- 2.77 BI- 3.44 BI- 4.62
Mixed- Moderate lateral Moderate lateral Laterally Laterally Laterally
load- stability stability active active active
dominated
channels Meandering channels Braided channels

Sand
Channel description key
Sin = sinuosity
= L/
L = length of channel
along thalweg
 = length of valley
along axis
1 = straight channel
Sin- 1.28 Sin- 1.55 Sin- 1.26 Sin- 1.05 Sin- 1 Sin- 1
Meandering rivers >1.5
BI- 0 BI- 0.49 BI- 0 BI- 1.11 BI- 0 BI- 4.20
Moderate lateral Laterally Laterally Laterally Laterally Laterally
BI = braided index
stability active active active active inactive = 2(total bar length)
reach length
Straight Anastomosing
Meandering channels >5 intensly braided/divided
channel channel
1 = single thread channel

Fine sand,
silt Key

Suspended- Channel
load-
dominated Gravel bars
channels Sand bars
Sand bedforms
Vegetation
Sin- 1.05 Sin- 1.37 Sin- 2.46 Sin- 1.17 Sin- 1.39
BI- 0 BI- 0 BI- 0 BI- 3.56 BI- 4.03
Laterally Moderate lateral Moderate lateral Laterally Laterally
inactive stability stability inactive inactive

Figure 1 Channel planform continuum of suspended, mixed, and bedload types, based on Schumm (1981, 1985), amended with anabranching
rivers from Nanson and Knighton (1996). Measurements of sinuosity, braided index, and lateral stability are included as channel planform
descriptors. Braided index refers to the Brice (1960) method.

terrestrial laser scanning or creating digital elevation models Therefore, a conscious effort should be made to address the
(DEMs) from differential global positioning system (GPS) trade-off between resolution, scale, and time, so that the in-
survey can provide more comprehensive data sets. However, formation collected reflects the specific needs of the task at
the time it takes to collect and post process data is far greater. hand.
76 Methods in Geomorphology: Investigating River Channel Form

N Man-made
Cross-sectional view
Planform view stopbank
Compound
mid-channel bar Mid-channel bars
Reach 1-Wandering river A
Riffle Pool Pool
Sinuosity- 1.20
Braided index- 2.577
Key
Gravel point bar
Gravel
A Sand
B
Pool Normal flow levels
Bankfull flow level
Levees

Wetland vegetation
Glide Glide
C
Reach 2- Transitional Reach
B
Sinuosity- 1.11
Braided index- 0

Reach 3- Meandering river

Sinuosity- 2.57
Braided index- 0.11
Key
Wetted channel
Long-profile view Unvegetated gravel
370 C Partly vegetated bars/island
Vegetated bars/islands
368
Reach 4- multi-channeled, delta river
366
Sinuosity- 1.20
Elevation (m)

364 Braided index- 0.62

362

360
1 km
358

356 Wandering Transitional

Meandering river
river reach Delta
354
0 2000 4000 6000 8000 10 000 12 000
Distance downstream (m)

Figure 2 Components of river channel form across the wandering, meandering threshold and into the delta along the Tongariro River, New
Zealand. This represents channel form through the 2D planform view, displaying the spatial distribution in units including relative areas of bars
and wetted channel. The cross-sectional view illustrates differences in heights for those units and the long-profile view shows how slopes, and,
as a result, sediment connectivity change downstream for each river type. Braided index uses the Brice (1960) method.

14.7.2 History/Background research has long concentrated upon the distribution of

rivers of different forms. Davis (1899, 1909) conceptualized
Classifying channel form provides a simple technique for a qualitative ‘cycle of erosion’ that differentiated among
analyzing river systems. Rivers can be classified into youthful, mature, and old-age landscapes, and associated
groups based on similar characteristics and governing pro- stages of hillslope and river adjustments. Gilbert (1914)
cesses (Kondolf et al., 2003). Increases in system-specific provided detailed quantitative measurements of channel
understanding, however, have markedly enhanced the devel- forms and small-scale processes. Much of our subsequent
opment of empirically based laws and understanding. understanding and measurement of fluvial geomorphology
Historically, river courses have been an important con- has been influenced by these two pioneers (Simon and
sideration in the earliest maps, providing the basis for political Castro, 2003). Small-scale quantitative measurements
and property boundaries and communications links. Fluvial provide inferences to advance understanding of large-scale
 Methods in Geomorphology: Investigating River Channel Form 77

Valley
confinement

Confined Partly confined Unconfined

valley setting valley setting valley setting

Channel
planform

No channel
Straight Low Meandering Single Multiple Braided
discontinuous
sinuosity channel channels channel
watercourse
Sinuosity Number of channels

Valley Valley
Valley
margins margins Deadwater
Intact valley margins
Pool
fill covered with
Step Riffle Lateral
Floodplain native vegetation
bar
Geomorphic Pool Run
pocket

units Step-run
sequence Pool
Glides
(examples of for Pool
Lateral
Riffle bar 2m
specific river styles) Bedrock
Recirculating
Run
flow 2m
outcrop 1m
Run1 m

Confined, Partly confined, Unconfined, Intact valley

low-sinuosity, low-sinuosity, low-sinuosity, fill
fine-bed river gravel-bed river tidal influenced river

Bed material
texture

Boulders Cobbles Gravels Sand Fine-grained material

Figure 3 Variables that describe channel form attributes as used to classify rivers in the River Styles Framework. This open-ended classification
structure means that rivers are classified based on attributes observed in the field, rather than being pigeonholed into existing river types. Valley
setting determines the ability of a system to adjust its boundaries. Channel planform and geomorphic units are a function of the energy regime
(e.g., sinuosity as the adjustment of slope and riffles as dissipational units) and sediment-transport storage (e.g., number of channels and extent
of bar features). Bed material indicates the size of sediment the reach is competent to transport. This provides a form-based foundation to
assess river processes acting upon the system.

processes, whereas larger-scale qualitative assessments provide Recent classification systems have increased the level of
inferences to advance understanding of smaller-scale pro- detail and the link to the processes that underlie the form.
cesses. Small-scale process and large-scale form are thus in- Schumm (1963) differentiated among bed-load, mixed-load,
tricately linked. and suspended-load channels using % silt–clay (and therefore
In the late 1950s and 1960s, the push toward process bed and bank cohesivity) in the channel boundary. Mont-
geomorphology, which underpins so much of modern-day gomery and Buffington (1997) developed a process-based
understanding of river channel process and form, began. This classification for seven river types in mountain drainage
was pioneered in the USA by Leopold and Wolman (1957) basins. Plots of grain size, slope, bankfull shear stress,
(Leopold et al., 1964), as well as Schumm (1960, 1963), Brice roughness (Manning’s n), and drainage area were used to
(1964), and Dury (1964). The work of these pioneers, based provide a quantitative underpinning with which to compare
on observations of stream channels in the USA working out of rivers of different forms. These river types were then displayed
or in conjunction with the US Geological Survey (USGS), along a continuum from supply to transport-limited con-
paved the way for further progress in linking process and form ditions, linking the form to the processes creating them.
in the 1970s by seminal work done in the UK by researchers Rosgen (1994) classified 41 types of river form in terms of
such as Richards (1976) and Lewin (1976) and in North entrenchment ratios, bed material, sinuosity, width–depth
America by Hickin (1974) and Hickin and Nanson (1975), as ratios, and slope. Figure 4 illustrates characteristics of nine
well as Schumm (Schumm and Khan, 1972; Schumm, 1977). examples using planform, cross sections, and longitudinal
A profusion of subsequent process-based analyses of river profiles to differentiate form. This simplistic, cookbook ap-
channel form built upon these foundations (e.g., Petts, 1979; proach to river analysis lacks insight into the processes driving
Hooke, 1980; Hey et al., 1982). river form, acting to pigeonhole rivers into preexisting groups
78 Methods in Geomorphology: Investigating River Channel Form

Aa+ Bankfull stage

Dominant >10% A Flood-prone area

slope B
range 4−10% C D DA E
2−4% F G
<2% <4% <0.5% <2% <2%
<2−4%

Cross
section

Plan
view

Stream
Aa+ A B C D DA E F G
types

Figure 4 Longitudinal, cross-sectional, and planform views of major channel types in Rosgen’s (1994) classification scheme. This scheme
measures planform variables and then places each reach into a predefined classification category. Reproduced from Figure 7.27 in Thorne, C.R.,
Hey, R.D., Newson, M.D., 1997. Applied Fluvial Geomorphology for River Engineering and Management, Wiley, Chichester.

and failing to appreciate the diversity in processes and controls 14.7.3 Methods
that may create similar forms (see Simon et al., 2007; Lave,
2009). Indeed, Goodwin (1999) argued that classification is a This section provides a brief overview of some commonly used
‘Neanderthal necessity’ and considered process to be more techniques to investigate channel planform, cross section,
important before classifying stream character (see Chapter long profile, and 3D form. These are summarized in Table 1.
9.36). The River Styles Framework (Brierley and Fryirs, 2005;
Figure 3) applies an open-ended set of procedures to classify
rivers in terms of their character and behavior based upon
14.7.3.1 Planform
process–form relationships (see case study 1).
Hydraulic geometry relationships provide an important The planform of a river reflects the quasi-equilibrium form of
mathematical descriptor of channel form, relating at-a-site and a channel in response to concentration/dissipation of energy
downstream changes in width, depth, velocity, slope, and and associated transfer and storage of sediment. If the river is
roughness to changes in discharge (Leopold and Maddock, not competent to transport its load, deposition most com-
1953). Knighton (1998) demonstrated the influence of bank monly forms bars within or at the margins of the active
material attributes upon these relationships, while Hey and channel. Channel planform reflects the assemblage of meso-
Thorne (1986) and Millar (2000) showed how vegetation to macro-scale morphological units within the active channel.
cover affects these relationships. Eaton et al. (2004, 2010a) Measurement of channel planform involves 2D planimetric
went further with this, formulating thresholds as power laws mapping of channel morphology at an appropriate reach
that relate the critical slope associated with a change in scale, where reach is defined by Eyquem (2007) as being a
channel pattern to dimensionless discharge and relative bank geomorphologically homogeneous watercourse. The length
strength. Thus, regime equations should account for differing studied should capture the diversity of morphologic (geo-
boundary conditions. morphic) units that make up the reach – a length of river with
Recent advances in the quantification of river form similar slope, bed and bank material, channel form, and
relate morphological attributes to the driving and resisting bedforms. Representative reaches typically extend to a length
forces that created them (Simon and Castro, 2003). For equivalent to 30 times channel width, where channel width is
example, Church (2006) linked the morphology of alluvial defined as the perpendicular bank-to-bank distance (Simon
rivers to bed material transport relationships using Shields and Castro, 2003). This distance is notably greater than the
number, a nondimensional value of shear stress calculated 5–7 times channel width relationships that equate to pool–
using sediment and fluid densities, gravity, water depth, pool or riffle–riffle spacing (Richards, 1976) and bar spacing
and grain size. These process-based studies have been ex- (Leopold et al., 1964).
tended to include analyses of sediment budgets (e.g., Ham and Channel planform is readily investigated using georectified
Church, 2000), rehabilitation assessments (e.g., Brookes and aerial photography, orthophotos, and/or satellite imagery.
Shields, 1996; Brooks and Brierley, 2004; Brooks et al., 2004; Topographic maps may provide useful information to classify
Raven et al., 2000), and modeling applications (Coulthard broad channel character, although details of discrete mor-
et al., 2007). phological units may not be evident dependent upon map
 Methods in Geomorphology: Investigating River Channel Form 79

Table 1 Summary of methods that analyze form including progression over time, with reference to selected authors that either describe the tool
and its use or present a case study of it in use over time

Method Key authors/years Method applied

Planform de- Brice (1960) Braided index which describes the proportion of bar length to channel length (see Figure 5 for
scription of details on calculating the Braided index).
planform Knighton (1998) Ways of assessing meander geometry including sinuosity, wavelength, and radius of
curvature (see Figure 5 for details on calculating sinuosity).
Planform over time Hooke (2007) Uses aerial photography to plot the distribution of erosion and deposition along the reach.
used to infer Brierley and Fryirs Assessment of how the channel planform has changed, through use of aerial photography
processes (2000) supplemented by field study including floodplain stratigraphy.
Technological Lewin et al. (1977) Stereosketch method for assessing planform. Channel outlines are plotted from old maps and
advances aerial photographs, then viewed in stereoscope allowing tilt and scale adjustments.
Provides a cheap and fast way to compare images.
Leys and Werritty (1999) Describes using ER Mapper, a software package that allows items of planform data to be
spatially referenced using a warping algorithm so that data are in the same coordinate space
and scale. This allows planform data from different times, with different accuracy in
mapping to be compared.
Gilvear and Bryant Provides an overview of aerial photography and other data sets used for remote sensing,
(2003) including a critique of their relative advantages or disadvantages.
Wishart et al. (2008) Provides an example of using Arc GIS to calculate volumes of sediment moved over time
using planform adjustment derived from aerial photographs.
Lejot et al. (2007) High-resolution aerial photography through use of QuickBird and Ikonos satellites proving
submeter accuracy.
Cross section Jowett (1998) Use of hydraulic geometry relationships to describe variation in width, depth, and velocity at a
specific discharge for downstream hydraulic geometry and within a reach for at-a-station
relationships.
Hicks (1991) Assessing Manning’s ‘n’ roughness value for rivers, so that energy expenditure and
resistance can be calculated for a cross section.
Hoyle et al. (2010) HEC-RAS 1-D modeling where cross sections are extracted and used to create a quasi 2D
model which can model flood inundation for events of differing magnitudes.
Fuller et al. (2002) Measuring changes to cross sections to create a sediment budget, based around calculating
areas of erosion and deposition.
Griffiths (1979) Uses mean bed level (a horizontal line with the same proportion of the bed above as below it)
to provide a datum with which changes in scour and deposition and the changes in
sediment storage (the mean of the cross sections at the up and downstream extent of the
reach, multiplied by the reach length) can be compared across time.
Simon and Castro Summary of the tools used to measure cross sections, such as total stations and level lines,
(2003) including some rationale and guidance for which units to include when surveying.
Long-profile uses Knighton (1999) Slope can be derived from long profiles and used to calculate stream power, to express
changes in energy downstream.
Brierley and Fryirs Uses slope and stream power from long profiles to understand the controls upon the
(2005) distribution of rivers with different forms.
Techniques for Flug et al. (1998) Using GPS to map the water surface and combining this with sonar to measure bed elevation.
measuring long This provides information about the depth and size of geomorphic units as well as the slope.
profiles Simon and Castro Provides a good overview of the tools used for measuring long profiles and how the
(2003) information from long profiles can be used.
Carbonneau et al. Uses high-resolution airborne digital imagery to derive grain-size data. This was used to
(2005a) create a longitudinal median grain size profile for an 80-km-long river.
3D Lane, 1998 Uses GIS-based DEMs to provide a 3D representation of channel morphology.
Heritage et al. (1998) Using ground-based photogrammetry to generate DEMs. This processes data from a series of
photographs and uses it to create 3D coordinates for a set feature (e.g., point bar).
Brasington et al. (2000) Using a real-time kinematic GPS to create DEMs so that local-scale changes can be recorded
and volumes of sediment moved calculated.
Fuller et al. (2003a) Comparison of two methods used to record river channel form so that spatial and temporal
changes and volumes eroded deposited calculated. These methods are (1) using a total
station and electronic distance measurement (EDM) to record cross sections and channel
planform to calculate change to cross-sectional volume and (2) digital elevation model
creation from theodolite (EDM) survey which generates x, y, z coordinates.
(Continued )
80 Methods in Geomorphology: Investigating River Channel Form

Table 1 Continued

Method Key authors/years Method applied

Carbonneau et al. High-resolution aerial remote-sensing surveys, specifically LiDAR, create high-resolution
(2005b) DEMs (1–2-m resolution). Analysis of terrain under water is calculated using illumination
variation corrected by feature-based image processing.
Heritage and Using a terrestrial laser scanner (TLS) to create very accurate DEMs (1600 points per m2).
Hetherington (2007)
Hilldale and Raff (2008) LiDAR that uses red and green light which can penetrate the water surface providing accurate
measurements of subaqueous surfaces. Equipment is still expensive and difficult to obtain.

scale and the skill of the cartographer. These data sources are ratio of stream length to straight-line valley length between two
dimensionally accurate and permit quantification of channel points. Braiding index is quantified as 2 times the total bar
planform/morphology (see, e.g., Werritty and Ferguson, 1981; length, divided by the reach length (Brice, 1960). Figure 5
Passmore et al., 1993; Gurnell et al., 1994; Gurnell, 1997; presents examples of how these variables are measured and
Brewer and Lewin, 1998; Nykanen et al., 1998; Lane, 2000; braided index and sinuosity output values for conceptual
Winterbottom, 2000; Wishart et al., 2008). Technological de- reaches. The geometry of meandering channels can be further
velopments in this field have increased the capacity for de- appraised using measures such as wavelength, radius of curva-
tailed and extensive interrogation of channel planform with ture, and amplitude. For example, Hickin and Nanson (1984)
relative ease: in the 1970s, the stereosketch was de rigeur (e.g., showed how the ratio of radius of curvature to channel width
Lewin et al., 1977), now ArcMap geographic information influences rates of lateral channel migration. Just as important
system (GIS) permits on-screen digitization of digital imagery as appraisal of these morphological attributes, however, is in-
run from a desktop PC (e.g., Wishart et al., 2008). Further- terpretation of the manner (and rate) of channel adjustment
more, image analysis in software such as ER Mapper can be (i.e., behavior) in any given reach (see Brierley and Fryirs, 2005).
deployed to automate this process (e.g., Leys and Werritty, Measurement of planform changes over time provides
1999). This reduces positional accuracy errors, typically in the useful information on system dynamics and magnitude and
order of 75 m (Hughes et al., 2006). rates of channel change (e.g., see Hooke, 2007). Assessment of
Gilvear and Bryant (2003) provided a useful overview recent and historic planform change also provides an im-
of the application of remotely sensed data to investigate river portant context for river management (Brierley and Fryirs,
form, with more recent advances reviewed by Marcus and 2000; Fryirs et al., 2009; Spink et al., 2009).
Fonstad (2008) and a broader review provided by Smith and
Pain (2009). As scale is a key issue, data sources must be
appropriate for the purposes to which they are to be
14.7.3.2 Cross Section
used. Satellite imagery has historically been at a lower reso- Channel geometry records the balance of erosional and de-
lution than aerial photography. However, these data are spa- positional processes that shape bed and bank (Brierley and
tially more extensive. As such, they are especially useful when Fryirs, 2005). Channel cross-section size and shape is
investigating large rivers at the landscape scale (e.g., Gupta dependent on discharge and channel boundary conditions
et al., 2002; Boruah et al., 2008). Smaller-scale rivers and (Knighton, 1998). Hydraulic geometry relationships, which
features must be analyzed using high-resolution imagery. This define adjustment of cross-sectional form with discharge at-a-
is better supplied using aerial photography. However, in the station and downstream, have been established (e.g., Hey and
last decade high-resolution satellite imagery has become Thorne, 1986; Jowett, 1998; Wohl and Wilcox, 2005). Eaton
available with the launch of Quick Bird and Ikonos satellites. et al. (2004) derived a theoretical model for predicting equi-
This imagery offers a submeter-resolution image (Lejot et al., librium alluvial channel form that uses discharge, sediment
2007). Nevertheless, it is arguably less flexible than aerial supply (caliber and quantity), and valley gradient as in-
photographs, which can be flown to order. They can be used, dependent governing variables of form. The inherent strength
for example, to capitalize on the immediacy of the aftermath and stability of banks and bed condition the sensitivity of the
of flood events (e.g., Fuller, 2007), or simply flown when the channel to adjustment in both width and depth (Eaton et al.,
weather allows acquisition of cloudless imagery. Furthermore, 2010b).
supply of aerial photographs no longer necessarily requires the The bankfull channel is normally adjusted to flows with
platform of a survey aircraft. Relatively cheap and more flex- recurrence intervals of c. 1.5 years (Leopold et al., 1964).
ible procedures can be used, such as helium balloons or un- However, Williams (1978) found that although this repre-
manned aerial vehicle (UAV) equipment (see respectively sented average recurrence interval of bankfull discharge, this
Vericat et al., 2009; Lejot et al., 2007). Gilvear and Bryant interval could range from 1.01 to 32 years, with 62% of re-
(2003) discussed the relative advantages and disadvantages of currence intervals lying between 1 and 2 years. The cross sec-
aerial photography, airborne multi- and hyper-spectral im- tion measured is most likely to be adjusted to more frequent,
agery, satellite, and space-borne imagery. low-magnitude floods, but the role of larger events should not
Channel planform is typically measured using parameters be overlooked and may result in dramatic changes to channel
such as sinuosity and braiding index. Sinuosity is defined by the shape and form (e.g., Fuller, 2008).
 Methods in Geomorphology: Investigating River Channel Form 81

Sinuosity Sinuosity Sinuosity

1.06 1.51 2.29
Sinuosity = L/
L = Length of the channel along thalweg
>1.5 = Meandering river
 = Length of valley along the axis

Braided index Braided index Braided index

0.63 1.26 4.50

Braided index = 2 (total bar length)/reach length

Bar length
Reach length
1 = Single thread channel, 2−4 = Moderatly braided, >5 = Intensely braided channel
Figure 5 Assessment of planform attributes sinuosity and braided index used to describe river type. Braided index uses the Brice (1960)
method.

Channel cross sections are generally measured to provide Wilcox, 2005), channel geometry can be related to bed and
information for computing a range of channel processes in- bank composition and the wider-scale processes that formed
cluding discharge, stream power, and shear stress (e.g., see them (Figure 6). For example, asymmetrical channels are
Fuller et al., 2002). Channel roughness is commonly assessed formed by meander thalweg shift, with point bars forming
at a cross-section scale, using Manning’s n based on variables on the inside of the bend and erosion on the outside
such as bed material, geomorphic units (e.g., pool depth), and (concave) bank. Coarse, gravelly streams with wide, shallow
vegetation. This is necessary for calculations of resistance and channels and composite or noncohesive banks are subjected
energy expenditure (see Hicks, 1991; Rutherfurd et al., 2000; to greater degrees of lateral reworking than finer-grained
Chow, 1959; Cowan, 1956 for methodologies). Channel channels with cohesive banks. 1D modeling can be used
cross-section data can be used to provide information on to model energy and inundation of floods of differing
the form and dynamics of the channel. In addition to magnitudes using channel cross-section data (e.g., Patro et al.,
hydraulic geometry relationships (see Jowett, 1998; Wohl and 2009).
82 Methods in Geomorphology: Investigating River Channel Form

Compound Asymmetrical

Floodplain

Floodplain Lateral bar Inset feature/ Floodplain Concave

Point bar
bench bank

Symmetrical Irregular

Floodplain Sand sheet Floodplain Bedrock, boulders, gravel

Figure 6 Channel geometry for asymmetrical, symmetrical, compound, irregular, bed, and bank attributes. Geometry reflects the formative
processes and can thus be used to assess channel adjustment. Compound channels have a stepped morphology with the channel being inset
within a macrochannel and can be depositional (as seen here with benches deposited) or erosional with ledges (Brierley and Fryirs, 2005).
Asymmetrical channels are the product of lateral migration, as the channel erodes the concave bank and forms a depositional point bar on the
convex bank. Symmetrical channels have a homogeneous bed and can be a result of either incision or aggradation such as sand sheet
deposition. Irregular channels are commonly a product of the influence of bedrock, coarse material, or vegetative forcing, creating a diverse
structure.

Assessment of lateral river channel adjustment in response banks. Monumented pegs or benchmarks should therefore be
to bank erosion is usefully reviewed by Lawler (1993) and set well back from riverbanks and preferably above flood
Lawler et al. (1997). However, to address cross-sectional form levels. Ideally, cross-section surveys should bisect a represen-
of the channel as a whole, repeat measurement of cross sec- tative array of morphological units as identified from channel
tions from monumented pegs or benchmarks on the banks planform assessments. Spacing of channel cross sections
permits a quantitative assessment of change in channel shape should be scaled according to channel size, as noted for reach
in response to erosion and/or deposition in the active chan- identification. In many instances, documentation of the range
nel. Thus, measurement of channel cross section can be used of channel geometry in a given reach is just as important as
to perform sediment budgets in gravel-bed channels that are detailed measurement of a given cross section.
dominated by bedload movement (Leopold, 1992; Martin Cross sections are most commonly surveyed using survey
and Church, 1995; Brewer and Passmore, 2002; Fuller et al., equipment such as total stations, differential GPS, or auto-
2002; Raven et al., 2009, 2010). Regular measurement of matic levels. Clinometers or stretching a tape across the
channel cross sections is routine in many parts of the world. channel and measuring the height at set distances can be used
Data from these riverbed surveys are used to quantify mean for simple surveys, but the degree of error is relatively high.
bed level (MBL) at each cross section and provide valuable GPS and total station measurement errors are in the order of
data for river management, such as monitoring flood capacity 70.005 m (e.g., Raven et al., 2009). Surveys should be taken
and assessing gravel resource. The MBL represents a horizontal at any breaks in slope/depth, recording the array of units
straight line across the channel, positioned so there is as much found in a reach. Cross-section data can also be extracted from
bed above the line as below it (Griffiths, 1979). With suc- DEMs or they can be derived from high-resolution ground or
cessive surveys of a particular cross section, the same datum aerial survey (see below).
and width are used for compatibility between years. Griffiths Cross-section analyses are critical in assessments of bed/
(1979) used the effective channel width, defined by the flood bank stability (e.g., Murgatroyd and Ternan, 1983), the nature
capacity of the channel (bankfull width). Volumes of scour of channel adjustment (e.g., Griffiths, 1979), habitat avail-
and fill, and the net change in sediment storage, can be de- ability, and rehabilitation design (e.g., Thomson et al., 2001).
termined as the mean of the upstream and downstream cross-
section area times the length of a reach between two cross
sections. Changes in cross-section area between consecutive
surveys are based on changes in MBL. 14.7.3.3 Long Profile
The endpoint method can be used to calculate volumes of
sediment transfer in the channel between cross sections as per Long profiles are used to understand changes in slope and
the following equation: energy along a river course. These surveys enable assessment of
  slope, along with bed features, residual pools, flow ob-
A1 þ A2 structions, knickpoints, and gradient changes (Simon and
V¼ L ½1
2 Castro, 2003). A minimum survey length of 6–30 channel
widths, incorporating at least two complete pool–riffle se-
where A1 is the area of cross section at downstream end of quences, is recommended to measure gradient. Long-profile
reach (m2), A2 the area of cross section at upstream end of data can be combined with discharge to create a model of
reach (m2), and L the distance between the sections (m). stream power variability along river’s course (Knighton, 1999;
Investigations of channel form based upon cross sections Jain et al., 2006). This can be used to interpret downstream
should incorporate the adjacent floodplain and not end at the changes in river type (e.g., Jain et al., 2008) and identify
 Methods in Geomorphology: Investigating River Channel Form 83

reaches most likely to be responsive to high-magnitude floods in Section 14.7.4. Data acquisition needs to be terrain sensitive,
(e.g., Reinfelds et al., 2004). that is, the survey should be designed to generate data that will
Numerous techniques can be used to generate long pro- model reach morphology as accurately as possible, increasing
files. Method selection is scale dependent. Long profiles of survey point densities at breaks in slope. One drawback with
local, reach-scale sections are best investigated using accurate tacheometric survey is the field time required to map the
and precise survey equipment such as differential GPS or total channel. Many person-hours are required to complete a de-
station (see Raven et al., 2009). Long profiles should record tailed ground survey. Nevertheless, this is probably the most
the type of geomorphic units as well as instream hydraulic accurate and highest precision method of channel form in-
units (pools and riffles). Ideally, survey points should trace the vestigation. Ground-based photogrammetry has been used
channel thalweg. However, where water levels are too deep, alongside tacheometric survey to generate detailed DEMs of
channel slope can be derived based on edge-of-water surveys channel form permitting assessment of change on a diurnal
(Simon and Castro, 2003). If an approximate channel gradient basis (e.g., Lane et al., 1996; Heritage et al., 1998). More re-
is required, abney levels or clinometers can be deployed. These cently, ground-based terrestrial laser scanning (TLS) enables
types of surveys are generally not feasible on larger rivers, al- assessment of daily changes in channel form at a resolution of
though the combination of GPS mapping water surface and 1600 points per m2 (Heritage and Hetherington, 2007; Milan
sonar point measurements can provide detailed long profiles et al., 2007). TLS can produce very accurate DEMs of river
(e.g., Flug et al., 1998). reaches. From this, DEMs of difference between surveys
Long-profile data inevitably become more generalized over measure daily change in dynamic channels at an extremely
longer distances. At the simplest and most generalized level, high resolution. The problem with TLS is limited laser pene-
long profiles may be constructed using height information tration of water. Its utility to date has therefore been best
from topographic maps. Greater precision is afforded where demonstrated in shallow gravelly streams (Milan et al., 2007;
catchment DEMs are available. Airborne LiDAR surveys and Brasington et al., 2007). Each TLS survey generates millions of
high-resolution remote sensing are especially useful at this data points which require considerable post-processing time
scale (e.g., Carbonneau et al., 2005a). and/or computer power. Helpful advice in setting up a TLS
survey of a river channel is provided by Heritage and Hether-
ington (2007). Field survey positions should maximize scan
angle across a surface. Care should be made to provide scan
14.7.3.4 3D Form
overlap to best capture 3D form of the surface.
Technological advances have taken assessment of channel High-resolution aerial remote sensing and photogrammetry
form to another level of sophistication. Rumsby et al. (2008, as well as airborne LiDAR surveys have increased the spatial
p. 41) suggested that, ‘‘The growing interest and attention coverage made possible in the acquisition of detailed topo-
given to three dimensional morphological analysis has been graphic information in river channels (Winterbottom and
driven in part by developments in survey technology and data Gilvear 1997; Lane et al., 2003; Carbonneau et al., 2005b;
processing software, and by the increasing need for high Leckie et al., 2005). Deeper water continues to present a
resolution topographic data to better model fluvial systems.’’ problem to data acquisition, as does vegetation (Charlton
Previously, 3D analyses of river form had to integrate data et al., 2003). However, multispectral image analysis (e.g.,
from channel cross sections with data collected on channel Winterbottom and Gilvear, 1997; Whited et al., 2002) or image
planform (see Ashmore and Church, 1998). This is typified in analysis of water color (Westaway et al., 2003) or illumination
studies on morphological budgeting in gravelly rivers (e.g., variation corrected by feature-based image processing (Car-
Popov, 1962; Goff and Ashmore, 1994; Martin and Church, bonneau et al., 2005b) has made sub-aqueous, high-resolution
1995; McLean and Church, 1999; Brewer and Passmore, 2002; data generation feasible from large-scale aerial photography.
Fuller et al., 2002). However, the implicitly cross-stream em- Accordingly, the dynamics of extended reaches of channel can
phasis in these studies gives rise to a high degree of uncertainty be assessed (e.g., Westaway et al., 2003). The latest generation
in reach-scale sediment budgets (Brasington et al., 2000). of airborne LiDAR, which emits red and green light, overcomes
These uncertainties can be overcome by reducing distance the difficulties posed by first-generation (red only) LiDAR,
between cross sections (e.g., Raven et al., 2009). 3D analyses since while the red is absorbed by water, the green penetrates to
utilizing GIS-based DEMs of river channel morphology pro- the bed, generating a return from sub-aqueous surfaces. This
vide more rigorous measures of channel morphology (Lane, has been used successfully in extended channel mapping
1998). Data used to generate DEMs in GIS software such as (Hilldale and Raff, 2008), but its application to date has been
ArcGISs can be acquired in a variety of ways, both ground restricted by expense and availability. Bathymetric LiDAR, like
based and airborne. conventional LiDAR, is not (yet) a high-precision survey tool,
Advanced tacheometric equipment (e.g., electronic Total but the advantage of LiDAR is in acquiring data at a c. 1–2-m
Stations and real-time kinematic differential global positioning spatial resolution across large tracts of river corridor, with ele-
system (RTK-dGPS)) has increased the ground survey potential vation accuracies within about 0.2 m. This permits generation
for rapid acquisition of data to build detailed DEMs of river of DEMs, which, while not as accurate as terrestrially derived
morphology at a reach scale (see Brasington et al., 2000, 2003; data, nevertheless allow for gross change detection between
Fuller et al., 2003a). This provides opportunities to link change surveys (e.g., Procter et al., 2010). When using LiDAR data, the
in form with processes over timescales measured in months second return should be used, since the first return pulse will
(Fuller et al., 2003b; Fuller and Hutchinson, 2007; Fuller and reflect vegetation where present. Aerial survey companies
Marden, 2011). A case study using these procedures is discussed should be asked to supply post-processed ground return data,
84 Methods in Geomorphology: Investigating River Channel Form

as these data are of most use to fluvial geomorphologists in- specifically given the nature and history of the catchment.
vestigating river channel form. These catchment-framed assessments of river channel form
support more targeted proactive management applications
(Brierley et al., 2002).
The River Styles Framework was applied in Twin Streams
14.7.4 Case Studies
catchment, an urban catchment in West Auckland, New Zea-
land (see Reid et al., 2008; Gregory et al., 2008 for full details).
Two case studies are presented here to reflect the variability in
Geoindicators were specifically designed to assess variables
techniques used to assess river channel form. Techniques can be
that become degraded due to changes in urban hydrology and
placed along a continuum with descriptive, qualitative techni-
land use. For example, channel geometry was assessed with the
ques at one end and quantitative, technologically advanced,
understanding that urban channels are commonly over-
data rich techniques at the other. The case studies given below
widened due to larger and more frequent flood peaks.
provide examples from each end of this continuum, illustrating
Catchment-scale maps illustrating the distribution of river
how research aims dictate the type and nature of the data set
type and condition were produced, with the distribution of
collected to assess river form. The first summarizes the use of
geomorphic condition found to be driven by river type, land
the qualitative River Styles Frameworks to classify streams.
use, and catchment position. In general, good-condition rivers
These applications have been widely used to underpin man-
were found in the headwaters and channels declined in con-
agement and river rehabilitation at the catchment scale (Brier-
dition downstream. However, important exceptions to this
ley and Fryirs, 2005; Brierley et al., 2002, 2011). The second
trend were present. Surprisingly, the estuary of Huruhuru
uses recent technological advances to generate high-resolution
Creek was competent to flush the excess sediment delivered,
topographic data to assess process–form links and associated
maintaining good geomorphic condition, despite being the
sediment budgets in river systems. This provides an example of
receiving environment for many more degraded reaches up-
measurement of 3D channel form, wherein a detailed, reach-
stream. By contrast, many of the confined, fine-bed headwater
scale data set is used to answer more detailed scientific ques-
reaches were found to have formed through incision into
tions regarding morphological adjustment.
Intact Valley Fills (wetlands), resulting in a degraded,
homogeneous geomorphic structure (cf. Reid et al., 2010).
Recognizing the rarity and hydroecomorphological import-
14.7.4.1 Case Study 1: River Styles
ance (i.e., filtering and slowing the movement of water down
The River Styles Frameworks is a classification scheme that the catchment) of remaining sections of Intact Valley Fill,
aims to capture the range of rivers within a catchment and these areas are identified as strategic reaches that require tar-
understand the controls on their distribution (Brierley and geted management to protect them. This assessment also
Fryirs, 2000, 2005; Reid et al., 2008; Gregory et al., 2008). identified the causes and types of degradation suffered across
Channel form is classified based on valley confinement the catchment. Fine-grained material smothered the bed of
(confined, partly confined, or unconfined), channel planform many of the reaches in the lower catchment. Several small
(sinuosity, number of channels, and lateral channel adjust- tributaries undergoing rapid erosion were identified as major
ments on the valley floor), geomorphic units, and bed ma- sources contributing to this sedimentation. Sediment man-
terial (see Figure 3). Field visits complement air photograph, agement and replanting in these areas would reduce sediment
remote sensing, and DEM analyses. River Styles record the supply to the lower reaches, managing the cause rather than
character and behavior for each type of river. Analyses are not the effects of the degradation. This framework provided a
pigeonholed into a finite number of river types based on physical template upon which to base more targeted man-
preordained classifications; rather, the inherent variability of agement actions framed within a catchment-scale vision/plan.
any given riverscape is allowed to speak for itself. Field as-
sessment is carried out at a representative site for each River
Style to derive planform maps, cross sections, and assess
14.7.4.2 Case Study 2: Waipaoa
geomorphic units and bed material attributes. The distri-
bution of River Styles is explained at the catchment scale. Steep relief, high rainfall, and highly weathered and crushed
Subsequent analyses assess the condition of rivers across a Cretaceous shales in the upper part of the Waipaoa catchment
catchment. Geoindicators are selected for each River Style, have predisposed the catchment to extensive gully erosion
based on the range of channel adjustment. For example, as- (Figure 7). Erosion was greatly exacerbated by catchment de-
sessing bank erosion in a confined stream with bedrock banks forestation in the early twentieth century (Marden et al.,
would provide little insight into adjustment. Relevant geoin- 2005). One particular gully complex has been the focus of
dicators are selected, and appropriate questions asked for each research over the past c. 20 years. The Tarndale Slip feeds a
geoindicator. These questions are based upon attributes of this first-order tributary of the Waipaoa, which has been marked
variable in a good condition variant of this type of river. Once by the development of a fan emanating from the toe of the
more, a catchment-scale map of river condition is derived. gully complex. The channel on this fan has been observed to
Diagrams showing good-, moderate-, and poor-condition cut and fill its bed in response to sediment supply variability
variants of each River Style are generated. Finally, analyses of from the Tarndale gully mass-movement complex. Initially,
river evolution are performed to interpret the trajectory of river based on biannual cross-section surveys, it appeared that
adjustments and prospects for river recovery. From this, vi- cutting and filling was primarily a seasonal process, with
sions and target conditions for rehabilitation can be designed winter infilling due to accelerated supply of sediment from the
 Methods in Geomorphology: Investigating River Channel Form 85

Te rate rain
gauge
∗
N
177° 48′ E

Te Weraroa 178° E

400 m Mangatu Waingaromia

38°14′ 45″ S
(a)
Waihora

0 3 6 km Kanakanaia

38°30′ S
Tk Pacific
Cretaceous terrain ocean
Wh
Tasman
TG Tertiary terrain sea
North
island
Gully
Pacific
Aggraded riverbed ocean
0 10 20 km
(b)

Figure 7 Location map of Tarndale gully and fan in the context of the upper Waipaoa catchment: (a) LiDAR-derived digital elevation model of
the gully–fan complex (LiDAR datar Gisborne District Council, courtesy Dave Peacock), (b) upper Waipaoa catchment showing major lithologies
and gullies (after Marden et al., 2005). Reproduced from Fuller, I.C., Marden, M., 2010. Rapid channel response to variability in sediment supply:
cutting and filling of the Tarndale Fan, Waipaoa catchment, New Zealand. Marine Geology 270, 45–54.

gully mass-movement complex, whereas in summer the however, as survey data are not collected on a grid basis they
channel incised in response to reduced sediment supply in were interpolated to create a digital elevation surface. Data
drier conditions (Fuller and Marden, 2010). This demonstrates interpolation (DEM generation) in this study used triangula-
the utility of cross sections in assessment of channel form to a tion with linear interpolation (TLI). TLI is a grid-based version
point; however, the speed and complexity of processes in this of a triangulated irregular network (TIN). Further technical
dynamic landscape limited the capacity of a network of six details are outlined in Fuller et al. (2011).
cross sections spaced between 100 and 200 m apart to capture Enhanced resolution of the behavior of the channel flow-
the detail. To better understand patterns and processes of ing across the fan afforded by analysis of these DEMs presents
sediment delivery and channel response, the channel along its a picture of complex, rapid, a-seasonal cutting, and filling
entire c. 1-km length from gully toe to confluence with the Te (Figure 8). Generation of sediment from the gully complex by
Weraroa Stream was surveyed using differential GPS and Total large-scale mass movements, including landslides and debris
Station at 10 intervals over a 3-year period between December flows, causes rapid channel and fan infilling. This sediment is
2004 and April 2008 (Fuller and Marden, 2011). The min- conveyed downstream and most effectively delivered to the
imum acceptable vertical accuracy of observations was set at Te Weraroa Stream by trimming of the lower fan during high
0.05 m. Average vertical accuracy was 0.02 m. Points were flows. This means the upper part of the channel may be in-
surveyed to a precision of 0.001 m. Each survey took about 60 filled at the same time as the downstream portion incises,
person hours, which, with a three- or four-person team, meant since trimming produces a headcut which extends to mid-fan,
2–3 days’ fieldwork per survey, depending upon the per- incising the downstream section of channel. The behavior of
formance of the GPS units, which is dependent upon satellite the channel on the lower fan is thus conditioned by both
constellation. Topographic data were used to construct DEMs interaction with the Te Weraroa Stream and sediment supplied
using Surfers spatial analysis software (Surfer, 2002). These from up-fan. Channel form on Tarndale Fan is therefore
DEMs (Figure 8) are based on 1-m grids (Fuller et al., 2003a); strongly conditioned by processes of sediment transfer, which
86 Methods in Geomorphology: Investigating River Channel Form

December 2004 April 2005 August 2005

December 2005 May 2006 August 2006

November 2006 April 2007 August 2007

April 2008
N

300 m

(a) (b) (c) (d)

(e) (f) (g) (h)

Qualitative assessment of
sediment supplied from gully
Minimal sediment supply
Moderate sediment supply Elevation change (m)
Substantial sediment supply
−0.2
−0.6

−1.4
−1.8
−2.2
−2.6

−3.4
−3.8
−4.2
−4.6
4.6
4.2
3.8
3.4

2.6
2.2
1.8
1.4

0.6
0.2

−1

−3

−5

N
5

(i) 200 m
Figure 8 (Top) DEMs for each survey of the Tarndale Fan. (Bottom) DEMs of difference for successive surveys showing elevation changes in
the channel on the Tarndale Fan and inferred qualitative assessment of sediment supplied from the Tarndale Gully to the channel: (a) December
2004–April 2005, (b) April 2005–August 2005, (c) August 2005–December 2005, (d) December 2005–May 2006, (e) May 2006–August 2006,
(f) August 2006–November 2006, (g) November 2006–April 2007, (h) April 2007–August 2007, (i) August 2007–April 2008. Reproduced from
Fuller, I.C., Marden, M., 2011. Slope–channel coupling in steepland terrain: a field-based conceptual model from the Tarndale gully and fan,
Waipaoa catchment, New Zealand, Geomorphology 128, 105–115.
 Methods in Geomorphology: Investigating River Channel Form 87

in turn is conditioned by the extent of activity in the Tarndale are required to better understand process–form interactions.
gully complex, as well as lateral erosion at the confluence with The more frequent the surveys, the better the understanding of
Te Weraroa Stream. Detailed assessment of 3D form in this system dynamics. Ideally, more extensive use of bathymetric
example demonstrates the detail and complexity of processes LiDAR will fill the gaps left by deeper water. These approaches
operating in this system. Selected techniques in this case are generally prohibitively expensive and cheaper methods of
studies enabled data to be acquired rapidly at short notice. acquiring similar detailed data are being explored. The po-
tential for lighter-than-air blimps and UAVs is enormous.
These methods are especially useful to investigate the inter-
14.7.5 Future Work and Direction mediate scale reach (1–10 km) in small-medium gravelly
rivers.
All too often, analyses of river channel form are applied at a Future work is required to determine and communicate
particular scale for a particular purpose. Comprehensive, limitations to methods and tools used, finding techniques to
cross-scalar analyses are notable for their scarcity. Ideally, in- overcome some of these shortcomings. Each approach to in-
tegrative understandings of channel planform, cross section, vestigating channel form has its own limitations, be it accur-
long profile, and 3D morphology will be applied at the acy, spatial extent, technological requirement, time taken, or
catchment scale, providing direct insight into reach-to-reach expertise required. Assessment of channel form to characterize
(and tributary-to-trunk stream) linkages. In this manner, ap- process can be misleading and needs to be combined with
preciation of spatial variability can be meaningfully tied to assessment of process in order to obtain a reliable and not
appraisals of river dynamics, and associated evolutionary tra- misleading data set (Simon and Castro, 2003). Recognition
jectories (see Surian et al., 2009). that channels with the same form can result from different
Investigation and understanding of channel form and processes operating from different initial conditions is also
process–form interactions have benefited from technological important (the principle of equifinality). Multiple hypotheses
advancement over the past 20 years. This is likely to continue. should be applied to create the best explanation of process–
Whereas annual surveys of channel form were once considered form interactions for a specific reach (Haines-Young and
a high temporal resolution, the increasing availability of Petch, 1983). Conclusions drawn from only assessing form are
equipment such as TLS, capable of rapid data acquisition, difficult to compare and lack the depth of insight that is re-
means that daily reach-scale surveys are now feasible and in- quired to draw valid conclusions. Investigation of change in
deed required to best understand process–form interactions in channel form provides the opportunity to link process and
very dynamic environments (e.g., proglacial streams; Milan form.
et al., 2007). Advances in this field can be seen in pushing the Care should also be applied in the use of classification
data collection capability of this technology to the limit, with schemes to assess river channel form. In some instances, this
increasingly dense spatial and temporal resolution surveys of has led to misunderstanding or misinterpreting the processes
channel form. Williams et al. (2011) analyzed channel change that generate a specific channel form. Some techniques are
in a 2.5  0.7 km reach of the braided Rees River, in the South oversimplistic. For example, there are inherent limitations in
Island of New Zealand. High-resolution DEMs (reduced to a application of schemes that fit rivers into preordained cat-
0.05-m resolution grid) using B300 TLS scans were created egories. Assessments of river form should incorporate ap-
following 10 flood events over 8 months to record the distri- praisals of channel change over time. The evolutionary history
bution and quantity of aggradation and erosion for events of of a river provides insight into how processes have changed
different magnitudes. Although these types of study generate over time (see Simon et al., 2007; Kondolf et al., 2003).
increasingly detailed data sets offering important scientific Over-reliance upon computer-derived appraisals of river form
insights, the substantial data collection and processing time can also be misleading. For example, the river environment
makes it impractical for many management applications. classification (REC), which classifies rivers in New Zealand
Whereas such advances in technology can offer seemingly using a 25-m DEM and other large-scale GIS-derived variables
simple solutions to assessing channel form, in practice they (e.g., geology and land use), has been found to not generate
may be more difficult, making it important to design data sets biophysically meaningful differentiation of stream classes
(in terms of data scale and resolution) that relate to the spe- (Inglis et al., 2008). Thus, site visits should be an essential
cific aims of the study. Future automated techniques for pro- part of any investigation of channel form. System-specific
cessing and cleaning these data sets will decrease the time knowledge is required to assess process–form interactions and
required, increasing the practicality of applying such techni- associated management applications (Brierley and Fryirs,
ques more widely. 2005).
In less dynamic channel environments, it is certainly de- Better data management in the future will improve the ease
sirable to provide as high a temporal resolution of survey as with which river channel form can be assessed. At present, the
possible. Ideally, event-based analyses are applied to constrain distribution of data can be fragmented among people with dif-
the role played by discrete floods of varying magnitude as ferent interests, with historic data commonly poorly docu-
determinants of channel form adjustment. From this, better mented and difficult to obtain. Creating centrally held databases
predictions of channel response to events of a given magni- that pull together different information sources on catchments
tude can be derived. At a larger spatial scale, increasing avail- would greatly increase the speed at which meaningful assess-
ability of LiDAR allows better modeling of channel form at ment of river form can be created. Understanding past influences
reach to catchment scales. This is of immediate benefit for and adjustments provides an important underpinning for as-
flood modeling and hazard assessment. Repeat LiDAR surveys sessment. In addition, it can shape the types of data that are
88 Methods in Geomorphology: Investigating River Channel Form

collected, so that new data sets are either complementary or Brasington, J., Langham, J.A., Rumsby, B.T., 2003. Methodological sensitivity of
build upon past work. Gaining large bodies of data is relatively morphometric estimates of fluvial coarse sediment transport. Geomorphology 53,
299–316.
easy, but each set of data acquired needs to be underpinned by a
Brasington, J., Rumsby, B.T., McVey, R.A., 2000. Monitoring and modelling
clear purpose of what is going to be analyzed. Focused goals and morphological change in braided river systems using the Global Positioning
outcomes, created relative to a specific system and question, System. Earth Surface Processes and Landforms 25, 973–990.
mean that time is utilized more effectively, creating more co- Brasington, J., Wheaton, J.M., Vericat, D., Hodge, R., 2007. Modelling braided river
herent and relevant research outputs. morphodynamics with terrestrial laser scanning. Eos Transactions AGU 88(52),
Fall Meeting Supplement, Abstract H51L-02.
Developments into predictive modeling of channel dy- Brewer, P.A., Lewin, J., 1998. Planform cyclicity in an unstable reach: complex
namics provide the best potential for advancing assessment fluvial response to environmental change. Earth Surface Processes and
and understanding of river channel form into the future. How Landforms 23, 989–1008.
rivers will respond to changing flood regimes in a climatically Brewer, P.A., Passmore, D.G., 2002. Sediment budgeting techniques in gravel bed
rivers. In: Jones, S., Frostick, L.E. (Eds.), Sediment Flux to Basins: Causes,
changing world is a critical question to address. It is also im-
Controls and Consequences. Geological Society Special Publication, London,
portant to understand the impact of land-use change on river UK, vol. 191, pp. 97–113.
behavior. For example, deforestation of catchments in the East Brice, J.C., 1960. Index for description of channel braiding. Geological Society of
Coast Region of New Zealand has contributed to substantial America Bulletin 85, 581–586.
valley floor aggradation and transformation of floodplain Brice, J.C., 1964. Channel patterns and terraces of the Loup Rivers in Nebraska.
United States Geological Survey Professional Paper P 0422-D, pp. D1–D41.
morphology in rivers such as the Waiapu (Page et al., 2008).
Brierley, G.J., 2010. Landscape memory: the imprint of the past on contemporary
The better our understanding of current dynamics, the stron- landscape forms and processes. Area 42, 76–85.
ger is our position to predict the future direction of channel Brierley, G.J., Fryirs, K.A., 2000. River Styles, a geomorphic approach to catchment
change in response to both climate and land-use change. characterization: implications for river rehabilitation in Bega Catchment,
Future work will therefore both develop and improve existing New South Wales, Australia. Environmental Management 25, 661–679.
Brierley, G.J., Fryirs, K.A., 2005. Geomorphology and River Management:
methods as new and emerging technology becomes increas- Applications of the River Styles Framework. Blackwell, Oxford.
ingly available. Brierley, G.J., Fryirs, K.A., Boulton, A., Cullum, C., 2008. Working with change: the
importance of evolutionary perspectives in framing the trajectory of river
adjustment. In: Brierley, G.J., Fryirs, K.A. (Eds.), River Futures: An Integrative
Scientific Approach to River Repair. Island Press, London, pp. 65–84.
14.7.6 Conclusions Brierley, G.J., Fryirs, K.A., Cook, N., Outhet, D., Raine, A., Parsons, L., Healey, M.,
2011. Geomorphology in action: linking policy with on-the-ground actions
Emphasis in investigating channel form has moved toward an through applications of the River Styles framework. Applied Geography 31,
integrated understanding of process–form relationships in the 1132–1143.
river channel domain. This has been driven in the last Brierley, G.J., Fryirs, K.A., Outhet, D., Massey, C., 2002. Application of the River
styles framework as a basis for river management in New South Wales,
c. 20 years by a dramatic increase in the power and availability Australia. Applied Geography 22, 91–122.
of survey technology for use in channel form investigation. Brooks, A.P., Brierley, G.J., 2004. Framing realistic river rehabilitation targets in
Investigations are no longer restricted to 1D or 2D assessments light of altered sediment supply and transport relationships: lessons from East
of particular channel components or processes using planform Gippsland, Australia. Geomorphology 58, 107–123.
Brooks, A.P., Gehrke, P.C., Jansen, J.D., Abbe, T.B., 2004. Experimental
or cross section or long profile. Rather, the availability and
reintroduction of Woody Debris on the Williams River, NSW: geomorphic and
application of high-resolution survey technology have facili- ecological responses. River Research and Applications 20, 513–536.
tated a tremendous growth in the possibilities of data Brookes, A., Shields, F., 1996. River Channel Restoration: Guiding Principles for
acquisition along a river corridor in a 3D capacity, markedly Sustainable Projects. Wiley, New York, NY.
enhancing our ability to appraise adjustments over time in Carbonneau, P.E., Bergeron, N., Lane, S.N., 2005a. Automated grain size
measurements from airborne remote sensing for long profile measurements of
an effective manner. This is not to denigrate the value of fluvial grain sizes. Water Resources Research 41, W11426. http://dx.doi.org/
more simple investigations of channel form, which still have 10.1029/2005WR003994.
their place in providing cheap and accessible means of in- Carbonneau, P.E., Lane, S.N., Bergeron, N., 2005b. Feature based image processing
vestigation. However, it is now possible to build high- methods applied to bathymetric measurements from airborne remote sensing in
fluvial environments. Earth Surface Process and Landforms 31(11), 1413–1423.
resolution models, at the grain scale, of whole lengths of
Charlton, M.E., Large, A.R.G., Fuller, I.C., 2003. Application of airborne LiDAR in
channel (e.g., Carbonneau et al., 2005a). These developments river environments: the River Coquet, Northumberland, UK. Earth Surface
permit increasingly detailed understanding of the interaction Processes and Landforms 28, 299–306.
between channel processes and the resulting channel form, Chow, V.T., 1959. Open Channel Hydraulics. McGraw Hill, New York, NY.
which will be of critical use in predictive modeling of channel Church, M., 2006. Bed material transport and the morphology of alluvial river
channels. Annual Review of Earth and Planetary Science 34, 325–354.
process and form. Coulthard, T.J., Hicks, D.M., Van De Wiel, M.J., 2007. Cellular modelling of river
catchments and reaches: advantages, limitations and prospects. Geomorphology
90, 192–207.
References Cowan, W.L., 1956. Estimating hydraulic roughness coefficients. Agricultural
Engineering 37, 473–475.
Davis, W.M., 1899. The geographical cycle. Geographical Journal 14, 478–504.
Ashmore, P.E., Church, M.A., 1998. Sediment transport and river morphology: a Davis, W.M., 1909. The systematic description of landforms. Geographical Journal
paradigm for study. In: Klingemann, P.C., Beschta, R.L., Komar, P.D., Bradley, 34, 300–326.
J.B. (Eds.), Gravel-Bed Rivers in the Environment. Water Resources Publications, Dury, G.H., 1964. Principles of Underfit Streams. United States Geological Survey
Highlands Ranch, CO, pp. 115–148. Professional Paper 452-A.
Boruah, S., Gilvear, D., Hunter, P., Sharma, N., 2008. Quantifying channel Eaton, B.C., Church, M., Millar, R.G., 2004. Rational regime model of alluvial channel
planform and physical habitat dynamics on a large braided river using morphology and response. Earth Surface Processes and Landforms 29, 511–529.
satellite data – the Brahmaputra, India. River Research and Applications 24, Eaton, B.C., Millar, R.G., Davidson, S., 2010a. Channel patterns: braided,
650–660. anabranching, and single-thread. Geomorphology 120, 353–364.
 Methods in Geomorphology: Investigating River Channel Form 89

Eaton, B., Moore, R., Giles, T., 2010b. Forest fire, bank strength and channel Heritage, G.L., Fuller, I.C., Charlton, M.E., Brewer, P.A., Passmore, D.G., 1998. CDW
instability: the ‘unusual’ response of Fishtrap Creek, British Columbia. Earth photogrammetry of low relief fluvial features: accuracy and implications for
Surface Processes and Landforms 35, 1167–1183. reach-scale sediment budgeting. Earth Surface Processes and Landforms 23,
Eyquem, J., 2007. Using fluvial geomorphology to inform integrated river basin 1219–1233.
management. Water and Environment Journal 21, 54–60. Heritage, G.L., Hetherington, D., 2007. Towards a protocol for laser scanning in
Flug, M., Seitz, H., Scott, J., 1998. Measurement of stream habitat using sonar. fluvial geomorphology. Earth Surface Processes and Landforms 32, 66–74.
Regulated Rivers: Research and Management 14, 511–517. Hey, R.D., Bathurst, J.C., Thorne, C.R., 1982. Gravel Bed Rivers: Fluvial Processes,
Fryirs, K., Brierley, G.J., 2010. Antecedent controls on river character and behaviour Engineering and Management. Wiley, Chichester.
in partly confined valley settings: upper Hunter catchment, NSW, Australia. Hey, R.D., Thorne, C.R., 1986. Stable channels with mobile gravel beds. Journal of
Geomorphology 117, 106–120. Hydraulic Engineering 112, 671–689.
Fryirs, K., Spink, A., Brierley, G., 2009. Post-European settlement response Hickin, E.J., 1974. The development of meanders in natural river channels.
gradients of river sensitivity and recovery across the upper Hunter catchment, American Journal of Science 274, 414–442.
Australia. Earth Surface Processes and Landforms 34, 897–918. Hickin, E.J., Nanson, G.C., 1975. The character of channel migration on the Beaton
Fuller, I.C., 2007. Geomorphic work during a ‘150-year’ storm: contrasting River, Northeast British Columbia, Canada. Geological Society of America
behaviour of river channels in a New Zealand catchment. Annals Association of Bulletin 86, 487–494.
American Geographers 97, 665–676. Hickin, E.J., Nanson, G.C., 1984. Lateral migration rates of river bends. Journal of
Fuller, I.C., 2008. Geomorphic impacts of a 100 year flood: Kiwitea Stream, Hydraulic Engineering 110, 1157–1167.
Manawatu catchment, New Zealand. Geomorphology 98, 84–95. Hicks, D.M., 1991. Roughness Characteristics of New Zealand Rivers: A handbook
Fuller, I.C., 2011. Using morphological adjustments to appraise sediment flux. for Assigning Hydraulic Roughness Coefficients to River Reaches by the ‘‘Visual
Journal of Hydrology (New Zealand) 50, 59–79. Comparison’’ Approach. Water Resources Survey, DSIR Marine and Freshwater,
Fuller, I.C., Hutchinson, E., 2007. Sediment flux in a small gravel-bed stream: Wellington.
response to channel remediation works. New Zealand Geographer 63, Hilldale, R.C., Raff, D., 2008. Assessing the ability of airborne LiDAR to map river
169–180. bathymetry. Earth Surface Processes and Landforms 33(5), 773–783.
Fuller, I.C., Large, A.R.G., Charlton, M.E., Heritage, G.L., Milan, D.J., 2003a. Hooke, J.M., 1980. Magnitude and distribution of rates of river bank erosion. Earth
Reach-scale sediment transfers: an evaluation of two morphological Surface Processes 5, 143–157.
budgeting approaches. Earth Surface Processes and Landforms 28, Hooke, J.M., 2007. Spatial variability, mechanisms and propagation of change in an
889–903. active meandering river. Geomorphology 84, 277–296.
Fuller, I.C., Large, A.R.G., Milan, D.J., 2003b. Quantifying channel development and Hoyle, J., Brooks, A., Spencer, J., 2010. Modelling reach-scale variability in
sediment transfer following chute cutoff in a wandering gravel-bed river. sediment mobility: an approach for within-reach prioritization of river
Geomorphology 54, 307–323. rehabilitation works. River Research and Applications, http://dx.doi.org/10.1002/
Fuller, I.C., Marden, M., 2010. Rapid channel response to variability in sediment rra.1472.
supply: cutting and filling of the Tarndale Fan, Waipaoa catchment, New Hughes, M.L., McDowell, P.F., Marcus, W.A., 2006. Accuracy assessment of
Zealand. Marine Geology 270, 45–54. georectified aerial photographs: implications for measuring lateral channel
Fuller, I.C., Marden, M., 2011. Slope–channel coupling in steepland terrain: a field- movement in a GIS. Geomorphology 74, 1–16.
based conceptual model from the Tarndale gully and fan, Waipaoa catchment, Inglis, L., Boothroyd, I.K.G., Brierley, G.J., 2008. Effectiveness of the river
New Zealand. Geomorphology 128, 105–115. environment classification in the Auckland Region. New Zealand Geographer 64,
Fuller, I.C., Passmore, D.G., Heritage, G.L., Large, A.R.G., Milan, D.J., Brewer, P.A., 181–193.
2002. Annual sediment budgets in an unstable gravel bed river: the River Jain, V., Fryirs, K.A., Brierley, G.J., 2008. Where do floodplains begin? The role of
Coquet, northern England. In: Jones, S., Frostick, L.E. (Eds.), Sediment Flux to total stream power and longitudinal profile form on floodplain initiation
Basins: Causes, Controls & Consequences. Geological Society Special processes. Geological Society of America Bulletin 120, 127–141.
Publication, London, UK, vol. 191, pp. 115–131. Jain, V., Preston, N., Fryirs, K.A., Brierley, G.J., 2006. Comparative assessment of
Gilbert, G.K., 1914. The transportation of debris by running water. US Geological three approaches for deriving stream power plots along long profiles in the
Survey Professional Paper 86. upper Hunter River Catchment, New South Wales, Australia. Geomorphology 74,
Gilvear, D., Bryant, R., 2003. Analysis of aerial photography and other remotely 297–317.
sensed data. In: Kondolf, G., Piegay, H. (Eds.), Tools in Fluvial Geomorphology. Jowett, I.G., 1998. Hydraulic geometry of New Zealand rivers and its use as a
Wiley, Chichester, pp. 135–170. preliminary means of habitat assessment. Regulated Rivers: Research and
Goff, J.R., Ashmore, P.E., 1994. Gravel transport and morphological change in Management 14, 451–466.
braided Sunwapta river, Alberta, Canada. Earth Surface Processes and Landforms Knighton, A.D., 1999. Downstream variation in stream power. Geomorphology 29,
19, 195–213. 293–306.
Goodwin, C.N., 1999. Fluvial classification: Neanderthal necessity or needless Knighton, D., 1998. Fluvial Forms and Processes: A New Perspective. Arnold,
normalcy. In: Olson, D.S., Potyondy, J.P. (Eds.), Wildland Hydrology. American New York, NY.
Water Resources Association, TPS-99-3, Herndon, VA, pp. 229–236. Kondolf, G.M., Montgomery, D.R., Piegay, H., Schmitt, L., 2003. Geomorphic
Gregory, G.E., Reid, H.E., Brierley, G.J., 2008. Geomorphic river recovery in an classification of rivers and streams. In: Kondolf, G., Piegay, H. (Eds.), Tools in
urban catchment: Twin Streams Catchment, Auckland, New Zealand. Physical Fluvial Geomorphology. Wiley, Chichester, pp. 171–204.
Geography 29, 222–246. Lane, S.N., 1998. The use of digital terrain modelling in the understanding of
Griffiths, G.A., 1979. Recent sedimentation history of the Waimakariri River, dynamic river systems. In: Lane, S.N., Richards, K.S., Chandler, J.H. (Eds.),
New Zealand. Journal of Hydrology (New Zealand) 18, 6–28. Landform Monitoring, Modelling and Analysis. Wiley, Chichester,
Gupta, A., Hock, L., Xiaojing, H., Ping, C., 2002. Evaluation of part of the Mekong pp. 311–342.
River using satellite imagery. Geomorphology 44, 221–239. Lane, S.N., 2000. The measurement of river channel morphology using digital
Gurnell, A.M., 1997. Channel change on the River Dee meanders, 1946–1992, from photogrammetry. Photogrammetric Record 16, 937–957.
the analysis of air photos. Regulated Rivers: Research and Management 13, Lane, S.N., Richards, K.S., Chandler, J.H., 1996. Discharge and sediment supply
13–26. controls on erosion and deposition in a dynamic alluvial channel.
Gurnell, A.M., Downward, S.R., Jones, R., 1994. Channel planform change on the Geomorphology 15, 1–15.
River Dee Meanders, 1876–1992. Regulated Rivers: Research and Management Lane, S.N., Westaway, R.M., Hicks, D.M., 2003. Estimation of erosion and
9, 187–204. deposition volumes in a large, gravel-bed, braided river using synoptic remote
Haines-Young, R.H., Petch, J.R., 1983. Multiple working hypotheses: equifinality sensing. Earth Surface Processes and Landforms 28, 249–271.
and the study of landforms. Transactions of the Institute of British Geographers Lave, R., 2009. The controversy over Natural Channel Design: substantive
New Series 8, 458–466. explanations and potential avenues for resolution. Journal of the American Water
Ham, D.G., Church, M., 2000. Bed-material transport estimated from channel Resources Association 45, 1519–1532.
morphodynamics: Chilliwack River, British Columbia. Earth Surface Processes Lawler, D.M., 1993. The measurement of river bank erosion and lateral channel
and Landforms 25, 1123–1142. change: a review. Earth Surface Processes and Landforms 18, 777–821.
Hart, M.G., 1986. Geomorphology: Pure and Applied. George Allen and Unwin, Lawler, D.M., Thorne, C.R., Hooke, J.M., 1997. Bank erosion and instability.
London. In: Thorne, C.R., Hey, R.D., Newson, M.D. (Eds.), Applied Fluvial
90 Methods in Geomorphology: Investigating River Channel Form

Geomorphology for River Engineering and Management. Wiley, Chichester, Popov, I.V., 1962. Application of morphological analysis to the evaluation of the
pp. 137–172. general channel deformations of the Obi River. Transactions of the State
Leckie, D.G., Cloney, E., Jay, C., Paradine, D., 2005. Automated mapping of Hydrological Institute 94, 22–86 (Translation: Soviet Hydrology 1962, 267–324).
stream features with high-resolution multispectral imagery: an example of Procter, J., Cronin, S.J., Fuller, I.C., Manville, V., Lube, G., 2010. Quantifying
the capabilities. Photogrammetric Engineering and Remote Sensing 71, Geomorphic Impacts of a Lake-Breakout Lahar from Mt. Ruapehu, New Zealand.
145–155. Geology 38, 67–70.
Lejot, J., Delacourt, C., Piégay, H., Fournier, T., Trémélo, V., Allemand, P., 2007. Raven, E.K., Lane, S.N., Ferguson, R., 2010. Using sediment impact sensors to
Very high spatial resolution imagery for channel bathymetry and topography improve the morphological sediment budget approach for estimating bedload
from an unmanned mapping controlled platform. Earth Surface Processes and transport rates. Geomorphology 119, 125–134.
Landforms 32, 1705–1725. Raven, E.K., Lane, S.N., Ferguson, R.I., Bracken, L.J., 2009. The spatial and
Leopold, L.B., 1992. Sediment size that determines channel morphology. In: Billi, temporal patterns of aggradation in a temperate, upland, gravel-bed river. Earth
P., Hey, R.D., Thorne, C.R., Tacconi, P. (Eds.), Dynamics of Gravel-bed Rivers. Surface Processes and Landforms 34, 1181–1197.
Wiley, Chichester, pp. 297–311. Raven, P.J., Holmes, N.T.H., Naura, M., Dawson, F.H., 2000. Using river habitat
Leopold, L.B., Maddock, T., 1953. The hydrologic geometry of stream channels and survey for environmental assessment and catchment planning in the U.K.
some physiographic implications. United States Geological Survey Professional Hydrobiologia 422/423, 359–367.
Paper 252. Richards, K.S., 1976. The morphology of riffle-pool sequences. Earth Surface
Leopold, L.B., Wolman, M.G., 1957. River channel patterns – braided, meandering Processes 1, 71–88.
and straight. United States Geological Survey Professional Paper 282B, pp. Rosgen, D.L., 1994. A classification of natural rivers. Catena 22, 169–199.
39–85. Reid, H.E., Brierley, G.J., Boothroyd, I.K.G., 2010. Influence of bed heterogeneity
Leopold, L.B., Wolman, M.G., Miller, J.P., 1964. Fluvial Processes in and habitat type on macroinvertebrate uptake in peri-urban streams. International
Geomorphology. W.H Freeman, San Francisco. Journal of Sediment Research 25, 203–220.
Lewin, J., 1976. Initiation of bed forms and meanders in coarse-grained sediment. Reid, H.E., Gregory, C.E., Brierley, G.J., 2008. Measures of physical heterogeneity in
Geological Society of America Bulletin 87, 281–285. appraisal of geomorphic river condition for urban streams: Twin Streams
Lewin, J., Hughes, D., Blacknell, C., 1977. Incidence of river erosion. Area 3, Catchment, Auckland, New Zealand. Physical Geography 29, 247–274.
177–180. Reinfelds, I., Cohen, T., Batten, P., Brierley, G., 2004. Assessment of downstream
Leys, K.F., Werritty, A., 1999. River channel planform change: software for historical trends in channel gradient, total and specific stream power: a GIS approach.
analysis. Geomorphology 29, 107–120. Geomorphology 60, 403–416.
Marcus, W.A., Fonstad, M.A., 2008. Optical remote mapping of rivers at sub-meter Rumsby, B.T., Brasington, J., Langham, J.A., McLelland, S.J., Middleton, R.,
resolutions and watershed extents. Earth Surface Processes and Landforms 33, Rollinson, G., 2008. Monitoring and modelling particle and reach-scale
4–24. morphological change in gravel-bed rivers: applications and challenges.
Marden, M., Arnold, G., Gomez, B., Rowan, D., 2005. Pre- and post-reforestation Geomorphology 93, 40–54.
gully development in Mangatu Forest, East Coast, North Island, New Zealand. Rutherfurd, I.D., Jerie, K., Marsh, N., 2000. A rehabilitation Manual for Australian
River Research and Applications 21, 757–771. Streams. Land and Water Resources Research and Development Corporation,
Martin, Y., Church, M., 1995. Bed-material transport estimated from channel Canberra.
surveys: Vedder River, British Columbia. Earth Surface Processes and Landforms Schumm, S.A., 1960. The shape of alluvial channels in relation to sediment type.
20, 347–361. United States Geological Survey Professional Paper 352-B, pp. 17–30.
McLean, D.G., Church, M., 1999. Sediment transport along lower Fraser River 2. Schumm, S.A., 1963. A tentative classification of alluvial river channels. United
Estimates based on the long-term gravel budget. Water Resources Research 35, States Geology Circular 477.
2549–2559. Schumm, S.A., 1977. The Fluvial System. Wiley, New York, NY.
Milan, D.J., Heritage, G.L., Hetherington, D., 2007. Application of a 3D laser Schumm, S.A., 1981. Evolution and response of the fluvial system, sedimentologic
scanner in the assessment of erosion and deposition volumes and channel implications. Society of Economic Palaeontologists and Mineralogists Special
change in a proglacial river. Earth Surface Processes and Landforms 32, Publication 31, 19–29.
1657–1672. Schumm, S.A., 1985. Patterns of alluvial rivers. Annual Review of Earth and
Millar, R.G., 2000. Influence of bank vegetation on alluvial channel patterns. Water Planetary Science 13, 5–27.
Resources Research 36, 1109–1118. Schumm, S.A., Khan, H.R., 1972. Experimental study of channel patterns. Bulletin
Montgomery, D.R., Buffington, J.M., 1997. Channel–reach morphology in of the Geological Society of America 83, 1755–1770.
mountain drainage basins. Geological Society of America Bulletin 109, Simon, A., Castro, J., 2003. Measurement and analysis of alluvial channel form.
596–611. In: Kondolf, G., Piegay, H. (Eds.), Tools in Fluvial Geomorphology. Wiley,
Murgatroyd, A.L., Ternan, J.L., 1983. The impact of afforestation on stream Chichester, pp. 292–322.
bank erosion and channel form. Earth Surface Processes and Landforms 8, Simon, A., Doyle, M., Kondolf, M., Shields, F.D., Rhoades, B., McPhillips, M.,
357–369. 2007. Critical evaluation of how the Rosgen classification and associated
Nanson, G.C., Huang, H.Q., 2008. Least action principle, equilibrium states, iterative ‘‘Natural Channel Design’’ methods fail to integrate and quantify fluvial processes
adjustment and the stability of alluvial channels. Earth Surface Processes and and channel response. Journal of the American Water Resources Association 43,
Landforms 33, 923–942. 1117–1131.
Nanson, G.C., Knighton, A.D., 1996. Anabranching rivers: their cause, character and Smith, M.J., Pain, C.F., 2009. Applications of remote sensing in geomorphology.
classification. Earth Surface Processes and Landforms 21, 217–239. Progress in Physical Geography 33, 568–582.
Nykanen, D.K., Foufoula-Georgiou, E., Sapozhnikov, V.B., 1998. Study of spatial Spink, A., Fryirs, K., Brierley, G., 2009. The relationship between geomorphic
scaling in braided river patterns using synthetic aperture radar imagery. Water river adjustment and management actions over the last 50 years in the Upper Hunter
Resources Research 34, 1795–1807. Catchment, NSW, Australia. River Research and Applications 25, 904–928.
Page, M., Marden, M., Kasai, M., et al., 2008. Changes in basin-scale sediment Surfer, 2002. Manual. Golden Software, Colorado.
supply and transfer in a rapidly transformed New Zealand landscape. Surian, N., Ziliani, L., Comiti, F., Lenzi, M.A., Mao, L., 2009. Channel adjustments
In: Habersack, H., Piegay, H., Rinaldi, M. (Eds.), Gravel-Bed Rivers VI: and alteration of sediment fluxes in gravel-bed rivers of North-Eastern Italy:
From Process Understanding to River Restoration. Elsevier, Amsterdam, pp. potentials and limitations for channel recovery. River Research and Applications
337–358. 25, 551–567.
Passmore, D.G., Macklin, M.G., Brewer, P.A., Lewin, J., Rumsby, B.T., Newson, Thomson, J., Taylor, M.P., Fryirs, K.A., Brierley, G.J., 2001. A geomorphological
M.D., 1993. Variability of late Holocene braiding in Britain. In: Best, J.L., framework for river characterisation and habitat assessment. Aquatic
Bristow, C.S. (Eds.), Braided Rivers. Geological Society Special Publication 75, Conservations: Marine and Freshwater Ecosystems 11, 373–389.
London, pp. 205–229. Vericat, D., Brasington, J., Wheaton, J., Cowie, M., 2009. Accuracy assessment of
Patro, S., Chatterjee, C., Singh, R., Raghuwanshi, N.S., 2009. Hydrodynamic aerial photographs acquired using lighter-than-air blimps: low cost tools for
modelling of a large flood-prone river system in India with limited data. mapping river corridors. River Research and Applications 25, 985–1000.
Hydrological Processes 23, 2774–2791. Werritty, A., Ferguson, R.I., 1981. Pattern changes on a Scottish braided river over
Petts, G.E., 1979. Complex response of river channel morphology subsequent to 1, 30, and 200 years. In: Cullingford, R.A. (Ed.), Timescales in Geomorphology.
reservoir construction. Progress in Physical Geography 3, 329–362. Wiley, Chichester, pp. 53–68.
 Methods in Geomorphology: Investigating River Channel Form 91

Westaway, R.M., Lane, S.N., Hicks, D.M., 2003. Remote survey of large-scale Winterbottom, S.J., Gilvear, D.J., 1997. Quantification of channel bed morphology
braided rivers using digital photogrammetry and image analysis. International in gravel-bed rivers using airborne multispectral imagery and aerial
Journal of Remote Sensing 24, 795–816. photography. Regulated Rivers: Research and Management 13, 489–499.
Whited, D., Stanford, J.A., Kimball, J.S., 2002. Application of airborne multispectral Wishart, D., Warburton, J., Bracken, L., 2008. Gravel extraction and planform
digital imagery to quantify riverine habitats at different base flows. River change in a wandering gravel-bed river: The River Wear, Northern England.
Research and Applications 18, 583–594. Geomorphology 94, 131–152.
Williams, G.P., 1978. Bankfull discharge of rivers. Water Resources Research 14, Wohl, E.E., Wilcox, A., 2005. Channel geometry of mountain streams in
1141–1158. New Zealand. Journal of Hydrology 300, 252–266.
Williams, R., Brasington, J., Vericat, D., Hicks, M., Labrosse, F., Neal, M., 2011.
Monitoring braided river change using terrestrial laser scanning and
optical bathymetric mapping. In: Smith, M., Paron, P., Griffiths, J. (Eds.),
Geomorphological Mapping: Methods and Applications. Elsevier, Relevant Websites
Amsterdam.
Winterbottom, S.J., 2000. Medium and short-term channel planform changes on the www.riverstyles.com
Rivers Tay and Tummel, Scotland. Geomorphology 34, 195–208. River Styless website.

Biographical Sketch
Ian Fuller is a senior lecturer in physical geography at Massey University. His research interests in fluvial geo-
morphology span the Holocene by addressing (1) slope–channel coupling in eroding steepland environments
and the contribution of gully mass movement complexes to the catchment sediment cascade at a seasonal to
annual timescale, (2) channel dynamics and sediment transfers in gravelly rivers over annual and multi-decadal
timescales and their application to river management, and (3) alluvial histories and river responses to climate
forcing at centennial and millennial timescales.

Helen Reid is a PhD student at Auckland University. Her research interests include sensitivity and evolutionary
trajectories involving looking at the controls governing the rate and extent of planform adjustment, sediment
budgets and connectivity at the reach and catchment scales, and process-based frameworks to underpin re-
habilitation and management in rivers.

Gary Brierley is professor of physical geography at the University of Auckland. He is co-developer, with Kirstie
Fryirs, of the River Styles framework. His research focuses on the application of river science to river rehabilitation
and management activities. He is particularly interested in the application of coherent river science in guiding the
management of river futures. Much of his present work is being undertaken in western China. He hopes to
develop further professional short courses that demonstrate the use of geomorphology as an integrating platform
for river management activities.
 14.8 Methods in Geomorphology: Mapping Glacial Features
OM Hyatt, University of Canterbury, Christchurch, New Zealand
J Shulmeister, University of Queensland, Brisbane, QLD, Australia
r 2013 Elsevier Inc. All rights reserved.

14.8.1 Introduction 92
14.8.2 Types of Maps 92
14.8.2.1 Process Inference Maps 93
14.8.2.2 Age-Correlation Maps 93
14.8.3 Identification of Features 94
14.8.4 Production of a Base Map or Image 94
14.8.4.1 What Constitutes a Good Base Map or Base Image? 95
14.8.4.2 Images 95
14.8.4.3 Stereoscope Use 95
14.8.4.4 Scale of Mapping 96
14.8.4.5 Map Extents 96
14.8.5 Field Mapping 96
14.8.5.1 Examining Outcrop in Association with Mapping 97
14.8.6 Mapping in Different Glacial Settings – Case Studies 98
14.8.6.1 Temperate Valley Glaciers – East Coast South Island, New Zealand (Rakaia and Rangitata Valleys) 98
14.8.6.2 Former Ice-Cap Covered Regions 98
14.8.6.3 Antarctica and Other Dry Polar/Alpine Environments 98
14.8.6.4 Karst Areas 98
14.8.7 Map Production/Cartography 100
14.8.7.1 Map Compilation – Digitizing Data and Producing the Map 100
14.8.7.2 Map Elements 100
14.8.7.3 Publication 100
Acknowledgments 103
References 103

Abstract

The process of mapping glacial geomorphological features is broken down into its component elements and each is
described. Two types of glacial geomorphological maps are defined and the identification of glacial features discussed. The
basic steps to producing a glacial geomorphological map are detailed, including production of a base map and field
mapping. Short case studies of different mapping settings are included. Finally map production and publication are
discussed.

14.8.1 Introduction authors present a basic description of approaches aimed pri-

marily at those embarking on producing a glacial geo-
Less than 20 000 years ago glaciers covered approximately morphological map for the first time.
30% of the Earth’s land surfaces (e.g., Sugden and John, There are two main parts in mapping glacial features,
1976). Now they cover approximately 10%. As a consequence, identification of landforms and drafting the map. The field
in mid- and high-latitude regions glaciers have modified vast area location, size, and final map destination influences how
areas of the landscape and in many locations are the last mapping is undertaken and the type of map produced. For
major agent of landscape modification (see Figure 1). Con- example, a small field area of 10 km2, with a scale of 1:10 000,
sequently, the mapping of glacial landscapes both modern would likely be mapped using aerial photographs and exten-
and relict is a significant field in geomorphology. Here the sively verified on the ground. In contrast, a map of a sub-
continental area with a scale of 1:2 000 000 would be largely
mapped from satellite images, with very little, if any, field
mapping. In this chapter the authors focus on small area,
relatively large scale maps with final outputs in the range of
Hyatt, O.M., Shulmeister, J., 2013. Methods in geomorphology: mapping
glacial features. In: Shroder, J. (Editor in Chief), Switzer, A.D., Kennedy,
1:20 000 to 1:50 000 as opposed to larger area maps at smaller
D.M. (Eds.), Treatise on Geomorphology. Academic Press, San Diego, CA, scales. This is because these scales are applicable to most field-
vol. 14, Methods in Geomorphology, pp. 92–104. based glacial mapping projects.

92 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00375-4

 Methods in Geomorphology: Mapping Glacial Features 93

Truncated valley spurs

Kame terraces
Kettle
holes

Alluvial fan
Lake benches

Abandoned
channels Terminal
Outwash moraines

Figure 1 A block sketch of part of a valley system that has been glaciated showing some of the main features of valley glaciation and
associated outwash systems including moraines, kettle holes, kame terraces, and abandoned braidplains. Truncated spurs are also a key indicator
of past glaciation, whereas lakes commonly form in the former glacial trough, so lake beaches may be diagnostic also.

14.8.2 Types of Maps mapped straightforwardly. The identification of features such

as ‘ablation moraine’ and even ‘till’ (sic) (e.g., Keser, 1982)
Although some geomorphological maps may aim to be en- requires understanding of the subsurface sediments as well as
tirely descriptive most are at least in part interpretive. It is the morphology of the feature. This is true for most con-
virtually impossible to produce a final map that is purely de- structional ice-derived landforms. Equifinality of shape forms
scriptive. To be descriptive the authors mean the grouping and means that precise interpretation requires outcrop. There are
recognition of features on morphological criteria only and many examples in the literature, which combine geomorph-
without the filter of inferred mode of formation (process in- ology and sedimentology (e.g., Evans and Twigg, 2002; Evans
ferences). Purely descriptive maps have the advantage that et al., 2007; Hambrey et al., 2008).
they allow subsequent researchers to apply their own inter- The construction of an inference map is a crucial tool for
pretation but this is of limited value to the general user who understanding the glacial history and environment of an area.
will be typically using the maps for environmental (e.g., The spatial distribution of features and of process environ-
habitat management) or for geotechnical (gravel resources, ments allows the mapper to gain a better insight into the
construction related) purposes. In practice most maps are in- processes associated with the mapped features. This will guide
terpretive and the interpretation is based on either ‘mode of the mapper into the interpretation of the overall glacial history.
formation’ inferences (e.g., land system models) or on age
inferences. The focus on the type of inference relates to the
purpose for which the map has been drawn. 14.8.2.2 Age-Correlation Maps
For many studies, the primary interest in glacial deposits relate
to the age of events and subsequent links to climate or other
14.8.2.1 Process Inference Maps
forcings. Age-correlation maps should only be produced after
Glacial features are identified from a combination of morph- a process inference map has been constructed.
ology and ground verification that generally includes exam- The advantage of this sort of map is that it systematizes the
ination of available outcrop. Features such as circular glacial sequences and makes the field interpretation easier
depressions on an outwash plain are related to the mode of for future workers. A classic example is the mapping of sets
formation (in this case the melting of buried ice) and can be of moraines and outwash terraces in the German Alps to
94 Methods in Geomorphology: Mapping Glacial Features

provide a glacial sequence for the North German Plain that settings. Glacial Land Systems (Evans, 2005) provides a good
generated the famous, Gunz, Mindel, Riss, Wurm series of overview of the major types. The advantage of a land system
glaciations (Penck and Brückner, 1909). Schemes of this type approach is that it not only identifies what sorts of landforms
dominated world glacial literature for many decades. For ex- you might find in any particular glacial reach but it also in-
ample, early work in New Zealand adopted a very similar dicates how the landscape elements might be assembled.
terrace and moraine mapping format (Gage, 1958; Soons, This becomes particularly important in dynamic locations
1963). The disadvantage with these earlier efforts was that the such as the Himalaya, which have high relief from rapid uplift.
age correlations were unsecured. More recently, the arrival of In this environment there are ever-present mass movements and
dating techniques that allow the direct dating of glacial de- catastrophic landslides, landforms of which have similar forms
posits (e.g., surface exposure dating using cosmogenic radio- to moraines (Hewitt, 2009). Multiple techniques have been
nuclides and optically stimulated luminescence) has resulted used to understand the glacial history of these areas, which
in renewed interest in this type of mapping for correlative include sedimentological descriptions (e.g., Benn and Owen,
purposes. This is generally tied to linking climate or other 2002; Hambrey et al., 2008; Seong et al., 2009), coupled with
regional events to the chronology (e.g., http://maps.gns.cri.nz/ geomorphic studies, many of which are at modern margins.
website/csigg/; Schäfer et al., 2006). A number of additional resources can be useful, such as
The process of composing an age-correlative map means teaching sets that have aerial photograph stereo pairs that can
that the mapper has to rank the landforms so that features be compared against a completed map (e.g., Evans and Twigg,
such as moraines are subdivided into arbitrary categories 2002). Published sets of aerial stereo-photographs also have
based on age inferences. These groupings may or may not been preinterpreted. For example, the ‘Interpretation of Land-
reflect the geomorphic significance of the landforms. These forms from Aerial Photographs’ (Keser, 1982) provides a basic
maps commonly focus on lateral/terminal moraines and their guide to the interpretation of glacial landforms and all other
associated outwash plains and terraces. Scales are character- common landform types in British Columbia. The authors
istically small and hence detail is lost in the final map pro- particularly recommend this kind of book as a supplement to
duction. Significant degradation of geomorphic information what they write here, as they cover some glacial landforms that
occurs where the correlations are made. Consequently, the are not present in our New Zealand-based examples.
authors recommend that age-correlation maps are published Photographs from modern and historic glacier margins in
in tandem with the process-inference maps. similar climatic and tectonic settings can help visualize the
glacier in the field area (e.g., NSIDC/WDC for Glaciology,
Boulder, compiler, 2002). In many parts of the world National
14.8.3 Identification of Features Parks and other sources provide historical photographic se-
quences and often easy access to aerial photographs that allow
Before mapping can be started, the mapper needs to have the reader to examine a changing margin through time and
some understanding of what glacial features look like. Texts, observe glacial features forming and being destroyed. These
such as ‘Glaciers and Landscapes’ (Sugden and John, 1976) or modern margins are by themselves not a complete record of
the more recent ‘Glaciers and Glaciation’ (Benn and Evans, all possible landforms that can occur. For example, in New
1998) and ‘Glacial Geology: Ice Sheets and Landforms’ Zealand many modern glaciers currently terminate into lakes
(Bennett and Glasser, 2009), are a good place to start. The and are retreating. As a result there are very few modern gla-
types of features will depend largely on the field location; ciers that form substantial end moraines (e.g., Fox and Franz
therefore assessing the local literature is also important. Josef Glaciers). Hence it is important to consult the literature
Compiling a list and short descriptions of features likely to in conjunction with studying modern margins.
occur in the specific field area, from this existing literature, is Take advantage of modern technologies such as Google
recommended. EarthTM, which is an excellent facility to look at global ex-
Note on terminology: Glacial geomorphology and the re- amples of glacial features and moreover allows viewing from
lated field of glacial sedimentology is jargon rich. Commonly both vertical and low-angle oblique perspectives. Once you are
multiple terms exist for similar features and the use of a spe- familiar with the literature and have looked at images of the
cific term over another depends on the training and back- area, you can proceed to mapping.
ground of the individual scientist. The authors suggest that
time should be taken to understand the range of terms and the
different circumstances in which they are used. This allows the 14.8.4 Production of a Base Map or Image
compiler of the map to compare these to the field area. It is
imperative to be consistent with the terms chosen. Describing Generally a base map is compiled onto which the landforms
the mapped features in accompanying text (or on map) is are drawn. This will typically show topography and major
useful for comparison to other field areas where different geographic features such as roads and streams. Alternatively,
terminology is employed. and in our view preferably, a photogrammetrical or satellite
The authors strongly recommend that glacial land systems image can be used. For most people, visual images are easier to
literature is read before commencement of fieldwork. Glacial map onto than a topographic base-map but the choice de-
features vary depending on climatic (polar vs. temperate; high pends on personal preference and available resources. The
precipitation vs. low precipitation), topographic (valley vs. reason why imagery is normally preferred is that many of the
piedmont vs. continental), geological (sediments vs. hard features can be observed and mapped directly onto the image.
crystalline rock), and tectonic (rapidly uplifting vs. stable) For a base map an extra interpretation step is required.
 Methods in Geomorphology: Mapping Glacial Features 95

14.8.4.1 What Constitutes a Good Base Map or Base changes on modern glaciers so long as rigorous quality as-
Image? surance procedures are followed.
Aerial photographs remain the primary mapping tool for
It should have a scale that is appropriate to the mapping (i.e.,
glacial geomorphology. The scale is a function of the camera
roughly twice the scale of the intended final product). It must
lens and the elevation flown, but aerial photographs with
have features that are easily locatable in the field – linear
ground scales of 1:15 000–1:30 000 are appropriate for typical
features such as roads, rivers, fence lines, and treelines are very
field mapping scales and both larger and smaller scale images
valuable tools for locating yourself on a map. It should have
can be obtained. It is not prohibitively expensive to get aerial
registration points, which are map coordinates of a type that
photograph runs taken for a specific project. This potentially
can be read from a GPS and should preferably be derived from
will become more cost effective with the development of
the local topographic map coordinate system.
remote-controlled aircraft. Both false color and black and
white photos are commonly available and both are useful.
False color tends to provide data in infra-red wavelengths,
14.8.4.2 Images
which are optimal for distinguishing vegetation. In areas
Satellite images have been available for general users since the where vegetation type is linked to geomorphology (quite
1970s. There are many different satellites and many different commonly) this can provide help in distinguishing features.
scanning systems in operation. They share the characteristic Black and white photos tend to be sharper and easier to in-
that they involve scanners at a variety of different wavelengths terpret. In ideal circumstances, the images should be ortho-
(mostly infra red and optical) and they have coarse pixel sizes rectified, which will allow the images in a GIS program to be
(typically in the tens to hundreds of meters). Landsat is the overlain with other images and vector data. This becomes
oldest and best known of these satellite platforms. The current important if high precision is required, as scales will change
Landsat 7 has an operating elevation of 705 km and follows a away from the center of the image and is especially important
sun synchronous polar orbit. It has mesoscale resolution be- in high-relief areas.
tween 15 and 60 m using eight different spectral bands. One One of the valuable aspects of using aerial photographs is
of the advantages is repeat cover (16 day intervals for Landsat that multiple series of photographs commonly were taken
7). This feature is useful for monitoring seasonal and annual over an extended period of time. In many areas of the world,
changes on modern glaciers but of lesser value for mapping the first good sets of aerial photographs date to the 1940s. This
formerly glaciated areas. Overall the limited resolution of is helpful because there are widespread changes in land use
satellite images make them ancillary rather than primary since that time and in many formerly glaciated areas, land has
mapping tools, but they have important roles to play in large been converted to forestry (due to its low agricultural value),
area mapping and in the mapping of larger scale features. which obscures the geomorphology. These older aerial
Associated with satellite data are a series of widely available photographs tend to be flown at lower elevations than mod-
geospatial image platforms. Perhaps the best known is Google ern runs and are consequently at larger scales. In addition,
EarthTM, which provides access to satellite images and geo- different sets of aerial photographs highlight different features.
graphic information system (GIS) overlay information for the If multiple sets are available, use them all. Different runs are
whole planet. The ability of this platform to allow the user to taken during different light conditions and different seasons,
view features from different elevations (scales) and per- which may result in some landforms being more easily dis-
spectives is very helpful. In base format it is also currently free. tinguishable in some runs than others. This is also true for
GeoEye and Ikonos are two commercial satellite platforms satellite images. If using aerial photos to create a base image,
that provide similar images and a wide range of geospatial choose aerial photographs with the clearest image that show
information. Radarsat and Terra-SAR X are commercial plat- the majority of the landforms and have the largest scale.
forms that provide access to radar images. There are other Like oblique ground photos, oblique aerial photographs
platforms and more are added continuously. Access and are an excellent tool to support mapping. These should be
charges for data vary widely. taken in the late afternoon or early morning, as shadows
LiDAR (light detection and ranging) is based on laser re- highlight subtle morphological features (see Figure 2). They
flection of the ground surface with the time of the reflection are, however, completely unsuitable for producing base im-
pulse measured to determine distance to the surface. LiDAR ages as the image is distorted by the field of view of photog-
can be used to generate a digital elevation model (DEM) for raphy, which gives a strong parallax error.
an area with cm–dm accuracy. LiDAR can distinguish between
foliage responses and bare earth responses and provide con-
tour information from the bare earth DEM, which is signifi-
14.8.4.3 Stereoscope Use
cant in areas with tree cover, but very dense foliage will defeat
LiDAR as it requires some reflection from the ground to make The stereoscope is a critical tool for mapping landforms. Using
its measurements. It is very expensive data to obtain but gives the stereoscope with stereo pair of aerial photographs enables
high-precision results. There are still some significant issues of a 3D visualization of the photos, like viewing an area from a
data quality assurance (e.g., Choma et al., 2005) that mean plane. In glacial systems this allows the easy distinction of
that LiDAR is not a replacement for aerial photography or morphological features such as moraines and eskers. A light
planimetric mapping of ancient glacial deposits. It can provide directed onto the aerial photographs is useful in most cases, as
high-quality inferences about surface shapes and is excellent the stereoscope throws shadows on the photos from overhead
for the construction of detailed surface maps to measure lights. An adjustable table lamp is perfect for this job.
96 Methods in Geomorphology: Mapping Glacial Features

iii
ii
i
iii

i ii

(a) (b)

iiii

ii

(c)

Figure 2 Two photographs taken from Mount John and a satellite image, near Lake Tekapo, South Island, New Zealand. Photograph (a) is a
photo taken in average light conditions, whereas (b) was taken in the early evening, just before a summer sunset. Note how landforms are
highlighted with the contrast with the shadows in (b) which allows subtle features to be seen in the landscape. Image (c) is a satellite image
taken after a snowfall and sourced from Google Earth. Features labelled are as follows; (i) is a meltwater channel, with braided channels,
(ii) moraine ridge, (iii) streamlined ridges/flutes.

One point that needs emphasis is that aerial photographs should be) revised regularly. The outcomes of the initial
and satellite imagery should be constantly revisited during the mapping will direct in most cases the scale and extent of the
course of the mapping project. The authors guarantee that we final map. For example, if after initial mapping more features
will see many more features on our final inspection than we were identified that originally estimated, an increase in scale
did on the first. It is recommended to take a stereoscope on might be necessary to adequately convey all the details.
field trips, so aerial photos can be consulted in the evening
and during bad weather. This can assist in deciding which
areas in the field to visit on foot and is a useful comparison as 14.8.4.5 Map Extents
generally different landforms are identified in the field verses
The map extent might seem obvious initially, but it is another
through the stereoscope.
common part of the mapping process that changes during a
project. As with the map scales, when defining (and re-
defining) the map extent, consider the target age (if any) of the
14.8.4.4 Scale of Mapping landforms, spatial characteristics of landforms, and map detail
required. In many circumstances, not all of the mapping is
Always map at a larger scale than your intended final product.
included in the final map. This can result for a variety of
As a rule of thumb, map at roughly twice the scale that you
reasons: cartography, does not fit on the destined media,
intend the final map to be. If we want to map an area at
change in scale, insert orientation maps or other figures,
1:50 000, produce 1:25 000 scale field maps. There are two
location of legend, etc.
reasons for this; (1) by mapping at a larger scale we are able to
provide four times the detail, and (2) when the final map is
produced, field errors are reduced by half as the scale doubles.
If the mapping scale is too much larger than the final map 14.8.5 Field Mapping
scale, however, it becomes infeasible to transfer all the detail to
the final map. The first task is to define features and landform groups.
If undertaking mapping in an area with very little previous Numerous schemes exist for marking up glacial features, but
literature or a remote location, deciding on an initial scale is there are fundamentally two types of map markings – symbols
not always straightforward. In such cases aim for as large as for individual features such as moraine ridges and drumlins;
scale as initially feasible. The scale of the map can be (and and block colors or codes for surface cover. Some maps
 Methods in Geomorphology: Mapping Glacial Features 97

Figure 3 Examples of symbols for glacial features and landforms.

employ symbols instead of colors for surface cover as well. For Try not to get too bogged down in identifying each exact
detailed maps it is highly recommended to use color, as a map landform everywhere. Elements in the field can change rapidly.
with careful use of complementary colors convey geomorphic In our experience a ridge on the side of a glacial valley can (and
information better. A basic glacial geomorphology symbol does commonly) change from a kame terrace to a lateral mo-
scheme is provided in Figure 3. raine and back again over a 100 m or less. Furthermore, the
The authors recommend us to map onto MylarTM or other same ridge(s) can rapidly switch from constructional features
transparent waterproof media rather than directly onto the such as lateral moraines to erosional ice contact features such as
image or basemap. This is because field mapping is an iterative trimlines. New mappers characteristically struggle with this
process and the basemap or image will rapidly become de- variability. Remember what the final scale of the map is going
graded. In addition the overlay provides rainfall protection for to be – most of this sort of variability will not be captured by
the image/map. Laminating the base map is desirable but it the final map unless it is exceptionally detailed, so do not stress.
does interfere with high-precision visualization of the image. Do not be timid in initial feature identification. It is best to
If used, a matt lamination is better than a glossy lamination. make a guess at what the feature is, which can then be changed
They also recommend that all mapping done with a pencil later. Difficult areas or individual features generally resolve
and a good quality eraser is very valuable. Expect numerous themselves when observing the landform associations and
iterations. patterns across the whole map.
When first visiting the field area, it is very useful to get an
elevated view of the area, from a hilltop or valley slope. This
14.8.5.1 Examining Outcrop in Association with Mapping
allows a perspective over the field area or parts thereof, where
you can observe landforms, spatial relationships, and patterns. Views on the role of geomorphological mapping vary, but the
These vantage points are especially valuable in the early authors strongly support the concept that surface morpho-
morning or late afternoon, when the sun angle in the sky is low. logical mapping is greatly enhanced if it is supplemented by
The shadows cast by landforms, are especially good at making sedimentary observations of outcrop (if available). Although
subtle features visible. For example, photograph (b) in Figure 2 morphologically incorrect, they suggest that features may be
is taken in the late afternoon at Mt John near Lake Tekapo in subdivided by sedimentary characteristics if the outcrop is
the Southern Alps of New Zealand; braided channels and sufficient to this purpose (e.g., Evans and Twigg, 2002).
moraine ridges are highlighted by the low-light angles. It is It is important that careful consideration is given to the
worthwhile moving around and changing your perspective – context of the landform contents. For morphology such as a
different features become visible from different angles. small hill, three types of content exist. Firstly, the hill may be an
Once an area has been mapped remotely (in the lab or the erosional form where the hill shape has been carved out of
top of the nearby hill) it is critical to validate the mapping by bedrock. In this case the hill will form a roche moutonée or
walking the areas mapped. This is best achieved by examining equivalent landform. Hills can also be formed from the de-
examples of all types of features mapped. Particular emphasis position of materials either directly from the ice or from sub- or
should be placed on areas of complexity and on features that englacial flows. The most challenging case is where the form is
are relevant to the study for which the mapping is being erosional but the fill is sedimentary, especially if the fill is de-
undertaken. It is not necessary to walk every square meter of a rived from glacial or glacifluvial sediments. This is a widespread
mapped area. problem in mid-latitude continental deposits and becomes
98 Methods in Geomorphology: Mapping Glacial Features

exceptionally challenging in some temperate valley glacier set- extended periods as there is a marked change in mass balance
tings where ice has over-run proglacial sediments (outwash or required to extend beyond the confined valley opening (see,
lake beds) or previous glacial deposits (Shulmeister et al., e.g., Anderson and Mackintosh, 2006).
2010). Geomorphological maps will certainly distinguish be- The loss of features from postglacial fluvial erosion or
tween the bedrock and sedimentary forms and in most cases burial is highlighted in New Zealand by contrasting the out-
the mapper will want to distinguish between all the erosional standing preservation of features in the Heron intermontane
and depositional forms. basin (see Figures 4 and 5) with the preservation of features in
the floor of the adjacent main Rakaia Valley. The former area
preserves almost every glacial feature of the last recession be-
cause the Rakaia glacier fed distributary lobes into this basin,
14.8.6 Mapping in Different Glacial Settings – Case
which became isolated from the main valley when incision
Studies
occurred in the main valleys during the early stages of ice
recession. In contrast, the main Rakaia Valley has incised by
14.8.6.1 Temperate Valley Glaciers – East Coast South
tens of meters since the deglaciation and melt-water features
Island, New Zealand (Rakaia and Rangitata
dominate much of the landscape, even on the preserved
Valleys)
higher terraces (Figure 6).
A common setting for glacial mapping is in formerly
(or presently) glaciated mountain valleys and associated
piedmont lobe settings beyond the mountain fronts. 14.8.6.2 Former Ice-Cap Covered Regions
These systems include most of the best known and described One special case is the mapping of formerly glaciated areas
glacial landforms including cirques, arêtes, and horns in under the source (distributary) area of continental ice caps. At
upper valley systems and truncated spurs, lateral moraines, the heart of the ice cap, flow rates are very low and preglacial
terminal moraines, and trimlines in the lower valleys and landforms at the base have a high probability of being pre-
in the upper valleys where recession has occurred. The served. This has been demonstrated for the heart of the Fenno-
major morphological challenge is the distinguishing of Scandian ice cap in northern Europe where preglacial
mass movement deposits from glacial landforms and this periglacial features such as tors are preserved in regions where
may be far from straightforward as equifinality of morpholo- kilometers of ice cover are inferred (e.g., Fabel et al., 2002).
gies characteristically occur (Hewitt et al., 2008; Hewitt, This type of landscape is very difficult for glacial re-
2009). constructions and this example leads to two other settings
One of the classic features of glacial valleys is the vertical where glacial geomorphological mapping is challenging: polar
nesting of glacial features whereby features that relate to areas and karst landscapes.
younger advances are located lower in the valley floor and
further up valley than features associated with older and
bigger advances. This is because bigger advances destroy 14.8.6.3 Antarctica and Other Dry Polar/Alpine
the evidence of smaller ones so that the preserved record Environments
is a time-size sequence. In fact, statistical analyses show
that this process of glacial overprinting will typically In areas where dry (cold) based ice is inferred, the preservation
leave evidence for only two to three phases of glaciation of glacial features tends to be very limited. Cold-based ice
(Gibbons et al., 1984) irrespective of how many have moves slowly and mainly by internal deformation (e.g., Fitz-
occurred. Figure 4 displays a relative age correlation map for simons, 2005). As a result it erodes very little (Atkins et al.,
the Heron lobe of the former Rakaia Glacier (from Pugh, 2002) and if it is detached from slopes where freeze–thaw can
2008). Note the concentric arrangement of moraines and provide supraglacial debris, it will carry very little debris also.
outwash pairs. Polar glaciers may leave no trace, or a diffuse trail of ground
In areas with substantial fluvial activity, preservation of moraine or a range of terminal and lateral features, but except
glacial deposits after recession is limited by postglacial fluvial in areas where fast-moving ice streams occur, the glacial land-
erosion. Burial of features under alluvial outwash fans and scapes are subtle and the landscape itself often palimpsest.
fans from valley sides can also occur. This is a minor problem
in more arid areas and in areas with gradual release of
14.8.6.4 Karst Areas
melt-water (high polar regions). It is very significant in high-
rainfall, high-sediment flux areas such as South Island, New The recognition of glacial features in karst (limestone solution
Zealand. The consequence is very limited preservation of gla- dominated) areas is challenging. The issues relate to the effi-
cial features at lower elevations in valley floors due to fluvial ciency of subglacial drainage on the karst that results in re-
reworking and a tendency for moraines to become larger up duced ice flow and limits the efficiency of glacial processes and
valley as the catchment becomes smaller, because the thor- to the modification of the landscape by karst processes after
ough flow of water is reduced. Another feature is the strong glaciations has ceased.
preservation of glacial deposits at the point where the valley Glacial erosional features such as cirques, ice-smoothed
breaks out of its mountain constraint. Again, this is due in the bedrock, and U-shaped valleys are commonly preserved. There
main to the dissipation of the fluvial discharge at this location, are, however, karst processes that produce similar features such
allowing for the preservation of deposits. There is also a ten- as box canyons that are not easy to distinguish from glacial
dency for the glacier to remain close to the valley mouth for features. There are also a number of karst features such as
 Methods in Geomorphology: Mapping Glacial Features 99

Lake

800
0 0.5 1 2
km
PF moraine

800
DH moraine
EM moraine

700
Mt Sugarloaf EM outwash
JS moraine
JS outwash
700 LH 1 moraine
LH 1 outwash
Lake LH 2 moraine
heron
LH 2 outwash
LH 3 moraine
LH 3 outwash
Post LH moraine

800
Ricki spur
–700– Contour (100 m)

900
Isolated hill

80
0
1100
800

Emily hill

900
700

800
Lake
emily

0
90
700

11
00

Maori
Lakes
Clent hills
700

Figure 4 Glacial formation map of the Lake Heron lobe of the Rakaia Glacier, South Island, New Zealand from Pugh, J., 2008. Late Quaternary
environmental history of the Lake Heron Basin, Mid-Canterbury, New Zealand. MSc Thesis, University of Canterbury, New Zealand, unpublished.
This map identifies morphological pairs of moraines and outwash to generate a relative chronosequence of glacial positions. Younger positions
are nested inside the older larger advances. The sequence of advances from oldest to youngest is Pyramid Formation (PF), Dogs Hill (DH),
Emma (E), and the Lake Heron (LH) 1–3 positions.

limestone pavements that are regarded as glacio-karst in generally associated with inliers of other lithologies. In short,
origin (e.g., Williams, 1966; Hyatt et al., 2007). Depositional the recognition of glacial morphologies on karst is extremely
features tend to be extremely rare on pure karst as solution challenging (see Hyatt et al., 2007) and as a result there is a
dominates the subglacial processes and few sediments relative paucity of mapping investigations in karst areas (see
are produced. If constructional features occur they are Figure 7).
100 Methods in Geomorphology: Mapping Glacial Features

capabilities but CAD-based systems are somewhat simpler to

master and also give good results.
One of the most useful aspects of using GIS programs is the
ability to display multiple images (rasters) and vector layers at
the same time. GIS is a good tool in composing the base
image/map, in which the user is able to overlay a number of
layers, such as an orthophotograph with overlain topographic
contours.
Other options for smaller maps is to use graphics programs
such as Adobe Photoshop and Corel. These are not georefer-
enced, hence are not as easily manipulated as vector maps in
GIS programs, but some maps created in these graphics pro-
grams can be imported into GIS programs and georeferenced.
These graphics programs are particularly useful in the final
map production stage, such as inserting location images and
titles. The final map can be exported from a GIS program and
Figure 5 Terminal ridges with kettle holes and localized drainage imported into a graphics program, where it is often easier to
channels in the Heron Basin at the north end of the valley combine multiple elements and produce a polished final
immediately east of Ricki Spur (see Figure 4). Note the exceptional product.
preservation of low amplitude features. Reproduced with permission
from Pugh, J., 2008. Late Quaternary environmental history of the
Lake Heron Basin, Mid-Canterbury, New Zealand. MSc Thesis, 14.8.7.2 Map Elements
University of Canterbury, New Zealand, unpublished.
As well as the glacial map, a number of other recommended
and optional elements are included in the final map presen-
tation. The final map should have a title, map legend/
key, scale, and details of the mapping production such
14.8.7 Map Production/Cartography
as coordinate system and materials used (e.g., aerial
photographs and digital data). It is a good practice to have a
Cartography is an important consideration. A wide range of
small map insert that shows the location of the mapped area
glacial map styles occur in the literature and it can be difficult to
in respect to the region/country/world. Locations of these
decide on a style. A useful resource when determining the final
elements can be located outside or within the map area, de-
map style is to visit Journal of Maps (www.journalofmaps.com),
pending on the map characteristics. Some maps have areas
which have a good number of glacial maps of various styles.
that have sparse or absent glacial features (e.g., ocean,
The style of map will largely be determined by scale, size in final
mountain tops), which can be good areas for the legend or
publication (paper, poster, thesis etc.), detail, and spatial char-
location map.
acteristics of mapped features, mapping program, and the main
Many maps only include these basic map elements
focus of the mapping undertaken. All these criteria influence
(e.g., Kjaer et al., 2008; Sahlin and Glasser, 2008; Evans et al.,
not just the final format of the map but also how the mapping
2009), but many maps also include other material (e.g., Evans
is undertaken.
et al., 2007). These maps include aerial photographs, DEMs,
sketches, sediment sketches and photographs, and short de-
scriptions of the landforms and features included in the map.
These are ideal for conference presentation and standalone
14.8.7.1 Map Compilation – Digitizing Data and Producing publications.
the Map
Although the publication of analog maps continues, most
14.8.7.3 Publication
users will want to convert their field maps into digital format,
as this greatly enhances the ability to adapt map data to spe- Although some glacial geomorphology maps are profession-
cific purposes. Two of the most widely used platforms for map ally published (which is beyond the scope of this chapter) the
compilation are ARCinfo/ARCmap and OCAD. OCAD majority of glacial geomorphology maps will be published in
(http://www.ocad.com) is a dedicated cartographic software thesis, papers, and conference posters. The authors strongly
system based on a computer-aided design (CAD) program that urge that the maps be made available as digital data and in
allows both scanning and conversion of analog map data and such a manner that the layers and individual features can be
full data handling of vector data. ARCinfo (http://www.esri.- edited so that maps can be modified and upgraded by future
com) is GIS designed for mapping a wide variety of data types. users. At this end we highly recommend that the maps are
ARCmap is the main geospatial data-handling component of produced using a widely available software system and stored
ARCinfo and is the mapping tool in ARCinfo. Many other as vector (shape) files in a data depository. A PDF version only
systems exist, including some open source options such as of a map is inadequate.
GRASS GIS (http://grass.itc.it/). Most systems used are GIS- There is one specialized venue for the publication of
based and provide much more than simple mapping geomorpological maps and associated data. This is the
 Methods in Geomorphology: Mapping Glacial Features 101

495
500 m

(a) (b)

(c)
Glacial landforms

Hummocky moraine Push moraine

Meltout Flutings
Overridden bedrock and sediment Kame terrace

Outwash Ice-contact slope

Former proglacial lake deposits Lake bench or beach

Kettle holes
Alluvial fan

Modern lakes and ponds Melt water channels

Bedrock and unmapped areas Abandoned channels

Terraces
Miscellaneous
314 Spot heights in metres
Streams and rivers Road •

400 Contours (20 metres)

Contours (100 metres)
Figure 6 (a) An example of a superficial glacial geomorphological map from the Rakaia Valley, South Island, New Zealand with (b) an aerial
photograph of map extent and (c) a photograph taken from across the river, looking east toward the map area of A. Note the extensive terrace
sequence with outwash channels and the consequently limited preservation of direct glacial features. The figure modified from Hyatt, O.M., 2010.
Insights into New Zealand glacial processes from studies of glacial geomorphology and sedimentology in Rakaia and other South Island Valleys.
PhD Thesis, University of Canterbury, New Zealand, unpublished.

Journal of Maps (http://www.journalofmaps.com/). The Journal of Maps. The journal operates on a creative commons
authors highly recommend that if good-quality glacial geo- copyright that means open distribution. The journal has a
morphology maps are produced that are not going to be strong focus on glacial mapping and also has special graduate
substantively published elsewhere, that they are submitted to student map issues. It was set up to compliment other data,
102 Methods in Geomorphology: Mapping Glacial Features

Scale 1:34 000 Grid references, map and topographic features based on topographic sheet 260 - M28

Glacial landforms Topographic features

Lateral and/or terminal moraine 1732 Height (metres) Lake and ponds

Kame terrace Track Native forest

U-shape valley sides Cliff

Diamicton Contour (20 m)

Other landforms
Suffused alluvium Contour (100 m)
Karst depression
Ice rounded knobs Streams and rivers
Scree
Swamp
Figure 7 Glacial mapping in karst area. Simplified glacial and karst geomorphology map of the Mount Owen Massif in northwest Nelson, South
Island, New Zealand. Map modified from Hyatt, O.M., Shulmeister, J., Smart, C.C., 2007. A reconnaissance study of glaciation on the Owen
Massif, Northwest Nelson, New Zealand. Quaternary Australasia 24, 11–18, with permission from AQUA.
 Methods in Geomorphology: Mapping Glacial Features 103

descriptions, and interpretations published in traditional Gage, M., 1958. Late Pleistocene glaciations of the Waimakariri Valley, Canterbury,
journals. New Zealand. New Zealand Journal of Geology and Geophysics 1, 123–155.
Gibbons, A.B., Megeath, J.D., Pierce, K.L., 1984. Probability of moraine survival in
a succession of glacial advances. Geology 12, 327–330.
Hambrey, M.J., Quincey, D.J., Glasser, N.F., Reynolds, J.M., Richardson, S.J.,
Acknowledgments Clemmens, S., 2008. Sedimentological, geomorphological and dynamic context
of debris-mantled glaciers, Mount Everest (Sagarmatha) region, Nepal.
The authors thank coworkers especially Dave Evans for advice Quaternary Science Reviews 27, 2361–2389.
and assistance in field mapping over the last few years. Jeremy Hewitt, K., 2009. Catastrophic rock slope failures and late Quaternary developments
in the Naga Parbat-Haramosh Massif, Upper Indus basin, northern Pakistan.
Pugh made figures from his MSc thesis available to us.
Quaternary Science Reviews 28, 1055–1069.
Hewitt, K., Clague, J.J., Orwin, J.F., 2008. Legacies of catastrophic rock slope
failures in mountain landscapes. Earth Science Reviews 87, 1–38.
References Hyatt, O.M., Shulmeister, J., Smart, C.C., 2007. A reconnaissance study of
glaciation on the Owen Massif, Northwest Nelson, New Zealand. Quaternary
Anderson, B., Mackintosh, A., 2006. Temperature change is the major driver of late- Australasia 24, 11–18.
glacial and Holocene glacier fluctuations in New Zealand. Geology 34, 31–34. Keser, N., 1982. Interpretation of Landforms from aerial Photographs. Province of
Atkins, C.B., Barrett, P.J., Hicock, S.R., 2002. Cold glaciers do erode and deposit: British Columbia, Ministry of Forests, Victoria, 217 pp.
evidence from Allan Hills, Antarctica. Geology 30, 659–662. Kjaer, H.H., Korsgaard, N.J., Schomacker, A., 2008. Impact of multiple glacier
Benn, D.I., Evans, D.J.A., 1998. Glaciers and Glaciation. Arnold, London. surges – a geomorphological map from Bruarjokull, East Iceland. Journal of
Benn, D.I., Owen, L.A., 2002. Himalayan glacial sedimentary environments and Maps 2008, 5–20. http://dx.doi.org/10.4113/jom.2008.91.
dating former extent of glaciers in high mountains. Quaternary International NSIDC/WDC for Glaciology, Boulder, compiler, Boulder, compiler, 2002. Updated
97–98, 3–25. 2009. Glacier Photograph Collection. Boulder. National Snow and Ice Data
Bennett, M., Glasser, N., 2009. Glacial Geology: Icesheets and Landforms. Second Center/World Data Center for Glaciology, Colorado USA, Digital media.
ed. Wiley, London, 400 pp. Penck, A., Brückner, E., 1909. Die Alpen im Eiszeitalter. Tauchnitz, Leipzig, . 1199p.
Choma, A. Ratcliff, C. Frisina, R., 2005. Evaluation of remote sensing technologies Pugh, J., 2008. Late Quaternary environmental history of the Lake Heron Basin,
for high-resolution terrain mapping. Proceedings of the SSC 2005 Spatial Mid-Canterbury, New Zealand. MSc Thesis, University of Canterbury, New
Intelligence, Innovation and Praxis: The National Biennial Conference of Spatial Zealand, unpublished.
Sciences Institute, September 2005. Spatial Sciences Institute, Melbourne, Sahlin, E.A.U., Glasser, N.F., 2008. Geomorphological map of Cadair Idris, Wales.
Australia. ISBN:0-9581366-2-9. Journal of Maps 2008, 299–314. http://dx.doi.org/10.4113/jom.2008.1011.
Evans, D.J.A. (Ed.), 2005. Glacial Land Systems. Oxford University Press, Oxford, Schäfer, J.M., Denton, G.H., Barrell, D.J.A., et al., 2006. Near-synchronous
542 pp. interhemispheric termination of the last glacial maximum in mid-latitudes.
Evans, D.J.A., Twigg, D.R., 2002. The active temperate glacial landsystem: a model Science 312, 1510–1513.
based on Breidemerkurjokull and Fjallsjokull, Iceland. Quaternary Science Seong, Y.B., Bishop, M.P., Bush, M., et al., 2009. Landforms and landscape
Reviews 21, 2143–2177. evolution in the Skardu, Shigar, and Braldu valleys, central Karakoram.
Evans, D.J.A., Twigg, D.R., Rea, B.R., Orton, C., 2009. Surging glacier landsystem Geomorphology 103, 251–267.
of Tungnaarjokull, Iceland. Journal of Maps 2009, 134–151. http://dx.doi.org/ Shulmeister, J., Thackray, G.D., Rieser, U., Hyatt, O.M., Rother, H., Smart, C.C.,
10.4113/jom.2009.1064. Evans, D.J.A., 2010. The stratigraphy, timing and climatic implications of pre-
Evans, D.J.A., Twigg, D.R., Rea, B.R., Shand, M., 2007. Bruarjokull: Surficial
LGM glaciolacustrine deposits in the middle Rakaia Valley, South Island, New
Geology and Geomorphology of the Surging Glacier Landsystem. 1:30 000 Scale
Zealand. Quaternary Science Reviews 29, 2362–2381.
Map Poster. Universities of Glasgow, Durham, Loughborough and Aberdeen.
Soons, J.M., 1963. The glacial sequence in part of the Rakaia Valley, Canterbury,
Fabel, D., Stroeven, A.P., Harbor, J., Kleman, j., Elmore, D., Fink, D., 2002.
New Zealand. New Zealand Journal of Geology and Geophysics 6, 735–756.
Landscape preservation under fennoscandian ice sheets determined from in situ
Sugden, D., John, B.S., 1976. Glaciers and Landscape: A Geomorphological
produced 10Be and 26Al. Earth and Planetary Letters 201, 397–406.
Fitzsimons, S.J., 2005. Mechanical behaviour and structure of the debris-rich basal Approach. Wiley, New York, 376 pp.
ice layer. In: Knight, P. (Ed.), Glacier Science and Environmental Change. Williams, P.W., 1966. Limestone pavements with special reference to western
Blackwell, pp. 251–259. Ireland. Transactions of the Institute of British Geographers 40, 155–172.

Biographical Sketch

Olivia Hyatt is a glacial geomorphologist and sedimentologist, with a keen interest in glacial interactions with
climatic and geological processes. She recently completed her PhD in early 2010, which investigated the nature of
glacial processes operating through time in South Island, New Zealand.
She is currently working as Climate Change and Occupational Health consultant based in Nelson, New Zea-
land, but continues glacial research in New Zealand in conjunction with colleagues around the world.
104 Methods in Geomorphology: Mapping Glacial Features

James Shulmeister is a geomorphologist and paleoclimatologist with a long-standing interest in glaciers and
landscapes. He is currently Professor and Head of School of Geography, Planning, and Environmental Man-
agement at the University of Queensland in Brisbane, Australia but worked for nearly 20 years in New Zealand. He
has published widely on glacial geomorphology, sedimentology, and geochronology focusing on New Zealand
but with published work from Antarctica and Canada also. He is currently involved in glacial projects in New
Zealand and the western United States Recent work has targeted the causes of glacial advances and retreats in New
Zealand.
 14.9 Techniques and Methods for the Field: An Introduction and Commentary
AD Switzer, Earth Observatory of Singapore, Nanyang Technological University, Singapore
DM Kennedy, The University of Melbourne, Parkville, VIC, Australia
r 2013 Elsevier Inc. All rights reserved.

14.9.1 Introduction 105

14.9.2 What’s on Top? – Studying the Surface 105
14.9.2.1 Collecting Information Before You Leave for the Field 105
14.9.2.2 Once You’re Out There 106
14.9.3 What Lies Beneath? – Subsurface Investigations in the Field 106
14.9.4 Back in the Laboratory 107
14.9.5 Never Ignore Safety 107
14.9.6 Value of Fieldwork in Educational Aspects of Geomorphology 107
14.9.7 Conclusions 108
References 108

Abstract

As a subject that deals with the evolution of the landscape, geomorphology is classically considered to be a field discipline.
Although field studies are increasingly supplemented with laboratory (particle size, component analysis, and chemical
techniques) or computer-based studies (e.g., remote sensing, digital elevation models, and virtual globes) much of the data
means very little when applied without fieldwork and observation. This paper provides a brief introduction to the field
studies discussed in this volume, a discussion of preparing for fieldwork, a summary of the main drivers for investigating
landforms both from the perspective of on top and below the ground and a brief commentary on safety in the field and
some philosophical reasons for the importance of fieldwork in geomorphology education.

14.9.1 Introduction limited by the costs and time available to carry out the re-
search. Ideally, a thorough desk study will make the fieldwork
Before starting any field investigation the researcher generally component (generally a far more expensive exercise) more
needs to define the particular area to be investigated. In many efficient and productive.
cases this choice of the site to be investigated may already be A detailed site investigation will generally require the sur-
prescribed, for example, in relation to a specific construction face and subsurface characteristics of a location(s) to be in-
project, hazard assessment or development project. Other vestigated. Fookes et al. (2003) provided an excellent overview
studies involve a regional approach that commonly requires the of the process of preparing for a field study in geomorphology.
balance between competing forces of resources versus com- Other technique-orientated texts and reviews by Goudie et al.
prehensiveness and therefore generally requires the selection of (1994), Slaymaker (1991), McCarthy (2007), and Goudie and
‘representative field sites.’ In the latter situations a selection of Viles (2010) provide good overviews of the process of field-
representative locations to be investigated needs to be made, as work. It is however, generally recognized (and evident through
it will not usually be physically or financially possible to cover the case studies in this treatise) that applied and problem-
the whole terrain. Site selection is clearly a question of sampling orientated questions (as opposed to observational and
and generally an appropriate sampling framework will need to philosophical) require field data that will generally differ ac-
be designed for the desired research questions. In many cases cording to the nature of the investigation. For example, an
the sampling framework may be designed according to the investigation of surface sediment pollution in an estuary (e.g.,
existing, available resources (e.g., maps). see Chapter 14.24) may require sampling to only a few
For the most part, good site selection and investigation is centimeters depth; a study of estuary infill associated with
most effective if it is preceded by a detailed desk study (this Holocene sea level will require sampling at much greater
often includes studying maps, local studies, as well as books or depth, generally several meters (e.g., Whanganui Inlet, New
scientific journals). Of increasing usefulness is the gray lit- Zealand, Kennedy et al., 2008).
erature generally occurring in electronic sources but add-
itionally much information can also be gleaned from reports
of studies that have previously been carried out by planning 14.9.2 What’s on Top? – Studying the Surface
offices, government departments or private companies. Almost
all studies and their associated sampling frameworks are 14.9.2.1 Collecting Information Before You Leave for the
Field
Switzer, A.D., Kennedy, D.M., 2013. Techniques and methods for the
field: an introduction and commentary. In: Shroder, J. (Editor in Chief),
Before leaving the lab or office it is generally a good initial
Switzer, A.D., Kennedy, D.M. (Eds.), Treatise on Geomorphology. Academic exercise to assess terrain conditions in the study area in
Press, San Diego, CA, vol. 14, Methods in Geomorphology, pp. 105–109. terms of the morphometry of the surface (Goudie et al., 1994;

Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00376-6 105

106 Techniques and Methods for the Field: An Introduction and Commentary

see Chapter 14.3). It is generally appropriate to take transported. Studies of such material underpin our knowledge
basic measurements both in terms of sections through the of past climates around the globe and associated changes in
landscape, and to produce geomorphological maps (see wind, storms, atmospheric chemistry, runoff, and flooding to
Chapter 14.4). In addition to investigating the terrain, take name just a few elements of the geomorphic cycle. The sub-
note of the local climate as seasonal changes commonly affect surface is also a key component of the surface; it forms the
the dominant geomorphic processes (e.g., seasonal floods, boundary conditions (i.e., the material) on what the hydro-
tropical cyclones, monsoonal rains, changes in wave energy). logical cycle operates. For example, the damage and impact of
Such information on climatic variability may enable a re- landslides is determined by the actual material that is moving,
searcher to concentrate observations on appropriate features hence the fundamental reason why geomorphologists in-
more effectively, interpret their finding in light of this vestigate the character of subsurface material.
seasonality (e.g., seasonal bed level changes in the Changjiang Studying what lies beneath is a limited exercise in a desk
(Yangtze) River, China; Yang et al., 2001) or visit sites study. Although previous reports (e.g., drilling reports) and
when access is more favorable (e.g., dry season in the tropics). investigations (e.g., environmental impact statements) provide
Another example would be the use of aerial photographs an invaluable study framework, for the most part new
or high resolution satellite images to look at surface relief geomorphological studies will involve some investigation of
features (e.g., channel offsets indicating fault displacement subsurface information. The detail to which investigations
that can be an important indicator of subsurface process are conducted will be determined by the research question
(e.g., Klinger et al., 2011)). In many cases surface conditions that is being asked and commonly data collected for one
provide an important control on geomorphic or geological purpose may be of limited use for another. For example, in-
processes. Within and across sites one generally will be faced vestigations of groundwater permeability may pay little at-
with significantly different soils or regolith. In such cases it is tention to the sedimentological details needed to unravel
important to characterize the landscape in terms of surface the depositional history of a landscape, or the area’s sus-
cover taking into account other variables such as vegetation ceptibility to landsliding. Good note taking, and storage of
and slope. Such information should be noted in the desk data in a repository, can alleviate such problems but most
study and direct measurements or detailed descriptions of type commonly the data collected are restricted to the narrow
and characteristics of the land cover in a study site should confines of the original research project. In many geo-
be made. morphological studies soil type is often recorded; however,
detailed investigations of soil science require specific meth-
odologies (see Chapter 14.15). Generally investigations of
14.9.2.2 Once You’re Out There soils are conducted through the excavation of pits and where
such excavations are expanded over tens of meters width and
Once in the field the depth of such coverage will generally
depth, significant geomorphological information can be ex-
depend on the end user of the information. For example,
tracted. The chapter by McAlpin (see Chapter 14.12) highlights
someone conducting an environmental impact assessment
the potential of large excavations in the investigation of slope
may need a detailed botanical study to assess the potential
stability or for reconstructing past earthquake events. Deep
impacts on vegetation, whereas in contrast, someone looking
trenches are characteristically time consuming to excavate as
at fluvial channel migration in a gravel bed stream may only
well as costly. They are also limited in areas which are per-
need passing mention of vegetation types present (if any). In
manently inundated. As a result many subsurface investi-
this case the levels of treatment contrasts considerably, al-
gations are carried out through coring (see Chapters 14.11 and
though the importance of vegetation as indirect measure of
14.15).
other characteristics, including the water regime, can be of
Coring and trenching are limited to providing information
vital importance in terms of mapping the landscape. Detailed
at the particular site of sampling. The question therefore arises
mapping and surveying in the field is essential; however, a
as to how representative a single core less than three inches
balance always exists between the time required to collect
wide is of an entire floodplain. To overcome this problem
accurate data and the amount of data that is needed to pro-
many contemporary subsurface investigations are combined
duce scientifically valid results. Such considerations are high-
with geophysical techniques such as ground penetrating radar.
lighted in the chapters by Kennedy (see Chapter 14.10) and
Ground penetrating radar is a relatively new technique that
Hyatt and Shulmeister (see Chapter 14.8), which explore the
provides data on surface structure that can be gathered very
methodology of surveying in a general sense and specific
quickly and over a large area. The chapter by Bristow (see
mapping of glacial environments, respectively.
Chapter 14.16), outlines the basic principles of this techni-
que and its application to a range of geomorphologic
environments.
14.9.3 What Lies Beneath? – Subsurface Unless the field site is in an area of particularly thick soil or
Investigations in the Field alluvium cover (e.g., a mega-delta), or the study is over a very
small temporal scale (e.g., measuring river flow) the geology of
Investigation of the materials that lie below the surface of the the area also needs evaluation. For most studies a simple de-
land is one of the fundamental methodologies of geo- scription of the rock types and physical characteristics will
morphology. The accumulation of sediments through time in suffice although the inherent effects of geology on weathering
depositional landscapes acts as a preserver of the surface potential and weathering profiles can make such investi-
conditions that occurred when the materials ceased to be gations a focal part of a study (e.g., landslide studies). Selby
 Techniques and Methods for the Field: An Introduction and Commentary 107

(1993) provided a six-way classification of the ‘grade’ of rock, be minimized. In many institutions a risk evaluation is re-
which when used with a description of soil and weathering quired before any fieldwork is undertaken. In any sense, re-
profiles in the field, can provide a reasonable summary of quired or not, some attempt at assessing risk should be carried
subsurface physical characteristics of an area. Additional in- out as part of the definition of the project. One good oppor-
formation on properties (e.g., descriptions of bedding planes, tunity of assessing risk may come during the planning of
cleavage planes or fractures, if present) that affect rock strength sampling strategies, although in essence despite very good
can also be noted, and this plays a critical role when assessing planning, it is often necessary to modify the latter in the light
the strength of a material (see Chapter 14.30). Where rock of local conditions or situations where safe access is too
outcrops, or is exposed in section, strength can also easily be difficult.
tested in the field using a device such as the Schmidt hammer
(see Chapter 14.29).
Each field study and report will vary in structure, de- 14.9.6 Value of Fieldwork in Educational Aspects of
pending on specific requirements, but in general most field Geomorphology
reports should contain an introduction to the problem, a
summary of what is already known about the problem and the Geomorphologists should not let safety issues deter them
area to be investigated (the results of the desk study); a de- from a desire to conduct field studies, especially as with correct
scription of the methods, analyses that have been carried out preparations, fieldwork can be done with minimal risk. In a
(or are needed), notes on the implications for the problem statement that reflects the importance traditionally attached to
investigated and generally some basic conclusions (or working fieldwork within the discipline of geography, Sauer stated that
hypotheses) based on field observations. ‘‘the principal training of the geographer should come,
wherever possible, by doing field work’’ (Sauer, 1956, p. 296).
In a UK study of the importance of fieldwork in education,
14.9.4 Back in the Laboratory Job et al. (1999) provided five broad educational purposes
that geographical fieldwork serves; conceptual development,
Any good field campaign is of course useless unless it is backed skills, esthetics, values, social, and personal development. Al-
up with equally detailed laboratory analysis. All observations though there is considerable overlap between the categories,
need to be supported by precise, and unbiased, measurements this framework provides a good opportunity to discuss the
of the material that is collected. Such critical parameters include educational value of fieldwork in geomorphology.
the grain size and character of material, its chemistry as well the In the context of conceptual development fieldwork clearly
microfossils that are contained within the sediment. Laboratory provides the bridge between the classroom and the real world
methods are generally standardized in science so the data ob- (Fuller, 2006; Ham and Flood, 2009) that in turn provides a
tained from these techniques can be compared both within a unique opportunity for educators to query a students’ under-
given study but also with other studies. The need for standard- standing of the terminology and processes studies undertaken
ization has lead to the development of a range of methodo- in geomorphology (Job et al., 1999) and reinforce concepts
logical procedures for measuring each aspect of material with examples from the landscape. With reference to skill
collected in the field. Several chapters later in this volume are development, students and practitioners in the field are given
therefore entirely developed to the laboratory methods needed the opportunity to experience a type of research rarely ex-
to analyze materials (see Chapters 14.19, 14.20, 14.21, 14.22, perienced (Hall et al., 2002). Skills learnt in the field include
14.23, 14.24, 14.25, 14.26, 14.27, 14.28, 14.29, 14.30, 14.31, enquiry skills such as observational techniques, data col-
and 14.17). lection, data analysis, mapping techniques, and investigative
skills. Although it is obvious that fieldwork provides students
with the opportunity to practice and apply technical skills
14.9.5 Never Ignore Safety learnt in a ‘classroom or lecture style,’ it may also play a key
role in developing secondary skills in communication and
In this volume authors have provided detailed descriptions of numerical analysis (Nowicki, 1999).
methods for the field along with numerous case studies. It is If one considers esthetics and an appreciation of the
important to note that an underlying but generally neglected natural environment as a key education outcome of field-
theme of fieldwork and site investigations in modern geo- work then the value of fieldwork cannot be understated. It is
morphological studies is health and safety issues. In many clearly apparent that students in the field characteristically
cases the terrain or climate in which a geomorphologist plies develop an appreciation of, and sensitivity to, a variety of
their trade is dangerous and a long list of inherent risks exists different environments (Job et al., 1999). An appreciation of
in terms of external factors (e.g., climatic extremes, falling geomorphology can assist in the development of a respect for
rocks, unstable ground, flowing water). The impact of the environment and the processes that shape it. Working in
such risk factors increase if the area to be covered is unfamiliar the field also provides a natural platform for educators to
and/or remote. In addition to the preceding list there may facilitate experiential learning (Hall et al., 2002).
also be risks from the use of equipment (either those in use Fieldwork may also contribute to the values and develop-
for the investigation, or from other sources, e.g., those used ment of participants. In their study Job et al. (1999) also
in construction sites, quarries or road cuttings). It is ther- suggested that learning also entails an appreciation of values
efore fundamentally important that all field investigations and fieldwork enables participants to develop their under-
are properly planned with an underlying aim that all risks standing of different perspectives on social, political, or
108 Techniques and Methods for the Field: An Introduction and Commentary

ecological issues. In turn such developments can enable them Geomorphology remains a field-based endeavor and as such it
to clarify and justify their own values even while learning to is important that current practitioners pass on the field ex-
acknowledge and respect the values of others. perience to the younger generations. From the role fieldwork
Fieldwork has also been a demonstrated part of the plays in the education to the value of fieldwork in develop-
social and personal development of students as students ment, engineering and planning the role of fieldwork cannot
learn the importance of personal responsibility for their be underestimated.
learning (Hall et al., 2002), whereas students may build
confidence and increased resilience as they overcome physical
and mental tests that the many challenges of fieldwork gen- References
erally provide (Job et al., 1999). Outside of the pedagogical
realm, fieldwork will commonly help to break down barriers Fookes, P.G., Lee, E.M., Griffiths, J.S., 2003. Engineering Geomorphology. Theory
between researchers and provide an opportunity to develop and Practice. Whittles Publishing, Caithness, 288 pp.
Fuller, I.C., 2006. What is the value of fieldwork? Answers from New Zealand using
leadership, teamwork and communication skills, further
two contrasting undergraduate physical geography field trips. New Zealand
contributing to the personal and social development of field Geographer 62(3), 215–220.
party participants. Goudie, A. (Ed.), 1994. Geomorphological Techniques, second ed. Routledge,
London, UK, 570 pp.
Goudie, A.S., Viles, H.A., 2010. Landscapes and Geomorphology: A Very Short
Introduction. Oxford University Press, UK, 144 pp, ISBN-10: 0199565570.
14.9.7 Conclusions Hall, T., Healey, M., Harrison, M., 2002. Fieldwork and disabled students:
discourses of exclusion and inclusion. Transactions of the Institute of British
The process of choosing a study site, defining an objective or Geographers New Series 27(2), 213–231.
objectives, building a sampling strategy and conducting a field Ham, N.R., Flood, T.P., 2009. International field trips in undergraduate geology
curriculum. Philosophy and Perspectives Special Paper of the Geological Society
campaign may be an iterative process for many working in
of America 461, 99–104.
geomorphology but it remains at the core of our discipline; Job, D., Day, C., Smyth, T., 1999. Beyond the Bikesheds: Fresh Approaches to
especially where assessing landscape hazard. The only limi- Fieldwork in the School Locality. Geographical Association, Sheffield, UK, 80 pp.
tation on fieldwork comes when investigating the landforms Kennedy, D.M., Paulik, R., Millar, M., 2008. Infill of a structurally-controlled
of other planets and even then with the recent (and planned) estuary: an example from southern Whanganui Inlet, New Zealand. New Zealand
Geographer 64, 20–33.
robotic rovers traversing the surface of Mars (e.g., NASA’s Spirit Klinger, Y., Etchebes, M., Tapponnier, P., Narteau, C., 2011. Characteristic slip for
and Opportunity rovers) fieldwork is possible. The investi- five great earthquakes along the Fuyun fault in China. Nature Geoscience 4,
gation of data from all the terrestrial rocky planets in our Solar 389–392.
System as well as the large moons of Jupiter and Saturn may McCarthy, D.F., 2007. Essentials of Soil Mechanics and Foundations: basic
Geotechnics, seventh ed. Reston Publishing/Prentice Hall, USA, 864 pp.
be leading geomorphology into a new golden intellectual age.
Nowicki, M., 1999. Developing key skills through geography fieldwork. Teaching
Modern geomorphology does, however, currently exist in a Geography 24(3), pp.116–121.
world with restricted resources and increasing demands. When Sauer, C.O., 1956. The education of a geographer. Annals of the Association of
working in such a paradigm, a well prepared field campaign is American Geographers 46(3), 287–299.
an essential component of delivering the best results for your Selby, M.J., 1993. Hillslope Materials and Processes. Oxford University Press,
Oxford, UK, 466 pp.
investment (of time, money, and other resources). The desk- Slaymaker, O. (Ed.), 1991. Field Experiments and Measurement Programmes in
top study has become increasingly important and the wealth Geomorphology. Rotterdam, Holland, 236 pp.
of knowledge online combined with recent advances in Yang, S.-I., Ping, P.-X., Chen, S.-L., 2001. Changes in progradation rate of the tidal
mapping and technology mean many researchers are spending flats at the mouth of the Changjiang (Yangtze) River, China. Geomorphology 38,
167–180.
more time at their desk and less time in the field. Despite these
restrictions, we as a field must not lose sight of our roots.

Biographical Sketch

Adam Switzer is a sedimentologist and geomorphologist, who holds the positions of Singapore NRF Fellow,
Nanyang Assistant Professor and Principal Investigator at The Earth Observatory of Singapore and Division of
Earth Sciences in SPMS at Nanyang Technological University.
He and his research group are investigating the geological record of coastal hazards (storms and tsunamis) in
Southeast Asia. One of the driving aims of Dr Switzer’s research group is to make the coastlines of Asia safer places
to live, work, and holiday. Currently, Asia is the focus of his research program and Singapore has the resources to
enable him to effectively study the sedimentary and geomorphic effects of tsunamis and storm surges.
Dr Switzer’s 10 strong research group currently works in nine different countries throughout the world.
Dr Switzer obtained a BSc and PhD in Geosciences from the University of Wollongong, Australia, before
completing his postdoctoral studies at the University of Hong Kong.
 Techniques and Methods for the Field: An Introduction and Commentary 109

Dr David M. Kennedy is a geomorphologist who specializes in the response of coastal systems to climatic and
environmental change. He particularly focuses on the Holocene, researching estuarine, coral reef, and rocky coast
landforms. He completed his doctorial study at the University of Wollongong, Australia under the guidance of
Prof. Colin Woodroffe following undergraduate studies at the University of Sydney, Australia. After a postdoctoral
research position at Wollongong he was based at Victoria University of Wellington, New Zealand. He is currently a
senior lecturer in the Department of Resource Management and Geography at The University of Melbourne,
Australia.
 14.10 Topographic Field Surveying in Geomorphology
DM Kennedy, The University of Melbourne, Parkville, VIC, Australia
r 2013 Elsevier Inc. All rights reserved.

14.10.1 Introduction 110

14.10.2 Basic Survey Principles 111
14.10.2.1 Pre-Field Planning 111
14.10.2.1.1 What is the instrumental accuracy required? 111
14.10.2.1.2 What are the field conditions? 111
14.10.2.1.3 What is the availability/reliability of benchmarks/survey points? 111
14.10.2.2 In the Field 112
14.10.2.2.1 How accurate do surveys need to be? 112
14.10.2.2.2 Weather conditions and instrumental ruggedness 112
14.10.2.3 Post-Field Analysis 112
14.10.3 Common Types of Instruments 112
14.10.3.1 Auto/Engineer’s Level 112
14.10.3.2 Electronic Distance Meter 114
14.10.3.3 Total Station 114
14.10.3.4 Terrestrial Laser Scanner 114
14.10.4 Summary and Conclusions 117
References 117

Abstract

Measurement of landscape form is a fundamental technique for geomorphologists. A range of surveying equipment is
available, with this chapter focussing on the common surveying systems encountered by researchers and students, namely
engineer’s/auto level, electronic distance meter, total station, and terrestrial laser scanner. The choice of which type of
instrument is to be used in the field will depend on: (1) the accuracy required; (2) field conditions; (3) reliability of
benchmarks; (4) instrument ruggedness; and (5) facilities for post-field data analysis. This chapter is not a substitute for a
professional surveying degree, but aims to outline the basic principles of surveying for a geomorphological audience.

14.10.1 Introduction measurement of the land, with the Roman Frontinus in the
first century AD being a pioneer in the field. During the eight-
Measurement of the form of a landscape is a fundamental skill eenth and nineteenth centuries surveying advanced significantly
of the geomorphologist. Whether it is the shape of a contin- as nations sought to define their own borders, as well as those
ental shelf slope, a river channel cross section or the micro- of their dominions (Wolf and Ghilani, 2006). In modern times,
morphology of fractures in rock surfaces, quantifying form is the miniaturization of electronics means the tools available to
one of the most basic techniques of research methodologies. surveyors have increased exponentially in the past few decades.
Surveying is a wide-ranging discipline and one that cannot be Most of the standard instruments available in the mid-twentieth
summarized in a single book chapter. University degrees in century are now obsolete with it now being possible to survey at
surveying are offered globally, with the discipline underpin- up to sub-meter scale on other planets (e.g., Mars, Smith et al.
ning all human construction. Societies for professional sur- (2001)) and their orbiting moons in our solar system (e.g.,
veyors exist in most developed countries, and, generally, such Titan, Elachi et al. (2005); Soderblom et al. (2007)).
professionals have to be registered to practice in their field. This chapter does not intend to provide the reader with an
The oldest surveying records date to ancient Egypt. Here, alternative to a professional degree, nor can it provide a suc-
Herodotus recorded that Sesostris (c. 1400 BC) divided Egypt cinct coverage of all the types of equipment existing, due to
into plots for taxation purposes. As the annual Nile floods the wide variety of instruments available to the researcher.
would remove the property boundaries, surveyors were ap- It aims to provide a summary of the common instruments
pointed to reestablish them each year (Wolf and Ghilani, 2006). encountered in the majority of Geography/Geomorphology/
As human civilization grew, so did the need for accurate Earth Science departments and those which students are most
likely to encounter in their studies: namely an automatic level,
total station (TS), and terrestrial laser scanner. The reader is
Kennedy, D.M., 2013. Topographic field surveying in geomorphology.
In: Shroder, J. (Editor in Chief), Switzer, A.D., Kennedy, D.M. (Eds.),
referred to the many texts on surveying to provide a detailed
Treatise on Geomorphology. Academic Press, San Diego, CA, vol. 14, understanding of surveying principles as well as the major
Methods in Geomorphology, pp. 110–118. manufacturers of surveying equipment (e.g., Sokkia, Trimble,

110 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00377-8

 Topographic Field Surveying in Geomorphology 111

and Lecia) to gain information on the latest specifications and give a representation of the downstream changes that may
capabilities of surveying instruments. occur as a result of channel meandering, flood plain sedi-
mentology, inchannel bedrock outcropping, or pool and riffle
sequences (Knighton, 1998). It may also be the case that the
time taken to set-up and collect a high-resolution dataset is
14.10.2 Basic Survey Principles
too long for a planned field program.
The level of accuracy is also important if the morphological
For geomorphologists, surveying of the landscape primarily
data set that is being collected is to provide the base layer for
involves the measurement of cross sections or the construction
other analyses. For example, hydrological modeling programs
of digital elevation models (DEMs). Such activities aim to
have specific requirements of data density to produce a re-
precisely quantify the shape and form of the land for the
liable result (Blöschl and Sivapalan, 2006). If the topographic
purposes of measuring change or to provide the boundary
detail required for a model is not available, then the results
conditions on which earth-surface processes operate. Before a
produced by any subsequent analyses are inaccurate. In such
field campaign of surveying is undertaken, several consider-
instances, simple line surveys are not appropriate and more
ations need to be taken into account so that the data are
detailed DEMs obtained through laser scanning, Light De-
collected in a timely manner, the equipment is kept in a safe
tection and Ranging (LiDAR), or differential global pos-
condition, and the precision and accuracy obtained in the
itioning systems (GPS) are required.
field matches that which is required to solve the research
question.

14.10.2.1.2 What are the field conditions?

14.10.2.1 Pre-Field Planning
The area in which fieldwork is undertaken will also determine
14.10.2.1.1 What is the instrumental accuracy required? what type of survey gear is to be used. Instruments capable of
Instrumental accuracy is a fundamental property of surveying. obtaining detailed data sets are generally more sensitive to
Engineer’s levels can be accurate to within centimeters, environmental factors (e.g., precipitation, temperature, and
whereas terrestrial laser scanners will provide scale resolution humidity), and are more bulky, and therefore more difficult to
at the millimeter level. A common mistake is that researchers transport. For example, research work on coral reefs com-
aim for the highest possible accuracy without considering the monly requires transport in small open boats between mul-
research question that is being solved. For example, repeat tiple islands. Engineering levels and compact GPSs are
surveying of beach profiles may be possible at a millimeter- therefore more generally used, as they are rugged and easily
scale level; however, this detail is not necessary if the rate of moved when exiting boats in rough sea settings. For sites that
movement is on a scale many orders of magnitude higher. require arduous hikes to access, lighter gear is generally
A profile at millimeter-scale accuracy will provide information favored, with the same applying to remote field areas where
of sand-ripple dimensions in the surf. These will change form strict weight limits may apply to cargo transport. An additional
over a tidal cycle of hours (e.g., Masselink, 1998), but such consideration is how often the survey gear will be moved.
fine-scale variability is not required if the investigation is fo- More technical instruments require a longer setup time and
cused on beach- envelope change. Such is illustrated on the may not be as mobile as simpler equipment. Alternatively,
southeastern Australian beach of Moruya where profiling over differential GPS systems are designed to be extremely mobile
30 years has indicated a vertical envelope of change of 46 m and have a high degree of accuracy, but their accuracy depends
(McLean and Shen, 2006). Another example of the effect of on the availability of established benchmarks.
differing survey detail on geomorphic investigations can be
observed in fluvial channels of Northumberland, England.
Over a 1-km river reach, Fuller et al. (2003) compared sedi-
ment budgets calculated from 21 cross sections with those 14.10.2.1.3 What is the availability/reliability of
derived from a DEM of over 2600 points. They found the DEM benchmarks/survey points?
to be much more accurate in determining sediment budgets Accurate surveying generally requires the elevational data to be
for the river; however, this increased accuracy required a more linked to an established benchmark so as to relate the col-
intense field program. lected data to a nationally relevant and accurate datum. In
Instruments with a high accuracy also require a greater many remote areas, this is not possible; hence, researchers
amount of time to use. Terrestrial laser scanners or robotic TSs commonly use a local datum, such as tidal elevation or lake
may scan a land surface to millimeter scale but are limited in level. In the case of repeat surveys of landscape change, fixing
the extent of the view from the station they are set-up on. For the elevation to a national-level datum may not be needed as
surveying areas of high relief, multiple stations are therefore the relative change between points is the critical information
required; however, this level of detail may not be needed. For needed for the researcher. For surveys where elevational data
example, for understanding morphological changes on shore with respect to an established datum are required, bench-
platforms, simple line surveys are the most common approach marks need to be created at the research site. This can be done
used (such as in Britain, Canada, and Japan, Trenhaile (1999); by surveying from an established mark, or using differential
New Zealand, Stephenson (2001); or Sydney, Australia, GPS systems. In either case, the distance from the survey
Kennedy (2010)). Additionally, a detailed DEM of a single benchmark to the field site will determine how easily a reliable
section of river bank may be highly accurate, yet it does not benchmark can be established in the area of interest.
112 Topographic Field Surveying in Geomorphology

14.10.2.2 In the Field All TSs now have the capability to download survey data into
software packages that allow the user to quickly reduce their
14.10.2.2.1 How accurate do surveys need to be?
data. Terrestrial laser scanner and differential GPS systems,
As with all scientific analyses, the more data that are collected,
producing very large and complex datasets, require a large
the longer the experiment will take. The same principle applies
amount of post-processing. The availability of computer
to surveying. A single line survey, or DEM, will be more ac-
power and specialist software means that the more complex
curate where more points are taken; however, this will also
surveying techniques are avoided by those who simply want to
require more time in the field. The researcher therefore needs
construct a profile of a surface.
to balance the detail of the survey with the research question
that is being investigated. For example, when surveying a tidal
flat, how many points are appropriate to fully quantify its
elevation, width, and morphology? As shown in Figure 1, 14.10.3 Common Types of Instruments
tidal flats generally appear relatively flat and featureless, which
can lead the surveyor to minimize the number of survey 14.10.3.1 Auto/Engineer’s Level
points taken. The survey data in Figure 1 are of Kawhia Inlet
Leveling is defined as the measurement of the difference
on the west coast of the North Island of New Zealand, and are
in relative heights between points on the Earth’s surface
collected using a Sokkia 3030R Total Station. Two points at
(Knight, 1963; Davis et al., 1966; Foxall, 1970). This is one
the beginning and end of the profile give the total fall between
of the most common methods of cross-sectional surveying
the edge of the road and the edge of the tidal channel as well
as instruments for leveling are highly durable, portable, and
as the width of the flat, but little other information. At a detail
relatively inexpensive. Where only the height difference
of six data points, the convex profile of the tidal flat is shown;
between two points is needed, rather than the horizontal
however, with 15 data points, a shallow trough at the land-
distance that separates them, then taping or barometric
ward edge of the profile and a dip at its seaward edge become
leveling is possible (Wolf and Ghilani, 2006). Taping involves
apparent. Only with the full survey detail of 42 data points can
the simple measurement using a tape measure or graduated
the small bar at the profile’s seaward edge be fully quantified,
staff to measure the height (or depth) of an object, and elec-
as well as a beach on the landward edge of the profile below
tronic range finders are commonly used for this purpose.
the road. The advantage of the detailed profile is that the full
Barometric leveling uses instruments that convert differences
tidal flat morphology is quantified, although such a survey
in air pressure to estimate height. This latter method works
takes about twice as long as one that had only 10 data points.
on the principle that atmospheric pressure decreases with
For research activities that require more survey data, it is
increased elevation. As this method relies on measurements
therefore likely the instruments that have a greater capability
of air pressure, it will be affected by temperature, humidity,
for surveying quickly and storing data will be used. As a
and atmospheric waves, and, as a result, it is generally only
comparison, the survey undertaken in Figure 1 took ap-
accurate to within 2–5 m (SRBV, 1997). This level of accuracy
proximately 20 min to complete using a TS; however, if
is still useful in reconnaissance studies, such as mapping of
completed with an engineer’s level, it would likely take at least
terraces on remote Pacific Islands (e.g., Niue Island; Nunn and
an hour.
Britton, 2004; Box 1).
An auto level, also known as an engineer’s level, is the most
14.10.2.2.2 Weather conditions and instrumental common instrument used for leveling. Some are precise
ruggedness enough for second-order work, and first-order surveying is
An additional consideration when in the field is what weather possible if a parallel-plate micrometer is attached to the in-
conditions are likely to be encountered. The more compli- strument (Wolf and Ghilani, 2006). Auto levels are based on
cated the instrument, generally, the more sensitive it is to the dumpy levels. Dumpy levels were originally designed by
weather at the particular site. High amounts of precipitation William Gravatt, which, at the time, represented a revolution
will require the instrument to be waterproof, and cold weather in surveying instrumentation design from the spirit level
will limit battery efficiency. In extreme climates such as in (Davis et al., 1966). All instruments are composed of a tele-
Antarctica, specialist equipment is required as most standard scope, leveling screws, as well as clamps, which allow the
instruments will not operate for prolonged periods. Atmos- surveyor to sight along a horizontal plane (Figure 2(a)). The
pheric haze can also limit the optical capabilities of an in- advantage of auto levels is that once the main bulls’-eye level
strument, both for the reflection of light (in the case of laser bubble is set, an automatic compensator, commonly a wire-
electronic distance meters (EDMs)) as well as for the operator hung prism or mirror, takes over, which levels the line of sight
being able to physically site the reflector or staff holder at and keeps it level throughout the survey (Delano, 2003)
distance. In addition, in areas of high wind, heavier and larger (Figure 3). The compensator generally has a range of 10–20
tripods for mounting the survey instruments are required. arc-minutes; hence, unless the instrument is properly leveled
using the bulls’-eye bubble, it will simply ‘‘bottom out’’ against
the maximum swing limit (Delano, 2003). The advantage of
14.10.2.3 Post-Field Analysis
this compensator is that it ensures that the line of sight is
The final consideration when choosing a survey method is the always level, as, previously, the surveyor would generally have
amount of data analysis that is required after the field work to relevel the instrument especially to maintain the per-
has been completed. Essentially, the more simple the method pendicular orientation of the telescope with respect to the
(e.g., engineer’s level), the more simple the calculation of data. vertical axis of the survey line (Delano, 2003).
 Topographic Field Surveying in Geomorphology 113

Box 1 Engineer’s/auto level surveying

The auto level consists of a (1) level, (2) tripod, and (3) survey staff (stadia rod). The level is basically a telescope which is used to read heights from the staff.
The main principle of auto-level surveying is that you are measuring cross section of the landscape. The view through the level is the ‘line of sight’ that is surveyed.

Procedure
1. set up the tripod firmly on the ground and attach the level;
2. level the instrument using the three screws on its base;
3. choose the line that you wish to survey and point the level along that line; and
4. place the staff on a point of interest on the profile and read off its measurements through the level.

When looking through the level you should see three cross-hairs: an upper (U), middle (M), and lower (L).

For each point on the profile, all three readings must be taken. The middle reading is the height, the upper and lower give the distance
((U–L)  100 ¼ distance in meters). In the above example, the staff is 20 m away from the level.

Recording the Survey

As the profile is surveyed, a table similar to this is generally completed.

• Each line should contain only 1 set of readings, plus a description of the point in the remarks section.
• The first point on the profile is the back sight, the second an intermediate site, and the last the foresight.

For example
For the following survey profile:

Backsight
Level Foresight

Intermediate
B/S Intermediate
F/S
114 Topographic Field Surveying in Geomorphology

The survey form should look something like this:

When the furthest point along the survey is reached (i.e., the foresight), the level and tripod are moved further down the profile, a process called
‘changing station’. Once the level is set up, the profile is sited back to the previous foresight and a reading taken. This reading becomes the backsight for
the continuation of the profile. It is important to sight back to the foresight as the reading of this point will tie the second level plane back to the first level
plane.

Reducing the Data

To convert the field data to a profile, the distances from the level need to be converted to a cumulative distance and heights related to either the start point
or a known vertical point. A range of procedures are available for reducing the data, and the reader is referred to the many introductory texts of surveying
for specific instructions.

14.10.3.2 Electronic Distance Meter 14.10.3.3 Total Station

An EDM is an instrument that measures the distance between TSs are a surveying instrument that combines an EDM, elec-
two objects, based on the propagation of electromagnetic tronic-angle-measuring component, and a computer (Wolf
energy. The principle is that electromagnetic energy propagates and Ghilani, 2006). These are probably the most common
at a known speed (speed of light 299 792 458 m s–1 in a instruments used by researchers. They are also able to be used
vacuum), which is directly related to the modulated frequency in a range of environmental conditions from  20 to 50 1C
of the energy and its wavelength (Rüeger, 1996). An EDM does (Sokkia, 1996, 2006). An additional advantage over EDMs
not directly measure the travel time of the signal it emits but and engineer’s levels is that the data recorded in the field can
rather determines the number of wavelengths that are required be downloaded and analyzed at a later convenient time. TSs
to transmit the signal over a given distance. This precise dis- are highly accurate (millimeter scale), and therefore require
tance measured is actually the distance from the EDM to the care when setting up. The instruments also require delicate
target and back; hence, the measurement is divided by two. As handling, as improper use will cause damage. For example, it
the distance is the unknown variable, the machine counts the is recommended that TSs be transported using their carry case
number of full (and partial) wavelengths that are transmitted, or with the handle (Figure 2(b)) rather than attached to the
and converts this to a distance measurement. This is done tripod (Sokkia, 1996, 2006).
through transmitting a reference signal of precisely known The range of TSs available mean that specific instructions
wavelength superimposed onto the carrier signal emitted for the operation of the machines are best obtained from the
from the EDM (Kavanagh, 2007; Wolf and Ghilani, 2006) manufacturer. For all machines, some common elements exist.
(Figure 4). The carrier signal for most modern instruments is All are mounted on a tribrach, which allows the TS to be
visible or infrared light. As the distance between the instru- leveled using a bulls’-eye bubble, and, for precision surveying,
ments and the target will generally not result in the return additional level bubbles are present on the TS as well as in
of full wavelength, the fractional lengths returned are quan- digital form within the software. Optical sights and telescopes
tified through the phase shift in the signal. The original in- provide the surveyor with the ability to sight the target. Ad-
struments developed from this technique were designed in vanced models also allow the researcher to operate a TS re-
the late 1940s; however, they were expensive and bulky motely using Bluetooth or similar wireless link. Such robotic
(Wolf and Ghilani, 2006). Modern EDMs are highly portable, TSs allow surveying to be completed by a single operator.
and, for surveying, are most commonly incorporated with
digital theodolites and microprocessors to produce a TS.
14.10.3.4 Terrestrial Laser Scanner
Technology has also advanced so that now most instruments
can reflect directly off any surface, thus dispensing with the Advances in reflectorless laser technology have led to the de-
need for a purposely built reflector to return the signal to the velopment of automated optical systems that allow for the
machine. rapid collection of topographical data. These systems can be
 Topographic Field Surveying in Geomorphology 115

(a)

2 Data points

3 Data points

(b)

6 Data points 10 Data points

Distance (m)
0 200 400 600
0
Height below EDM (m)

15 Data points 42 Data points

1

(c)

Figure 1 (a) A single profile survey across a tidal flat in Kawhia Inlet, New Zealand; the set of footprints follows the line; (b) a Sokkia 3030R
total station used in the survey; and (c) a survey profile across the tidal flat based on 2, 3, 6, 10, 15, and 42 points; the level of detail will
depend on the number of points surveyed: as data density increases so does the amount of geomorphic information that can be extracted from
the field survey.
116 Topographic Field Surveying in Geomorphology





Reflector

Total
station
Figure 4 A schematic diagram of the phase principle of electronic
distance measurement. A reference signal is superimposed on the
carrier signal emitted from the instrument with the distance being
calculated based on the number of wavelengths (l) transmitted.
Figure 2 (a) An engineer’s/auto level and (b) a typical total station.
The model shown is a Sokkia SET3030R.
For all TLS instruments, the survey data are collected
through rotation of the laser in both horizontal and vertical
Level line planes. Large data sets are rapidly collected with 4250
of sight points s–1 (or up to 1 million points h–1) being collected in
Viewer
cliff scans in North Yorkshire, UK (Rosser et al., 2005) or up to
12 000 points s–1 for geological outcrop mapping in British
Wire hung compensator Columbia, Canada (Buckley et al., 2009). These data sets are
termed point clouds where each point is represented in space
Figure 3 A schematic diagram of a wire-hung compensator by X, Y, and Z coordinates measured in relation to the scan-
occurring within engineer’s/auto levels. The mirror provides fine ner’s position as well as a reading of the intensity (I) of the
adjustments to keep the survey line level.
beam reflections (Slob and Hack, 2004). Some scanner
models also record reflections in passive color (red, green, and
blue) reflection values (Slob and Hack, 2004) or the scanner is
mounted either on a plane, termed ‘LiDAR system’ or in a TS- used in conjunction with digital photography to reconstruct a
like package termed a ‘terrestrial laser scanner (TLS)’. This color image of the landform being investigated (Buckley et al.,
section focuses on TLS measures that are the technological 2009; Sturzenegger and Stead, 2009). Such an ability to rap-
advancement of TS systems; the reader is referred to volume 3 idly collect data has meant that TLS methodologies are used in
of this treatise for a more detailed analysis of LiDAR systems as a range of geomorphic environments, and also for a range of
well as such textbooks as Popescu (2012). other research purposes, such as measuring individual trees in
TLS systems can be categorized into three different systems forestry inventories (Pfeifer et al., 2004), scanning building
based on the principle of how distance is calculated, they facades in urban centers (Brenner et al., 2008), or as part of
being: (1) time of flight; (2) phase measurement; and (3) engineering assessments of roads (Burton, 2007).
optical triangulation (Fröhlich and Mettenleiter, 2004). Time- The collection of point clouds is influenced by the degree
of-flight systems measure distance through the time taken for of surface reflectivity as well as issues related to shading
the laser to reach its target and be reflected back to the receiver. and orientation of the surface being scanned. Moisture on a
Each point measured generally represents a burst of several rock surface will significantly reduce its reflective capability.
hundred laser pulses with the return pulse analyzed by a Standing water or even a veneer caused by precipitation
digital signal processor to produce a single distance measure generally interferes with the laser to such an extent that
(Bauer et al., 2003). Time-of-flight systems are the most scans cannot be made, which has also been noted to occur in
common used in geomorphology (e.g., Callidus, Leica, Mensi, the calibration of aerial (LiDAR) laser studies (Kaasalainen
Optech, and Riegel) as they allow the greatest distance et al., 2009). This is an important issue for studies on the
to be measured. For example, readings up to 2 km distant coast as intertidal areas generally do not dry out between
have been achieved in rock-slide scanning studies in Austria tidal cycles. Whereas different wavelength lasers available for
(Paar et al., 2000). Phase measurement (e.g., IQSun, Lecia, LiDAR can penetrate water surfaces, these are not currently
VisImage, and Zoller þ Frisöhlich) works on a principle simi- available for TLS systems. Another significant consideration
lar to how most EDMs measure distance. Although the range is for TLS use is the degree of shading that may occur in a
generally less than for time-of-flight systems, their accuracy is landscape. The laser can only measure those points that are
much higher, with millimeter resolution being possible in a direct line of site of the instrument. Cliff overhangs,
(Fröhlich and Mettenleiter, 2004). The third principle of boulders on platforms, or even channels will often be out of
measurement is used in the manufacturing industry and is sight of the TLS and therefore be shaded in the data set. This
based on optical triangulation. Micrometer accuracy is pos- shading effect means that, for nearly all studies, multiple scans
sible but this is only achieved over meter-scale distances of the area of interest must be undertaken so all surfaces can
(Fröhlich and Mettenleiter, 2004). be measured.
 Topographic Field Surveying in Geomorphology 117

The rapid collection of very large data sets mean that their dunes to surf-zone bars. Such a technological advance-
processing the point-cloud data is complex and compu- ment, when it occurs, will likely provide a revolution in
tationally demanding (Rosser et al., 2005). This is particularly geomorphological understanding.
the case for multiple scans where it is necessary to align each
point cloud to a common reference frame or co-ordinate
system. Artificial targets (e.g., reflective disks, globes, or cy-
References
linders) of a known location are typically manually placed
within the environment, surveyed using a TS, and sub-
Bauer, A., Paar, G., Kaufman, V., 2003. Terrestrial laser scanning for rock glacier
sequently used to align the point cloud. As a result, generally, monitoring. In: Phillips, M., Springman, S.M. (Eds.), Permafrost: Proceedings of
the process of establishing known points can exceed the time the 8th International Conference on Permafrost. International Permafrost
taken to scan the area by a factor of five (Brenner et al., 2008). Association, Zurich, pp. 55–60.
After the data are collected, the point cloud needs to be edited Blöschl, G., Sivapalan, M., 2006. Scale issues in hydrological modelling: a review.
Hydrological Processes 9, 251–290.
to remove erroneous points as well as trimmed to the area of Brenner, C., Dold, C., Ripperda, N., 2008. Coarse orientation of terrestrial laser
interest. As raw point-cloud data have no topological context, scans in urban environments. ISPRS Journal of Photogrammetry and Remote
they are generally converted to a mesh using a view-dependent Sensing 63, 4–18.
triangulation algorithm (see Lim et al. (2005) for a fuller Buckley, S.J., Schwarz, E., Terlaky, V., Howell, J.A., Arnott, R.W.C., 2009. Terrestrial
laser scanning combined with photogrammetry for digital outcrop mapping. In:
description).
Bretar, F., Pierrot-Deseiligny, M., Vosselamn, G. (Eds.), Laserscanning09. ISPRS,
Despite the complexity and expense of TLS equipment, its Paris, vol. XXXVIII, part 3/W8, pp. 1–6.
accuracy and speed of data collection means it has the po- Burton, G., 2007. Terrestrial laser scanner. Queensland Roads 4, 45–48.
tential to greatly advance geomorphology. For example, it is Davis, R.E., Foote, F.S., Kelly, J.W., 1966. Surveying: Theory and Practice, Fifth ed.
now being used to monitor cliff retreat rates at centimeter McGraw-Hill, New York, NY.
Delano, M., 2003. Automatic level compensators. Professional Surveyors Magazine
accuracy over monthly scales (Rosser et al., 2005) or for real- March, 2.
time monitoring of glacier movement (Bauer et al., 2003). Elachi, C., Wall, S., Janssen, M., et al., 2005. Titan Radar Mapper observations
from Cassini’s Ta and T3 fly-bys. Nature 441, 709–713.
Foxall, H.G., 1970. Handbook for Practising Land and Engineering Surveyors,
Second ed. The Institution of Surveyors, Sydney.
14.10.4 Summary and Conclusions Fröhlich, C., Mettenleiter, M., 2004. Terrestrial laser scanning – new perspectives in
3D surveying. In: Thies, M., Koch, B., Spiecker, H., Weinacker, H. (Eds.), Laser
A wide range of equipment is now available to the researcher Scanners for Forest and Landscape Assessment. International Archives of
for the surveying of natural landscapes. These range from basic Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. XXXVI-
optical methods, such as engineer’s levels, to more complex 9/W2, pp. 1–7.
Fuller, I.C., Large, A.R.G., Charlton, M.E., Heritage, G.L., Milan, D.J., 2003. Reach-
laser systems, such as terrestrial laser scanners. Simple optical scale sediment transfers: an evaluation of two morphological budgeting
systems are often more rugged and easy to use, but the rate of approaches. Earth Surface Processes and Landforms 28, 889–903.
data collection is low, requiring field campaigns to be longer. Kaasalainen, S., Hyyppä, H., Kukko, A., et al., 2009. Radiometric calibration of
The more complex terrestrial laser systems are, at present, LIDAR intensity with commercially available reference targets. IEEE Transactions
on Geoscience and Remote Sensing 47, 588–598.
generally too expensive for every department to purchase;
Kavanagh, B.F., 2007. Surveying with Construction Applications, Sixth ed. Pearson,
however, their potential for rapid collection of very large data Upper Saddle River, NJ.
sets makes them very appealing for researchers across a range Kennedy, D.M., 2010. Geological control on the morphology of estuarine
of disciplines. As with all large data sets, the processing of shore platforms: Middle Harbour, Sydney, Australia. Geomorphology 114,
terrestrial laser scanner data is very complex and requires ex- 71–77.
Knight, B.H., 1963. Surveying and Levelling for Students, Third ed. C.R. Books,
pert knowledge. London.
The type of equipment that researchers will use for sur- Knighton, D., 1998. Fluvial Forms and Processes. Arnold, London.
veying is dependent on a spectrum of factors ranging from the Lim, M., Petley, D.N., Rosser, N.J., Allison, R.J., Long, A.J., 2005. Digital
durability, precision and accuracy required, and availability of photogrammetry and time-of-flight laser scanning as an integrated approach to
monitoring cliff erosion. Photogrammetric Record 20, 109–129.
post-processing facilities to the physical conditions of the area
Masselink, G., 1998. The effect of sea breeze on beach morphology, surf zone
to be surveyed. As a result, although systems such as terrestrial hydrodynamics and sediment resuspension. Marine Geology 146, 115–135.
laser scanners may provide the most detailed data sets, their McLean, R., Shen, J.-S., 2006. From foreshore to foredune: foredune development
complexity means basic surveying techniques, such as pro- over the last 30 years at Moruya Beach, New South Wales, Australia. Journal of
filing, using an engineer’s level will remain commonplace. Coastal Research 22, 28–36.
Nunn, P.D., Britton, J.M.R., 2004. The long-term evolution of Niue Island. In: Terry,
A preference by researchers for simpler systems is also driven J.P., Murray, W.E. (Eds.), Geographic Perspectives on the Rock of Polynesia.
by equipment affordability. Insula, pp. 31–73.
As the affordability and durability of laser systems in- Paar, G., Nauschnegg, B., Ullrich, A., 2000. Laser scanning monitoring – technical
creases, their use will become more widespread. Technological concepts, possbilities and limits. Workshop on Advances Techniques for the
Assessment of Natural Hazards in Mountain Areas. Igls, Austria.
advances in laser reflection, such as through water, have to
Pfeifer, N., Gorte, B., Winterhalder, D., 2004. Automatic reconstruction of single
potential to revolutionize geomorphological research across trees from terrestrial laser scanner data. International Archives of
the terrestrial–aqueous boundary. Airborne laser systems are Photogrammetry and Remote Sensing XXXV, 114–119.
now able to penetrate the water column to shallow (o50 m) Popescu, S.C., 2012. LiDAR: Remote Sensing of Terrestrial Environments. Taylor
water depths. The application of this technology to land- and Francis, London (in press).
Rosser, N.J., Petley, D.N., Lim, M., Dunning, S.A., Allison, R.J., 2005. Terrestrial
based instruments will give the researcher the capacity to laser scanning for monitoring the process of hard rock coastal cliff erosion.
simultaneously survey river channels filled with water, lake Quarterly Journal of Engineering Geology and Hydrogeology 38, 363–375.
bottoms and their shorelines, as well as beach systems from Rüeger, J.M., 1996. Electronic Distance Measurement, Fourth ed. Springer, Berlin.
118 Topographic Field Surveying in Geomorphology

Slob, S., Hack, R., 2004. 3D terrestrial laser scanning as a new field measurement SRBV, 1997. Surveyors Registration Board of Victoria, Survey Practice Handbook.
and monitoring technique. In: Hack, R., Azzam, R., Charlier, R. (Eds.), LNES http://www.surveyorsboard.vic.gov.au/publications/surveypracticehandbook.htm
104. Springer, Berlin, pp. 179–189. (accessed November 2010).
Smith, D.E., Zuber, M.T., Frey, H.V., et al., 2001. Mars orbiter laser altimeter: Stephenson, W., 2001. Shore platform width – a fundamental problem. Zeitschrift
experiment summary after the first year of global mapping of Mars. Journal of für Geomorphologie 45, 511–527.
Geophysical Research 106, 23689–23722. Sturzenegger, M., Stead, D., 2009. Close-range terrestrial digital photogrammetry
Soderblom, L.A., Tomasko, M.G., Archinal, B.A., et al., 2007. Topography and and terrestrial laser scanning for discontinuity characterization on rock cuts.
geomorphology of the Huygens landing site on Titan. Planetary and Space Engineering Geology 106, 163–182.
Science 55, 2015–2024. Trenhaile, A.S., 1999. The width of shore platforms in Britain, Canada, and Japan.
Sokkia, 1996. Sokkia PowerSet series electronic total station: SDR software Journal of Coastal Research 15, 355–364.
reference manual. Sokkia Company, Olathe, KS. Wolf, P.R., Ghilani, C.D., 2006. Elementary Surveying: An Introduction to
Sokkia, 2006. Sokkia Series030R electronic total station: SDR software reference Geomatics. Eleventh ed. Pearson, Upper Saddle River, NJ.
manual. Sokkia Company, Olathe, KS.

Biographical Sketch

David is a geomorphologist who specializes in the response coastal systems to climatic and environmental
change. He particularly focuses on the Holocene, researching estuarine, coral reef, and rocky coast landforms. He
completed his doctoral study at the University of Wollongong, Australia under the guidance of Prof. Colin
Woodroffe following undergraduate studies at the University of Sydney, Australia. After a postdoctoral research
position at Wollongong, he was based at Victoria University of Wellington, New Zealand. He is currently a senior
lecturer in the Department of Resource Management and Geography at The University of Melbourne, Australia.
 14.11 Coring and Augering
CA Woodward, The University of Queensland, Brisbane, QLD, Australia
CR Sloss, Queensland University of Technology, Brisbane, QLD, Australia
r 2013 Elsevier Inc. All rights reserved.

14.11.1 Introduction 119

14.11.2 The Principles of Coring 120
14.11.3 Corer Types: Designs and Operation 122
14.11.3.1 Hand Operated Auger 122
14.11.3.2 Chamber Corer: ‘D-section’ or ‘Russian’ Corer 123
14.11.3.3 Gravity Corers 124
14.11.3.4 Extruding Gravity Cores 127
14.11.4 Corers for Taking Long Cores 128
14.11.4.1 Livingstone-Type 128
14.11.4.2 Percussion and Vibracorers 128
14.11.4.3 Percussion Corers 128
14.11.4.4 Vibracorers 129
14.11.4.5 Mackereth Corer 130
14.11.4.6 Permafrost Corers 130
14.11.4.7 Freeze Corer 131
14.11.4.8 The Geo-Slicer 132
14.11.5 Core Handling and Contamination Control 133
14.11.5.1 Core Handling and Identification 133
14.11.5.2 Contamination Control 135
14.11.6 Conclusion 135
References 135

Glossary plant and animal remains and fine nonorganic sediment

Diagenesis The physical or chemical transformation of deposited in standing water.
deposits such as sediments or soil by geological processes. Indurated To make hard or harden. A sedimentary rock
Heat and pressure are the two main geological forces that that is hard may be said to be indurated, for example, an
can alter sedimentary deposits altering their physical and indurated limestone.
chemical composition. Thixotropic Thixotrophy is a property of gels or fluids
Gyttja A term derived from Swedish meaning dredging, that are thick (or viscous), but become less viscous over
mud, slush, or silt. In geomorphology this term is used to time when shaken, agitated, or stressed.
refer to sediment composed of (partially) decomposed

Abstract

This chapter introduces the principles of coring, including the objectives for taking a good core and the major factors that
should be considered to ensure the collection of nondisturbed, representative core samples. The chapter also provides an
overview of the design, function, and operation of the main types of coring equipment that can be used to sample
sediments from a variety of settings including lakes, the ocean, peat bogs, soils, and permafrost. The major advantages and
disadvantages of each type of corer are also discussed.

14.11.1 Introduction
Woodward, C.A., Sloss, C.R., 2013. Coring and augering. In:
Sedimentary deposits are the natural archives of the earth,
Shroder, J. (Editor in Chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise
on Geomorphology. Academic Press, San Diego, CA, vol. 14, Methods in providing evidence for the processes that have shaped and
Geomorphology, pp. 119–137. continue to shape the landscape we see today. In some places

Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00378-X 119

120 Coring and Augering

these sediments are exposed in outcrops (e.g., a river cutting) tube is pushed or vibrated into the sediment (Figure 1(a)),
and one can easily study the structure of the deposit in the either by the force of gravity (e.g., gravity corers) or with
field and collect samples for later laboratory analysis (Chapter mechanical (manual) force. With most types of coring tech-
14.12). In other cases the sediments that one may be inter- niques the sediment enters the core tube at the same time and
ested in studying may still be buried beneath the earth’s sur- ideally the same speed as the core tube is pushed into the
face: in a peat bog, at the bottom of a lake, in coastal and sediment. Problems occur when the sediment enters the core
floodplain successions, or in the depths of the ocean. This tube faster or slower than the core tube is pushed into the
chapter provides an introduction to the various equipment ground (Figures 1(b) and 1(c)). The process of pushing a core
and techniques that are available to the geomorphologist who tube into sediment is analogous to an open-ended foundation
wishes to sample, or collect cores from, sedimentary deposits pile being driven into sediment, and consequently the physics
that are not exposed on the earth’s surface as outcrops. The of this process has been well studied by engineers (Randolph
focus of this chapter is on techniques that are available for et al., 1991; Paik and Lee, 1993; de Nicola and Randolph,
coring unconsolidated or poorly consolidated sediments that 1997). In engineering terminology, a pile is plugged when the
have undergone no, or minor diagenesis (e.g., heat and pres- sediment can’t enter the inside of the pile as fast as the pile is
sure). We refer the reader to publications by the Australian pushed into the ground. Skinner and McCave (2003) provide
Drilling Industry Training Committee (1996), Robello (2007), a detailed discussion of the factors that may lead to plugging
and Lyons (2009), on drilling equipment and techniques for (or under-sampling) in the context of coring. As the core tube
indurated sediments. Similarly, researchers who are interested is pushed into the sediment friction impedes the flow of
in ice-core drilling equipment and techniques are referred to material into the core tube and past the outside of the core
publications by Johnsen et al. (1994, 2007) and Bentley and tube (Figure 1(a)). Friction between the sediment and the
Koci (2007). core barrel is mainly a function of the shear strength of the
This chapter begins with an introduction to the principles sediment (Chapter 14.28) and the total contact area between
of coring, including the objectives for taking a good core and the sediment and the corer. The effect of contact area means
the major factors that should be considered to ensure the that friction increases with respect to the amount of sediment
collection of nondisturbed, representative core samples. The in the core tube. Therefore, we might have a representative
main types of coring techniques are introduced, including core at the top, but as the core tube becomes plugged (due to
corers that have a simple design and require basic operation increased friction) we begin to under sample (Skinner and
(augers and chamber type corers) and coring techniques with McCave, 2003).
more elaborate designs and that are more difficult to operate Sediment under-sampling is sometimes called ‘core short-
and maintain (e.g., the Mackereth type pneumatic corer). ening’ (Morton and White, 1997) which should not be con-
A summary of each corer type and the main advantages and fused with core compression. Core shortening usually occurs
disadvantages of each is listed in Table 1. What is clear is that during gravity or piston coring when plugging occurs. As the
there is no single piece of coring equipment that is suitable for flow of sediment into the core tube is impeded, the sediment
every depositional environment and the researcher will have bypasses the core tube (Figure 1(b)). True compression of
to make a careful and informed decision about which coring sediments during coring will only occur in dry porous sedi-
method is most appropriate for their specific geomorpho- ments, or in the case of wet sediments, when the pore water is
logical situation. displaced. This is important as true core compression would
lead to a core that contains the correct amount of sediment,
but has reduced water content (Blomqvist, 1991). Core
14.11.2 The Principles of Coring shortening due to sediment bypassing results in the loss of
some of the sediment from the core tube. An excellent dis-
In general there are three main objectives that should be cussion on the topic of core shortening and compression is
strived for when retrieving a core sample: provided by Cumming et al. (1993).
Over-sampling or ‘core thickening’ is a common problem
1. The core should be representative of the depositional en-
with piston corers (McCoy, 1985; Buckley et al., 1994; Lotter
vironment, landscape, or event that is being investigated
et al., 1997). Skinner and McCave (2003) state that over-
(i.e., as many sedimentary units or layers collected as
sampling occurs in piston corers when the negative pressure
possible).
generated by the piston, overcompensates for the friction
2. The core should faithfully preserve the sedimentary struc-
between the sediment and the core tube. Effectively the sedi-
tures (as much as possible) in their original context (i.e.,
ment is sucked up by the core tube. Ideally the piston stays
there is no or minimal mixing of different layers).
at the sediment surface during the sampling procedure
3. There should be no, or minimal distortion of the sedi-
(Figure 7(b)), but sometimes the piston might move upward
mentary structures (e.g., stretching or compression).
due to cable recoil when the cable is rapidly unloaded
Different types of coring methods (detailed below) will (McCoy, 1985; Buckley et al., 1994), or if too much tension
affect the ability to achieve these objectives. For now, it is is applied to the piston cable during the coring process
useful to simplify the coring process to its basic elements (Figure 1(c)).
to understand the factors that affect our ability to achieve Over- and under-sampling the sediment stratigraphy can
the desirable properties of a representative core sample. seriously affect sediment accumulation rate and chemical and
A core tube or barrel is a common element in most corer biological indicators, e.g., heavy metals (see Chapter 14.24)
types (see later figures). During the sampling process the core and fossil pollen (see Chapter 14.26), respectively. Both
 Coring and Augering 121

Table 1 Summary of coring techniques discussed in this chapter. There are many different types of corer available, each with its own particular
set of advantages and disadvantages. No single corer is suitable for all settings and applications

Corer type Setting Sediment type Depth of penetration Selected references

Hand operated auger Usually terrestrial, Different heads useful for Depends on sediment Stein (1991)
especially useful for a variety of sediment type.
soils. types. Typicallyo10 m
Russian or ‘D’ Corer Best in peat bogs or Best for peat, can also Depends on sediment Belokopytov and
terrestrialized lake penetrate gyttja and thin type. Beresnevich (1955);
margins sand and clay layers Typicallyo10 m Jowsey (1966)
Gravity corers
Flow actuated valve o50 m water depth Unconsolidated clay to o1.5 m Stager (1992)
fine sand
Kajak-Brinkhurst (‘KB’) o50 m water depth Unconsolidated clay to o1.5 m Kajak (1965); Brinkhurst
fine sand (1969)
Modifications to the KB At least 100 m water Unconsolidated clay to o1.5 m Glew (1991); Glew
type depth fine sand (1995); Cushing et al.
(1997)
Renberg and Hansson
(2008)
Multiple corer o50 m water depth Unconsolidated clay to o1.5 m Jahnke and Knight
fine sand (1997); Black et al.
(2002)
Deep ocean gravity corer Can operate in deep Unconsolidated clay to o5 m Hvorslev and Stetson
(41000 m) water fine sand (1946)
HAPS corer o200 m water depth Unconsolidated clay to o50 cm Kanneworff and
fine sand Nicolaisen (1973)
Corers for longer cores
Livingston-type o20 m water depth Unconsolidated clay to Depends on water depth Livingstone (1955);
(limited by rods) sand and sediment type. Wright (1991)
Typically o10 m
Percussion corer Terrestrial or in standing Can penetrate a variety of Depends on setting and Gilbert and Glew (1985);
water at least 100 m sediment types sediment, can easily Reasoner (1993);
water depth exceed 10 m Chambers and Cameron
(2001)
Vibracorers Terrestrial and standing Water saturated Depends on setting and Smith (1998)
water. Limited by unconsolidated fine- sediment, can easily
airhose. At least up grained sediments such exceed 10 m
to 150 m water as sand, silt, clay, peat
depth
Mackereth Standing water o250 m Unconsolidated clay to o6 m Mackereth (1958)
fine sand
Permafrost corers
Modified concrete drill-bit Permafrost Permafrost (frozen peat o7 m Calmels et al. (2005)
with some sand and
clay). Can cut
through pebbles
o20 cm Ø
Modified ice corer Permafrost Permafrost (frozen peat o10 m Hughes and Terasmae
with some sand and (1963); Blake (1964);
clay) Nørnberg et al. (2004)
Freeze corer o50 m water depth Unconsolidated clay to Gravity version o1 m Miskimmin et al. (1996);
fine sand percussion version Lotter et al. (1997);
o1.3 m Verschuren (2000)

over- and under-sampling can also result in distortion of to detect using traditional sounding (weighted measuring line)
sedimentary structures and mixing of sediment in the core or SONAR (Sound Navigation and Ranging) as it is sometimes
tube. A simple check for core stretching or shortening is to an indistinct transition composed of flocculent organics and
compare the depth of penetration to the length of core re- thixotropic sediments. Distortion and mixing of sediments may
covered. This can be difficult sometimes when sampling not be evident in homogenous sediments such as gyttja (organic-
in deep water as determining the depth of penetration rich lake sediment). Sediment mixing may be detected in these
is dependent on knowing an accurate depth of the sediment– circumstances based on isotopic dating methods (especially
210
water interface. The sediment–water interface can be problematic Pb and 137Cs) but this is an expensive method to determine
122 Coring and Augering

Core
barrel Piston

Fo Fl Fl Fo

Sediment

Forces acting on the Core shortening Core thickening

(a) core tube and sediment (b) (Under-sampling) (c) (Over-sampling)
Figure 1 (a) Forces acting on the core tube and sediment during the process of coring. The core tube is pushed into the sediment (downward
arrows) and internal and external friction (FI and FO respectively) can inhibit the flow of sediment past the outside and inside of the core tube.
(b) Friction between the sediment and the inside of the core tube can inhibit the flow of sediment into the core tube and result in plugging or
under-sampling. Friction between the sediment and the inside of the core barrel can increase with depth, so the top of the sample may be
representative and the bottom less so. As plugging occurs, sediment is displaced away from the leading edge of the core tube (white arrows)
rather than entering the core tube. (c) A piston is supposed to counteract the resistance to sediment flow provided by FI. If the piston is not held
stationary during coring (e.g., due to cable recoil) then we may over sample the sediment and distort the sedimentary layers.

that a nonrepresentative core has been collected. The loss of the rotating motion (usually clock wise) combined with down-
sediment–water interface (Figure 6(a)) can be a problem when ward pressure to drill the augering head into the sediment.
sampling sediment in standing water bodies and this will be During this process the drilling, or cutting action of the blades,
discussed in the section on gravity corers. and the downward pressure push the auger head into the
Another consequence of friction between the sediment and sediment with sediment being retained in the auger head (i.e.,
the inside of the core tube is the tendency for the sediment within tube or chamber type corers).
near the core tube surface to become smeared during coring A major advantage of this coring technique is that it is
and extrusion (Skinner and McCave, 2003). During coring, highly portable, versatile, and relatively low cost. It is simple to
shallower layers are smeared into deeper layers; during extru- use and cores as deep as 10 m can be realistically obtained by a
sion deeper layers are smeared into shallower layers. A core few operators depending on the sediment being cored. This
that has been smeared during extrusion is depicted in coring method is portable and can be easily transported to
Figure 2. Because smearing occurs, researchers should never remote areas by individuals or a small number of operators. In
sample the surface of any core for any type of laboratory an- addition, there are a variety of auger heads that can be used in
alysis. It is a good practice to sample from the centre of the a variety of sediment types ranging from sandy substrates to
core or remove at least a few millimeters from the outside of relatively compact soils, clays, and peat deposits.
the core before sampling using a scalpel or razor blade. Ac- There are five main types of auger systems: spiral, helical,
cordingly it is fundamental that a careful sampling strategy be Dutch (Edelman), Iwan (posthole), and bucket-type auger
developed that minimizes contamination and maximizes heads. The spiral augering head is basically a large drill bit,
sample collection. An excellent example of a sampling strategy usually 50–60 mm in diameter. The spiral augering head is
is presented in Chivas et al. (2001). usually used to penetrate hard compacted sediment layers such
as wet and compacted clays (to allow sampling to continue)
and is seldom used to study sediment stratigraphy or soil pro-
files. There are two main types of helical auger heads: short
14.11.3 Corer Types: Designs and Operation
flight and continuous flight. Both types comprise a helical blade
(or flighting) that encircles a central rod or shaft. Continuous
14.11.3.1 Hand Operated Auger
flight helical augers have a flight that (as the name implies) is
A hand operated auger (Figure 3(a)) consists of a handle and continuous to the surface of the earth. Sediment is carried from
augering head attached to a rod; with the capability of adding the bottom of the borehole to the surface as drilling progresses.
extension rods to increase the depth of penetration. The Short flight helical augering heads must be removed from the
handle and augering head can be disconnected from the rod bore-hole once the blades are full of sediment.
and additional rods can be attached in various ways de- The Dutch (or Edelman) and Iwan (or posthole) augering
pending on the system (e.g., thread, notch, or hex coupling). heads are both based on cutting blades that can be useful
The hand-auger is simple to operate; the operator uses a in fibrous and heavily rooted soil. The thin blades of the
 Coring and Augering 123

The major disadvantage of most hand and motor driven

augering systems is that they tend to produce disturbed samples,
that is, samples that are mixed, compressed and commonly

Core top
broken up into smaller pieces. These pieces can be arranged in
successive piles to give an overall impression of changes in
sediment stratigraphy or a soil profile (Figure 3(c)). Utilizing
split-tube samples in place of the auger head will overcome this
problem to a degree, but is generally restricted to a penetration
of less than a meter. This makes the augering system most useful
as a tool for site reconnaissance and general surveys of sedi-
ment/soil stratigraphy. If more detailed information on sedi-
mentary structures and precise depths of sediment boundaries is
required, then sampling systems that are based on a tube
sediment sampler (e.g., Livingstone, percussion, and vibra-
corers; discussed in Sections 14.11.4.1, 14.11.4.3, and 14.11.4.4
respectively) are required. Another issue is that auguring
below the water table may also cause problems such as the
sediment slumping out of the core tube and the auger hole
itself can collapse. Utilizing core casing to reinforce the
Streaks of clay

auger hole and collecting the sample inside a removable soft

clear plastic tube using a plastic sample retainer will enable
undisturbed samples to be collected in water saturated
materials.
Smearing

14.11.3.2 Chamber Corer: ‘D-section’ or ‘Russian’ Corer

An excellent tool for taking cores from unconsolidated sedi-
ment in terrestrial settings – especially peat bogs, is known as
the ‘D-section’ or ‘Russian’ corer (Figure 4). These names are
due to the ‘D’ (semicylindrical, Figures 4(a) and 4(b)) shape
of the sampling chamber in cross-section and the origin of the
concept for this type of corer in the former Soviet Union
Core base

(Belokopytov and Beresnevich, 1955). The semicylindrical

sample chamber varies in size from 4 to 10 cm in diameter
and 25 to 100 cm in length. This type of corer is also some-
times referred to as the ‘Jowsey,’ as Jowsey (1966) was the first
to publish detailed schematics of a corer based on the concept
described by Belokopytov and Beresnevich (1955). The system
Figure 2 An extruded Livingstone core from the west of Ireland will be referred to as a Russian corer for the rest of this chapter.
(Watson et al., 2010). The sediment was deposited at the end of the The Russian corer is physically pushed into the sediment
last ice age and the grey sediment represents an abrupt cooling by pushing on a handle attached to the extension rods (usu-
event known as the Younger Dryas period. As the core has been
ally 50 cm in length) that attach to the coring unit. The corer is
extruded (top to bottom) friction between the core tube and the
pushed into the ground in the closed position (Figure 4(b1)),
outside of the sediment has resulted in smearing or spreading some
of each adjacent sediment layer into the next. with the hemispherical coring chamber flushed with the an-
chor plate. Once the corer is pushed to the desired depth, the
coring unit is pivoted around the anchor plate (now anchored
clay-type Dutch auger also make it much easier to remove the in the ground) in a clock-wise rotation of the handle attached
sample (due to reduced friction). The bucket-type augering to the extension rods. The leading edge of the sample chamber
head (Figures 3(a) and 3(b)) comprises a sampling chamber then slices a hemispherical sediment sample adjacent to
with cutting blades at the base. The blades cut into the base of the borehole. The corer is then pulled out and opened
the bore-hole and the cut sediment then passes into the (Figure 4(c)) and the retrieved sediment is transferred to half-
sampling chamber as the auger unit is pushed into the ground. piping (usually plastic or PVC piping), wrapped in plastic
This method is particularly useful in sandy sediments. Al- cling film and labeled. A detailed guide to coring and sub-
though some mixing still occurs with bucket type auger heads sampling peatlands for paleoenvironmental research is pro-
(samples are sometimes referred to as ‘semidisturbed’) we can vided by De Vleeschouwer et al. (2010).
generally get a better impression of the stratigraphy as the Similar to the hand-auger technique, the Russian corer is
sample is retained in the sample chamber (Figure 3(b)). Some also a relatively inexpensive method of retrieving subsurface
bucket-auger heads are also hinged and allow the sample to be data. Another significant advantage of the Russian corer is that
taken out without breaking it up. it is portable and easy to use. The unit and necessary
124 Coring and Augering

(b) (c)

Handle Surface
sample

Rod

Augering
head

(a)

Figure 3 (a) Example of a hand driven augering system with one extension rod and a bucket augering head. (b) A bucket augering head with
a sample of soil in the sampling chamber (soil surface is at the top). The sequence of soil is coherent, but will be mixed or slightly disturbed.
(c) Augering often results in disaggregated samples as the blades (and removal from the sample chamber) break the sediment up. Nevertheless,
an impression of the overall stratigraphy can be made by making successive piles of sediment. This is a Ferrosol sample from southeast
Queensland, Australia.

equipment can be easily carried into remote sites by an indi- straightforward in environments with sediment units that
vidual or small team (ideally two or three people). This change quickly laterally. In addition, this type of corer is not
number allows for easy sediment collection and preservation suitable for sampling the sediment–water interface or un-
(wrapping of the core sections) and for the recovered core to consolidated wet sediment, as aqueous sediment can leak out
be carried out from the field site. This coring method can also of the core chamber.
be used to collect long core sections by attaching extension
rods to the coring unit and collecting contiguous samples at
successively greater depths. The contiguous samples can be
14.11.3.3 Gravity Corers
taken from the same borehole, but it is preferable to take the
samples from a close (B40 cm apart) pair of boreholes. This is A Gravity corer presents the simplest way of sampling sedi-
because the nose of the corer (Figure 4(a)) will deform sedi- ment from the bottom of the standing bodies of water and is
ments immediately beneath it where the upper part of the next commonly used to sample sediments from lakes, estuaries,
sample would be taken. It is a usual practice to allow some and ocean basins. The basic common elements of any gravity
overlap (at least 10 cm) for the contiguous samples from the corer are a core tube and mechanism for sealing the top of the
boreholes. core tube to ensure the sample is retained in the tube
The main disadvantage of the Russian corer is that (de- (Figure 5(a)). The sealing mechanism works like a finger
pending on how compact the sediments are) it can be difficult placed over a drinking straw in a glass of water. The water stays
to penetrate sandy or clay rich sediments, especially when in the drinking straw when it is lifted out of the water due to a
using a larger diameter core chamber. The other disadvantage reduction in pressure at the top of the straw. The same prin-
stems from the fact that paired boreholes are necessary to ciple (along with sediment/core tube friction) helps to keep
acquire undeformed sediment sequences. Generally it is a the sediment in the gravity core tube. A rubber stopper is
simple matter to correlate between the adjacent core samples usually placed in the bottom of the core tube for small gravity
using a palaeo-environment proxy such as loss-on-ignition corers used in lakes when the corer is pulled to the water
(Chapter 14.22) or magnetic susceptibility. This may not be surface.
 Coring and Augering 125

Handle

Handle
Pivot

Rod

Pivot

Chamber

A B
Anchor
Anchor plate
plate

Pivot
Nose

Pivot Anchor
Section
plate 5 cm
AB Chamber
(a)

1
Section AB
showing operation
2 of the corer Pivot
2 cm
Nose
(b) (c)

Figure 4 (a) Schematic diagram of a ‘Russian’ or ‘D Section’ corer. (b) Cross section through the sampling chamber showing how the corer
takes a sample (air is white, sediment is grey). The corer is pushed to the correct depth in position (1) and then rotated to take a sample (2).
(c) A photograph of an early Holocene sediment sample taken using a ‘Russian’ corer from a terrestrialized lake margin on the South Island, New
Zealand.

There are two types of sealing mechanisms; flow-actuated slowly lowered into the sediment, and then the seal can be
(e.g., Stager, 1992), and messenger-operated ‘KB’ type corers activated using a messenger (a weight) that triggers the closing
(Kajak, 1965; Brinkhurst, 1969; Figure 5). The flow actuated mechanism (Figure 5(b)). The gravity coring system has
seal is a one-way valve that opens due to the pressure exerted continued to be modified through time to meet the challenges
by the water flowing through the core tube as the core tube of taking representative cores in standing water. Many of the
descends toward the sediment surface. It is generally necessary modifications focus on controlling the speed of impact of the
for the corer to descend at a considerable velocity so that there gravity corer (e.g., a parachute assisted gravity core; Cushing
is enough pressure to open the flow actuated seal. If the seal et al., 1997), positioning the coring unit, and activating the
does not open during descent, the resistance to through-flow seal at the right place and time when the gravity corer is being
of the water through the core tube can create a bow-wave used in deep water. It can be a simple matter of feeling (or
below the core-tube that can ‘blast’ away the sediment–water even seeing) when the corer has sunk into the sediment in
interface (Stephenson et al., 1996). Even if the seal does open, shallow lakes. In deep water environments (particularly with a
a rapidly descending corer can cause disturbance and dis- diffuse sediment–water interface) we cannot feel when the
tortion in the sediments, and it is difficult to control the depth coring unit hits the sediment–water interface. Glew (1995)
of penetration. outlines modifications that can be made to the basic KB corer
The KB-type corer presents one solution to the limitations for successful operation in deep water. Deep ocean gravity
of the flow actuated seal system. The gravity corer can be corers are triggered using a pilot weight release mechanism
126 Coring and Augering

Rope

1 2 3 4
Messenger

Triggering
mechanism

Water
Spring loaded
seal

Tube clamp
Weight

Core tube
Sediment

(a) (b)

Figure 5 (a) Schematic diagram of the construction of a ‘KB’ type messenger triggered gravity corer. The core tube is usually about 5 cm in
diameter. (b) A diagram depicting the coring sequence of a ‘KB’ corer in standing water. 1 and 2: The corer is lowered into the sediment, with
the seal in the open position (armed). 3: Once the corer is in position the messenger is dropped to trigger the sealing of the core tube to hold
the sediment in place. 4: The corer is then lifted to the water surface. A rubber stopper is placed in the bottom of the core tube, the seal is lifted
and the tube clamp loosened. The core tube (and sample) is then disconnected from the rest of the coring unit.

that hits the sediment before the core tube (Hvorslev and There are generally visual checks that can be performed in the
Stetson, 1946). Other modifications have focused on de- field to ensure that a representative and nondisturbed gravity
veloping a smaller lightweight version of the KB (Glew, 1991), core has been taken. For example, the preservation of algal
using new materials to construct the corer and improving mats (Figure 6(a)) or chironomid (nonbiting midge) larval
the design of the core extruder (see Section 14.11.3.4) (Renberg tubes can be used as an indicator that we have captured the
and Hansson, 2008). sediment–water interface. The preservation of fine scale sedi-
This method of coring is particularly useful where rela- mentary structures (Figure 6(a)) (e.g., varves or laminations)
tively undisturbed short cores are required for environmental is also useful indicator that the core has not mixed or dis-
impact studies (e.g., Smol et al., 1998) establishing modern torted. However, researchers need to be aware that a mixed
calibration sets for fossil species (e.g., Woodward and Shul- sediment–water interface is also a natural feature in many
meister, 2006) and establishing relatively recent sedimen- shallow lakes, where there has been mixing due to the action
tation rates. This is because this coring method retains the of the wind or bioturbation by invertebrates (e.g., chironomid
sediment–water interface (Figure 6(a)). Accordingly, it is larvae) (Larsen and MacDonald, 1993).
possible to sample modern sediments that are currently being One common requirement in palaeoenvironmental studies
deposited in the sediment sinks and receiving basins today. is having enough sediment to perform the multiple analyses
 Coring and Augering 127

Core tube

Support
disk Core tube
support

Extruder
interval
spacer

Extruder
key

(a)

Extruder
Extruding tray rod

Base

(c) (b)

Figure 6 (a) A close-up of the sediment–water interface of an undisturbed gravity core from a lake in southern Ontario, Canada. Note the
presence of a thin green algal mat on the sediment surface. This would be blasted away or mixed with the underlying mud by a bow-wave in
front of the core tube if we dropped the corer too quickly and if the seal was not opened properly. The presence of fine-scale sedimentary
structures (such as the light band) is also a good indication of an unmixed or undisturbed gravity core. (b) A specialized extruding device for
extruding gravity cores in the field. (c) Sediment intervals are extruded into the extruding tray and scraped into sterile, labeled sample bags using
a clean spatula.

that are now common in multidisciplinary, multiproxy stud- 14.11.3.4 Extruding Gravity Cores
ies. One possible solution is to build a large diameter core
tube. The dimensions of the core tube are limited by the When extruding (recovering) the gravity core it is best to dis-
ability of the seal mechanism and sediment friction to prevent turb the core as little as possible and to keep it in a vertical
the weight of a large diameter sediment column from falling position in the boat until returning to the shore or the des-
out of the tube. The HAPS corer (Kanneworff and Nicolaisen, ignated sampling station. Specially designed gravity core
1973) uses a metal plate to close off the bottom of a large extruders are available (Figure 6(b)) that can be used to ex-
diameter tube to prevent the sample from falling out. Another trude and subsample a gravity core in the field. The core (with
solution to the sediment supply problem is to take multiple rubber stopper still in place) is placed in the core tube support
cores from the same place at the same time. Jahnke and Knight on top of the support disk.
(1997) and Black et al. (2002) have published designs for The extruder interval spacer controls the distance the core
gravity corers intended for taking multiple sediment cores at tube support drops for each sampling interval, and this can be
the same time. set prior to extruding and subsampling a core (usually 0.5 or
128 Coring and Augering

1 cm intervals). It is important to record the length of all cores When operating in a standing body of water, a coring
in the field and describe the major sediment transitions prior platform is required to core from. There are many different
to extruding the gravity core. It is possible for the extruder types of coring platform used, but a ‘cataraft’ (a platform
interval spacer (Figure 6(b)) to slip slightly as extrusion of a with inflatable pontoons) is a lightweight, portable, and
gravity core proceeds (it should be checked periodically) and stable option. It is also important that a casing is used to keep
the depth of our last bagged sample interval may fall short of the extension rods vertical (especially in water depths 45 m).
our originally measured core length. It is generally a good idea The casing generally comprises threaded PVC piping that
to note down when major sediment transitions are flush (or screw together to form a continuous coring shaft from the
close to flush) with the top of the extruding tray for this water surface to the sediment surface.
reason. The coring procedure for the Livingstone (Figure 7(b)) is
Before subsampling, water is carefully siphoned out and similar to that described for the Russian corer; as extension
then the extruder tray (Figure 6(c)) is mounted to the top of rods continue to be added to the coring unit to take pro-
the core tube. The core tube support and core tube is then gressively deeper samples. The corer is pushed into the sedi-
lowered until the top of the sediment column is flush with the ment until the core tube is full, and then brought back to the
surface of the extruding tray. The extruder interval spacer is surface. Each sample collected in this way is sometimes re-
then used to drop the core by the preset interval and the ferred to as a ‘push.’ It is important when collecting Living-
corresponding amount of sediment is extruded into the ex- stone cores to measure the water depth and the depth that the
truding tray. This sediment is then scraped into a labeled core was pushed into the sediment. These measurements are
sterile sample bag. Some researchers use a small piece of used to check for core extension or compression. Once the
plastic piping on top of the extruding tray to stop the aqueous core is brought back to the surface, the core tube can be dis-
sediment from the sediment–water interface from being lost connected from the rods and the collected sediments extruded
out the end of the extruding tray. The issue of core handling from the core tube into PVC piping. Alternatively, if there is no
and contamination control is of utmost importance when need to reuse the core tube, the sediment can be retained in
extruding and subsampling in the field. We refer readers to the the core tube until subsampling in the laboratory. A Living-
section dealing with these subjects toward the end of this stone type system where different core tubes are used to keep
chapter. the sediment stored in the core tube until analysis is some-
One other point that may be considered when sub- times referred to as a ‘Bolivia’ system. It is preferable to use a
sampling a gravity core is the presence of live invertebrates in transparent core tube (e.g., constructed from polycarbonate) if
the surface sediments. The invertebrates may be negatively we wish to retain the sediment in the core tube so we can see
phototaxic (move away from a light source) and will con- that a good core has been collected. Wright (1991) provides a
sequently burrow deeper away from the top of the core. This detailed guide to taking good cores using a Livingstone
potential problem was recognized by Wilhelm and Lasenby type corer.
(1997) who designed a specialized core tube that can prevent
the post sampling redistribution of invertebrates.
14.11.4.2 Percussion and Vibracorers
The depth of penetration of hand-driven coring systems such
as the Livingstone corer can be limited in sediments that
14.11.4 Corers for Taking Long Cores
are difficult to penetrate such as clays and sands. In these
situations a percussion corer or vibracorer may be a better
14.11.4.1 Livingstone-Type
option.
A Livingstone corer (Livingstone, 1955; Figure 7) provides the
opportunity to take longer cores than a gravity corer in
standing water, and can usually penetrate deeper into bog
14.11.4.3 Percussion Corers
sediments, clay layers or sandy substrates than a Russian corer.
The Livingstone type corer has a smaller cross-sectional area A percussion corer uses the hammering action of a weight to
than the Russian corer and unlike the gravity corer the Liv- drive or hammer the core tube into the sediment. The per-
ingstone corer is physically pushed into the sediment using a cussion system designed by Gilbert and Glew (1985) provides
handle and extension rods. However, in contrast to other push an example of a simple percussion or hammer mechanism.
core methods the Livingston corer has the addition of a piston The weight is a 9 kg piece of metal cut from a 10.2 cm diam-
to prevent plugging of the core tube (Figure 7). Greater eter steel bar, center bored to slide onto a central rod, or ‘jar
penetration and sediment recovery is achieved, as the piston staff.’ The jar staff is attached to a metal boss or cap assembly
assists the sediment to flow into the core tube by creating a that fits onto the top of the core tube. The weight is lifted up
negative pressure at the top of the core tube, analogous to and down on the jar staff and strikes the boss cap to hammer
sucking up fluid using a syringe. This is particularly important the core tube into the ground. When coring on land, the
for longer and deeper cores where friction between the sedi- system requires the use of handles on the hammer mech-
ment and the core tube can provide resistance to the sediment anism. When coring in standing bodies of water, a rope
flowing into the core tube. Problems (and coring artifacts) can or cable is required to lift and drop the weight. The latter
occur if the piston over-compensates for the friction (see dis- system is sometimes called a rod-less or remotely operated
cussion in Section 14.11.2 above). percussion corer.
 Coring and Augering 129

Handle 1 2 3 4

Water
Rod

Core tube
(sectioned)
Piston
remains
stationary

Cable
Sediment

Piston with
gasket rings

Nose-piece
(a) (b)

Figure 7 (a) Schematic showing the main features of a Livingstone type piston corer. (b) Diagram depicting the sampling sequence using a
Livingstone corer to sample in shallow water (without casing). 1: The piston is at the bottom of the core tube as the corer is lowered on
extension rods to the sediment surface. 2: Constant tension is kept on the piston cable to keep the piston stationary as the core tube is pushed
into the sediment. Once the push is completed. 3: The core tube is brought back to the surface by disconnecting the extension rods. 4: The
sediment is then extruded into a PVC half-pipe and the core tube reused or the core-tube disconnected and another tube attached (as per the
‘Bolivia’ system).

There have been many variations in the design of remotely 14.11.4.4 Vibracorers
operated or rod-less percussion corers since the early versions
produced by Huttenen and Meriläinen (1975) and Digerfeldt Vibracoring is a technique for collecting unconsolidated fine-
(1978). Earlier versions required two ropes or cables, one lifting grained sediments such as sand, silt, clay, peat, and gyttja (lake
or recovery line and one line to lift and drop the driving weight. sediment) from a variety terrestrial settings (e.g., coastal and
Because of the tendency for these lines to tangle, Gilbert and flood plains) to subaqueous and saturated sediments (e.g.,
Glew (1985) simplified the system to require only one line. The estuarine, wetlands, and shallow marine settings). With the
line attaches to the driving weight. To pull the corer out of the use of submersible electric and hydraulic systems this techni-
mud, the driving weight is simply pulled to the top of the jar que is also applicable to water depths of approximately 150 m.
staff where there is a retaining bolt stopping the weight sliding Smith (1998) provides an excellent review of the development
off the shaft. The system designed by Gilbert and Glew (1985) of the modern vibracoring system.
and a later version designed by Reasoner (1993) does not have Early devices (Grumenskii and Kamarov, 1961; Sanders
a piston, but relies on a device called a core catcher (Figure 8) and Imbrie, 1963; Pierce and Howard, 1969) relied on a
to prevent sediment falling out the core tube during lifting. vertical ‘pile-driving’ effect from a modified jack hammer.
Other examples of the percussion type corer do have a piston Lanesky et al. (1979) mounted a concrete vibrator onto alu-
(Chambers and Cameron, 2001) or a piston and a core catcher minum irrigation pipe which resulted in a true vibration
(Nesje, 1992). The disadvantage of a piston is that it can be mechanical effect rather than the ‘pile driving’ effect. Another
difficult to control the position of the piston during coring major innovation was the development of rigid core tube
when using a percussion corer (Glew et al., 2001). couplers (Smith, 1992) that now allow more flexibility for
130 Coring and Augering

core tube. The tube and sediment within acts like a solid rod
being driven into the sediment. No more sediment is collected
unless the tube penetrates to a harder layer and the friction
inside is exceeded once more. This ‘plugging’ effect can also
result in additional compaction. That is why it is very im-
portant to record the measured depth of the core in com-
parisons of retrieved thickness of the core section to make an
accurate assessment of plugging and compaction (e.g., Sloss
et al., 2005, 2006, 2010).

14.11.4.5 Mackereth Corer

A Mackereth corer (Figure 9) is a pneumatically operated
corer for taking long (up to 6 m long) cores independent of
water depth up to a limit of approximately 250 m (Mackereth,
1958). A Mackereth corer uses a system of hoses, chambers,
and compressed air to secure the ‘anchor chamber’ to the
sediment surface, drive the core tube into the sediment, and
1 cm lift the entire apparatus to the water surface using the
buoyancy of the air-filled anchor chamber (Figure 9(b)).
Figure 8 Photograph of a plastic core catcher (5 cm diameter). The There are three main phases of operation of the Mackereth
core catcher would be mounted at the bottom of the core tube, on corer. When the device settles into the mud, water is pumped
top of the nose-piece (Figure 7(a)) in this orientation. The ‘fingers’ out of the anchor chamber using hose number 1
can spread to allow sediment to flow into the core tube (upward and (Figure 9(b1)) creating a negative pressure that sucks the
outward). Downward force of the sediment column will push the
anchor chamber into the sediment. Water is pumped out of
‘fingers’ down and close the core catcher inhibiting the outward flow
the anchor chamber until traces of mud are evident in the
of sediment.
pumped water. Compressed air is then pumped into the
chamber above the driving piston (Figures 9(b2) and 9(b3))
through hose number 2, pushing the core tube into the
taking long cores from deep lakes and shallow marine en-
sediment. Once the driving piston is lower than hose number
vironments. Another recent peculiar modification to the
3 (Figures 9(a)) and 9(b3), air spills through hose number 3
vibracoring system by Aqua Survey Inc. in the USA incorpor-
into the anchor chamber. The air-filled anchor chamber be-
ates a nose-cone with an electromagnetic detector. This allows
comes buoyant and the coring unit will rise to the water sur-
coring in areas with submerged or buried utilities, unexploded
face. The corer can rise extremely rapidly to the water surface,
ordnance (UXO), cultural resources, or other obstacles to be
and an extension arm connected to a weight and chain acts as
avoided.
a release valve to control the ascent of the coring apparatus
The main difference between a percussion corer and a
(Figure 9(b4)). The main disadvantages of the Mackereth
vibracorer (sometimes referred to as a vibrocorer) is that (as
corer are that it is not very portable, and that the rapidly rising
the name implies) the coring tube or pipe is vibrated rather
core apparatus can be hazardous. This is not a problem if we
than hammered into the ground. High frequency, low-amp-
have hoses that are sufficiently long enough to allow members
litude (0.1–1.0 mm) waves propagate along the pipe from the
of the coring team to get a safe distance from the breaching
power source and transfer energy to the sediment, liquefying a
core unit.
1–2 mm thick layer of sediment inside and outside the coring
tube. This high frequency vibration and liquefaction greatly
reduces wall friction both inside and outside of the tube and
14.11.4.6 Permafrost Corers
result in longer and more representative cores. In addition the
sediments recovered are also relatively undisturbed (i.e., Coring permafrost presents a set of additional challenges to
maintain stratigraphic integrity and may preserve sedimentary successful core retrieval, including the harsh environmental
structures). conditions and the requirement to avoid partial melting of the
The main disadvantage of the vibracoring system is the size permafrost during coring. The composition of permafrost can
and weight of the components. The basic vibracoring system be highly variable, including solid ice inclusions, sand, and
requires a power source (petrol or electric generator), tripod, other mineral particles. There are two main types of coring
winch, and piping. Therefore the vibracoring system is best apparatus available for coring permafrost: a modified concrete
used for more accessible sites, unless helicopter or a boat for drill-bit (Figure 10(a)), and a modified ice corer with teeth
subaqueous environments is available (and is within the (Figure 10(b)). Both of these coring units are powered by a
budget of the project!). Another disadvantage is the ‘plugging’ light engine available commercially that is designed to power
or ‘rodding’ effect. This is where the increasing wall friction an earth auger.
inside the tube exceeds the cohesion of the sediment. When A modified concrete drill-bit permafrost corer is described by
plugging occurs, the core tube will continue to penetrate the Calmels et al. (2005). A commercially available, off-the-shelf,
sediments; however sediment is no longer drawn inside the concrete drill-bit is regularly used in industry to pierce holes
 Coring and Augering 131

1 2 3 4

Hose 2 Hose 1

Air

Air

Air
Driving

Water
piston

Water
External
tube

Piston
tube

Core
tube

Extension
arm

Sediment
Hose 3
Piston
Tube

Anchor
Weight
chamber

(a) (b)
Figure 9 (a) Schematic showing the main features of a pneumatically operated Mackereth corer (Mackereth 1958). (b) The sequence of
operation for sediment sampling using this corer. 1: The corer is lower so the anchor chamber settles into the sediment. Water is then pumped
out from Hose 1, creating suction in the anchor chamber. 2: Once the anchor chamber is in place, air is pumped into Hose 1, which drives the
driving piston down and the core tube into the sediment. 3: Once the driving piston passes Hose 3, air can flow into the anchor chamber. The
anchor chamber then becomes buoyant and the entire unit floats (rapidly) to the surface. 4: Once the height of the unit above the sediment
surface exceeds the chain length on the extension arm, the arm drops and opens a tube that releases air from the anchor chamber, slowing the
rapid ascent of the coring unit. The entire unit is usually taken to shore to disconnect the core tube for transport.

into concrete walls. The ring-shaped bit is made of diamonds 14.11.4.7 Freeze Corer
set in a heavy-duty carbide paste.
Hughes and Terasmae (1963) described an early version of Although excellent lake sediment cores can be collected with
a permafrost corer based on a modified ice corer with teeth of gravity corers, it can be difficult to prevent the mixing of
steel. Blake (1964, 1974, 1977) replaced the steel teeth with aqueous surface sediments during core retrieval and sub-
more durable teeth which are made from Carboloy (an alloy sampling. Freezing of surface lake sediments in situ is a useful
containing cobalt, tungsten, and carbon). Nørnberg et al. method for high-resolution studies of recent lake sediments,
(2004) described a Teflons-coated coring system that can particularly for the analysis of varved lake sediments
continuously core frozen peat up to 10 m deep in 70 cm long (Figure 11(b)). The basic freeze core design includes a con-
sections of 9.7 cm diameter. The Teflons coating helps pre- tainer filled with cooling agent that causes a layer of sediment
vent the samples or chips of peat from freezing to the tube. to freeze to the outside. The cooling agent is generally a
This unit includes spring-loaded cutting blades on the inside mixture of dry ice (solid CO2) and alcohol, but liquid nitro-
of the cutting head, which are designed to cut horizontally gen is occasionally used (Pachur et al., 1984; Schmidt et al.,
across the bottom of the core section when the rotation is 1995). The most basic type is a gravity freeze corer that sinks
reversed. into the sediments under its own weight. The shape of the
132 Coring and Augering

Both units connect duration of effective freezing from approximately 20–45 min,
to a motor sufficient to form an adequate thickness of frozen crust in
warm (422 1C) lakes and prevent its premature melting
Extension rod during retrieval through the water column.
Extension rod
A disadvantage of the freeze coring method is the possible
effect of ice crystal formation on the concentration of chemical
elements in aqueous sediment. Stephenson et al. (1996) re-
port that freeze core sediments have higher water content than
tube-cored sediment; which affected the concentration of
210
Pb and 137Cs in sediments from the experimental lakes area
in Canada. Stephenson et al. (1996) hypothesize that ice
crystal formation increases the water content of freeze cores,
resulting in lower 210Pb and 137Cs concentrations.

14.11.4.8 The Geo-Slicer

Most coring techniques capture a small (o10 cm) diameter
Core barrel column of sediment. In some cases, this is enough to capture
Core barrel
finer scale structures within a deposit, such as the deformation
of sediment layers by earthquakes (Monecke et al., 2004), but
not enough to capture larger scale structures and variation
within a deposit. If we need to see a larger sample of the
deposit we are studying to get a sense for larger structures (such
Cutting head as faults; see Chapter 5.10) Fault scarps and associated land-
forms) and variation within the deposit (such as lenses).
Cutting bit Digging a trench into the deposit is one obvious way to pro-
Cutting blade
10 cm vide us with a wider view of the deposit of interest (Chapter
(a) (b) 14.12). However, digging a trench is not possible or desirable
Figure 10 Schematic of the two main types of permafrost coring in many situations. These situations may include marine or
head. (a) A unit that is based on a commercially available diamond lacustrine settings or archeological sites of a sensitive nature
impregnated concrete drilling bit. (b) A modified ice drill with where disturbance must be kept to a minimum.
specialized cutting head for coring and cutting the base of the core Haraguchi et al. (1998) have developed a method for taking
after coring is completed. large wedges of relatively undisturbed sediment using a device
called a Geo-slicer. The Geo-slicer (Figure 12) comprises a
container may be a tube (Shapiro, 1958; Wright, 1980); large metal wedge-shaped chamber or pile that is pushed into
letter-box shaped (Huttunen and Meriläinen, 1978), or the target deposit using a vibrating head or ‘vibrohammer.’ The
wedge-shaped (Renberg, 1981; O’Sullivan, 1983). The thin wedge is pushed into the deposit and the inner cavity of the
wedge-shaped type (Figure 11(a)) produces the least sediment wedge fills with sediment. A ‘shutter’ or metal plate is then
disturbance during sediment penetration (Lotter et al., 1997). pushed in to the deposit adjacent to the wedge using the
Miskimmin et al. (1996) have produced a percussion freeze vibrohammer to trap the sediment in the wedge cavity.
corer with a tube-shaped container that is hammered into the The wedge and shutter are then pulled together to the surface.
sediment by a weight attached to a cable. The shutter is then removed to expose the sediment sample.
Modifications to the basic freeze corer type have been in- This method was designed specifically to study deposits
tended to provide a more controlled penetration of the lake containing evidence of past earthquake events. He et al. (2008)
sediments (prevent tilting), prevent lake water from freezing on used the Geo-slicer method to examine evidence for earthquake
the coring container, and facilitate freeze coring in warm tro- events in the Beijing area. He et al. (2008) took several Geo-slicer
pical lakes. Renberg and Hansson (1993) designed a system samples and compared the observations from these samples to
where the dry-ice/alcohol mixture was stored separately from data from a trench face. He et al. (2008) found that there was
the wedge, which avoided the formation of an ice crust prior to limited sample compression using this method (approximately
penetration. Once the wedge has penetrated the sediment, an 5%) and that the Geo-slicer preserved the structures (including
electrical pump circulated chilled alcohol into the wedge. faults) in the unconsolidated sediment that was sampled.
Lotter et al. (1997) improved the design of Renberg and Matsuoka and Ho Shin (2010) also used a Geo-slicer to
Hansson (1993) by including a tripod and hydraulic system for take samples from Ariake Sound in Japan in order to study
stability and controlled lowering of the corer into the sediment. environmental change over the last 150 years. This study
Verschuren (2000) made modifications to the basic freeze illustrated the utility of the Geo-slicer for sampling sediment
corer design that allows coring in warm tropical lakes. The in standing water and the advantages of taking a wide Geo-
design by Verschuren (2000) combines a standard wedge- slicer sample. Many studies of this nature require a lot of
shaped freeze-on container with a 20 l thermos (Figure 11(a)) sediment in order to provide enough for multiple analyses.
from which pellets of solid CO2 descend into the wedge to Wide, thick Geo-slicer samples could provide a larger sedi-
replace those that have sublimated. This system extends the ment volume for each layer than many other types of corer.
 Coring and Augering 133

Thermos
(sectioned)

Freeze wedge

(a) (b)

Figure 11 (a) A Schematic of a gravity driven freeze corer with a wedge-shaped freeze chamber. This type is designed for operation in warmer
tropical lakes (Verschuren 2000). Dry ice (frozen CO2) pellets drop from the insulated thermos at the top to replace sublimated pellets in the
freeze wedge. (b) Wedge with frozen sediment and sediment–water interface from the deepest part of Baldeggersee (66 m), Switzerland. Note the
presence of fine laminations or annual varves. Reproduced from Figure 4 in Lotter, A.F., Renberg, I., Hansson, H., Stöckli, R., Sturm, M., 1997. A
remote controlled freeze corer for sampling unconsolidated surface sediments. Aquatic Science 59, 295–303, with permission from Springer.

14.11.5 Core Handling and Contamination Control Nevertheless, it is still good practice to clearly label the top
and base in case of accidents.
14.11.5.1 Core Handling and Identification Each core, or core section, should also be given a unique
identifier (core code). Sample depth information should also
There are some simple, yet fundamental, procedures that need be written on the core and if necessary the core sample should
to be adhered to when dealing with retrieved cores in order to be cling wrapped or sample bags used to store subsamples
avoid contamination and ensure accurate data collection. from extruded gravity cores. Record the same core identifier
Firstly, retrieved cores should immediately be clearly labeled and core depth information along with core location (GPS
using permanent marker on the piping housing, marking the coordinates) and any other notable aspects that may be
top and base of every core or core section (i.e., this way up). relevant (e.g., depositional environment, water depth, core
Identifying the top (shallow) versus bottom (deeper) portion measured depth vs. core penetration etc.) in a field note-book.
of a gravity core is usually not a problem due to the need to Once the retrieved cores are back in the laboratory the
keep the gravity core vertical to prevent sediment mixing. cores or core subsection is generally split in two and the
134 Coring and Augering

Vibrohammer

Shutter Wedge

Figure 12 Inserting the Geo-slicer into the trench at Tule Pond (Tyson Lagoon), Hayward fault, Fremont, California. The wedge is being vibrated
into the ground by the vibrating head that grips the top of the wedge (the ‘vibrohammer’). The Shutter sits to the left of the wedge in the trench.
The shutter will be vibrated into position in front of the wedge cavity once the wedge is in the ground to hold in the sample. Photo taken by
Jennifer Adleman (USGS) and provided courtesy of the United States Geological Survey.

stratigraphy logged. Dating materials for different methods sediments from dehydrating (and shrinking) and prevents
need to be collected before and during the subsample mould from growing on the sediment. Freezing is not rec-
(Chapter 14.30). One half is then used for analyses and an- ommended for aqueous samples as the growth of large
other half is kept as an archive (in case further ice crystals may affect fine scale sedimentary structures or
analyses is required in the future). This is not always the delicate fossils. It is also a good idea to weigh bags containing
case as large amounts of sediment are commonly required to subsamples from a gravity core on a digital balance (correc-
perform all the desired analyses (e.g., sediment analysis, ting for the bag) as soon after coring as possible, as it is
geochemistry, geochronology, other environmental proxies inevitable that some moisture will be lost with long-term
etc.). Cores and core samples are generally wrapped to be air- storage. Moisture loss will affect sediment weights; sediment
tight using cling wrap and stored in a cool room. This prevents weights and water content can be crucial for many dating
 Coring and Augering 135

techniques, loss on ignition, and other environmental proxy in their original context (i.e., minimal mixing, compaction,
studies. or deformation during core collection).
However, the coring techniques detailed in this chapter
involve pushing a core tube into the sediment and friction on
14.11.5.2 Contamination Control the inside and outside of the core tube can potentially result in
Cross-contamination within the core when subsampling and sediment distortion and plugging (under-sampling or core
between cores when coring should always be considered shortening). A vibracore or piston corer can be used to com-
during logging and sampling. Coring equipment should al- pensate for internal core tube friction, but over-compensation
ways be cleaned prior to and in between each core/push being by the piston can result in core stretching or oversampling and
taken. The equipment should be rinsed with distilled water if plugging remains an issue in vibracoring. Friction between the
geochemical studies are required, but it is more often (for sediment and the core tube also causes smearing of the sedi-
practical reasons) rinsed in lake or sea water. We should be ment on the surface of the core. Both the deformation during
aware of that lake and sea water and even tap water contain the coring procedure and the contamination of the outer
pollen and diatoms (microscopic algae) that will cause surface due to smearing are common issues associated with
contamination if pollen and diatom analysis is required. All the coring methods illustrated in this chapter. Deformation
equipment (scalpel, spatula etc.) used for subsampling should and smearing need to be taken into account when attempting
also be cleaned thoroughly in between sample collection. For to reconstruct an accurate picture of landscape evolution or
sample collection we may use a modified syringe (with the changes in a depositional environment.
end sliced off) to subsample known volumes of sediment Core handling and contamination control is another im-
from cores in the laboratory. Sampling measured sediment portant consideration to ensure the integrity of the data col-
volumes will allow the calculation of influx rates for chemical lected from the retrieved core. It is fundamental that the
or biological proxies later. For an example of a sampling physical aspects of the retrieved core be accurately recorded
strategy, see Chivas et al. (2001). Researchers also need to be (top and base of core, core identifier, water depth, location,
aware of other sources of potential contamination. For ex- measured depth vs. retrieved thickness of core, etc.). The
ample, clothing fibers or even stray hairs may be enough to quality of the sediment core and subsamples can be com-
contaminate and severely compromise an AMS (Accelerator promised by improper handling, storage, and contamination
Mass Spectrometry) radiocarbon sample, especially an old of the sediment samples. Accordingly, before a researcher
sample where the concentration of 14C is very low. Laboratory obtains cores from a depositional environment they first need
analyses can be time consuming and expensive, a little bit of to choose the most appropriate means of collecting that core
effort in establishing an effective sampling protocol will re- and develop a rigorous sampling stratagem to ensure the
duce the risk of contamination, avoid expensive mistakes and quality of the data and minimize contamination.
produce better results.

References

14.11.6 Conclusion Australian Drilling Industry Training Committee Limited, 1996. Drilling: The
Manual of Methods, Applications, and Management. CRC Press, Boca Raton,
Obtaining subsurface (stratigraphic) data is commonly a USA, 615 pp.
Belokopytov, I.E., Beresnevich, V.V., 1955. Giktorf’s peat borers. Torfânaâ
fundamental component of developing a detailed and accur-
Promyslennost 8, 9–10.
ate understanding of the geomorphological evolution of a Bentley, C.R., Koci, B.R., 2007. Drilling to the beds of the Greenland and Antarctic
specific landform, depositional environment, or even. One ice sheets: a review. Annals of Glaciology 47, 1–9.
way of collecting this stratigraphic data is by collecting sedi- Black, H.J., Dainat, M., Köster, M., Meyer-Reil, L.-A., 2002. A multiple corer for
ment cores. The type of coring technique used will be taking virtually undisturbed samples from shallow water sediments. Estuarine,
Coastal and Shelf Science 54, 45–50.
dependent on the particular type of sediment to be cored, the
Blake, Jr. W., 1964. Preliminary account of the glacial history of Bathurst Island,
depth of penetration required, the desired preservation of the Arctic Archipelago. Geological Survey of Canada Paper 64–30, 8.
cored material (undisturbed or semidisturbed), and the cost Blake, Jr. W., 1974. Periglacial features and landscape evolution, central Bathurst
effectiveness of utilizing the desired coring technique. This Island, District of Franklin. Report of Activities, Geological Survey of Canada
chapter has described the common coring techniques, the Paper 74-1B, 235–244.
Blake, Jr. W., 1977. Radiocarbon age determinations from the Carey Islands,
types of settings that they are most appropriate for and the
Northwest Greenland. Report of Activities, Part A, Geological Survey of Canada
advantages and disadvantages of each. In addition there are Paper 77-1A, 445–454.
some fundamental aspects that need to be considered to en- Blomqvist, S., 1991. Quantitative sampling of soft-bottom sediments: problems and
sure that representative cores are collected including: solutions. Marine Ecology Progress Series 72, 295–304.
Brinkhurst, R.O., Chua, K.E., Batoosingh, E., 1969. Modifications in sampling
• Sedimentary core sections should be as representative procedures as applied to studies on the bacteria and tubificid oligochaetes
as possible of the landscape, depositional environment, or inhabiting aquatic sediments. Journal of the Fisheries Research Board of Canada
event being investigated (i.e., as many layers of strata col- 26, 2581–2593.
Buckley, D.E., MacKinnon, W.G., Cranston, R.E., Christian, H.A., 1994. Problems
lected as possible so there is no time missing from the
with piston core sampling: mechanical and geochemical diagnosis. Marine
record). Geology 117, 95–96.
• The core section should, where possible, faithfully preserve Calmels, F., Gagnon, O., Allard, M., 2005. A portable earth-drill system for
sedimentary structures, facies changes, and biostratigraphy permafrost studies. Permafrost and Periglacial Processes 16, 311–315.
136 Coring and Augering

Chambers, J.W., Cameron, N.G., 2001. A rod-less piston corer for lake sediments: Lyons, W.C., 2009. Air and Gas Drilling Manual. Gulf Publishing, Burlington, USA,
an improved, rope-operated percussion corer. Journal of Paleolimnology 25, 776 pp.
117–122. Mackereth, F.J.H., 1958. A portable sampler for lake deposits. Limnology and
Chivas, A.R., Garcia, A., van der Kaars, S., 2001. Sea-level change since the last Oceanography 3, 181–191.
interglacial in the Gulf of Carpentaria, Australia: an overview. Quaternary Matsuoka, K., Ho Shin, H., 2010. Environmental changes in the inner part of Ariake
International 83–85, 19–46. Sound, West Japan Recorded in Dinoflagellate Cyst Assemblages. In: Ishimatsu,
Cumming, B.F., Glew, J.R., Smol, J.P., Davis, R.B., Norton, S.A., 1993. Comment A., Lie, H.-J. (Eds.), Coastal Environmental and Ecosystem Issues of the East
on ‘‘core compression and surficial sediment loss of lake sediments of high China Sea. TERRAPUB and Nagasaki University, Tokyo, pp. 111–120.
porosity caused by gravity coring’’ (Crusius and Anderson). Limnology and McCoy, F.W., 1985. Mid-core flow-in: implications for stretched stratigraphic
Oceanography 38, 695–699. sections in piston cores. Journal of Sedimentary Petrology 55, 608–610.
Cushing, S., Desiardins, J., Filion, J.-M., 1997. Parachute-assisted gravity sediment Miskimmin, B.M., Jefferson Curtis, P., Schindler, D.W., Lafaut, N., 1996. A new
corer (Algonquin Corer). Journal of Paleolimnology 18, 307–311. hammer-driven freeze corer. Journal of Paleolimnology 15, 265–269.
De Nicola, A., Randolph, M.F., 1997. The plugging behavior of driven and jacked Monecke, K., Anselmetti, F.S., Becker, A., Sturm, M., Giardini, D., 2004. The record
piles in sand. Geotechnique 47, 841–856. of historic earthquakes in lake sediments of central Switzerland. Tectonophysics
De Vleeschouwer, F., Chambers, F.M., Swindles, G.T., 2010. Coring and sub- 394, 21–40.
sampling of peatlands for palaeoenvironmental research. Mires and Peat 7, Morton, R.A., White, W.A., 1997. Characteristics of and corrections for core
1–10. shortening in unconsolidated sediments. Journal of Coastal Research 13,
Digerfeldt, G., 1978. A simple corer for sediment sampling in deep water. 761–769.
Department of Quaternary Geology, University of Lund, Report 14, 8 pp. Nesje, A., 1992. A piston corer for lacustrine and marine sediments. Arctic and
Gilbert, R., Glew, J.R., 1985. A portable percussion coring device for lacustrine and Alpine Research 24, 257–259.
marine sediments. Journal of Sedimentary Research 55, 607–608. Nørnberg, T., Goodsite, M.E., Shotyk, W., 2004. An improved motorized corer and
Glew, J.R., 1991. Miniature gravity corer for recovering short sediment cores. sample processing system for frozen peat. Arctic 57, 242–246.
Journal of Paleolimnology 5, 285–287. O’Sullivan, P.E., 1983. Annually-laminated lake sediments and the study of
Glew, J.R., 1995. Conversion of shallow water gravity coring equipment for deep quaternary environmental changes – a review. Quaternary Science Reviews 1,
water operation. Journal of Paleolimnology 14, 83–88. 245–313.
Glew, J.R., Smol, J.P., Last, W.M., 2001. Sediment core collection and extrusion. Pachur, H.-J., Denner, H.-D., Walter, H., 1984. A freezing device for sampling the
In: Last, W.M., Smol, J.P. (Eds.), Tracking Environmental Change Using Lake sediment–water interface of lakes. Catena 11, 65–70.
Sediments. Kluwer Academic Publishers, Dordrecht, Basin Analysis, Coring, and Paik, K., Lee, S., 1993. Behaviour of soil plugs in open-ended model piles driven
Chronological Techniques, pp. 73–105. into sands. Marine Georesources and Geotechnology 11, 353–373.
Grumenskii, B.M., Kamarov, N.S., 1961. Soil Drilling by Vibrations (Translated from Pierce, J.W., Howard, J.D., 1969. An inexpensive portable vibracorer for sampling
Russian). Consultants Bureau, New York, 80 pp. unconsolidated sands. Journal of Sedimentary Petrology 55, 605–606.
Haraguchi, T., Nakata, T., Shimazaki, K., Imaizumi, T., Kojima, K., Ishimaru, K., Randolph, M.F., Leong, E.C., Houlsby, G.T., 1991. One-dimensional analysis of soil
1998. A new sampling method of unconsolidated sediments by long Geo-slicer, plugs in pipe piles. Geotechnique 41, 587–598.
a pile-type soil sampler. Journal of the Japan Society of Engineering Geology Reasoner, M.A., 1993. Equipment and procedure improvements for a lightweight,
39, 306–314. inexpensive, percussion core sampling system. Journal of Paleolimnology 8,
He, H., Min, W., Haraguchi, T., 2008. Testing Geo-slicer on the rupture of the M8 273–281.
Sanhe-Pinggu earthquake of. Journal of Geography 106, 59–69 (In Japanese Renberg, I., 1981. Improved methods for sampling, photographing and varve-
with English abstract and figure captions). counting of varved lake sediments. Boreas 10, 255–258.
Hughes, O.L., Terasmae, J., 1963. SIPRE ice-corer for obtaining samples from Renberg, I., Hansson, H., 1993. A pump freeze corer for recent sediments.
permanently frozen bogs. Arctic 16, 270–272. Limnology and Oceanography 38, 1317–1321.
Huttenen, P., Meriläinen, J., 1975. Modifications of a rodless core sampler for Renberg, I., Hansson, H., 2008. The HTH sediment corer. Journal of Paleolimnology
investigating lake sediments. Publications of the University of Joensuu, Series 40, 655–659.
BII 4, 4 pp. Robello, S.G., 2007. Downhole Drilling Tools: Theory and Practice for Engineers
Huttunen, P., Meriläinen, J., 1978. New freezing device providing large unmixed and Students. Gulf Publishing, Burlington, USA, 664 pp.
sediment samples from lakes. Annales Botanici Fennici 15, 128–130. Sanders, J.E., Imbrie, J., 1963. Continuous cores of Bahamian calcerous sands
Hvorslev, M.J., Stetson, H.C., 1946. Free-falling coring tube: a new type made by vibrodrilling. Geological Society of America Bulletin 74, 1287–1292.
of gravity bottom sampler. Geological Society of America Bulletin 57, Schmidt, R., Hollerer, H., Wallner, G., 1995. A vacuum sampler for subsampling
935–950. freeze-dried laminated sediments with the application to in situ frozen varves of
Jahnke, R.A., Knight, L.H., 1997. A gravity-driven, hydraulically-damped multiple Mondsee, Austria. Journal of Paleolimnology 14, 93–96.
piston corer for sampling fine-grained sediments. Deep Sea Research I 44(4), Shapiro, J., 1958. The core-freezer – a new sampler for lake sediments. Ecology
713–718. 39, 748.
Johnsen, S.J., Hansen, S.B., Sheldon, S.G., 2007. The Hans Tausen drill: design, Skinner, L.C., McCave, I.N., 2003. Analysis and modeling of gravity- and piston
performance, further developments and some lessons. Annals of Glaciology 47, coring based on soil mechanics. Marine Geology 199, 181–204.
89–98. Sloss, C.R., Jones, B.G., Murray-Wallace, C.V., 2005. Holocene sea level change
Johnsen, S.J., Gundestrup, N.S., Hansen, S.B., Schwander, J., Rufli, H., 1994. The and the evolution of a barrier estuary: a case study, Lake Illawarra, NSW,
new improved version of the ISTUK ice core drill. Memoirs of the National Australia. Journal of Coastal Research 21, 943–959.
Institute of Polar Research, Special Issue, 49, 9–23. Sloss, C.R., Jones, B.G., Switzer, A.D., Nichol, S.P., Clement, A.J.H., Nicholas, A.W.,
Jowsey, P.C., 1966. An improved peat sampler. New Phytologist 65, 245–248. 2010. The Holocene infill of Lake Conjola, a narrow incised valley system on
Kajak, Z., 1965. Analysis of quantitative benthic methods. Ekologia Polska 11, the southeast coast of Australia. Quaternary International 221, 23–35.
1–56. Sloss, C.R., Murray-Wallace, C.V., Jones, B.G., 2006. Aminostratigraphy of two
Kanneworff, E., Nicolaisen, W., 1973. The ‘‘Haps’’. A frame-supported bottom corer. Holocene wave-dominated barrier estuaries in southeastern Australia. Journal of
Ophelia 10, 119–129. Coastal Research 22, 113–136.
Lanesky, D.E., Logan, B.W., Brown, R.G., Hine, A.C., 1979. A new approach to Smith, D.G., 1992. Vibracoring: recent innovations. Journal of Paleolimnology 7,
portable vibracoring underwater and on land. Journal of Sedimentary Petrology 137–143.
49, 654–657. Smith, D.G., 1998. Vibracoring: a new method for coring deep lakes.
Larsen, C.P.S., MacDonald, G.M., 1993. Lake morphometry, sediment mixing and Palaeogeography, Palaeoclimatology, Palaeoecology 140, 433–440.
the selection of sites for fine resolution palaeoecological studies. Quaternary Smol, J.P., Cumming, B.F., Dixit, A.S., Dixit, S.S., 1998. Tracking recovery patterns
Science Reviews 12, 781–792. in acidified lakes: a paleolimnological persepective. Restoration Ecology 6,
Livingstone, D.A., 1955. A lightweight piston sampler for lake deposits. Ecology 36, 318–326.
137–139. Stager, J.C., 1992. Simple gravity corer suitable for educators. Journal of
Lotter, A.F., Renberg, I., Hansson, H., Stöckli, R., Sturm, M., 1997. A remote Paleolimnology 7, 253–255.
controlled freeze corer for sampling unconsolidated surface sediments. Aquatic Stein, J.K., 1991. Coring in CRM and archaeology: a reminder. American Antiquity
Science 59, 295–303. 56, 138–142.
 Coring and Augering 137

Stephenson, M., Klaverkamp, J., Motycka, M., Baron, C., Schwartz, W., 1996. Wilhelm, F.M., Lasenby, D.C., 1997. A sediment core sectioning method to
Coring artifacts and contaminant inventories in lake sediment. Journal of minimisze the post-sampling redistribution of organisms. Aquatic Sciences 59,
Paleolimnology 15, 99–106. 34–38.
Verschuren, D., 2000. Freeze coring soft sediments in tropical lakes. Journal of Woodward, C.A., Shulmeister, J., 2006. New Zealand chironomids as proxies for
Paleolimnology 24, 361–365. human-induced and natural environmental change: transfer functions for temperature
Watson, J.E., Brooks, S.J., Whitehouse, N.J., Reimer, P.J., Birks, H.J.B., Turney, and lake production (chlorophyll a). Journal of Paleolimnology 36, 407–429.
C.M., 2010. Chironomid-inferred late-glacial summer air temperatures from Wright, H.E., 1980. Coring of soft lake sediments. Boreas 9, 107–114.
Lough Nadourcan, Co. Donegal, Ireland. Journal of Quaternary Science 25, Wright, H.E., 1991. Coring tips. Journal of Paleolimnology 6, 37–49.
1200–1210.

Biographical Sketch

Dr. Craig Woodward completed his PhD in Environmental Science at the University of Canterbury, New Zealand
in 2006 and subsequently took up research positions in Ireland and Canada. He is currently based in the School
of Geography, Planning and Environmental Management at The University of Queensland, Australia. Dr.
Woodward specialiszes in limnology (the study of lakes) and paleolimnology - the study of past changes in
aquatic environments. He uses a wide variety of the techniques described in this chapter to collect sediment
records from lakes and bogs that can tell us about past climate change and human impacts on aquatic ecosystems.

Dr. Craig Sloss completed his PhD in Science at the University of Wollongong, NSW, Australia in 2005. He took
up an academic position as a coastal geomorphologist at Massy University, Palmerston North from 2006–2008.
He is currently employed as a sedimentologist at the Queensland University of technology. The research interests
of Dr. Sloss research interests include sea-level and climate change over the last glacial cycle, coastal geo-
morphology and sedimentology, and extreme events (storm surge and/or tsunami). He is also interested in the
anthropogenic impacts on coastal and marine environments and the role that of humans have had on the historic
modification of coastal geomorphology.
 14.12 Trenching and Exposed Faces
JP McCalpin, GEO-HAZ Consulting, Inc., Crestone, CO, USA
r 2013 Elsevier Inc. All rights reserved.

14.12.1 The Purpose of Trenching and Mapping Exposed Faces 138

14.12.2 Creating an Exposed Face (Trenching) 139
14.12.2.1 Introduction 139
14.12.2.2 Improving a Natural Exposure 139
14.12.2.3 Digging a Trench 139
14.12.2.3.1 Location, orientation, and pattern of trenches 139
14.12.2.4 Shape and Depth of the Excavation 139
14.12.2.5 Trench Safety 139
14.12.2.5.1 Dewatering the trench 141
14.12.3 Preparing the Exposed Face for Mapping (Logging) 141
14.12.4 Logging the Exposed Face 141
14.12.4.1 Identifying Mappable Units and Marking Their Contacts 141
14.12.4.2 Mapping Soil Horizons in Trenches 142
14.12.4.3 Subjective versus Objective Logging 142
14.12.4.4 Techniques of Trench Logging 143
14.12.4.4.1 Photomosaic logging (2-D) 143
14.12.4.4.2 Photogrammetric logging (3-D) 143
14.12.5 Applications of Trenching in Geomorphology 144
14.12.5.1 Structural Targets 144
14.12.5.1.1 Active fault studies (paleoseismology) 144
14.12.5.1.2 Landslide studies 147
14.12.5.1.3 Sackung (deep-seated gravitational spreading) studies 147
14.12.5.1.4 Sinkhole studies 148
14.12.5.2 Stratigraphic Targets 148
14.12.6 Summary 148
References 149

Glossary Paleoseismology The study of prehistoric earthquakes

Accommodation space The volume of a topographic using preserved geologic evidence of fault rupture and
depression formed by downward displacement along a ground shaking.
surface fault, landslide headscarp, or sackung scarp. Sackung Antislope scarps and grabens formed by deep-
Event horizon An unconformity produced by a surface seated gravitational spreading of mountain ridges.
displacement event. Unconformity A planar surface between geologic deposits
Fault-angle depression An asymmetrical accommodation that represents a missing interval of geologic time, produced
space, formed by dropping and tilting the ground surface either by an interruption in deposition or by the erosion of
toward a surface deformation scarp. strata followed by renewed deposition.

Abstract

In geomorphology, trenching refers to creating an artificial exposure of the geologic deposits and structures in the shallow
subsurface. Most trenches in modern studies are excavated with machinery, so the choices of trench length, width, depth,
and orientation are limited only by the imagination of the investigator. The primary output of a trenching study is a scaled
map of the vertical trench walls, called a trench log. Trenching is widely used in applied geomorphological studies, for
example, paleoseismology, landslide, and sinkhole hazards.

McCalpin, J.P., 2013. Trenching and exposed faces. In: Shroder, J. (Editor in
Chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise on Geomorphology.
Academic Press, San Diego, CA, vol. 14, Methods in Geomorphology,
pp. 138–149.

138 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00379-1

 Trenching and Exposed Faces 139

14.12.1 The Purpose of Trenching and Mapping a deposit are dug either parallel or perpendicular to the in-
Exposed Faces ferred flow direction (or perhaps both). The depth of tren-
ching is limited, either by the mode of excavation (by hand
Several reasons exist to map (or create) exposed vertical and machinery), by the system of wall stabilization (shoring),
faces. The first is to relate the landform to the underlying geo- or by encountering groundwater.
logic stratigraphy and structure; this is commonly critical for In trench studies of active structures (faults, landslide
determining the origin of a landform. A second reason is to planes, sackungen, and sinkhole margins), trench placement
collect samples to date the formation of the landform. A third is such a critical element that success or failure of the in-
reason is to determine if the landform was created by a process vestigation characteristically depends on it. Stated another
that, if repeated in the future, would constitute a geologic hazard. way, once a trench is sited, the stratigraphy and structure ex-
Unfortunately, clean natural exposures are rare, so in the posed are predetermined. A poorly located trench will yield
past few decades the trenching technique has been developed. poor results; the data may prove insufficient to test the hy-
In this usage, ‘trenching’ refers to creating an artificial vertical potheses and, thus, defeat the purpose of trenching.
exposure by excavating into a landform. To date, trenching Trench locations may also be restricted by nongeological
has been utilized mainly in geologic hazard studies, includ- considerations, such as road access, land ownership, and
ing paleoseismic studies (see Chapter 5.12); landslide and previous ground-surface disturbance. Whenever possible,
debris-flow investigations (see Chapters 7.14, 7.15, 7.16, and trenching sites should have undergone minimal prior surface
7.17); and sinkhole/karst studies (see Volume 6). disturbance (grading and filling), which might destroy the
critical relationships of faults to shallow deposits, especially
the modern soil. However, trenching has been successful
14.12.2 Creating an Exposed Face (Trenching) where the original ground surface and underlying structural
target were buried beneath artificial fill, and thus preserved.
14.12.2.1 Introduction
The term trenching herein is used as a generic term for making
14.12.2.4 Shape and Depth of the Excavation
an artificial exposure, normally (but not always) with one or
more vertical walls. The techniques for mapping (logging) the As pointed out by Hatheway and Leighton (1979: p. 178), ‘‘the
vertical walls can also be applied to a natural exposure which method of excavation that proves to be least disturbing to the
has been cleaned off. In either case, the goal is the same: to host soil and/or rock will also likely prove to be the least
create a vertical or near-vertical exposure that is clear enough expensive.’’ Four basic types of trenches exist, excavated by
that details of subsurface lithology, stratigraphy, and structure hand, or by machines (rubber-tired backhoe loaders, tracked
can be mapped and recorded. hydraulic excavators (‘trackhoes’), rubber-tired (wheel) load-
Machine-dug trenches are typically long and narrow, but ers, scrapers, and track-type tractors (‘bulldozers’); Table 1).
the investigator can make them any shape, depending on the Trenches may be excavated completely before logging
local topography and the objective of trenching. Since the begins, or dug incrementally as logging proceeds ahead of
1980s, trenches have become more widely used, more diverse, backfilling. In most cases, complete excavation is advantageous
and customized in shape and larger. because the entire trench can be viewed at once and critical
areas reexamined. However, due to the instability of trench
walls or lack of shoring, some trenches may have to be dug and
14.12.2.2 Improving a Natural Exposure
logged in increments. ‘Incremental trenching’ refers to the pro-
Making an artificial excavation is generally logistically difficult gressive excavation of closely spaced, parallel trench walls. Once
and expensive, so geomorphologists look first for fortuitous the initial trench is excavated and logged, the logged trench wall
natural exposures. Unfortunately, most natural exposures are is excavated back 20–50 cm, and the new exposure logged. That
not vertical and are partly covered with weathered materials wall is then cut back 20–50 cm parallel to the previous wall,
that obscure their stratigraphic and structural details. However, and is again logged. As the trench wall is progressively cut back,
steep natural exposures can be improved to a mappable con- many successive wall positions are mapped. The resulting
dition by cutting them to a steeper angle and removing the closely spaced, parallel trench logs can then be used to create a
covering deposits. This can be done manually (Figures 1(a) three-dimensional diagram of structural and stratigraphic re-
and 1(c)) or by excavating machinery (Figures 1(b) and lations in the subsurface. At present, this technique has been
1(d)–1(f)). mainly used on strike-slip fault traces, where a three-dimen-
sional (3-D) representation is needed to calculate displacement
14.12.2.3 Digging a Trench vectors. This method of trenching is totally destructive, however,
because excavation consumes the entire feature being mapped.
14.12.2.3.1 Location, orientation, and pattern of
trenches
Trench location and orientation depend on the purpose of
14.12.2.5 Trench Safety
trenching, a topic described more fully at the end of this
chapter. In brief, trenches targeting planar subsurface struc- Natural geologic exposures have evolved over geologic time
tures (e.g., an active fault) are sited where the structure is be- and have generally degraded to (or below) the angle of repose,
lieved to be and oriented perpendicular to its inferred strike. so are relatively safe to work on. The same is not true of an
Trenches dug to describe the stratigraphy or sedimentology of artificially cut vertical face that can collapse without warning.
140 Trenching and Exposed Faces

Spoil pile
Terrace T4 Terrace T4
Graben Trench 1

Trench 2 Trench 3
Graben Graben
margin margin
Trench 4

(a) (b) (c)

(d) (e)
(f)

Figure 1 Types of trenches. (a) Natural streamcut exposure, cleaned off with shovels to expose the structures within a Quaternary graben. (b)
Telephoto view of four ‘half-trenches’ excavated along the outcrop of a normal fault on a steep, eroded hillslope. Note the person at center of
trench 4 for scale. Each trench overlaps the other in elevation, to expose a continuous 17-m-high stratigraphic section of the faulted strata on
both sides of the fault (shown by arrows). (c) Hand-dug trench across a sackung scarp in south coastal Alaska (5.5-m long, 0.75-m wide, up to
1.5-m deep); the string grid has a 0.5-m spacing. (d) Multibenched trench dug for a geologic hazard study in the USA. The trench, with 13 levels
of vertical walls each 1.5-m high, was excavated by scrapers (see inset, lower right). (e) Open-pit trench dug in the Venezuelan style by
Audemard (2005). Note the exposed fault plane (white) on the left wall. (f) Single-slot trench shored on 2-m centers with hydraulic aluminum
expandable shores, stacked in two tiers (each shore is 2.15-m high). From the Kaysville 1988 trench on the Wasatch fault, Utah, USA. (a)
Cliffhanger exposure of Carver, G.A., Bemis, S.P., Solie, D.N., Castonguay, S., Obermiller, K.E., 2010. Active and potentially active faults in or
near the Alaska Highway corridor, Dot Lake to Tetlin Junction, Alaska. Preliminary Interpretive Report 2010-1, Alaska Division of Geological and
Geophysical Surveys, Fairbanks, AK, 42 pp., with permission from Geophysical Survey. (b) Reproduced from McCalpin, J.P., Olig, S.S., Harrison,
J.B.J., Berger, G.W., 2006. Quaternary faulting and soil formation on the County Dump fault, Albuquerque, New Mexico. New Mexico Bureau of
Geology and Mineral Resources, Circ 212, 38 pp., with permission from Geology and Mineral Resources. (c) Reproduced from McCalpin, J.P.,
Bruhn, R.L., Pavlis, T.L., Gutierrez, F., Guerrero, J., Lucha, P., 2011. Antislope scarps, gravitational spreading, and tectonic faulting in the western
Yakutat microplate, south coastal Alaska: Geosphere, Special Issue 7(5), 1143–1158, with permission from Geosphere. (d) Reproduced from
McCalpin, J.P., Shlemon, R.J., 1996. Trenching as a tool in geologic investigations. The Professional Geologist (American Institute of Professional
Geologists) 33(12), 19–22, with permission from American Institute of Professional Geologists. (e) Photo courtesy of Franck Audemard,
FUNVISIS, Venezuela. (f) Reproduced from McCalpin, J.P., Forman, S.L., Lowe, M., 1994. Reevaluation of Holocene faulting at the Kaysville site,
Weber segment of the Wasatch fault zone, Utah. Tectonics 13(1), 1–16, with permission from Tectonics.

Table 1 Types of trenches and their characteristics

Trench type Depth Machinery Advantage Disadvantage

Manual c. 2 m (Carver None; hand tools only Only choice where machinery Limited area and depth; slow to
et al., 2008) cannot be brought to site dig
Machine, narrow c. 4.3 m Backhoe-loader; excavator Digging and backfilling is fast; Can only see small areas of wall
slot (trackhoe) minimum surface area at a time; reviewing with large
disturbance; easy to shore groups is difficult; difficult to
photograph
Machine, benched unlimited Excavator; front-end loader; Depth is nearly unlimited; Due to government regulations,
bulldozer; scraper digging and backfilling is fast; width must be at least twice
shape can be changed as the depth; large surface area
excavation progresses; disturbed
shoring not required; walls
are easily seen and
photographed; trench has
long-term stability
Machine, open pit unlimited Excavator; bulldozer Same as above, but less Large surface area disturbed;
volume need be excavated to log of non-vertical walls must
achieve the same depth be rectified
 Trenching and Exposed Faces 141

Wall collapses are caused by weak materials and/or seepage that faithfully reproduces the salient features. Trench logging
pressure from groundwater entering the trench, so dewatering was originally done by freehand drawing on gridded paper,
is critical for improving safety. To ensure a safe working but recent advances in technology have created additional
environment, the trench must be designed deliberately to options.
minimize the possibility of wall collapse, a subject typically
regulated by national health and safety agencies.
14.12.4.1 Identifying Mappable Units and Marking Their
14.12.2.5.1 Dewatering the trench Contacts
Dewatering trenches increases their safety and makes the
trench a more pleasant place in which to work. Dewatering The lithologic units on an exposed face are normally Qua-
strategies can be passive or active. On sloping ground, passive ternary sediments, and are differentiated as discrete deposits
dewatering is achieved by digging a shallow ditch downslope characterized by a consistent texture, sorting, bedding, fabric,
from the toe of the trench, such that groundwater entering the or color. Soil horizons, by contrast, are postdepositional
trench will simply continue flowing down the trench axis and weathering zones that may be developed on a single lithologic
out the lower end. Audemard (2005) described this method of unit, or may be developed across multiple lithologic units.
draining sag ponds. On flat ground there are two dewatering Identifying lithologic units and soil horizons on trench walls
options, both active. Periodically pump the water out of the is facilitated if lithologic contrasts are emphasized by using
trench and release it some distance away. However, in this some wall-treatment technique. Slight differences in deposit
method groundwater will continue to intersect the trench cohesion are accentuated if the trench wall is left to ‘weather’
walls and seep out, so the bottom of the trench will always for several days or weeks. Wind and rain can then etch
have some water in it. The part of the wall submerged will out differential relief between different strata and even reveal
have to be recleaned before logging. A preferable approach is subtle structures such as cross-bedding in loose sands. Similar
to dig shallow pits or drill relief wells around the trench and to relief can sometimes be created by repeated brushing of the
lower the groundwater level by pumping, until it falls below face with brooms or paintbrushes. Conversely, some contacts
the trench floor. Although this method keeps the trench walls appear sharper when moist, so walls can be sprayed or misted
stable and the trench floor dry, it is more expensive than with a portable water sprayer immediately before logging.
simply pumping the water out of the trench floor. Vague stratigraphic contacts with little textural or color
contrasts can commonly be seen better in diffuse-lighting
conditions than in direct sunlight. At mid-day, the sun is
14.12.3 Preparing the Exposed Face for Mapping nearly vertical, which accentuates (via shadowing) minor ir-
(Logging) regularities and tool marks on a vertical wall at the expense of
subtle textural or color variations. Frost et al. (1991) suggested
Before a trench can be logged, the walls must be cleaned well logging trenches at night with artificial illumination, the dir-
enough to expose the structures, stratigraphy, and soil horizons ection and intensity of which can be controlled. A corollary
(Table 2). In most trench investigations, in the USA, only one use is that of ultraviolet illumination of trench walls at night,
wall is cleaned and mapped (logged). This wall is usually which accentuates the fine structure of soils containing cal-
chosen to be the shaded one, because it is easier to trace con- cium carbonate.
tacts on a wall which is uniformly lit. Because of the extra labor Contacts identified visually are commonly accentuated in
involved in cleaning the opposite trench wall, it is usually done some way before logging, for example, by scribing a line on
in selected places to confirm important features seen in the the trench wall with a knife or edge of a trowel (in finer
logged wall (e.g., the trend of faults and paleochannels). sediments). In coarser sediments, one marks contacts with
nails and attached colored flagging, or with spray paint, using
unique colors for soil horizons, depositional contacts, ero-
14.12.4 Logging the Exposed Face sional contacts, faults, etc. In the corresponding trench log,
target features of the highest importance for the particular
To log the stratigraphy and structure of an exposed face, it is study (e.g., faults, tension cracks, liquefaction features, land-
necessary to define mappable units, mark them on the face, slide shear planes, sinkhole collapse zones, and angular
and produce some type of scaled drawing or image of the face unconformities) are rendered by the thickest lines, lithologic

Table 2 Recommended methods for cleaning trench walls and erecting string grids

Task Material Recommended techniques

Cleaning the walls Very coarse-grained soils Blow with compressed air
Coarse-grained, cohesionless soils Broom and brush
Fine-grained, cohesive soils Scrape with hoes and hand-tools (trowels)
Very cohesive soils Pluck with a claw hammer, or small mattock; Blast with a water
pressure washer
Constructing the reference grid A single trench Attach horizontal and vertical string lines, 0.5 to –1 m apart
A group of trenches Relate each trench’s grid to a common surveyed datum
142 Trenching and Exposed Faces

contacts by thinner lines, and soil horizon boundaries or differentiated the subjective versus objective approaches to
facies boundaries within major (genetic) depositional units by trench logging. In subjective logging, the logger first observes
very thin or dashed lines. the trench wall and makes a geologic interpretation of the
structural and stratigraphic relations exposed in the wall. The
correctly scaled log is then made to illustrate the primary
14.12.4.2 Mapping Soil Horizons in Trenches
geologic features. The rock or soil matrix is added in secondary
Soil horizons are important markers in trench exposures be- importance; small features that do not bear on the major in-
cause they indicate the location of past ground surfaces in the terpreted structures or strata may not be logged at all. The log
stratigraphic sequence, and their degree of development may is, thus, schematic (Figure 3) but planimetrically accurate. The
indicate the length of time that surface was stabilized. How- subjective approach to trench logging developed during nu-
ever, the interaction of soil profiles with lithologic units and clear power plant investigations (Hatheway and Leighton,
structures is generally critical to understanding the sequence 1979) where the log was meant to answer specific regulatory
of depositional events versus deformation events and their questions, such as ‘is a fault present?’ and, if so, ‘is the age of
relative timing (Shlemon, 1985). For example, if a soil is faulting older than some predefined regulatory criterion?’
developed on tectonically displaced strata and is truncated at The advantages of a subjective log are that it can be made
the fault scarp along with those strata, the faulting postdates rapidly and is easy to interpret with respect to regulatory cri-
both stratal deposition and the development of its soil profile. teria, because all extraneous features that do not bear on the
In contrast, if strata are displaced but the soil horizons extend major interpretation have been omitted. The disadvantage of
unbroken across the scarp face, the faulting postdates stratal this type of log is that it is difficult to advance alternative
deposition but predates development of the soil profile. interpretations of the log, because the interpretation was in-
Figure 2 shows an example of the complex relations created tegral to drafting the log and, thus, many details (which may
by soil formation that began before the initial faulting, and conflict with the interpretation) have been omitted.
continued throughout a history of repeated faulting, fault- By contrast, objective logging ‘‘attempts to portray equally
induced erosion, and colluvial deposition. all physical features of the trench face, larger than a threshold
To accurately identify and map soil horizons separately of resolution, in an impartial manner and without regard to
from lithologic units on an exposed face requires some formal relative importance. Both obvious and subtle features are
training in pedology, something that many geomorphologists shown with equal resolution, and little subjective interpret-
lack. Techniques for recognizing and delineating soil horizon ation is made during the recording process’’ (Hatheway and
contacts are beyond the scope of this chapter; see Birkeland Leighton, 1979: p. 173). This approach seeks only to docu-
(1999) for an excellent summary. Birkeland et al. (1991) also ment what the trench wall looks like (Figure 4). The most
described some applications of pedology to fault trenching. extreme example of an objective log would be an unannotated
photograph of the trench wall. The advantage of an objective
trench log is that several interpretations can be proposed and
14.12.4.3 Subjective versus Objective Logging tested against the stratigraphic relations portrayed on the log.
For a trench log to communicate information to its intended The log also acts as an archival record of how the trench wall
audience, the map units must be defined in a way that helps appeared, which may be recognized in the future as con-
this communication. Hatheway and Leighton (1979) taining some newly discovered phenomenon. The

Relict soil = 100% (130 000 yr) Colluvial soils: 1 = 5% (6000 yr)
2 = 15% (20 000 yr)
3 = 20% (26 000 yr)
4 = 10% (13 000 yr)
Buried soil = 50% (65 000 yr)
Relict soil Colluvial soil 1

Colluvial soil 2

Colluvial soil 3

Colluvial soil 4
Buried soil

Figure 2 Example of complex soil profiles on a fault scarp. On the downthrown block, the buried soil predates faulting. Colluvial soils 4, 3, and
2 were developed on scarp-derived colluvium deposited after the first three faulting events; they are all truncated at the fault plane. Colluvial soil
1 postdates the most recent faulting event and is not truncated. On the upthrown block, the relict soil developed over the entire pre- and
postfaulting time span. Reproduced from Birkeland, P.W., Machette, M.N., Haller, K.M., 1991. Soils as a tool for applied Quaternary geology.
Utah Geological Survey, Miscellaneous Publication 91–3, 1–63, with permission from Utah Geological Mineral Survey.
 Trenching and Exposed Faces 143

East

11 Cca 11A
Geochronology:
0
Sample Sample 11
number type Date 10
ITL-50 TL 1000 ± 100 1
14C 3
Schwartz and Coppersmith 7300 ± 1000
West (1984)
2
ITL-50 25 20a
2
11A 20b
3 3
11Cca
11
10
Height (m)

4 4
1
3

5 5
Bas
e of 2
expo
6 sure 1 6
Note: General location of Schwartz and Coppersmith’s
(1984) date of 7300 ± 1000 yr B.P

12 11 10 9 8 7 6 5 4 3 2 1
Distance (m)
Figure 3 Example of a subjective trench log. Stratigraphic units and the fault are outlined, but no details of deposit sedimentology or
sedimentary structures are portrayed. Unit numbers were changed for this book to represent time gaps associated with erosional unconformities
(between units 3 and 10) and soil horizons (units 11 A/11 Cca). The number gap between units 11 and 20 represents the time required to
develop the soil horizons. Reproduced from Jackson, M., 1991. The number and timing of Holocene paleoseismic events on the Nephi and Levan
segments, Wasatch fault zone. Utah Special Study – Utah Geological Mineral Survey 78, 1–23, with permission from Utah Geological Mineral
Survey.

disadvantage of this approach is that objective logs may not be to take a small number of big photographs that cover rela-
readily interpretable as drawn; even an expert may have to tively large areas of the trench wall (e.g., 3-m high by 2-m
study the relations in a log for a time before deducing an wide), whereas many others prefer to take smaller photo-
interpretation. graphs (0.5 m  0.5 m to 1 m  1 m) in greater numbers
In practice, a judicious combination of the subjective and (Figure 5). The smaller the photos the less rectification is
objective approaches produces trench logs that are sufficiently required for each photo, but the more photos must be
detailed to act as archival records, yet have sufficient inter- mosaicked. The choice is ultimately a matter of personal
pretational emphasis that the major elements can be separated preference.
at a glance from the minor elements. One easy way to make an
objective/subjective log is to make a photomosaic of the
trench wall and add annotations to it (described later).
14.12.4.4.2 Photogrammetric logging (3-D)
14.12.4.4 Techniques of Trench Logging The photogrammetric method is the application of terrestrial
photogrammetry to trench logging, and may be required for
Over the past 40 years, trench logging techniques have evol- very large, irregular fault exposures that cannot be accessed
ved from simple sketching on graph paper, to using more everywhere. The method was developed in the late 1980s by
sophisticated digital equipment (Table 3). Nevertheless, all Fairer et al. (1989) and Coe et al. (1991), using analytical
geomorphologists should be able to make a trench log using stereoplotters. Twenty-five years later the method now utilizes
the old, manual method. The current state-of-the-art method 3-D digital technology, but many principles remain the same.
is photomosaic logging. The overlapping photographs are analyzed in commercial
3-D photogrammetric software (e.g., Sirovision; JointMetrix).
14.12.4.4.1 Photomosaic logging (2-D) These programs were designed for mapping open-pit mine
The 2-D photomosaic method became the standard for walls (typically 25-m high and inaccessible) and the structures
research-grade studies around a decade ago. There are two exposed on them, particularly faults and joints. Photogram-
end-member approaches for taking wall photographs, the metric logging has many advantages in smaller trenches also
‘big-photo’ and ‘small-photo’ approaches. The author prefers and may soon replace 2-D logging.
144 Trenching and Exposed Faces

East
3

5 B3
B2 G
B1j
B1h West
B1i F
6
B1g
B1f
F
Distance (m)

B1c D4
B1b
7
D1b D4
B1a Crack
Shear
8 C
C
F1 Shear
Crack D1a
FZ
B4 C
S3 D1b
9 OFPC RC2
F2
N 18° E 66°−74° NW
Main fault zone
10 B5
N 32° E 77° NW C a
D1

11

21 22 23 24 25 26 27 28 29 30 31
Distance (m)
Figure 4 Example of an objective trench log with added linework emphasizing stratigraphic and structural features. Note how the accurate
representation of clasts defines sedimentary structures within mappable units, and shear fabric in the main fault zone (in box). Reproduced from
Olig, S.S., Lund, W.R., Black, B.D., 1994. Large mid-Holocene and late Pleistocene earthquakes on the Oquirrh fault zone, Utah. Geomorphol. 10,
285–315.

14.12.5 Applications of Trenching in the investigator wishes to observe. Trenches across small
Geomorphology scarps created by deformation of very young Quaternary sur-
faces may expose evidence of only one or two deformation
Trenching in geomorphic studies generally has one of two events. Trenches across higher scarps on progressively older
targets: (1) to expose and characterize a structure (fault, surfaces will expose the cumulative deformation from many
landslide plane, and sinkhole margin), in order to assess its events, but the effects of earlier displacements may be ob-
past movement history and future hazard (structural targets), scured by those of later displacements.
or (2) to expose a Quaternary deposit in section, in order to Workers must determine if unconformities on the trench
assess its stratigraphy, sedimentology, weathering profile, or wall were created by paleoearthquakes, or by some other
to collect samples for dating (stratigraphic targets). type of deformational, erosional, or depositional event. If the
characteristics of an unconformity unambiguously indicate an
earthquake origin, then it is interpreted as an earthquake
horizon, the horizon is bracketed by geochronology samples,
14.12.5.1 Structural Targets
and its resulting age is assigned to a paleoearthquake, hope-
14.12.5.1.1 Active fault studies (paleoseismology) fully associated with a mapped fault.
The ultimate goal of paleoseismic trenching is to expose de- Since 1970, the field of paleoseismology has grown con-
formation related to Quaternary faulting as well as to identify siderably, due to government regulations requiring fault-
and date unconformities in the stratigraphic sequence. Trench trenching studies for building or retrofitting various critical
location is also dictated by the number of displacement events structures (dams, power plants, pipelines, etc.). Since 1990,
Table 3 Comparison of the four methods of trench logging. Adapted from McCalpin (2009)

Method How to do it Equipment Advantages Disadvantages

needed

Manual One person measures the horizontal and vertical Tape measure, (1) It is inexpensive and does not rely on (1) The log does not contain an actual image of
distances from trench features to the nearest gridded paper, electronic equipment; all paleoseismologists the trench wall, but is an interpretive
grid line, a second person plots the position of pencil should be able to log a trench manually; (2) (subjective) drawing; (2) the hard copy log has
these control points on the trench log. one person can do the logging if necessary, to be digitized for publication or distribution;
Contacts are then drawn by hand on the log, although two should be used for safety; (3) the (3) if the reference grid lines lie at some
by connecting the control points with lines that detail of the log is limited only by the time and distance from the logged wall, there are
mimic the natural, irregular nature of contacts artistic ability of the logger; (4) it is easy to possible parallax errors in the drawing; (4)
on the trench wall. make revisions by erasing and redrawing; (5) considerable time is consumed by measuring
the log is drawn as the trench is traversed, and hundreds of points by hand.
is in nearly final form when the end of the
trench is reached.
Electronic Survey the positions of stratigraphic contacts Total station (1) No string line reference grid is needed. (2) (1) Equipment is expensive and fragile; (2)
and structures on the trench wall with an survey instru- faster and more accurate the manual method; plotting programs connect control points with
electronic surveying instruments. One person ment, software (3) trench logs can be printed at any scale; (4) straight line segments, so many more control
holds the EDM reflector against the trench wall surveyed points can be projected onto a points must be measured on curving or
at unit contacts, while the instrument operator vertical plane from irregular or nonvertical irregular contacts than in the manual method,
surveys the angle and distance to that point trench walls; (5) surveyor can enter unique for lines to appear realistic; (3) need to have a
(however, many newer EDMs do not need codes for each feature type (fault, stratigraphic printer or plotter in the field so that a hardcopy
reflectors). contact) for use in plotting trench log. log can be printed for on-site use.
Photomosaic, 2D Mark all lithologic, soil, and structural contacts Digital camera, (1) The photomosaic is the ultimate objective (1) Depending on the skill of the mosaicker,
on the wall clearly enough to be seen on a commercial log; (2) lines (geologic contacts, structures) there can be spatial errors of centimeters to
photograph. Photograph the entire trench wall photo-editing and areas (stratigraphic units, shear zones) decimeters in the trench log; (2) if individual
with a digital camera. Mosaic the photographs. software can be added by the logger as vector overlays mosaic photos are not tied/rectified to a grid,
To ease mosaicking, photographs should be in graphics software; these define the spatial errors will accumulate from the
taken perpendicular to the trench wall, with subjective part of the log; (3) the objective beginning to end of the mosaic; (3) in poor
frame edges aligned parallel to the reference (raster) and subjective (vector) layers can be lighting conditions, most contacts on the wall
grid; this minimizes scale differences and stored and viewed separately or together in may not be apparent in the photographs, so
parallax within each individual photograph. graphics software, making interpretation more the photomosaic backdrop adds little to the

Trenching and Exposed Faces

Each photo must overlap adjacent photos efficient; (4) the photomosaic is an archival vector line log.
slightly to avoid gaps in coverage. record of the appearance of the trench wall,
unaffected by anyone’s interpretation.
Photogrammetric, Mark all lithologic, soil, and structural contacts Digital camera, (1) Its ability to capture and visualize the (1) Cost of the specialized software (e.g.,
3D on the wall clearly enough to be seen on a specialized exposed face in true-scale, georeferenced 3D; Sirovision, Jointmetrix); (2) learning curve to
photograph. At least four surveyed control photogram- (2) very high spatial accuracy; (3) ability to master its capabilities.
points (with x, y, and z coordinates) must lie metric software measure the strike and dip of any planar
within each stereo pair of photographs and feature visible; (4) can statistically analyze
must be visible on the photos. The trench wall populations of planar structures such as faults
must be photographed from a constant and joints.
distance (c. 10–100 m) such that adjacent
photographs overlap by 60%.

Source: Adapted from McCalpin, J.P. (Ed.), 2009. Paleoseismology, Second ed. Elsevier Publishing, Amsterdam, 805 pp.

145
146 Trenching and Exposed Faces

15

U500
Fill

U500

U450

Swirls
U400
14
E1
U350 BSC3

U300

Shore
U250
00A22
00A21
E2
U200
00A10

U150 BSC2
00A66
13

BSC1
gc 00A74 gc
E3
gc
U130 00A39
gc
E4 gc
U110 Lower
gc gc
U100
U95
U90

U80
12

No
photo
Detail,
image
south wall
trench 00A

m
East 7 6 5 West

11
Figure 5 Example of a trench log made with the 2-D photomosaic method. Individual photographs in the mosaic cover a 0.5-m by 0.5-m area.
Trench units (labeled) are highlighted by thin lines, and faults (such as at right center) are highlighted by slightly thicker lines. Selected units can
be further emphasized by semi-transparent color fills (see color version in McCalpin, 2009). Log is from the Hayward fault, a dextral fault east of
the San Andreas fault in northern California. Reproduced from Lienkaemper, J.J., Dawson, T.E., Personius, S.F., Seitz, G.G., Reidy, L.M.,
Schwartz, D.P., 2002. A record of large earthquakes on the southern Hayward fault for the past 500 years. Bulletin of the Seismological Society
of America 92(7), 2637–2658, with permission from Bulletin of the Seismological Society of America.
 Trenching and Exposed Faces 147

trenching has been expanded to locating and analyzing widespread dating of episodes of landslide movement, to de-
earthquake-associated deformation caused by strong ground termine how those compare with periods of climatic
shaking, for example, liquefaction, sand blows, clastic dikes, fluctuation.
earthquake-triggered lateral spreads, and other types of land-
slides. Commonly, these deformation features lie far from any
mapped fault and, although their past deformation episodes 14.12.5.1.3 Sackung (deep-seated gravitational
can be dated, those cannot be positively associated with a spreading) studies
known fault. McCalpin (2009) provided an in-depth de- Sackungen are closed depressions and upslope-facing scarps
scription of trenching related to paleoseismology. found high on mountain ridges, the result of deep-seated
gravitational spreading of rocks at the ridge crest due to
downslope extrusion of rock masses lower on the ridge. Prior
14.12.5.1.2 Landslide studies to 1980, the published literature on sackungen was based
Only recently has trenching been applied to landslide studies solely on surface observations, but in the 1980s and 1990s
to supplement the more traditional methods of obtaining paleoseismic-style trenching was begun on sackung troughs
subsurface information (drilling and geophysics). As pointed and their adjacent antislope scarps. A pioneering study
out by Cotton (1999), the structures that define the head, was performed by McCalpin and Irvine (1995) in Colorado,
flanks, and toe of a landslide are essentially identical to where they demonstrated that a swarm of sackung troughs at
the normal, strike-slip, and reverse faults (respectively), which the ridge crest had begun forming thousands of years after the
have long been the subject of paleoseismic-trenching studies. steep valley wall below them was debuttressed by melting of
In the past few years, landslide workers such as Gutierrez the valley glacier. Using paleoseismic techniques, and inter-
et al. (2010) have demonstrated the advantages of trench- pretive models such as that shown in Figure 6, they concluded
ing for answering previously unanswered questions about that the trough formed by aseismic creep and that the creep
the past behavior of landslides (Figure 6). Perhaps in the rate was high early in its development and had slowed down
future landslide-trenching investigations will permit more in the late Holocene.

(a) (b)

(c) (d)

Figure 6 Theoretical stratigraphic arrangements in the ‘fault-angle depression’ developed at the head of a slump (diagram at center), each
recording a different kinematic behavior. (a) A single episode of rapid rotational slumping, followed by horizontal sediment infill; (b) continuous
creep and rotation over a long tie period, with continuous sedimentary infill; (c) two discrete episodes of rapid rotational slumping, each followed
by a period of horizontal sediment infill, the dashed line is an angular unconformity; (d) two episodes of creep movement, either of different
velocity, or separated by a hiatus (dashed line). Reproduced from Gutierrez, F., Lucha, P., Galve, J.P., 2010. Reconstructing the geochronological
evolution of large landslides by means of the trenching technique in the Yesa Reservoir (Spanish Pyrenees). Geomorphology 124(3–4), 124–136.
148 Trenching and Exposed Faces

Gradual subsidence Episodic subsidence

(c)

(a)

(d)

(b) (e)

Figure 7 Five types of subsurface relationships found on the edges of sinkholes. (a) Cumulative wedge-out at the margin of a sagging sinkhole
recording progressive synsedimentary subsidence. (b) Conformable and monotonous sequence with no evidence of sedimentary breaks
juxtaposed to a creeping fault at the margin of a collapse sinkhole. (c) Two stacked colluvial wedges associated to the edge of a sinkhole
recording two collapse and scarp rejuvenation events. The dashed lines indicate the event horizons. (d) Two overlapped fissure fills associated to
two unconformable sedimentary packages recording two subsidence events that involve vertical displacement and tilting of the hanging wall. (e)
Two sedimentary units bounded by an angular unconformity that truncates a fault in the lower sedimentary package. These geometrical
relationships can be explained by a minimum of two subsidence events; the one which created the accommodation space in which the lower unit
was deposited and a subsequent one that deformed the lower unit and created the depression where the younger unit was accumulated.
Reproduced from Gutierrez, F., Galve, J.-P., Lucha, P., Casteneda, C., Bonachea, J., Guerrero, J., 2011. Integrating geomorphological mapping,
trenching, InSAR and GPR for the identification and characterization of sinkholes in the mantled evaporite karst of the Ebro Valley (NE Spain).
Geomorphology, doi:10.1016/j.geomorph.2011.01.018.

14.12.5.1.4 Sinkhole studies historic and prehistoric floods (e.g., Alcoverro et al., 1999).
Karst sinkholes are another landform that forms episodically Where main stream paleofloods had deposited fine-grained
along subsurface structures, and thus yields valuable data ‘slackwater deposits’ in tributaries, these were trenched to look
when trenched. These data include: (1) confirming or ruling out for unconformities (indicating multiple floods and their crest
anomalies in the geophysical profiles (e.g., ground-penetrating heights), and to collect dating samples for paleoflood chron-
radar) attributable to subsidence; (2) precisely locating the edge ologies. Alluvial fans have been trenched to expose the num-
of some filled sinkholes; (3) obtaining information on sub- ber of debris flow deposits at various parts of the fan, their
sidence mechanisms recorded by various deformation structures average thickness (proxy for flow depth), and, to date their
and cumulative subsidence magnitude; and (4) calculating recurrence interval (e.g., Giraud, 2005).
minimum long-term subsidence rates using radiocarbon dates
obtained from deformed sinkhole deposits. Trenching sink-
holes to reconstruct their past behavior is in its infancy, but
preliminary results are promising. For example, Gutierrez et al. 14.12.6 Summary
(2009, 2011) have shown that the most-favorable sinkhole types
for trenching are depressions generated by sagging and collapse Trenching, the creation of an artificial exposure into the geo-
mechanisms, with the greatest information occurring at sink- logical subsurface, was originally developed for applied geo-
hole margins (Figure 7). Once the trench has been logged, the morphological and geohazard studies such paleoseismology
log is retrodeformed in order to reconstruct the chronological and landslide hazards. Trenches expose cleaner stratigraphic
sequence of deformation and sedimentary episodes. and structural sections than are available in natural exposures.
As a result, increased detail can be observed on cleaned trench
walls, permitting an enhanced identification of short-lived
geologic events (e.g., faulting, folding, erosion, deposition,
14.12.5.2 Stratigraphic Targets
and soil formation). In the past decade, trenching techniques,
Early stratigraphic trenching of fluvial deposits aimed to in- originally developed for paleoseismic studies, have been
vestigate river processes and flood hazards. For example, river applied to a growing spectrum of geoscience investigations.
floodplains and low terraces were trenched to measure the size This includes stratigraphic studies related to the effects of cli-
of boulders, and thus compare flood competence between mate change, and of human impact, on Earth-surface systems.
 Trenching and Exposed Faces 149

In the future, trenching may find a place in most types of Frost, E.G., Baker, E.D., Lindvall, S.C., Smilo, S.A., Rockwell, T.K., 1991. Computer
geological and geomorphological investigations of terrestrial enhanced imaging of neotectonic features exposed within trench faces:
techniques for seeing the unseen being developed on the Rose Canyon fault
deposits.
zone. In: Abbott, P.L. (Ed.), Environmental Perils, San Diego Region. Annual
Meeting of the Geological Society of America. San Diego Association of
Geologists, San Diego, CA, pp. 47–60.
References Giraud, R.E., 2005. Guidelines for the geologic evaluation of debris-flow hazards
on alluvial fans in Utah. Utah Geological Survey, Miscellaneous Publication
05–6, 16.
Alcoverro, J., Corominas, J., Gomez, M., 1999. The Barranco de Aras of
Gutierrez, F., Galve, J.P., Lucha, P., Bonachea, J., Jorda, L., Jorda, R., 2009.
7 August 1996 (Biescas, Central Pyrenees, Spain). Engineering Geology 51(4),
Investigation of a large collapse sinkhole affecting a multi-storey building by
237–255.
means of geophysics and the trenching technique (Zaragoza city, NE Spain).
Audemard, A., 2005. Paleoseismology in Venezuela; objectives, methods,
Environmental Geology 58(5), 1107–1122.
applications, limitations, and perspectives. Tectonophysics 408(1–4), 29–61.
Gutierrez, F., Galve, J.-P., Lucha, P., Casteneda, C., Bonachea, J., Guerrero, J.,
Birkeland, P.W., 1999. Soils and Geomorphology. Oxford University Press, Oxford,
2011. Integrating geomorphological mapping, trenching, InSAR and GPR for the
448 pp.
identification and characterization of sinkholes in the mantled evaporite karst of
Birkeland, P.W., Machette, M.N., Haller, K.M., 1991. Soils as a tool for applied
the Ebro Valley (NE Spain). Geomorphology, doi:10.1016/
Quaternary geology. Utah Geological Survey, Miscellaneous Publication 91–3,
j.geomorph.2011.01.018.
1–63.
Gutierrez, F., Lucha, P., Galve, J.P., 2010. Reconstructing the geochronological
Carver, G.A., Bemis, S.P., Solie, D.N., Obermiller, K.E., 2008. Active and potentially
evolution of large landslides by means of the trenching technique in the Yesa
active faults in or near the Alaska Highway corridor, Delta Junction to Dot Lake,
Reservoir (Spanish Pyrenees). Geomorphology 124(3–4), 124–136.
Alaska. Preliminary Interpretive Report 2008-3d, Alaska Division of Geological
Hatheway, A.W., Leighton, F.B., 1979. Exploratory trenching. Geological Society of
and Geophysical Surveys, Fairbanks, AK, 32 pp.
America Reviews in Engineering Geology 4, 169–195.
Coe, J.A., Taylor, E.M., Schilling, S.P., 1991. Close-range geophotogrammetric
Jackson, M., 1991. The number and timing of Holocene paleoseismic events on the
mapping of trench walls using multi-model stereo restitution software.
Nephi and Levan segments, Wasatch fault zone. Utah Special Study – Utah
ACSM-ASPRS Annual Convention. Primary Data Acquisition 5, 30–43.
Geological Mineral Survey 78, 1–23.
Cotton, W.R., 1999. Faults aren’t always what they’re cracked up to be. In: Hanson,
McCalpin, J.P. (Ed.), 2009. Paleoseismology, Second ed. Elsevier Publishing,
K.L., Kelson, K.I., Angell, M.A., Lettis, W.R. (Eds.), Techniques for Identifying
Amsterdam, 805 pp.
Faults and Determining Their Origins. U.S. Nuclear Regulatory Commission,
McCalpin, J.P., Irvine, J.R., 1995. Sackungen at Aspen Highlands Ski Area, Pitkin
Washington, DC, pp. A-29–A-50 (contract report NUREG/CR-5503, Appendix A).
County, Colorado. Environmental and Engineering Geoscience 1, 277–290.
Fairer, G.M., Whitney, J.W., Coe, J.A., 1989. A close-range photogrammetric
Shlemon, R.J., 1985. Application of soil stratigraphic techniques to engineering
technique for mapping neotectonic features in trenches. Bulletin of the
geology. Bulletin of the Association of Engineering Geologists 22(2), 129–142.
Association of Engineering Geologists 26, 521–530.

Biographical Sketch

Dr. McCalpin received his degrees in geology (BA, University of Texas, 1972; MS, University of Colorado, 1975;
PhD, Colorado School of Mines, 1981) and taught geology at Utah State University from 1982 to 1991. In 1991,
he founded GEO-HAZ Consulting, a consultancy in geohazard mapping and mitigation. His book Paleo-
seismology (1996; Second edition, 2009) has become the standard reference for investigations of active faults.
 14.13 Working with Gravel and Boulders
EA Felton, The Australian National University, Canberra, ACT, Australia
r 2013 Elsevier Inc. All rights reserved.

14.13.1 Introduction 150

14.13.2 Background 151
14.13.3 Methodology 152
14.13.3.1 The CAF Method 154
14.13.3.2 The Sedimentary FMB Method 155
14.13.4 Problems, Pitfalls, and Limitations 156
14.13.5 Case Studies 159
14.13.6 Future Work and Direction 162
14.13.7 Conclusions 162
References 162

Glossary Sedimentary structure Any structure in a sedimentary

Bed geometry The three-dimensional shape of a bed. rock. Sedimentary structures may be primary, that is,
Tabular (two dimensions much larger than the third) and developed during deposition, such as cross-bedding, or
lenticular (lens-shaped) are common bed geometries. secondary, formed after deposition, such as animal
Fabric In sediments and sedimentary rocks, the burrows.
orientation in space of the component particles. Size mode When the amounts of particles of different
Framework In a gravel, the framework comprises the sizes in a sediment are plotted on a frequency diagram, one
larger particles, which are often in mutual contact (see or more size classes usually predominate. This size class is
‘matrix’). termed the ‘mode’ of the sediment.
Lithofacies The rock record of a sedimentary Texture In sediments, texture comprises the collective
environment. attributes or characteristics of the component particles,
Littoral zone The area on or near a shoreline. including size, shape, and arrangement relative to one
Matrix In a rock, in which certain particles are much larger another.
than others, the particles of smaller size comprise the matrix Turbidite A sediment deposited by a turbidity current,
(see ‘framework’). that is, a dense, sediment-laden water flow.
Petrofacies The composition of the components of a
sediment body defines its petrofacies, and is related to the
provenance, or sources, of these components.

Abstract

Two complementary descriptive methods serve to systematize and integrate the wide range of bed and clast attributes that
characterize complex deposits of coarse gravel. The composition and fabric method has a materials focus, providing
detailed descriptions of attributes of the constituent clasts, their petrology, the proportions of gravel, sand, and mud, and
the ways in which these materials are organized. The method of sedimentary facies model building emphasizes the
organization of a deposit on a bed-by-bed basis so as to identify facies and infer depositional processes. Use of a com-
prehensive gravel fabric and petrography log ensures systematic collection of data. Vertical profiles provide visual repre-
sentations of deposit characteristics and enable correlations.

14.13.1 Introduction they consist of a wide range of clast sizes, from pebbles to
Coarse gravelly sedimentary deposits are more complex to blocks and even larger, but constituent clasts may be derived
describe and characterize than finer sediments. Not only do from diverse sources that include bedrock, soil, or older gravel
deposits. Pore spaces between the gravel framework may be
empty, contain secondary cementing minerals, or be filled
Felton, E.A., 2013. Working with gravel and boulders. In: Shroder, J. (Editor
in chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise on Geomorphology.
with finer sediment, including sand and mud. The attributes
Academic Press, San Diego, CA, vol. 14, Methods in Geomorphology, of coarse gravelly sediments reflect the sources and transport
pp. 150–163. history of the materials present (clast attributes), the mode of

150 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00380-8

 Working with Gravel and Boulders 151

deposition (bed attributes), and their subsequent modifi- size data, together with surface features such as percussion
cation (bed and clast attributes). marks, provide a starting point for constraining transport
How best to describe such complex mixtures? The de- options.
scriptive methods detailed here were devised to meet the re-
quirements of the study of a diverse suite of enigmatic gravels
and boulders of unknown origin, the Hulopoe Gravel, 14.13.2 Background
on the lower slopes of Lanai, Hawaiian Islands (Felton, 2002).
The methods are objective and make no assumptions about In the last 200 years, description of gravels and boulders has
the origins of the deposits, for which a number of competing developed from observations recorded in careful and detailed
hypotheses have been proposed (see below). They are termed written accounts (see Box 1) to today’s measurements and
the methods of ‘composition and fabric’ (CAF) and ‘sedi- characterization using sophisticated instruments.
mentary facies model building’ (FMB). Both methods aim to Throughout the twentieth century, emphasis has been
describe and characterize gravelly sediments on a bed-by-bed placed on developing and testing methods for describing the
basis, but have different objectives. The CAF method has a various attributes of sediment, to standardize data collection.
materials focus, providing detailed descriptions of a wide Attributes comprise both the characteristics of individual
range of attributes of the constituent clasts, their petrology, particles such as composition, rounding, shape, or surface
the proportions of gravel, sand, and mud, and the ways in features, and the characteristics of the bed or deposit (color,
which these materials are organized. It is the basis for classi- fabric, sorting, sedimentary structure, inclination of clasts).
fying each bed in a stratigraphic sequence comprising gravels. Some of these attributes can be assigned numerical values in
In modern deposits, it is the basis for description of a sedi- order to apply statistical tests that measure the variability of
mentary unit such as the surface layer of a gravel beach. The each attribute.
FMB method emphasizes the organization and geomorphic In earth sciences/geomorphological usage, gravel is a par-
context of a deposit to identify facies and infer depositional ticle-size term. The commonly used Udden–Wentworth grain-
processes. size scale, developed by the 1930s, defined particle sizes from
Both methods rely on comprehensive and systematic col- clay to boulder. The scale establishes the size grades of gravel,
lection of clast and bed attribute data, using a gravel fabric and from granules (2–4 mm) through pebbles, cobbles, and
petrography log (GFPL) developed during the Hulopoe Gravel boulders (256 mm to 4.1 m; see Table 1). Recognizing that
study. These data are integrated in graphic logs (vertical pro- boulder sizes in natural environments on Earth can range well
files) to provide visual representation of each bed in each beyond 4.1 m, Blair and McPherson (1999) proposed an ex-
stratigraphic section. A series of data-rich graphic logs facili- panded size scale to include megagravel, with clast sizes ran-
tates identification of facies and enables detailed correlations ging from 4.1 m up to 1075 km (Table 1).
along and across depositional strike. Together, the CAF and The Udden–Wentworth scale is a logarithmic scale in
FMB methods provide for well-supported interpretations which each size grade limit is twice as large as the next smaller
about the sources of the materials present, and their transport grade limit, resulting in a simple geometric progression of
and depositional histories. For specific problems, such as sizes. Krumbein (1934) devised the phi (f) scale using the
transport mechanisms of individual boulders, the shape and logarithm to the base 2 of the grain diameter instead of simply

Box 1 Early descriptions of gravels and boulders

Two early studies show how careful observations and written records describing the attributes of gravel clasts, the character of deposits, and the geomorphic setting
lead to insightful interpretations of sedimentary processes.

• Charles Darwin observed the sizes, shapes, roundness, and geomorphic settings of boulders in the coastal zone of southern South America during
the ‘Beagle’ voyage 1831–36, deducing two modes of ice transport of coastal boulders based on their roundness. He wrote: ‘‘Masses of floating ice
are produced in two ways. Boulders in floating sea ice show evidence of being worked on a beach and are somewhat rounded, but boulders
embedded in icebergs are always angular, and have been incorporated in glacial ice prior to its breaking off to form icebergs.’’ (Darwin, 1842).
• Amadeus William Grabau (1913) described a characteristic size zonation of littoral sediment with increasing distance from the shoreline.
He recognized three classes of gravel associated with rocky shores, that he called the rocky cliff facies, the bouldery facies, and the gravel facies. He
distinguished these facies by the size and angularity of gravel detritus, and described the processes of abrasion by waves or ice that result in the
distinctive character, or facies, of the associated gravel. Much detail of associated organisms, the hardness of source materials and geomorphic
details of the bedrock round out his detailed descriptions and support his interpretations.

References

Darwin, C., 1842. XXVII. On the Distribution of the Erratic Boulders and on the Contemporaneous Unstratified Deposits of South America. Transactions of the
Geological Society of London 1842; ser. 2; 6(2), 415–431.
Grabau, A.W., 1913. Principles of Stratigraphy. A. G. Seiler & Co., New York, 1185 pp.
http://www.archive.org/stream/principlesofstra00grab#page/n5/mode/2up.
152 Working with Gravel and Boulders

Table 1 Modified Udden–Wentworth particle size scale for coarse gravel

Particle length (intermediate axis) Grade Class Fraction

km M mm f Unlithified Lithified

1075  30 Very coarse Megalith Megagravel Megaconglomerate

538  29 Coarse
269  28 Medium
134  27 Fine
67.2  26 Very fine
33.6  25 Very coarse Monolith
16.8  24 Coarse
8.4  23 Medium
4.2  22 Fine
2.1  21 Very fine
1.0 1048.6  20 Very coarse Slab
0.5 524.3  19 Coarse
0.26 262.1  18 Medium
131.1  17 Fine
65.5  16 Very coarse Block
32.8  15 Coarse
16.4  14 Medium
8.2  13 Fine
4.1 4096  12 Very coarse Boulder Gravel Conglomerate
2.0 2048  11 Coarse
1.0 1024  10 Medium
0.5 512 9 Fine
0.25 256 8 Coarse Cobble
128 7 Fine
64 6 Very coarse Pebble
32 5 Coarse
16 4 Medium
8 3 Fine
4 2 Granule

Source: Reproduced from Blair, T.C., McPherson, J.G., 1999. Grain-size and textural classification of coarse sedimentary particles. Journal of Sedimentary Research 69(1), 6–19.

the diameter itself (Table 1). For example, 4 mm is 22, 32 mm to sedimentary models developed strongly. Facies, in gravel, is
is 25: phi ¼  log2 diameter (mm). The negative sign avoided ‘‘the ‘aspect’ of a [gravelly] deposit [that] includes lithological,
negative numbers for the finer size grades that were the focus textural and organic aspects detectable in the field’’ (de Raaf
of most early studies of sediment. For all gravel size grades, et al., 1965: p. 13). Facies models that embody general de-
however, negative numbers apply (see Table 1). Using the phi scriptions of the nature and sequence of coarse gravelly sedi-
scale simplifies statistical computations. ments were devised for coarse alluvial deposits: that is, braided
Folk (1954, 1965) and Folk et al. (1970) developed and re- stream gravelly alluvium and alluvial fans, glacial deposits,
fined a sediment textural classification scheme to describe mix- volcaniclastic deposits, and turbidites as well as associated
tures of gravel, sand, and mud (Figure 1(a)). Objective and coarse gravel deposits that form on marine basin slopes (see
practical, the Folk classification uses a texture triangle with gravel, chapters on these topics in Walker, 1984). General and specific
sand, and mud as end members, subdivided into textural classes. models for sedimentation in each geomorphic setting were
These are the basis of descriptive names for the sediment. depicted visually in vertical profiles (stratigraphic sequence
The Folk classification is simple to use and apply in the field logs) that showed typical facies successions. These models
(see Box 2). It is also adaptable to description of mixtures of served as guides and predictors of processes and the resulting
sediment with a high proportion of gravel from most geo- sedimentary products, for example, a specific sequence of
morphological settings. Blair and McPherson (1999), needing sediments (see Box 3).
better representation of coarse gravelly sediment while working
with alluvial gravel deposits, modified the Folk classification to
increase the number of gravel and muddy gravel subdivisions.
Felton and colleagues (see case study, below) used a boulder: 14.13.3 Methodology
cobble:pebble texture triangle (Figure 1(b)) to characterize a
diverse suite of sand-poor coarse gravel and boulder deposits, To describe complex mixtures of gravel comprehensively, a
the Hulopoe Gravel. much greater range of bed and clast attributes and parameters
General models for sedimentation based on systematic must be documented than for finer sediments; Table 2 is a
description began to appear in the 1970s. The ‘facies’ approach comprehensive list of these parameters. Measuring and
 Working with Gravel and Boulders 153

Key
Gravel
(100%) G gravel
sG sandy gravel
msG muddy sandy gravel
G mG muddy gravel
gS gravelly sand
80% gmS gravelly muddy sand
gM gravelly mud
(g)S slightly gravelly sand
(g)mS slightly gravelly

l
ve
muddy sand

g ra
(g)M slightly gravelly mud
nt
rce mG msG S sand
mS muddy sand
Pe

sG sM sandy mud
30% M mud

gM gmS gS
5%
(g)M (g)mS (g)S
Trace
M sM ms S
Mud 9:1 1:1 1:9 Sand
(100%) (100%)
Mud : sand ratio
(a)

Key
Boulder BG boulder gravel
(100%) (p)BG slightly pebbly boulder gravel
(c)BG slightly cobbly boulder gravel
BG pBG pebbly boulder gravel
90
(p)BG (c)BG cBG cobbly boulder gravel
80 cpBG cobbly pebbly boulder gravel
pcBG pebbly cobbly boulder gravel
bPG bouldery pebble gravel
ers

bCG bouldery cobble gravel

uld

bcPG bouldery cobbly pebble gravel

bo

bpCG bouldery pebbly cobble gravel

nt

(b)PG slightly bouldery pebble gravel

rce

pcBG cpBG
(b)CG slightly bouldery cobble gravel
Pe

(b)cPG slightly bouldery cobbly pebble gravel

pBG cBG (b)pCG slightly bouldery pebbly cobble gravel
30

bPG bcPG bpCG bCG

5

(b)cPG (b)pCG
(b)PG (b)CG
Pebble 9:1 1:1 1:9 Cobble
(100%) (100%)
Pebble:cobble ratio
(b)

Figure 1 (a) Triangular diagram illustrating basic textural classes for gravel, sand, and mud (Folk et al., 1970). (b) Triangular diagram
illustrating basic textural classes for boulder:cobble:pebble mixtures with o20% fine sediment. Reproduced from Felton, E.A., 2002.
Sedimentology of Rocky Shorelines: 1. a review of the problem, with analytical methods, and insights gained from the Hulopoe Gravel and the
modern rocky shoreline of Lanai, Hawaii. Sedimentary Geology 152(3–4), 221–245.

estimating bed and clast attributes employs standard field each bed in each stratigraphic section. Key data are readily
techniques, as outlined in Box 4. extracted from the GFPL to use in graphical plots of com-
A GFPL (Figure 2) standardizes the large range of sedi- ponent characteristics, and to assist in formulating a descrip-
mentological data required to characterize complex gravel fa- tive name for each bed (see Box 2). Data collection using the
cies. This log also serves as an essential checklist for recording GFPL is a prerequisite to application of the CAF and FMB
data at the outcrop, with one log sheet being completed for methodology.
154 Working with Gravel and Boulders

Box 2 Key concepts: Sediment description and classification

Folk’s ‘Petrology of Sedimentary Rocks’ is a classic ‘back to basics’ reference manual for sediments and sedimentary rocks (Folk, 1968).The fundamental questions
‘what are sediments, and how are they studied?’ are clearly answered. The focus is on fine sediment (sand and mud); but the principles apply equally to working
with gravels and boulders. Folk’s classification of sediments, succinctly summarized, is treated in greater detail in journal articles (e.g., Folk, 1954).
The purpose of a sediment/sedimentary rock classification is to provide a form of shorthand that conveys a conception of the physical character of the material
(Folk et al., 1970). Folk’s general principles for classifying sediments and sedimentary rocks are objectivity and practicality. The Folk classification can be easily
determined in the field, with no need for laboratory analysis, and the sediment or rock given a name immediately. For these reasons, the Folk classification is widely
used.
Texture is the basis for the Folk classification of sediments and sedimentary rocks. Folk used the term texture to mean particle size distribution, that is, the
proportions of gravel, sand, and mud. The Folk classification is based on the weight per cent of gravel, sand, and mud (particle size terms) as three end members
depicted on a triangular diagram, the texture triangle (Figure 1(a) herein). Subdivisions within the triangle are based on the proportions of each end member, and
define basic textural classes. For example, muddy gravel applies to sediment with a composition between 30% and 80% gravel, 0–50% mud, and o50% sand.
Sediment description consists of the basic textural class, and may be qualified by other attributes of the deposit and its constituent particles.
Blair and McPherson (1999) adapted the Folk classification to accommodate a need for better representations of coarse gravelly sediment, adding further
subdivisions to the Folk texture triangle. They also proposed an expanded size scale beyond the Udden–Wentworth size limit of 4.1 m for boulders, to encompass
megagravel clasts up to 1075 km in size. Felton (2002) used a boulder: cobble: pebble texture triangle (Figure 1(b) herein) to classify sand-poor gravels.

References

Blair, T.C., McPherson, J.G., 1999. Grain-size and textural classification of coarse sedimentary particles. Journal of Sedimentary Research 69(1), 6–19.
Felton, E.A., 2002. Sedimentology of Rocky Shorelines: 1. A review of the problem, with analytical methods, and insights gained from the Hulopoe Gravel and the
modern rocky shoreline of Lanaia, Hawaii. Sedimentary Geology 152(3-4), 221–245.
Folk, R.L., 1954. The distinction between grain size and mineral composition in sedimentary-rock nomenclature. Journal of Geology 62, 344–359.
Folk, R.L., 1968. Petrology of Sedimentary Rocks. Hemphill Publishing Company, Austin, Texas, 170 pp.
Folk, R.L., Andrews, P.B., Lewis, D.W., 1970. Detrital sedimentary rock classification and nomenclature for use in New Zealand. New Zealand Journal of Geology and
Geophysics 13, 937–968.

Box 3 Key concepts: Facies and facies models

Application of the facies concept to sedimentary environments became a strong research focus in the 1970s. Facies is defined as ‘‘the sum total of the lithological
and palaeontological aspects of a stratigraphic unit’’ (Walker, 1984, p 1); ‘‘lithological structural and organic aspects detectable in the field’’ (de Raaf, 1965). Facies,
or aspect, distinguishes one body of sediment, or part of it, from another.
Walker (1975) recognized distinctive features, including repeated facies sequences, in turbidite deposits on continental submarine slopes. These features
formed the basis of a facies model for deep marine turbidites. Facies models, second edition (Walker and James, 1992), contains chapters describing sedimentary
features and models for gravelly sedimentary environments, including alluvial, volcaniclastic, glacial, and marine slope.
In his general introduction to Facies models, Walker (1984) provided guidelines and caveats concerning the formulation and purpose of facies models:
‘‘Subdivision of a sediment body into constituent facies is essentially a classification procedure and the degree of subdivision must . . . be governed by the
objectives of the study’’ (p. 1). A facies model is a summary of a specific sedimentary environment, based on many studies of ancient rocks and recent sediments
(p. 5). A facies model must fulfill four important functions: (1) act as a ‘norm’, for purposes of comparison; (2) act as a ‘framework’ and ‘guide’ for future
observations; (3) act as a ‘predictor’ in new geological situations; (4) act as an integrated ‘basis for interpretation’ of the environment or system that it represents.

References

Walker, R.G., 1975. Generalized facies models for resedimented conglomerates of turbidite association. Geological Society of America, Bulletin 86, 737–748.
Walker, R.G. (Ed.), 1984. Facies Models. Second ed. Geoscience Canada Reprint Series 1, 315 pp.
Walker, R.G., James, N.P. (Eds.), 1992. Facies Models. Response to Sea-Level Change. Geological Association of Canada, Toronto, ON, 409 pp.

14.13.3.1 The CAF Method separates the attributes of the deposit that are related to its
most recent transport and depositional processes from those
In the CAF method, as developed for the case study described attributes that pertain to the nature of the material from which
below, the basic textural class of each bed is determined using the deposit is derived. Some bed attributes, such as bio-
an appropriate texture triangle such as those in Figures 1(a) turbation, rooting, and cementation, may be due to post-
and 1(b); see also Box 2. This is placed first in the name, so as depositional modification. Color, which should include a
to establish the fundamental nature of the sediment. Other Munsell chart color (Munsells, 1992), is placed first in the
descriptors distinguished as bed attributes or clast attributes descriptive string, followed by attributes due to postdeposi-
follow the basic textural class, with bed attributes preceding tional modification, then other bed attributes, then clast at-
clast attributes in the description (see notes c, d, Table 2). This tributes. This information can be summarized using letter
 Working with Gravel and Boulders 155

Table 2 Letter symbols for bed and clast attributes used to describe Table 2 Continued
coarse gravelly sediment
Sorting
Basic textural class (upper case)a ws well sorted
GG granule gravel ms moderately sorted
PG pebble gravel ps poorly sorted
CG cobble gravel us unsorted
BG boulder gravel
KG block gravel Roundness (in italics)
b wr well rounded
Adjectival qualifier (lower case)
sr sub-rounded
m muddy
sa sub-angular
s sandy
an angular
g granular
p pebbly a
Reproduced from Felton, E.A., 2002. Sedimentology of Rocky Shorelines: 1. a review
c cobbly of the problem, with analytical methods, and insights gained from the Hulopoe Gravel
b bouldery and the modern rocky shoreline of Lanai, Hawaii. Sedimentary Geology 152(3–4),
k blocky 221–245.
b
Bed attributesc When adjectival qualifier is bracketed, signifies that quality is present to only a small
Color extent.
c
Postdepositional modification Bed attributes follow the textural class in the order shown, with letter symbols for
Biological geometry underlined and letter symbols for fabric in italics. For example, slightly sandy
bt bioturbation pebbly cobble gravel (s)pCG, pebbly cobble gravel, thick sheet, weakly organised,
bu burrows inversely graded pCG sh tk (o)v.
d
rt roots Clast attributes follow bed attributes, from which they are separated by a vertical bar.
For example, pebbly cobble gravel, thick sheet, weakly organised, inversely graded;
Cement clasts compact, poorly sorted, subangular pCG sh tk (o)v9c pssa.
cm cemented e
Abbreviations for combinations of these three basic shapes can be used: For
un uncemented example, ce, compact-elongate; vp, very platy; etc. (see Sneed and Folk, 1958).

Geometry
wd wedge
bp parallel bands
bn anastomosing bands symbols (Table 2), for example, bouldery pebbly cobble
sh sheet/tabular gravel, rooted, reddish (Munsell color), (carbonate) cemented,
cl cluster sheet, plane bedded, clast supported, inversely graded, closely
ic isolated clast packed; cobbles compact (some platy), moderately sorted,
Bed thickness subangular to subrounded: bpCG rt 5YR/6 cm sh h lt v c 9 c(p)
tk thick ms sa-sr. A vertical bar separates bed and clast attributes in the
tn thin letter symbol.
Sedimentary structure
h plane bedded
w wavy bedded
14.13.3.2 The Sedimentary FMB Method
Organization
o organised The purpose in building a facies model for the deposit is to
(o) weakly organised focus on the distinctive features of the deposit in order to
d disorganised begin interpretation of those features. A facies model can also
be compared with facies models developed for coarse gravelly
Fabric (in italics)
lt clast support deposits in a range of geomorphologic settings to assist with
m matrix support interpretation (see Walker, 1984; Walker and James, 1992 for
i imbricate clasts chapters on a number of settings).
a aligned clasts The range of attributes associated with coarse gravel mix-
v inversely graded tures could potentially result in a very large number of facies
n normally graded being derived from deposit descriptions. The key task is to
Packing identify attributes that best distinguish and characterize con-
l loosely packed stituent gravel facies, by isolating the particular features – bed
c closely packed and clast attributes – that define the essential nature or dis-
tinctiveness of each bed (see Box 4 for steps to take, and the
Clast attributesd
case study below for a discussion on how this was done for the
Shape
c compacte Hulopoe Gravel).
e elongate The vertical profile is the starting point for building a facies
p platy model for the gravel deposit. Bed-by-bed characteristics are
b bladed represented visually on vertical profiles (stratigraphic logs)
(Continued ) that depict clast compositions (petrology) and the range of
156 Working with Gravel and Boulders

Box 4 Step by step from description to interpretation

In the field:
1. Use the gravel fabric and petrography log (GFPL) to systematically record data from each bed in the deposit or stratigraphic section. For clast sizes,
use measuring tapes, rulers, and/or calipers. Comparative charts are used to determine clast shape and roundness, and bed sorting. A compass or
inclinometer measures bedding dip and inclination of clasts.
TIP: a sorting comparator, devised for sands, works for mixed gravels: from a few metres away, hold the comparator up and compare visually with the
outcrop.
Composition and fabric (CAF) – from the GFPL
2. Determine the ‘basic textural class’ of each bed (the ‘Folk descriptor’). Use an appropriate ‘texture triangle’ (e.g., see Figures 1(a) and 1(b)).
3. Add ‘descriptors’ (bed and clast attributes; see Table 2). Record the full descriptive name.
4. Assign shorthand ‘letter symbols’ to each description (Table 2).

Facies model building (FMB) – using vertical profiles

5. Use the letter symbols and descriptive data to select those bed and clast attributes that identify discrete lithofacies and petrofacies.
6. List lithofacies and petrofacies using their distinguishing features, grouping them by textural class (e.g., all boulder lithofacies).
7. Draw ‘vertical profiles’ for each location to illustrate the stratigraphic succession of lithofacies. Use a computer drawing program or draw by hand.
TIP: the vertical profile should, where possible, be constructed in the field. Use a template incorporating common elements of the vertical profile, such as
the labelled columns. Use the template as a starting point for each new profile.
8. Add as much other data as possible to the vertical profile: from field data for bed relationships, highlight any breaks in succession (unconformities);
add age data. Record the numbers of photographs taken for each bed. Aim for the most complete visual representation of data.
9. Examine the vertical profiles to identify common elements or patterns in the succession, such as repeated lithofacies sequences. These are the basis
of a facies model for the deposit (see Box 3).
TIP: to help identify common elements, assign a number to each lithofacies (1, 2, ...) and a letter to each unconformity (a, b, c, ...) on the vertical profiles
and look for repeated patterns. Start with the longest profile, likely to be the most complete stratigraphic sequence, and do this exercise for each vertical
profile.

10. Correlate unconformities and lithofacies along and across depositional strike. Be alert for gaps in deposition.
11. Use the data depicted in the correlated vertical profiles to make interpretations about gravel sources, transport and depositional processes.

gravel bed textures (Figures 3 and 4). Descriptive shorthand rounded boulders is not consistent with the current wave
letter symbols highlight bed-by-bed changes. Other data such regime.
as age determinations are added as they come to hand.
The vertical profile should be a complete visual representation
of data pertaining to each bed. 14.13.4 Problems, Pitfalls, and Limitations
In studying the successions of lithofacies and petrofacies for
identifiable patterns, if unconformities are present, one begins Recording the complete range of data from outcrops of com-
with the facies successions above each unconformity. This is plex gravel and boulders requires extended time to be spent in
done for each vertical profile. A recurring or consistent lithofacies the field. Using photo-interpretation to estimate attributes for
succession is the basis for a facies model for the deposit. This the GFPL should be done with caution, and field-checked for
model can be used to make interpretations about gravel sources, accuracy where possible. Differential cementation in outcrops
and transport and depositional processes. These are clues to the can obscure details in a photograph. Dark shadows cast by
depositional environments for the deposit. strong lighting can mimic particular rock types in photo-
Finally, the vertical profiles can be correlated along and graphs. This can be problematic either for vertical sections in
across depositional strike, starting with unconformities and terrestrial outcrops or for the bed surfaces of submarine
going on to match up lithofacies/lithofacies successions. These gravels when lit obliquely.
correlations will yield further insights into the depositional Clearly for deposits with a more limited range of attributes,
processes and depositional environments. such as a boulder beach with well-rounded clasts of uniform
The methodology described above was developed for very size, derived locally from a limited source rock, the description
diverse gravelly deposits of unknown origin: an extremely will be much simpler, with fewer boxes to be filled in on the
complex problem to resolve as any or all deposit character- GFPL. However, where working from a grid along and across
istics may be required. Most studies of coarse gravel are sim- an apparently uniform boulder beach, subtle variations will be
pler, because the deposits themselves are less complex. Many highlighted if the most complete range of data is collected
studies aim to characterize modern deposits where the de- from each gridline.
positional environment is known, but transport and de- The approaches and methods developed during the
positional processes are not clear. Commonly, for example, Hulopoe Gravel study described below and detailed above
the energy of transporting waves implied by the sizes of have yet to be widely accepted and tested; yet, there is
 Working with Gravel and Boulders 157

Logged by:
Gravel fabric and petrography log
(version 1.1)
date:
(Format prepared by Anne Felton and Keith Crook)

Graphic Log #: Unit #: Log basis: outcrop/ photo/ video/.....................

Orientation of face: bedding normal / bedding \\ (viewed from above/below) / other..............................

Face: transects clasts / does not transect clasts
Framework: clast supported / matrix supported / matrix type:
Grainsize: Max. (5 largest) Minimum:
(b-axis) 1: 2: 3: 4: 5:

Size Modes: I: II: III:

Estimated %: gravel: sand: mud: Folk descriptor:
Estimated %: Mode I popn: Mode II popn: Mode III popn: Mode IV popn:

Sorting: Overall: Mode I popn: Mode II popn: Mode III popn: Mode IV popn:
(comparator)
Determined by:
Clast types: Rock type 1 Rock type 2 Rock type 3 Rock type 4 Rock type 5 # count /pt. count
/visual estimate
Name: /other.
Surface textures:
weathered/dull/
Mode I %: % % % % % polished/striae/
percussion marks/
sarface text: smooth/rough/not
visible/other
Special shapes: flat
special shape irons/semilunate/
dreikanter/other
Determined by:
Clast types: Rock type 1 Rock type 2 Rock type 3 Rock type 4 Rock type 5 # count /pt. count
/visual estimate
Name: /other.
Surface textures:
weathered/dull/
Mode II %: % % % % % polished/striae/
percussion marks/
sarface text: smooth/rough/not
visible/other
Special shapes: flat
special shape irons/semilunate/
dreikanter/other
Determined by:
Clast types: Rock type 1 Rock type 2 Rock type 3 Rock type 4 Rock type 5 # count /pt. count
/visual estimate
Name:
/other.
Surface textures:
weathered/dull/
Mode III %: % % % % % polished/striae/
percussion marks/
sarface text: smooth/rough/not
visible/other
Special shapes: flat
special shape irons/semilunate/
dreikanter/other
Determined by:
Clast types: Rock type 1 Rock type 2 Rock type 3 Rock type 4 Rock type 5 # count /pt. count
Name: /visual estimate
/other.
Surface textures:
weathered/dull/
Mode IV %: % % % % % polished/striae/
percussion marks/
sarface text: smooth/rough/not
visible/other
Special shapes: flat
special shape irons/semilunate/
dreikanter/other
Clast Sand: mud:
compositions

Figure 2 Gravel fabric and petrography log. Reproduced from Felton, E.A., 2002. Sedimentology of Rocky Shorelines: 1. a review of the
problem, with analytical methods, and insights gained from the Hulopoe Gravel and the modern rocky shoreline of Lanai, Hawaii. Sedimentary
Geology 152(3–4), 221–245.
158 Working with Gravel and Boulders

Clast Rock type 1: Rock type 2: Rock type 3: Rock type 4: Rock type 5:
Roundness
Mode I ρ: ρ: ρ: ρ: ρ:
Clast Rock type 1: Rock type 2: Rock type 3: Rock type 4: Rock type 5:
Roundness
Mode II ρ: ρ: ρ: ρ: ρ:
Clast Rock type 1: Rock type 2: Rock type 3: Rock type 4: Rock type 5:
Roundness
Mode III ρ: ρ: ρ: ρ: ρ:
Clast Rock type 1: Rock type 2: Rock type 3: Rock type 4: Rock type 5:
Roundness
Mode ρ: ρ: ρ: ρ: ρ:

Clast Sand:
roundness ρ:
Clast shape Rock type 1: Rock type 2: Rock type 3: Rock type 4: Rock type 5:
(measured) a: a: a: a: a:
Mode Ι: b: b: b: b: b:
c: c: c: c: c:
Clast shape Rock type 1: Rock type 2: Rock type 3: Rock type 4: Rock type 5:
(measured) a: a: a: a: a:
Mode II: b: b: b: b: b:
c: c: c: c: c:
Clast shape Rock type 1: Rock type 2: Rock type 3: Rock type 4: Rock type 5:
(measured) a: a: a: a: a:
Mode III: b: b: b: b: b:
c: c: c: c: c:
Clast shape Rock type 1: Rock type 2: Rock type 3: Rock type 4: Rock type 5:
(measured) a: a: a: a: a:
Mode : b: b: b: b: b:
c: c: c: c: c:
Clast shape Rock type 1: Rock type 2: Rock type 3: Rock type 4: Rock type 5:
(comparator)
Mode I:
Clast shape Rock type 1: Rock type 2: Rock type 3: Rock type 4: Rockt ype 5:
(comparator)
Mode II:
Clast shape Rock type 1: Rock type 2: Rock type 3: Rock type 4: Rock type 5:
(comparator)
Mode III:
Clast shape Rock type 1: Rock type 2: Rock type 3: Rock type 4: Rock type 5:
(comparator)
Mode :
Clast shape
(comparator)
Sand:

Apposition Fabric (bedding normal/parallel): imbricate (∠= °) / edgewise (∠= °) / random a-axes /
a-axes \\ lamination / other :
Vector Fabric 1: orientation: ; based on:
Vector Fabric 2: orientation: ; based on:

Figure 2 Continued
 Working with Gravel and Boulders 159

Petrofacles

Texture
Unit #
Bed #
Ist:ba
Bcl |cb ps wr
750 1 16
Bic |cb ps wr

700
H 15 CG w lt (i ) v|cb ms sr
650

600 14 sgCG w lt|cb ms sr

550 13 bpCG lt v(i ) |cb (p) ps sr

500 G

12
450 cPG lt |cb ps sr

11 bCG lt |cb ps sr
400
10 sPG lt |cb ms sr
9 sCG lt |cb ps sr
350 F

8
300 bCG lt (i )v|cb ms sr-sa

D 7 sPG (h) lt v|cb ps sa

250 PG lt |cb ps sa
6
5 (s)PG lt v|cb ws sr
200 C 4 (s)cPG (h) lt v|cb ps sr
3 Bic (s)cPG (h) lt v|c-cb ms sa-sr
150 2
pcBG lt |c ps sr
100

Figure 3 Example of vertical profile used for facies model building. Reproduced from Felton, E.A., Crook, K.A.W., Keating, B.H., Kay, E.A., 2006.
Sedimentology of rocky shorelines: 4. coarse gravel lithofacies, molluscan biofacies, and the stratigraphic and eustatic records in the type area of
the Pleistocene Hulopoe Gravel, Lanai, Hawaii. Sedimentary Geology 184(1–2), 1–76.

particular need to apply rigorous descriptive methods to other Comprehensive descriptions of a range of coastal gravels are
modern and ancient deposits of coarse gravel, boulders, and needed from which to develop such models. Such compara-
blocks in the coastal zone. In the 1990s, boulder and gravel tive studies, of modern and ancient environments, exemplify
deposits on rocky shorelines were ascribed to tsunami with the effectiveness of the descriptive approach taken to new,
equivocal data presented in support of this interpretation. unstudied sedimentary environments, as described by Walker
Methods of distinguishing between boulders deposited by (1975) for modern marine turbidites. From these detailed
storms, tsunami, or other processes had yet to be developed. descriptions, well-supported inferences can be made con-
For example, a 1990s series of papers by Robert Young, cerning depositional processes.
Edward Bryant, and colleagues at the University of Wollon-
gong, brought together in successive editions of Tsunami – the
Underrated Hazard (Bryant, 2001, 2008), ascribe boulder de-
posits and geomorphic features on rock coasts in SE Australia 14.13.5 Case Studies
to one or more very large tsunami. Tsunami impacts on the SE
Australian coast were a novel suggestion at the time. Sub- The two related papers by Felton (2002) and Felton et al.
sequent work has raised doubts about the tsunami interpret- (2006) describe and demonstrate approaches to character-
ation of some of these deposits and features (e.g., Felton and izing, representing, and interpreting a complex and diverse
Crook, 2003; Switzer and Burston, 2010). However, Young’s suite of gravel/conglomerate deposits of unknown origin, the
and Bryant’s pioneering work, together with several large Pleistocene Hulopoe Gravel, on Lanai, one of the Hawaiian
tsunami worldwide post-2000, has stimulated interest in both Islands (Figure 5).
tsunami deposits (Spiske et al., 2010; Goto et al., 2010) and in This study aimed to distinguish between competing hy-
high-energy processes on rock coasts that generate coarse potheses, and clarify the origin of the Hulopoe Gravel. Hy-
gravel (Hall et al., 2008; Stephenson and Naylor, 2010). potheses included elevated beach gravels, one or more
However, adequate methods have still to be developed to deposits, resulting from catastrophic events (Shiki, 1996) such
differentiate the tsunami and storm-wave origin of many as tsunami or hurricane, abandoned rocky shorelines, Native
coastal boulder deposits (see Goto et al., 2010 for historical Hawaiian house sites and gardens for parts of the deposit, and,
review). most notably, deposition by one or more mega-tsunami
Given the wave hazards that affect coastlines, it is essential upslope of the island’s modern shoreline (Box 5). Each hy-
to interpret deposits correctly, especially where undertaking pothesis suggested very different relative sea-level histories and
coastal risk assessments. Facies models for the littoral zone, processes on Lanai’s south shore. The deposits resulting from
especially of rocky shorelines, are needed to assist in this task. different processes should also appear very different from one
160 Working with Gravel and Boulders

mud sand gravel

Petrofacies
pbl bld

Scale (cm)
granule cbl

Unit #
Texture
Bed #
φ +4 0 −4 −8

Structures Ist:ba
C

4 Contact sharp, planar

200 Contact sharp, irregular.

3 Heavy line signifies
B disconformable contact
150 Clast size of largest mode;
increases upwards
2 A
100

Size modes of
1 bimodal beds
50

Bed contact obscure or

approximate
0
Texture column Structures

Mud; sand Parallel lamination

proportions indicated

Pebble gravel/ Rhizoliths, peds, caliche

conglomerate

Cobble gravel/ Coral, mollusk, bioclasts

conglomerate

Boulder gravel/ Petrofacies column

conglomerate

Large boulder gravel/ Limestone, basalt

conglomerate with proportions indicated

Basalt

Figure 4 Key to Figure 3. Reproduced from Felton, E.A., 2002. Sedimentology of Rocky Shorelines: 1. a review of the problem, with analytical
methods, and insights gained from the Hulopoe Gravel and the modern rocky shoreline of Lanai, Hawaii. Sedimentary Geology 152(3–4),
221–245.

another. For example, a tsunami deposit upslope from a Clearly, a detailed sedimentological study was needed to
gravelly shoreline involves a few very large waves over a few help resolve some competing hypotheses. Uncertainty about
hours, whereas a deposit from a hurricane in the same upslope which gravel and boulder particle or bed parameters might
setting is the product of thousands of waves over several days. yield critical information about sediment sources and trans-
Theoretically, the hurricane deposit should be better sorted port and depositional processes required collection of com-
and the constituent gravel particles better rounded than a prehensive and detailed gravel data. A GFPL (Figure 2 herein)
tsunami deposit. was developed for the project, and was used in the field to
No comprehensive description and analysis had been record a full range of sedimentary data from each bed in each
undertaken prior to the study by Felton and colleagues. stratigraphic section. These data included both clast and bed
Resolution of the deposits’ origin (or multiple origins) bears attributes, the basis of the CAF and FMB approaches to deposit
not only on the depositional processes proposed, but also on description and interpretation (see Box 2).
the relative sea-level history, including issues concerning the The Hulopoe Gravel consists entirely of gravel facies. Four-
tectonic and isostatic (uplift and subsidence) history of Lanai teen coarse gravel lithofacies – eight boulder and six pebble/
and adjacent Hawaiian hot-spot oceanic islands. cobble lithofacies – were distinguished using a selection of bed
 Working with Gravel and Boulders 161

? ? ?

N
17
G N

G
14
70 25
G
?

60
19
G
50

1
40
?G
? 21
G ?G
13

30
?G
9 ?
12
G G
11 ? G
3
27

20 10

G
G
2 26
2 G

4 G G
5 10 Kapihaa Bay
G 20 3 1
?G
G
Hulopoe Bay
E
6 7 G

4
Pebble/cobble 2 Stratigraphic log location
conglomerate (unit G) and number
Boulders, facies 1 Gully number
N B3 (unit H)
Modern cobble and boulder
Boulders, facies
B4 and B5 (unit H) pocket beach/bar
Eroded edge of basalt flow
Basalt bedrock
0 250 m Top of palaeocliff

Figure 5 Geological map of Hulopoe Bay area, south Lanai, Hawaii, showing locations of logged stratigraphic sections. Base air photo is from
image posted at http://www.coralreef.gov/mapping/. Box in center right modified from Moore and Moore (1988). Reproduced from Felton, E.A.,
Crook, K.A.W., Keating, B.H., Kay, E.A., 2006. Sedimentology of rocky shorelines: 4. coarse gravel lithofacies, molluscan biofacies, and the
stratigraphic and eustatic records in the type area of the Pleistocene Hulopoe Gravel, Lanai, Hawaii. Sedimentary Geology 184(1–2), 1–76.

Box 5 Competing hypotheses for the origins of the Hulopoe Gravel on Lanai, Hawaii

Hypothesis Proponent Basis

Abandoned shorelines Stearns (1938) Similar deposits found earlier on adjacent islands and ascribed to
eustatic sea-level changes
‘Giant wave’ deposits during Pleistocene Moore and Moore (1984, Large submarine landslides identified during mapping Hawaiian
1988); McMurtry et al. Islands’ slopes
(2004)
Native Hawaiian house sites and gardens Kaschko and Athens (1987) Archaeological survey
(parts of deposit)
Uplifted shoreline deposits Grigg and Jones (1997); Isostatic adjustments of Lanai due to lithospheric flexure
Keating and Helsley (2002)
Multiple depositional events (?tsunami) Rubin et al. (2000) Isotopic dating of coral clasts in deposits; dates range from 250
to 130 kyr.
Normal deposition in a high-energy, Felton et al. (2000; 2006); Sedimentological and geomorphological comparison of deposits
rocky shoreline sedimentary Felton (2002) with modern sedimentary environment of south Lanai coast
environment
Mega-tsunami deposits Webster et al. (2007) Geochronology of drowned coral reefs on Lanai
162 Working with Gravel and Boulders

and clast attributes. Size modes of framework clasts together potential for preserving subtidal boulder fields during sea-
with the amount of matrix present provided an initial broad level changes?
subdivision. Sedimentary structures and fabric best dis- Key questions are: What are the preconditions of the bed-
tinguished the different pebble/cobble facies, whereas bed rock for boulders to be generated, detached, and transported
geometry was the most distinctive feature differentiating the in the rocky shoreline environment? Do coastal boulder de-
various boulder lithofacies. Clast roundness distinguished posits, above and below the shoreline, reflect the actions of
subfacies in two of the boulder facies. both tsunami and storm waves during wave events that are
Vertical profiles constructed to depict the lithofacies pre- separated in time? Some of the distinguishing criteria for
sent incorporated petrofacies (bed composition), bed contact tsunami versus hurricane boulders may well be teased from a
(short or long breaks in deposition between successive comprehensive data set collected using the GFPL.
beds), and age-dating details (e.g., Figure 3). This visual rep- Application of the CAF and FMB methodology in studies of
resentation of data enabled detailed correlation parallel and gravels around coastlines identified on high-resolution im-
normal to depositional strike over the 2 km2 study area. agery of planets and moons is an exciting prospect. Some of
Examination of the vertical profiles of several of the thicker the possible coastlines imaged on Mars should be rocky
stratigraphic sections revealed multiple repetitions of similar shorelines with associated coarse gravel and boulder deposits
stratigraphic successions separated by long depositional that should be visible on the high-resolution imagery now
breaks (unconformities). Not only did these occurrences in- available. The methodology is fully adaptable for use with
dicate cyclicity in sediment deposition, but also enabled a imagery. Using the GFPL will ensure that the widest possible
facies model to be developed for the deposit. range of data is collected.
Felton et al. (2006) interpreted the deposits as rocky For the many boulder and megagravel deposits appearing in
shoreline littoral deposits, a poorly known sedimentary en- high-resolution imagery on extraterrestrial bodies, comprehen-
vironment. An earlier paper (Felton, 2002) included obser- sive data collection will be needed to support interpretations of
vations of geomorphic features and gravelly sediment processes that may differ from those on Earth (see Takeuchi et al.,
distribution on the modern rocky shoreline of south Lanai, 2010). To this end, a renewed interest in boulders and gravel
adjacent to the study area, with which to compare the Hulo- among planetary geologists has led to compilation of a photo-
poe Gravel. The repeated cyclic boulder and gravel deposition graphic atlas of rock breakdown features, using examples of
was interpreted as a response to relative sea-level rises and falls surface features of boulders on Earth (Bourke and Viles, 2007).
related to both sea-level eustatic highstands and the uplift of The atlas will assist planetary geologists in interpreting rock
Lanai. Felton et al. (2006) commented that (mega)tsunami surface features on extraterrestrial bodies.
deposits might form part of this depositional record, but they
were unable to identify parts of the Hulopoe Gravel that could
unequivocally be ascribed to mega-tsunami, either above
14.13.7 Conclusions
or below sea level, because of the lack of definitive sedi-
mentological and geomorphological criteria for identifying
Although the CAF and FMB methods were devised for a par-
mega-tsunami deposits.
ticularly complex deposit, they are applicable to any deposit of
The conjoint application of the CAF and FMB methods
coarse gravel and/or megagravel. The GFPL is comprehensive
provided a reliable basis for systematic study and interpret-
and designed for maximum utility in describing modern and
ation of these unusual gravel deposits. The study provided a
ancient gravels in a range of terrestrial and submarine settings.
methodology for detailed sedimentological and geomorphic
The GFPL is an essential checklist for ensuring that the max-
comparison of elevated gravel deposits on other tropical
imum available information is recorded.
oceanic volcanic islands that have also been attributed to high
tsunami run-ups. The study and methodology used demon-
strate the range of data that need to be collected and critically
assessed to resolve significant issues, not only for coastal References
gravels but also for coarse gravel and boulder deposits in other
Blair, T.C., McPherson, J.G., 1999. Grain-size and textural classification of coarse
geomorphic settings.
sedimentary particles. Journal of Sedimentary Research 69(1), 6–19.
Bourke, M.C., Viles, H.A. (Eds.), 2007. A Photographic Atlas of Rock Breakdown
Features in Geomorphic Environments. Planetary Science Institute, Tucson, AZ,
88 pp.
14.13.6 Future Work and Direction Bryant, E.A., 2001. Tsunami – The Underrated Hazard, First ed. Cambridge
University Press, Cambridge, 320 pp.
Bryant, E.A., 2008. Tsunami – The Underrated Hazard, Second ed. Springer Praxis
Felton (2002) stated: ‘‘rocky shorelines may well be the last Books, Chichester, 330 pp.
frontier of sedimentary environments for study on planet Felton, E.A., 2002. Sedimentology of Rocky Shorelines: 1. a review of the problem,
Earth.’’ Much work remains to be done in the subtidal zone of with analytical methods, and insights gained from the Hulopoe Gravel and the
modern rocky shorelines to describe and develop gravel facies modern rocky shoreline of Lanai, Hawaii. Sedimentary Geology 152(3–4),
models that will be the basis for interpretation of the range of 221–245.
Felton, E.A., Crook, K.A.W., 2003. Evaluating the impacts of huge waves on rocky
wave-related and other depositional processes that occur there,
shorelines: an essay review of the book ‘Tsunami, the Underrated Hazard’.
such as mass flows. For boulder deposits in supratidal settings, Marine Geology 197, 1–12.
there are still no adequate criteria for distinguishing tsunami- Felton, E.A., Crook, K.A.W., Keating, B.H., 2000. The Hulopoe Gravel, Lanai,
deposited from hurricane-deposited boulders. What of Hawaii: new sedimentological data and their bearing on the ‘‘giant wave’’ (mega-
 Working with Gravel and Boulders 163

tsunami) emplacement hypothesis. Pure and Applied Geophysics 157, Munsells, 1992. Munsells Soil Color Charts. Macbeths Division of Kollmorgen
1257–1284. Instruments, New York, NY, 9 charts.
Felton, E.A., Crook, K.A.W., Keating, B.H., Kay, E.A., 2006. Sedimentology of rocky de Raaf, J.F.M., Reading, H.G., Walker, R.G., 1965. Cyclic sedimentation in the
shorelines: 4. coarse gravel lithofacies, molluscan biofacies, and the Lower Westphalian of North Devon, England. Sedimentology 4, 1–52.
stratigraphic and eustatic records in the type area of the Pleistocene Hulopoe Rubin, K.H., Fletcher III, C.H., Sherman, C., 2000. Fossiliferous Lanai deposits
Gravel, Lanai, Hawaii. Sedimentary Geology 184(1–2), 1–76. formed by multiple events rather than a single giant tsunami. Nature 408,
Folk, R.L., 1954. The distinction between grain size and mineral composition in 675–681.
sedimentary-rock nomenclature. Journal of Geology 62, 344–359. Shiki, T., 1996. Reading of trigger records of sedimentary events – a problem for
Folk, R.L., 1965. Petrology of Sedimentary Rocks. Hemphill Publishing, Austin, TX, future studies. Sedimentary Geology 104, 244–255.
159 pp. Sneed, E.D., Folk, R.L., 1958. Pebbles in the Lower Colorado River, Texas, a study
Folk, R.L., Andrews, P.B., Lewis, D.W., 1970. Detrital sedimentary rock classification in particle morphogenesis. Journal of Geology 66, 114–150.
and nomenclature for use in New Zealand. New Zealand Journal of Geology and Spiske, M., Weiss, R., Bahlburg, H., Roskosch, J., Amijaya, H., 2010. The
Geophysics 13, 937–968. TsuSedMod inversion model applied to the deposits of the 2004 Sumatra and
Goto, K., Shinozaki, T., Minoura, K., Okada, K., Sugawara, D., Imamura, I., 2010. 2006 Java tsunami and implications for estimating flow parameters of palaeo-
Distribution of boulders at Miyara Bay of Ishigaki Island, Japan: a flow characteristic tsunami. Sedimentary Geology 224, 29–37.
indicator of the tsunamis and storm waves. Island Arc 19(3), 412–426. Stearns, H.T., 1938. Ancient shorelines on the island of Lanai, Hawaii. Geological
Grigg, R.W., Jones, A.T., 1997. Uplift caused by lithospheric flexure in the Society of America, Bulletin 49, 615–628.
Hawaiian Archipelago as revealed by elevated coral deposits. Marine Geology Stephenson, W.J., Naylor, L., 2011. Geological controls on boulder production
141, 11–25. in a rock coast setting: insights from South Wales, UK. Marine Geology 283,
Hall, A.M., Hansom, J.D., Jarvis, J., 2008. Processes, patterns and rates of erosion 12–24.
by storm waves on hard rock headlands: the grind of the Navir, Shetland, Switzer, A.D., Burston, J.M., 2010. Competing mechanisms for boulder deposition
Scotland. Marine Geology 253(1–2), 36–50. on the southeast Australian coast. Geomorphology 114, 42–54.
Kaschko, M.W., Athens, J.S., 1987. Archaeological inventory survey of the Hulopoe Takeuchi, H., Miyamoto, H., Maruyama, S., 2010. Origins of bright spots on the
Bay and Manele Bay areas, island of Lanai. Hawaii. Rport prepared for M & E surface of boulders covering Asteroid Itokawa. 41st Lunar and Planetary Science
Pacific, Inc., Honolulu, HI, 161 pp. Conference 1578.
Keating, B.H., Helsley, C.E., 2002. The ancient shorelines of Lanai, Hawaii, Walker, R.G., 1975. Generalized facies models for resedimented conglomerates of
revisited. Sedimentary Geology 150(1–2), 3–15. turbidite association. Geological Society of America, Bulletin 86, 737–748.
Krumbein, W.C., 1934. Size frequency distributions of sediments. Journal of Walker, R.G. (Ed.), 1984. Facies Models, Second ed. Geoscience Canada Reprint
Sedimentary Petrology 4, 64–77. Series, Toronto, ON, 1, 315 pp.
McMurtry, G.M., Fryer, G.J., Tappin, D.R., et al., 2004. Mega-tsunami deposits on Walker, R.G., James, N.P. (Eds.), 1992. Facies Models. Response to Sea-Level
Kohala volcano, Hawaii, from flank collapse of Mauna Loa. Geology 32, 741–744. Change. Geological Association of Canada, 409 pp.
Moore, G.W., Moore, J.G., 1988. Large scale bedforms in boulder gravel produced Webster, J.M., Clague, D.A., Braga, J.-C., 2007. Support for the Giant Wave (Mega-
by giant waves in Hawaii. Geological Society of America, Special Paper 229, Tsunami) Hypothesis: evidence from the submerged terraces off Lanai, Hawaii.
101–110. International Journal of Earth Sciences 96(3), 517–524.
Moore, J.G., Moore, G.W., 1984. Deposit from a giant wave on the island of Lanai,
Hawaii. Science 226, 101–110.

Biographical Sketch

Dr. E Anne Felton graduated BSc (Hons., Geology) from The Australian National University and has a PhD in
sedimentology from the University of Wollongong, Australia. In her early career, she worked for the Geological
Survey of New South Wales (Australia) and the Bureau of Mineral Resources, Geology, and Geophysics (now
Geoscience Australia).
In 1995, Anne joined the Hawaii Institute of Geophysics and Planetology, School of Ocean and Earth Science
and Technology at the University of Hawaii. Here, she developed a research interest in the sedimentology of rocky
shorelines, particularly sand-poor shorelines where gravelly sediments dominate the depositional system. This
research has expanded to include impacts of tsunami and storm waves and paleo-sea level records on rocky
shorelines.
Currently, a school visitor at the Research School of Earth Sciences at the Australian National University,
Canberra, Australia, Anne is continuing her rocky shoreline research, with a focus on the ages of rock coastal
landforms and deposits in SE Australia. She is active in community education about coastal erosion and sea-level
changes and serves on the Board of Directors of the Sapphire Coast Marine Discovery Centre in Eden, New South
Wales.
 14.14 The Micro and Traversing Erosion Meter
W Stephenson, University of Otago, Dunedin, New Zealand
r 2013 Elsevier Inc. All rights reserved.

14.14.1 Introduction 164

14.14.2 The Microerosion Meter 164
14.14.3 The Traversing Microerosion Meter 166
14.14.4 Rates of Erosion and Swelling 167
14.14.5 Comparisons with other Methods 168
14.14.6 Conclusions 168
References 168

Abstract

The microerosion meter and variant, the traversing microerosion meter, are described. This instrument is used for meas-
uring erosion on a millimeter scale of bare rock surfaces. The technique has also revealed that rock surfaces swell as well as
lower. The most widely used application has been to determine the rate at which shore platforms lower. Contributions to
understanding rates of erosion in noncoastal environments are fewer, and there is a scope for further use of the technique in
terrestrial environments.

14.14.1 Introduction 14.14.2 The Microerosion Meter

The microerosion meter (MEM) was originally invented by The microerosion meter consists of an equilateral triangular
Hanna (1966) as a technique for measuring rates of erosion base, with legs located at each corner and an engineering dial
on limestone in caves. High and Hanna (1970) introduced the gauge located on a central pillar (Figure 1). The spindle of the
method to the wider geomorphological community. The MEM gauge extends through the base plate. Readings are taken by
was later modified by Trudgill et al. (1981) to allow a greater placing the MEM on three bolts permanently fixed into a rock
number of individual measurements (37 rather than just 3) to surface (a bolt site) (Figure 2). These bolts are made from
be made. The modified meter is known as the traversing marine grade stainless steel so as to withstand environmental
microerosion meter (TMEM). Both the MEM and TMEM conditions, particularly the corrosive effects of salt. Using the
(collectively referred to below as T/MEM) have been widely MEM three readings can be taken on each bolt site by rotating
used to measure surface lowering rates on shore platforms
(Robinson, 1976, 1977a, b, c) and to a lesser extent on
bedrock in terrestrial settings, particularly on limestone
(Stephenson and Finlayson, 2009). The use of the TMEM has
also revealed that rock surfaces not only undergo erosion but
also swelling where the rock surfaces rise when compared to
the previous measurements (Stephenson and Kirk, 2001;
Stephenson et al., 2004; Trenhaile, 2006; Gómez-Pujol et al.,
2007). In more recent times, the advent of laser scanning
technologies has opened new ways to measure rock erosion,
although these are not always an improvement on the T/MEM,
Since the MEM measures erosion at the scale of microns to
millimeters, terrestrial laser scanning offers a way to overcome
scale problems associated with the T/MEM but this laser
scanning still needs to be integrated with the microscale to
allow for measurement across the complete range of scales
(microns to meters) at which erosion can occur.

Stephenson, W., 2013. The micro and traversing erosion meter. In: Shroder,
J. (Editor in Chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise on
Geomorphology. Academic Press, San Diego, CA, vol. 14, Methods in
Geomorphology, pp. 164–169. Figure 1 The microerosion meter located on a bolt site.

164 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00381-X

 The Micro and Traversing Erosion Meter 165

Figure 2 MEM bolt site consisting of three bolts permanently fixed Figure 3 Positioning template for locating the spirit level to check
into a rock surface. In this case the bolts are exposed above the that all three bolts are in the horizontal plane.
surface 32 years after they were installed by Gill and Lang (1983) on
the Victorian coast of southeastern Australia.
can be used if this method is preferred. This method of in-
stillation also requires an additional step where a larger hole
the instrument 1201 as the spindle of the dial gauge is off- has to be drilled over the expansion bolt hole to accommodate
center. The three-point readings can be averaged to gain a rate the washer of the expansion bolt. The washer holds the bolt at
of surface lowering for the site. Each time the instrument is the desired depth below the rock surface. Typically this larger
placed on the three bolts the exact same position is relocated. hole can be drilled with a brace and bit. The depth of this hole
Exact relocation is achieved through the Kelvin Clamp prin- determines the depth to which the bolt is recessed below the
ciple; the end of each leg is machined differently: one has a rock surface and allows for very precise positioning, but care
cone-shaped depression, one a v-notch depression, and the must be taken not to overdeepen the larger hole. If epoxy is
third is flat (High and Hanna, 1970). Each leg end opposes used then the depth of the bolt is set but pushing the stainless
movement in three, two, and one directions respectively; thus rode down to the required depth below the rock surface.
preventing movement of the plate when placed on a bolt site. Installation of a MEM bolt site requires a template to
The effectiveness of the Kelvin clamp is reduced if the tops of position the three holes correctly (the MEM itself can serve this
the three bolts are not on a horizontal plane. Some workers purpose), a quality masonry drill and one with a SDS (Special
have modified the original design by replacing the engineering Direct System) chuck is advisable, as well as a suitable epoxy.
dial gauge with a micrometer; however, the principles of op- Epoxy that sets in wet conditions are available and when bolts
eration remain the same (Cucchi et al., 1987; Drysdale and are being installed on coastal rock surfaces close to or in the
Gillieson, 1997; Allred, 2004). intertidal zone this can be a significant advantage. Stephenson
In early designs of the MEM, bolts consisted of a masonry and Thornton (2005) used bolts made from 8 mm threaded
expansion plug with a stainless steel bolt and washer with a 316 marine grade stainless steel rod set into 16 mm diameter
hollow machined into the top of the bolt into which a holes drilled twice the depth of the bolt. The length of the bolt
stainless steel ball bearing was fixed with epoxy. It soon be- is not necessarily important but a length exceeding 65 mm is
came apparent to early users that the ball bearings could de- recommended in cases where lowering rates are in excess
tach or be removed by vandals. A subsequent design used a 1 mm yr 1. Depending on the hardness of the rock being
single piece of threaded steel rod with the top machined to a drilled a portable generator may be required, but for softer
hemisphere and a masonry expansion bolt. A nut was added rocks modern battery powered drills can be used effectively.
to carry a washer to create the complete bolt. Gill and Lang Ideally the tops of the three bolts once installed should lie in
(1983) used a Portland lime mortar to fix bolts rather than the same horizontal plane to ensure the Kelvin Clamp works
expansion bolts. More recently, MEM bolts have been installed correctly. Gill and Lang (1983) used a spirit level attached to
using just the stainless steel rod and marine epoxy an MEM frame in order to check the level of all three bolts
(Stephenson and Thornton, 2005; Foote et al., 2006; Trenhaile (Figure 3). It is also possible to design a template for the
et al., 2006). One advantage in using epoxy rather than ma- positioning of the three bolts with adjustable legs and a spirit
sonry bolts is that a high degree of accuracy in the initial level to aid in positioning and leveling of the studs.
location of the three holes is not needed because the precise In order to maximize the length of time that bolts can be
position of the bolts can be adjusted using the MEM base read it is desirable to counter sink the bolts as deep as possible
before the epoxy hardens. Another advantage is that using into the rock surface. The depth to which the bolts are recessed
epoxy greatly speeds up instillation, which can be an issue below the surface depends on the design of the MEM, par-
where working on intertidal rock surfaces. The use of expan- ticularly the length of the legs and the length of travel in the
sion bolts, although still applicable, requires a high degree of probe of the dial gauge. Typical dial gauges have a precision of
precision in the placement of the three holes and a template 0.001 mm having 25 mm of travel. Thus when the MEM is
166 The Micro and Traversing Erosion Meter

placed on a set of bolts, initial readings of around 23–24 mm offer detailed discussion with regard to these issues. Partly this
are ideal and this means that 23–24 mm of vertical erosion is due to the unique nature of each study and different ques-
can occur before the site is redundant or extensions need to be tions that are being addressed so there are relatively few
added to the gauge probe. guidelines for future users of the T/MEM. Intuition seems to
A critically important component of operating a T/MEM is have been how people have addressed the issue of sampling
a calibration plate to check the accuracy of the instrument and representativeness of erosion rates. Mottershead (1989)
through time. For example, Stephenson and Kirk (1996) were and Stephenson and Kirk (1996) have considered the number
able to calibrate readings taken 20 years after the installation of individual readings required to gain representative erosion
of bolts sites on Kaikoura Peninsula, New Zealand. Several data, and in both cases 30 points (10 bolt sites) were deemed
designs have used steel or alloys with low thermal conductivity appropriate for the settings in which they worked. Stephenson
to construct plates that the MEM is placed on and the in- and Kirk (1996) and Stephenson et al. (2010) have demon-
strument can be checked regularly. Gill and Lang (1983) used strated that erosion data collected over a period of 2 years is
a polished glass plate and ball bearings of the same size as the not statistically different from data collected over 20 and
hemisphere of the installed studs. This light weight design can 30 years. Thus short-term studies over 2 years appear to be
be used in the field (Figure 4). representative of decadal time periods. It remains to be de-
Trudgill (1977) and Spate et al. (1985) have examined termined if MEM rates are representative of longer timescales.
sources of error and limitations of the MEM. Errors associated Careful consideration to the number of bolts sites and meas-
with thermal expansion, probe erosion, and operator error can urements and the length of time over which measurements
be reduced using equations devised by Spate et al. (1985). occur are required.
Regardless, a high degree of precision is possible with A final consideration is the long-term relocation and
Gómez-Pujol et al. (2007) claiming measurement error of reading of bolt sites. Several studies have been able to extend
70.005 mm. Errors are a concern with relatively short ob- the length of observation from typical periods of 1–3 years to
servation periods or where very slow rates of change are ob- decades because the bolt sites, MEM, and original data could
served. With increasing use of the TMEM at very short time be located. Longer term records of erosion measured with the
scales (hours) errors need to be considered (e.g., Gómez-Pujol T/MEM are desirable. Some guidelines for the establishment
et al., 2007; Hemmingsen et al., 2007). At the longest time of a study utilizing the T/MEM should ensure:
scales of measurement and where erosion rates are relatively
1. the locations of bolts are recorded using GPS,
high error becomes insignificant against measured erosion
2. the original data recorded from the instrument are kept
(e.g., Stephenson and Kirk, 1998; Furlani et al., 2010;
with the MEM
Stephenson et al., 2010).
3. the calibration plate is stored with the MEM to allow for
Another consideration when using the MEM is the number
later use.
and location of bolts sites, the number of readings, and the
period of time of observation that is needed to gain repre- Stephenson and Kirk (1996) and Stephenson et al. (2010)
sentative data of erosion rates. Relatively few studies appear to have successfully read bolt sites 20 and 30 years after sites were
established in 1973 because of the integrity of the instrument,
bolt location, and original data. Stephenson et al. (2011) have
been able to re-read 46 of the 50 bolt sites established by Gill
and Lang (1983) 31.8 years after the first readings, but they
reported that it took 10 years to locate all three components
(bolts, MEM, and original data).

14.14.3 The Traversing Microerosion Meter

Trudgill et al. (1981) first described the traversing microero-

sion meter (TMEM), a modified version of the MEM. The
TMEM differs in that the dial gauge is separated from the base
and is mounted on a block with three arms separated at 1201
intervals (Figure 5). The center of the base plate is cut out to
allow the dial gauge to be moved within the area defined by
the permanent bolts mounted on the rock surface. The dial
gauge can be moved to a number of different positions within
the area covered by the base, by locating each horizontal
arm between ball bearings fixed along the sides of the base
(Figure 5). So long as each arm is at right angles to a side of
the triangular base, a precise location is obtained each time the
Figure 4 Calibration plate being used in the field. The plate is made instrument is placed on a bolt site and the dial gauge is located
from polished glass with ball bearings, the same diameter as the correctly on the base. The number of locations that the dial
head of the bolts installed in the rock, placed under each leg. gauge can be moved to on the base depends on the size of the
 The Micro and Traversing Erosion Meter 167

Figure 5 The traversing microerosion meter. The dial gauge is

supported on three arms separated at 1201. So long as the arms rest
between ball bearings and are at right angles to the base, exact
relocation is achieved.

base, the number of ball bearings fixed along each edge, and Figure 6 The traversing microerosion meter connected to a laptop
for data collection in the field with foot switch to send data from the
on the size and configuration of the dial gauge and arms, but
dial gauge to a spreadsheet. A foot switch allows the operator to
up to 255 individual measurements within the area of the
control the position of the instrument and greatly speeds up data
three bolts possible (Neves et al., 2001). The TMEM can utilize capture.
the same bolt sites as the MEM if it has the same leg spacing.
Alternatively the TMEM can be used as an MEM only recording
range of values for erosion derived from various studies.
three readings. In the case where interest is only in the erosion
Typically erosion rates on shore platforms average 1.48 mm
rate, Stephenson and Kirk (1998) demonstrated that there was
yr 1 but a wide range of values occurs from almost zero up to
no statistical difference between erosion rates calculated from
70 mm yr 1 (Foote et al., 2006). Erosion rates at noncoastal
120 individual point measurements at a bolt site and a mean
sites range again from zero to 13 mm yr 1 and average
rate calculated from three individual point readings.
0.655 mm yr 1 (Stephenson and Finlayson 2009). A key point
The replacement of the analog gauge with a digital one on
made by Stephenson and Finlayson. (2009) is the wide range
the TMEM allowed readings to be recorded to a laptop computer
of values reported from various studies that means caution
via a cable in the field (Figure 6) (Stephenson, 1997), or
must be used when interpreting erosion rates represented by
this connection can be achieved wirelessly (Stephenson and
averages.
Finlayson, 2009). Data from the TMEM can be plotted as a
In addition to measured erosion rates the TMEM has been
digital elevation model using appropriate software, which allows
used to examine the phenomenon of rock swelling. Surfaces
visualization of the surface to aid the interpretation and iden-
that were expected to be lowering between measured intervals
tification of processes causing lowering and swelling. Surface
have been observed to rise. Early reports of surface swelling
plotting software also enables the volume of material eroded
occur in most MEM studies (but not all), Mottershead (1989)
from a bolt site to be estimated and these data can be used to
was the first to examine the phenomenon in detail but found
determine a sediment budget (e.g., Stephenson et al., 2010).
no correlation with climatic variables (temperature, precipi-
Furlani et al. (2010) developed a method where bolts can be
tation, and wind) that might have been expected. Stephenson
installed on a slab of rock that was placed vertically in the
and Kirk (2001) report swelling of limestone of 3–4 mm be-
intertidal region and then removed to allow readings to be
tween readings taken several months apart. In their study
taken. This particularly novel approach overcomes the limitation
swelling was correlated with season with greater swelling
of the T/MEM of only being placed on horizontal surfaces.
during summer months when wetting and drying was more
pronounced. Laboratory experiments investigating swelling
and contraction utilizing the TMEM have been made by
14.14.4 Rates of Erosion and Swelling Trenhaile (2006), Trenhaile et al. (2006), and Porter and
Trenhaile (2007). In those experiments different rock types
Stephenson and Finlayson (2009) provided a review of the were subjected to wetting and drying revealing that argillite
use of the T/MEM in geomorphology and reported on the underwent swelling whereas sandstone did not.
168 The Micro and Traversing Erosion Meter

14.14.5 Comparisons with other Methods Gómez-Pujol, L., Stephenson, W.J., Fornós, J.J., 2007. Short term surface change
on supra-tidal rock coast (Marengo, Victoria, Australia). Earth Surface Processes
and Landforms 32, 1–12.
Comparisons between the T/MEM and laser scanners have
Hanna, F.K. 1966. A technique for measuring the rate of erosion of cave passages.
been made for measuring erosion and mapping weathering Proceedings University of Bristol Speleology Society 11, 83–86.
morphologies (Williams et al., 2000, Swantesson et al., 2006). Hemmingsen, S.A., Eikaas, H.S., Hemmingsen, M.A., 2007. The influence of
Although laser scanners like those described by Swantesson seasonal and local weather conditions on rock surface changes on shore
et al. (2006) offer a variety of benefits, the T/MEM is more platforms, Kaikoura Peninsula, South Island, New Zealand. Geomorphology 87,
239–249.
versatile and robust. The T/MEM is inexpensive to construct, High, C.J., Hanna, F.K., 1970. A method for the direct measurement of erosion on
easy to operate, more portable (than a laser scanner), readings rock surfaces. British Geomorphological Research Group Technical Bulletin 5,
can be taken rapidly, bolt sites can be established in locations 1–25.
inaccessible to a laser scanner and the T/MEM provides more Kirk, R.M., 1977. Rates and Forms of Erosion on Intertidal Platforms at Kaikoura
Peninsula, South Island, New Zealand. New Zealand Journal of Geology and
precise results. It does not, however, provide as many indi-
Geophysics 20, 571–613.
vidual readings as a laser scanner per site. The MEM can Mottershead, D.N., 1989. Rates and patterns of bedrock denudation by coastal salt
provide a greater spatial coverage because more bolt sites can spray weathering: a seven-year record. Earth Surface Processes and Landforms
be established over a wider area. A limitation of the MEM 14, 383–398.
technique (and laser scanning) is that it does not capture Naylor, L.A., Stephenson, W.J., 2010. On the role of discontinuities in mediating
shore platform erosion. Geomorphology 114, 89–100.
erosion at larger scales associated with block removal (Kirk
Naylor, L.A., Stephenson, W.J., Trenhaile, A.S., 2010. Rock coast geomorphology:
1977; Stephenson and Kirk 1998, Dornbusch, et al., 2005; Recent advances and future research directions. Geomorphology 114, 3–11.
Rosser et al., 2005; Naylor and Stephenson, 2010). Techniques Neves, M., Pereira, A.R., Laranjeira, M., Trindade, J., 2001. Desenvolviment de um
such as photogrammetry, LiDAR, and terrestrial laser scanning microerosómetro do tipo TMEM para aplicac- ão em platformas rochosas litorais.
can be used to measure larger scale erosion. The measurement Finisterra 36, 83–101.
Porter, N.J., Trenhaile, A.S., 2007. Short-term rock surface expansion and
of erosion across all scales on bedrock therefore requires the contraction in the intertidal zone. Earth Surface Processes and Landforms 32,
use of a number of techniques, including the T/MEM. 1379–1397.
Robinson, L.A., 1976. The Micro-erosion meter technique in a littoral environments.
Marine Geology 22, M51–M58.
Robinson, L.A., 1977a. The Morphology and development of the north east
14.14.6 Conclusions Yorkshire shore platform. Marine Geology 23, 237–255.
Robinson, L.A., 1977b. Marine erosive processes at the cliff foot. Marine Geology
The MEM has been in wide use for 45 years and the TMEM for 23, 257–271.
35 years. They have provided useful data on the rates of rock Robinson, L.A., 1977c. Erosive processes on the shore platforms of Northeast
Yorkshire. England. Marine Geology 23, 339–361.
erosion, particularly on shore platforms. If it was not for the
Rosser, N.J., Petley, D.N., Lim, M., Dunning, S.A., Allison, R.J., 2005. Terrestrial
resurgence of interest in rock coasts (Naylor et al. 2010), and laser scanning for monitoring the process of hard rock coastal cliff
particularly shore platforms, the method may well have been erosion. Quarterly Journal of Engineering Geology and Hydrogeology 38,
consigned to history. The application of the method to ter- 363–375.
restrial-based rock erosion remains an open avenue for re- Spate, A.P., Jennings, J.N., Smith, D.I., Greenaway, M.A., 1985. The micro-erosion
meter: use and limitations. Earth Surface Processes and Landforms. 10,
searchers. The limited scale of measurement of the T/MEM 427–440.
means that it is best used with other techniques that capture Stephenson, W.J., 1997. Improving the traversing micro-erosion meter. Journal of
the range of scales at which erosion can occur. Coastal Research 13, 236–241.
Stephenson, W.J., Finlayson., B.L., 2009. Measuring erosion with the micro-erosion
meter – contributions to understanding landform evolution and building stone
decay. Earth Science Reviews 95, 53–62.
References Stephenson, W.J., Kirk, R.M., 1996. Measuring erosion rates using the micro-
erosion meter: 20 years of data from shore platforms, Kaikoura Peninsula, South
Allred, K., 2004. Some carbonate erosion rates of Southeast Alaska. Journal of Island New Zealand. Marine Geology 131, 209–218.
Cave and Karst Studies 66, 89–97. Stephenson, W.J., Kirk, R.M., 1998. Rates and patterns of erosion on shore
Cucchi, F., Forti, F., Finocchairo, F., 1987. Carbonate surface solution in the classic platforms, Kaikoura, South Island, New Zealand. Earth Surface Processes and
karst. International Journal of Speleology 16, 125–138. Landforms 23, 1071–1085.
Dornbusch, U., Moses, C. Robinson, D.A. Williams, R. 2005. Soft copy Stephenson, W.J., Kirk, R.M., 2001. Surface swelling of coastal bedrock on inter-
photogrammetry to measure shore platform erosion on decadal time scales. tidal shore platforms, Kaikoura Peninsula, South Island, New Zealand.
Geomorphology 41, 5–21.
Proceedings of CoastGIS 2005, 6th International Symposium on GIS and
Stephenson, W.J., Kirk, R.M., Hemmingsen, S.A., Hemmingsen, M.A., 2010.
Computer Cartography for Coastal Zone Management, 21–23 July 2005,
Decadal scale micro erosion rates on shore platforms. Geomorphology 114,
Aberdeen. pp. 72–84.
22–29.
Drysdale, R., Gillieson, D., 1997. Micro-erosion meter measurements of travertine
Stephenson, W.J., Kirk, R.M., Kennedy, D.M., Finlayson, B.L. and Chen, Z. 2011.
deposition rates: a case study from Louie Creek, Northwest Queensland,
Long term shore platform surface lowering rates: revisiting Gill and Lang 31
Australia. Earth Surface Processes and Landforms 22, 1037–1051. years later. Abstracts Australia New Zealand Geomorphology Group Conference
Foote, Y., Plessis, E., Robinson, D.A., Henaff, A., Costa, S., 2006. Rates and 1–4 February, 2011, Oamaru, New Zealand.
patterns of downwearing of chalk shore platforms of the Channel: comparisons Stephenson, W.J., Taylor, A.J., Hemmingsen, M.A., Tsujimoto, H., Kirk, R.M., 2004.
between France and England. Zeitschrift für Geomorphologie, Supplementbände Short-term microscale topographic changes of coastal bedrock on shore
144, 93–115. platforms. Earth Surface Processes and Landforms 29, 1663–1673.
Furlani, S., Cucchi, F., Odorico, R., 2010. A new method to study Stephenson, W.J., Thornton, L.E., 2005. Australian rock coasts: review and
microtopographical changs in the intertidal zone: one year of TMEM prospects. Australian Geographer 36, 95–115.
measurements on a limestone removable rocl slab (RRS). Zeitschrift für Swantesson, J.O.H., Moses, C.A., Berg, G.E., Jansson, K.M., 2006. Methods for
Geomorphologie, Supplementbände 54, 137–151. measuring shore platform micro erosion: a comparison of the micro-erosion
Gill, E.D., Lang, J.G., 1983. Micro-erosion meter measurements of rock wear on the meter and laser scanner. Zeitschrift für Geomorphologie, Supplementbände 144,
Otway Coast of Southeast Australia. Marine Geology 52, 141–156. 1–17.
 The Micro and Traversing Erosion Meter 169

Trenhaile, A., 2006. Tidal wetting and drying on shore platforms: An experimental Trudgill, S.T., 1977. Problems in the estimation of short-term variations in
study of surface expansion and contraction. Geomorphology 76, 316–331. limestone erosion processes. Earth Surface Process and Landforms 2, 251–256.
Trenhaile, A.S., Porter, N.J., Kanyaya, J.I., 2006. Shore platform processes in Williams, R.B.G., Swantesson, J.O.H., Robinson, D.A., 2000. Measuring rates of
eastern Canada. Geographie Physique et Quaternaire 60, 19–30. surface lowering and mapping microtopography: The use of the micro-erosion
Trudgill, S., High, C.J., Hanna, F.K., 1981. Improvements to the micro-erosion meter and laser scanners in rock weathering studies. Zeitschrift für
meter. British Geomorphological Research Group Technical Bulletin 29, 3–17. Geomorphologie, Supplementbände 120, 51–66.

Biographical Sketch
Wayne Stephenson is currently a Senior Lecturer in Geography in the Department of Geography at the University of Otago in Dunedin, New Zealand. From
1999 to 2010 he was a member of the Geography Department at the University of Melbourne in Australia. He holds a PhD and MSc from the University of
Canterbury. His research is focused on the geomorphology of rock coasts.
 14.15 Soil Description Procedures for Use in Geomorphological Studies
RJ Morrison and RA Lawrie, University of Wollongong, Wollongong, NSW, Australia
r 2013 Elsevier Inc. All rights reserved.

14.15.1 Introduction 170

14.15.2 A Brief History of Soil Survey and Descriptions 171
14.15.3 Methodology 171
14.15.3.1 Study Objectives 171
14.15.3.2 Site Selection 171
14.15.3.3 Noninvasive On-Ground Techniques 172
14.15.3.4 Site Description 172
14.15.3.5 Exposing the Soil Profile 172
14.15.3.5.1 Manual techniques 172
14.15.3.5.2 Mechanical soil excavation 173
14.15.3.5.3 Special sites 173
14.15.3.5.4 Natural and man-made exposures 174
14.15.3.5.5 Recognizing soil disturbance 174
14.15.3.6 Steps to Describe a Soil Profile 175
14.15.3.6.1 Using a pit or relatively undisturbed core 176
14.15.3.6.2 Other observations 178
14.15.3.6.3 Level of detail required in descriptions 178
14.15.3.7 Soil Field Tests 178
14.15.3.8 Sampling Soils for Laboratory Analysis 178
14.15.4 Problems, Pitfalls, and Limitations 179
14.15.5 Case Study 179
14.15.6 Future Work and Directions 179
14.15.7 Conclusions 180
References 181

Glossary distance of 75 mm or less; in the Australian Soil

Soil boundary terms There is sometimes a difference in Classification (Isbell, 2002), it is 5–20 mm.
terminology between the soil classification systems used in Soil color terms Hue refers to a color such as red or
different countries. For example, an abrupt textural change yellow; value refers to the amount of gray shading; and
in Soil Taxonomy (USA soil classification, Soil Survey Staff, chroma refers to the intensity of the color.
1999) is one where the change occurs within a vertical

Abstract

This chapter focuses on describing soils (sites and profiles) in the field, working from a geomorphic perspective. Topics
considered include placing soils in the landscape, selecting sites for description, accessing soil profiles, describing sites and
profiles, field testing that can be done, and soil sampling for laboratory analysis. Aspects such as level of detail required
(with examples) and recent advances are also considered. Some frequently encountered problems are highlighted and
current approaches discussed.

14.15.1 Introduction and the potential management of a region and its soil re-
sources (Birkeland, 1999; Gerard, 1992; Anderson and
Soils are an integral component of the terrestrial landscape. Anderson, 2010). Geomorphologists ask questions such as:
They form the outer skin of most of the land surface, and an What processes led to this soil being in this location?
understanding of soils – the types, distribution, and properties How long has this soil been here? How will the soil behave
– is often invaluable in determining landscape evolution under different conditions (natural or anthropogenic, e.g.,
erosion)? Soil properties are determined to a considerable
extent by their geomorphological location (Brady and Weil,
Morrison, R.J., Lawrie, R.A., 2013. Soil description procedures for use in
2007). An accurate description of soils in the field is, therefore,
geomorphological studies. In: Shroder, J. (Editor in Chief), Switzer, A.D.,
Kennedy, D.M. (Eds.), Treatise on Geomorphology. Academic Press, San critical to the interpretation of the landscape in most locations
Diego, CA, vol. 14, Methods in Geomorphology, pp. 170–182. on Earth.

170 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00382-1

 Soil Description Procedures for Use in Geomorphological Studies 171

A soil can be defined (after Joffe, 1949) as ‘‘a natural body Information that can be obtained by noninvasive techniques
of animal, mineral and organic constituents differentiated into is also outlined.
horizons of variable depth which differ from the material below
in morphology, physical make-up, chemical properties and
composition, and biological characteristics.’’ The term horizons
14.15.3.1 Study Objectives
is vitally important as the formation of horizons distinguishes
soil from weathered rock, which does not exhibit horizons. Soils can be described using a very wide range of techniques.
The succession of horizons exposed as a vertical cut is made Many soil profile features are readily and easily assessed in the
through the soil comprises the soil profile, which extends from field. They can generally be combined with some basic prop-
the surface to a depth where nonsoil material is encountered. erties tested in the laboratory and then used to draw inferences
Soils are open systems (materials are added at the surface and about many different soil-related issues. The objectives of the
lost from the profile by leaching) that are continually chan- study (some are listed in Table 1) provide the best guide in
ging in both the short term (e.g., water content) and the long deciding which particular properties need to be assessed. The
term (profile features). cost and availability of the necessary equipment are also fac-
Since a soil profile is effectively a two-dimensional section, tors when choosing a particular method, as well as the type of
while soils are three dimensional in the landscape, it is ne- terrain in the study area.
cessary to establish a soil entity that can be described. The unit The soils at a large number of sites characteristically need
adopted is the soil ‘pedon’, the smallest volume that can be to be described (e.g., when compiling a soil map). This generally
recognized as a soil individual. According to Soil Survey Staff means a limited range of features are assessed at most sites,
(1960) (as reported in Buol et al., 2003), a pedon is three although a few type profiles, representative of wide areas or
dimensional with its lower limit at some arbitrary limit be- having special significance to the study, will commonly be
tween the soil and nonsoil. The lateral dimensions are large examined in detail. The number and location of inspection
enough to allow study and description of any horizons present sites are generally a logistical decision as much as a technical
as they may be of variable thickness or even discontinuous. or scientific one.
The area of a pedon may range from 1 to 20 m2, depending on
the variability of the horizons.
This chapter presents a review of the methodology for
14.15.3.2 Site Selection
description of the position of soil pedon in the landscape and
for describing the actual profile. Methods of soil testing in the Selection of the sites at which to study soils is an important
field and soil sampling procedures for laboratory analysis are decision and should be made with care as poor site selection
also described. These procedures are used in landscape geo- can lead to imperfect results or a significant wastage of re-
morphology and evolution studies, especially for Quaternary sources. Remote sensing (aerial photography, satellite imagery,
investigations, soil surveys, soil genesis and management and much of it now available online) and other noninvasive
research, and for engineering developments. techniques (airborne laser scanning and various geophysical
techniques) are mostly used for mapping or to select sites
worthy of further study. Scrutiny of any previous studies may
also help in deciding on both the number and location of
14.15.2 A Brief History of Soil Survey and
sites. Filling gaps in existing data for large area studies can
Descriptions
sometimes produce a skewed distribution of new description
sites, although this is clearly a more efficient way to use scarce
Soils have been described for hundreds of years (Buol et al.,
resources.
2003), but this moved into a more formal scientific mode in
the second half of the nineteenth century. Considerable early
work was completed by Dokuchaev, Glinka, and Neustruyev Table 1 Types of soil-related studies
in Russia, and Muller in Germany; this was built upon by
Hilgard, Marbut, and others in the USA. The original de- Soil mapping
scriptions were based on visual observations that contained a • pedology/soil landscape development studies
degree of subjectivity involved. There has been a gradual • soil fertility/salinity/acidity surveys
shift to more objective and quantitative measurements of Land evaluation/suitability/capability studies
soil properties, for example, the use of Munsell color charts. • soil suitability for irrigation/cropping/horticulture/forestry/urban
This has been matched by a greater incorporation of object- development
ivity in soil classification, and the significant integration of Soil erosion surveys
geomorphology and soils, especially since the 1930s. • water-quality/catchment management studies
• estuary/wetland studies
Site assessment for waste application (liquid/solid)
• soil contamination studies
14.15.3 Methodology • soil monitoring
Urban soil studies
This section describes the various steps and procedures for
• urban parks and sports fields
soil descriptions, including site selection, profile exposure,
Soils of archeological sites
field activities, and sampling for supporting laboratory work.
172 Soil Description Procedures for Use in Geomorphological Studies

Table 2 Some non-invasive techniques used for continuous sensing of variation in the soil properties marked with an asterisk

Electro-magnetic Ground Visible/NIR Time-domain Nuclear magnetic Electrical resistance/

induction penetrating radar reflectance reflectometry resonance capacitance

Moisture      
Texture   
Topsoil depth  
Horizon boundaries/  
claypans
Salinity 
Organic matter 
Source: Adapted from Whelan, B.M., 1999. Field technologies: ‘‘On-the-Go’’ sensors. In: Cattle, S.R., George, B.H. (Eds.), Describing, Analysing and Managing Our Soil.
Proceedings of the DAMOS ’99 Workshop. University of Sydney and the Australian Soil Science Society Inc., NSW, pp. 385–400.

14.15.3.3 Noninvasive On-Ground Techniques minimize delays by carrying to the field sufficient equipment
able to cope with a range of ground conditions. In general,
A number of techniques that do not require excavation of the
two options are available – either to dig a hole in the ground
soil mantle have been used to measure particular soil prop-
(inspection pit) or to extract a soil profile core using manual
erties (Table 2). Some, like electromagnetic induction, can be
or mechanical means.
used to cover wide areas (up to several square kilometers) for
Soil inspection pits are generally dug with hand tools when:
mapping or to assist in selecting sites with particular features
that need detailed study. 1. machinery is unavailable or access is difficult;
2. the depth of inspection is shallow;
14.15.3.4 Site Description 3. only surface samples are needed;
4. ground conditions are very soft or waterlogged; and
In order to give a reader a clear indication of the profile 5. a natural exposure is not available.
position in the landscape, several site factors that affect soil
properties are generally described briefly when the profile is
being studied in some detail (e.g., see McDonald et al., 1990; 14.15.3.5.1 Manual techniques
Soil Survey Staff, 1993; USDA-NRCS, 2010). The location of The most useful tool for digging a shallow inspection pit is a
the site and the modern climate of the area are included. strong spade with a long narrow blade (tiling spade) (see
Other useful information includes the physiography of the Figure 1). Samples of undisturbed material can be prized away
surrounding landscape, and particular topographic features at from the side walls, and major profile features revealed by
the profile site are generally the most important; local drainage, gently picking back the exposed face with a knife or small pick.
vegetative cover (original or altered by human activity), and Digging a pit deeper than half a meter requires considerable
the type of soil parent material or underlying bedrock are also time and effort, and retrieval of undisturbed subsoil samples
characteristically described (examples of the types of infor- from the sidewall can be a problem unless the pit is wide
mation gathered are shown in the modern profile description enough to stand in.
for the pedon RQ-1 given later in this chapter). Small-diameter (20–50 mm in dry sites, up to 100 mm in
very soft or wet sites) steel tubes can be pushed in or ham-
mered in; they need to be levered out of hard ground by hand
14.15.3.5 Exposing the Soil Profile
(Figure 2), and this generally limits the depth of inspection to
A full and accurate soil description normally requires exam- about a meter. If the tip of the tube (Figure 3) has an outside
ination of a soil profile in the field – generally involving some bulge 1–2 mm wider than the rest of the tube, pulling it out of
invasive techniques. The profile has to be exposed for exam- the ground is much easier. In soft ground, tapered tubes can
ination, and a number of techniques are now available for be pushed in deeper (i.e., gouge augers of various designs) and
provision of a profile if none are available through natural then pulled up by hand, usually after giving them a half twist
processes (e.g., landslips). In determining how, what, and when they have penetrated to the desired depth (Figure 4).
where to dig, a number of points need consideration, in- Gouge augers have openings cut into the side wall of the tube
cluding equipment and tool selection, the use of natural ex- so that the soil can be removed with a knife or spatula.
posures, identification of profile disturbance, and issues with Straight tubes need a special tip (Figure 3) with an internal
special sites (wetlands and sands). The depth of the profile diameter slightly smaller than the rest of the tube to prevent
inspected will depend on the purpose of the study, for ex- the soil core from sticking to the inside wall. The core is re-
ample, for some taxonomic work a 2-m depth is required, for moved by pushing it out the top end with a rod, generally
many agricultural activities about 1 m is required, whereas for onto a narrow tray or half tube about the same or slightly
some engineering operations more than 2 m may be needed. larger diameter. A sheet of corrugated roofing material can
Conditions on the ground must be considered when se- hold several long cores next to each other, making it easier to
lecting an inspection technique. Delays in completing field- compare profile features from different sites (Figure 3).
work can be lengthy if the equipment most appropriate for the A hand auger (bucket-type, 100-mm diameter is common)
job is unavailable at the site when it is needed. One can is the tool of choice if disturbed material is satisfactory for
 Soil Description Procedures for Use in Geomorphological Studies 173

Auger
Spade

Meter
rule

Figure 3 Corrugated roofing material supporting equipment used for

examination of soil cores – from left, soil core from which samples
Figure 1 Hand tools used for digging soil inspection pit including have been taken, meter rule, undisturbed soil core, gouge auger and
tiling spade and meter rule (on wheelbarrow) and hand auger. a 50-mm diameter steel tube with a modified tip which has a
Excavated soil from the hole is placed on the mat in separate heaps narrower mouth than the inside diameter and a bulge above the tip.
according to depth. Site is an urban park where machinery would Foreground – samples being tested with peroxide for presence of
cause unacceptable damage. Photograph courtesy: NSW Department sulfide minerals. Photograph courtesy: NSW Department of Primary
of Primary Industries. Industries.

pulled out by hydraulic or pneumatic rams up to 2 m long,

mounted on a trailer or on the back of small truck or utility
(Figure 5). With a portable generator, an electrically powered
jackhammer can also be used in some situations. Hydraulic
pushing without a hammer is useful for shallower depths or
softer ground. On hard ground, the vehicle that these ma-
chines are mounted on can in some cases be lifted up off its
wheels unless it is very heavy. If a jackhammer is fitted, a small
trolley-mounted 1-m corer (Figure 6) can be used to collect
and retrieve cores from sensitive sites where compaction or
damage to vegetation must be minimized. Deep penetration is
possible where a nested coring technique is employed, that is,
Figure 2 (a)–(c) Hammering in and pulling out a small diameter if the mechanical (or hand) corer has a 1-m travel, a second
steel tube for soil sampling. Photograph courtesy: NSW Department coring tube 2 m long but of smaller diameter can be dropped
of Primary Industries. down the 1-m hole, hammered in, and then used to retrieve
material from deeper horizons.
subsoil examination (Figure 1). When fitted with an extension Excavating with a backhoe (sometimes trailer-mounted) is
or two, depths down to 3 m or more are achievable in dry another technique, generally employed where soil physical
ground. Material retrieved from the hole is best laid out in properties are to be examined or large sample volumes are
small individual heaps arranged in rows, spread out on a needed. Hand augering or coring through the floor of the pit
plastic sheet or tarpaulin (Figure 1). This makes it easier to can reach deeper layers if necessary. A less disruptive technique
backfill the hole, after any samples have been collected for is to collect large undisturbed soil cores using a hollow-flight
further study. auger (Wells, 1959a, 1959b), generally 150 mm, but some-
A combination of excavation techniques can be employed times 300 mm, in diameter. By using extensions, cores have
to cope with varying ground conditions. For example, in valley been retrieved from depths up to 6 m (Figure 7).
floors or floodplains the harder, dryer surface layers can be Mechanical screw augers of various diameters up to
excavated with a spade or hand auger, and a gouge auger used 600 mm can drill down rapidly to several meters, but their
for deeper examination of any sloppy waterlogged material in main purpose is to create a hole rather than collect soil
the subsoil. All the tools described above can be carried in a suitable for examination. They are useful when horizon depth
light aircraft, helicopter, or small boat, as well as on foot, to and thickness do not need to be accurately recorded as they
reach sites inaccessible to vehicles. generate mixed and heavily disturbed samples.

14.15.3.5.2 Mechanical soil excavation 14.15.3.5.3 Special sites

Steel coring tubes can be hammered in quickly to retrieve Soil properties in waterlogged floodplains and wetlands are
relatively undisturbed cores using hydraulic or pneumatic difficult to examine with conventional equipment. Provided
jackhammers (e.g., Riddler and Thomas, 1966). The tube is the surface is firm enough to walk on, a gouge auger (and
174 Soil Description Procedures for Use in Geomorphological Studies

(a) (b)

Figure 4 (a) Using a tapered gouge auger (b) in soft waterlogged ground. Photograph courtesy: NSW Department of Primary Industries.

parked on the water’s edge, can retrieve a core under deeper

water close to the shore using a long sampling tube.
Inspecting profiles in sand dunes can also involve a range
of alternative techniques, particularly when the sand is dry
and loose or soft and waterlogged. Special vibrating corers
have proved to be useful in these situations, but sometimes a
hand auger can be modified to retrieve incoherent sand from
the subsoil. A thin-walled steel sleeve (can be made if neces-
sary from a cylindrical food can of the right diameter) that sits
above the teeth of the auger, covering the area where it joins
the shaft, will often retain enough loose sand to be brought up
to the surface (Figure 8), provided the side walls of the hole
remain stable during excavation. When driven into loose sand,
steel tubes sometimes block up if the sand fails to rise up
inside them, effectively converting the hollow tube into a solid
rod. To check if this has occurred, the length of the core has to
be compared with the depth of the hole.

14.15.3.5.4 Natural and man-made exposures

Sometimes soil profiles are exposed in road or railway cut-
tings, or in riverbanks or cliffs. Provided there is little or no
vegetative cover, these exposures can sometimes be used as a
substitute for pits dug for soil survey purposes. Their use for
more detailed study generally requires a considerable amount
of site preparation to remove possible surface contaminants.
This is not normally an issue in freshly dug pits or trenches at
construction sites or quarries.

Figure 5 Trailer-mounted soil corer fitted with a hydraulically 14.15.3.5.5 Recognizing soil disturbance
powered jackhammer (1.8-m travel). Photograph courtesy: NSW Soil profiles are in some places subject to major disturbance
Department of Primary Industries. by human activities. This can produce conspicuous features
such as abrupt breaks in the lateral continuity of horizons;
extensions) can be carried out into a swamp and used to buried, truncated, or missing horizons; the blending or mixing
collect profile samples under shallow water. The soil retrieved of horizons; or the presence of anomalous or extraneous
is left inside the auger, and carried back onto dry ground for material inside the profile (i.e., man-made objects). Less ob-
description and sampling. Sometimes a truck-mounted corer, vious are accumulations of natural debris, such as charcoal or
 Soil Description Procedures for Use in Geomorphological Studies 175

(a) (b)

Figure 6 (a) Small trolley-mounted soil corer (1-m travel) fitted with a hydraulically powered jackhammer used for sampling sites where there
is no vehicle access. (b) The same corer mounted on a 4WD truck. Photograph courtesy: NSW Department of Primary Industries.

(a) (b)

Figure 7 (a) Components of a 150-mm hollow-flight auger, with (right) a 1-m extension. (b) Truck-mounted 300-mm diameter hollow-flight
auger. Photograph courtesy: NSW Department of Primary Industries.

as unsuitable for the study. On the other hand, detailed

observations of disturbance produced by human activity
are characteristically essential when profiles are described at
archeological sites.

14.15.3.6 Steps to Describe a Soil Profile

The most common soil properties described are the depth and
thickness of horizons, color, texture (assessment of particle-
size distribution), structure, presence of coarse fragments,
Figure 8 A hand auger adapted for use in loose sandy soils. In use, presence of major features such as hardpans, and type of
the steel sleeve is pushed down against the head of the auger. boundary between horizons. A full description of how to as-
Photograph courtesy: NSW Department of Primary Industries.
sess basic features such as color, texture, and structure is
available in many manuals and textbooks (e.g., McDonald
fired clay, disrupted mottle patterns, or soil structural damage et al., 1990; Taylor and Pohlen, 1970; Soil Survey Staff, 1993;
produced by heavy vehicles. Undisturbed large cores or pits Brady and Weil, 2007; USDA-NRCS, 2010), and, hence, full
generally reveal these features much more clearly than hand details will not be provided here. It is best to check which
auger samples; mechanically augered soil samples can easily reference/manual is used for soil descriptions in the area
obscure these signs of human activity. Knowledge of prior where you are working.
land use can help greatly not only in assessing the likely extent The basic steps are described in the following, although the
of human disturbance, but also in deciding to reject a site order may vary depending on the outcomes required.
176 Soil Description Procedures for Use in Geomorphological Studies

14.15.3.6.1 Using a pit or relatively undisturbed core 5. Record the dominant color (hue, value, and chroma,
using a Munsell color book) on the surface of a freshly
broken aggregate from each layer.
1. Prepare the pit by picking back any smeared soil from the
6. Look for any mottles (features associated with prolonged
least shaded side wall, or split open the soil core (after
wetness) and record their size, abundance, and color.
pushing it out of the tube) (see, e.g., Eldridge, 2003).
7. Assess the texture of each layer by moistening (if required)
2. Find the boundaries between the horizons; this can be
and kneading a small handful of soil so that it just fails to
helped by selecting a clod typical of each layer or horizon
stick to the fingers (Figure 9). Note the behavior when
and placing it level with the layer above and below it.
squeezed between thumb and forefinger, referring to
Moving the clod up and down outside its own horizon
Figure 10 and/or Table 3. This enables an experienced
will reveal where the appearance of the soil changes.
practitioner to assess the proportions of different-sized
3. Record the depth and the thickness of each layer or
particles in the sample. Although there are small vari-
horizon and the nature of the boundaries between them
ations in the size classes used in different countries (see
(whether sharp, abrupt, clear, gradual, or diffuse, and
McDonald et al., 1990), the general ranges are: coarse
straight or wavy).
fragments, 42 mm; sand, 2 mm to 50 mm; silt 50–2 mm;
4. Record the soil moisture status of each layer as dry,
clay o2 mm.
moderately moist, moist, or wet.
Once the various proportions have been determined, they
are generally converted into a shorthand notation (or vice
versa) using a texture triangle such as in Figure 11 (see also
USDA-NRCS, 2010).
8. Note the abundance, size, shape, lithology, and distri-
bution of any coarse fragments in each layer.
9. Describe the structure of each layer in terms of the dis-
tinctness, size, and shape of the peds (McDonald et al.,
1990; USDA-NRCS, 2010). The major structural types are
shown in Figure 12.
10. Note whether the fabric is earthy, sandy, smooth ped, or
rough ped (McDonald et al., 1990); check for any clay
coatings on peds (clayskins/argillans/cutans).
11. Note the size and abundance of cracks and pores; look for
old worm channels or other soil fauna burrows.
Figure 9 Assessing soil texture. Photograph courtesy: NSW 12. Record the consistence (the strength of cohesion and ad-
Department of Primary Industries. hesion) of a 20-mm piece of soil from each layer, and

Moist
soil

Will not form a Will easily form a

ribbon or ribbon ribbon longer than
shorter than 25 mm 25 mm

Soil feels Ribbon cannot Ribbon can

Soil feels Soil feels
smooth and be turned be turned
gritty silky
spongy into a ring into a ring

Soil will Soil will Soil feels Soil does not Soil easy Soil stiff
not make make a gritty feels gritty to mold to mold
a firm ball firm ball

Sand Sandy Silty Loam Sandy Clay Light Medium to

loam loam clay loam loam clay heavy clay

Increasing clay content

Increasing sand content
Figure 10 Chart for assessing soil texture. Reproduced from Rengasamy, P., North, S., Smith, A., 2010. Diagnosis and Management of Sodicity
and Salinity in Soil and Water in the Murray Irrigation Region. School of Agriculture Food and Wine, the University of Adelaide, South Australia.
 Soil Description Procedures for Use in Geomorphological Studies 177

Table 3 Broad field texture classes

Field texture Description Approximate clay content

Sand Nil to slight coherence. Ribbon of 0–15 mm. Less than 10%
Sandy loam Coherent but very sandy to touch. Ribbon of 15–25 mm. 10–20%
Loam Coherent, spongy, and greasy feel with no obvious sandiness or About 25%
silkiness. Ribbon of about 25 mm.
Silt loam Coherent, very smooth to often silky when manipulated. Ribbon of About 25% and with silt 25% or more
about 25 mm.
Sandy clay loam Strongly coherent, sandy to touch with medium size sand grains 20–30%
visible in finer matrix. Ribbon of 25–40 mm.
Clay loam Coherent plastic bolus. Smooth to touch with no obvious sand 30–35%
grains. Ribbon of 50–75 mm.
Light clay Plastic bolus. Smooth to touch with slight resistance to shear. 35–40%
Ribbon of 50–75 mm.
Medium to heavy Plastic bolus. Smooth to touch. Feels like normal to stiff plasticine. 40% or more
clay Moderate to firm resistance to shear. Ribbon of 75 mm or more.

Source: Reproduced from Rengasamy, P., North, S., Smith, A., 2010. Diagnosis and Management of Sodicity and Salinity in Soil and Water in the Murray Irrigation Region. School
of Agriculture Food and Wine, the University of Adelaide, South Australia.

100

10
90

80 20
30
70
Pe
y

Clay
Cla

rce
40
t-

nt
igh

60 by
we

we
50
by

igh
50
nt

t-
Silty
rce

Sandy
Sil

clay
60
Pe

clay
t

40
Silty Clay
Clay loam
70

loam
30
Sandy clay loam
80

20
Loam
Silt loam
90

Sandy Loam
10
Loamy Silt
0

Sand Sand
10
10
20
10

40
70

30
60

50
80
90
0

Percent by weight - Sand

Figure 11 Particle-size summary terminology using the texture triangle.

check for any hardpans or cemented layers in the profile texture). When using the auger, it is essential to note the depth
(McDonald et al., 1990; Soil Survey Staff, 1993). that each small heap of excavated soil came from. Removing
13. Examine the nature of the soil surface, particularly noting soil in small, uniform-depth increments (i.e., 4 or 5 cm) will
the presence of any cracks. help in keeping track of depths and reduce the incidence of
14. Identify any pedogenic segregations (e.g., lime con- missing or blurring the boundary between adjacent horizons.
cretions, gypsum crystals, and manganese nodules) and The field observations must be recorded immediately in a field
record their form, abundance, and distribution. notebook, a preformatted data collection sheet, or a laptop
computer. Early transcription of field notes into a standard profile
Disturbed samples excavated with a hand auger can be used for description with notes on sample numbers and additional infor-
some of the above-mentioned observations (e.g., color and mation is highly recommended.
178 Soil Description Procedures for Use in Geomorphological Studies

Spheroidal often sufficient when detailed knowledge of the various soils

Commonly found in within the unit was rudimentary.
Granular
A horizons and also For example, a typical profile of the Beelbangera clay
Crumb
in B horizons
loam (Taylor and Hooper, 1938) was described elegantly and
succinctly by van Dijk (1961), as follows:
Blocklike
0–7 in: Light brown clay loam, no aggregation, sometimes
Angular
weakly platy.
Blocky
Commonly found in 7–14 in: Reddish brown heavy clay, moderate grade of
B horizons, particularly aggregation, angular prismatic 1½  3 in easily breaking
in more humid regions down to angular blocky 1–1/2 in.
Subangular
Blocky 14–42 in: Brown to yellowish brown, clay, well aggregated,
angular blocky 1½ to 1/2 in slight degree of sub-plasticity,
with light to moderate amounts of soft and concretionary
Prismlike
lime and usually the fine earth effervescences to dilute acid.
Prismatic
(level tops) This was sufficient to recognize the profile taxonomically as
Both normally found in a red-brown earth (Stephens 1956). A more recent description
B horizons, especially will need more detail. To allocate this profile to the broadest
in soils of arid and
grouping (soil order) of the Australian Soil Classification
semi-arid regions
Columnar (Isbell, 2002), for example, information on the nature of the
(rounded boundary between the A and B horizons is required, and on
tops)
the color of the B horizon to allocate the suborder.
Platelike
May be found in any part
Platy-leafy of profile, commonly derived 14.15.3.7 Soil Field Tests
from the parent material
Tests commonly performed include pH, presence of lime (ef-
fervescence with a few drops of hydrochloric acid), presence of
Figure 12 The main structural units occurring in soils. Modified
sulfides (reaction with hydrogen peroxide, Figure 3), and ag-
from Brady, N.C., Weil, P.R., 2007. The Nature and Properties of
Soils, Fourteenth ed. Prentice Hall, Upper Saddle River, NJ, 960 pp.
gregate stability (behavior of small dry aggregates when
dropped in distilled water, Figure 13) (USDA-NRCS, 2010).

14.15.3.6.2 Other observations

Other features commonly described include root depth and 14.15.3.8 Sampling Soils for Laboratory Analysis
distribution, signs of soil faunal activity, presence of any water
table, nature of the underlying regolith or substrate, nature of Depending on the project objectives, soil samples may be col-
profile drainage, extent and likely duration of water logging, lected and sent to the laboratory for analysis. Common tests
and the nature of any overburden covering the profile. carried out include particle-size distribution, bulk density,
There are few instances when all of the above-mentioned structural stability, moisture-retention characteristics, pH, cat-
features are described at many sites. A few profiles may be ion exchange capacity, exchangeable bases, organic carbon,
selected as special reference sites for detailed examination, but, nitrogen, and phosphorus. For geomorphic studies, total
normally, only the particular features most relevant to the element analysis, mineralogy, and thin sectioning may be re-
purpose of the study are described. quired. For many analyzes, collection of a 250–500-g sample
When making a taxonomic description, the requirements without any special precautions may be sufficient, as the ma-
of the soil classification system must be recognized; samples terial collected will be air-dried and ground to a desired particle
may be needed for laboratory testing. Occasionally, field ob- size for the particular technique (commonly o2 mm,
servations can act as surrogates for particular chemical prop- o250 mm, or o63 mm). For the physical analyzes, more care is
erties used as taxonomic criteria, for example, the diagnostic generally required. Some tests should be carried out on field-
criteria for a Ferrosol in the Australian Soil Classification moist (undried) materials (e.g., particle-size analysis, moisture
(Isbell, 2002). characteristics, and structural stability). It is, therefore, common
practice to collect samples in plastic bags (or other container)
and double or triple bag them to minimize moisture loss (in
the laboratory, such samples can be subsampled to get material
14.15.3.6.3 Level of detail required in descriptions for drying); for bulk density, accurately measured (in the field)
Over the years, the number of features described has in- soil volumes have to be collected. It is recommended that
creased, most probably prompted by the requirements of the specifically designed containers be used for samples being
particular soil classification system used in the study. For taken for structure stability assessments.
mapping purposes, it was sufficient to describe a soil only to In the field, the actual sampling should be done from the
the detail needed to separate the different mapping units, for lowest horizon upward to minimize any contamination of sam-
example, soil series, associations, or broader groupings. Allo- ples. In general, samples should be taken with a small pick or
cating a profile to a broad international taxonomic unit was trowel so that they are representative of the whole horizon (or
 Soil Description Procedures for Use in Geomorphological Studies 179

Figure 13 Soil aggregate stability testing – samples from different horizons after (right to left) 2 min, 2 h, and 20 h.

other selected zone) and not just the middle. Samples are then after a pit has been dug. Recognizing the disturbance, and
placed in the plastic bag or other container and immediately deciding on the location of a suitable replacement, can de-
labeled (inside and outside of the container) in a way that will pend on the extent of the disturbance and also requires some
not be lost during transportation (permanent markers and strong experience and familiarity with soil profile features.
card). For bulk density, a common procedure is to have steel Deciding on the depth to which a soil should be investigated
tubes that can be driven into each horizon and then removed is an ongoing question. The ultimate criterion is the objective of
with a known volume of soil inside – these tubes can be sealed the project. In general, the deeper the better, but time, effort, and
and sent to the laboratory for weighing of the soil mass therein. logistical considerations are also important. If a pit is too shal-
In the planning of any project, it is important to discuss with low, some key soil features (e.g., discontinuities and water table)
the testing laboratory what tests will be completed, what size may be missed, compromising any geomorphic interpretation of
(mass) of samples is required, what types of samples are needed the site. Too deep a pit obviously means inefficient use of time
(moist/dry), and any special equipment needed for collection of and resources. This latter point is one of many issues that relates
specific test samples. Allowance should be made for replication to balancing resources and objectives in soil investigations, and
and storage of reference material for future work. the allocation of sufficient time to the overall planning of a
project cannot be over emphasized.

14.15.4 Problems, Pitfalls, and Limitations

14.15.5 Case Study
As many aspects of describing soils in the field are relatively
standard around the world, there are few specific problems to In Section 14.15.3.6, a soil profile description of the Beel-
be considered. Several issues, however, do require careful bangera clay loam was presented as being sufficient for the
consideration. In different countries there are differences in purpose at the time of the study. This should be compared
terminology, for example, soil classification systems; particle- with the site and profile description for the pedon RQ-1 given
size classes/ranges vary slightly between countries (even though in Table 4. This latter description was developed for a project
there is an international system) and the ability to accurately investigating the potential application of dairy effluent to
assess particle size in the field does require some experience. soils, but it does indicate the detail that is provided in modern
The selection of the number of profiles (or auger holes) to soil descriptions. In particular, a site description is provided
examine in order to assess soil variability requires some ex- and, in the profile description, details on biological activity
perience, as does determination of the level of detail required (ants, worms, and roots), boundaries between horizons, and
in descriptions to match the requirements of the project ob- consistency are given, and there is more information on soil
jectives. Assessing the local landscape is critical, as many clues structure. These features assist in interpretation of the soil
to the formation and properties of soils can be obtained from from both a geomorphic and an agronomic perspective.
a good understanding of the landscape position. Such an
understanding is also invaluable in the selection of sites for
more detailed inspection and sampling. 14.15.6 Future Work and Directions
Sometimes there is major disturbance to the profile that
has the potential to compromise the findings of the study. This The most significant directions in soil survey work are the
may not be apparent on the surface, and may only be revealed attempts to try to gain more information about soil profiles
180 Soil Description Procedures for Use in Geomorphological Studies

Table 4 The site and profile description for the pedon RQ-1

Pedon RQ-1
Site Description
Classification: Brown Dermosol (Isbell, 2002)
Typic Argiudoll (Soil Survey Staff, 1999)
Location: The property of Mr. Barry Leaney, corner of Woodstock and Croobyar Roads, Milton, NSW, Australia. The pit was located in the corner of a
paddock, approximately 20 m from the fence intersection. Grid Reference KF657877, NSW 1:25 000 Topographic Map Sheet 8927-2-N.
Physiography: The landscape consists of undulating hills, with the study site at an elevation of just over 50 m above mean sea level. Hills of greater
elevation are located approximately 1 km to the north, and Kingmans Ridge is approximately 8 km to the west at 300 m above mean sea level.
Topography: The pit was on the upper third portion of a convex 300-m-long west-facing slope; approximately 100 m from the crest with an on-site
slope of 3–51; the surface is flat with little microtopography, in certain sections of the paddock there are a few moderately sized subrounded stones,
no exposed bedrock; and there is no apparent erosion.
Drainage: The site and profile were well drained. The site drains to the west and eventually to Stoney Creek at the foot of the slope. Free water was not
reached at 1.2 m.
Vegetation: The site has been completely cleared, sown to pasture, and cultivated. Cultivation in the 1970s occurred to a depth of 12 cm each year and
in the 1990s has been cultivated to 2–3 cm every 5 years. The immediate site is covered with grasses only. A non-native hedge borders the paddock
to the east along Woodstock Road.
Parent material : Laccolith of porphyritic monzonite of Late Permian age (Ward, 1991).
Climate: Temperate maritime climate with variable rainfall, averaging 1246 mm per year. Average annual temperature is 17 1C, with the maximum
temperature averaging 24 1C and 16 1C in summer and winter months, respectively.
Soil profile description (by P.H. Walker and R. Quinane)
Ap1 0–8 cm Black (10YR 2/1; 3/2, dry); silty clay loam (approximately 30% clay); weak polyhedral structure with peds to 100 mm; slightly sticky and
slightly plastic consistence; few small stones; many fine and medium grass roots; field pH 5.5; abrupt straight boundary to
Ap2 8–22 cm Black (2.5Y 2/0; 10YR 3/1, dry); silty clay loam to silty clay; strong polyhedral structure with peds 50–200 mm; slightly sticky and
slightly plastic consistence; few small stones; common grass roots; ant and worm activity evident; field pH 6; gradual straight boundary to
AB 22–32 cm Dark reddish brown (5YR 3/2) with 30% very dark brown (10YR 3/1) mottles; medium clay; weak angular and subangular blocky
structure with peds to 100 mm; slightly sticky to moderately sticky and slightly to moderately plastic consistence; frequency of roots decreasing;
field pH 5.5–6; gradual, slightly wavy boundary to
Bt 32–50 cm Dark brown (7.5YR 3/4); medium to heavy clay; moderate angular and subangular structure with peds to 60 mm; a few discontinuous
darker (5YR 3/2) cutans on ped faces; very sticky and very plastic consistence; few grass roots but evidence of old tree root at this level elsewhere in
the pit; field pH 6–6.5; gradual, slightly wavy boundary to
BC 50–65 cm Dark brown (7.5YR 3/4); light to medium clay with approximately 20% less clay than Bt; massive with a slight tendency to break into
subangular and granular peds; slightly sticky and slightly plastic consistence; few fine roots, occasional rounded monzonite stones to 400 mm
diameter, with evidence of spheroidal weathering; field pH 6; gradual, slightly wavy boundary to
C1 65–120 cm This horizon consists of two distinct components: dominant are ‘ghosts’ of weathered monzonite boulders up to 500 mm diameter,
mottled dark brown (7.5YR 3/4 60%) and strong brown (7.5YR 4/6 40%), gritty light clay; minor features are ‘tongues’ of dark grayish brown (2.5Y
4/2) medium clay; ‘ghost’ field pH 6, ‘tongue’ field pH 7
C24120 cm Substrate of weathered monzonite
Soil colors are moist unless otherwise specified

Source: Reproduced from Quinane, R., 1997. Sustainability of the land application of effluent in the Southern Shoalhaven Region, NSW. Unpublished BEnvSc Honours Thesis,
University of Wollongong, New South Wales, Australia, 90 p þ Appendices.

using sensors, both near and remote, with an even greater an indication of soil properties and as an inhibitor of remote-
interest in the latter. As will be noted from the comments sensing applications, also requires greater research effort.
above, the collection of profile information is quite manpower Another direction in which pedologists are moving is to
intensive, requiring a profile to be excavated and then exam- enhance the testing of soils that can be done in the field. New
ined. Even an experienced group can examine only a limited and improved techniques now permit completion of numer-
number of profiles per day. There has been some progress ous tests limiting the cost and time delays associated with
made in the use of noninvasive techniques (see Sec- laboratory work. For some information, especially for detailed
tion 14.15.3.3), including studies where two remote techni- taxonomic data and research purposes, laboratory testing is
ques are used in combination (e.g., Castrignano et al., 2010). essential; however, test kits are now available that give data on
Another area currently receiving some attention is the devel- infiltration, bulk density, conductivity, pH, nitrate, aggregate
opment of rapid methods of reliably assessing the amount of stability, earthworms, respiration, and some associated water-
carbon and organic matter, and some remote-sensing appli- quality information.
cations have been developed to give assessments of mainly
surface soil properties (e.g., Gomez et al., 2008). Further
evaluation of these, and other noninvasive techniques, is 14.15.7 Conclusions
needed before they are widely adopted.
The capacity to gather more information about subsurface The description of soils in the field has not changed dramat-
properties, apart from those directly affected by moisture, ically over the last century. The main features and procedures
needs further attention. The influence of vegetation, both as still involve careful examination of the soil profile. What has
 Soil Description Procedures for Use in Geomorphological Studies 181

changed is the level of detail and to some extent the degree of Quinane, R., 1997. Sustainability of the land application of effluent in the Southern
quantification that is now included. Significant research effort Shoalhaven Region, NSW. Unpublished BEnvSc Honours Thesis, University of
is trying to develop new procedures for soil assessment that Wollongong, New South Wales, Australia, 90 p þ Appendices.
Rengasamy, P., North, S., Smith, A., 2010. Diagnosis and Management of Sodicity
can be done remotely without the need to excavate a soil
and Salinity in Soil and Water in the Murray Irrigation Region. School of
profile for examination. Developments to date in this regard Agriculture Food and Wine, the University of Adelaide, South Australia.
can provide limited information, especially about upper soil Riddler, A.M.H., Thomas, A.E., 1966. A power driven thin walled core sampler.
layers, but a full understanding of soils and their position in Agricultural Gazette of New South Wales 77, 627.
the landscape still requires an accurate and detailed de- Soil Survey Staff, 1960. Soil Classification, a Comprehensive System – 7th
Approximation. US Department of Agriculture. US Government Printing Office,
scription in the field, together with appropriate supporting
Washington, DC.
laboratory analyzes. Soil Survey Staff, 1993. Soil Survey Manual. USDA Handbook No.18, US
Government Printing Office, Washington, DC, 437 pp.
Soil Survey Staff, 1999. Soil Taxonomy: A Basic System of Soil Classification for
References Making and Interpreting Soil Surveys, Second ed. Soil Conservation Service,
USDA-NRCS Agricultural Handbook No. 436, Washington, DC, 871 pp.
Stephens, C.G., 1956. Manual of Australian Soils, Second ed. CSIRO, Melbourne.
Anderson, R.S., Anderson, S.P., 2010. Geomorphology: The Mechanics and
Taylor, J.K., Hooper, P.D., 1938. Soil Survey of the Horticultural Soils in the
Chemistry of Landscapes. Cambridge University Press, Cambridge, 637 pp.
Murrumbidgee Irrigation Areas, NSW. CSIRO Bulletin No. 118, CSIRO,
Birkeland, P.W., 1999. Soils and Geomorphology, Third ed. Oxford University Press,
Melbourne.
New York, NY, 430 pp.
Taylor, N.H., Pohlen, I.J., 1970. Soil Survey Method. Soil Bureau Bulletin 25, New
Brady, N.C., Weil, P.R., 2007. The Nature and Properties of Soils, Fourteenth
Prentice Hall (ed.), Upper Saddle River, NJ, 960 pp. Zealand Department of Scientific and Industrial Research, Lower Hutt, 242 pp.
Buol, S.W., Southard, R.J., Graham, R.C., McDaniel, P.A., 2003. Soil Genesis and USDA-NRCS, 2010. U.S. Department of Agriculture, Natural Resources Conservation
Classification. Fifth ed. Iowa State Press, Ames, 494 pp. Service. National Soil Survey Handbook, Title 430-VI. Available online at: http://
Castrignano, A., Wong, M.T.F., Guastaferro, A., 2010. Assessment of soil variation soils.usda.gov/technical/handbook/ (accessed June 2011).
by multivariate geostatistical analysis of EMI and gamma-radiometric data. Ward, P.J., 1991. Landforms developed on a Monzonitic Intrusion at Milton on the
Proceedings of the 19th World Congress of Soil Science, Brisbane. South Coast of New South Wales. Unpublished BSc Honours thesis, Department
van Dijk, D.C., 1961. Soils of the Southern Portion of the Murrumbidgee Irrigation of Geography, University of Wollongong, NSW, Australia.
Areas. Australian Soil Publication No. 11, CSIRO, Melbourne. Wells, C.B., 1959a. Core samplers for soil profiles. Journal of Agricultural
Eldridge, S.M., 2003. Sugar soils: A Guide to Characterising Australian Sugarcane Engineering Research 4, 260–266.
Soils. CRC Sugar Technical Publication, CRC for Sustainable Sugar Production, Wells, C.B., 1959b. Description and comments on a power driven soil profile core
Townsville, 180 pp. sampler. Division of Soils Report 5/59, CSIRO, Melbourne.
Gerard, A.J., 1992. Soil Geomorphology: An Integration of Pedology and Whelan, B.M., 1999. Field technologies: ‘‘On-the-Go’’ sensors. In: Cattle, S.R.,
Geomorpohology. Chapman and Hall, London, 269 pp. George, B.H. (Eds.), Describing, Analysing and Managing Our Soil. Proceedings
Gomez, C., Viscarra Rossel, R.A., McBratney, A.B., 2008. Soil organic carbon of the DAMOS ’99 Workshop. University of Sydney and the Australian Soil
prediction by hyperspectral remote sensing and field vis–NIR spectroscopy: an Science Society Inc., NSW, pp. 385–400.
Australian case study. Geoderma 146, 403–411.
Isbell, R.F., 2002. The Australian Soil Classification (Revised Edition). CSIRO
Publishing, Melbourne, 2002 pp.
Joffe, J.S., 1949. Pedology. Second ed. Somerset Press, New Brunswick, NJ,
662 pp. Relevant Websites
McDonald, R.C., Isbell, R.F., Speight, J.G., Walker, J., Hopkins, M.S., 1990.
Australian Soil and Land Survey: Field Handbook, Second ed. Inkata Press, http://soils.usda.gov
Melbourne, 198 pp. US Department of Agriculture: NRCS, Soil Quality Test Guide.

Biographical Sketch

John Morrison is BHP Professor of Environmental Science at the University of Wollongong. He was originally
trained as a chemist but, since the late 1970s, his research interests have focused on land and water science and
management, particularly the land–ocean interface and the environmental impacts of agricultural activities. He
has extensive experience in environmental issues in the Pacific Islands where he resided for almost 20 years. His
main areas of research activity are currently: ion-exchange properties and mineralogy of soils; salinity impacts on
the environment; agriculture impacts on small island ecosystems; and waste management in Australia and the
Pacific Islands.
182 Soil Description Procedures for Use in Geomorphological Studies

Roy Lawrie is an Australian soil scientist with the Department of Primary Industries and Honorary Research Fellow
at the University of Wollongong, New South Wales. He has worked in many areas of soil science, including soil
mapping and land classification, soil characterization and sampling of research sites, acid sulfate soil studies, land
suitability for irrigation, land application of a range of organic wastes, the development of soil sampling ma-
chinery, and soil studies of archaeological sites. The work has taken him from the semi-arid inland to the high-
rainfall coastal areas of eastern Australia. He currently works in communicating soils information to rural
landholders (see photo, author on right), using soil monoliths or undisturbed soil cores.
 14.16 Ground Penetrating Radar
CS Bristow, Birkbeck University of London, London, UK
r 2013 Elsevier Inc. All rights reserved.

14.16.1 History of Ground Penetrating Radar (GPR) 183

14.16.2 GPR Principles 183
14.16.3 Equipment 184
14.16.4 Processing 186
14.16.5 Survey Design 187
14.16.6 Radar Profiles as Cross-Sections and Ground Truth 187
14.16.7 Radar Facies 188
14.16.8 Radar Stratigraphy 188
14.16.9 3-D Date and 2.5D Grids 189
14.16.10 Problems, Pitfalls, and Limitations 189
14.16.11 Side Swipes and Airwaves 189
14.16.12 Examples: Fluvial Geomorphology 190
14.16.13 Sand Dunes 192
References 193

Abstract

Ground penetrating radar (GPR) provides almost continuous, high resolution profiles of the shallow subsurface which
cannot be achieved by any other geophysical technique. Applications of GPR in geomorphology have increased significantly
in recent years due to an increased awareness of the technique, combined with improvements in the equipment which
make it more portable, more reliable, and easier to use. The ease of use has presented some problems due to unrealistic
expectations, inappropriate survey design, and questionable interpretations. Examples are included to illustrate common
‘problems’ as well as successful GPR surveys.

14.16.1 History of Ground Penetrating Radar (GPR) Commercial GPR systems became available in the 1970s
(Morey, 1974; Annan and Davis, 1976). Since then, the devel-
RADAR is an acronym for radio detection and ranging and was opment of electronics and computing have seen GPR systems
developed during the early twentieth century becoming a getting smaller, faster, and more reliable so that GPR is now a
practical technique for detecting aircraft during the Second technique that is widely used by archeologists, engineers, fo-
World War. However, radar had already been used for depth rensic scientists, geologists, and geomorphologists. The increased
sounding on a glacier in Austria by Stern (1929, 1930) and use of GPR is reflected in the literature which has blossomed
there was a slow development of applications through the during recent years. The bar chart (Figure 1) is compiled from a
following decades that are reviewed by Annan (2002). One of popular search engine using the terms GPR and Geomorphology
the key developments followed reported errors by aircraft and although not necessarily strictly accurate it is probably a
pilots using pulsed radar altimeters over the Antarctic ice in realistic representation of the state of research with very few
1946 (Waite and Schmidt, 1961), who reported that the al- publications in the early 1990s, rapid growth in the mid-1990s
timeter of an aircraft ‘flying low over 500-ft-thick ice would as new digital equipment became available, and then a second
read 900 ft of altitude when the plane’s wheels were about to growth in the 2000s as GPR was adopted more widely by the
hit the surface.’ Waite and Schmidt were most concerned about geomorphological community. Several key texts provide reviews
the safety implications of their results but their experiments of relevant data including: Ground Penetrating Radar in Sediments
opened the way for airborne surveys of ice thickness across the (Bristow and Jol, 2003), Stratigraphic Analyses Using GPR (Baker
polar icecaps in Antarctica and Greenland, where pulsed radar and Jol, 2007), and Ground Penetrating Radar: Theory and Appli-
was used to determine the thickness of the Antarctic and cations (Jol, 2009) as well as the review of Neal (2004). These
Greenland icecaps (Bailey et al., 1964; Walford, 1964). How- texts are recommended for further details of the principles
ever, GPR remained an experimental technique for many years underlying GPR as well as case studies and relevant reviews of
before it became commercially viable (Annan, 2002). applications in geomorphology and other disciplines.

14.16.2 GPR Principles

Bristow, C.S., 2013. Ground penetrating radar. In: Shroder, J. (Editor in
Chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise on Geomorphology.
Academic Press, San Diego, CA, vol. 14, Methods in Geomorphology, GPR is an active geophysical survey technique that transmits an
pp. 183–194. electromagnetic (EM) signal into the ground. The electrical

Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00383-3 183

184 Ground Penetrating Radar

250 14.16.3 Equipment

There are two common types of GPR which differ in the way
that data are acquired, either in the time domain (impulse
200
radar) or the frequency domain (continuous-wave and stepped-
frequency radars). The GPR systems described here are impulse
radar systems, where a pulse is transmitted and the time taken
Number of papers

150 for reflected energy to be received by the antennas is recorded.

The resulting waveform indicates the amplitude of the energy
scattered from subsurface objects in the time domain. The re-
corded signal is displayed in two-way traveltime (twt), the time
100 taken for the pulse to travel down through the ground to the
subsurface object and to return back to the surface where it is
recorded. Most modern impulse GPR systems transmit multiple
signals and then sequentially sample the return waveform for
50 display, data storage, and postprocessing so that the signal that
is displayed is actually a composite of repeated transmissions
and reflections. Prototype systems have been built with a flash
conversion or complete waveform sampling where the entire
0 waveform, after one complete pulse is digitized. This reduces
1993

2000

2010
the time taken for data collection but these systems are not yet
commercially available.
Year of publication It is not the intention of this chapter to provide an ex-
Figure 1 The number of publications that include GPR and haustive comparison of the pros and cons of different
geomorphology listed by a popular search engine have increased manufacturers or their products. In addition, technology is
significantly in the past decade. developing rapidly and radar systems are becoming smaller,
faster, and easier to operate, so that equipment on the market
today is likely to be superseded by newer products very shortly.
properties of natural geological materials is complex and beyond For example, output has moved from thermal plotters to
the scope of this chapter. Readers are referred to papers by magnetic tapes, laptop computer hard-drives, and video data
Olhoeft (1981), and the review by Cassidy (2009a) which pro- loggers in a few decades and there is a system that records on a
vides an accessible summary that is strongly recommended. GPR high-end mobile phone, and although this is not available
operates in a signal-loss environment where EM waves are at- commercially, it is indicative of the way that radar systems
tenuated, scattered, or absorbed. Part of the signal that is might develop in the near future.
transmitted into the ground will be attenuated or scattered de- Several equipment manufacturers market GPR, among
pending on the ground conditions. However, part of the signal is which the leading manufacturers are Geophysical Survey Sys-
reflected from layers in the subsurface where there is a contrast tems Inc., Sensors and Software Inc. who make the PulseEKKo
in dielectric properties and returns to the surface where it is system, and MALÅ. Each system has slight differences in the
recorded and it is this part of the signal that is used to image configuration but several features are in common. A typical
subsurface stratigraphy. The surfaces in the ground that reflect GPR system will comprise of one antenna or a pair of an-
the radar signal are called reflectors and the signal that is re- tennas connected to a console which controls the timing of
corded at the surface is called a reflection. The frequency of the the GPR signal transmission and subsequent sampling of the
EM signals transmitted by GPR systems typically range between reflected signal, a data logger to record the signal with an
10 and 1000 MHz. The frequency of the transmitted signal and analog to digital conversion (A/D) and a display screen to
the velocity of the signal through the ground control the show the results in the field, as well as a power source, gen-
wavelength of the radar signal. The wavelength of the radar erally a 12 V battery (Figure 2).
signal affects the resolution and there is a trade-off between the When selecting a radar system for a survey one critical
depth of penetration of the radar signal and the resolution be- choice is the antennas which come in a variety of shapes and
cause low-frequency signals with a longer wavelength have a sizes. First is the choice of antenna frequency which together
greater depth of penetration but less resolution whereas higher- with the velocity of the radar signal through the ground de-
frequency signals with a short wavelength have higher resolution termines the wavelength of the signal, resolution, and depth
but less penetration into the ground. In geomorphological in- of penetration. Lower-frequency antennas, for example,
vestigations typical radar frequencies are between 50 and 10–50 MHz produce longer wavelength signals with lower
200 MHz. The higher-frequency systems (500–1000 MHz) are resolution and increased depths of penetration; whereas higher-
generally used for engineering applications such as the detection frequency antennas, for example, 200–500 MHz produce
of reinforcing bars in concrete or infrastructure studies where shorter wavelength signals with higher resolution but reduced
resolution, to accurately determine the location of a hidden depths of penetration. Having selected an appropriate wave-
object, is more important than the depth of penetration into the length for the objectives of the survey (see Jol and Bristow,
subsurface. Lower-frequency GPR systems have been developed 2003), there are different types of antennas available. There are
for geological applications such as mining surveys. at least four different types of antennas in common use: bistatic
 Ground Penetrating Radar 185

Power Display

Control and record

Timing control Delay

Transmitting Receiving electronics

Analog to digital A/D
electronics sample reflected signal

Transmitting Receiving
antenna antenna

Figure 2 Simplified diagram for an impulse radar system. Modified from Koppenjan, S., 2009. Ground-penetrating radar systems and design.
In: Jol, H.M. (Ed.), Ground Penetrating Radar Theory and Applications. Elsevier, Amsterdam, pp. 73–97.

(a) (b) (c)

Figure 3 Common types of commercially available GPR antennas used for field surveys: (a) rough terrain antennas 100 MHz unshielded, (b)
shielded bistatic 100 MHz, and (c) unshielded ski type 200 MHz antennas.

antennas, monostatic antennas, shielded/unshielded antennas, researchers. The disadvantages with these types of antennas is
and rough terrain antennas (RTA) (Figure 3). Monostatic an- that they are unshielded so that they are subject to ‘noise’ from
tennas have a single antenna that transmits and receives. Ex- external sources of EM radiation such as very high frequency
amples of unshielded bistatic antennas are shown in (VHF) radio waves as well as reflections from above the ground
Figure 3(c), there are two antennas which look like skis, one (see Section 14.16.11). In addition, the antennas are generally
transmitting and the other receiving, which are moved across connected to the console via fiber optic cables, the fiber-optic
the ground at a constant separation in steps that can be deter- cables improve the signal-to-noise ratio but they are fragile and
mined by the user. Strictly speaking this configuration is a easily damaged in the field, making this type of system less
common offset survey with the antennas arranged per- robust. It may also require more man-power to operate in dif-
pendicular broadside (Figure 4). These types of antennas rest ficult terrain with separate operators required to handle the
directly on the ground, are relatively lightweight, easy to antennas, guide the fiber-optic cables as well as the console, for
transport, and can be set up in a variety of configurations so example, Jol and Bristow (2003).
that they can be deployed in parallel or perpendicular broad- Shielded antennas may be monostatic or bistatic and
side, endfire, or cross polarized (Figure 4). In addition, they can housed within a case which usually incorporates a Faraday
be deployed with increased separation for common midpoint cage or absorbing material that helps to reduce interference
(CMP) surveys that are used to determine the velocity of the from external sources of EM radiation. The advantages are
radar signal through the ground. The light weight and flexibility that the shielded antennas are generally more robust
of these types of antennas, and the ability to calculate radar (Figure 3(b)), and can be towed across the ground more easily
velocity from CMP surveys has made them popular with than the unshielded ski type of antennas that need to be
186 Ground Penetrating Radar

Perpendicular broadside Parallel broadside Parallel endfire Cross polarization

(c) (a) (e)
(b) Perpendicular endfire
Survey direction (d)

Antenna
Figure 4 Configurations of radar antennas oriented with respect to the survey direction marked by an arrow: (a) parallel broadside, (b) parallel
endfire, (c) perpendicular broadside, (d) perpendicular endfire, and (e) cross polarized.

modified or attached to some sort of cart or sledge to be towed mean zero level. Dewow is generally an automatic function on
across the ground. The use of shielded antennas is advised for most radar systems. Time-zero correction picks the first arrival
areas with high levels of EM noise such as urban areas or along of the direct air wave arrival at the receiver to correct for drift
highways. However, the shielding adds bulk and weight to the due to instability in the electronics or small changes in an-
antennas and although this is not a significant problem for tenna configurations or air gaps where the antennas are not
higher-frequency antennas that are relatively small, it does directly on the ground. Filters can be applied that average
become an increasing problem with lower-frequency antennas down the trace (temporal) or from trace to trace (spatial) with
as the mass and volume of the antennas become increasingly the effect of smoothing the data and removing high-frequency
difficult to transport and deploy in the field. At lower noise. These filters should be used with discretion because
frequencies there are RTA (Figure 3(a)). excessive averaging between traces or down the trace will
As the name implies, the RTA are designed to be dragged smooth the data. Additional high-pass filters allow the higher-
across the ground over rough terrain or through scrub. RTA frequency components to pass, removing low-frequency
antenna are arranged end to end and joined together in par- components. In contrast low-pass filters can be used to remove
allel endfire configuration (Figure 4) that is known colloqui- high-frequency noise while allowing the lower-frequency
ally as a GPR snake. The advantages of this type of antennas is component to pass. High-pass and low-pass filters can be
that they are more robust, easier to use single-handed in the combined in band-pass filters that let through frequencies
field and relatively quick to cover the ground. However, on the either side of the peak frequency of the transmitted radar
down-side they are only available as low-frequency antennas signal. In order to convert the GPR profiles from the time
(currently 50 and 100 MHz) with a relatively long wavelength domain in which the data are recorded to depth beneath the
and, due to the endfire configuration, and wide separation of surface requires the velocity. Estimate of the average subsurface
the antennas, the direct air- and ground-waves cover a sig- velocity can be extracted from ground truthing, collecting data
nificant section of the shallow subsurface making them more above an object of known depth, or more commonly col-
suitable to geological investigations where the depth of lection of CMP surveys. In addition, velocities can be esti-
interest is 10 m or more. In addition they are liable to roll mated from matching modeled hyperbolic curves to
sideways if dragged across steep slopes which can twist cables diffractions or hyperbolas in the observed data. Field data will
and damage connections so they are better dragged straight up require static corrections for elevation changes along the
or down slopes rather than across slopes. profile. Topographic data should be collected in the field at the
same time as the GPR profiles are collected using levels, laser
levels, total stations, global positioning system (GPS) or dif-
14.16.4 Processing ferential global positioning system (DGPS) equipment. Gains
are applied to the GPR data to enhance the appearance of later
Some form of signal processing will almost certainly have to arrivals and compensate for the effects of signal attenuation
be applied to GPR data and many functions can be performed and geometrical spreading loss. There are a variety of gain
during data processing (Figure 5). However, there is an old functions but the most common are constant, spherical and
adage that should be remembered ‘rubbish in equals rubbish exponential compensation (SEC) and automatic gain control
out.’ No amount of processing can compensate for poor data (AGC). Constant gain applies a multiplier to the data en-
collection in the first place. If the data collected in the field is hancing the signal and usually retaining attributes such as
not good, all the processing in the world will not turn it into amplitude information. SEC gains correct the signal for geo-
good data, and there is a lot to be said for the use of minimum metrical spreading effects. AGC enhances relatively weak, later
processing in order to retain the integrity of the data. With this arrivals, and is widely used as an aid to interpretation. Mi-
proviso in mind there are some basic processing steps that gration to collapse hyperbolas and correct the position of
should be followed and the reader is directed to Cassidy dipping reflections is strongly recommended by Neal (2004)
(2009b) for further details. The first step is to edit the data for GPR profiles in sediments but Cassidy (2009b) urges
files, removing bad races or start positions if files need to be caution in applying migration because it can generate artifacts
merged together to form longer lines. Dewow filters remove a that could be misinterpreted. It is important to compare the
low-frequency ‘wow’ and DC signal, reducing the data to a profiles before and after each step in the processing to check
 Ground Penetrating Radar 187

Data processing in
GPR data acquisition
the field commonly
automated Check data quality
Adjust acquisition setup
CMP data collection
-Editing Initial interpretation
-Dewow
Topographic survey data -Data display and gains

Postcollection data
processing

-Editing
-Rubber banding
-Dewow
-Time zero correction
-Filtering Interpretation
-Deconvolution
-Velocity analysis
-Migration
-Elevation correction
-Depth conversion
-Gains
-Data display

Figure 5 Shaded box shows typical data collection including three sets of data: the GPR survey data, a common midpoint (CMP) survey for
computation of velocity, and topographic survey data. Display of GPR data in the field provides an opportunity for initial interpretation as well
as a check on data quality allowing adjustment of acquisition parameters or systems set-up, for example, changing antennas, step size, or time
window. The unshaded section shows more advanced processing steps that can be undertaken in the lab or office to produce a final
interpretation using the processed CMP data to calculate the velocity for depth conversion and incorporating the topographic survey data for
static elevation correction. Modified from Cassidy, N.J., 2009b. Ground-penetrating radar data processing, modeling and analysis. In: Jol, H.M.
(Ed.), Ground Penetrating Radar: Theory and Applications. Elsevier, Amsterdam, pp. 141–176.

that the profile is being enhanced by processing and to be DGPS, and is used for static corrections of the surface ele-
aware that processing can generate artifacts. vation along the line of section, effectively raising or lowering
the time-zero position where a recorded signal starts to correct
for undulations in the surface topography. Further details with
14.16.5 Survey Design practical guidelines for survey design and data collection can
be found in Jol and Bristow (2003).
GPR surveys need to be designed to collect data that is re-
quired to answer the research question. A prerequisite is to
determine the likely size, shape, and depth of the survey target. 14.16.6 Radar Profiles as Cross-Sections and
The depth to the target is most important because, given the Ground Truth
trade-off between range and resolution (see Section 14.16.2),
this will constrain the choice of antennas. The size and shape Some of the earlier GPR stratigraphic studies were conducted in
of the target will constrain the survey size and density of sand and gravel pits where it was possible to ground truth the
sampling, both step size along GPR profiles as well as the GPR profiles by comparison with exposed sections (Bristow,
spacing between survey lines. In some cases a simple two di- 1994, 1995a; Huggenberger, 1993; Jol et al., 2003) or excavated
mensional (2-D) cross-section may be sufficient; in others a trenches (Bristow et al., 1996). Field observations showed that
full 3-D survey might be required, or where time is limited GPR is able to image primary sedimentary structures, most
some compromise of intersecting survey lines or survey grid likely due to changes in moisture content which were associated
can be used. A CMP survey should be conducted at each lo- with changes in grain size where fine grained sediments have
cation where a continuous horizontal reflection can be iden- slightly higher water saturation than coarser grained sands and
tified in the GPR profile. CMP data are required for the it was these minor changes in water that controlled the di-
computation of radar velocity through the ground, which is electric properties of the sediments (Bristow et al., 1996). Fur-
used to convert GPR profiles from twt in nanoseconds (ns) to ther detailed body of work by Remke van Dam (van Dam and
depth in meters. Velocities can also be determined by curve- Schlager, 2000; van Dam et al., 2002a, b, 2003) investigated the
matching, parabolic reflections but these may not always be effects of iron minerals (van Dam et al., 2002a), organic matter
present on GPR profiles. It is also essential to collect topo- (van Dam et al., 2002b), and determined that water content
graphic elevation data along the GPR survey lines. Topography was the most significant control of dielectric properties in
can be measured using a level, laser level, total station, GPS or sediments (van Dam et al., 2003).
188 Ground Penetrating Radar

14.16.7 Radar Facies 14.16.8 Radar Stratigraphy

Classifying GPR profiles on the basis of reflection patterns, The concepts of radar stratigraphy were introduced by Beres and
radar facies, is one method for interpreting GPR profiles. A Haeni (1991) and Jol and Smith (1991). The terms radar se-
radar facies can be defined as a mappable, three-3-D sedi- quence and radar sequence boundary were introduced by
mentary unit composed of GPR reflections whose parameters Gawthorpe et al. (1993) and are based on the terminology de-
differ from adjacent units (definition modified from Mitchum veloped for seismic interpretation by Mitchum (1977). Radar
(1977), by Jol and Bristow (2003)). Different authors have stratigraphy is similar to seismic stratigraphy but it is at a much
identified a range of radar facies, van Overmeeren (1998) higher resolution (10’s of cm instead of 10’s of m). Following
identified combinations of inclined, parallel continuous, and seismic stratigraphic interpretation principles, it is necessary to
undulating reflections within aeolian dune sand. And other identify reflection terminations to identify radar sequence
authors have developed their own radar facies interpretation (Gawthorpe et al., 1993). The identification of reflection ter-
for fluvial (Best et al., 2003; Huggenberger, 1993), and coastal minations is the basis for constructing a relative chronology
sediments (van Heteren et al., 1998). However, there are because reflections are regarded as isochronous surfaces (Vail
limitations to this approach because first, radar facies are not et al., 1977), and terminations or truncations of reflections
unique to a specific sedimentary structure (Jol and Bristow, represent breaks in time, chronostratigraphic gaps. Successive
2003), and second, a 3-D object like a set of trough cross- radar sequences can be used to construct a relative chronology
strata can appear as a different shaped reflection on a following the laws of superposition and cross-cutting relations.
GPR profile depending on orientation of the cross-strata and Radar stratigraphy is a very useful approach to interpreting
direction of the GPR profile. Inclined reflections in the shallow stratigraphy and the generation of a relative chronology
dip direction can be trough shaped in the strike direction from GPR profiles can be used to optimize the collection of
(see Figures 3, 5, and 6 in Bristow et al., 1996). This reinforces samples for relatively expensive drilling and dating campaigns.
the requirement that radar facies should be defined as 3-D This interpretation methodology has been applied to radar
objects and not just 2-D reflection patterns as is commonly profiles across dunes in the Namib Sand Sea resulting in a re-
the case. construction of dune migration (Bristow et al., 2005) (Figure 6).

Distance (m)
30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240
20 Profile SD1 20
Elevation (m)

Elevation (m)
15 15
10 10
5 5
20 20
Elevation (m)

Elevation (m)
15 15
10 10
5 5
20 20

Elevation (m)
Elevation (m)

21 22
13
15 16 4
15
11 10 8 7 6 5
19 15 14 12 3
10 20
18 17 9
2 1 10
5 Base of dune Bounding surfaces Reflection truncation Borehole Sample point 5
Dune
Face

Profile location 50 m 84 m 139 m 234 m

Distance 34 m 55 m 95 m
OSL age 0.34 ± 0.02 ka 0.99 ± 0.05 ka 1.57 ± 0.07 ka
Calendar age 1662 AD 1012 AD 432 AD
Years 340 yrs 650 yrs 580 yrs
Linear rate 0.1 m yr−1 0.08 m yr−1 0.16 m yr−1
Linear rate 0.12 m yr−1
Linear rate 0.12 m yr−1

Figure 6 Top panel, 100 MHz GPR profile across a sand dune in Namibia with topographic correction applied. Inclined reflections are
interpreted as cross-strata and bounding surfaces within the dune, whereas the continuous reflection at the base of the dune is from the contact
with the underlying bedrock. Middle panel, interpretation of bounding surfaces (red lines) based on truncation of reflections marked by arrows
used to construct a relative chronology of the dune strata. Lower panel, relative chronology of stratigraphic units within the dune defined by
cross-cutting relationships and superposition with borehole locations that were selected for sampling and dating of the dune using optically
stimulated luminescence giving dates from which rates of migration could be calculated. Modified from Bristow C.S., Lancaster, N., Duller,
G.A.T., 2005. Combining ground penetrating radar survey sand optical dating to determine dune migration in Namibia. Journal of the Geological
Society 162(Part 2), 315–322.
 Ground Penetrating Radar 189

In addition, the relative chronology derived from radar profiles sand bars within rivers (Mumpy et al., 2007), or sand dunes
can be used to select optimal locations for boreholes and (Bristow et al., 1996; Pedersen and Clemmensen, 2005).
sample sites for the collection of dates in the same manner that Pedersen and Clemmensen (2005) used a loose grid of 2-D
seismic surveys are used by the oil and gas industry to select sites surveys and interpolated between reflections on the GPR
for drilling just on a much smaller scale. profiles to visualize the shallow subsurface stratigraphy in a
dune field at Thy in Denmark with isopach maps of strati-
graphic units constrained by boreholes.
14.16.9 3-D Date and 2.5D Grids

Unless a GPR profile is oriented perpendicular to the dip 14.16.10 Problems, Pitfalls, and Limitations
direction of strata in the subsurface all inclined reflections on
GPR profiles are apparent dips and not a true representation of Some of the best features of GPR are also its Achilles’ heel.
the depositional dip. In addition, sediments vary laterally and Modern GPR systems are easy to assemble and operate; data
it is commonly the geometry or architecture of sediment de- collection is rapid and provides subsurface data in real time.
posits that can be imaged by GPR that are critical in deter- The relative ease of operating a GPR system and rapid data
mining depositional environments within river channels collection means that inexperienced users have taken GPR into
(Bridge, 2009) coastal sediments (Buynevich et al., 2009). In the field and returned with mixed or poor quality data, due to
order to capture the true depositional geometry, a 3-D survey simple errors in system set-up which compromises the data, for
is required. The advantages of 3-D data sets is that they can be example, maximizing ‘footprint’ instead of minimizing it, or
cropped or sliced to reveal strata at different depths, or rotated more often unrealistic expectations of what GPR can provide. It
to achieve the optimum viewing angle for interpretation. For a is important to realize that a GPR profile is not a picture or a
true 3-D representation the step size between measurements direct image of the subsurface per se, but, the time-dependant,
or trace spacing has to be dense enough to sample entire recorded response of the subsurface materials to the propa-
diffractions (Grasmueck et al., 2003) who recommend a step gation of EM energy across a relatively narrow range of radio
size of a quarter wavelength or less in order to avoid spatial wave frequencies (Cassidy 2009a). GPR works well in soils with
aliasing of steeply dipping reflections. Examples of 3-D surveys a high resistivity but where the ground is conductive such as
include imaging of sand bars in oolitic limestones (Grasmueck areas with saline ground waters as well as some clay soils there
and Weger, 2002) (Figure 7), and cross-strata in modern is significant attenuation and limited depths of penetration.
and ancient desert sand dunes (Bristow et al., 2007; Jol et al.,
2003). The small step size along profiles and the close spacing
of profiles required for true 3-D surveys means that they are 14.16.11 Side Swipes and Airwaves
time consuming to collect and generally cover very limited
areas (e.g., Jol et al. (2003) and Bristow et al. (2007)). For A common misconception of GPR is that the signal is trans-
example, the 3-D data cube in Figure 7, which covers an area mitted directly down into the ground and returns nearly ver-
24 m  46 m, took two people 9 h to collect with a trace every tically back to the receiver on the surface. Unfortunately this is
0.1 m along 121 parallel lines that were spaced 0.2 m apart not the case; the signal spreads out in 3-Ds into the air as well
(Grasmueck and Weger, 2002). In some cases it might not be as the ground. Certain designs of antennas try to focus the radar
time or cost effective to collect sufficient data to generate a true signal and shielding around antennas helps to reduce trans-
3-D model of the shallow subsurface, in which case grids of mission through the air but the signal still spreads out under-
2-D data with perpendicular lines along and across strike, or ground. The spread of the signal through the air or through the
along paths or tracks where access is limited, can be used to ground can result in some spurious interpretations because the
construct 3-D representations of the sediment architecture of radar profile is not an image of the subsurface directly beneath

E dipping crossbeds
of migrating sandwaves N
SSW dipping progradation
sets of the barrier bar

3-D GPR cube

W

Ingraham outcrop
E
Figure 7 100 MHz 3-D data cube 24 m  46 m and a 29 m GPR profile with correlation to an outcrop of cross-stratified oolitic limestone in
Florida. Modified from Grasmueck, M., Weger, R., 2002. 3D GPR reveals complex internal structure of Pleistocene oolitic sandbar. The Leading Edge
21(7), 634–639.
190 Ground Penetrating Radar

Distance (m)
0 10 20 30 40

0 0

50
5

100

10
150

Data spike
15 200
Depth
(m) Two-way traveltime
twt (ns)
(a) Distance (m)
0 20 40 60 80

0 0

50
5

100
10
150

15 200

250
20
Depth Two-way traveltime
(b) (m) (ns)
Figure 8 (a) Parabolic reflections centered around 12 m along the GPR profile with an apparent depth of 6 m are actually airwave reflections
from a tree above ground. This profile has not been processed and includes a data spike that is attributed to a communications error between
the GPR console and the computer. (b) Overlapping parabolic reflections at around 50 m along the GPR profile come from overhead powerlines.
Metallic objects, vehicles, and wire fences can act as secondary antennas producing spurious results.

the line of survey. For example, in Figure 8(a) a parabolic re- fences, or overhead wires from telephone cables or electricity
flection appears on a profile at 12 m and an apparent depth of pylons acting as secondary antennas. Figure 8(b) shows a re-
6 m. But the signal is actually from a tree about 20 m to the east flection pattern which is the product of overlapping parabolic
of the GPR profile and in this example from Australia the radar reflections from overhead electric cables. Other sources of
has picked up reflections from above the ground. The reflection external EM ‘noise’ include motor vehicles as well as microwave
from the tree appears in the radar profile because the antennas and VHF radio transmissions. These can generally be dis-
are not shielded, the radar signal propagates through the air tinguished on radar profiles by their high frequency and pres-
above ground as well as through the ground, and the radar is ence within the record before the first arrival of the GPR signal.
not able to differentiate between a signal that has been reflected
from above or beneath the surface.
In the GPR profile (Figure 8(a)) the tree appears to be at a 14.16.12 Examples: Fluvial Geomorphology
relatively shallow depth because the velocity of radar through
air is close to the speed of light, and faster than the velocity Rivers are generally good environments for GPR surveys be-
through the ground. Objects like trees and buildings above cause the sediments are resistive and river deposits commonly
ground can act as reflectors and a reflection from an object contain fresh water. GPR has been used to image cross-strata,
outside the plane of section above or beneath the ground is sand bars, and scour hollows and has contributed to the de-
known as a side swipe. Another problem is metallic objects, velopment of new facies models for fluvial deposits (e.g.,
 Ground Penetrating Radar 191

Gawthorpe et al., 1993; Huggenberger, 1993; Bridge et al., are primarily related to differences in water saturation. Be-
1995; Best et al., 2003; Skelly et al., 2003; Lunt et al., 2004; neath the water table the sediments are saturated but water
Bridge, 2009). It has proved particularly useful for imaging the content still varies with porosity, the space between the grains.
deposits of recent and Quaternary river sediments above and In both situations it is water content that is the primary con-
below the water table. Above the water table radar reflections trol on the dielectric permittivity of the sediments and changes

GPR grid

Flow

(a)
Distance (m)
Flow
0 10 20
0 0
X horizon
1
50
2
Z horizon

100 3

4
150
5
(b) Two-way traveltime (ns)
Depth
(m)
X horizon Z horizon

Flow

(c) (d)

Figure 9 (a) Oblique aerial photograph of the Niobrara River near Niobrara Town in Nebraska, view toward the north. A 20 m  20 m grid of
200 MHz GPR profiles was collected on one of the sand bars and used to investigate the architecture of shallow sand-bed braided rivers by
Skelly et al. (2003). (b) 200 MHz GPR profile along the sand bar showing sets of cross-strata and bounding surfaces. The morphology of the
sand bars and scour surfaces defined by the blue reflection (X horizon) and the green reflection (Z horizon) are shown as oblique elevation
models (c) and (d), respectively.
192 Ground Penetrating Radar

in sedimentary fabrics, grain size, and sorting, associated with central frequency of 100 MHz will be 0.6 m, with resolution at
primary sedimentary structures that are imaged by GPR. The 1/2–1/4 of the wavelength of 0.3–0.15 m. The lower velocity
water table, and the capillary zone above the saturated zone in water saturated sands gives a correspondingly shorter
create significant changes in dielectric permittivity and often wavelength and increased resolution. In the example illus-
form a continuous, horizontal, high amplitude reflection on trated in Figure 9 from the Niobrara River in Nebraska the
GPR profiles. Although the water table forms a high amplitude water table is at or close to the surface. A detailed 3-D survey
reflection this does not mean that one cannot use GPR to with 200 MHz antennas images cross-strata, sand bars, and
image beneath the water table. In fact imaging down from the scour surfaces showing the architecture of the shallow sand-
water table gives very good results. Above the water table dry bed braided river (Figure 9).
or damp sands typically have velocities of 0.15–0.12 mns 1,
with this range of velocities the wavelength of GPR with a
central frequency of 100 MHz will be 1.5–1.2 m, giving a 14.16.13 Sand Dunes
resolution at 1/2–1/4 of the wavelength of 0.75–0.3 m. Be-
neath the water table the velocity in saturated sands is typically GPR can be used to produce 3-D visualizations of the sedi-
0.06 mns 1; with this velocity the wavelength of a GPR with a mentary structures within sand dunes, and from these

N
VD3
LVD3

VD2

VD1
LVD5

GPR profile
OSL sample

(a) (b)

Northwest Distance (m) Southeast

320 300 280 260 240 220 200 180 160 140 120 100 90

0 GPR profile VD2 60

50
250 40
30
500 20
10
750 0
−10
1000
twt Depth
(ns) (m)

320 300 280 260 240 220 200 180 160 140 120 100 90

Inclined reflections dipping Unconformity

0 towards the northwest Inclined reflections dipping 60
towards the southeast 50
250 40
30
500 20
10
750 0
−10
1000
twt Base of dune Depth
(ns) (m)
(c)

Figure 10 (a) Photograph of data collection in Antarctica with unshielded 100 MHz antennas on the GPR and a total station to measure the
topography. (b) Lidar image of the Lake Vida dune showing location of GPR profiles. (c) 100 MHz GPR profile VD2 showing reflections from
strata within the dune to depths of around 70 m which is exceptional. The dipping reflections record the migration of the dune from west to east
with a recent reversal at the crest.
 Ground Penetrating Radar 193

structures it is possible to reconstruct the history of dune de- Bristow, C.S., Jol, H.M., 2003. Ground Penetrating Radar in Sediments. The
velopment and migration (e.g., Bristow et al. 2000a, 2005, Geological Society, London, Special Publication 211, 330 pp.
Bristow, C.S., Lancaster, N., Duller, G.A.T., 2005. Combining ground penetrating
2007). Dune sands are suitable targets for GPR surveys because
radar surveys and optical dating to determine dune migration in Namibia.
they generally have low conductivity and low magnetic per- Journal of the Geological Society 162(Part 2), 315–322.
meability allowing good depths of penetration (low attenu- Bristow, C.S., Pugh, J., Goodall, T., 1996. Internal structure of aeolian dunes in
ation). The depth of penetration of GPR in dune sands can Abu Dhabi determined using ground penetrating radar. Sedimentology 43,
exceed 25 m, which is deeper than most trenches and a recent 995–1003.
Buynevich, I.V., Jol, H.M., FitzGerald, D.M., 2009. Coastal environments. In: Jol,
survey of sand dunes in Antarctica reached depths of around H.M. (Ed.), Ground Penetrating Radar: Theory and Applications. Elsevier,
75–80 m (Figure 10). Before the advent of GPR the internal Amsterdam, pp. 299–322.
structure of sand dunes was investigated from trenches dug by Cassidy, N.J., 2009a. Electrical and magnetic properties of rocks, soils and fluids.
hand (e.g., McKee and Tibbitts, 1964), or mechanical excav- In: Jol, H.M. (Ed.), Ground Penetrating Radar: Theory and Applications. Elsevier,
Amsterdam, pp. 41–72.
ation (McKee, 1966, 1979), or extrapolated from observations
Cassidy, N.J., 2009b. Ground-penetrating radar data processing, modeling and
of partially deflated dune surfaces (Tsoar, 1982). Digging or analysis. In: Jol, H.M. (Ed.), Ground Penetrating Radar: Theory and Applications.
mechanical excavation of trenches through dunes are effective Elsevier, Amsterdam, pp. 141–176.
but destructive, and the depth of investigation is limited by the van Dam, R.L., Schlager, W., 2000. Identifying causes of ground-penetrating radar
stability of dry or slightly damp sand. Furthermore, the ex- reflections using time-domain reflectometry and sedimentological analyses.
Sedimentology 47, 435–449.
cavation of trenches in unconsolidated sands can be danger-
van Dam, R.L., Schlager, W., Dekkers, M.J., Huisman, J.A., 2002a. Iron oxides as a
ous due to the risk of the trench walls collapsing. In contrast, cause of GPR reflections. Geophysics 67, 536–545.
GPR is a noninvasive technique that can be used to image the van Dam, R.L., Schlager, W., Dekkers, M.J., Huisman, J.A., 2002b. Influence of
internal structure of dunes leaving no more than a set of organic matter in soils on radar-wave reflection: sedimentological implicatations.
footprints on the dune surface. GPR offers a fast and efficient Journal of Sedimentary Research 72, 341–352.
van Dam, R.L., Van Den Berg, E.H., Schaap, M.G., Broekema, L.H., Schlager, W.,
method for the collection of high resolution, almost con- 2003. Radar reflections from sedimentary structures in the vadose zone.
tinuous images of the shallow subsurface structure in aeolian In: Bristow, C.S., Jol, H.M. (Eds.), GPR in Sediments. Geological Society,
sands and has proven to be very useful in the interpretation London, Special Publication 211, pp. 257–273.
of desert and coastal sand dunes. GPR surveys of sand dunes Gawthorpe, R.L., Collier, R.E.L., Alexander, J., Bridge, J.S., Leeder, M.R., 1993.
Ground penetrating radar: application to sandbody geometry and heterogeneity
in Antarctica have proven successful especially imaging
studies. In: North, C.P., Prosser, D.J. (Eds.), Characterisation of Fluvial and
cross-strata to depths of around 80 m (Bristow et al., 2010) Aeolian Reservoirs. The Geological Society, London, Special Publication 73,
(Figure 10). GPR works particularly well in ice and frozen pp. 421–432.
sediments because water molecules form ionic bonds when Grasmueck, M., Weger, R., 2002. 3D GPR reveals complex internal structure of
frozen which prevents molecular rotation, which reduces Pleistocene oolitic sandbar. The Leading Edge 21(7), 634–639.
Grasmueck, M., Weger, R., Horstmeyer, H., 2003. How dense is dense enough for
losses, but results like this should not be anticipated a‘real’ 3D GPR survey? Society of Exploration Geophysicists 73rd International
elsewhere. Annual Meeting Extended Abstracts, 1180–1183.
van Heteren, S., Fitzgerald, D.M., McKinley, P.A., Buynevich, I.V., 1998. Radar facies
of paraglacial barrier systems: coastal New England, USA. Sedimentology 45,
181–200.
References Huggenberger, P., 1993. Radar facies: recognition of facies patterns and
heterogeneities with Pleistocene Rhine gravels, NE Switzerland. In: Best, J.L.,
Annan, A.P., 2002. GPR-History, trends, and future developments. Subsurface Bristow, C.S. (Eds.), Braided Rivers. The Geological Society, London, Special
Sensing Technologies and Applications 3, 253–270. Publication 75, pp. 163–176.
Annan, A.P., Davis, J.L., 1976. Impulse radar sounding in permafrost. Radio Jol, H.M. (Ed.), 2009. Ground Penetrating Radar: Theory and Applications. Elsevier,
Science 11, 383–394. Amsterdam, 524 p.
Bailey, J.T., Evans, S., Robin, G de Q., 1964. Radio echo sounding of polar ice Jol, H.M., and Bristow, C.S., 2003. GPR in sediments: advise on data collection,
sheets. Nature 204, 420–421. basic processing and interpretation, a good practice guide. In: Bristow, C.S.,
Baker, G.S., Jol, H.M., 2007. Stratigraphic analyses using GPR. Geological Society Jol, H.M. (Eds.), Ground Penetrating Radar in Sediments. The Geological
of America Special Paper 432, 181. Society, London, Special Publication 211. pp. 9–27.
Beres, M., Haeni, F.P., 1991. Application of ground-penetrating-radar methods in Jol, H.M., Bristow, C.S., Smith, D.G., Junck, M.B., Putnam, P., 2003. Stratigraphic
hydrogeologic studies. Ground Water 29, 375–386. imaging of the Navajo Sandstone using ground-penetrating radar. The Leading
Best, J.L., Ashworth, P.J., Bristow, C.S., Roden, J., 2003. Three dimensional Edge 22(9), 882–887.
sedimentary architecture of a large, mid-channel sand braid bar, Jamuna River, Jol, H.M., Smith, D.G., 1991. Ground penetrating radar of northern lacustrine
Bangladesh. Journal of Sedimentary Research 73, 516–530. deltas. Canadian Journal of Earth Sciences 28, 1939–1947.
Bridge, J., 2009. Advances in fluvial sedimentology using GPR. In: Jol, H.M. (Ed.), Koppenjan, S., 2009. Ground-penetrating radar systems and design. In: Jol, H.M.
Ground Penetrating Radar: Theory and Applications. Elsevier, Amsterdam, pp. (Ed.), Ground Penetrating Radar Theory and Applications. Elsevier, Amsterdam,
323–359. pp. 73–97.
Bridge, J.S., Alexander, J., Collier, R.E.L., Gawthorpe, R.L., Jarvis, J., 1995. Lunt, I.A., Bridge, J.S., Tye, R.S., 2004. A quantitative, three-dimensional
Ground-penetrating radar and coring used to document the large-scale depositional model of gravely braided rivers. Sedimentology 51, 377–414.
structureog point bar deposits in 3-D. Sedimentology 42, 839–852. McKee, E., Tibbitts, Jr. G.C., 1964. Primary structures of a seif dune and associated
Bristow, C.S., 1994. A new look at the Lower Greensand using ground-penetrating deposits in Libya. Journal of Sedimentary Petrology 34, 5–17.
radar. Geology Today 10, 24–27. McKee, E.D., 1966. Structures of dunes at White Sands National Monument, New
Bristow, C.S., 1995a. Facies analysis in the Lower Greensand using ground Mexico (and a comparison with structures of dunes from other selected areas).
penetrating radar. Journal of the Geological Society, London 152, 591–598. Sedimentology 7, 1–69.
Bristow, C.S., Augustinus, P.C., Wallis, I.C., Jol, H.M., Rhodes, E.J., 2010. McKee, E.D., 1979. A study of Global Sand Seas. US Geological Survey
Investigation of the age and migration of reversing dunes in Antarctica using Professional Paper, 1052.
GPR and OSL, with implications for GPR on Mars. Earth and Planetary Science Mitchum, Jr. R.M., 1977. Part eleven: glossary of terms used in seismic
Letters 289, 30–42. stratigraphy. In: Payton, C.E. (Ed.), Seismic Stratigraphy Applications to
Bristow, C.S., Duller, G.A.T., Lancaster, N., 2007. Age and dynamics of linear dunes Hydrocarbon Exploration. The American Association of Petroleum Geologists,
in the Namib desert. Geology 35, 555–558. Tulsa, OK, Memoir 26, pp. 205–212.
194 Ground Penetrating Radar

Morey, R.M., 1974, Continuous subsurface profiling by impulse radar. Skelly, R.L., Bristow, C.S., Ethridge, F.G., 2003. Architecture of channel-belt
Proceedings, Engineering Foundations Conference on Subsurface Exploration deposits in an aggrading shallow sandbed braided river: the lower Niobrara
for Underground Excavations and Heavy Construction. Henniker, NH, pp. River, northeast Nebraska. Sedimentary Geology 158, 249–270.
213–232. Stern, W., 1929. Versuch einer elektrodynamischen Dickenmessung von
Mumpy, A.J., Jol, H.M., Kean, W.F., Isbell, J.L., 2007, Architecture and Gletschereis. Gerlands Beitraege zur Geophysik 23, 292–333.
sedimentology of an active braid bar in the Wisconsin River based on 3D Stern, W., 1930. Uber Grundlagen, Methodik und bisherige Ergebnisse
ground penetrating radar. In: Baker, G.S., Jol, H.M. (Eds.), Stratigraphic Analysis elektrodynamischer Dickenmessung von Gletschereis. Zeitschrift fuer
Using GPR: The Geological Society of America, Boulder CO, Special Paper 432, Gletscherkunde 15, 24–42.
pp. 111–131, doi:10.1130/2007.2432(09). Tsoar, H., 1982. Internal structure and surface geometry of longitudinal seif dunes.
Neal, A., 2004. Ground-penetrating radar and its use in sedimentology: principles, Journal of Sedimentary Petrology 52, 823–831.
problems and progress. Earth Science Reviews 66, 261–330. Vail, P.R., Todd, R.G., and Sangree, J.B., 1977. Seismic stratigraphy and global
Olhoeft, G.R., 1981. Electrical properties of rocks. In: Touloukian, Y.S., Judd, W.R., changes in sea level, part 5: chronostratigraphic significance of seismic
Roy, R.F. (Eds.), Physical Properties of Rocks and Minerals. MacGaw-Hill, New reflections. In: Payton, C.E. (Ed.), Seismic Stratigraphy – Applications to
York, vol. II, 548. Hydrocarbon Exploration. The American Association of Petroleum Geologists,
van Overmeeren, R.A., 1998. Radar facies of unconsolidated sediments in The Tulsa, OK, Memoir 26, pp. 99–116.
Netherlands: a radar stratigraphy interpretation method for hydrogeology. Journal Waite, A.H., Schmidt, S.J., 1961. Gross errors in height indication from pulsed
of Applied Geophysics 40, 1–18. radar altimeters operating over thick ice or snow. Proceedings of the Institute of
Pedersen, K., Clemmensen, L.B., 2005. Unveiling past aeolian landscapes: a Radio Engineers 50, 1516–1520.
ground-penetrating radar survey of a Holocene coastal dunefield system, Thy, Walford, M.E.R., 1964. Radio echo sounding through an ice shelf. Nature 204,
Denmark. Sedimentary Geology 177, 57–86. 317–319.

Biographical Sketch

Dr. Charlie Bristow is the assistant dean and head of department at the Department of Earth and Planetary
Sciences, Birkbeck College, University of London. He has a BSc in Geology from the University of Sheffield and a
PhD in the ‘Sedimentology of braided rivers: ancient and modern’ from the University of Leeds. He has more than
20 years of experience in conducting GPR surveys in a wide range of environments including, the Brahamputra
River in Bangladesh and the Niobrara River in Nebraska. He has surveyed sand dunes in the deserts of America,
Africa, Australia, and Antarctica as well as beach ridges in Australia, New Zealand, and the UK and published
extensively on the subject including editing a book on GPR in Sediments with Harry Jol.
 14.17 Electronic Measurement Techniques for Field Experiments in
Process Geomorphology
DJ Sherman and C Houser, Texas A&M University, College Station, TX, USA
ACW Baas, Kings College London, London, UK
r 2013 Elsevier Inc. All rights reserved.

14.17.1 Introduction 196

14.17.2 Monitoring Geomorphic Systems Controlled by Hydrodynamic Processes 197
14.17.2.1 Eularian Flow Measurement 197
14.17.2.1.1 Impeller-type current meters 197
14.17.2.1.2 Electromagnetic current meters 198
14.17.2.1.3 Acoustic Doppler current meters and profilers 199
14.17.2.1.4 Acoustic noise 200
14.17.2.2 Water Level Measurement 201
14.17.2.2.1 Wave staffs and wires 201
14.17.2.2.2 Pressure transducers 201
14.17.2.3 Bed Elevation Change and Bedform Monitoring 201
14.17.2.3.1 Altimeters 201
14.17.2.3.2 Ultrasonic distance sensors 202
14.17.2.3.3 Photoelectronic erosion pins (PEEP) 202
14.17.2.3.4 Side-scan sonars 203
14.17.2.4 Sediment Detection 203
14.17.2.4.1 Optical sensors 203
14.17.2.4.2 Acoustic sensors 205
14.17.2.4.3 Magnetic sensors 205
14.17.2.4.4 Laser in situ scattering and transmissometry 206
14.17.2.4.5 Conductivity concentration profiler 206
14.17.2.4.6 Hydrophones 206
14.17.2.5 Flow and Transport Tracking Approaches 207
14.17.3 Monitoring Geomorphic Systems Controlled by Aeolian Processes 207
14.17.3.1 Airflow Measurement 207
14.17.3.1.1 Mechanical anemometers and vanes 207
14.17.3.1.2 Sonic anemometers 208
14.17.3.1.3 Thermal anemometers 209
14.17.3.2 Transport Measurement 210
14.17.3.2.1 Load cells 210
14.17.3.2.2 Laser and optical sensors 210
14.17.3.2.3 Impact sensors 211
14.17.3.2.4 Soil moisture measurement 211
14.17.3.3 Surface Shear Stress 212
14.17.4 Interpreting the Signal 213
14.17.5 Conclusions 214
References 214

Glossary The greater the dielectric constant of a substance, the greater its
Dielectric constant The dielectric constant is the ratio of the capacitance, or the lower its conductivity.
electromagnetic permittivity of a substance to the permittivity Eulerian The Eulerian perspective on fluid or sediment
of a vacuum. The dielectric constant of a vacuum is therefore 1. transport relies on a constant frame of reference through

Sherman, D.J., Houser, C., Baas, A.C.W., 2013. Electronic measurement

techniques for field experiments in process geomorphology. In: Shroder, J.
(Editor in Chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise on
Geomorphology. Academic Press, San Diego, CA, vol. 14, Methods in
Geomorphology, pp. 195–221.

Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00401-2 195

196 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

which a fluid parcel or sediment particle passes, thus follows a fluid parcel or sediment particle, thus producing
producing time series of flux. transport pathways.
Frequency-domain reflectometry (FDR) FDR uses Time-domain reflectometry (TDR) TDR uses changes in
changes in resonant electromagnetic frequencies to measure travel time of electromagnetic pulses to measure the
the dielectric constant of a soil, which is related to soil dielectric constant of a soil, which is related to soil moisture
moisture content. content.
Lagrangian The Lagrangian perspective on fluid or
sediment transport relies on a frame of reference that

Abstract

The processes that shape the Earth’s surface have a variation or rhythm that can be measured in the field using a number of
instruments and techniques. This chapter describes electronic field equipment that has increased our capacity to investigate
sediment transport and fluid flows in coastal, fluvial, and aeolian environments. This equipment includes: (1) anemometers,
velocimeters, profilers, and hydrophones for Eularian flow measurement; (2) traps, impact sensors, and backscatterance
sensors for sediment transport measurement; (3) drogues, particle image velocimeters, and electronic tracers for measuring
Lagrangian flow and transport; and (4) erosion pins, distance sensors, and sonar altimeters for measuring bed elevation
change. Modern deployment of these instruments and past applications are described with respect to the spatial and temporal
scale of the processes being examined, and the methodological and interpretative limitations inherent to field deployments.

14.17.1 Introduction (e.g., Aagaard and Masselink, 1999), the processes of greatest
importance for Earth surface forms either occur over an ex-
The geomorphologist has always relied on observation, simple ceptionally long period and spatial scale (Figure 1(a)) or at
measurement, deduction, and intuition to explain the devel- scales much smaller than visible with the naked eye
opment and evolution of surface features on the Earth (e.g., (Figure 1(c)). In addition to the ability to extend our obser-
Gilbert, 1886; Davis, 1909; Sauer, 1930). For example, Corn- vational range, the instrumented record is able to reveal sec-
ish (1934) developed an observational convention of signifi- ondary processes or secondary scales of primary processes that
cant wave height that relates the distribution of wave height to are not readily apparent to the most careful visual observations
the height most apparent to the human eye – the significant (Figure 2). These advances have not only improved our ability
wave height. Visually obvious descriptions of Earth surface to interpret the landscape through more robust links between
processes were sufficient for many applications and explan- process and form, but they have also strengthened the theory-
ations of a landscape, and the technique was common practice driven, empirical basis of geomorphology.
in geomorphology until the late 1900s. However, recent ad- The purpose of this paper is to describe several modern
vances in process geomorphology have only been possible instrumentation systems, to present examples of their appli-
through improvements and advances in the technology used cations and operations, to examine the methodological impli-
to measure the processes responsible for sculpting and re- cations of deploying instruments in the field, to discuss how
shaping the landscape. In large part, this new and improved these measurements contribute to our understanding of geo-
technology has allowed field scientists to measure processes at morphic environments, and to describe some of the analytical
increasingly smaller and larger spatial and temporal scales, approaches to time-series data analysis. This discussion is lim-
which in turn requires greater data storage capacity and better ited to instrument systems that include a sensor capable of
analysis techniques. Variations in the strength and in some producing an electronic signal and a digital data recorder, with
cases the direction of a process are recorded as a voltage or an emphasis on aeolian, fluvial, and coastal sediment systems.
pulsed signal that can be calibrated in the laboratory to a Our focus is restricted to instruments commonly used in field
known speed, direction, concentration, transport rate, ele- applications. We do not consider other important technologies
vation, or depth depending on the specific instrument. such as airborne or satellite remote sensing, geographical pos-
Once measured and recorded, the frequency and magnitude itioning systems, ground penetrating radar, ground-based
of the variations and the presence of trends can be explored LiDAR, and particle image velocimetry (Baas and Van den Berg,
through graphical representations and statistical analyses. 2010). Many of the sensors discussed herein (such as pitot
Figure 1 provides sample time series data collected from a tubes, anemometers, etc.) have a long history of use and de-
number of environments at increasingly smaller scales from velopment in geomorphology. Until about 1980, the output of
monthly regional current data collected along the mesotidal these systems was recorded on paper strip charts, audio or
American reef in southern Belize (Houser, unpublished data; video tape, or recorded manually from visual displays. For
Figure 1(a)) to high-frequency aeolian sediment transport data those lucky few, these instruments might have been cabled to
collected in Jericoacoara, Brazil (Figure 1(c)). With the unwieldy digital acquisition and recording systems located at a
exception of the wave data (Figure 1(b)), for which visual considerable distance from the sensor (e.g., Greenwood and
observations might be similar to those derived statistically Sherman, 1984). These systems also required the scientist to be
 Electronic Measurement Techniques for Field Experiments in Process Geomorphology 197

30

Current velocity (cm s–1)

20 North
10
0
–10
–20
–30
–40
–50 South
–60
0 1000 2000 3000 4000 5000 6000 7000
(a) Time (hr)

0.80
0.70
Water depth (m)

0.60
0.50
0.40
0.30
0.20
0.10
0.00
0 50 100 150 200 250
(b) Time (s)

0.30
0.25
Impact force (volts)

0.20
0.15
0.10
0.05
0.00
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
(c) Time (s)
Figure 1 Sample time series collected by electronic field equipment at three different scales. Shown are time series of: (a) north–south current
velocity along a barrier reef in southern Belize collected using an S4 current meter (Houser, unpublished data), (b) nearshore water depth and
waves measured at Burley Beach, Lake Huron (Houser and Greenwood, 2005), and (c) impact force of saltating grains measured at Jericoacoara,
Brazil (Sherman and Li, unpublished data).

able to develop custom computer/data acquisition programs, difficulties inherent in measuring directly the spatial dynamics
and the lack of a monitor made in situ trouble shooting and of either processes or sediment pathways. As a consequence,
preliminary evaluation of data in the field impractical. Only most instrumentation systems used in the field are designed
with the advent of the personal computer, low-cost and robust for the Eulerian perspective rather than the Lagrangian. We
data acquisition systems, and packaged software programs have organized this section around those two approaches and,
could the field geomorphologist measure and troubleshoot in within those sections, by the aspect of the system being
real or near-real time. This represented a fundamental change measured. In a general sense, we discuss instruments from
in the geomorphologist’s ability to scientifically examine how simplest to most complex, or the related technologies from
the Earth’s surface developed and evolved. This change, how- oldest to newest.
ever, was not without its own limitations.

14.17.2.1 Eularian Flow Measurement

14.17.2 Monitoring Geomorphic Systems Controlled 14.17.2.1.1 Impeller-type current meters
by Hydrodynamic Processes The ability to measure flow velocity at high spatial and tem-
poral resolutions and in three dimensions has been the basis
Hydrodynamic processes and related sediment transport are for significant advancements in our understanding of coastal
generally measured at a point, or series of points, to generate a and fluvial sediment transport and morphological change.
times series. This Eulerian approach is necessitated by the However, early sensors, such as the impeller-type (or propeller
198 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

1.24
1.22
1.20

Water depth (m)

1.18
1.16
1.14
1.12
1.10
1.08
1.06
1.04
0 50 100 150 200 250 300 350 400 450 500
(a) Time (s)

0.003
Infragravity waves
Gravity waves (<20 s)
Spectral density (m2ΔF)

(>20 s)
0.002 0.007 Hz
(143 s)
0.002 0.061 Hz
0.035 Hz (16 s)
(28 s)
0.001 0.072 Hz
(14 s)

0.001

0.000
0 0.05 0.1 0.15 0.2 0.25
(b) Frequency (Hz)
Figure 2 Time series of water surface elevation (a) and the associated spectral density plot showing the record to be composed of both
primary (high-frequenc) and secondary (low-frequency) components (b). Data sampled from behind a barrier reef in southern Belize using an
RBR pressure transducer sampling at 4 Hz (Houser, unpublished data).

or cup) current meters (see Hjulström, 1935), are only able to used extensively in coastal geomorphology. Based on Faraday’s
measure flow speed at a single point and provide no infor- law of electromagnetic induction, electrodes on the surface of
mation about flow direction. Despite these limitations, im- the probe measure the deformation of a controlled electro-
peller-type current meters are still used by the U.S. Geological magnetic field by the flow around the discoidal or spherical
Survey (USGS) as the standard for streamflow measurement sensor head (see Aubrey and Trowbridge, 1985 or Lane et al.,
(Hubbard et al., 2001). Similar to three-cup anemometers 1993). Sherman et al. (1993) used a single current meter to
used for measuring wind speed, these current meters consist of monitor current velocities associated with bedform migration
three or more cups (or a propeller), which rotate at a speed in a rip channel, whereas Houser and Greenwood (2005) used
proportional to the flow. An electronic contact is tripped with an array of ECMs to describe oscillatory and quasisteady
each revolution and the number and the time between signals components of flow responsible for rapid landward migration
is used to calculate the flow speed. These sensors are not able of a swash bar. Used with tracer-tagged sediments, the current
to instantaneously adjust to changing flow speeds due to the meters have been used to estimate longshore sediment trans-
inertia of the rotating assembly, and due to their size cannot port (Nordstrom et al., 2003) and the vertical structure of flow
be deployed near the bed or other instruments. Based on an and sediment transport (Osborne and Greenwood, 1993). The
early design by Woods (1968), bidirectional ducted flow lack of moving parts and generally rugged design allow the
meters (using propellers), capable of measuring orbital vel- instrument to be deployed for extended periods without
ocities and turbulent fluctuations on the order of 1 Hz, have mechanical failure, although biofouling common to such long
been developed for marine and lacustrine systems (Nielsen deployments may significantly degrade the signal. Fedderson
and Cowell, 1981; Brander and Short, 2000, 2001). However, and Guza (2003) used data from a two-dimensional (2D)
ducted flow meters are not truly bidirectional (measuring in array of 25 ECMs deployed at the U.S. Army Corps of Engin-
both x and y axes), and two current meters need to be installed eers Field Research Facility at Duck, NC, for almost four
side by side and in opposite direction to measure flow speeds months to obtain measurements of along and across shore
along the two axes. flows to describe nearshore circulation. The sensors have also
been used in fluvial environments to measure macro-
14.17.2.1.2 Electromagnetic current meters turbulence generated by bedforms (Kostaschuk and Church,
Impeller-type current meters have been largely abandoned by 1993; Kostaschuk and Villard, 1996; Carling et al., 2000) and
academic field scientists in favor of more sophisticated elec- flow dynamics at river confluences (Best and Roy 1991), and
tromagnetic and acoustic sensors capable of rapid and to assess coarse bedload transport (Whitaker and Potts, 2007)
multidirectional flow measurements. Electromagnetic current and the reach-scale variability of turbulent flow (Lamarre and
meters (ECMs), capable of measuring two orthogonal com- Roy, 2008). Example data from an Interocean S4 current meter
ponents of the flow at sample frequencies 41 Hz, have been deployed off the coast of Belize are provided in Figure 1(a).
 Electronic Measurement Techniques for Field Experiments in Process Geomorphology 199

14.17.2.1.3 Acoustic Doppler current meters and profilers Differential Global Positioning System (DGPS) to increase spa-
Acoustic Doppler current meters, sometimes referred to as tial coverage (Figure 3(b)) and to complete sample transects in
ultrasonic sensors, emit a focused acoustic beam that reflects little time (e.g., Kostaschuk et al., 2005). In another recent ap-
off sediment and bubbles in the water. Operating on the plication, a downward looking ADCP was used to characterize
principal of Doppler shift, the sensor measures the shift in the influence of large dunes in the Fraser River on mixing with
acoustic frequency when a sound source and receiver are in the migration of a salt wedge through the tide. As the salt wedge
motion relative to each other. Sound reflected from an object migrates into the dune field, a large internal wave develops over
moving away from the sensor will have a lower frequency the largest dunes, lifting the salt wedge to the water surface.
(compared to the sound first emitted by the sensor), whereas Because a downward looking ADCP may not accurately detect
objects moving toward the instrument will reflect back a the motionless bed through the moving sediment, errors can
higher frequency sound. The velocity is calculated from the arise in computing channel-scale properties such as discharge as
change in frequency as demonstrated by Callede et al. (2000) for the Amazon River.
  With an integrated power supply that can be supplemented
Fd ¼  2Fs VC ½1 with external battery packs and internal memory boards,
ADCPs can be deployed autonomously for an extended per-
where Fd is the frequency of the returned signal, Fs is the fre- iod. The system can either be programmed to sample and
quency of the original signal, V is the velocity difference between average the current profile at regular intervals (such as hourly),
the instrument and the flow, and C is the speed of sound. In or collect longer ‘burst’ samples over a designated period at the
their most simple form, acoustic velocimeters measure flow sampling frequency (1–2 Hz). For example, Austin et al.
velocity in two or three dimensions at a fixed point and have (2009) programmed an ADCP to average a current over 10 m
been used to measure flow velocities and sediment transport in every half hour to characterize tidal hydrodynamics on a
the nearshore (e.g., Austin et al., 2009, 2010; Ruggiero et al., microtidal wave-dominated delta. Sheremet et al. (2005)
2005), on tidal flats (e.g., Houser and Hill, 2010a, b; Chen et al., averaged a profile over 16 m every hour to quantify current
2010), in rivers (e.g., Frothingham and Rhoads, 2003; Stone and velocities during Hurricane Claudette, and to describe the
Hotchkiss, 2007; Houser, 2010), within vegetation canopies (e.g., formation of a high-density fluid mud layer during and fol-
Granata et al., 2001; Neumeier and Amos, 2006; Bradley and lowing the storm. Programmed in burst mode, an ADCP can
Houser, 2009), and across coral reefs (e.g., Lowe et al., 2005). provide information about high-frequency flow variations
The ability of acoustic Doppler velocimeters (ADVs) to measure (e.g., Shih et al., 2003) and estimate the directional wave field
close to the water–sediment interface at high frequencies allows (e.g., Terray et al., 1999; Storlazzi et al., 2004). Combined with
linkage of fluid and sediment systems and the detection and a pressure transducer and surface tracking technology, recent
characterization of coherent flow structures. A comparison of models provide up to four independent measures of the
ADV and ECM sensors suggests that they provide similar results nondirectional wave-height distribution and improved esti-
for a wide range of flow conditions, but that ECM sensors mates of the directional wave spectrum.
provide more reliable estimates flow under highly turbulent A number of important limitations apply to the use of
conditions (MacVicar et al., 2007). The flow velocities measured ADCP units. First, because the acoustic signal is strongly re-
by ADV systems can be corrupted by bubbles and suspended flected from boundaries such as the bottom and the water
sediment, biofouling or blockage of the signal by floating debris, surface, data cannot be collected close (within the blanking)
barnacles, algae, and an accreting seabed (Elgar et al., 2005). area of those boundaries. Upward looking units are unable to
Unlike an ADV that only measures the flow velocity at a measure close to the bed and thus cannot be used for sediment
single point, an acoustic Doppler current profiler (ADCP) can transport studies. Even for a downward looking (broadband)
measure the reflected signal at different time intervals that ADCP, the minimum cell size is not sufficient for the unit to
represent different travel distances between the moving re- resolve the near-bed flow for turbulence and sediment trans-
flectors and the sensor. The average velocity is calculated for port studies. These shortcomings can be addressed through the
spaced sample bins from the sensor to the water surface for use of a velicometer (ADV) or the newly developed pulse-
upward looking units, or to the bed for downward looking coherent ADCP. The cell size for a pulse-coherent (PC-ADP)
units, although in either case there is a near-boundary ‘blank- unit is an order of magnitude smaller than that of a traditional
ing’ zone (Gartner and Ganju, 2003; Lacy and Sherwood, ADCP, allowing for high-resolution, high-frequency profiles
2004). The number of sample bins and the depth of flow over a relatively short range (Lohrmann et al., 1990; Zedel
measured depend on the frequency of the specific unit. Al- et al., 1996) with a smaller averaging area (footprint) along the
though high-frequency sounds yield more precise information profile. The PC-ADP emits a pair of acoustic pulses a short time
about the flow, they do not travel as far as low-frequency apart and measures the phase shift between the reflected signals
sounds. In this respect, high-frequency current profiles received at the unit. The velocity of the reflector is calculated
(1–1.5 MHz) tend to be used in shallow water (e.g., MacMahan using the phase shift (Df) between the acoustic pairs:
et al., 2005; Masselink et al., 2007; Orzech et al., 2010) and low-
Df
frequency profiles (75 kHz–1 MHz) in deep water (e.g., Simp- v¼c pf t ½2
son and Oltman, 1993; Filizola and Guyot, 2004; Filizola et al., 4
2009). Not only can the sensor be deployed on a fixed frame at
the surface or the bottom, it can be oriented horizontally to where c is the speed of sound, f is the acoustic frequency, and t
measure the current profile across a channel (Figure 3(a)) and is the interval between emitted signals. The second pulse is only
mounted to the bottom of moving vessels equipped with a emitted once the first pulse has been returned, resulting in a
200 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

Cross section number 002

0
0.5
0.5
1
1.5 0.4

River depth (m)

Velocity (m/s)
2
0.3
2.5
3 0.2
3.5
4 0.1

4.5
0
5

–150 –100 –50 0 50 100 150

Across river distance (m)(negative = river left)

Velocity (ft/s)
Fast : > 1 N
0 0.5 1 2
Slow : 0 km

Figure 3 Cross-channel and along-channel velocity measured using a hull-mounted ADCP on the Flathead River in northwestern Montana.
Images provided by Dr. Mark Lorang, Flathead Lake Biological Station: unpublished data, imagery produced with proprietary River Analyzer
software developed by Lorang, University of Montana.

long lag with low instrument noise. However, as discussed in 14.17.2.1.4 Acoustic noise
great detail by Lacy and Sherwood (2004), the phase shift can A more recent application of acoustic sensing has been de-
only be measured up to some multiple of 2p, beyond which an veloped for the characterization of fluvial flow environments.
ambiguity error is introduced and the measured velocity is not The ‘noisiness’ of a system is believed to provide key habitat
accurate (see also Lohrmann et al., 1990). In this respect, the signals to aquatic organisms (Wysocki et al., 2007), in part
increase in spatial resolution comes at the expense of increased because sound can propagate relatively long distances through
errors depending on the sampling range and maximum velocity water. Animals rely on interpretation of aquatic sounds to
expected of the flow. The measurable velocity range (Va) and interpret their environment and to sense the presence of other
the maximum sampling range (rmax) can be calculated as organisms, friendly or otherwise (Amoser and Ladich, 2005).
Tonolla et al. (2009) have attempted to characterize fluvial
c2 soundscapes in a laboratory study. They found in their abiotic
Va rmax ¼ ½3
4f experiments that the ambient noise is controlled by flow vel-
ocities, the characteristics of depth and flow obstructions, and
Despite these limitations, the PC-ADP can provide esti- the generation of air bubbles. In other words, the turbulent
mates of near-bed turbulence (e.g., Lorke and Wuest, 2005; state of the flow is reflected in noise generation. In a follow-up
Rippeth et al., 2001) to measure gravity currents (in the study, Tonolla et al. (2010) performed field experiments where
Chicago River, for example Garcia et al., 2007), to estimate they demonstrated that the fluvial subenvironments of pools,
near-bed sediment concentrations (Bai et al., 2009), and runs with and without sediment transport, riffles, and step
to quantify near-bed turbulence and ripple evolution on the pools could be discriminated based on their acoustic charac-
continental shelf (Hay, 2008). teristics, controlled by the system attributes of Froude number,
 Electronic Measurement Techniques for Field Experiments in Process Geomorphology 201

relative roughness, and the ratio of flow velocity to depth. This 1.2
information provides important linkages between geo-
morphological and biological systems in the characterizations 1.0
of fluvial habitats.

Water depth (m)

0.8

0.6
14.17.2.2 Water Level Measurement
14.17.2.2.1 Wave staffs and wires 0.4
The elevation of a water surface can be measured using piercing
0.2
probes, stilling well floats, pressure transducers, and ultrasonic
distance sensors. The choice of instrument largely depends on 0.0
the amount and frequency at which water level changes. For 0 50 100 150 200 250 300
example, stilling wells are used for environments where rates of (a) Time (hr)
surface change are relatively slow, such as tide or changes in 0.5
river discharge. Surface piercing resistance and capacitance wave 0.4

Water surface fluctuation (m)

wires can be used to measure waves and tide in the nearshore
(e.g., Thornton and Guza, 1983; Greenwood and Sherman, 0.3
1984; Hughes, 1995) or in the swash zone if laid on their sides, 0.2
and called typically, run-up wires (e.g., Holland et al., 1995).
The wave wires consist of a staff fixed in a vertical or inclined 0.1
position and is a component of an electrical circuit in which the 0
voltage changes in proportion to the length of staff that is
submerged. These instruments generally produce continuous –0.1
voltage signals that allow detailed representation of water sur- –0.2
face changes. However, because they must pierce the water
–0.3
surface to be operational, they are also prone to damage or
destruction during severe wave events, or fouling by drifting –0.4
0 100 200 300 400 500 600 700 800
materials. Run-up wires are prone to burial or fouling by sea-
(b) Time (s)
weed, for example (Holman and Guza, 1984; Nelson and
Gonsalves, 1990). As such, applications of wave staffs, wave Figure 4 Example time series from a pressure transducer at
wires, and run-up wires are best suited for relatively short-term Pensacola Beach, Florida, showing (a) tidal-scale variations in water
level and (b) high-frequency wave-scale variations around the mean
experiments where onsite maintenance is possible.
water elevation (Houser, unpublished data).

14.17.2.2.2 Pressure transducers

14.17.2.3 Bed Elevation Change and Bedform Monitoring
The use of wave staffs and wires has been largely replaced by
pressure transducers (see Figure 1(b) for sample data). These 14.17.2.3.1 Altimeters
instruments avoid many of the problems associated with wave Changes in bed elevation (dh/dt) result from gradients in
wires and are also capable of providing precision measure- transport (dq/dx), and the ability to measure erosion and ac-
ment of water level changes. Measuring hydrostatic pressure, cretion at rates comparable to rates at which we can sample flow
the transducers record changes in water level as a change in and transport characteristics is invaluable to linking sediment
pressure measured over a mechanical diaphragm or piezo- transport with landform change. Traditionally, the net change in
electric wafer. As the sensing element is distorted, there is a bed elevation is surveyed after an extended period of hours or
proportional change in output voltage. The pressure changes even longer (using repetitive ground-based or airborne LiDAR,
associated with surface waves is attenuated with depth, as a for example) to allow a measurable change to be observed. Only
function of wavelength. This attenuation can be corrected recently have sensors been developed that are capable of meas-
using linear wave theory. As shown in Figure 4, pressure uring bed elevation change at frequencies commensurate with
transducers are capable of reproducing water surface fluctu- changes in sediment transport. By the very nature of their design,
ations in great detail and over a wide range of scales from tidal these sensors only provide a time series of bed elevation change
variations to short-period wind waves (e.g., Fedderson and at a single point and sets of multiple sensors need to be deployed
Guza, 2003; Houser et al., 2006; Houser and Greenwood, in order to quantify the areal change. The bed elevation sensors
2005; Ruessink, 2000). In addition to wave, tide, and river discussed herein are sonar altimeters, photoelectronic erosion
stage measurements, pressure transducers can be used to pins, and ultrasonic distance sensors.
measure the slope of the surface water elevation along rivers Sonar altimeters are based on the same technology as echo
(Meirovich et al., 1998), across coral reefs (Jago et al., 2007; sounders used for bathymetric surveys and for fish finders, but
Lowe et al., 2005), and shore platforms (Farrell et al., 2009). with the ability to measure bed elevation with great precision.
With burial within unconsolidated sediments, such as in a Altimeters are most commonly used in the nearshore en-
beach, high-precision pressure transducers can measure pres- vironment, where the sensor is positioned at a known ele-
sure gradients in the beach and elucidate infiltration and vation. The sensor generates a narrow-beam acoustic pulse
seepage processes along the beachface (Horn et al., 1998). that is transmitted at 25 Hz for a 1 MHz sensor, for example,
202 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

0.6 0.45
Depth of water
Waver height
Distance to bed
0.5 0.40

Wave height (m) or water level (m)

Distance to bed (m)

0.4 0.35

0.3 0.30

0.2 0.25

0.1 0.20

0 0.15
00

00

00

00

0
:0

:0

:0

:0
0:

0:

0:

0:
12

12

12

12
10

10

10

0
10

10

10

/1

0
7/

8/

9/

/1
10
7/

8/

9/
8/

8/

8/

10
8/

8/

8/

8/

8/
Figure 5 Example time series of water level, wave height, and bed elevation change over several tidal cycles on a nearshore terrace at
Pensacola Beach, Florida (Houser, unpublished data). The data show bed accretion as a smaller distance between instrument and sediment
surface through high tide and bed erosion toward low tide (Houser, unpublished data).

and the travel time of the pulse to the bed and back is used to submillimeter changes in bed elevation to be measured de-
calculate the distance between the instrument and the bed. pending on the sensor range and its distance from the sedi-
The first applications of altimeters were for the measurement ment surface. The sensor was first used in the field by Turner
of nearshore bedforms (Dingler et al., 1977) and as a remote et al. (2008) to measure bed elevation changes of B20 mm
bed profiler (Greenwood et al., 1985). The resolution of the following individual swash events. Houser and Barrett (2010)
sensors degrades with sediment or bubbles in suspension, were able to assess bed elevation changes driven by several
over a sloped bed, where there are variations in bed smaller scales of water motion, including gravity, infragravity, and
than the sonar footprint, and as a result of the dynamic nature tidal time-scales (Figure 6). Not only can the distance from
of the bed itself (Gallagher et al., 1996; Green and Boon, the sensor to the bed be measured between swash events, the
1988). Gallagher et al. (1996) suggested that altimeter reso- sensors can also provide swash height and water depth
lution is B2 cm seaward of the surf zone and B3 cm in the (Blenkinsopp et al., 2010; Houser and Barrett, 2010). The first
turbulent surf zone, although Green and Boon (1988) found measurements of bed elevation change in the swash zone were
that the resolution can be B0.5 cm in controlled laboratory made by Waddell (1973, 1976, 1980) using a capacitance
settings. Recent applications from the nearshore include the probe more commonly used to measure swash depth. Al-
monitoring of storm-based changes in nearshore morphology though the methods by which the water surface and bed
(Fedderson and Guza, 2003) and quantifying bedform surface could be differentiated are not clear, Waddell (1973)
morphology (Bell and Thorne, 1997; Gallagher et al., 2003; observed elevation changes of up to 50 mm resulting from
Feddersen et al., 2003). Beyond the nearshore, sonar altim- individual swash events with a period of B40 s. A similar
eters are also used in transport and bedform studies on the device was used by Ridd (1992), but the minimum resolution
continental shelf (e.g., Cacchione et al., 1999) on estuarine ranged from 2 to 10 mm, the latter of which is too coarse for
mudflats (e.g., Allison et al., 1995) and in fluvial environ- swash zone changes.
ments (e.g., Kuhnle and Willis, 1992; Kuhnle, 1992; Kuhnle
et al., 2006). These instruments are commonly integrated with
an acoustic-based current meter or other flow or sediment 14.17.2.3.3 Photoelectronic erosion pins (PEEP)
sensors (Figure 5). Subaerial bed elevation change can also be measured using
photoelectronic erosion pins that were first developed by
Lawler (1991, 1992), and used to estimate bank erosion along
14.17.2.3.2 Ultrasonic distance sensors the Severn River (Lawler et al., 1997) and in the Swale–Ouse
Grain-scale changes in bed elevation in the swash zone have river system (Lawler et al., 1999). A robust version of the PEEP
been measured using ultrasonic distance sensors (e.g., Turner was designed for deployment in the intertidal zone (Lawler
et al., 2008; Houser and Barrett, 2010). These sensors use a et al., 2001). The sensor contains a linearly arranged set of
piezoelectronic crystal that resonates at a set frequency photovoltaic cells, wired in series and housed in a water-tight
(420 kHz) and the electrical energy is converted into acoustic transparent tube. The cells generate a voltage signal that is
energy. The short wavelength of the acoustic pulse allows for proportional to the amount of light received, which in turn
 Electronic Measurement Techniques for Field Experiments in Process Geomorphology 203

1.5
1.4

Elevation (m)
1.3
1.2
1.1
1
0 100 200 300 400 500 600 700 800 900 1000
Time (s)
Figure 6 Example output of an ultrasonic distance sensor used by Houser and Barrett (2010) in a study of swash zone dynamics. The changing
of the bed surface is visible as the minimum of the time series whereas individual swash events appear as spikes in the data. Reproduced from
Houser, C., Hill, P., 2010a. Wave attenuation across an intertidal sand flat: implications for mudflat development. Journal of Coastal Research 26,
403–411, and Houser, C., Hill, P., 2010b. Sediment transport on Roberts bank: a sandy intertidal flat on the fraser river delta. Journal of Coastal
Research 26, 333–341, with permission from Coastal Education and Research Foundation (CERF).

depends on the length of the sensor that is exposed but also necessarily in equilibrium with the present hydrodynamic
changes in ambient light. It is calibrated through the use of forcing (Traykovski, 2007; Austin et al., 2007), which affects
reference cells at the ends of the sensor to monitor the max- the amount and direction of suspended sediment transport.
imum (exposed ambient) and minimum (completely buried)
light intensities. McDermott and Sherman (2009) deployed
the sensors in a cross-shore transect, colocated with pressure 14.17.2.4 Sediment Detection
transducers to measure bed elevation changes associated with Traditionally, geomorphologists relied on mechanical and
groupiness in the incident wave field. The authors observed static traps to measure sediment transport rates (Goudie,
bed elevation changes of between 10 and 15 mm over periods 1990). Devices such as the Helly–Smith bed-load sampler
of B300 s. Erlingsson (1991) used a similar sensor to detect (Helley and Smith, 1971), the DH-48 suspended sediment
changes in bed elevation and ripple dynamics and the sampler (Guy and Norman, 1970), for rivers, the streamer trap
deposition of mud within the sand matrix. (Kraus, 1987), for nearshore environments, the cylindrical
trap (Leatherman, 1978), or BSNE (e.g., Goossens et al., 2000)
14.17.2.3.4 Side-scan sonars for aeolian studies, became commonly employed. This was
Bedforms are both a manifestation of, and an important despite some concerns about their ability to accurately repre-
control on, the interaction between sediment transport pro- sent the actual sediment transport rates, and the fact that only
cesses and near-bed boundary hydrodynamics. Photography time and space discrete samples could be obtained. The recent
and direct-diver observations were an early approach to bed- development of electronic instruments capable of continuous
form measurement (Hale and McCan, 1982; Sherman and measurement of transport has solved the time constraint, and
Greenwood, 1984) and continue to be an important source of deployment of sets of sediment transport instruments has
data (e.g., Masselink et al., 2007). In general, bedforms are addressed some of the spatial issue.
more dynamic during storms (Hay and Wilson, 1994; Tray-
kovski et al., 1999; Hay, 2008), when the bed is obscured by 14.17.2.4.1 Optical sensors
suspended sediment and conditions are too rough for obser- Suspended sediments in water may be detected and quantified
vation via diving or snorkeling. Rotary side-scan sonars, also based on their reflection or attenuation of an optical or
called sector-scanning sonars, have been used in a number of acoustic signal. Both optical and acoustic sensors emit a signal
studies to produce images of the bottom topography during of known intensity that is scattered by sediment, air bubbles,
storm conditions and in areas inaccessible by divers (Hay and and even zoo-plankton in the water (Kringel et al., 2003). The
Wilson, 1994; Traykovski et al., 1999; Masselink et al., 2007; optical backscatterrance sensor (OBS) illuminates sediment in
Hay, 2008). The backscatter from surfaces facing toward the suspension using a high-intensity, infrared-emitting diode and
sonar is stronger, whereas surfaces facing away from the sonar a photodetector that converts the returned signal to a photo-
have weak backscatter or are completely shadowed. The al- current. Since the amount of signal returned to the sensor
ternating backscatter outlines the crests and troughs of the depends on the surface area of the particle, which is assumed
bedforms and can be used to determine wavelength, orien- to be proportional to the particle volume, the sensor provides
tation, and crest variability. Although the bedform dimensions an indirect estimate of suspended-sediment concentration
can be used to estimate the amplitude from past observations (Downing, 2006). However, the relative intensity of the scat-
and theory, a sonic altimeter should be used to directly tered signal that is received by the sensor needs to be cali-
measure the amplitude of the bedforms as they migrate be- brated against known sediment concentrations to translate the
neath the sensor. Alternatively, a pencil beam sonar can be voltage signals into sediment equivalents. Various calibration
used to directly measure bed elevation at centimeter scale experiments have shown that optical backscatterance is sen-
resolution along a short transect or even multiple transects at sitive to the sediment color (Sutherland et al., 2000), particle
regular intervals (see Traykovski et al., 1999; Ngusaru and Hay, size, shape, composition, and aggregation/flocculation (Green
2004). In several of these studies it has been shown that and Boon, 1988; Greenwood and Jagger, 1995; Hatcher et al.,
bedforms can persist for extended periods of time and are not 2000), bubbles (Downing et al., 1981; Sternberg et al., 1984;
204 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

Puleo et al., 2006), and chemical and biological fouling. As When installed vertically with the sensor flush with
shown by Battisto et al. (1999), the OBS response to clay the bed, an OBS can also be used to measure the rate of
(B2 mm) is 50 times greater than the response to sand sediment deposition (Ridd et al., 2001). An example of wave-
(B100 mm) in the same suspension. Regardless, the OBS re- induced sediment resuspension measured using an OBS
mains an important method of estimating suspended sedi- colocated with an ECM is presented in Figure 7 along with
ment concentration (see also review by Gray and Gartner, the cospectral density plot for calculating suspended sediment
2010) and combined with a current meter can be used to transport (Houser and Greenwood, 2005). The data
estimate the sediment transport rate in the nearshore (Beach show sediment resuspension at infragravity frequencies and a
and Sternberg, 1988; Osborne and Greenwood, 1993; Rues- net onshore transport of sediment. Development of fiber-
sink, 2000), with estuarine sediments (Kraus and Ohm, 1984; optic OBS sensors (see Beach et al., 1992) have permitted
Kineke and Sternberg, 1992; Fugate and Friedrichs, 2002), on finer-scale measurements of suspended sediment concen-
tidal flats (Houser and Hill, 2010a, b), rivers (Lewis, 1996), on tration in relation to near-bed turbulence (Ogston and
the continental shelf (Green et al., 1995; Kineke et al., 2006), Sternberg, 2002), to characterize incipient motion (Foster
and to measure boat wake erosion (Bauer et al., 2002; Houser, et al., 2006) and in shallow swash flows (Puleo et al., 2003;
2010). Puleo, 2009).

2.0 30

Suspended sediment concentration (kg m–2 s–1)

Velocity
1.5 Suspended sediment concentration
25
Cross-shore velocity (m s–1)

1.0
20
0.5

0.0 15

–0.5
10
–1.0
5
–1.5

–2.0 0
0 100 200 300 400 500 600
(a) Time (s)

100
Infragravity Gravity waves
waves
80
Cospectral density (kg m–2 s–1 Δf–1)

0.0078 Hz (128 s)

60

40 0.011 Hz (90 s)

20

Onshore
0
Offshore

–20
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
(b) Frequency (Hz)
Figure 7 Time series of water surface elevation (a) and the associated spectral density plot showing the record to be composed of both
primary (high-frequency) and secondary (low-frequency) components (b). Velocity measured using an electromagnetic current meter and
suspended sediment concentration measured using a colocated optical backscatterance sensor (OBS) at Burley Beach, Lake Huron (Houser,
unpublished data).
 Electronic Measurement Techniques for Field Experiments in Process Geomorphology 205

Table 1 Particle radius for peak sensitivity and minimum detectable radius for ADCP of
different frequency

ADCP frequency (kHz) Peak sensitivity radius (mm) Minimum detectable radius (mm)

3000 80 4
1500 160 8
750 320 15
500 480 25
250 960 50

Source: Modified from Kostaschuk, R.A., Best, J.L., Villard, P.V., Peakall, J., Franklin, M., 2005. Measuring flow velocity
and sediment transport with an acoustic Doppler current profiler. Geomorphology 68, 25–38.

14.17.2.4.2 Acoustic sensors 1100

The suspended sediment concentration and bedload velocity
can also be measured using an acoustic backscatterance sensor
1200
(ABS) that works on the principle of Doppler shift (see
Holdaway et al., 1999). Whether collected from an in-
dependent sensor or extracted from ADCP and ADV sensors, 1300
acoustic backscatter estimates of sediment concentration are
common in fluvial (e.g., Reichel and Nachtnebel, 1994;
1400
Alvarez and Jones, 2002; Kostaschuk et al., 2005; Kostaschuk
and Best, 2005), nearshore (e.g., Crawford and Hay, 1993;
Osborne and Greenwood, 1993; Vincent and Osborne, 1995; 1500
Webb and Vincent, 1999; Thorne et al., 2009), and estuarine
environments (e.g., Hamilton et al., 1998). The intensity of 1600
the acoustic backscatter is a function of the electronics of the
instrument and how it is deployed (frequency, transmit power,
receiver sensitivity, and distance to the measurement volume)
50

00

50

00

50

00

50

00

50

00

50
18

19

19

20

20

21

21

22

22

23

23
and the size, type, and concentration of the suspended sedi-
ment. The acoustic frequency of a particular ABS is most Figure 8 Along-channel variation in bedload transport measured
sensitive to a particular particle size, and the sensitivity de- using an ADCP on the Flathead River, Montana. Brighter colors
creases significantly for particles with a radius smaller than the correspond to faster transport rates and units along axes are meters.
minimum detectable radius (Table 1). Peak sensitivity occurs Images provided by Dr. Mark Lorang, Flathead Lake Biological
where the particle circumference is equal to the acoustic Station: unpublished data, imagery produced with proprietary River
wavelength of the sensor. It is not possible to distinguish Analyzer software, developed by Lorang, University of Montana.
changes in particle size from changes in the suspended sedi-
ment concentration, and as a result, acoustic sensors also need The estimate of bedload flux is, however, complicated by
to be calibrated against known sediment concentrations to sediment size (Rennie et al., 2002) and the depth of pene-
translate the voltage signals into sediment equivalents. tration of the acoustic pulse into the mobile layer (Villard et al.,
Osborne and Greenwood (1992) compared OBS and ABS and 2005). For example, predicted bedload transport from an
found that the concentrations, when averaged over several ADCP on the Flathead River, Montana, is provided in Figure 8.
minutes, agree within 710%, which suggests that both these
instruments probably give reasonable estimates of concen- 14.17.2.4.3 Magnetic sensors
tration. Both sensor types remain critical tools to the process There have been a number of attempts to measure bedload
geomorphologist (Gray and Gartner, 2010). using magnetic sensors that provide autonomous and non-
During periods of minimal suspended sediment transport, invasive sampling (Ergenzinger and Conrady, 1982; Ergen-
acoustic Doppler profilers are also capable of tracking the bot- zinger and Custer, 1983; Custer et al., 1987; Spieker and
tom, and the Doppler shift at the sediment surface can be used to Ergenzinger, 1990; Bunte and Ergenzinger, 1989; Tunnicliffe
estimate the velocity of the mobile bed (Kostaschuk et al., 2005; et al., 2000; Hassan and Ergenzinger, 2003). The most recent
Kostaschuk and Best, 2005), particularly since most acoustic version of this sensor includes a copper wound coil inside a
units are sensitive to the larger particles that are typically moving Toroidal (or doughnut-shaped) magnet and placed within an
as bedload (Villard et al., 2005). As presented by Kostaschuk iron casing that acts as a Faraday cage and ensures that the
et al. (2005), the bedload flux (qb) can be calculated as the signal is not contaminated by other magnetic objects (Tun-
product of the bedload layer thickness (db), the bedload con- nicliffe et al., 2000). According to Faraday’s law, an electro-
centration (cb), and the bed velocity (u) measured by the ADCP: motive force is induced when sediment moves through the
magnetic field. The sensor can be digitally sampled using an
analog to digital recorder to create a sharp-peaked oscillation.
qb ¼ udb cb ½4 The amplitude, width and integral of the oscillation are
206 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

proportional to the strength of the sensor magnet, as well as strength of the return at different beam travel times, concen-
the size, shape, and mineralogy of the sediment, and its speed trations at multiple distances from the instrument can be
and path relative to the sensor. Through controlled, laboratory obtained. It is thus possible to gather concentration profiles
testing, Hassan et al. (2009) found that it is possible to relate through a water column. With appropriate calibration, the
the width and amplitude of the signal to the volume and signal can also be used to estimate particle size (e.g., Law et al.,
speed of the bed sediment. However, the sensor has proven 1997; Schat, 1997). When the suspended sediment concen-
difficult to calibrate for use in the field. For example, a particle tration exceeds 100 g l1, nuclear turbiometric gauges measure
moving along the edge of the sensor produces a very different the attenuation of radiation intensity (gamma rays) either
signal than the same particle moving beneath the sensor. The backscattered or absorbed by the water and sediment (Tazioli,
calibration is further complicated by the need for consistent 1980, 1981). Instead of an acoustic or optical signal, an arti-
magnetic susceptibility of the bed sediment, which is difficult ficial source of radiation (137Cs, 60Co, 109Cd, or 241AM) is
to assume in the prototype. emitted into the water and the attenuation of the signal is
linearly dependent on the suspended sediment concentration.
14.17.2.4.4 Laser in situ scattering and transmissometry This device was tested at three gauging stations by Lu et al.
The concentration and size distribution of suspended sedi- (1981), who found good agreement of the nuclear gauge with
ment can be measured using a laser in situ scattering and traditional sampling methods over a wide range of concen-
transmissometry (LISST) sensor, which uses near-forward light trations (15–500 kg m3).
scattering (e.g., Traykovski et al., 1999; Agrawal and Pottsmith,
2000; Battisto, 2000). Specifically the sensor uses a collimated 14.17.2.4.6 Hydrophones
beam of light from a laser source (with wavelength l ¼ Hydrophone systems measure the sediment transport process
670 nm) that is emitted at 32 angles from the sensor and through either the direct impact of particles to generate a signal
scattered by the suspended sediment. The scattering intensity or by measuring ambient sounds from moving clasts. As dis-
depends on the physical characteristics of the particles, such as cussed in Bogen and Men (2003) and Gray et al. (2010), the
size, shape, and composition (Bohren and Huffman, 1983; impact sensors rely on microphones or piezoelectric sensors
Kirk, 1994; Traykovski et al., 1999; Mikkelsen and Pejrup, located adjacent or attached to metal plates or pipes or plastic
2000). The scattering is dominated by diffraction at small bases (see also Richards and Milne, 1979; Bänziger and Burch,
angles, which is most sensitive to the particle cross-section, 1990; Taniguchi et al., 1992; Mizuyama et al., 2010). Use of
such that the angular distribution of scattered light is a hydrophones to detect and quantify sediment transport in
measure of the particle size and shape. The concentration of fluvial and marine environments traces back at least to the work
sediment can then be calculated by converting the measured of Mühlhoffer (1933, cited in Bedeus and Ivicsics, 1963) who
size into an equivalent spherical diameter (Agrawal and installed a hydrophone to detect the impacts of bed load hitting
Pottsmith, 2000). Use of the LISST in Chesapeake Bay, USA, a steel plate on the bed of a river, although it is not clear how
has shown the bottom boundary layer to be composed of these sounds were used. Bedeus and Ivicsics (1963) developed
several particle populations (Fugate and Friedrichs, 2002) that a hydrophone system that was suspended above the bed (to
behave differently in response to the wave and current forcing minimize disturbance of the bed load) and recorded variations
(Greenwood and Xu, 2001). In a recent application, Shugar in bed load in the Danube River, Hungary, but did not translate
et al. (2010) used a LISST to examine the relationship between these measurements to absolute transport rates. Since then,
sediment resuspension over alluvial bedforms and large-scale there have been numerous improvements in design and in-
turbulence in the Rio Parana, Argentina, while Kostaschuk stallation of hydrophone systems, and there have been appli-
et al. (2009) characterized the variation in the suspended cations for detecting the initiation or cessation of motion,
sediment concentration and rate of sediment deposition over especially of coarser particles, in rivers (e.g., Bray, 1975; Rick-
the same dunes. Thonon et al. (2005) installed LISST-type enmann, 1997; or Turowski and Rickenmann, 2009) and bed
sensors on two floodplains in the Netherlands to detect flood load transport rates (e.g., Rouse, 1994; Froehlich, 2003; Krein
sedimentation. The OBS and ABS systems perform similarly to et al., 2008; Mizuyama et al., 2010), and in marine environ-
the LISST in environments where there is a uniform distri- ments (Thorne, 1986; Williams et al., 1990; Jagger and Hadisty,
bution of fine and medium-sized grains respectively (Creed 1991; Voulgaris et al., 1999).
et al., 2001). Whether using direct impact or ambient noise, hydro-
phones provide continuous and detailed sediment transport
14.17.2.4.5 Conductivity concentration profiler measurements with minimal disturbance to the transport
Puleo et al. (2010) described a new conductivity-based sensor system. In the Erlenbach stream in Switzerland, hydrophones
capable of measuring near-bed sediment concentrations. The have been measuring bedload transport since 1986 (Rick-
conductivity concentration profiler (CCP) uses the conduct- enmann and McArdell, 2007), with measurements comparing
ivity of the sediment–water mixture as a proxy for sediment favorably with data obtained from other methods, including
concentration. Although not yet ruggedized for field appli- sediment impoundment. A similar system, using hydrophone
cations, the eight electrode pairs span a vertical range of pipes, was installed in the Ashiarai-dani stream in Japan in
0.016 m, and if partly buried in the bed, can be used to 2004. Unlike static or pit-type sediment traps, the acoustic
measure bed mobilization. Alternative technologies use a systems can be installed with sensing elements flush with or
spatially separated source and receiver that measure the at- just below the transport surface. The ambient noise sensors
tenuation of the signal between them, as it is scattered by the can be mounted or suspended above the bed to listen for the
sediments (e.g., Waddel, 1976). Further, by measuring the self-generated sounds of moving sediment, which are
 Electronic Measurement Techniques for Field Experiments in Process Geomorphology 207

normally coarse clasts (Thorne et al., 1984; Voulgaris et al., postprocessing. As described by Dalrymple et al. (2010), surf
1999). In this respect, hydrophone-based systems can also be zone drifters have been instrumental in elucidating the struc-
used to detect the initiation of particle motion. ture of rip current flows (e.g., MacMahan et al., 2010) and in
The primary disadvantage to all hydrophone-based systems measuring rip currents that extend beyond the surf zone (e.g.,
is the difficulty in calibrating their signals to produce absolute Reniers et al., 2010).
transport rates, despite early attempts to relate particle size and Sediment tracking has been an important approach to
distribution with the frequency of the impact-generated sound understanding sediment transport for more than a century (e.g.,
(Jonys, 1976; Millard, 1976). It is generally difficult to separate Richardson, 1902; Lamarre et al., 2005) and has traditionally
transport-related sounds from the ambient noise resulting from been based on visual (e.g., Dornbusch et al., 2002) and re-
such things as flow turbulence, although this is less problematic motely detected tracers (e.g., Osborne, 2005). In the latter case,
in piezoelectric systems where frequency filtering can be used to radioactive, aluminum, magnetic, or electronic tracers are pas-
isolate the distinct signal produced by impacting sediment. sively tracked and do not provide information about the actual
Similar to the magnetic sensors, it is also difficult to associate sediment path through an environment or transport event. The
clast size with particular impacts so that impact counts can be electronic tracers can be attached to gravel-sized sediment and
translated into mass flux equivalents, although there have been can be sensed by a detector in close proximity, allowing for a
attempts to relate particle size with acoustic frequency (Thorne quick survey if the particle is not buried deeper than the sensing
et al., 1983, 1989; Rouse, 1994), or to measure particle size range (Osborne, 2005; Curtiss et al., 2009). Recently, active or
from concurrent video records (Thorne, 1986). Most hydro- smart tracers have been developed to track sediment through
phone systems are limited to environments where sediments the course of an event (Allan et al., 2006). The tracer agents
are larger than sand because most piezoelectric systems have include acoustic or radio-tagged artificial gravels that actively
impact thresholds below which particle collisions will not be transmit signals and can be detected at depths of up to 1 m
detected. Finally, under conditions of intense transport, it is not (Allan et al., 2006). The emitter and battery are encapsulated in
clear that all impacts can be identified. When sensing surfaces resin of similar size, shape, and weight as the real sediment.
are relatively large, extremely fast sample rates are necessary to Although expensive and restricted to subaerial environments,
allow the opportunity for discrete impacts to be detected, and both passive and active electronic tags have the potential to
there is always the possibility of simultaneous impact. Despite yield significant insight into sorting processes (Allan et al.,
the potential drawbacks, these surrogate measures of bedload 2006; Curtiss et al., 2009).
transport offer great potential for improving our understanding
of this mode of sediment transport.

14.17.3 Monitoring Geomorphic Systems Controlled

by Aeolian Processes
14.17.2.5 Flow and Transport Tracking Approaches
The flow and sediment transport monitoring techniques de- Measurement of wind velocity and aeolian sediment transport
scribed in the previous section involve fixed, single-point requires a different methodological approach compared to
(Eulerian) measurements of the flow. Only through the de- coastal and fluvial systems. Electronic instrumentation in the
ployment of an extensive instrument array can the complex water requires bulky and expensive containers that are watertight
flow patterns of large rivers and irregular bathymetry be de- and capable of withstanding significant forces by both waves and
scribed (see Schmidt et al., 2003; MacMahan et al., 2009). currents. Although the lower density and viscosity of air reduce
Such large arrays are expensive and only possible through the force applied, the sensors need to be much smaller to capture
extensive collaborative efforts such as the series of experiments the higher resolution flow and transport processes. A significant
conducted at the U.S. Army Corps of Engineers’ Field Research technological challenge is to develop ruggedized instruments to
Center at Duck, North Carolina, but these large-scale de- measure near-bed turbulence and saltation dynamics on scales
ployments carry with them a risk of altering the very flow of centimeters and fractions of seconds.
conditions being measured. To avoid this problem, GPS-based
drifters have been used to resolve spatially complex circulation
14.17.3.1 Airflow Measurement
patterns in some of the world’s largest rivers (e.g., Limeburner
et al., 1995; Banas et al., 2009), in estuarine and shelf en- 14.17.3.1.1 Mechanical anemometers and vanes
vironments (e.g., Ostrander et al., 2008; Stevens, 2010) and to The primary instrument for measuring wind speed is the an-
measure rip currents in the surf zone (e.g., Spydell et al., 2007; emometer, which is generally based on one of three basic
Reniers et al., 2009; MacMahan et al., 2010; Austin et al., working principles: mechanical, sonic, and thermal an-
2010). Recently designed drogues used in the surf zone include emometry. Regardless of the sensor type, airflow measure-
tubular PVC ballast with a PVC disk affixed to the bottom of ments are typically used to estimate the shear stress acting on
the ballast to dampen vertical motion and allow waves to pass the sediment surface and responsible for transport. The shear
over the unit with minimal displacement (Schmidt et al., stress, t, can either be estimated from the vertical profile of the
2005). GPS units acquire their position and transmit the data wind using the so-called law-of-the-wall, or estimated from
to receivers on the beach for real-time correction and data the fluctuating horizontal and vertical components of the
storage. Position errors in the nearshore are on the order of wind (u0 and w0, respectively) and the density of the air, r:
1 m when in an active surf zone, but submeter accuracy can be
achieved if the signal is corrected in real time or through tE  ru0 w0 : ½5
208 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

The latter approach can only be measured by fast-response element of a sonic anemometer consists of an ultrasonic
anemometers (see Sections 14.17.3.1.2 and 14.17.3.1.3). sound emitter and receiver at opposite ends of a sampling
Traditionally, mechanical anemometers have been the in- volume. The operating principle has been described in detail
strument of choice for field studies of aeolian processes, al- by Kaimal and Finnigan (1994), van Boxel et al. (2004), and
though the use of sonic or thermal anemometry has been Walker (2005), and is based on the transit times, t1 and t2, in
increasingly popular. Mechanical anemometers consist of a each direction between the transducers of the ultrasonic sound
rotating shaft driven by cups or a propeller, with the rate of signal over the path length, L, where the sonic speed in still air,
rotation proportional to the wind speed. They are of two C, is offset by the speed of the airflow component, V. The
general types, generating a signal using a photo-chopper, measurement of transit times across both directions of the
where the rotating shaft interrupts a light beam and the path length allows the explicit determination of the speed of
interruptions are counted, or using a dc-generator, where the sound, which is dependent on the density of the air. As air
spinning shaft produces an output voltage. Due to the inertia density is principally controlled by air temperature the sonic
of the rotating assembly in either type, these sensors are not measurement of C therefore also yields an estimation of the
able to adjust instantaneously to changes in the wind speed, temperature in conjunction with wind component speed,
but require a certain amount of airflow (expressed as a dis- which can be useful for atmospheric stability assessment.
tance) to pass the anemometer before the rate of rotation Three pairs of emitters and receivers are combined in a 3D
adjusts. Specifically, the distance is calculated as the length of anemometer to resolve the complete wind vector. Two basic
airflow required for the sensor speed to be 63% of the newly arrangements of transducer pairs are in use. First, the pairs can
imposed wind speed (Schubauer and Adams, 1954; Camp be mounted independently, measuring along individual path
et al., 1970; Stout and Zobeck, 1997). The most widely used lengths, in a mutually orthogonal framework. An example is
cup anemometer is the Gill-type three-cup anemometer, the ‘K’ probe manufactured by Applied Technologies Inc.,
which has a distance constant on the order of 2 m and a which is based on a design by Kaimal to minimize potential
threshold of initial motion of around 0.5 m s1. As a result, flow distortions of the mounting frame (Kaimal et al., 1990).
wind speed fluctuations can only be resolved up to a fre- Second, the pairs can be arranged so that they measure along
quency limit of around 5 Hz under wind speeds generally paths crossing through a single sampling volume, usually
encountered in the field. The sensors are also subject to optimized for a specific orientation to the wind so that the
fouling of the bearings of the rotating shaft by dust when transducer clusters are positioned above and below the hori-
deployed in field conditions with active sediment transport. zontal plane of the flow. The advantage of the latter design is
Other disadvantages are the relatively large size of the cup or that the 3D wind vector is obtained from a single, relatively
propeller assemblies (on the order of tens of centimeters), and small, sampling volume.
the restrictions this imposes to deployment near the surface or Sonic anemometers are commercially available over a
spaced close together with other instruments. However, range of design specifications, principally as a function of
mechanical anemometers are generally robust and inexpen- measuring frequency and resolution. Entry models start at 4 to
sive, do not require an external power supply, and can be left 10 Hz measurements, whereas research grade sensors like the
in the field for long deployments, collecting wind speed Gill Instruments HS-range can measure at 50 or even 100 Hz,
measurements at long time-scales. with a speed resolution on the order of 0.01 m s1 and a
Mechanical anemometers require colocated or integrated directional resolution of less than 11. The precision and
vanes if wind direction is also to be measured. Vane response measuring frequency make these anemometers particularly
to changes in wind direction can be described by the damping suitable for investigating many scales of turbulence and for
ratio and the vane delay distance, which are both independent determining shear stresses from Reynolds’ decomposition (see
of wind speed (Finkelstein, 1981). The damping ratio is an eqns [9]–[11] in Section 14.17.4). Furthermore, the sensor
(inverse) measure of the overshoot as the vane adjusts to a does not contain any moving parts and is therefore not prone
newly imposed wind direction, with low damping ratios in- to fouling by airborne dust or sediment.
dicating large overshoot with a change in wind direction. The In the recent decade sonic anemometers have been de-
delay distance is measured as the amount (or distance) of air ployed in a number of aeolian field investigations on beaches
required to pass a vane before it shifts is position 50% of the (most recently: Walker et al., 2009) as well as in desert and
way to the newly imposed wind direction. In a review of semi-arid environments (Wiggs et al., 1996; Leenders et al.,
commercially available wind vanes, Finkelstein (1981) found 2005; Weaver and Wiggs, 2010) in a variety of configurations. In
that delay distances are typically between 0.6 and 2 m, and most experiments the deployment has been restricted to a small
damping ratios between 0.2 and 0.5. Assuming a delay dis- number of sonics (and commonly a single one), although re-
tance of 1 m and a wind speed of between 4 and 10 m s1, the cently a comprehensive study of wind over coastal foredunes by
delay times range between 0.25 and 0.1 s, and the sampl- Jackson et al. (2011) used a total of 24 Gill-HS sonic an-
ing frequency of wind vanes reaches a limit between 5 and emometers distributed over a sequence of 16 m tall masts to
10 Hz, approximately commensurate with a typical three-cup explore the role of off-shore winds on sand transport on the
anemometer. beach, also enabling a detailed comparison with computational
fluid dynamics (CFD) modeling. Physical limitations on pos-
14.17.3.1.2 Sonic anemometers itioning of the instrument are few, apart from the size of the
Sonic anemometers work on the principle that the travel time transducer framework, and they have been deployed close to
of sound waves through air is affected by the wind speed the ground and just above marram grass vegetation on coastal
component parallel with the direction of travel. The basic foredunes. Van Boxel et al. (2004), however, reviewed
 Electronic Measurement Techniques for Field Experiments in Process Geomorphology 209

18
16

Wind speed (m s–1)

14
12
10
8
6
4
2
0
(a) 0 50 100 150 200 250 300

18
16
Wind speed (m s–1)

14
12
10
8
6
4
2
0
0 50 100 150 200 250 300
(b) Time (s)
Figure 9 Synchronous time series of wind speed obtained from a cup anemometer and an ultrasonic anemometer colocated 1.00 m above the
surface at Jericoacoara, Brazil (Sherman and Li, unpublished data).

limitations to the resultant turbulence power spectra as a (1999a). Because of their small size, these probes can be de-
function of the height of the instrument above the surface, the ployed on very small scales. Furthermore, thermal anemom-
path length between the transducers, and the average wind etry allows measurement of wind speed at high frequencies,
speeds involved, to establish minimum measurement run ranging from 50 Hz to 100 kHz, and are thus able to resolve
durations for restricting low-frequency losses and minimum details of turbulence not possible with the slower sample rates
sampling frequencies for restricting high-frequency losses. They of sonic anemometers (Marusic and Hutchins, 2008). Pairs of
suggest that, for typical sonic anemometers measuring at Z20 hot-films or wires can be mounted in mutually orthogonal
Hz high-frequency losses in turbulence spectra may become directions (hence termed ‘cross-probes’), to measure wind
significant when deployed too close to the ground. This as- speeds in two dimensions. These cross-probes allow for the
sessment is based, however, on the surface layer turbulence direct determination of shear stresses according to a Reynolds’
models derived from the Kansas meteorological experiment decomposition (e.g., Bauer et al., 1998).
(Kaimal et al., 1990). These results are supported by the analysis Regular hot-wire or hot-film probes are, however, very
of Kunkel and Marusic (2006) that demonstrated the loss of the fragile and are easily broken by the impact of saltating grains
high-frequency portions of the spectra. However, the relevant on the sensitive element, and cannot be deployed inside an
scalings in aeolian geomorphology settings may be sub- active saltation layer. For this purpose, metal-clad or armored
stantially different. Figure 9 presents two time series of wind hot-film probes have been developed, where the sensitive hot-
speed obtained simultaneously from a cup anemometer (top) film is wrapped with a protective coating of nickel or stainless
measuring at 5 Hz, and a collocated, ultrasonic anemometer steel. The resulting probe has a greatly increased diameter, on
hot-film probe (bottom) measuring at 20 Hz. the order of 0.5 mm, and due to thermal inertia the frequency
response of these sensors is greatly suppressed, down to a
14.17.3.1.3 Thermal anemometers frequency on the order of 5 to 15 Hz. Thermal anemometry is
Thermal anemometry is based on the concept that heat is considerably more expensive than sonic anemometry, requires
conducted away from a thin, electrically-heated metal wire or additional signal processing equipment, and necessitates a
film by passing airflow. Hot-film or hot-wire probes are gen- significant power supply. Furthermore, because the probes are
erally kept at a certain temperature (around 250 1C) and the fragile, thermal anemometry is rarely used in fieldwork. Re-
power necessary to keep the probe at constant temperature is a search with hot-film probes also imposes some serious cali-
measure of the speed of the airflow past the sensor. Hot-wire bration challenges. The response of the probes to wind speed
probes generally consist of a tungsten or platinum wire with a is not linear and therefore requires careful calibration in a
diameter on the order of 5 mm, whereas hot-film probes are wind tunnel. Hot-film probes also generally exhibit significant
composed of a thin metal film deposited on a core of quartz calibration-drift, so that calibrations must be repeated before
fiber. In both cases the sensitive element is usually only 1 to and after deployment. When thermal anemometry is deployed
2 mm long. An overview of various types of hot-wire and hot- successfully in atmospheric environments, however, it gener-
film probes used in aeolian research is provided by Butterfield ates a wealth of information on small time-scales otherwise
210 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

Speed
Transport
16 250

Transport intensity (relative)

14

Wind speed (m s–1)

200
12
10 150
8
6 100
4
50
2
0 0
0 50 100 150 200 250 300
Time (s)
Figure 10 Synchronous time series of wind speed (airflow forcing) obtained from an ultrasonic anemometer 1.00 m above the bed and a
colocated miniphone measuring the transport response 0.05 m above the bed. Both instruments were measured at 32 Hz. Data from the
miniphone are in terms of relative transport intensity. Data measured at Jericoacoara, Brazil (Sherman and Li, unpublished data).

inaccessible with cup-anemometry. The graphs show how subject to issues of unknown trapping efficiency, moist sand
the cup-anemometer measurements suffer from lags and conditions, and changing wind directions.
suppressed fluctuations due to a slow dynamic response,
whereas the thermal probe is able to capture gustiness in 14.17.3.2.2 Laser and optical sensors
greater detail. Clearly, such information is required for relating More recently, instrumentation has been developed that does
wind to sediment transport on small spatiotemporal scales not require the capture of sand with traps. Noninterfering
(see Figure 10, in Section 14.17.3.2.3). optical or remote-sensing techniques have been applied to the
More recently Ellis (2006) and Ellis et al. (2012) de- measurement of sand transport rates. Nickling and Ecclestone
ployed ruggedized Dantec Dynamics thermal probes similar to (1981) developed a laser-based system for the detection of the
those used by Butterfield (1999a) to explore details of the initiation of motion in a wind tunnel. Butterfield (1999a, b)
boundary layer near the surface, and to link small-scale fluc- pioneered the use of lasers in an optical mass-flux sensor that
tuations in wind speed with changes in saltation intensity. The is capable of determining transport rates on a scale of milli-
small size of the probes allows close spacing of the sensors. seconds with an accuracy close to the mass of individual sand
However, because the probes are armored for protection from grains. The system consists of a sheet of laser light that is
saltating sand, the instrument response rate is limited to about transmitted between two prism posts over a height of 2 cm
6 Hz. These probes also require careful calibration as their re- above the bed, perpendicular to the sand transport direction,
sponse to changes in wind speed is nonlinear. onto a detector. This detector subsequently measures the de-
gree of interference or obstruction of the laser sheet caused by
the movement of sand grains through it. Because the system is
nonintrusive there are no issues with trapping efficiency, al-
though it must be carefully calibrated with the appropriate
14.17.3.2 Transport Measurement
sediment. These systems have not yet been ruggedized for field
14.17.3.2.1 Load cells applications. More recently, Mikami et al. (2005) developed a
Over the past decades a number of electronic devices have been laser-based particle counter capable of measuring grain counts
developed to measure wind-blown sand transport at time scales and particle size, and thus sand transport at a given elevation.
similar to those at which wind measurements are obtained. The Davidson-Arnott et al. (2008) used a commercially available
use of load cells for weighing trapped sand on a continuous laser sensor (Wenglor YH03PCT8) to count grain passages in a
basis was initiated by Fryberger et al. (1979) and Lee (1987). field experiment. Most optical or remote-sensing techniques,
Load cells are able to weigh sand with a resolution of 0.1 to 1 g however, have been employed to investigate details of sal-
at frequencies on the order of 5 Hz. Jackson (1996) and Bauer tation dynamics rather than to obtain direct transport rates.
and Namikas (1998) refined the use of load cells by installing The use of film and still photography for determining saltation
tipping-bucket assemblies that periodically empty the weighing trajectories and grain impact and ejection processes has yiel-
container connected to the load cell to avoid overloading the ded a great wealth of insight into these small-scale dynamics
system. This allows the device to be operated over prolonged (Willetts and Rice, 1986; Rice et al., 1995; Foucaut and Sta-
periods of time. Namikas (2002) also used load cells in com- nislas, 1997; Zou et al., 2001). The velocity of grains in
bination with a horizontally compartmentalized trap and a moving sand clouds has also been measured with particle
vertical array trap for high-resolution transport measurements dynamic analysis or phase Doppler ‘anemometry’ (Dong et al.,
on both temporal and spatial scales. Although these devices 2002) at frequencies of over 40 Hz. Although the above
overcome the limitation of manual traps with regard to the nonintrusive techniques are able to provide high-resolution
scale of the measurements and ease of operation, they are still measurements, most of the equipment is relatively expensive,
 Electronic Measurement Techniques for Field Experiments in Process Geomorphology 211

fragile, and elaborate, so that much of their use has been traditional transport rates therefore requires either colocated
limited to wind tunnel studies. sand traps (with related problems) (Gillette et al., 1997) or
specific assumptions concerning vertical distributions of grain
14.17.3.2.3 Impact sensors size and flight velocity in the saltation layer together with
The third principal technique for measuring sand transport careful sensitivity determinations (Baas, 2004). A recent in-
rates is based on devices that respond to the impact of salt- novation, the Buzzer Disc (Sherman et al., 2011), uses an ex-
ating grains on a small sensitive element with electronic pul- posed piezoelectric wafer to detect grain impacts directly. This
ses; hence, these are termed impact responders. In the case of sensor has a low impact threshold and, because the sensing
the Saltiphone (Spaan and Van den Abele, 1991), one of the surface is flat, has fewer calibration issues than the Sensit or
earliest of such sensors, the sensing element consists of a Safire. This is also a low-cost device (self-built for less than
microphone that is able to detect and register discreet impacts $10). It is less sensitive than the miniphone, but more durable.
of grains larger than 50 mm at frequencies of up to 1 kHz. The Despite such calibration issues, the deployment of impact
physical design of the instrument includes a horizontal cy- sensors and colocated thermal anemometers allows for de-
linder positioned in front of the microphone to guide saltating tailed investigations of sediment transport response to chan-
grains and a wind vane to rotate the microphone. This design ges in wind speed on an unprecedented scale. Figure 10
introduces some issues, however, including clogging by moist presents two synchronous time series from a hot-film probe
sand, inertial directional response time to changing wind (top) and a colocated Safire (bottom), both measuring at
directions and fouling of bearings. Ellis et al. (2009) de- 32 Hz at an elevation 0.04 m above a sand surface. The spatial
veloped a modified version of the Saltiphone, called the scale of these measurements is on the order of a few square
miniphone, especially designed for high-frequency (45 kHz) centimeters in frontal area to the flow. The figure shows how
detection of grain impacts. The miniphone uses a modified individual transport events can be related to high-frequency
electret microphone with a very small sensing element (as fluctuations in wind speed to a level not attainable with tra-
small as 11 mm2) so that unidirectional details of the saltation ditional sand traps and cup anemometers. Such instru-
field can be measured. Although the miniphone is quite sen- mentation has been crucial for the understanding of the
sitive, it is also relatively delicate, and may last less than one formation and behavior of aeolian streamers (Baas and
hour in conditions of intense saltation. Its low cost (self-built Sherman, 2005) and is bound to further our fundamental
for less than $10) compensates for the short functional life of understanding of aeolian transport processes.
the sensor. The development of the different sensor types has led to a
The other forms of impact sensors have been developed number of attempts at comparison. These include laboratory
around piezoelectric technology. One of the earliest laboratory comparisons of the Saltiphone, Sensit, and Safire by Van Pelt
systems was developed by Hardisty (1993) using a piezo- et al. (2006), field comparison of the Wenglor particle counter
electric crystal attached to a phonograph needle. Two com- and the Safire by Davidson-Arnott et al. (2008), field com-
mercially available piezoelectric sensors have been developed parison of three Sensit models and the Safire by Barchyn and
for field applications: the Sensit (Stockton and Gillette, 1990) Hugenholtz (2010), and field comparison of the Wenglor,
and the Safire (Baas, 2004; Baas and Sherman, 2005). The Safire, miniphone, and the Buzzer Disc by Sherman et al.
Sensit consists of a piezoelectric sensing ring mounted around (2011). The general finding is that sensors such as the Wenglor,
a vertical post that can be positioned in the saltation layer. miniphone, and Buzzer Disc have the potential to count most
This omnidirectional sensor can detect grain impacts at fre- or all grains at a particular location. The Wenglor has a greater
quencies of up to 55 kHz and registers grains larger than potential for signal saturation, but is viable for longer de-
roughly 150 mm. The exposure of the piezoelectric crystal to ployments than the miniphone or Buzzer Disc. The Sensit and
abrasion by saltating sand may lead to deterioration of the Safire are much less sensitive and prone to signal saturation
signal, however, and the considerable size of the crystal can because of the relatively large areas of their sensing surfaces.
lead to internal reverberations of grain impacts. As a result the However, they are well suited for long-term deployments
Sensit is at present primarily used to detect the onset of sal- where measurements of relative transport intensities are sat-
tation. The Safire circumvents these problems by using a isfactory. As with all sensors, the selection of an appropriate
damped metal ring around a vertical post from which impact device must be guided by the intended application.
vibrations are transmitted by leads to an internally protected,
smaller piezoelectric element. The Safire can measure grain 14.17.3.2.4 Soil moisture measurement
impacts with frequencies of up to 12.5 kHz, filters the signal The presence of moisture in surface sediments has the po-
for vibrations stemming from rain drops or jolts, and registers tential to substantially alter aeolian transport rates because of
grains larger than approximately 200 mm (Baas, 2004). These surface tension-induced cohesion between particles (Sarre
types of sensors are fully weatherproof, omnidirectional, and 1987; Jackson and Nordstrom, 1997; Wiggs et al., 2004;
easy to deploy. Both types of impact responders suffer from Davidson-Arnott et al., 2008), and several models have been
one fundamental limitation, however, because the detection developed to account for that effect (e.g., reviews by Namikas
of grain impacts is based on the momentum, rather than the and Sherman, 1995; or Cornelis and Gabriels, 2003). Use of
mass, of the saltating grains. Their detection limit is therefore a these models required measurement or estimation of the
function of grain mass, velocity, and the angle of impact on moisture content (usually by weight) of the surface sediments.
the curved sensing surface. For a particular aeolian environ- The common protocol requires taking surface sand samples
ment this can and will lead to some unknown degree of that are weighed, dried, and reweighed to determine moisture
undercounting. A translation of such measurements to content. This is a tedious methodology and requires the
212 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

disruption of the surface of interest. The development of di- et al., 2001). This instrument measures the impedance of a
electric-based instruments obviates both problems. There are medium at a fixed oscillation frequency and is calibrated using
three main approaches based on the dielectric permittivity of known dielectric constants for different media (Robinson
sediments (Robinson et al., 1999): Time-domain reflectometry et al., 1999). The sensor relies on signals returned from a set of
(TDR); frequency-domain reflectometry (FDR or capacitance); four, 0.06 m long, probes. This instrument, in several versions,
and theta probes. The basic premise underlying these ap- has been used widely in aeolian research where sediment
proaches is that the speed, v, of electromagnetic waves through moisture content is a critical complicating factor in the
a medium is a function of the dielectric constant, K, of that modeling of sand transport rates (Atherton et al., 2001; Van-
medium: hee et al., 2002; Yang and Davidson-Arnott, 2005; Edwards
  and Namikas, 2009). Theta probes are minimally intrusive to
c
v ¼ pffiffiffiffi ½6 the sediments, easily mobile, produce stable measurements
K
quickly, and are relatively inexpensive. Because of the necessity
where c is the speed of light. The dielectric constants of air (K to measure moisture content only quite near the sand surface,
¼ 1), dry sand (K ¼ 3–5), and water (K ¼ 80) are different several studies have used a dielectric foam block to reduce the
enough that changes in moisture content cause readily effective length of the probes. Yang and Davidson-Arnott
discernable and calibratable changes in the dielectric constant (2005) reduced the effective length to 0.02 m without sig-
of the air/sand/water mixture. nificant loss of performance, and this modification has been
The use of TDR to measure water content in soils dates used in field studies (e.g., Davidson-Arnott et al., 2008).
back at least to the work of Davis and Chudobiak (1975), and Schmutz (2007) demonstrated that a length as short as
was popularized through numerous soil moisture applications 0.014 m resulted in only 0.9% error, and this length has been
in the 1980s (Topp et al., 1980; Dalton et al., 1984; Dasberg used by Edwards and Namikas (2009) in a field study.
and Dalton, 1985; Yanuka et al., 1988). It was recognized that
the concept was valid for mineral soils and sands, and was
14.17.3.3 Surface Shear Stress
suggested for use in measuring beach moisture above the
water table by Baird and Horn (1996). The TDR electronics In the presence of nonerodible roughness elements such as
comprise a pulse generator and an oscilloscope. A pulse is sent pebbles, cobbles, and vegetation partitions the total shear
along a coaxial cable to one or more probes. Part of the pulse stress (t) between the roughness elements (tR) and the sedi-
is reflected at the junction of the cable and the probe and part ment surface tS is
is reflected from the end of the probe. The differences in return
times, t, of the two reflections is a function of the probe t ¼ tR þ tS : ½8
length, L, and the dielectric constant (K), so that eqn [7] can
be rewritten as The amount of force dissipated by the roughness elements
 ct 2 and the degree of protection afforded to the surface have been
K¼ ½7 the focus of numerous wind tunnel studies (Marshall, 1971;
2L
Lyles and Allison, 1976; Musick and Gillette, 1990; Musick
The performance of the TDR probe is constrained by the dis- et al., 1996; Crawley and Nickling, 2003; Brown et al., 2008).
turbance of sediment packing and possible formation of air Measurement of stress partitioning in the field typically in-
pockets when the probe is inserted. Therefore, many TDR volves the measurement of the vertical wind profile using
applications use semifixed probe installations (Siddiqui and anemometers, Irwin sensors to measure both tS and t, and
Drnevich, 1995). We do not know of any TDR applications to drag balances to measure the surface shear attributed to the
measure sand surface moisture content for aeolian studies. roughness elements (tR). The Irwin sensors are simple omni-
This is likely due to problems using TDR within about 0.15 m directional, skin-friction meters capable of measuring the
of discontinuities such as soil horizons or the sand surface pressure gradient at the surface (Irwin, 1980) which is used to
(Robinson and Dean, 1993). However, there are other geo- estimate the surface shear stress in complex flows at frequen-
morphological applications where detailed knowledge of soil cies greater than 10 Hz (Gillies et al., 2007). The sensor is
moisture is important. Adaptations of the TDR concept are mounted flush with the surface and the differential pressure is
also frequently used to complement the use of ground pene- measured by a pressure transducer connected to ports at the
trating radar (Gerber et al., 2007; Sambuelli et al., 2010). surface and 0.00175 m above the surface (see Brown et al.,
FDR measures the ability of the soil (including soil mois- 2008).
ture) to complete a capacitance circuit between two electrodes. The drag force applied to nonerodible roughness elements
Variations in capacitance are related to the soil (or sand) di- is measured using an omnidirectional force balance consisting
electric constant. Capacitance probes are much less expensive of two shear plates mounted orthogonally on top of each
than TDR probes and are capable of sampling through rela- other (see Wyatt and Nickling, 1997; Gillies et al., 2000).
tively small volumes (e.g., Dean et al., 1987; or Eller and Electronic load cells measure the displacement of the rough-
Denoth, 1996). One drawback to the widespread use of cap- ness element and produce a voltage proportional to that dis-
acitance probes, especially in coastal environments, is sensi- placement. The load cell is calibrated in the laboratory using
tivity to salinity and temperature, typically requiring localized static loads of known weights pulling through the center of the
calibration (Kelleners et al., 2004). force plate. Use of both the Irwin and force balance sensors in
The theta probe was developed by the Macaulay Land Re- the field has permitted the development of drag coefficients
search Institute in Aberdeen, Scotland (according to Atherton, for desert shrubs (e.g., Wyatt and Nickling, 1997; Gillies et al.,
 Electronic Measurement Techniques for Field Experiments in Process Geomorphology 213

2000) and validation of drag relationships observed in la- Table 2 The distinctive hydrodynamic signatures of types of wave
boratory wind tunnels (e.g., Gillies et al., 2007). Namikas common in the surf zone, based on the relationships between periodic
(2002) successfully developed a drag (shear) plate that floated across shore (u) and alongshore (v) velocities, and water surface
on water and was linked to four load cells. In a field experi- fluctuations (Z)
ment, he found that the drag plate replicated shear stress Wave type u–v u–Z v–Z
measurements obtained from velocity profiles over averaging
intervals of several minutes. The drag plate, however, could Progressive gravity In phase In phase In phase
respond to shear stress fluctuations at intervals as short as 1 s. Standing gravity Quadrature Quadrature In phase
The difficulty in building and deploying such devices has Progressive edge Quadrature Quadrature In phase
limited their use, despite the advantages that they hold over Standing edge In phase Quadrature Quadrature
other means of shear stress measurement.
transport events (Ellis et al., 2012). Nonstationary time series
can be decomposed using wavelet analysis, which quantifies
14.17.4 Interpreting the Signal how the spectral signal varies over a range of temporal scales,
rather than the overall strength of the signal at specific scales.
With few exceptions, the instruments described above output a
Deployment of an instrument array and/or the colocation
time series of measurements of flow velocity, sediment trans-
of different instruments allows for the covariation of processes
port, or a closely related variable showing fluctuating, oscil-
to be examined in space and time, respectively. For example,
lating, or trending components. In this section we provide just a
estimates of suspended sediment transport requires the col-
few examples of analytical approaches to using and interpreting
location (or integration) of instruments to measure both the
such time series. Interpretation of such time series may be as
flow velocity (e.g., velocimeter) and the sediment concen-
straightforward as averaging the data, or it can be considerably
tration (e.g., backscatterance sensor). The covariation can be
more complex. Consider, for instance, the potential analysis of
estimated using cospectral analysis for stationary time series
anemometer records. For many applications it is just a measure
and cross-wavelet analysis for nonstationary time series. Both
of mean wind speed that is required. Obtaining estimates of
analysis techniques describe the strength of the relationship
shear velocity requires a little more analysis. If wind profiles are
and the phase, a measure of the spatial and temporal separ-
being used, then average wind speeds from several carefully
ation of the signal between instruments. As a further example,
measured elevations are necessary, an appropriate averaging
measurements of surface fluctuations can be paired with bi-
interval must be chosen (e.g., Namikas et al., 2003), and re-
directional current measurements to distinguish the nature of
gression analysis performed and interpreted (e.g., Bauer et al.,
these (and other) important motions because they have dis-
1992). If the estimates are to be obtained from a sonic an-
tinct signatures, such as standing, reflected or edge waves
emometer, then a different sequence of steps is necessary, in-
(Table 2, based on Holman and Bowen, 1979). Individual
cluding digital rotation of the axes of measurement and
pressure transducer records are capable of indicating the
assessment of an appropriate analysis interval (e.g., van Boxel
presence of wave motions at periods longer than that of the
et al., 2004). Sonic (or hot film) anemometry also allows for
incident wave field, but they cannot determine the nature of
more detailed analysis of wind records, including Variable
the motion. The terminology ‘in phase’ indicates that the
Interval Averaging (e.g., Bauer et al., 1998) and quadrant an-
maximum fluctuations from the mean occur simultaneously
alysis for burst sweep structures (Leenders et al., 2005). More
in the paired time series. For example, with a progressive
recently, detailed, 3D airflow data from an array of sonic an-
gravity wave (the waves you see on beaches), the maximum
emometers were used to evaluate the performance of CFD
across shore (u) and alongshore (v) velocities, and water sur-
models (Jackson et al., 2011).
face fluctuations (Z), occur at the same time. With quadrature,
A similar example of increasing complexity of analysis is the
the maximum fluctuations are offset by a period equal to one
processing of water surface elevation data. If we consider a time
half of the wave period (see Komar, 1998 for more details).
series from a pressure transducer, several analytic options are
High-frequency, 3D vector measurements from current
available. An estimate of mean water depth is obtained by
meters and anemometers are routinely used to investigate
averaging the time series over an appropriate interval. A series of
turbulence dynamics and in particular to determine the Rey-
such averages can be used to reconstruct tidal cycles if the
nolds shear stress based on quadrant analysis, usually inter-
sample duration is long enough. If waves are present during the
preted in the context of the burst-sweep model. Each
sample period, the same record can also be analyzed to obtain
directional component of the flow vector can be written as the
estimates of wave height and period (e.g., Ellis and Sherman,
sum of a time-averaged speed and an instantaneous fluctu-
2005). The periodicity of the oscillations can be calculated
ating part according to a Reynolds’ decomposition:
through spectral analysis that decomposes the periodic signal
into sine and cosine functions. The spectra presented in Fig- uðtÞ ¼ u þ u0 ðtÞ ½9
ure 2 exhibit distinct peaks at 16 and 14 s respectively, but also
low-frequency oscillations at 143 and 28 s that would be dif- vðtÞ ¼ v þ v0 ðtÞ ½10
ficult or impossible to quantify, or perhaps even discern,
through visual observation. Spectral analysis assumes that the
wðtÞ ¼ w þ w0 ðtÞ ½11
variation in the process is stationary (or unchanging)
throughout the record, which is not appropriate for such things where u, v, and w denote the streamwise (x), spanwise (y), and
as vessel-generated waves and very high frequency sediment vertical (z) flow vector components, respectively, the overbar
214 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

indicates a time average, and the prime (0 ) indicates the fluc- reflect a fundamental change and improvement in our ability
tuating part. The fluctuating parts carry the turbulent mo- to quantify earth surface processes. This has promoted the
mentum flux, thus exerting a shear stress in the fluid that is shift in geomorphology from qualitative descriptions of the
transferred to the sediment surface. This so-called Reynolds’ Earth’s surface to hypothesis testing of process response re-
shear stress is defined as sponsible for landform evolution. In large part, there has been
a tendency toward the development and utilization of sensors
t ¼  ra u0 w0 ½12 capable of measuring near-bed turbulence and transport
processes, which require instruments that are small, sensitive,
where ra is the density of the fluid and u0 w0 is the covariance and rugged. At the same time, there has also been develop-
between the horizontal and vertical fluctuations over a given ment of sensors that elucidate the development and behavior
averaging period. By definition, the shear velocity, U, based of coherent flow and transport structures across several scales
on the Reynolds decomposition, can then be directly ex- of interest.
pressed as Modern field instrumentation systems marry advances in
 1=2 sensor design, data acquisition systems, and robust data
U ¼ u0 w0 ½13 storage and processing (computer) systems. The twenty first
century geomorphologist is able to design prototype experi-
The estimation of turbulent fluxes based on the Reynolds ments that would have been difficult or impossible 50 years
decomposition is only strictly correct when the fluctuating ago. These same advances have allowed much greater and
components are perpendicular to the local streamline vector. certainly much easier access to sophisticated systems and a
The original flow measurements thus usually require a trans- much greater ability to examine complex experimental and
formation to correct for misalignments between the orien- theoretical questions. Because it is nature where our geo-
tation of the current meter or anemometer head and the flow morphological lenses must focus, it is nature that our experi-
streamline. The usual transformation method is based on data ments must examine. The best laboratory is outside.
rotations (also called the 2–3 rotation method) to correct for
the yaw, pitch, and roll of the flow. Yaw correction aligns the
u and v components to the streamline vector by forcing the References
mean spanwise component to zero (i.e., v 0 ¼ 0), and sub-
sequent pitch correction aligns the u and w components so Aagaard, T., Masselink, G., 1999. The surf zone. In: Short, A.D. (Ed.), Handbook of
that the mean vertical component is zero. While these two Beach and Shoreface Morphodynamics. Wiley, Chichester, pp. 72–118.
Agrawal, Y.C., Pottsmith, H.C., 2000. Instruments for particle size and
data rotations are generally always applied, roll correction
settling velocity observations in sediment transport. Marine Geology 168,
requires a more substantial assumption that the spanwise 89–114.
shear stress along the streamline should revert to zero Allan, J.C., Hart, R., Tranguili, J.V., 2006. The use of passive integrated transport
(v0 w0 ¼ 0), which may not always be justified in flows (PIT) tags to trace cobble transport in a mixed sand-and-gravel beach on the
over more complex surfaces (Kaimal and Finnigan, 1994; Van high-energy Oregon coast, USA. Marine Geology 232, 63–86.
Allison, M.A., Nittrouer, C.A., Kineke, G.C., 1995. Seasonal sediment storage on
Boxel et al., 2004). An alternative data transformation, the mudflats adjacent to the Amazon River. Marine Geology 125, 303–328.
planar-fit method, has been developed by Wilczak et al. Alvarez, L.G., Jones, S.E., 2002. Factors influencing suspended sediment flux
(2001), based on the notion that a series of mean velocity in the Upper Gulf of California. Estuarine Coastal Shelf. Science 54,
vectors from varying directions define a tangent plane for all 747–759.
Amoser, S., Ladich, F., 2005. Are hearing sensitivities of freshwater fish adapted to
local streamlines. This plane can be found statistically from
the ambient noise in their habitats. Journal of Exploratory Biology 18,
fitting a least-squares regression to the set of directionally 3533–3542.
variable vectors, and the turbulent fluxes can subsequently be Atherton, R.J., Baird, A.J., Wiggs, G.F.S., 2001. Inter-tidal dynamics of surface
calculated perpendicular to this plane. A comprehensive moisture content on a mesotidal beach. Journal of Coastal Research 17,
review and analysis of both data transformation methods is 482–489.
Aubrey, D.G., Trowbridge, J.H., 1985. Kinematic and dynamic estimates from
given by Finnigan (2004).
electromagnetic current meter data. Journal of Geophysical Research 90,
9137–9146.
Austin, M.J., Masselink, G., O’Hare, T., Russell, P.E., 2007. Relaxation time effects
14.17.5 Conclusions of wave ripples on tidal beaches. Geophysical Research Letters 34, L16606. http
://dx.doi.org/10.1029/2007GL030696.
Austin, M.J., Scott, T.M., Brown, J.W., Brown, J.A., MacMahan, J.H., 2009.
It is recognized that this summary may be incomplete, as there Macrotidal rip current experiment: circulation and dynamics. Journal of Coastal
are many other instruments that have been used in the field Research SI 56, 24–28.
and others that have the potential to become standards of field Austin, M.J., Scott, T.M., Brown, J.W., Brown, J.A., MacMahan, J., Masselink, G.,
work (such as laser applications, etc.). Further, we have given Russell, P.E., 2010. Macro-tidal rip current dynamics. Continental Shelf
Research 30, 1149–1165.
little attention to the critical advances in data acquisition
Baas, A.C.W., 2004. Evaluation of saltation flux impact responders (Safires) for
systems and related hardware and software, all of which measuring instantaneous aeolian sand transport intensity. Geomorphology 17,
contribute importantly to greater sampling versatility. This is 482–489.
especially the case with regard to increases in sampling rate Baas, A.C.W., Sherman, D.J., 2005. The formation and behavior of aeolian
and duration. The various types of electronic field instrument streamers. Journal of Geophysical Research 110, F03011. http://dx.doi.org/
10.1029/2004JF000270.
systems reviewed herein have increased our capacity to in- Baas, A.C.W., Van den Berg, F., (2010). Particle image velocimetry (PIV) monitoring
vestigate sediment transport and fluid flows in coastal, fluvial, of Aeolian sand transport over a beach. Abstract of the International Conference
and aeolian environments. More importantly, these sensors on Aeolian Research, 2.
 Electronic Measurement Techniques for Field Experiments in Process Geomorphology 215

Bai, X., Li, S., Gong, D., Xu, Y., Jiang, J., 2009. Estimation of suspended sediment Butterfield, G.R., 1999b. Near-bed mass flux profiles in aeolian sand transport:
concentrations using pulse-coherent Acoustic Doppler profiler (PCADP). Chinese high-resolution measurements in a wind tunnel. Earth Surface Processes and
Journal of Oceanology and Limnology 27, 260–265. Landforms 24, 393–412.
Baird, A.J., Horn, D.P., 1996. Monitoring and modelling ground water behaviour in Cacchione, D.A., Wiberg, P.L., Lynch, J., Irish, J., Traykovski, P., 1999. Estimates of
sandy beaches. Journal of Coastal Research 12, 630–640. suspended-sediment flux and bedform activity on the inner portion of the Eel
Banas, N.S., MacCready, P., Hickey, B.M., 2009. The Columbia River plume as continental shelf. Marine Geology 154, 83–97.
cross-shelf exporter and along-coast barrier. Continental Shelf Research 29, Callede, J., Kosuth, P., Guyot, J.-L., Guimarães, V.S., 2000. Discharge determination
292–301. by Acoustic Doppler Current Profilers (ADCP): a moving bottom error correction
Battisto, G.M., 2000. Field measurement of mixed grain size suspension in the method and its application on the River Amazon at Óbidos. Hydrological
nearshore under waves. MS thesis, Faculty of the School of Marine Science, Sciences Journal 45, 911–924.
College of William and Mary, 88 pp. Camp, D.W., Turner, R.E., Gilchrist, L.P., 1970. Response tests of cup, vane, and
Battisto, G.M., Friedrichs, C.T., Miller, H.C., Resio, D.T., 1999. Response of OBS to propeler wind sensors. Journal of Geophysical Research 75, 5265–5270.
mixed grain-size suspension during Sandyduck ’97. Coastal Sediments, Carling, P.A., Williams, J.J., Gölz, E., Kelsey, A.D., 2000. The morphodynamics of
297–312. fluvial sand dunes in the River Rhine, near Mainz, Germany. II. Hydrodynamics
Bauer, B.O., Lorang, M.S., Sherman, D.J., 2002. Estimating boat-wake-induced and Sediment Transport. Sedimentology 47, 253–278.
levee erosion using suspended sediment measurements. Journal of Waterway, Chen, S., Geyer, W.R., Sherwood, C.R., Ralson, D.K., 2010. Sediment transport and
Port, Coastal and Ocean Engineering 128, 152–162. deposition on a river-dominated tidal flat: an idealized model study. Journal of
Bauer, B.O., Namikas, S.L., 1998. Design and field test of a continuously weighing, Geophysical Research, 115. http://dx.doi.org/10.1029/2010JC006248.
tipping-bucket assembly for aeolian sand traps. Earth Surface Processes and Cornelis, W.M., Gabriels, D., 2003. A simple low-cost sand catcher for wind-tunnel
Landforms 23, 1171–1183. simulations. Earth Surface Processes and Landforms 28, 1033–1041.
Bauer, B.O., Sherman, D.J., Wolcott, J.F., 1992. Sources of uncertainty in shear Cornish, V., 1934. Ocean Waves and Kindred Phenomena. Cambridge University
stress and roughness length estimates derived from velocity profiles. Press, Cambridge, pp. 164.
Professional Geographer 44, 453–464. Crawford, A.M., Hay, A.E., 1993. Determining suspended sand size and
Bauer, B.O., Yi, J., Namikis, S.L., Sherman, D.J., 1998. Event detection and concentration multifrequency acoustic backscatter. Journal of the Acoustical
conditional averaging in unsteady aeolian systems. Journal of Arid Environments Society of America 94(6), 3312–3324.
39, 345–375. Crawley, D.M., Nickling, W.G., 2003. Drag partition for regularly-arrayed rough
Bänziger, R., Burch, H., 1990. Acoustic sensors (hydrophones) as indicators for bed surfaces. Boundary-Layer Meteorology 107(2), 445–468.
load transport in a mountain torrent. IAHS Publication 193, 207–214. Creed, E.L., Pence, A.M., Rankin, K.L., 2001. Inter-comparison of turbidity and
Barchyn, T.E., Hugenholtz, C.H., 2010. Field comparison of four piezoelectric sediment concentration measurements using an ADP, an OBS-3 and a LISST.
sensors for detecting aeolian sediment transport. Geomorphology 120, Oceans 3, 1750–1754.
368–371. Curtiss, G.M., Osborne, P.D., Horner-Devine, A.R., 2009. Seasonal patterns
Beach, R.A., Sternberg, R.W., 1988. Suspended sediment in the surf zone: response of coarse sediment transport on a mixed sand and gravel beach due
to cross-shore infragravity motion. Marine Geology 80, 61–79. to vessel wakes, wind waves, and tidal currents. Marine Geology 259,
Beach, R.A., Sternberg, R.W., Johnson, R., 1992. A fiber optic sensor for 73–85.
monitoring suspended sediment. Marine Geology 103, 513–520. Custer, S.G., Ergenzinger, P.E., Bugosh, N., Anderson, B.C., 1987. Electromagnetic
Bedeus, K., Ivicsics, L., 1963. Observation of the noise of bedload. In: land Erosion, detection of pebble transport in streams: a method for measurement of sediment
precipitations, hydrometry, soil moisture (Proc. Berkeley Assembly, August 1963). transport waves. In: Ethridge, F., Flores, R. (Eds.), Recent Developments in
Publication of International Association of Scientific Hydrology 65, 384–390. Fluvial Sedimentology. Society of Paleontologists and Mineralogists, Special
Bell, P.S., Thorne, P.D., 1997. Application of a high resolution acoustic scanning Publication, Tulsa, Oklahoma, vol. 39, pp. 21–26.
system for imaging sea bed microtopography. Seventh International Conference Dalrymple, R.A., MacMahan, J.H., Reniers, J.H.M., Nelko, V., 2010. Rip currents.
on Electronic Engineering in Oceanography, Southampton Oceanographic Centre Annual Review of Fluid Mechanics 43, 551–581.
23–25 June 1997. IEE Conference Publication No. 435, pp. 128–133. Dalton, F.N., Herkelrath, W.N., Rawlins, D.S., Rhoades, J.D., 1984. Time-domain
Best, J.L., Roy, A.G., 1991. Mixing-layer distortion at the confluence of unequal reflectometry: simultaneous measurement of soil water content and electrical
depth channels. Nature 350, 411–413. conductivity with a single probe. Science, . 989–990.
Blenkinsopp, C.E., Turner, I.L., Masselink, G., Russell, P.E., 2010. Validation of Dasberg, S., Dalton, F.N., 1985. Field measurement of soil water content and bulk
volume continuity method for estimation of cross-shore swash flow velocity. electrical conductivity with time-domain reflectometry. Soil Science Society of
Coastal Engineering 57, 953–958. America Journal 49, 293–297.
Bogen, J., Men, K., 2003. Bedload measurements with a new passive acoustic Davis, J.L., Chudobiak, W.J., 1975. In situ meter for measuring relative permittivity
sensor. Erosion and Sediment Transport Measurement in Rivers: Technological of soils. Geological Surveyof Canada, Paper 75-1, pp. 75–79 Part A.
and Methodological Advances: Workshop, Oslo, Norway, pp. 181–192. Davis, W.M., 1909. Geographical Essays. In: Johnson, D.W. (Ed.), Ginn and Co.,
Bohren, C.F., Huffman, D., 1983. Absorption and Scattering of Light by Small Boston, 777.
Particles. Wiley, New York, 530 pp. Davidson-Arnott, R.G.D., Bauer, B.O., 2009. Aeolian sediment transport on a beach:
van Boxel, J.H., Sterk, G., Arens, S.M., 2004. Sonic anemometers in aeolian thresholds, intermittency and high frequency variability. Geomorphology 105,
sediment transport research. Geomorphology 59, 131–147. 117–125.
Bradley, K., Houser, C., 2009. Relative velocity of seagrass blades: implications for Davidson-Arnott, R.G.D., Yang, Y., Ollerhead, J., Hesp, P.A., Walker, I.J., 2008.
wave attenuation in low energy environments. Journal of Geophysical Research, The effects of surface moisture on aeolian sediment transport threshold
Earth Surface 114, F01004. http://dx.doi.org/10.1029/2007JF000951. and mass flux on a beach. Earth Surface Processes and Landforms 33,
Brander, R.W., Short, A.D., 2000. Morphodynamics of a large-scale rip current 55–74.
system at Muriwai Beach, New Zealand. Marine Geology 165, 27–39. Dean, T.J., Bell, J.P., Baty, A.J.B., 1987. Soil moisture measurement by an improved
Brander, R.W., Short, A.D., 2001. Flow kinematics of low-energy rip current capacitance technique, Part I. Sensor design and performance. Journal of
systems. Journal of Coastal Research 17(2), 468–481. Hydrology 93, 67–78.
Bray, D.I., 1975. Representative discharges for gravel-bed rivers in Alberta, Canada. Dingler, J.R., Boylls, J.C., Lowe, R.L., 1977. A high-frequency sonar for profiling
Journal of Hydrology 27, 143–153. small-scale subaqueous bedforms. Marine Geology 24, 279–288.
Brown, S., Nickling, W.G., Gillies, J.A., 2008. A wind tunnel examination of shear Dong, Z., Wang, H., Liu, X., Zhao, A., 2002. Velocity profile of a sand cloud
stress partitioning for an assortment of surface roughness distributions. Journal blowing over a gravel surface. Geomorphology 45, 277–289.
of Geophysical Research 113, F02S06. http://dx.doi.org/10.1029/ Dornbusch, U., Williams, R.B.G., Moses, C.A., Robinson, D.A., 2002. Life
2007JF000790. expectancy of shingle beaches: measuring in situabrasion. Journal of Coastal
Bunte, K., Ergenzinger., P., 1989. New tracer techniques for particles in Research Special Issue 36, 249–255.
gravel bed rivers. Bulletin De La Societe Geographique De Liege 25. Downing, J., 2006. Twenty-five years with OBS sensors – the good, the bad, and
pp 85–90. the ugly. Continental Shelf Research 2006(26), 2299–2318.
Butterfield, G.R., 1999a. Application of thermal anemometry and high-frequency Downing, J.P., Sternberg, R.W., Lister, C.R.B., 1981. New instrumentation for the
measurement of mass flux to aeolian sediment transport research. investigation of sediment suspension processes in the shallow marine
Geomorphology 29, 31–58. environment. Marine Geology 42, 19–34.
216 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

Edwards, B.L., Namikas, S.L., 2009. Small-scale variability in surface moisture on a Gillette, D.A., Fryrear, D.W., Xiao, J.B., 1997. Large-scale variability of wind erosion
fine-grained beach: implications for modeling aeolian transport. Earth Surface mass flux rates at Owens Lake; 1. Vertical profiles of horizontal mass fluxes of
Processes and Landforms 34, 1333–1338. wind-eroded particles with diameter greater than 50 mm. Journal of Geophysical
Elgar, S., Raubenheimer, B., Guza, R.T., 2005. Quality control of acoustic Doppler Research 102, 25977–25987.
velocimeter data in the surfzone. Measurement Science and Technology 16, Gillies, J.A., Lancaster, N., Nickling, W.G., Crawley, D.M., 2000. Field determination
1889–1893. of drag forces and shear stress partitioning effects for a desert shrub
Eller, H., Denoth, A., 1996. A capacitive soil moisture sensor. Journal of Hydrology (Sarcobatus vermiculatus, greasewood). Journal of Geophysical Research 105,
185, 137–146. 24871–24880.
Ellis, J.T., 2006. Coherent structures and aeolian saltation. PhD dissertation, Texas Gillies, J.A., Nickling, W.G., King, J., 2007. Shear stress partitioning in large
A&M University, College Station, Texas, 139 pp. patches of roughness in the atmospheric inertial sublayer. Boundary-Layer
Ellis, J.T., Morrison, R.F., Priest, B.H., 2009. Measuring the transport of aeolian Meteorology 122, 367–396 http://dx.doi.org/10.1007/s10546-006-9101-5.
sand with a microphone system. Geomorphology 105, 87–94. Goossens, D., Offer, Z., London, G., 2000. Wind tunnel and field calibration of five
Ellis, J.T., Sherman, D.J., 2005. Effects of sampling frequency on wave aeolian sand traps. Geomorphology 35, 233–252.
characterization. Zeitschrift fur Geomorphologie, Supplementband, Coasts Under Goudie, A., 1990. Geomorphological Techniques. Routledge, New York, 570 pp.
Stress II SV141, 183–195. Granata, T.C., Serra, T., Colomer, J., Casamitjana, X., Duart, C.M., Garcia, E., 2001.
Ellis, J.T., Sherman, D.J., Farrell, E.J., Li, B., 2012. Horizontal variability of aeolian Flow and particle distributions in a nearshore seagrass meadow before and after
sand transport. Aeolian Research 3, 379–387. a storm. Marine Ecology Progress Series 218, 95–106.
Erlingsson, U., 1991. A sensor for measuring erosion and deposition. Journal of Gray, J.R., Gartner, J.W., 2010. Surrogate technologies for monitoring suspended-
Sedimentary Research 61, 620–623. sediment transport in rivers. In: Poleto, Cristiano, Charlesworth, Susanne (Eds.),
Ergenzinger, P., Conrady, J., 1982. A new tracer technique for measuring bedload in Sedimentology of Aqueous Systems. Wiley-Blackwell, London, pp. 3–45
natural channels. Catena 9, 77–80. Chapter 1.
Ergenzinger, P., Custer, S.G., 1983. Determination of bedload transport using Gray, J.R., Hornewer, N.J., Larsen, M.C., Fisk, G.G., Macy, J.P., 2010. Pressure
naturally magnetic tracers: first experiences at Squaw Creek, Gallatin County, difference. In: Gray, J.R., Gartner, J.W. (Eds.), Section 1.2.4, Chapter 1,
Montana. Water Resources Research 19, 187–193. Surrogate Technologies for Monitoring Suspended-sediment Transport in Rivers:
Farrell, E.J., Granja, H., Cappietti, L., Ellis, J.T., Li, B., Sherman, D.J., 2009. Wave Poleto, Cristiano, and Charlesworth, Susanne, eds., Sedimentology of Aqueous
transformation across a rock platform, Belinho, Portugal. Journal of Coastal Systems. Wiley-Blackwell, London, pp. 23–28.
Research SI56, 44–48. Green, M.O., Boon, J.D., 1988. Response characteristics of a short-range, high-
Fedderson, F., Guza, R.T., 2003. Observations of nearshore circulation: alongshore resolution digital sonar altimeter. Marine Geology 81, 197–203.
uniformity. Journal of Geophysical Research 108, 1–10. Green, M.O., Vincent, C.E., McCave, I.N., Dickson, R.R., Rees, J.M., Pearson, N.D.,
Filizola, N., Guyot, J.L., 2004. The use of Doppler technology for suspended 1995. Storm sediment transport: observations from the British North Sea shelf.
sediment discharge determinations in the Amazon. Hydrologic Science Journal Continental Shelf Research 15, 889–912.
49, 143–153. Greenwood, B., Dingler, J.R., Sherman, D.J., Anima, R.J., Bauer, B.O., 1985.
Filizola, N., Guyot, J.L., Guimaraes, V., 2009. Measuring the discharge of the Monitoring bedforms under waves using High Resolution Remote Tracking
Amazon River using Doppler technology (Manacapuru, Amazonas, Brazil). Sonar (HRRTS). Proceedings of the Canadian Coastal Conference. Ottawa,
Hydrological Processes 23, 3151–3156. Canada, pp 143–158.
Finkelstein, P.L., 1981. Measuring the dynamic performance of wind vanes. Journal Greenwood, B., Jagger, K., 1995. Sensitivity of optical sensors to grain-size
of Applied Meteorology 20, 588–594. variation in the sand mode: Implications for transport measurements.
Finnigan, J.J., 2004. A re-evaluation of long-term flux measurement techniques Part Proceedings of 1995 Canadian Coastal Conference, Dartmouth, Canadian
II: coordinate systems. Boundary-layer Meteorology 113, 1041. Coastal Science and Engineering Association, NRC, Ottawa, p. 383–398.
Foster, D.L., Bowen, A.J., Holman, R.A., Natoo, P., 2006. Field evidence of pressure Greenwood, B., Sherman, D.J., 1984. Waves, currents, sediment flux, and
gradient induced incipient motion. Journal of Geophysical Research, 111. http:// morphological response in a barred nearshore system. Marine Geology 60,
dx.doi.org/10.1029/2004JC002863. 31–61.
Foucaut, J.-M., Stanislas, M., 1997. Experimental study of saltating particle Greenwood, B., Xu, Z., 2001. Size fractionation by suspension transport: a
trajectories. Experiments in Fluids 22, 321–326. large scale flume experiment with shoaling waves. Marine Geology 176,
Froehlich, W., 2003. Monitoring of bedload transport by use of acoustic and 157–174.
magnetic device, in erosion and sediment transport measurement in rivers – Guy, H.P., Norman, V.M., 1970. Field methods for measurement of fluvial sediment:
technological and methodological advances, publication 283. IAHS, 201–210 U.S. Geological Survey Techniques of Water-Resources Investigations, book 3,
Frothingham, K.M., Rhoads, B.L., 2003. Three-dimensional flow structure and chapter C2, 59.
channel change in an asymmetrical compound meander loop, Embarras Riverr, Hale, P.B., McCan, S.B., 1982. Rhythmic topography in a mesotidal low-wave-
Illinois. Earth Surface Processes and Landforms 28, 625–644. energy environment. Journal of Sedimentary Research 52, 415–429.
Fryberger, S.G., Ahlbrandt, T.A., Andrews, S., 1979. Origin, sedimentary features, Hamilton, L.J., Shi, Z., Zhang, S.Y., 1998. Acoustic backscatter measurements of
and significance of low-angle eolian sand sheet deposits, Great Sand Dunes estuarine cohesive sediment concentration profiles. Journal of Coastal Research
National Monument and vicinity, Colorado. Journal of Sedimentary Petrology 49, 14, 1213–1224.
733–746. Hardisty, J., 1993. Monitoring and modelling sediment transport at turbulent
Fugate, D.C., Friedrichs, C.T., 2002. Determining concentration and fall velocity of frequencies. In: Clifford, N.J., French, J.R., Hardisty, J. (Eds.), Turbulence;
estuarine particle populations using ADV, OBS, and LISST. Continental Shelf Perspective on Flow and Sediment Ttransport. Wiley, New York, NY, pp. 35–59.
Research 22, 1867–1886. Hassan, M.A., Church, M., Rempel, J., Enkin, R.J., 2009. Promise, performance and
Gallagher, E.L., Boyd, W., Elgar, S., Guza, R.T., Woodward, B., 1996. Performance of current limitations of a Magnetic Bedload Movement Detector. Earth Surface
a sonar altimeter in the nearshore. Marine Geology 133(3–4), 241–248. Processes and Landforms 34, 1022–1032.
Gallagher, E.L., Thornton, E.B., Stanton, T.P., 2003. Sand bed roughness in the Hassan, M.A., Ergenzinger, P., 2003. Tracers in fluvial geomorphology. In: Kondolf,
nearshore. Journal of Geophysical Research 108, 1–8. G.M., Piegay, H. (Eds.), Fluvial Morphology. Wiley, Chichester, pp. 397–423.
Garcia, C.M., Oberg, K., Garcia, M.H., 2007. ADCP measurements of gravity currents Hatcher, A., Hill, P., Grant, J., Macpherson, P., 2000. Spectral optical backscatter of
in the Chicago River, Illinois. Journal of Hydraulic Engineering 133, 1356–1366. sand in suspension: effects of particle size, composition and, color. Marine
Gartner, J.W., Ganju, N.K., 2003. A preliminary evaluation of near-transducer Geology 168, 115–128.
velocities collected with low-blank acoustic Doppler current profiler. Proceedings Hay, A.E., 2008. Near-bed turbulence and relict wave-formed sand ripples:
of Hydraulic Measurements and Experimental Methods Conference, Estes Park, Observations from the inner shelf. Journal of Geophysics Research 113,
CO, ASCE, CD-ROM. C04040. http://dx.doi.org/10.1029/2006JC004013.
Gerber, R., Salat, C., Junge, A., Felix-Henningsen, P., 2007. GPRbased detection of Hay, A.E., Wilson, D.J., 1994. Rotary sidescan images of nearshore bedform
Pleistocene periglacial slope deposits at a shallow-depth test site. Geoderma evolution during a storm. Marine Geology 119, 57–65.
139, 346–356. Helley, E.J., Smith, W., 1971. Development and Calibration of a Pressure-Difference
Gilbert, G.K., 1886. The inculcation of the scientific method by example, with an Bed Load Sampler. US. Geological Survey Open File Report, Washington, USA.
illustration drawn from the Quaternary geology of Utah. American Journal of Hjulström, F., 1935. Studies of the morphological activities of rivers illustrated by
Science 3, 1–13. the river Fyris. Uppsala University Geological Institute Bulletin 25, 221–527.
 Electronic Measurement Techniques for Field Experiments in Process Geomorphology 217

Holdaway, G.P., Thorne, P.D., Flatt, D., Jones, S.E., Prandle, D., 1999. Comparison Kostaschuk, R.A., Best, J.L., Villard, P.V., Peakall, J., Franklin, M., 2005. Measuring
between ADCP and transmissometer measurements of suspended sediment flow velocity and sediment transport with an acoustic Doppler current profiler.
concentration. Continental Shelf Research 19, 421–441. Geomorphology 68, 25–38.
Holland, K.T., Raubenheimer, B., Guza, R.T., Holman, R.A., 1995. Runup kinematics Kostaschuk, R.A., Church, M.A., 1993. Macroturbulence generated by dunes: Fraser
on a natural beach. Journal of Geophysical Research 100, 4985–4993. River, Canada. Sedimentary Geology 85, 25–37.
Holman, R.A., Bowen, A.J., 1979. Edge waves on complex beach profiles. Journal Kostaschuk, R.A., Villard, P.V., 1996. Flow and sediment transport over large
of Geophysical Research 84, 6339–6346. subaqueous dunes: Fraser River, Canada. Sedimentology 43, 849–863.
Holman, R.A., Guza, R.T., 1984. Measuring run-up on a natural beach. Coastal Kostaschuk, R.A., Shugar, D., Best, J., Parsons, D., Lane, S., Hardy, R., Orfeo, O.,
Engineering 8, 129–140. 2009. Suspended sediment transport and deposition over a dune: Rio Parana,
Horn, D.P., Baldock, T., Baird, A.J., Mason, T., 1998. Field measurements of swash Argentina. Earth Surface Processes and Landforms 34, 1605–1611.
induced pressure gradients within a sandy beach. Proceedings, 26th Kraus, G., Ohm, K., 1984. A method to measure suspended load transport in
International Conference on Coastal Engineering (ASCE), pp. 2812–2825. estuaries. Estuarine, Coastal, and Shelf Science 19, 611–618.
Houser, C., 2010. Relative importance of vessel and wind waves to salt marsh Kraus, N.C., 1987. Applications of portable traps for obtaining point measurements
erosion at Fort Pulaski National Monument. Journal of Coastal Research 26, of sediment transport rates in the surf zone. Journal of Coastal Research 3,
230–240. 139–152.
Houser, C., Barrett, G., 2010. Divergent behavior of the swash zone in response Krein, A., Klinck, H., Eiden, M., Symader, W., Bierl, R., Hoffmann, L., Pfister, L.,
to different foreshore slopes and nearshore states. Marine Geology 271, 2008. Investigating the transport dynamics and the properties of bedload
106–118. material with a hydro-acoustic measuring system. Earth Surface Processes and
Houser, C., Greenwood, B., 2005. Hydrodynamics and sediment transport within the Landforms 33, 152–163.
inner surf zone of a lacustrine multiple-barred nearshore. Marine Geology 218, Kringel, K., Jumars, P.A., Holliday, D.V., 2003. A shallow scattering layer: high-
37–63. resolution acoustic analysis of nocturnal vertical migration from the seabed.
Houser, C., Greenwood, B., Aagaard, T., 2006. Divergent response of an intertidal Limnology and Oceanography 48, 1223–1234.
swash bar. Earth Surface Processes and Landforms 31, 1775–1791. Kuhnle, R.A., 1992. Bed load transport during rising and falling stages on two
Houser, C., Hill, P., 2010a. Wave attenuation across an intertidal sand flat: small streams. Earth Surface Processes and Landforms 17(2), 191–197.
implications for mudflat development. Journal of Coastal Research 26, 403–411. Kuhnle, R.A., Horton, J.K., Bennett, S.J., Best, J.L., 2006. Bedforms in bimodal
Houser, C., Hill, P., 2010b. Sediment transport on roberts bank: a sandy intertidal sand-gravel sediments: laboratory and field analysis. Sedimentology 53,
flat on the fraser river delta. Journal of Coastal Research 26, 333–341. 631–654.
Hubbard, E.F., Scwartz, G.E., Thibodeaux, K.G., Turcios, L.M., 2001. Price current- Kuhnle, R.A., Willis, J.C., 1992. Mean size distribution of bed load on Goodwin
meter standard rating development by the U.S. Geological Survey. Journal of Creek. Journal of Hydraulic Engineering 118(10), 1443–1446.
Hydraulic Engineering 127, 250–257. Kunkel, G.J., Marusic, I., 2006. Study of the near-wall-turbulent region of the high-
Hughes, M.G., 1995. Friction factors for wave uprush. Journal of Coastal Research Reynolds-number boundary layer using an atmospheric flow. Journal of Fluid
11, 1089–1098. Mechanics 548, 375–401.
Irwin, H.P.A.H., 1980. A simple omnidirectional sensor for wind tunnel studies of Lacy, J.R., Sherwood, C.R., 2004. Accuracy of a pulse-coherent acoustic Doppler
pedestrian level winds. Journal of Wind Engineering and Industrial profiler in a wave-dominated flow. Journal of Atmospheric and Oceanic
Aerodynamics 7, 219–239. Technology 21(9), 1448–1461.
Jackson, D.W.T., 1996. A new, instantaneous aeolian sand trap design for field use. Lamarre, H., Macicar, , Roy, A.G., 2005. Using passive integrated transponder (PIT)
Sedimentology 43, 791–796. tags to investigate sediment transport in Gravel-Bed rivers. Journal of
Jackson, D.W.T., Beyers, J.H.M., Lynch, K., Cooper, J.A.G., Baas, A.C.W., Delgado- Sedimentary Research 75(4), 736–741.
Fernandez, I., 2011. Investigation of three-dimensional wind flow behavior over Lamarre, H., Roy, A.G., 2008. A field experiment on the development of sedimentary
coastal dune morphology under offshore winds using computational fluid structures in a gravelbed river. Earth Surface Processes and Landforms 33(7),
dynamics (CFD) and ultrasonic anemometry. Earth Surface Processes and 1064–1081.
Landforms 36, 1113–1124. Lane, S.N., Richards, K.S., Warburton, J., 1993. Comparison between high
Jackson, N.L., Nordstrom, K.F., 1997. Effects of timedependent moisture content of frequency velocity records obtained with spherical and discoidal electromagnetic
surface sediments on aeolian transport rates across a beach,Wildwood, New current meters. In: Clifford, N.J., French, J.R., Hardisty, J. (Eds.), Turbulence:
Jersey, U.S.A. Earth Surface Processes and Landforms 22, 611–621. Perspectives on Flow and Sediment Transport. Wiley, Chichester, pp. 121–163.
Jagger, K.A., Hadisty, J., 1991. Higher frequency acoustic measurements of coastal Law, D.J., Bale, A.J., Jones, S.E., 1997. Adaption of focused beam reflectance
bedload transport. Proceedings of Coastal Sediments ’91. American Society of measurement to in situ particle sizing in estuaries and coastal waters. Marine
Civil Engineers, NY, pp. 2187–2198. Geology 140, 47–59.
Jago, O.K., Kench, P.S., Brander, R.W., 2007. Field observations of wave-driven Lawler, D.M., 1991. A new technique for the automatic monitoring of erosion and
water-level gradients across a coral reef flat. Journal of Geophysical Research deposition rates. Water Resources Research 27, 2125–2128.
112. http://dx.doi.org/10.1029/2006JC003740. Lawler, D.M., 1992. Design and installation of a novel automatic erosion monitoring
Jonys, C.K., 1976. Acoustic Measurement of Sediment Transport. Scientific Series system. Earth Surface Processes and Landforms 17, 455–463.
No. 66, Inland Waters Directorate, CCIW Branch. Burlington, Ontario. Lawler, D.M., Couperthwaite, J., Bull, L.J., Harris, N.M., 1997. Bankerosion events
Kaimal, J.C., Finnigan, J.J., 1994. Atmospheric Boundary Layer Flows, their and processes in the Upper Severn basin. Hydrology and Earth System Sciences
Structure and Measurement. Oxford University Press, Oxford, p. 289. 1, 523–534.
Kaimal, J.C., Gaynor, J.E., Zimmerman, H.A., Zimmerman, G.A., 1990. Minimizing flow Lawler, D.M., Grove, J.R., Couperthwaite, J.S., Leeks, G.J.L., 1999. Downstream
distortion errors in a sonic anemometer. Boundary-Layer Meteor 53, 103–115. change in river bank erosion rates in the Swale-Ouse system, northern England.
Kelleners, T.J., Soppe, R.W.O., Ayars, J.E., Skaggs, T.H., 2004. Calibration of Hydrological Processes 13, 977–992.
capacitance probe sensors in a saline silty clay soil. Soil Science Society of Lawler, D.M., West, J.R., Couperthwaite, J.S., Mitchell, S.B., 2001. Application of a
America Journal 68, 770–778. novel automatic erosion and deposition monitoring system at a channel bank
Kineke, G., Sternberg, R.W., 1992. Measurements of high concentration suspended site on the tidal River Trent, UK, Estuarine. Coastal and Shelf Science 53(2),
sediment using the optical backscatterance sensor. Marine Geology 108, 237–247.
253–258. Leatherman, S.P., 1978. A new aeolian sand trap design. Sedimentology 25,
Kineke, G.C., Higgins, E.E., Hart, K., Velasco, D., 2006. Fine-sediment transport 303–306.
associated with cold-front passages on the shallow shelf, Gulf of Mexico. Lee, J.A., 1987. A field experiment on the role of small scale wind gustiness in
Continental Shelf Research 26, 2073–2091. aeolian sand transport. Earth Surface Processes and Landforms 12, 331–335.
Kirk, J.T.O., 1994. Light and Photosynthesis in Aquatic Ecosystems, Second ed. Leenders, J.K., van Boxel, J.H., Sterk, G., 2005. Wind forces and related saltation
Cambridge University Press, Cambridge and New York. transport. Geomorphology 7, 357–372.
Komar, P.D., 1998. Beach Processes and Sedimentation, Second ed. Prentice-Hall, Lewis, J., 1996. Turbidity-controlled suspended sediment sampling for runoff-event
Upper Saddle River, NJ, 544 pp. estimation. Water Resources Research 32, 2299–2310.
Kostaschuk, R.A., Best, J.L., 2005. The response of sand dunes to variations in Limeburner, R., Beardsley, R.C., Soares, I.D., Lentz, S.J., Candela, J., 1995.
tidal flow. Journal of Geophysical Research 110, F04S04. http://dx.doi.org/ Lagrangian flow observations of the Amazon River discharge into the North
10.1029/2004JF000176. Atlantic. Journal of Geophysical Research 100, 2401–2415.
218 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

Lohrmann, A., Hackett, B., Røed, L.P., 1990. High-resolution measurements of Nelson, R.C., Gonsalves, J., 1990. A field study of reflections from an exposed
turbulence, velocity and stress using a pulse-to-pulse coherent sonar. Journal of dissipative beach. Coastal Engineering 14, 457–477.
Atmospheric and Oceanic Technology 7, 19–37. Neumeier, U., Amos, C.L., 2006. The influence of vegetation on turbulence and flow
Lorke, A., Wuest, A., 2005. Application of coherent ADCP for turbulence velocities in European salt-marshes. Sedimentology 53, 259–277.
measurements in the bottom boundary layer. Journal of Atmosphere and Oceanic Ngusaru, A.S., Hay, A.E., 2004. Cross-shore migration of lunate megaripples during
Technology 22, 1821–1828. Duck94. Journal of Geophysical Research 109(C02006), 16 pp.
Lowe, R.J., Falter, J.L., Bandet, M.D., Pawlak, G., Atkinson, M.J., Monismith, S.G., Nickling, W.G., Ecclestone, M., 1981. The effects of soluble salts on the threshold
Koseff, J.R., 2005. Spectral wave dissipation over a barrier reef. Journal of shear velocity of fine sand. Sedimentology 28, 505–510.
Geophysical Research- Oceans, 110. http://dx.doi.org/10.1029/2004JC002711. Nielsen, P., Cowell, P.J., (1981). Calibration and data correction procedures for flow
Lu, L., Liu, Y., Sun, L., Xu, X., Yang, Y., Kong, L., 1981. The development of meters and pressure transducers commonly used by the Coastal Studies Unit.
nuclear sediment concentration gauges for use on the Yellow River, Proceedings Coastal Studies Unit, Department of Geography, University of Sydney, Technical
of Florence Symp. Erosion Sediment Transp. Meas., IAHS, Publ. No. 133 Report 81/1, 31 pp.
(1981). Nordstrom, K.F., Jackson, N.L., Allen, J.R., Sherman, D.J., 2003. Longshore
Lyles, L., Allison, B.E., 1976. Wind erosion: the protective role of simulated sediment transport rates on a microtidal estuarine beach. Journal of Waterway,
standing stubble. Transactions of the American Society of Agricultural Engineers Port, Coastal and Ocean Engineering 129, 1–4.
19(1), 61–64. Ogston, A.S., Sternberg, R.W., 2002. Effect of wave breaking on sediment eddy
MacMahan, J., Brown, J., Brown, J., et al., 2010. Mean lagrangian flow behavior diffusivity, suspended-sediment, and longshore flux profiles in the surf zone.
on an open coast rip-channeled beach: a new perspective. Marine Geology 268, Continental Shelf Research 22, 633–655.
1–15. Orzech, M.D., Thornton, E.B., MacMahan, J.H., O’Reilly, W.C., Stanton, T.P., 2010.
MacMahan, J., Brown, J., Thornton, E.B., 2009. Low-cost handheld global Alongshore rip channel migration and sediment transport. Marine Geology 271,
positioning systems for measuring surf zone currents. Journal of Coastal 278–291.
Research 10, 2112/08–1000. Osborne, P.D., 2005. Transport of gravel and cobble on a mixed-sediment inner
MacMahan, J., Stanton, T., Thornton, E.B., Reniers, A.J.H.M., 2005. bank shoreline of a large inlet, Grays Harbor, Washington. Marine Geology 224,
RIPEX-rip currents on a shore-connected shoal beach. Marine Geology 218, 145–156.
113–134. Osborne, P.D., Greenwood, B., 1992. Frequency dependent cross-shore suspended
MacVicar, B.J., Beaulieu, E., Champagne, V., Roy, A.G., 2007. Measuring water sediment transport. II: a barred shoreface. Marine Geology 106, 25–51.
velocity in highly turbulent flows: field tests of an electromagnetic current meter Osborne, P.D., Greenwood, B., 1993. Sediment suspension under waves and
(ECM) and an acoustic Doppler velocimeter (ADV). Earth Surface Processes and currents: time scales and vertical structure. Sedimentology 40, 599–622.
Landforms 132, 1412–1432. Ostrander, C.E., McManus, M.A., DeCarlo, E.H., MacKenzie, F.T., 2008. Temporal
Marshall, J.K., 1971. Drag measurements in roughness arrays of varying densities and spatial variability of freshwater plumes in a semienclosed estuarine-bay
and distribution. Agricultural Meteorology 8, 269–292. system. Estuaries and Coasts 31, 192–203.
Marusic, I., Heuer, W.D., 2007. Reynolds number invariance of the structure Puleo, J.A., 2009. Tidal variability of swash zone sediment suspension and
inclination angle in wall turbulence. Physical Reviews Letters 99, 1–4. transport. Journal of Coastal Research 25, 937–948.
Marusic, I., Hutchins, N., 2008. Study of the log-layer structure in wall turbulence Puleo, J.A., Faries, J., Davidson, M., Hicks, B., 2010. A conductivity sensor for
over a very large range of Reynolds number. Flow, Turbulence and Combustion nearbed sediment concentration profiling. Journal of Atmospheric and Oceanic
81, 115–130. Technology 27, 397–408.
Masselink, G., Austin, M., O’Hare, T., Russell, P.E., 2007. Geometry and dynamics Puleo, J.A., Holland, K.T., Plant, N.G., Slinn, D.N., Hanes, D.M., 2003. Fluid
of wave ripples in the nearshore zone of a coarse sandy beach. Journal of acceleration effects on suspended sediment transport in the swash zone. Journal
Geophysical Research 112, C10022. http://dx.doi.org/10.1029/2006JC003839. of Geophysical Research 108, 3350. http://dx.doi.org/10.1029/2003JC001943.
McDermott, J.P., Sherman, D.J., 2009. Using photo-electronic erosion pins for Puleo, J.A., Johnson, R.V., Butt, T., Kooney, T.N., Holland, K.T., 2006. The effect of
measuring bed elevation changes in the swash zone. Journal of Coastal bubbles on optical backscatter sensors. Marine Geology 230, 87–97.
Research 25, 788–792. Reichel, G., Nachtnebel, H.P., 1994. Suspended sediment monitoring in a fluvial
Meirovich, L., Laronne, J.B., Reid, I., 1998. The variation of water-surface slope and environment: advantages and limitations applying an acoustic Doppler current
its significance for bedload transport during floods in gravel-bed streams. profiler. Water Research 28, 751–761.
Journal of Hydraulic Research 36, 147–157. Reniers, A.J.H.M., MacMahan, J.H., Beron-Vera, F.J., Olascoaga, M.J., 2010. Rip-
Mikami, M., Yamada, Y., Ishizuka, M., Ishimaru, T., Gao, W., Zeng, F., 2005. current pulses tied to Lagrangian coherent structures. Geophysical
Measurement of saltation process over gobi and sand dunes in the Taklimakan ResearchLetters, 37. http://dx.doi.org/10.1029/2009GL041443.
desert, China, with newly developed sand particle counter. Journal of Reniers, A.J.H.M., MacMahan, J.H., Thornton, E.B., et al., 2009. Surfzone Surface
Geophysical Research, 11. http://dx.doi.org/10.1029/2004JD004688. Retention on a rip channeled beach. Journal of Geophysical Research 114, C10010.
Mikkelsen, O.A., Pejrup, M., 2000. The use of a LISST-100 laser particle sizer for Rennie, C.D., Millar, R.G., Church, M.A., 2002. Measurement of bed load velocity
in-situ estimates of floc size. Geo-Marine Letters 20, 187–195. using an acoustic Doppler current profiler. Journal of Hydraulic Engineering
Millard, N.W., 1976. Noise generated by moving sediments. Proceedings of the 128(5), 473–483.
Conference on Recent Developments in Underwater Acoustics 31 March–1 April Rice, M.A., Willetts, B.B., McEwan, I.K., 1995. An experimental study of multiple
1976, London, UK: Institute of Acoustics. grain-size ejecta produced by collisions of saltating grains with a flat bed.
Mizuyama, T., Laronne, J.B., Nonaka, M., et al., 2010. Calibration of a passive Sedimentology 42, 695–706.
acoustic bedload monitoring system in Japanese mountain rivers. In: Gray, J.R., Richards, K.S., Milne, L.M., 1979. Problems in the calibration of an acoustic device
Laronne, J.B., Marr, J.D.G. (Eds.), Bedload-surrogate Monitoring Technologies. for the observation of bedload transport. Earth Surface Processes and Landforms
US Geological Survey Scientific Investigations Report 2010–5091. US 4, 307–317.
Geological Survey, Reston, VA, pp. 296–318. Richardson, N.M., 1902. An experiment on the movements of a load of brickbats
Musick, H.B., Gillette, D.A., 1990. Field evaluation of relationships between a deposited on the Chesil Beach. Proceedings Dorset Natural History and
vegetation structural parameter and sheltering against wind erosion. Land Antiquarian Field Club 23, 123–133.
Degradation and Rehabilitation 2, 87–94. Rickenmann, D., 1997. Sediment transport in Swiss torrents. Water Resources
Musick, H.B., Trujillo, S.M., Truman, C.R., 1996. Windtunnel modeling of the Research 37, 3295–3305.
influence of vegetation structure on saltation threshold. Earth Surface Processes Rickenmann, D., McArdell, B.W., 2007. Continuous measurement of sediment
and Landforms 21(7), 589–606. transport in the Erienbach stream using piezoelectric bedload impact sensors.
Namikas, S.L., 2002. Field evaluation of two traps for high-resolution aeolian Earth Surface Professes and Landforms 32, 1362–1378.
transport measurements. Journal of Coastal Research 18, 136–148. Ridd, P.V., 1992. A sediment level sensor for erosion and siltation detection. Est.
Namikas, S.L., Bauer, B.O., Sherman, D.J., 2003. Influence of averaging interval on Coast. Shelf Science 35, 353–362.
shear velocity estimates for aeolian transport modeling. Geomorphology 53, Ridd, P.V., Day, G., Thomas, S., et al., 2001. Measurement of sediment deposition
235–246. rates using an optical backscatter sensor. Estuarine, Coastal and Shelf Science
Namikas, S.L., Sherman, D.J., 1995. A review of the effects of surface moisture 52, 155–163.
content on aeolian sand transport. In: Tchakerian, V.P. (Ed.), Desert Aeolian Rippeth, T.P., Fisher, N.R., Simpson, J.H., 2001. The cycle of turbulent dissipation
Processes. Chapman and Hall, London, pp. 269–293. of tidal straining. Journal of Physical Oceanography 31, 2458–2471.
 Electronic Measurement Techniques for Field Experiments in Process Geomorphology 219

Robinson, D.A., Gardner, C.M.K., Cooper, J.D., 1999. Measurement of relative Stevens, C., 2010. Short-term dispersion and turbulence in a complex-shaped
permittivity in sandy soils using TDR, capacitance and theta probes: estuarine embayment. Continental Shelf Research 30, 393–402.
comparison, including the effects of bulk soil electrical conductivity. Journal of Stockton, P.H., Gillette, D.A., 1990. Field measurement of the sheltering effect of
Hydrology 223, 198–211. vegetation on erodible land surfaces. Land Degradation and Rehabilitation 2,
Robinson, M., Dean, T.J., 1993. Measurement of near surface soil water content 77–85.
using a capacitance probe. Hydrological Processes 7, 77–86. Stone, M.C., Hotchkiss, R.H., 2007. Evaluating velocity measurement techniques in
Rouse, H.L., 1994. Measurement of bedload gravel transport; the calibration of a self- shallow streams. Journal of Hydraulic Research 45, 752–762.
generated noise system. Earth Surface Processes and Landforms 19, 789–800. Storlazzi, C.D., Ogston, A.S., Bothner, M.H., Field, M.E., Presto, M.K., 2004. Wave-
Ruessink, B.G., 2000. An empirical energetics-based formulation for the cross-shore and tidally-driven flow and sediment flux across a fringing coral reef: south-
suspended sediment transport by bound infragravity waves. Journal of Coastal central Molokai, Hawaii. Continental Shelf Research 24, 1397–1419.
Research 16, 482–493. Stout, J.E., Zobeck, T.M., 1997. Intermittent saltation. Sedimentology 44,
Ruggiero, P., Kaminsky, G.M., Gelfenbaum, G., Voigt, B., 2005. Seasonal to 959–970.
interannual morphodynamics along a high-energy dissipative littoral cell. Journal Sutherland, T.F., Lane, P.M., Amos, C.L., Downing, J., 2000. The calibration of
of Coastal Research 21, 553–578. optical backscatter sensors for suspended sediment of varying darkness levels.
Sambuelli, L., Silvia, B., Corrado, C., Stefano, S., 2010. A GPR survey on a Marine Geology 162, 587–597.
morainic lake northerly Turin (Italy). 13th International Conference on Ground- Taniguchi, S., Itakura, Y., Miyamoto, K., Kurihara, J., 1992. A new acoustic sensor
Penetrating Radar. Lecce, Italy.10.1109/ICGPR.2010.5550240. for sediment discharge measurement. Erosion and sediment transport monitoring
Sarre, R.D., 1987. Aeolian sand transport. Progress in Physical Geography 11, programmes in river basins. Proceedings of the Oslo Symposium, August 1992,
157–182. . 135–142.
Sauer, C.O., 1930. Basin and range forms in the Chiricahua Area. University of Tazioli, G.S., 1980. The measurement of suspended sediment transport in natural
California Publications in Geography 3, 339–414. streams using automatic radioisotope gauges. Journal of Hydrology 47,
Schat, J., 1997. Multifrequency acoustic measurement of concentration and grain 173–185.
size of suspended sand in water. Journal of the Acoustic Society of America Tazioli, G.S., 1981. Nuclear techniques for measuring sediment transport in natural
101, 209–217. streams. Examples from instrumented basis. Proceedings of the Florence
Schmidt, W.E., Guza, R.T., Slinn, D.N., 2005. Surf zone currents over irregular Symposium on Erosion and Sediment Transport Measurement. IAHS publication
bathymetry: drifter observations and numerical simulations. Journal of 133.
Geophysical Research 110, C12015. http://dx.doi.org/10.1029/2004JC002421. Terray, E.A., Drennan, W.M., Donelan, M.A., 1999. The vertical structure of shear
Schmidt, W.E., Woodward, B.T., Millikan, K.S., Guza, R.T., Raubenheimer, B., and dissipation in the ocean surface layer. In: Banner, M.L. (Ed.), The Wind-
Elgar, S., 2003. A GPS-tracked surf zone drifter. Oceanic Technology 20, driven Air-sea Interface: Electromagnetic and Acoustic Aensing, Wave Dynamics
1069–1075. and Turbulent Fluxes.. Pub: School of Mathematics, U New South Wales,
Schmutz, P.P., (2007). Investigation of utility of delta-theta probe for obtaining Sydney 2052, Australia, pp. 239–245.
surficial moisture measurements on beaches. MSc thesis, Department of Thonon, I., Roberti, J.R., Middlekoop, H., van der Perk, M., Burrough, P.A., 2005.
Geography and Anthropology, Louisiana State University and Agricultural In situ measurements of sediment settling characteristics in floodplains using a
Mechanical College, 114 pp. LISST-ST. Earth Surface Processes and Landforms 30, 1327–1343.
Schubauer, G.B., Adams, G.H., 1954. Lag of anemometers. Washington, DC, NBS Thorne, P.D., 1986. An intercomparison between visual and acoustic detection of
Report 3245. seabed gravel movement. Marine Geology 72, 11–31.
Sheremet, A., Mehta, A.J., Liu, B., Stone, G.W., 2005. Wave-sediment interaction on Thorne, P.D., Davies, A.G., Bell., P.S., 2009. Observations and analysis of sediment
a muddy inner shelf during Hurricane Claudette. Estuarine Coastal and Shelf diffusivity profiles over sandy rippled beds under waves. Journal of Geophysical
Science 63, 225–233. Research 114, C02023. http://dx.doi.org/10.1029/2008JC004944.
Sherman, D.J., Greenwood, B., 1984. Boundary roughness and bedforms in the surf Thorne, P.D., Heathershaw, A.D., Troiano, L., 1984. Acoustic detection of
zone. Marine Geology 60, 199–218. seabed gravel movement in turbulent tidal currents. Marine Geology 54,
Sherman, D.J., Li, B., Farrell, E.J., Ellis, J.T., Cox, W.D., Maia, L.P., Sousa, P.H., M43–M48.
2011. Measuring aeolian saltation: a comparison of sensors. Journal of Coastal Thorne, P.D., Salkfield, A.P., Marks, A.J., 1983. Application of acoustic techniques
Research SI59, 280–290. in sediment transport research, Inst. of Acoustics, Proceedings of Conference on
Sherman, D.J., Short, A.D., Takeda, I., 1993. Sediment mixing depth and Acoustics and the Sea-Bed Bath University, pp. 395–402.
megaripple migration in rip channels. Journal of Coastal Research SI15, Thorne, P.D., Williams, J., Heathershaaw, D., 1989. In situ acoustic measurements
39–48. of marine gravel threshold and transport. Sedimentology 36, 61–74.
Shih, H.H., Bushnell, M. Tronvig, K., Mero, T., 2003. Wave and current Thornton, E.B., Guza, R.T., 1983. Transformation of wave height dstribution. Journal
measurements in the Chesapeake Bay. Proceedings of IEEE/OES Seventh of Geophysical Research 88, 5925–5938.
Working Conference on Current Measurement Technology, San Diego, CA, IEEE/ Tonolla, D., Acuna, V., Lorang, M.S., Heutschi, K., Tockner, K., 2010. A field-based
OES, pp. 264–270. investigation to examine underwater soundscapes of five common river habitats.
Shugar, D.H., Kostaschuk, R., Best, J.L., Parsons, D.R., Lane, S.N., Orfeo, O., Hydrological Processes 24, 3146–3156.
Hardy, R.J., 2010. On the relationship between flow and suspended sediment Tonolla, D., Lorang, M.S., Heutsch, K., Tockner, K., 2009. A flume experiment to
transport over the crest of a sand dune, Rı́o Paraná, Argentina. Sedimentology examine underwater sound generation by flowing water. Aquatic Science 71,
57, 252–272. http://dx.doi.org/10.1111/j.1365–3091.2009.01110.x. 449–462.
Simpson, M.R., Oltman, R.N., 1993. Discharge measurement using an Acoustic Topp, G.C., Davis, J.L., Annan, A.P., 1980. Electromagnetic determination of soil
Doppler current profiler. USGS. Water-Supply Paper 2395, . 34. water content: measurements in coaxial transmission lines. Water Resources
Siddiqui, S.I., Drnevich, V.P., 1995. A new method of measuring density and Research 16, 574–582.
moisture content of soil using the technique of time domain reflectometry. Traykovski, P., 2007. Observations of wave orbital scale ripples and a
Report No.: FHWA/IN/JTRP-95/9, Joint Transportation Research Program, nonequilibrium time-dependent model. Journal of Geophysical Research, 112.
Indiana Department of Transportation, Purdue University, February, 271 p. Traykovski, P., Latter, R.J., Irish, J.D., 1999. A laboratory evaluation of the laser
Spaan, W.P., Van den Abele, G.D., 1991. Wind borne particle measurements with in situ scattering and transmissometery instrument using natural sediments.
acoustic sensors. Soil Technology 4, 51–63. Marine Geology 159, 355–367.
Spieker, R., Ergenzinger, P., 1990. New developments in measuring bed load by the Tunnicliffe, J., Gottesfeld, A.S., Mohamed, M., 2000. High resolution measurement
magnetic tracer technique. Erosion, transport, and deposition processes. of bedload transport. Hydrological Processes 14, 2631–2643.
Proceedings of the Jerusalem Workshop, March-April 1987. IAHS Publ 189, Turner, I.L., Russell, P.E., Butt, T., 2008. Measurements of wave-by-wave bed levels
169–178. in the swash zone. Coastal Engineering 55, 1237–1242.
Spydell, M., Feddersen, F., Guza, R.T., 2007. Observing surf-zone dispersion with Turowski, J.M., Rickenmann, D., 2009. Tools and cover effects in bedload transport
drifters. Journal of Physical Oceanography 37, 2920–2939. observations in the Pitzbach, Austria. Earth Surface Processes and Landforms
Sternberg, R.W., Shi, N.C., Downing, J.P., 1984. Field investigations of suspended 34, 26–37.
sediment transport in the nearshore zone. 19th International Conference on Van Pelt, R.S., Zobeck, T.M., Peters, P., Visser, S., (2006). Wind Tunnel Testing and
Coastal Engineering, American Society of Civil Engineers, Houston, TX, Comparison of Three Saltation Impact Sensors. Abstracts, International
1782–1798. Conference on Aeolian Research. July 24–28, 2006, Guelph, Ont., Canada.
220 Electronic Measurement Techniques for Field Experiments in Process Geomorphology

Vanhee, S., Anthony, E.J., Ruz, M., 2002. Aeolian sand transport on a ridge and Webb, M.P., Vincent, C.E., 1999. Comparison of time-averaged acoustic backscatter
runnel beach: preliminary results from Leffrinckoucke Beach, northern France. concentration profile measurements with existing predictive models. Marine
Journal of Coastal Research SI36, 732–740. Geology 162, 71–90.
Villard, P.V., Church, M., Kostaschuk, R.A., 2005. Estimating bedload in sand-bed Whitaker, A.C., Potts, D.F., 2007. Analysis of flow competence in an alluvial gravel
channels using bottom tracking from an acoustic Doppler Profiler. In: Blum, bed stream, Dupuyer Creek, Montana. Water Resources Research, 43 http://dx.d
M.D., Marriott, S.B., Leclair, S.L. (Eds.), Fluvial Sedimentology VII, International oi.org/10.1029/2006WR005289.
Association of Fluvial Sedimentologists Special Publication 35. Blackwell, Wiggs, G.F.S., Atherton, R.J., Baird, A.J., 2004. Thresholds of aeolian sand
Malden, MA, pp. 197–209. transport: establishing suitable values. Sedimentology 51(1), 95–108.
Vincent, C.E., Osborne, P.D., 1995. Predicting suspended sand concentration Wiggs, G.F.S., Livingstone, I., Thomas, D.S.G., Bullard, J.E., 1996. Airflow and
profiles on a macrotidal beach. Continental Shelf Research 15, 1497–1514. roughness characteristics over partially vegetated linear dunes in the southwest
Voulgaris, G., Workman, M., Collins, M.B., 1999. Measurement techniques of Kalahari, Desert. Earth Surface Process and Landforms 21(1), 19–34.
shingle transport in the Nearshore Zone. Journal of Coastal Research 15(4), Willetts, B.B., Rice, M.A., 1986. Collision in aeolian transport: the saltation/creep
1030–1039. link. In: Nickling, W.G. (Ed.), Aeolian Geomorphology. Allen and Unwin, London,
Waddell, E., 1973. Dynamics of Swash and Implication to Beach Response, pp. 1–18.
Technical report 139, Coastal Studies Institute, Louisiana State University, Baton Williams, J.J., Butterfield, G.R., Clark, D.G., 1990. Rates of aerodynamic
Rouge, La, p. 49. entrainment in developing boundary layer. Sedimentology 37, 1039–1048.
Waddel, E., 1976. Swash-groundwater-beach interactions. In: Davis, R.A., Wilczak, J.M., Oncley, S.P., Stage, S.A., 2001. Sonic anemometer tilt correction
Etherington, R.L. (Eds.), Beach and Nearshore Sedimentation. Society of algorithms. Boundary-layer Meteorology 99, 127–150.
Economic Paleontologists and Mineralogists Special Publication, 24, pp. Woods, J.D., 1968. Wave-induced shear instability in the summer thermocline.
115–125. Journal of Fluid Mechanisms 32, 791–800.
Waddell, E., 1976. Swash – groundwater – beach profile interactions. In: Davis, Wyatt, V.E., Nickling, W.G., 1997. Drag and shear stress partitioning in sparse desert
R.L., Ethington, R.L. (Eds.), Beach and Nearshore Sedimentation. Society of creosote communities. Canadian Journal of Earth Sciences 34, 1486–1498.
Economic Paleontologists and Mineralogists, Special Publication, Tulsa, Wysocki, L.E., Davidson, III J.W., Smith, M.E., et al., 2007. Effects of aquaculture
Oklahoma, vol. 24, pp. 115–125. production noise on hearing, growth, and disease resistance of rainbow trout
Waddell, E., 1980. Wave forcing of beachgroundwater, Proceedings of 17th Oncorhynchus mykiss. Aquaculture 272, 687–697.
International Conference on Coastal Engineering, ASCE, Sydney, Australia, Yang, Y., Davidson-Arnott, R., 2005. Rapid measurement of surface moisture
pp. 1436–1452. content on a beach. Journal of Coastal Research 21(3), 447–452.
Walker, I.J., 2005. Physical and logistical considerations of using ultrasonic Yanuka, M., Topp, G.C., Zegelin, S., Zebchuk, W.D., 1988. Multiple reflection and
anemometry in aeolian sediment transport research. Geomorphology 68(1–2), attenuation of time domain reflectometry pulses: theoretical considerations for
57–76. applications to soil and water. Water Resources Research 24, 939–944.
Walker, I.J., Davidson-Arnott, R.G.D., Hesp, P.A., Bauer, B.O., Ollerhead, J., Zedel, L., Hay, A.E., Cabrera, R., Lohrmann, A., 1996. Performance of a single
2009. Mean flow and turbulence responses in airflow over foredunes: beam, pulse-to-pulse coherent Doppler Profiler. IEEE Journal of Oceanic
new insights from recent research. Journal of Coastal Research SI56, Technology 21, 290–297.
366–370. Zou, X.-Y., Wang, Z.-L., Hao, Q.-Z., Zhang, C.-L., Liu, Y.-Z., Dong, G.-R., 2001.
Weaver, C.M., Wiggs, G.F.S., 2010. Field measurements of mean and turbulent The distribution of velocity and energy of saltating sand grains in a wind tunnel.
airflow over a barchan sand dune. Geomorphology 128, 32–41. Geomorphology 36, 155–165.

Biographical Sketch
Dr. Douglas J Sherman is the Daniel Bullock Harris Prof. of Geosciences at Texas A&M University. His academic
credentials include a PhD in Geography from the University of Toronto and a postdoc in the Ocean Engineering
Department at Woods Hole Oceanographic Institution. He is a Fellow of the AAAS and Past Chair of the IGU’s
Commission on Coastal Systems. Prof. Sherman’s research expertise is in nearshore processes and sediment
transport in coastal and aeolian systems. He has produced more than 100 scholarly publications, and his research
has been supported by numerous state, federal, and international programs.
Electronic Measurement Techniques for Field Experiments in Process Geomorphology 221

Dr. Chris Houser is a coastal geomorphologist in the Department of Geography at Texas A&M University. Since
graduating in 2004 as PhD student of Dr. Brian Greenwood at the University of Toronto, Chris has focused his
research on coastal morphodynamics with a focus on the response and recovery of barrier islands along the Gulf
of Mexico to extreme storms. This involves research in a range of spatial and temporal scales from swash zone
processes, to nearshore bar behavior and aeolian sediment transport from beach to dune.

Dr. Andreas Baas achieved his BSc in Physical Geography and Soil Science and MSc in Process Geomorphology at
the University of Amsterdam (UvA), including work at the Desert Research Institute in Reno, Nevada. He then
moved to the University of Southern California (USC) where he earned his PhD on investigating the formation
and behavior of aeolian streamers, supported by the National Science Foundation (NSF) and a Tyler Environ-
mental Fellowship. In 2003 he was appointed Lecturer and then Senior Lecturer at the Department of Geography
at King’s College London. Andreas is an editor for Nonlinear Processes in Geophysics, an editorial board member for
Geomorphology, and an active session convener for the Geomorphology Division of the European Geosciences
Union (EGU). Andreas’ research is supported by The Leverhulme Trust, The Nuffield Foundation, and the UK
Natural Environment Research Council, and spans all aspects of aeolian geomorphology across a range of spa-
tiotemporal scales, including the physics of aeolian sediment transport, in particular field experiments on the
interactions between airflow turbulence and sand transport, and dunes and dunefields, particularly modeling the
effects of vegetation on the development of dune landscapes in coastal environments.
 14.18 Laboratory Techniques for Geomorphologists: An Introduction
AD Switzer, Earth Observatory of Singapore, Nanyang Technological University, Singapore
r 2013 Elsevier Inc. All rights reserved.

14.18.1 Investigating the Size and Shape of Particles 222

14.18.2 Chemical Techniques for Geomorphological Investigations 222
14.18.3 Micropaleontology: Sometimes it’s the Little Things that Count 222
14.18.4 Dates and Rates: Dating Geomorphic Processes 223
References 223

Abstract

This chapter introduces the laboratory techniques covered in this section of the volume. The modern geomorphologist uses
a range of laboratory-based techniques from basic laboratory equipment like pH meters to mass spectrometry for
investigating the chemical proxies that often reflect environmental change.

A myriad of laboratory techniques are available to the modern 14.18.2 Chemical Techniques for Geomorphological
geomorphologist, from sieve shakers and cam sizers for par- Investigations
ticle analysis to inductively coupled plasma mass spec-
trometers for investigating chemical proxies of environmental The use of chemical techniques in geomorphology is ex-
change. The last few decades have been a period of incredible tremely broad and varied. Chemical treatments vary from
technical advances in the laboratory techniques used in geo- ‘bucket’ chemistry such as using large amount of hydrochloric
morphology; particular examples include the increased use of acid to remove carbonate from sediments, to the use of mass
chemical proxies as indicators of past environmental change spectrometry techniques on samples to investigate records of
(see Chapter 14.23) and the increased precision and appli- past environmental change. In the first paper from this section
cation of a variety of new or radically refined dating techni- Kennedy and Woods look at the different ways geomorph-
ques (see Chapters 14.30 and 14.31). In this section of the ologists determine organic and carbonate content in
volume the authors have compiled papers on a diverse range sediments (see Chapter 14.22). They provide a detailed com-
of laboratory techniques spanning physical measurements of mentary on the different techniques applied in one of the
soils and sediment properties (see Chapters 14.19, 14.20, and most common treatments in geomorphology. From the
14.22), basic chemical techniques (see Chapter 14.22), common to the diverse we then move to a complementary
chemical proxies (see Chapters 14.23 and 14.24), micro- paper by Chague-Goff and Wong (see Chapter 14.23) who
palaeontology (see Chapter 14.25), pollen (see Chapter look at a series of common analytical techniques including
14.26), and dating (see Chapters 14.30 and 14.31). colorimetry and atomic absorption spectrophotometry. Fol-
lowing this paper is a synthesis paper by Birch based on the
application of heavy metals in sediments (see Chapter 14.24).

14.18.1 Investigating the Size and Shape of Particles

The measurement of particle size and shape is fundamental to

14.18.3 Micropaleontology: Sometimes it’s the
many disciplines of geomorphology as well as the allied dis- Little Things that Count
ciplines of Environmental Engineering and Earth Sciences. In
Although the advancement in chemical techniques has been
the first two papers in this section Switzer examines particles
sizing techniques and data presentation (see Chapter 14.19) particularly noteworthy, the traditional techniques of micro-
before Bullard examines particle shape in a geomorphic con- paleontology have also seen significant advances in the past
text (see Chapter 14.20). These two syntheses lead well to the decade. In this section two micropaleontological techniques
are reviewed and recent advances are highlighted. The first
paper of Bull and Morgan, who provide an extensive summary
of the scanning electron microscope before providing a de- paper by Kemp and colleagues (see Chapter 14.25) examines
tailed case study based on their extensive work in the relatively the use of microfossils and associated transfer functions to
reconstruct sea-level change, paleoearthquakes, tsunamis, and
new field of geoforensics (see Chapter 14.21).
hurricane recurrence intervals. The second paper by Moss (see
Chapter 14.26) looks at the role of palynology (pollen stud-
Switzer, A.D., 2013. Laboratory techniques for geomorphologists: an
introduction. In: Shroder, J. (Editor in Chief), Switzer, A.D., Kennedy, D.M.
ies) in geomorphology. Pollen studies have a long and com-
(Eds.), Treatise on Geomorphology. Academic Press, San Diego, CA, vol. 14, prehensive history in the various fields of geomorphology and
Methods in Geomorphology, pp. 222–223. Moss uses primarily Australasian examples to provide a

222 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00384-5

 Laboratory Techniques for Geomorphologists: An Introduction 223

synthesis of some of the common techniques, applications, radiocarbon methodologies. Kemp et al. (see Chapter 14.31)
and treatments of pollen data. examine the process involved in developing a reliable
chronological framework in organic deposits. Using examples
from the west coast of North America they describe the process
14.18.4 Dates and Rates: Dating Geomorphic of obtaining accurate chronology using the techniques pion-
Processes eered by Alan Nelson as an example (Nelson, 1992; Nelson
et al., 1995).
Although the above-mentioned techniques provide many The papers in this section of the volume were solicited to
proxies for the analysis of landform evolution, such infor- reflect the depth and diversity of the field and this section of
mation is generally of little use if it is not put into a chrono- the volume is by no means exhaustive. There are numerous
logical framework. To this end the geomorphologists reviews of laboratory techniques in the key texts of soil sci-
commonly rely on a variety of dating techniques to provide ence, sedimentology, archaeology, and paleontology that
such a framework. To review and analyze the development, provides a wealth of knowledge for a geomorphologist who is
application, and myriad of dating techniques available to the about to head into the lab.
modern geomorphologist is well beyond the scope of this
volume. After careful consideration the editors solicited a team
to provide an introduction to dating techniques that would be
helpful to the undergraduate or postgraduate geomorph-
ologist. Sloss and coauthors summarize three main techni- References
ques, radiocarbon, amino acid racemization (AAR), and
optically stimulated luminescence (OSL) (see Chapter 14.30). Nelson, A.R., 1992. Discordant 14C ages from buried tidal-marsh soils in the
They review each technique and examine the pros and cons of Cascadia subduction zone, southern Oregon coast. Quaternary Research 38,
74–90.
each within the context in numerous case studies. Following Nelson, A.R., Atwater, B.F., Bobrowsky, P.T., et al., 1995. Radiocarbon evidence for
this review the focus moves on to a more in-depth study of the extensive plate-boundary rupture about 300 years ago at the Cascadia
processes of obtaining good dates from organic deposits using subduction zone. Nature 378, 371–374.

Biographical Sketch

Adam Switzer is a sedimentologist and geomorphologist, who holds the positions of Singapore NRF Fellow,
Nanyang Assistant Professor, and Principal Investigator at The Earth Observatory of Singapore and Division of
Earth Sciences in SPMS at Nanyang Technological University.
He and his research group are investigating the geological record of coastal hazards (storms and tsunamis) in
Southeast Asia. One of the driving aims of Dr. Switzer’s research group is to make the coastlines of Asia safer
places to live, work, and for holiday. Currently, Asia is the focus of his research program and Singapore has the
resources to enable him to effectively study the sedimentary and geomorphic effects of tsunamis and storm surges.
Dr Switzer’s 10 strong research groups currently works in nine different countries throughout the world. Dr.
Switzer obtained a BSc and PhD in Geosciences from the University of Wollongong, Australia, before completing
his postdoctoral studies at the University of Hong Kong.
 14.19 Measuring and Analyzing Particle Size in a Geomorphic Context
AD Switzer, Earth Observatory of Singapore, Nanyang Technological University, Singapore
r 2013 Elsevier Inc. All rights reserved.

14.19.1 Introduction 225

14.19.1.1 Measuring the Size of a Particle 226
14.19.1.2 Particle Size, Grain Size, and Particle Sizing 226
14.19.1.3 The Imperfect Sphere: A Universal Problem in Geomorphology 226
14.19.1.4 Overcoming the Irregularity of Particle Shape 226
14.19.1.5 Choosing the Right Particle-Sizing Technique or Instrument 226
14.19.2 Sample Preparations: A General Note on Labeling and the Selection of Materials for Particle-Size
Analysis 227
14.19.3 Grain (Particle) Size Scales: The Udden–Wentworth Scale 227
14.19.4 Analytical Techniques 228
14.19.4.1 Grain-Size Analysis of Gravel, Cobble, and Boulder Material 228
14.19.4.1.1 Two techniques for gravel and cobble analysis 230
14.19.4.1.1.1 Cobble cam: An example of new gravel-size analysis technologies 230
14.19.4.2 Analytical Techniques for Materials Composed Primarily of Sand, Silt, and Clay 231
14.19.4.2.1 Sieving 231
14.19.4.2.1.1 Dry sieving 231
14.19.4.2.1.1.1 Potential errors in dry sieving 231
14.19.4.2.1.2 Wet sieving 232
14.19.4.2.2 Sedimentation methods for sand, silt, and clay 232
14.19.4.2.2.1 The pipette method 232
14.19.4.2.2.2 The sedimentation tube 233
14.19.4.2.3 Laser diffraction analysis (LDA) 233
14.19.4.2.3.1 General process for LDA analysis 234
14.19.4.2.4 Other PSA techniques 234
14.19.5 Interpretation of Particle-Size Data 235
14.19.5.1 Use of Bivariate Plots (Scattergraphs) 235
14.19.5.2 Scattergraphs and Ternary Diagrams 236
14.19.5.3 Recent Advances in Data Presentation 237
14.19.5.4 Using Modern PSD Datasets 237
14.19.6 The Same but Different: A Concluding Note on Comparing Different Techniques 238
References 239

Glossary intensity of light scattered by a particle is directly

ASTM ASTM International, (formerly the American proportional to the size of the particle. Commonly, the laser
Society for Testing and Materials (ASTM)) is an is passed through the substance being examined, and the
international standards organization that develops and diffracted light focused onto a detector which measures the
publishes voluntary consensus technical standards for a angular distribution of the intensity of the scattered light.
wide range of materials, products, systems, and services. PSA In geomorphology, particle-size analysis (PSA) can be
ISO The International Organization for Standardization considered as any measurement of the size distribution of
(ISO), is an international standard-setting body composed individual particles in a soil or sediment sample.
of representatives from various national standards PSD The particle-size distribution (PSD) (also known as
organizations. grain-size distribution) of a granular material (e.g.,
LDA Laser diffraction analysis (LDA) is a technique which sediment) is a list of values or a mathematical function that
utilizes properties of the diffraction patterns of a laser beam defines the relative amounts of particles present, sorted
passed through a substance, to measure the size of particles. according to size.
Laser diffraction analysis is based on the principle that the SEM Scanning electron microscopy.

Switzer, A.D., 2013. Measuring and analyzing particle size in a geomorphic

context. In: Shroder, J. (Editor in Chief), Switzer, A.D., Kennedy, D.M.
(Eds.), Treatise on Geomorphology. Academic Press, San Diego, CA, vol. 14,
Methods in Geomorphology, pp. 224–242.

224 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00385-7

 Measuring and Analyzing Particle Size in a Geomorphic Context 225

Abstract

Measuring the particle-size distribution of sediments is a fundamental tool in geomorphology. Boulder- and cobble-size
material is generally measured by direct measurement whereas a wide variety of techniques are available to determine grain-
size distributions for sand- to clay-dominated sediments. Commonly, a combination of methods is needed and no par-
ticular technique can be considered ’more accurate’ or ’more precise’ than another. This chapter reviews the main tech-
niques used to obtain the particle-size distribution of sediments samples and outlines some of the uses and limitations of
each technique along with a commentary on popular methods to present grain-size data.

14.19.1 Introduction space (e.g., bedforms and landforms) and time, thereby
exerting control over physical transport and sedimentation
The grain or particle-size distribution (PSD) of sediments re- processes (Warrick et al., 2009). The variety of grain sizes
flects relative energy and is one of the fundamental environ- (Figure 1) encountered by geologists, geomorphologists, and
mental factors that control the erosion on the surface, archeologists means that no one technique can be applied to
transport in the atmosphere, and water bodies and deposition all environments and their associated landform components.
of sediment particles, on the Earth and other planetary bodies. It is generally desirable to have a convenient, repeatable,
Particle size is also relevant to many aspects of engineering, precise, accurate, and representative means by which to assess
petroleum, and agriculture industries (e.g., controls on aquifer sediment grain sizes when reconstructing palaeoenvironments
porosity and permeability in petroleum and groundwater or investigating processes (see discussions of Syvitski et al.,
studies). In most landform systems, the PSD of constituent 1991a, b; Cheetham et al., 2008; Donato et al., 2009). Of
sediments reflects morphological characteristics associated course, the variety of landforms and sediments occurring in
with the physical processes of landform development (e.g., depositional systems means that this desire must be traded
Klingeman and Emmett, 1982; McCave and Syvitski, 1991; against logistics of collection/analysis, sample sizes, repre-
Mason and Coates, 2001; Finkl, 2004; Buscombe and Masse- sentativeness of samples collected versus results obtained, and
link, 2006; Cheetham et al., 2008; Warrick et al., 2009). At any in many cases the economics of analysis. A myriad of PSA
site of sediment mobility the PSD varies considerably through techniques are available to the geomorphologist, particularly

0 10 20 30 40 50 60 70 80 90 100 110 120 130

Small pebbles

Medium pebbles

Large pebbles

Very large pebbles

Small cobbles
Granules

>256 mm: Boulders 128−256 mm: Large cobbles

Sand fraction

39−63 μm 125 μm 250 μm 500 μm 1000 μm 2000 μm

Silt Very fine Fine sand Medium sand Coarse sand Very coarse sand
sand

Figure 1 A grain-size comparator chart (to scale). The chart shows the different size fractions from silt (63 mm) through to large cobbles
(128–256 mm). Such charts are useful for field comparisons.
226 Measuring and Analyzing Particle Size in a Geomorphic Context

for those working with sand-sized material. A number of within particular disciplines, rather than a result of applied
papers exist that compare different techniques, and in many research or a technical comparison of applicability to par-
cases, comment on the limitations or inaccuracies of one ticular situations.
technique with respect to another (e.g., Agrawal et al., 1991;
McCave and Syvitski, 1991; Syvitski et al., 1991b; Konert and
Vandenberghe, 1997; McCave et al., 2006; Cheetham et al., 14.19.1.3 The Imperfect Sphere: A Universal Problem in
2008). Geomorphology
Applying a single method or technique to particle size that is
overlapping and comparable is implausible as the universal
14.19.1.1 Measuring the Size of a Particle technique would need to be equally applicable to regular and
Providing a definition for a ‘particle’ is not an easy task. Ac- irregular shapes. The simplest particle shape with respect to
cording to the Dictionary of Geological Terms, second editon three-dimensional geometry is a sphere, and visual inspection
(1984: 370), a particle is ‘‘a general term, used without re- (microscopy or image analysis) is the most straightforward
striction as to shape, composition, or internal structure, for a measurement technique. When examining a sphere, its per-
separable or distinct unit of rock e.g. a fragment or grain.’’ This imeter, projected cross-sectional area, surface area, and vol-
differs slightly from ‘clasts,’ which are defined as ‘‘an indi- ume can be described unambiguously by one linear
vidual constituent grain, or fragment of a detrital sediment or dimension (the diameter of the projected cross section) pro-
sedimentary rock, produced by the physical disintegration of a vided you have a cross section through the mid-point of the
larger rock mass’’ by the same source (p. 90). In geomorph- sphere. Furthermore, a sphere is the only shape that is geo-
ology, the particles measured are generally composed of in- metrically isotropic, i.e., the projected cross-sectional diameter
organic mineral grains or rock fragments, although in some remains constant regardless of the angle of view. The problem
cases they may be organic. Some particles will exhibit rela- in the natural world is that almost all natural particles project
tively molecular homogeneity (e.g., quartz grains) whereas a different cross section at all angles of view. As a result, for
others such as rock fragments will be inhomogeneous in any nonspherical particle neither surface area nor volume can
composition. Invariably, particles in most natural sediments be inferred from the projected cross section.
or soils will also be of varied density and shape (see Chapter
14.20). This is problematic, particularly when analyzing platy
sediments, carbonates, evaporates, and sediments of highly 14.19.1.4 Overcoming the Irregularity of Particle Shape
variable density. One example is the analysis of carbonate The fact that an irregular particle can present a different cross
grains (see reviews in Folk and Robles, 1964; Kench and section depending on orientation is only one of the many
McLean, 1997; Kench, 1997) where the vastly different dens- measurement problems faced when determining particle size.
ities and shapes occurring in biogenic sediments mean that A second problem is that the cross section of an irregularly
measurement devices and techniques will respond differently shaped particle has different ‘diameters’ depending on where
to the same particle. It is therefore very important that any the chord is drawn (Webb, 2008). To overcome this limitation
researcher compares the results from two different types of and to simplify statistics many techniques assume sphericity
sizing instruments. This chapter is not an attempt at an ex- and as such produce an equivalent sphere value as all particle-
haustive review of the problems and issues of particle-size sizing techniques aim to provide a single number that is in-
analysis (PSA). Throughout the chapter problems that arise dicative of particle size. Most sizing techniques therefore as-
with particular techniques or types of analysis will be con- sume that the material being measured is spherical and report
sidered. In such cases some topical references will be cited. In the particle size as the diameter of the ‘equivalent sphere.’ In
order to contain the word length of this paper this work will most cases the equivalent sphere approximation for an ir-
primarily deal with an assumption that the basic analysis regularly shaped particle is dependent on the physical prop-
needs of the reader pertain to near-spherical grains of homo- erty measured by the chosen technique (Figure 2). For
genous composition, for example, quartz sands. Grain shape is example, laser diffraction techniques provide a diameter of the
also covered in detail elsewhere in this volume (see Chapter sphere that yields an equivalent light scattering pattern to the
14.20). particle being measured.

14.19.1.2 Particle Size, Grain Size, and Particle Sizing 14.19.1.5 Choosing the Right Particle-Sizing Technique or
Instrument
A comprehensive text edited by Syvitski (1991a, b) shows that
although there are many methods for grain size, none have The particle-sizing instrument and technique selected must be
been adopted as a comprehensive standard in sedimentology suitable for the material to be measured and must also provide
and geomorphology. One could argue that the use of proto- data to meet the specific needs of the user. In some cases,
cols of the International Standards Organization (ISO) or this might mean fast, repeatable analyses, or it may mean
American Society for Testing and Materials (ASTM) are com- high-resolution and very accurate or precise results. In geo-
prehensive standards (e.g., ISO 11277, 2009; ISO 9276-6, morphology and sedimentology, the determination of the
2008 and ISO 13320, 1999a, b; ASTM, 2010) however, it is particle-size distribution alone is seldom the ultimate object-
apparent that these standards are primarily the result of a ive; determining how a particle was transported or investi-
desire for consistent and comparable reporting procedures gating the process of deposition or landform development is
 Measuring and Analyzing Particle Size in a Geomorphic Context 227

Sphere of same
minimum length Sphere of same
weight
Sphere of same dmin
dmax dw
maximum length

Sphere of same
dv volume
Sphere having same dsed
sedimentation rate

Sphere of same
ds
Sphere passing same dsieve surface area
sieve aperture

Figure 2 The primary goal of particle-sizing techniques is to provide a number indicative of the particle size. However, particles are three-
dimensional objects for which at least three parameters (length, breadth, and height) are required for a complete description. Most sizing
techniques therefore assume that the material being measured is spherical and report the particle size as the diameter of the ‘equivalent sphere’
which would give the same response as the particle being measured. Courtesy of Alan Rawle, Malvern Instruments Ltd.

generally the reason for the measurement. In this regard, the undertaken. For example, cobbles and pebbles may require
characteristic actually being measured and related to size is cleaning before examination. In most cases, washing in water
commonly more important than the actual particle size. is sufficient to remove mud coatings (assuming data from the
However, particle size still remains one of the key components muds is unwanted). Conversely, sands derived from oceanic,
in landscape reconstructions and process studies (Cheetham hypersaline or brackish conditions may contain quantities of
et al., 2008). For example, sediment samples taken from river salt within their pore spaces that can become deposited during
paleochannels are commonly analyzed for PSD. Here, the drying, cementing adjacent particles together, and thus pro-
primary interest is to characterize the deposition mechanism ducing misleading analyses (see Loring and Rantala, 1992).
and palaeoenvironments or to link process and landform Where fine (mud) material is present, some techniques (e.g.,
together through physical properties. For example, Chen et al. sieve/pipette) require a user to separate them by wet sieving so
(1996) used grain size as a proxy for sedimentation velocity to that the sands and mud fractions can be examined separately
reconstruct the energy regime of paleochannels in northern using different techniques (see note in Box 1). In such cases it is
China. All different particle-sizing techniques are likely to important that fine separation be carried out before drying of
produce different size results for the same particle, and all of the sediment because silts and clays can produce crusts or
them are likely to be technically ‘correct.’ The best instrument durable pellets on heating (McManus, 1988). Although these
for the application (the best size definition) may be the one crusts or pellets may break down to individual particles during
that most closely relates the ‘particle size’ to the application of analysis, there is no guarantee that individual flakes of clay
the user. would become separated by purely physical means. Combining
these techniques may require dividing the sample into two
fractions to permit later recombination of the different ana-
lytical results.
14.19.2 Sample Preparations: A General Note on
Labeling and the Selection of Materials for
Particle-Size Analysis 14.19.3 Grain (Particle) Size Scales: The
Udden–Wentworth Scale
In most geomorphology, sedimentology, or soil laboratories,
numerous samples enter the laboratory at any one time. A first The Udden–Wentworth scale (Table 1) has a relatively simple
step is to systematically check the material to ensure legible geometric progression of sizes that are based on a logarithmic
identification numbers and labels are in place by laying out scale (phi (f) scale) developed by Krumbein (1934). The f
cores, sample bags, bottles, or blocks of material sequentially. scale uses the (negative) logarithm to the base 2 of the grain
It is important to note external sample identification and diameter (D, in millimeters) instead of simply the diameter
numbering, unless engraved on indestructible material, may itself, expressed as
become unreadable (McManus, 1988). As a result relabeling
with external or additional labels is almost always needed f ¼  log2 ðDÞ, ðmmÞ
(this is particularly the case for sandy samples). During this
checking process any gaps in a sampling sequence may also be This means that each size grade limit is twice as large as the
identified and samples lacking clear labels can be returned to next smaller grade limit. For example, 32 mm is 25, 64 mm is
their correct sequential position. If misplaced samples cannot 26, and so forth.
be identified with certainty, they have little value for sub- Using the phi scale also simplified the statistical compu-
sequent analysis and should be discarded. tations popularized in the 1950s and 1960s (e.g., Folk, 1954,
In terms of PSD determination, further sample preparation 1955; Folk et al., 1970). Folk (1954) developed and later re-
depends on the form of the size determination to be fined a sediment textural classification scheme to describe
228 Measuring and Analyzing Particle Size in a Geomorphic Context

Box 1 A note of caution on salts and clays in saline environments

Fine-grained (silt and clay) samples collected from saline environments (e.g., estuaries, hypersaline lakes) can be problematic
as the behavior of the finest size fractions is partially determined by the relative concentration, and composition of salts in
the interstitial waters. The hydrochemistry partially determines if clays remain as discrete particles or cluster together
(forming flocs or aggregates) with diameters much larger than the individual component particles. In such cases, washing the
sample changes the salt concentrations possibly inducing separation or aggregation of the particles (that in the natural
system would have been transported in a floc or aggregate). Hence the size analysis may not be a true indicator of the
relationship between the particles and their dynamics. Although this issue is widely recognized, most researchers remove salts
before subjecting samples to water based treatments (e.g., pipette, hydrometer, or laser PSA). Removal of water salts from fine
sediments is best achieved using dialysis, although, Buller and McManus (1979) suggested that washing sands three times
with 1 L of deionizer or distilled water will clean a 200 g sediment sample of salts. The issue of removing salts is also
discussed succinctly in the landmark paper on treating marine sediments by Loring and Rantala (1992). This issue can also
be partially overcome by adding a dispersing agent to assist the separation of individual particles. However, a caveat exists as
some particles are transported and deposited as flocs. For example, on the Cooper floodplain in Central Australia sediments
in river channels and dunes are characterized by clay-rich mud aggregates (or pellets) that exist in the river channels (Nanson
et al., 1986, 1988; Rust and Nanson, 1989) and in lunettes on the lee side of playa lakes (e.g., Butler, 1956; Bowler, 1973)).
In such a case the treatment and process of disaggregating the sediments would not be reflective of the process of bedload
and aeolian transport of the aggregates. Those using settling characteristics (e.g., settling tube) for clays also face an additional
hurdle, as clay particles will commonly combine naturally to produce flocs or aggregates giving a misleading guide to the
dynamics of the depositional environment as flocs settle at greater rates than their individual component particles. It is
worthwhile to note that in almost all situations size analysis of fine particles comes with inherent assumptions that present
problems, as the pre-analytical preparations can largely determine the results obtained.

mixtures of mud, sand, and gravel that define the basic de- Hesse (2003). The majority of geomorphologists place the silt/
scriptive names for both unconsolidated and lithified sedi- sand boundary at 63 mm B4 Phi (see Table 1).
ments. A modification of the phi scale in the 1960s generally
agreed that by adding a nominal value of Do ¼ 1 the phi sys-
tem becomes dimensionless allowing it to be correctly used 14.19.4 Analytical Techniques
for statistical analysis to derive factors such as the inclusive
standard deviation, skewness, and kurtosis of grain-size dis- The following section reviews a number of common techni-
tributions (McManus, 1963; Krumbein, 1964) ques for the analysis of materials of different grain sizes. In
some cases a step-by-step process is described, for others ref-
d ¼  log 2  D=Do
erences are supplied to key studies where step-by-step ex-
where Do is the diameter of a 1 mm grain. planations can be found.
The Udden–Wentworth scale was further modified in the
late 1990s by Blair and McPherson (1999) who upon recog-
14.19.4.1 Grain-Size Analysis of Gravel, Cobble, and
nizing that boulder sizes in natural environments can range
Boulder Material
well beyond the 4.1 m maximum of the Udden–Wentworth
scale, proposed an expanded size scale to include mega-gravels The task of obtaining accurate grain-size data from coarse
up to an astonishing 1075 km in diameter (see Chapter 14.13). sediments 464 mm can be time-consuming and challenging.
Although the original and modified forms of For example, Kellerhals and Bray (1971) and Adams (1979)
Udden–Wentworth scale remain the most popular scale in recommended that a sample mass for statistically accept-
geomorphology, it is becoming apparent that the advent of able sieve analysis of cobble size material would range from
laser-based technologies, primarily developed from the ma- tens to hundreds of kilograms depending on grain size. Ob-
terials engineering and pharmaceutical industries, is reflected viously, this is far more material than most researchers would
in a move to the practice of using microns (mm) (e.g., Pye and like to extract from the landscape. This appears excessive al-
Blott, 2004; Cheetham et al., 2008). though it makes sense as sampling for PSA in some environ-
Despite the general consensus on the Udden–Wentworth ments (gravel-bed rivers, gravel beaches) is generally aiming to
scale, a considerable and long-running inconsistency remains. characterize an extremely large population of sedimentary
This relates to the finer fractions where despite agreement on particles for which a complete census is impractical. Sampling
the form of the scale and grade boundaries in the coarser of coarse sediments must be random, comprise enough grains
ranges, different authors still place the silt-clay boundary for an adequate sample size, and be drawn from a relatively
variously at 2 mm (Briggs, 1977; Friedman and Sanders, 1978), homogenous environment that is representative of the land-
which is a size commonly used by soil scientists, or at 4 mm, as form in question. For environments dominated by sand-size
in the original Udden–Wentworth system (Tanner, 1969; sediments and finer, it is relatively easy to obtain a large
Pettijohn, 1975), or at 3.9 mm. The same can be said for silt/ enough sample that can be analyzed in the lab. In contrast,
sand boundary as summarized by a thought provoking, al- PSA analysis in environments dominated by gravel-, cobble-,
though not terribly conclusive, discussion on this issue by and boulder-size material commonly needs requisite sample
 Measuring and Analyzing Particle Size in a Geomorphic Context 229

Table 1 Comprehensive correlation table showing the relationships between phi sizes, millimeter diameters, size classifications (Wentworth,
1922), and ASTM and Tyler sieve sizes. The table also shows the corresponding intermediate diameters, grains per milligram, settling velocities,
and threshold velocities for traction

PHI − mm

decimal inches
Fractional mm
coversion Size terms Sieve Settling Threshold
Number

equivalent to sieve size

Intermediate diameters
 = log2 (d in mm) (after sizes velocity velocity
of grains

and
1 μm = 0.001 mm

of natural grains
wentworth, 1922) (quartz, for traction
per mg
20 °C) cm s−1

(U.S. standard)
 mm

(Gibbs, 1971)
ASTM No.

Hjuistrom, 1939)
mesh no.

(modified from
(Novin, 1946)
−8

Crushed
Tyler
256 10.1″

Sphere
spheres

Natural
Quartz

sand
200 Boulders
(≥−8)
−7 128 5.04″
cm s−1
100 Cobbles
200 1m
−6 64.0 2.52″ 2 1/2″ above
53.9 2.12″ 2″ bottom
50 Very
45.3 150
40 33.1 coarse 1 1/2″ 1 1/2″
−5 30 32.0 1.26″ 1 1/4″
26.9 1.06″ 1.05″
22.6 Coarse 50
20 17.0 3/4″ .742″ 100
−4 16.0 0.63″ 5/8″ 90 40 100
Pebbles

13.4 1/2″ .525″

11.3 Medium 7/16″ 80 90
10 9.52 3/8″ .371″ 30
70 80
−3 8.00 0.32″ 5/16″
6.73 .265″ 3 60
5.66 Fine 70
5 4.76 4 4 50
−2 4 4.00 0.16″ 5 5 20 60 100
3.36 Very 6 6 40
3 2.83 fine 7 7
2.38 granules 8 8 30 50
−1 2 2.00 0.08″ 10 9
1.63 inches Very 12 10
1.41 14 12 20 50
mm coarse 16 40
1.19 14 10
0 1 1.00 1 18 16 1.2 .72 .6 9 40
.840 20 20 8
.707 Coarse 25 24 .86 2.0 1.5 10 7
.545 30 28 6 30
1 .5 1/2 35 32 .59 5.6 4.5 8
.500 7
.420 40 35 5 30
6
Sand

.4 Medium
.354 45 42 .42 15 13 5 4
.3 .297 50 48 4
2 .250 1/4 60 60 .30 43 35 3 3
.2 .210 70 65 20 26
.177 Fine 80 80 .215 120 91 2 2
.149 100 100 Minimum
3 .125 1/8 120 115 .155 350 240 1 (Inman, 1949)
.1 .105 Very 140 150 1.0
.088 170 170 .115 1000 580
.074 fine 200 200 0.5 0.5
4 .062 1/16 230 250 .080 2900 1700 0.329
.05 .053 270 270
.044 Coarse 325 325
.04 400
Note: The relation between the beginning

.037 0.1
depends on the height above the bottom
that the velocity is measured, and on

5 .03 .031 1/32 0.085

of traction transport and the velocity

Medium
.02
Note: Some sieve openings differ

Note: Sieve openings differ by as

much as 2% from phi mm scale

Note: Applies to subangular to

Silt

Stokes law (R = 6rv)

0.023
Note: Applies to subangular to

6 .016 1/64
slightly from phi mm scale

subrounded quartz sand

subrounded quartz sand

other factors

Fine 0.01
.01
7 .008 1/128 0.0057
(in mm)

Very
.005 fine
8 .004 .004 1/256 0.0014
Clay/silt 0.001
.003 boundary
for mineral
analysis
Clay

9 .002 .002 1/512 0.00036

0.0001
10 .001 .001 1/1024
Source : Reproduced from Poppe, L.J., Eliason, A.H., Fredericks, J.J., Rendigs, R.R., Blackwood, D., Polloni, C.F., 2000. Grain-size
analysis of marine sediments: methodology and data processing. In: Poppe, L.J., Polloni, C.F. (Eds.), USGS East-coast Sediment
Analysis: Procedures, Database, and Georeferenced Displays. U.S. Geological Survey, Open-File Report 00-358, CD-ROM.
230 Measuring and Analyzing Particle Size in a Geomorphic Context

L
I
S

L = Longest axis
I = Intermediate axis
(a) S = Shortest axis

(b)

Figure 3 (a) The measuring approach to pebbles for a Wolman pebble count. (b) Students conduct Wolman pebble counts in the Ashop River,
Derby County, England. Photograph courtesy of Farah Alamgir, Loughborough University.

sizes that are too large or impractical to be transported off-site beaches (Jennings and Shulmeister, 2002; Miller et al., 2011)).
for measurement or logistically impractical to sieve in Generally a grain-size distribution is developed as the cumu-
the field. lative frequency of numbers of clasts of different size classes. If
The analysis of particle sizes of coarse unconsolidated the sampled clasts are of the same density, which will generally
sediments may be achieved through direct measurement only be true if sampling clasts derived from one homogenous
of individual pebbles, by pebble counts and more recently lithology, the results obtained will be comparable to a distri-
by camera-based analysis (Warrick et al., 2007, 2009). bution by weight (Kondolf, 1997; Kondolf et al., 2003). If the
Traditionally the lengths of representative diameters or axes target material (landform) is composed of particle sizes of
are determined with the aid of vernier calipers (e.g., Briggs, heterogeneous-composition material, or recognizable facies,
1977) by placing the particle (e.g., pebble) on a flat surface the analysis may be improved by collecting populations of
and measuring the length of the intermediate axis, I. In this particles from each facies before calculating a weighted average
text we will use L, I, and S for large, intermediate, and of grain-size distribution with estimated proportions of the
short axes, respectively (many studies use A, B, and C (e.g., sampled landform occupied by each facies. A detailed dis-
Switzer and Burston, 2010)). The length of the largest axis, L, cussion of this issue is in Kondolf et al. (2003).
is always at right angles to I. A further rotation of the
particle by 901 about that axis reveals the shortest axis, S.
Three mutually perpendicular axes may be used to characterize
the particle (Figure 3(a)). Averaging the three lengths, I, L, 14.19.4.1.1.1 Cobble cam: An example of new gravel-size
and S, yields a mean diameter (DM) for the particle. Alter- analysis technologies
natively, a mean diameter can be derived by immersing Cobble cam is relatively a new autocorrelation technique de-
the pebble in water to determine the volume of water dis- veloped by Rubin (2004) to measure grain-size of fluvial and
placed, from which the volumetric diameter is obtained coastal gravel bars using digital photographs (Warrick et al.,
(McManus, 1988). 2009). This technique uses digital photos that are obtained
from B1 m above the ground surface of a landform con-
14.19.4.1.1 Two techniques for gravel and cobble taining granule- to cobble-sized sediment (e.g., a gravel bar).
analysis The cobble cam is calibrated with physical measurements of
To negate the problem of transporting large amounts of gravel the intermediate and long dimensions of clasts in the field that
material to the laboratory, geomorphologists have developed are compared to the short and long axes, respectively, meas-
various field-sampling techniques and protocols. The most ured from the digital photographic images. In most cases
enduring of such protocols is the Wolman Pebble Count calibration curves for the autocorrelation technique are gen-
(Figure 3(b)), primarily used in gravel-bed rivers (Wolman, erated from a series of the ‘best-sorted’ (well-sorted) samples
1954). This technique involves randomly collecting and in the digital photographs (Warrick et al., 2007, 2009). Further
measuring at least 100 particles from a homogeneous area of a analysis of gravel and boulder materials in a geomorphic
streambed (similar analysis can be applied to mixed or gravel context is provided in the paper of Felton (see Chapter 14.13).
 Measuring and Analyzing Particle Size in a Geomorphic Context 231

Figure 4 Brass sieves of different sizes and a RetchTM sieve shaker. Sieving has been a fundamental grain-size technique for the best part of a
century.

14.19.4.2 Analytical Techniques for Materials Composed determination of sand-sized material or greater (McManus,
Primarily of Sand, Silt, and Clay 1988). Sieve screens for dry sieving are commonly made of
strong wire mesh of stainless steel or brass with finer wire
Geomorphological investigations of landforms composed
meshes used for smaller particles. In contrast, wet sieving
primarily of sand, silt, and clay material, generally require the
equipment is generally stainless steel or plastic with Nylon
routine application of PSA (see reviews of Syvitski et al.,
sieves.
1991a, b; Cheetham et al., 2008).
Several methods of obtaining particle-size distribution data
are presently available for use in sandy/muddy sediments. 14.19.4.2.1.1 Dry sieving
These include sieve/hydrometer (combined), X-ray attenu- The coarsest sieve required is placed at the top of the stack in
ation, scanning electron microscopy (SEM), sedimentation, which the square screen openings become progressively
and laser diffraction techniques. The particle-size distribution smaller down the stack. Sieving generally gives the inter-
of sediments has many applications in geomorphology, for mediate measure of a particle because of the way particles
example, in fluvial geomorphology the PSA of a candidate orient themselves to pass through the mesh. The particle sieve
deposit allows for the determination of the paleocurrent size can either be defined as the smallest sieve size through
conditions required to quarry and transport sediment grains which a particle can pass (Dpass), or as the largest sieve size
(Hjulstrom, 1939; Moss, 1962, 1963; Miller et al., 1977). through which the particle did not pass, the retaining sieve
Clearly this is not the only consideration, as grain morphology size (Dret). The number of sieves reflects the number of size
(see Chapter 14.20) also affects a particles resistance to en- fractions and hence the relative level of analytical detail.
trainment in a fluid and its time in suspension (Folk, 1980; Once the sample is in place, the sieve stack is generally
McCave et al., 2006; Cheetham et al., 2008). In almost all agitated by a mechanical shaker for a predetermined time
aspects of geomorphology PSA along with determinants of interval, usually 15–20 min. The material retained in each
grain morphology and the analysis of sedimentary structures, sieve is commonly emptied onto a sheet of paper, by tapping
provide a primary diagnostic tools for the interpretation and gently in a direction diagonal to the mesh and sweeping with
reconstruction of palaeoenvironments (e.g., Leopold et al., appropriate sieve brushes to release particles that are stuck in
1964; Moss, 1972; Reineck and Singh, 1975; Schumm, 1977; the mesh (McManus, 1988). Each fraction of sediment
Folk, 1980; McCave and Syvitski, 1991; Lario et al., 2001; obtained is weighed (usually to B0.01 g). The sieve mesh
Miall, 2006; Cheetham et al., 2008). sizes, raw weights, weight percentages, and cumulative per-
centages, finer or coarser than the specific sieve, can then be
calculated.
14.19.4.2.1 Sieving
Sieving is perhaps the oldest and most traditional of the 14.19.4.2.1.1.1 Potential errors in dry sieving Dry sieving
analytical techniques for sand and gravel sized material is subject to potential errors from many sources such as;
(Krumbein and Pettijohn, 1961; Folk, 1980 Buller and particle aggregation where aggregates form clusters con-
McManus, 1979). With an appropriate stack of sieves siderably larger than the original single component grains
(Figure 4), particles between 0.002 and 250 mm in size may (this can be checked by hand lens or binocular microscope);
be separated into regular size class intervals. Although it is incomplete cleaning of the mesh (leftover particles can
possible to sieve silt particles, sieving is primarily used for size restrict the aperture spaces) and overloading of sieves that
232 Measuring and Analyzing Particle Size in a Geomorphic Context

can restrict the opportunity for particles to progress down the by the equation:
nest to the appropriate mesh. McManus (1965) suggested
that a sieve load should not exceed 4–6 grain diameters D2 ðrs  rÞg
Vs ¼
high. Overloading may also cause mesh distortion which 18m
also introduces error. For sand samples B100 g of material
is generally adequate for sieving but larger weights are re- where Vs is the settling velocity, D the particle diameter (mm),
quired for coarse gravelly deposits (McManus, 1988). rs and r the densities of the grain and water, respectively, g is
Considerable disagreement also exists on the standardi- the gravitational acceleration, and m is the dynamic viscosity of
zation of sieving time (Lewis, 1994a, b; McManus, 1988; the fluid. The size can then be determined by reorganizing to
Dalsgaard et al., 1991). For most purposes acceptable sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
18mVs
reproducibility of analyses is obtained after 20 min D¼
(Dalsgaard et al., 1991). ðrs  rÞg

14.19.4.2.1.2 Wet sieving 14.19.4.2.2.1 The pipette method

Since many types of sediments contain mixtures of coarse One inexpensive sizing technique for the mud fraction is the
and fine particles, the dry sieving technique may not be pipette method that relies on the principle that particles in a
appropriate for examination of the finer parts of the dilute suspension settle through a column of water at vel-
sample. Traditionally, the finer particles require other ocities that are dependent on their size (see Krumbein and
methods of analysis (e.g., pipette analysis) for which the Pettijohn, 1961). If the sample behaves according to the
separation of coarse and fine fractions was customarily made Stokes Law then repeatedly sampling at a constant depth in
at the silt/sand boundary. To wet sieve samples, some workers the water column yields progressively finer and finer sedi-
dry the whole sediment sample to a constant weight at ments. Investigating the temporal variations of particle con-
B110 1C before resaturating in water containing a dispersant centrations at the set depth will indicate the relative
such as sodium hexametaphosphate (Calgon). The sample is abundance of particles whose diameters may then be calcu-
then periodically stirred in water for B1 h before being lated. Standard tables of timing are available (Table 2) al-
washed through 2000 and 63 mm sieves until the water runs though when applying Stokes’ Law the dynamic viscosity of
clear. The total content of different fractions can be deter- the water is important and it is therefore important to try
mined as the difference between the initial and the retained to ensure thermal consistency as the analysis may take
material weights. Although this analysis comes with a cau- several days.
tionary note, some clay sized particles can become structurally Generally a pipette is inserted slowly and gently into the
altered at B100 1C; so if any further study is to be undertaken fluid until the inlet is at a given distance below the surface. An
on the clay fraction, this method may not be appropriate. If initial volume is withdrawn (e.g., 20 ml) at the given time
further analysis is to be undertaken one can use two identical from the start of settling (based on Stokes law). The sediment-
subsamples (one dried and one not). This will allow the col- laden fluid is then released into a small labeled beaker with
lection of an undisturbed mud fraction, whose proportional excess particles washed into the beaker using distilled water.
contribution to the whole sediment is now known from the Successive samples are extracted from the suggested depth in
other subsample. the column at time intervals which calculations reveal particles
of known diameter (see Table 2 after Krumbein and Pettijohn,
1961). The number of aliquots extracted relates to the com-
14.19.4.2.2 Sedimentation methods for sand, silt, and pleteness of the dataset. To determine silt and clay content at
clay 20 1C one needs only withdraw samples at 20 s at 20 cm for
Before the advent of particle-size analysis by laser, the grain- silt (4 phi) and 2 h and 3 min at 10 cm for clay (8 phi). The
size analysis of silts and clays was primarily based on indirect beakers containing the sampled fluids are then dried to con-
computations of diameters based on observation of the grains stant weight before weighing. A small correction may be
in fluids or the response of the fluids to grain-induced dis- needed for dispersant and this is achieved by comparing the
placement. These methods are collectively known as sedi- weights of successive withdrawals as both contain the same
mentation methods and are based on the speed with which amount of dispersant (B2% of the original amount of dis-
particles settle through fluids. Such information yields settling persant powder). Since each aliquot is B2% of the original
velocities from which equivalent grain diameters are then sediment water mixture a simple calculation will give the full
computed. size distribution based on continued withdrawals to the finest
Computation of the equivalent diameters from settling required size. The total weight of sediment undergoing an-
velocities of particles is based on Stokes’ Law of settling which alysis should be 50 times the sediment extracted in the initial
assumes that sediment particles are dominantly spheres of (58 s) sample. Commonly a stable temperature may be pos-
identical densities (almost certainly not true for most natural sible but not to the required 20 1C. Tables have been estab-
sediments). Stokes’ law is based on the principle that when a lished for other temperatures (Table 2) where the different
particle is in static water it settles at a constant velocity during viscosities require the reduction or extension of sampling
which the gravitational force exerted on the particle is bal- times. The quantities of initial sample used vary, although
anced by fluid resistances represented by viscosity and the generally 10 and 20 g l1 give the most satisfactory reprodu-
particle-drag coefficient. The balance is normally represented cibility. Higher concentrations can be hindered by settling
 Measuring and Analyzing Particle Size in a Geomorphic Context 233

Table 2 Withdrawal times and depth tables for 20, 24, and 32 1C based on Krumbein and Pettijohn (1961) and Milner (1962a, b). Times are in
hours (h) minutes (min) and seconds (s)

Analysis at 20 1C Analysis at 24 1C Analysis at 32 1C

Phi Depth (cm) Time Depth (cm) Time Depth (cm) Time

4 20 20 s 20 20 s 20 20 s
5 10 1 min 56 s 10 1 min 45 s 10 1 min 30 s
6 10 7 min 44 s 10 6 min 58 s 10 6 min
7 10 31 min 10 28 min 10 23 min
8 10 2 h 3 min 10 1 h 51 min 10 1 h 34 min
9 5 4 h 6 min 10 7 h 24 min 5 2 h 58 min
10 5 16 h 24 min 5 14 h 50 min 5 6 h 11 min
11 5 62 h 15 min 5 59 h 20 min 5 24 h 43 min
12 5 237 h 20 min
13 5 949 h

convection where upward motion of circulating waters im- single sizes and combinations of sizes (e.g., Schlee, 1966).
pedes settling (Kuenan, 1968). The main advantage of this Recently international comparative calibration exercises have
simple technique is that it requires little specialist equipment occurred (e.g., Dyer et al., 1996; Agrawal and Pottsmith, 2000
and is relatively easy to do. for marine and estuarine sedimentation).
Generally the particle sizes in the sediment are calculated
from the settled sample weight at specific time intervals. The
14.19.4.2.2.2 The sedimentation tube
measured settling velocities are converted into ‘equivalent
The second technique based on settling velocity is the sedi-
sedimentation diameters,’ or the diameters of spheres settling
mentation tube, which unlike the pipette technique that is
at the same rate as the natural particles being tested (e.g., Gibbs
confined to analysis of silts and clays, can be applied to sand
et al., 1971). In a noted advance Baba and Komar (1981) set
samples as well. The roots of this popular technique can be
about investigating the settling of natural quartz sands in an
traced back to the early twentieth century (e.g., Emery, 1938).
attempt to derive intermediate diameters from settling tubes
The technique is based on the principle that particles released
that could be compared and combined with sieve analysis.
simultaneously at the top of a large, broad water-filled tube
Their work resulted in a paper by Komar and Cui (1984),
settle out into a smaller diameter tube at the base. Initially the
which extended the work of Gibbs et al. (1971) and enabled
heights of accumulation at known time intervals are measured
the calculation of intermediate grain diameters (McManus,
by optical micrometer, pan collection, or cameras and the
1988). Although some early comparisons of sizes determined
particle sizes calculated (Dyer et al., 1996). The introduction
by sieving and settling techniques suggested that settling
of pressure transducers and electrical recorders to determine
techniques overestimated the size of fines and under-estimated
the temporal variations of weight in the water column were
the size of sands (e.g., Sengupta and Veenstra, 1968; Sanford
also significant advances (e.g., Zeigler et al., 1960). The
and Swift, 1971) and the application of correction factors
growing popularity of laser- and image-based grain-size an-
allowed the comparison of intermediate diameters from both
alysis techniques has meant that this method is not as
sieving and settling techniques within appropriate error
prominent as it used to be although several companies still
(Komar and Cui, 1984). These authors did note some devi-
make instruments of this type (e.g., the sedimentometer by
ations related to heavy minerals grains with densities con-
Topas).
siderably greater than quartz and to minerals whose shapes
As in the pipette method, the particles are released from
differ from the spherical (e.g., mica flakes) (Box 2).
the water surface, generally by holding 2–5 g of wet sample on
a plate and lowering into the water column. The sample is
kept small to minimize hindered settling which can decrease 14.19.4.2.3 Laser diffraction analysis (LDA)
settling velocities and give unrepresentative results (Richard- LDA has become increasingly popular over the past 15 years as
son and Zaki, 1954). The tube diameter is also important, a method for the analysis of particle-size distribution of
particularly for coarse sediments (Channon, 1971). Sedimen- sediments and soils. A variety of instruments have been de-
tation tubes also need static water conditions and since most veloped by different manufacturers (e.g., Beckman-Coulter,
are at least 2 m in height they can develop internal convection Malvern, Retsch, Horiba), which vary in terms of the range of
currents unless maintained in thermally stable conditions (see sizes that can be analyzed and the diversity of options for data
discussion of Dyer et al., 1996). processing. Standard operating procedures for laser diffraction
The interpretation of data from sedimentation tubes is analysis have also been published (e.g., ISO, 1999a,b; ASTM,
complex because although individual particles may permit 2010), although they are fairly generalized and aimed mainly
confident calculation of particle diameters, the behavior of at industrial powders rather than natural soils and sediments.
clusters of particles of a range of sizes is less easy to calculate The LDA technique works on the principle that a laser directed
(Dyer et al., 1996). Traditionally individual tubes are cali- through a cell will be diffracted by particles that pass through
brated against well characterized particles, such as single the beam. The instrument measures the particle-size charac-
spheres (e.g., Zeigler and Gill, 1959) or clusters of spheres of teristics of a sediment sample using the principles of laser
234 Measuring and Analyzing Particle Size in a Geomorphic Context

Box 2 Problems with sedimentation techniques and fine-grained natural material

Unfortunately, as particles decrease in diameter, they become increasingly cohesive as surface ionic charges grow in relative
significance. In rivers and estuaries, these materials commonly form flocs or aggregates with strong inter-particle cohesion. A
solution lies in the use of the natural waters in which sediment and ionic concentrations may be high (Peirce and Williams,
1966; Dalsgaard et al., 1991). Other problems are encountered where organic particles interfere in sedimentation (Duck,
1983). This limitation can be partially overcome by using a dispersing agent during sample preparation. Sodium hexame-
taphosphate solution is perhaps the most popular dispersing agent, and it can be made up in the laboratory or one can use
commercially available mixtures, like Calgon.

diffraction and relies on the fact that the diffraction angle is 14.19.4.2.4 Other PSA techniques
directly proportional to the particle size. In all instruments the Other techniques used in geomorphology include electrical
angle and intensity of laser light scattered by a suspended zone sensing (EZS), X-Ray sedimentation and scanning electron
sediment sample are selectively measured and are generally microscopy (SEM). The EZS approach is based on the Coulter
converted to a volume distribution based primarily on the Mie principle (e.g., Coulter counter) where a sample is dispersed at
optical theory (some machines e.g., Malvern Mastersizer 2000 low concentration in an electrolytic (i.e., conducting) solution,
can be set to use Fraunhofer) (de Boer et al., 1987). Previous and is then drawn through a small aperture (sensing zone) that
work has shown that, if appropriate sample preparation and has electrodes on either side of it. As each particle enters the
handling procedures are employed, laser diffraction provides a sensing zone it causes a temporary change in the measured
precise method for the analysis and comparison of sediments, electrical impedance across the opening. The amplitude of this
soils and similar material (Blott and Pye, 2006; Pye and Blott, impedance pulse is proportional to the particle’s volume and
2004; Cheetham et al., 2008). However, less attention has hence, can be used to infer size. Potential errors come from the
been given to the question of ‘accuracy’, although it has long coincident passage of two or more particles through the sensing
been recognized that there will be significant differences be- zone as this can cause the instrument to count the combined
tween results obtained using laser diffraction compared with pulse height of multiple small particles as a single large particle,
other methods (Agrawal et al., 1991; Agrawal and Pottsmith, thereby skewing the size distribution. A second potential
2000). Many authors have also shown that laser diffraction problem in EZS, which is also a concern in LDA, is particle
underestimates the amount of clay compared with pipette and asymmetry where flaky particles (e.g., clays) rotate as they pass
hydrometer analysis (e.g., McCave et al., 1986; Loizeau et al., through the aperture leading to potential oversizing (Milligan
1994; Konert and Vandenberghe, 1997; Beuselinck et al., and Krank, 1991).
1998, 1999a, b), and significant differences have also been The X-ray sedimentation technique for determining the
noted between laser diffraction and sieving results for some relative mass distribution of a sample by particle size is based
sands (e.g., Shillabeer et al., 1992; Konert and Vandenberghe, on two physical principles: sedimentation theory and the ab-
1997; Hayton et al., 2001). In contrast, recent work by sorption of X-radiation. These two theories are embodied in
Cheetham et al. (2008) showed that the use of LDA for fluvial an analytical instrument called the SediGraph. Any particle
sands produced precise and repeatable results when compared settling in a liquid will achieve a terminal velocity when the
to other techniques. gravitational force balances the buoyancy and drag forces on
the particle. This is dependent on the size and the density of
14.19.4.2.3.1 General process for LDA analysis the particle, and the density and viscosity of the liquid.
Background measurements are generally taken on the A beam of photons (X-rays, in this case) passing through a
suspension medium (characteristically water) based on rec- medium is attenuated in proportion to the path length
ommendations from the manufacturer. To assess the com- through the medium, its concentration, and the extinction
parability of samples prepared using different methods of coefficient of the medium (Coakley and Syvitski, 1991). Using
dispersion, two sample sets can be analyzed. The first set is Stokes and Beer-Lambert laws mean that interpretation of raw
agitated under ultrasound for 1 min or until the sample is data is achieved to determine the relationship between the
visibly disaggregated. The other group can be treated with basic measurements and the reported size distribution
50 ml of 0.1% sodium hexametaphosphate and soaked (Loveland and Whalley, 2001).
overnight. SEM can be used to count the number of particles in dif-
Commonly, the samples are added to the instrument using ferent size fractions (Ly, 1978; Goldberg and Richardson,
a large beaker or dispersion bath and agitated by ultrasound 1989). This type of analysis is a time-consuming technique,
until the laser obscuration is in the optimum range for the but it can be sped up by using a grid of set scale and choosing
instrument. Too high obscuration levels results in inter-par- a field of view with a specific magnification then counting
ticle interference whereas too low obscuration leads to poor particles of a particular size fraction (Cheetham et al., 2008).
signal-to-noise ratio. The data output can then be grouped Automated microscopy and image-analysis techniques can
according to standard outputs including, standard sieve-size provide a means of repeatable and routine characterization of
fractions for correlation statistics (Folk, 1980; Tanner 1991a, b; particle size and shape using automated microscopy and
Christiansen and Hartmann, 1991), Percent volume data can image-analysis techniques. Commonly, particle shape infor-
also be used (see Section 14.19.5.3). mation is generated from the analysis of thousands of particles
 Measuring and Analyzing Particle Size in a Geomorphic Context 235

Table 3 Grain-size parameters and their common indicative meaning

Parameter Description Main indicative meaning in PSA

Mode On a size frequency histogram the size class in which the The modal size represents the commonest grain size in a
greatest percentage is recorded provides the modal class. distribution. Frequency curves with several peaks indicate
On the size frequency distribution plot the highest point on polymodal distributions potentially indicating the presence
the curve provides the modal value of more than one population of grains
Median (Md) Half of the grains are coarser and half finer than the median Although useful for unimodal sediments, in polymodal
diameter. This is most readily determined from the 50% distributions (most natural sediments are polymodal) the
line of the cumulative distribution curve median may fall in the tails of two subpopulations of grains
Mean (M) The measure of average grain size, the mean is computed The mean is the most commonly presented parameter of
from sizes of particles spread through a range of percentile sediment size and is often presented with a measure of
values. In its simplest form the Graphic Mean, M can be standard deviation to express the basic properties of a
calculated by averaging the 16th, 50th and 84th percentiles sediment (e.g., poorly sorted fine sand)
in a cumulative frequency curve. This assumes that three
values alone are sufficient to give a useful mean. More
percentage values may be averaged which approximates to
the mean of moment statistics
Sorting (s) There are two common measures of sorting; the graphic Sorting relates to the range of grain sizes. Well-sorted
standard deviation (s) (Inmam, 1952), which provides a samples have a small range (aeolian sediments are
measure of the spread of one standard deviation on either commonly well sorted (i.e., grain sizes are very similar)
side of the mean and the inclusive graphic standard hence it can be said that winds sort sediments well). Glacial
deviation (Folk and Ward, 1957) which defines a spread of sediments are commonly poorly sorted (meaning there is a
1.65 standard deviations either side of the mean large spread of grain sizes in glacial deposits)
Skewness (a) In a normal distribution with a bell-shaped frequency curve Skewness has a positive or negative value when more fine or
the median and mean values coincide. Any tendency for a more coarse materials are present. Skewness is a positively
distribution to lean to one side, i.e., to deviate from or negatively signed dimensionless number; it has neither
normality, leads to differences between the median and metric nor phi value and lies within the range  1 to þ 1
mean values. These differences are used to characterize
the asymmetry or skewness of the curve and are
determined from the value of the mean less the median, all
divided by the range used in defining the mean
Kurtosis A measure of the ‘peakedness’ of the distribution. Although Generally kurtosis is a second indicator of sorting where very
frequently calculated kurtosis rarely receives much flat curves equate to poorly sorted sediments or those with
attention. The kurtosis is a measure of the dispersion and bimodal frequency curves that are platykurtic, whereas
the normality of the distribution. Kurtosis is a ratio of the strongly peaked curves are considered leptokurtic and
spreads of the tails and center of the distribution (and is exhibit very good sorting of the central part of the
therefore also dimensionless) distribution

and displays of particle size and shape data are supported by size distributions (i.e., sorting, skewness, and kurtosis) (see
images of all the particles to provide further visual under- Table 3 and glossary).
standing of the measurement data (Goldberg and Richardson,
1989). A number of shape parameters can also be calculated
for each particle in order to increase the sensitivity of the 14.19.5.1 Use of Bivariate Plots (Scattergraphs)
analysis to subtle variations in particle morphology (Chee-
Numerous approaches have been utilized to interpret particle-
tham et al., 2008).
size data in geomorphology. Commonly, summary statistics
(mean grain size, sorting, skewness, and kurtosis) of particle-
size distributions (Folk, 1980; Folk and Ward, 1957) have
been plotted on bivariate scattergrams, from which researchers
14.19.5 Interpretation of Particle-Size Data have identified graphic envelopes within which deposits of
particular environments are plotted (Mason and Folk, 1958;
Although the analysis of particle size in sediments is Friedman, 1961, 1967, 1979; Moilola and Weiser, 1968; Buller
commonly used in the identification of sedimentary en- and McManus, 1972; Tanner, 1991a, b; Duck, 1994; Lario
vironments, landforms, and facies in a variety of depositional et al., 2002). Several other approaches are reviewed in detail
settings (e.g., Folk and Ward, 1957; Huang and Goodell, 1967; by Syvitski (1991a, b) and Gale and Hoare (1991). Despite the
De Falco et al., 2006; Nichol et al., 2007; Donato et al., 2009) extent and detail of PSA investigations over the last 100 years,
a considerable variability remains in presentation techniques. most attempts to determine the environment of deposition
The pioneering work by Udden (1914), Wentworth (1922), from particle-size data are primarily site specific and are gen-
Krumbein (1934, 1938), Inmam (1952), and Folk and Ward erally inadequate as an unequivocal means of palaeoenvir-
(1957) established descriptive statistics that are still used in onmental reconstruction. However, it is clear that particle-size
geomorphology and sedimentology to characterize particle- data obtained by any means can be used in conjunction with
236 Measuring and Analyzing Particle Size in a Geomorphic Context

Shepard’s classification system

Modified by Schlee (1973)
Clay Gravel

Clay
nd 25 Gravel
%

l
ve
sa

gra
of

Sandy Silty

of
nt

50 50

nt
Clay Clay
rce

% %

rce
Pe

Pe
Sand
75 Clayey Silt Clayey Gravelly sediment
% Sand Silt
Clay

Sand Silty sand Sandy silt Silt

Sand, silt and clay (gravel <10%)

Sand Silt Sand Silt and clay

(a)
Folk’s classification system

Gravel Sand
90
% S
80 G
% G, gravel; g, gravelly S, sand; s, sandy

nd
l
ve

(g), slightly gravelly

sa
Z, silt; z, silty
gra

S, sand; s, sandy
of
cS mS zS M, mud; m, muddy
of

50
nt

M, mud; m, muddy C, clay; c, clayey

%
nt

rce

mG msG
rce

30
Pe

% sG
Pe

5% gM gMS sC sM sZ
gS
Tr 10
ac (g)mS (g)S 1% %
e (g)M (g)sM 0.0
M sM mS S C M Z
Mud 1:9 1:1 9:1 Sand Clay Silt
Sand : Mud ratio
(b)

Figure 5 Tripartite sediment classification schemes for sediments. Most geomorphologists use one of the systems described either by Shepard
(1954) or Folk (1954, 1980). The original scheme by Shepard (1954) does not allow for sediments with significant amounts of gravel and was
subsequently modified by the addition of a second ternary diagram to account for the gravel fraction (Schlee, 1973). The system devised by
Folk (1954, 1980) is also based on two triangular diagrams and has 21 categories. Folk’s scheme stresses gravel focusing on high velocity
flows whereas Shepard’s classification scheme emphasizes the ratios of sand, silt, and clay because they reflect sorting and reworking
(Poppe et al., 2000). Courtesy of USGS.

other evidence to determine the environment of deposition of environments using these graphic presentation approaches
a sedimentary body or landform. For a discussion on bivariate (McManus, 1988; Lario et al., 2002). Samples are generally
plots the reader is referred to Tanner (1991a) who acknow- plotted onto ternary plots (Figure 5) or scattergraphs
ledges the inherent problems faced when using such plots and (bivariate plots). Attempts to discriminate between different
examined methods that may assist in the resolution of the depositional settings, via bivariate plots, are based on the
environment of deposition. primary assumption that the statistical parameters reliably
reflect differences in sediment transportation and deposition
(Sutherland and Lee, 1994a, b). Many studies have tried to
14.19.5.2 Scattergraphs and Ternary Diagrams make environmental sense from bivariate plots of parameters
that describe the sample size spectrum (Stewart, 1958; Fried-
For many years sedimentologists and geomorphologists have man, 1961, 1967; Buller and McManus, 1972; Friedman and
attempted to use scattergraphs of grain-size data to distinguish Sanders, 1978; Tanner, 1991a, b) and the success has been
between different depositional environments or landforms. varied for several reasons, such as over-simplified discrimin-
No universal models exist to distinguish past depositional ation, for example, beach versus river where dunes and
 Measuring and Analyzing Particle Size in a Geomorphic Context 237

6.0 Those familiar with grain-size analysis statistics are fully

aware that the use of these statistics was not developed for
Hillslope use in a particular facies or landform analysis framework but
5.0 material
was primarily developed for the ease of calculation and
characterization of the PSD from graphical cumulative fre-
4.0 Estuarine
Inc Mean ()

Flood tide delta

muds
quency curves. As such, the derived statistics were simply a
sediments convenient way of working with the data (primarily from
3.0
sieving and pipette at the time). In geomorphology, the stat-
istical parameters used in the middle of the twentieth century
2.0 have persisted. This is in spite of significant advances in
Sandsheet
technology (i.e., laser particle-size analysis) that have greatly
1.0 sediments enhanced the ability to characterize large quantities of PSD
data. This can now be done with considerable statistical rigor
0 using modern computing.
0 2.0 4.0 6.0 8.0
Inc Std Dev ()
14.19.5.4 Using Modern PSD Datasets
Figure 6 A scattergram of inclusive mean (Inc Mean f) and
inclusive mean standard deviation (Inc Std Dev f) of samples taken Most modern instrumentation (e.g., LDA) can provide the
from a Holocene estuary fill that contains two overwash sandsheets complete PSD in a high resolution, precisely measured and
(Switzer et al., 2005). Grain populations indicative of overwash easily comparable form that is then available for a variety of
sandsheets, estuarine muds, hillslope materials, and floodtide deltaic statistical analysis using multivariate techniques (e.g., cluster
sediments are identified. analysis) that make the identification of PSA determined facies
and environments possible. Rapid analysis of PSDs can now
other depositional settings are ignored (Socci and Tanner, also provide abundant data for high-resolution grain-size
1980). For example, inclusive mean and inclusive mean studies on cores and surface samples that with the benefit of
standard deviation were plotted from samples taken from multivariate statistics and 3D surface plots allows better char-
a Holocene paleoestuary that contained two overwash sand- acterization of sedimentary processes (e.g., Beierle et al., 2002;
sheets (Switzer et al., 2005). The analysis clearly distinguished van Hengstum et al., 2007; Donato et al., 2008, 2009) Com-
the overwash sandsheets from estuarine muds, hill-slope ma- bining the PSD data with other relevant data (e.g., texture,
terials and floodtide deltaic sediments (Figure 6). This is not carbonate content, faunal data) can adequately discriminate
always the case and the comparison between mean grain size landforms and environments (e.g., Switzer et al., 2005; Switzer
and sorting is generally over simplified and in most cases there and Jones, 2008; Pilarczyk et al., 2011; Van Hengstum et al.,
is a clear covariance between mean grain size and sorting 2011).
(Tucker, 1990). Griffiths (1967) explained that as both mean In one recent example Donato et al. (2009) tested a new
grain size and sorting are hydraulically controlled in most technique using PSD data and multivariate cluster analysis (Q-
environments the best-sorted sediments in almost all en- mode) to define modern environmentally defined facies dis-
vironments will have mean grain size in the range of fine tributions determined in Sur Lagoon, Oman. In this example
sand sizes. they used blind clustering of PSD data and compared it to
surface topography, remote sensing and visual field surveys.
This allowed a comparison of the PSD data to the mapped
14.19.5.3 Recent Advances in Data Presentation
distribution of sedimentary facies to determine the effective-
Where PSA plays a major role in the analysis most researchers ness of PSD for determining lagoon sub-environments in the
will find that conventional graphical summary statistics such geological record (see previous work by Loizeau and Stanley,
as graphical mean, median, mode, standard deviation, skew- 1993, 1994). Donato et al. (2009) found that surface plots of
ness, and kurtosis are still the preferred method of displaying PSD data allowed a qualitative interpretation of the charac-
PSDs (e.g., Folk, 1966, 1980; Griffiths, 1967; Friedman and teristics of the entire PSD that can provided key insights into
Sanders, 1978; Wang and Ke, 1997; Al-Zamel et al., 2007; depositional processes and the dynamics of modern en-
Mwakumanya and Bdo, 2007). This is despite the noted vironmental conditions. This method is especially useful for
limitations of such statistics to adequately characterize the distinguishing multiple sedimentary processes, which can
grain-size distribution (Blott and Pye, 2001). appear as additional modes within the PSD (see studies of
Recent work has seen some notable attempts at new an- Pilarczyk et al., 2011; Van Hengstum et al., 2011). Given the
alysis and display techniques (e.g., Donato et al., 2009). This ease at which large amounts of particle size data can now be
is primarily a result of recent advances in instrumentation obtained, such techniques can now be applied to most de-
(e.g., LDA) that mean it is now possible to analyze many positional environments and landforms at high spatial or
samples economically. Many researchers have noted that PSD temporal resolution. Donato et al. (2009) noted that a com-
from a considerable number of sedimentary environments are bination of conventional summary statistics with PSD surface
required before a statistically significant application of derived plots increases the utility of PSA as a paleoenvironmental
statistical parameters could be used to characterize environ- proxy for identifying changes in clastic and organic de-
mental trends (e.g., Ward, 1980; McLaren, 1981; McLaren and positional processes in lake sequences. Similar approaches
Bowles, 1985; Donato et al., 2009; Pilarczyk et al., 2011). have since been applied to other environments. One example
238 Measuring and Analyzing Particle Size in a Geomorphic Context

C1 C8 C10 C5 C9 C12
0
Slack-
water
Carbonate Carbonate
10 mud mud

20
Core depth (cm)

30
Diamict
Slack- Clay Silt Sand
water
10 5 0
40 Diamict
Diamict Slack- Phi()
Clay Silt Sand water
0 2.5 5
10 5 0 Volume (%)
50 Phi()
Clay Silt Sand
0 2.5 5 Anchialine
Volume (%) 10 5 0 facies
Phi()
60
0 3 6
Volume (%)
Clay Silt Sand Littoral
70 Anchialine Clay Silt Sand facies
10 5 0
facies
10 5 0 Phi()
Phi() Vadose
Clay Silt Sand Vadose facies
0 2.5 5
facies
10 5 0 0 2.5 5 Volume (%)
Phi()
%

Volume (%)
e
um
l
Vo

0 1 2 3
Volume (%)
ze

De
cl e

pth
r ti od
si
e
pa M

in
co
re

Figure 7 Sediment particle size data from a coastal karst basin in Bermuda from Van Hengstum et al. (2011). Sediment samples are plotted
down core using size as the x axis, depth down core as the y axis and volume as the z axis. This technique is based on the work of Bierle
(2001) and shows particular promise for presenting data where high resolution sampling has been conducted on polymodal sediments.
Reproduced from van Hengstum, P.J., Scott, D.B., Gröcke, D.R., Charette, M.A., 2011. Sea level controls sedimentation and environments in
coastal caves and sinkholes. Marine Geology 286(1–4), 35–50.

of recent application is in the investigation of paleo-landforms analytical techniques, Webb (2008) stated that ‘‘Provided that
in coastal karstic basins (Figure 7) by Van Hengstum et al. the instruments used are capable of producing high-quality
(2011). In this example, sediment particle size is plotted down data, the pertinent questions, then, are, ‘was the sample prop-
core using size as the x axis, depth down core as the y axis and erly prepared and properly presented to the instrument,’ and
volume as the z axis. The different modes are visible in many ‘were the analytical parameters applied correctly’ ’’ (Webb,
cases and this technique shows particular promise for poly- 2008: 3). If the answer to both is ‘yes,’ then both analytical
modal sediments. results probably are equally correct; they are just expressed in
different terms. Recent reviews clearly show that some Earth
scientists still express concern about comparisons of particle-
14.19.6 The Same but Different: A Concluding Note sizing results by different techniques (see discussion of Chee-
on Comparing Different Techniques tham et al., 2008). The techniques commonly compared are
sieving, sedimentation, imaging (including microscopy and
In a recent commentary on the ‘correctness’ of the results ob- machine vision), EZS, and light scattering (laser diffraction).
tained for particle-size measurements by two or more different Invariably, the determination of any particle-size distribution
 Measuring and Analyzing Particle Size in a Geomorphic Context 239

on the same sediment sample by all of these techniques and Cheetham, M.D., Keene, A.F., Bush, R.T., Sullivan, L.A., Erskine, W.D., 2008. A
others will, in all cases where the samples are not perfect comparison of grain-size analysis methods for sand-dominated fluvial
sediments. Sedimentology 55, 6.
spheres of homogenous weight and composition, yield differ-
Chen, W., Qinghai, X., Xiuqing, Z., Yonghong, M., 1996. Palaeochannels on the
ent results for mean size, modal size, and quantity distribution North China Plain: types and distributions. Geomorphology 18(1), 5–14.
by size. So which technique is ‘correct’ or provides the most Christiansen, C, Hartmann, D., 1991. The hyperbolic distribution. In: Syvitski,
‘accurate’ representation of the sediment population? If the J.P.M. (Ed.), Principles, Methods, and Application of Particle Size Analysis.
sample is prepared properly and the analytical parameters are Cambridge University Press, Cambridge, pp. 237–248.
Coakley, J.P., Syvitski, J.P.M., 1991. Sedigraph technique. In: Syvitski, J.P.M. (Ed.),
correctly applied then all the answers are ‘correct,’ but they are Principles, Methods, and Applications of Particle Size Analysis. Cambridge
just measuring different things and applicants of these techni- University Press, New York, pp. 129–142.
ques must be aware of the limitations of each technique. It is Dalsgaard, K., Jensen, J.L., Sorensen, M., 1991. Methodology of sieving small
clear that geomorphologists must consider factors other than samples and calibration of sieve set. In: Syvitski, J.P.M (Ed.), Theory, Methods
and Applications of Particle Size Analysis. Cambridge University Press, New
size or sample preparation that could affect the reported size
York, pp. 64–75.
value. Generally errors from these sources are associated with de Boer, G.B.J., de Weerd, C., Thoenes, D., Goossens, H.W.J., 1987. Laser
nonspherical particles, variation in particle density or even the diffraction spectrometry: Fraunhofer versus Mie scattering. Particle
inappropriate application of the measuring instrument (e.g., Characteristics 4, 14–19.
Sieves for coralline sands (see discussion of Kench, 1997)). De Falco, G., Baroli, M., Murru, E., Piergallini, G., Cancemi, G., 2006. Sediment
analysis evidences two different depositional phenomena influencing seagrass
distribution in the Gulf of Oristano Sardinia, Western Mediterranean. Journal of
Coastal Research 22, 1043–1050.
References Donato, S.V., Reinhardt, E.G., Boyce, J.I., Pilarczyk, J.E., Jupp., B.P., 2009. Particle-
size distribution of inferred tsunami deposits in Sur Lagoon, Sultanate of Oman.
Adams, J., 1979. Gravel size analysis from photographs. ASCE Journal of the Marine Geology 257(1-4), 54–64.
Donato, S.V., Reinhardt, E.G., Boyce, J.I., Rothaus, R., Vosmer., T., 2008. Identifying
Hydraulics Division 10510, 1247–1255.
tsunami deposits using bivalve shell taphonomy. Geology 36(3), 199–202.
Agrawal, Y.C., Pottsmith, H.C., 2000. Instruments for particle size and settling
Duck, R.W., 1983. Settling velocity analysis of fine grained, freshwater sediments: a
velocity observations in sediment transport. Marine Geology 168(1–4), 89–114.
cautionary note. Journal of Water Resources 2, 23–29.
Agrawal, Y.C., McCave, I.N., Riley, J.B., 1991. Laser diffraction style analysis.
Duck, R.W., 1994. Application of the QDa-md method of environmental
In: Syvitski, J.P.M. (Ed.), Principles, Methods and Applications of Particle Size
discrimination to particle size analyses of fine sediments by pipette and
Analysis. Cambridge University Press, Cambridge, pp. 119–128.
sedigraph methods: a comparative study. Earth Surface Processes and
Al-Zamel, A., Al Sarkawi, M., Khader, S., 2007. Coastal geomorphology,
Landforms 19(6), 525–529.
hydrodynamics, and biofacies in the intertidal and subtidal areas of Umm Al-
Dyer, K.R., Cornelisse, J., Dearnaley, M.P., et al., 1996. A comparison of in situ
Namil Island, Kuwait Bay, Kuwait. Journal of Coastal Research 23, 501–514.
techniques for estuarine floc settling velocity measurements. Journal of Sea
ASTM Standard C33E112 – 2010 Standard Test Methods for Determining Average
Research 36(1-2), 15–29.
Grain Size Using Semiautomatic and Automatic Image Analysis ASTM
Emery, K.O., 1938. Rapid method of mechanical analysis of sands. Journal of
International West Conshohocken, PA, USA. doi: 10.1520/E1382-97R10
Sedimentary Petrology 8, 105–111.
Baba, J., Komar, P.D., 1981. Measurements and analysis of settling velocities of
Finkl, C.W., 2004. Coastal classification: systematic approaches to consider in the
natural quartz sand grains. Journal of Sedimentary Petrology 51, 631–640. development of a comprehensive scheme. Journal of Coastal Research 20,
Beierle, B.D., Lamoureux, S.F., Cockburn, J.M.H, Spooner, I., 2002. A new method 166–213.
for visualizing sediment particle size distributions. Journal of Paleolimnology. Folk, R.L., 1954. The distinction between grain size and mineral composition in
27, 279–283. sedimentary rock nomenclature. Journal of Geology 62(4), 344–359.
Beuselinck, L., Govers, G., Steegen, A., Hairsine, P.B., Poesen, J., 1999a. Evaluation Folk, R.L., 1955. Student operator error in determination of roundness, sphericity
of the simple settling theory for predicting sediment deposition by overland and grain size. Journal of Sedimentary Petrology 25, 297–301.
flow. Earth Surface Processes and Landforms 24(11), 993–1007. Folk, R.L., 1966. A review of grain-size parameters. Sedimentology 6, 73–93.
Beuselinck, L., Govers, G., Poesen, J., 1999b. Assessment of micro-aggregation Folk, R.L., Andrews, P.B., Lewis, D.W., 1970. Detrital sedimentary rock classification
using laser diffractometry. Earth Surface Processes and Landforms 24(1), 41–49. and nomenclature for use in New Zealand. New Zealand Journal of Geology and
Beuselinck, L., Govers, G., Poesen, J., Degraer, G., Froyen, L., 1998. Grain-size Geophysics 13, 937–968.
analysis by laser diffractometry: Comparison with the sieve-pipette method. Folk, R.L., Robles, R., 1964. Carbonate sands of Isla Perez, Alacran Reef complex,
Catena 32(3–4), 193–208. Yucatan. The Journal of Geology 72, 255–292.
Blair, T.C., McPherson, J.G., 1999. Grain-size and textural classification of coarse Folk, R.L., Ward, W.E., 1957. Brazos River bar: a study in the significance of grain
sedimentary particles. Journal of Sedimentary Research 691, 6–19. size parameters. Journal of Sedimentary Petrology 27, 3–26.
Blott, S.J., Pye, K., 2001. Gradistat: a grain size distribution and statistics package Folk, R.L., 1980. The Petrology of Sedimentary Rocks. Hemphill Publishing
for the analysis of unconsolidated sediments. Earth Surface Processes and Company, Austin, Texas, 184 pp.
Landforms 26, 1237–1248. Friedman, G .M., 1967. Processes and statistical parameters compared for size
Blott, S.J., Pye, K., 2006. Particle size distribution analysis of sand-sized particles frequency distribution of beach and river sands. Journal of Sedimentary
by laser diffraction: an experimental investigation of instrument sensitivity and Petrology 37, 327–354.
the effects of particle shape. Sedimentology 53, 671–685. Friedman, G.M., 1961. Distinction between dune, beach and river sand from their
Bowler, J.M., 1973. Clay dunes: their occurrence, formation and environmental textural characteristics. Journal of Sedimentary Petrology 31, 514–529.
significance. Earth Science Reviews 9, 315–338. Friedman, G.M., 1979. Differences in size distributions of populations of particles
Briggs, D., 1977. Sediments. In: Sources and Methods in Geography. Butterworth- among sands of various origins. Sedimentology 26, 3–32.
Heinemann, London, UK, 128 pp. Friedman, G.M., Sanders, J.E., 1978. Principles of Sedimentology. Wiley, New York.
Buller, A.T., McManus, J., 1972. Simple metric sedimentary statistics used to USA.
recognize different environments. Sedimentology 18, 1–21. Gale, S.J., Hoare, P.G., 1991. Quaternary Sediments. Wiley, Chichester and New
Buller, A.T., McManus, J., 1979. Sediment sampling and analysis. In: Dyer, K.R. York, USA, 323 pp.
(Ed.), Estuarine Hydrography and Sedimentation. Cambridge University Press, Gibbs, R.J., Mathews, M.D., Link, D.A., 1971. The relationship between sphere size
UK, pp. 87–130. and settling velocity. Journal of Sedimentary Petrology 41, 7–18.
Buscombe, D., Masselink, G., 2006. Concepts in gravel beach dynamics. Earth Goldberg, R., Richardson, D., 1989. The influence of bulk shape factors on
Science Reviews 79(1-2), 33–52. settling velocities of natural sand-sized sedimentary suites. Sedimentology
Butler, B.E., 1956. Parna: an aeolian clay. Australian Journal of Science 18, 145–151. 36, 125.
Channon, R.D., 1971. The Bristol fall column for coarse sediment grading. Journal Griffiths, J.C., 1967. Scientific Method in Analysis of Sediments. McGraw-Hill,
of Sedimentary Petrology 41, 867–870. New York, USA, 508 pp.
240 Measuring and Analyzing Particle Size in a Geomorphic Context

Hayton, S., Nelson, C.S., Ricketts, B.D., Cooke, S., Wedd, M.W., 2001. Effect of Lewis, D.W., McConchie, D., 1994b. Analytical Sedimentology. Chapman and Hall,
mica on particle-size analyses using the laser diffraction technique. Journal of London, UK, 197 pp.
Sedimentary Research 71(3), 507–509. Loizeau, J.L., Stanley, D.J., 1994. Bottom sediment patterns evolving in polluted
Hesse, P.P., 2003. The bigness question. Quaternary Australasia 211, 6–8. Mariut lake, Nile delta, Egypt. Journal of Coastal Research 102, 416–439.
Hjulstrom, F., 1939. Transportation of detritus in moving water. In: Trask, P.D. (Ed.), Loizeau, J.L., Stanley, D.J., 1993. Petrological-statistical approach to interpret recent
Recent Marine Sediments. American Association of Petrology and Geology, and subrecent lagoon subfacies, Idku, Nile delta of Egypt. Marine Geology
Tulsa, USA, pp. 5–31. 111(1–2), 55–81.
Huang, T.C., Goodell, H.G., 1967. Sediments of Charlotte harbor, Southwestern Loizeau, J., Arbouille, D., Santiago, S., Vernet, J., 1994. Evaluation of a wide range
Florida. Journal of Sedimentary Petrology 37, 449–474. laser diffraction grain size analyzer for use with sediments. Sedimentology 412,
Inmam, D.L., 1952. Measures for describing the size distribution of sediments. 353–361.
Journal of Sedimentary Petrology 22, 125–145. Loring, D.H., Rantala, R.T.T., 1992. Manual for the geochemical analyses of marine
International Standard Organization, 2009. Soil quality – Determination of particle sediments and suspended particulate matter. Earth Science Reviews 324, 235–283.
size distribution in mineral soil material – Method by sieving and Loveland, P.J., Whalley, W.R., 2001. Particle size analysis. In: Smith, K.A., Mullins,
sedimentation. ISO 11277:2009. International Standard Organization, Geneva, C.E. (Eds.), Soil and Environmental Analysis, Physical Methods. Marcel Dekker
(Reproduced by the British Standards Institution, Milton Keynes, 34 pp). Inc., New York, pp. 271–281, 637 pp.
International Standard Organization, 1999a. Particle Size Analysis – Laser Diffraction Ly, C.K., 1978. Late Quaternary deposits of the Newcastle-Port Stephens area as
Method. Part 1. General Principles. ISO 13320-1. International Standard revealed by grain size analysis and scanning electron microscopy. Journal and
Organization, Geneva, (Reproduced by the British Standards Institution, Milton Proceedings of the Royal Society of New South Wales 111, 77–88.
Keynes, 34 pp). Mason, T., Coates, TT., 2001. Sediment transport processes on mixed beaches: a
International Standard Organization, 1999b. Particle Size Analysis – Laser review for shoreline management. Journal of Coastal Research 17, 645–657.
Diffraction Method. Part 2. Validation of Inversion Procedures. ISO 13320-2. Mason, C.C., Folk, R.L, 1958. Differentiation of beach, dune, and aeolian flat
International Standard Organization, Geneva, (Reproduced by the British environments by size analysis, Mustang Island, Texas. Journal of Sedimentary
Standards Institution, Milton Keynes). Petrology 28, 211–226.
International Standard Organization 2008 ISO 9276-6: Representation of results of McCave, I.N., Hall, I.R., Bianchi, G.G., 2006. Laser versus settling velocity
particle size analysis – Part 6: Descriptive and quantitative representation of differences in silt grainsize measurements: estimation of palaeocurrent vigour.
particle shape and morphology. Sedimentology 53, 919–928.
Jennings, R., Shulmeister, J., 2002. A field based classification scheme for gravel McCave, I.N., Syvitski, J.P.M., 1991. Principles and methods of geological particle
beaches. Marine Geology 186(3–4), 211–228. size analysis. In: Syvitski, J.P.M. (Ed.), Principles, Methods and Applications of
Kench, P.S., 1997. Contemporary sedimentation in the Cocos (Keeling) Islands, Particle Size Analysis. Cambridge University Press, Cambridge, UK, pp. 3–21.
Indian Ocean: interpretation using settling velocity analysis. Sedimentary McCave, I.N., Bryant, R.J., Cook, H.F., Coughanowr, C.A., 1986. Evaluation of a
Geology 114, 109–130. laser-diffraction-size analyzer for use with natural sediments. Journal of
Kench, P.S., McLean, R.F., 1997. A comparison of settling and sieve techniques Sedimentary Reearch 56, 561–564.
for the analysis of bioclastic sediments. Sedimentary Geology 109(1–2), McLaren, P.A., 1981. Interpretation of trends in grain-size measurements. Journal of
111–119. Sedimentary Petrology 51, 611–624.
Kellerhals, R, Bray, DI., 1971. Sampling procedures for coarse fluvial sediments. McLaren, P.A., Bowles, D., 1985. The effects of sediment transport on grain-size
American Society Civil Engineers Journal of the Hydraulics Division 97, distributions. Journal of Sedimentary Petrology 55(4), 457–470.
1165–1179. McManus, J., 1988. Grain size determination and interpretation. In: Tucker, M (Ed.),
Klingeman, PC, Emmett, WW., 1982. Gravel bedload transport processes. In: Hey, Techniques in Sedimentology. Blackwell, Oxford, UK, pp. 63–85.
RD, Bathurst, JC, Thorne, CR (Eds.), Gravel Bed Rivers. Wiley, Chichester, pp. McManus, D.A., 1965. A study of maximum load for small diameter sieves. Journal
141–179. of Sedimentary Petrology 35, 792–796.
Komar, P.D., Cui, B., 1984. The analysis of grain-size measurements by McManus, D.A., 1963. A criticism of certain usage of the phi notation. Journal of
sieving and settling-tube techniques. Journal of Sedimentary Petrology 54, Sedimentary Petrology 33, 670–674.
603–614. Miall, A.D., 2006. The Geology of Fluvial Deposits: Sedimentary Facies, Basin
Kondolf, G.M., Lisle, T.E., Wolman., G.M., 2003. Bed sediment measurement. Analysis and Petroleum Geology. Springer-Verlag, Berlin, Germany.
In: Kondolf, G.M., Piégay, H. (Eds.), Tools in fluvial geomorphology. Wiley, Miller, M.C., McCave, I.N., Komar, P.D., 1977. Thresholds of sediment motion
New York USA, Chapter 13. under unidirectional currents. Sedimentology 24, 507–527.
Kondolf, G.M., 1997. Application of the pebble count: Notes on purpose, method, Miller, I.M., Warrick, J.A., Morgan, C., 2011. Observations of coarse sediment
and variants. Journal of the American Water Resources Association 33(1), movements on the mixed beach of the Elwha Delta, Washington. Marine
79–87. Geology, 201–214.
Konert, M., Vandenberghe, J., 1997. Comparison of laser grain size analysis with Milligan, T.G., Krank, K., 1991. Electro-resistance particle size analyzers. In: Syviski,
pipette and sieve analysis: a solution for the underestimation of the clay fraction. J.P.M. (Ed.), Principles, Methods and Applications of Particle Size Analysis.
Sedimentology 44(3), 523–535. Cambridge University Press, New York, pp. 109–118.
Krumbein, W.C., 1934. Size frequency distributions of sediments. Journal of Milner, H.B., 1962a. Sedimentary Petrography, Methods in Sedimentary Petrography
Sedimentary Petrology 10, 57–69. Vol. I:. Allen and Unwin, London, UK.
Krumbein, W.C., 1938. Size frequency distributions of sediments and the normal Milner, H.B., 1962b. Sedimentary Petrography, Vol. II: Principles and Applications.
phi curve. Journal of Sedimentary Petrology 8, 84–90. Allen and Unwin, London, UK.
Krumbein, W.C., Pettijohn, F.L., 1961. Manual of Sedimentary Petrography. Moilola, R.L., Weiser, D., 1968. Textural parameters: an evaluation. Journal of
Appleton-Century-Crofts, New York, USA. Sedimentary Petrology 38, 45–53.
Krumbein, W.C., 1964. Some remarks on the phi notation. Journal of Sedimentary Moss, A.J., 1962. The physical nature of common sandy and pebbly deposits. Part
Petrology 34, 195–196. I. American Journal of Science 260, 337–373.
Kuenan, P.H., 1968. Convection and grain size analysis. Journal of Sedimentary Moss, A.J., 1963. The physical nature of common sandy and pebbly deposits. Part
Petrology 38, 817–831. II. American Journal of Science 261, 297–343.
Lario, J., Zazo, C., Plater, A.J., et al., 2001. Particle size and magnetic properties of Moss, A.J., 1972. Bed-load sediments. Sedimentology 18, 159–219.
holocene estuarine deposits from the doñana national park SW iberia: evidence Mwakumanya, M.A., Bdo, O., 2007. Beach morphological dynamics: A case study
of gradual and abrupt coastal sedimentation. Zeitschrift fur Geomorphologie of Nyali and Bamburi beaches in Mombasa, Kenya. Journal of Coastal Research
45(1), 33–54. 23(2), 374–379.
Lario, J., Spencer, C., Plater, A.J., Zazo, C., Goy, J.L., Dabrio, C.J., 2002. Particle Nanson, G.C., Rust, B.R., Taylor, G., 1986. Coexistent mud braids and
size characterisation of Holocene back-barrier sequences from North Atlantic anastomosing channels in an arid zone river: Cooper Creek, Central Australia.
coasts SW Spain and SE England. Geomorphology 42, 25–42. Geology 14, 175–178.
Leopold, L.B., Wolman, M.G., Miller, P., 1964. Fluvial Processes in Geomorphology. Nanson, G.C., Young, R.W., Price, D.M., Rust, B.R., 1988. Stratigraphy,
W.H. Freeman and Company, 522 pp. sedimentology and Late Quaternary chronology of the Channel Country of
Lewis, D.W., McConchie, D., 1994a. Practical Sedimentology, Second ed. Chapman Western Queensland. In: Warner, R.F. (Ed.), Essays in Australian Fluvial
and Hall, London, UK, 213 pp. Geomorphology. Academic Press, Sydney, pp. 151–175.
 Measuring and Analyzing Particle Size in a Geomorphic Context 241

Nichol, S.L., Lian, O.B., Horrocks, M., Goff, J.R., 2007. Holocene record of gradual, embayment from southeastern New South Wales, Australia. Journal of Coastal
catastrophic, and human-influenced sedimentation from a backbarrier wetland, Research 21(4), 655–663.
northern New Zealand. Journal of Coastal Research 23, 605–617. Syvitski, J.P.M. (Ed.), 1991a. Principles, Methods and Application of Particle Size
Peirce, T.J., Williams, D.J .A., 1966. Experiments on certain aspects of Analysis. Cambridge University Press, New York, p. 368.
sedimentation of estuarine muds. Proceedings of Institute of Civil Engineers 54, Syvitski, J.P.M., 1991b. Factor analysis of size frequency distributions: significance
391–402. of factor solutions based on simulation experiments. In: Syvitski, J.P. (Ed.),
Pettijohn, F.J., 1975. Sedimentary Rocks, Third edition Harper and Row, London, Principles, Methods and Application of Particle Size Analysis. Cambridge
UK. University Press, NY, pp. 249–263.
Pilarczyk, J.E., Reinhardt, E.G., Boyce, J.I., Schwarz, H.P., Donato, S.V., 2011. Syvitski, J.P.M., Asprey, K.W., Clattenburg, D.A., 1991a. Principles, design and
Assessing surficial foraminiferal distributions as an overwash indicator in Sur calibration of settling tubes. In: Syvitski, J.P.M. (Ed.), Principles, Methods and
Lagoon, Sultanate of Oman. Marine Micropaleontology 80(3–4), 62–73. Application of Particle Size Analysis. Cambridge University Press, New York, pp.
Poppe, L.J., Eliason, A.H., Fredericks, J.J., Rendigs, R.R., Blackwood, D., Polloni, 45-63.
C.F., 2000. Grain-size analysis of marine sediments: methodology and data Syvitski, J.P.M., LeBlanc, K.W.G., Asprey, K.W., 1991b. Inter-laboratory, inter-
processing. In: Poppe, L.J., Polloni, C.F. (Eds.), USGS East-coast Sediment instrument calibration experiment. In: Syvitski, J.P.M. (Ed.), Principles, Methods
Analysis: Procedures, Database, and Georeferenced Displays. U.S. Geological and Application of Particle Size Analysis. Cambridge University Press, New York,
Survey, Open-File Report 00-358, CD-ROM. pp. 174–193.
Pye, K., Blott, S.J., 2004. Particle size analysis of sediments, soils and related Tanner, W.F., 1969. The particle size scale. 1. Sedimentary Petrology 39, 509–512.
particulate materials for forensic purposes using laser granulometry. Forensic Tanner, WF., 1991a. Suite statistics: the hydrodynamic evolution of the
Science International 144(2004), 19–27. sediment pool. In: Syvitski, JPM (Ed.), Principles, Methods and Applications
Reineck, H.E., Singh, LB., 1975. Depositional Sedimentary Environments. Springer- of Particle Size Analysis. Cambridge University Press, Cambridge,
Verlag, Berlin, Germany. pp. 225–236.
Richardson, J.F., Zaki, W.N., 1954. Sedimentation and fluidisation. Transactions of Tanner, WF., 1991b. Application of suite statistics to stratigraphy and sea-level
Institute Chemical Engineers 32, 35–51. changes. In: Syvitski, JPM (Ed.), Principles, Methods and Applications of
Rubin, D.M., 2004. A simple autocorrelation algorithm for determining grain Particle Size Analysis. Cambridge University Press, Cambridge, pp. 283–292.
size from digital images of sediment. Journal of Sedimentary Research 74, Tucker, M., 1990. Techniques in sedimentology. Wiley - Blackwell Publications,
160–165. 408 pp.
Rust, B.R., Nanson, G.C., 1989. Bedload transport of mud as pedogenic aggregates Udden, J.A., 1914. Mechanical composition of clastic sediments. Geological Society
in modern and ancient rivers. Sedimentology 36, 291–306. of America Bulletin 25, 655–744.
Sanford, R.B., Swift, D.J.P., 1971. Comparison of sieving and settling techniques for Van Hengstum, P.J., Reinhardt, E.G., Boyce, J.I., Clark, C., 2007. Changing
size analysis, using a Bepthos rapid sediment analyser. Sedimentology 7, sedimentation patterns due to historical land-use change in Frenchman’s Bay,
257–264. Pickering, Canada: Evidence from high-resolution textural analysis. Journal of
Schlee, J., 1966. A modified Woods Hole rapid sediment analyser. Journal of Paleolimnology 37(4), 603–618.
Sedimentary Petrology 36, 404–413. van Hengstum, P.J., Scott, D.B., Gröcke, D.R., Charette, M.A., 2011. Sea level
Schlee, J., 1973. Atlantic continental shelf and slope of the United States – controls sedimentation and environments in coastal caves and sinkholes. Marine
sediment texture of the northeastern part. US Geological Survey Professional Geology 286(1–4), 35–50.
Paper 529-L, 64 pp. Wang, Y., Ke, X., 1997. Grain-size characteristics of the extant tidal flat
Schumm, S.A., 1977. The Fluvial System. John Wiley and Sons, New York, 338 pp. sediments along the Jiangsu coast, China. Sedimentary Geology 112(1-2),
Sengupta, S., Veenstra, H.J., 1968. On sieving and settling techniques for sand 105–122.
analysis. Sedimentology 11, 83–98. Ward, P.R.B., 1980. Laboratory measurement of sediment by turbidity. Journal of
Shepard, F.P., 1954. Nomenclature based on sand-silt-clay ratios. Journal Hydraulic Division – American Society of Civil Engineers 106(HY6),
Sedimentary Petrology v. 24, 151–158. 1041–1053.
Shillabeer, N., Hart, B., Riddle, A.M., 1992. The use of a mathematical model to Warrick, J.A., George, D.A., Stevens, A.W., et al., 2007, revised 2010. Beach
compare particle size data derived by dry-sieving and laser analysis. Estuarine, morphology monitoring in the Elwha River littoral cell, 2004–2009: U.S.
Coastal and Shelf Science 35(1), 105–111. Geological Survey, Data Series 288, v. 1.1., 59 pp, and data folder [http://
Socci, A., Tanner, W.F., 1980. Little known but important papers on grain-size pubs.usgs.gov/ds/288/].
analysis. Sedimentology 1980(27), 231–232. Warrick, J.A., Rubin, D.M., Ruggiero, P., Harney, J.N., Draut, A.E., Buscombe., D.,
Stewart, Jr. H.B., 1958. Sedimentary reflections on depositional environments in 2009. Cobble cam: grain-size measurements of sand to boulder from digital
San Miguel Lagoon, Baja, California, Mexico. Bulletin of the American photographs and autocorrelation analyses. Earth Surface Processes and
Association of Petroleum Geologists 42, 2567–2618. Landforms 34(13), 1811–1821.
Sutherland, R.A., Lee, C.T., 1994a. Application of the log-hyperbolic distribution to Webb P.A., 2008. Interpretation of Particle Size Reported by Different Analytical
hawaiian beach sands. Journal of Coastal Research 102, 251–262. Techniques Micromeritics Instrument Corp. 10 pp.
Sutherland, R.A., Lee, C.T., 1994b. Discrimination between coastal subenvironments Wentworth, C.K., 1922. A scale of grade and class terms for clastic sediments.
using textural characteristics. Sedimentology 416, 1133–1145. Journal of Geology 30, 377–392.
Switzer, A.D., Burston, J.M., 2010. Competing mechanisms for boulder Wolman, M.G., 1954. A method of sampling coarse river-bed material. Transactions
deposition on the southeast Australian coast. Geomorphology 114(1-2), 42–54. of the American Geophysical Union EOS 35, 951–956.
Switzer, A.D., Jones, B.G., 2008. Large-scale washover sedimentation in a Zeigler, J.M., Gill, A., 1959. Tables and graphs for the settling velocity of quartz in
freshwater lagoon from the southeast Australian coast: sea-level change, tsunami water, above the range of Stoke’s Law. Woods Hole Oceanographic Institution,
or exceptionally large storm? Holocene 18(5), 787–803. Refs No. 59-36.
Switzer, A.D., Pucillo, K., Haredy, R.A., Jones, B.G., Bryant, E.A., 2005. Sea level, Zeigler, J.M., Whitney, G.G., Hayes, C.R., 1960. Woods hole rapid sediment
storm, or tsunami: Enigmatic sand sheet deposits in a sheltered coastal analyser. Journal of Sedimentary Petrology 30, 490–495.
242 Measuring and Analyzing Particle Size in a Geomorphic Context

Biographical Sketch

Adam Switzer is an NRF Singapore Fellow who holds concurrent positions as Nanyang Assistant Professor at NTU
and Principal Investigator at the Earth Observatory of Singapore (EOS). He is primarily a sedimentologist with
interests in coastal geomorphology, palaeoenvironments, and natural hazards. The overarching driving force of
his research is a desire to use geomorphological and sedimentological techniques to solve contemporary prob-
lems at local, regional, and international scales. His main research interest lies in using coastal stratigraphy to
define the recurrence interval of catastrophic marine inundation events (tsunami or large storms) and investigate
the relationship between tectonics and climate change in coastal systems. After BSc and PhD degrees from the
University of Wollongong, Australia he accepted an Endeavour Australia Cheung Kong fellowship to study at The
University of Hong Kong where he held positions as Post-Doctoral Fellow and Centenary Research Assistant
Professor before joining Nanyang Technological University in 2009. An internationally recognized expert in
tsunami and storm deposits his most significant contributions to the field include the first study of modern storm
deposits from the Australian southeast coast; the recognition that immature heavy mineral suites in coastal
sandsheets may indicate tsunami deposition rather than storm deposition in coastal settings; the recognition of an
erosional signature of large scale washover of coastal dunes using ground penetrating radar; and initial evaluation
of the sedimentary processes associated with the 2004 Indian Ocean tsunami.
 14.20 Examining Particle Shape
JE Bullard, Loughborough University, Loughborough, UK
r 2013 Elsevier Inc. All rights reserved.

14.20.1 Introduction 243

14.20.2 Background 247
14.20.3 Methodology 247
14.20.4 Limitations 254
14.20.5 Conclusions 255
References 255

Abstract

Particle shape can be described in a variety of ways but is commonly assessed using the interrelated variables of sphericity,
roundness, and surface texture. A number of different techniques has been developed to assess particle shape and these
increasingly make use of digital imaging and computer analysis. Although characterizing particle shape is relatively
straightforward, many different factors contribute to determining shape and this can make the geomorphological inter-
pretation of shape difficult.

14.20.1 Introduction For larger particles, shape is generally a description of

the dimensional characteristics of a particle defined using
The characterization of particle shape is a very common some ratio of its axis lengths. For example, the ratio of the
and relatively straightforward geomorphological technique, intermediate (I- or B-axis) and longest (L- or A-axis) axes in-
although as will be discussed, ascribing geological and geo- dicates the degree of grain elongation; the ratio of the shortest
morphological meaning to shape raises some challenges. (S- or C-axis) and intermediate axes indicates the degree of
Particle shape is of interest because it can reflect the transport
and environmental history of a grain or clast, it also affects
the operation of geomorphological processes, for example, Form
it typically requires more energy to entrain flat grains than
spherical grains (e.g., Landon, 1930; Mazzullo et al., 1986;
Graf, 1988). Shape also affects the porosity and compress-
Roundness
ibility of sedimentary deposits (e.g., Gray, 1968; Masselink
and Hughes, 2003).
Debates surrounding the characterization and subsequent
interpretation of particle shape abound in the literature. The
concept of particle shape depends in part on the scale of ob-
servation, but typically comprises three elements: ‘shape,’ or
overall external expression in two or three dimensions of the
individual particle (sometimes referred to as sphericity, e.g.,
Barrett, 1980); ‘roundness’ (or angularity), which describes
shape at the scale of the edges of a grain; and ‘surface texture’
(Figure 1). Commonly, the combined expression of all the
three elements is described as the ‘form’ of the particle. In
most instances, form is primarily determined by shape, then
by roundness, and finally by surface texture (cf. Orford, 1991); (a) Roughness
however, these three elements are interrelated. ‘Form’ and
‘shape’ (and rarely ‘morphology’) can be used interchangeably I L L
I
in the literature – some argue that shape is a subset of form
(e.g., Whalley, 1972; Orford, 1991), whereas others argue that S
the form is a subset of shape (e.g., Sneed and Folk, 1958; Blott S
and Pye, 2008). (b)

Figure 1 (a) Different components of a particle’s shape: form,

Bullard, J.E., 2013. Examining particle shape. In: Shroder, J. (Editor in roundness, and roughness (surface texture). (b) Examples of long
Chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise on Geomorphology. (L), intermediate (I), and short (S) particle axes. Reproduced from
Academic Press, San Diego, CA, vol. 14, Methods in Geomorphology, Barrett, P.J., 1980. The shape of rock particles, a critical review.
pp. 243–256. Sedimentology 27(3), 291–303, with permission from Wiley.

Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00386-9 243

244 Examining Particle Shape

grain flattening. The use of ratios makes the shape value in- Crowden Grinds Lady Jaggers Peakshole
dependent of particle size. Such ratios also give an indication brook brook booth clough water
of sphericity – the degree to which a particle approximates to a brook
true sphere. Sphericity is important in controlling entrain-
ment, sediment transport, and deposition (settling velocity)
and is closely related to the surface area to volume ratio of 300
the particle (Komar and Reimers, 1979; Wilson and Huang,

Mean roundness
1979; Cui et al., 1983); the sphere is also a useful reference
and many formulae used by geomorphologists initially as-
sume that particles are spherical (e.g., Stoke’s law and Shield’s 200
Curve) (Tucker, 2001; Gordon et al., 2004).
Roundness is conceptually and geometrically distinct from
sphericity and is a measure of the relative roundness or an-
gularity of a particle at the scale of the edge or corners of a 100
grain (Figure 1). For most particles, roundness is affected by
(a)
weathering and transport processes to a greater extent than is
sphericity, which is characteristically more strongly influenced
by the original shape of the particle (Orford, 1991; Tucker,
2001). Chemical weathering and mechanical abrasion increase 300

Mean roundness
particle roundness such that in some environments there is a
clear gradient of increasing roundness with distance from the
sediment source (e.g., Knighton, 1982) (Figure 2).
The surface texture of a particle refers to its small-scale 200
surface characteristics and these can arise from interaction
with other particles through impacts, abrasion, and attrition,
from weathering, or they may be a property of the particle
lithology. On larger particles, surface texture may be visible
100
with the naked eye, examples include striations on glacial
boulders, pits and flutes on ventifacts, and vesicles in basalts. (b)

On small particles, such as sand grains, surface texture is only

visible using a microscope (Figure 3).
Mean roundness

Although attempts have been made to differentiate these

three aspects of particle form (e.g., Wadell, 1932; Whalley, 200
1972), they are typically interdependent. Particle shape can
also be strongly related to particle size (Figure 4). Different
size fractions within a sediment population may have con-
trasting shape characteristics and it is worth noting that the 100
shape of a particle can influence the results of size analysis
depending on the methods of size analysis that are chosen. For 0 5 10 15 20
sand-sized particles, laser diffraction particle-size analysis (c) Distance downstream (km)
may cause overestimation of the proportion of particles at Figure 2 Increasing roundness with distance downstream of particles
the extremes of the distribution (i.e., relatively fine or coarse) (mean grain size 20–100 mm) in the River Noe, Derbyshire, England.
compared with that indicated by the sieving and sieve–pipette (a) Total sample, (b) sandstone group, and (c) shale group. Although
methods and this has largely been attributed to the influence both lithologies show increasing roundness, the change downstream is
of particle shape on the measurement method (Beuselinck more rapid for sandstone. Discontinuities in roundness are associated
et al., 1998; Blott and Pye, 2006; Rodriguez and Uriarte, 2009; with sediment input from tributaries from the locations, which are
marked with labeled arrows. Roundness is determined using the index
Di Stefano et al., 2010). Similarly, the fact that particle shape
of Cailleux (1947). Reproduced from Knighton, A.D., 1982.
has an influence on settling velocity means that particle Longitudinal changes in the size and shape of stream bed material:
size distribution analyses undertaken using settling tubes evidence of variable transport conditions. Catena 9, 25–34.
will differ from sieve measurements (Kench and McLean,
1997). A comparison of sieve analysis and image-based an-
alysis of aggregated sediments in the range 32–64 mm suggests
that image-based analysis gives the more accurate measure of addition to descriptions of size, sorting, color, and mineralogy.
particle axis in long and intermediate dimensions (Fernlund, Given that shape can be affected by processes that have been
1998). operating on the particles during the formation of a landform,
Prior to selecting which aspect(s) of particle shape to char- grain-shape analyses can be useful for interpreting sediments
acterize and what methods to use, it is important to have a clear and landforms of uncertain origin. For example, Boulton
understanding of the relevance of particle shape to a study. It is (1978) identified fundamental differences in clast shape among
commonly useful to characterize sediments for facies analysis passively transported angular supraglacial debris, englacial
and environmental reconstruction, and shape can be a useful debris (subrounded to rounded with cross-cutting striae
 Examining Particle Shape 245

(a) (b)

(c)

Figure 3 Surface texture (roughness) at different scales. (a) SEM image of a sand grain from the Simpson Desert, Australia; (b) Wind-eroded
boulder (ventifact) in the Mojave Desert, CA, USA, showing the development of flutes and helical forms on the rock surface; and (c) Glacial
striations on the surface of a small boulder in southeast Iceland.

4.0
Mean roundness of very fine sand (3−4 )

3.5

3.0

UAE
2.5
Kuwait
Simpson
USA
2.0
Thar
Kalahari
Namib
1.5
Tunisia
Bahrain

1.0
2.0 2.5 3.0 3.5 4.0 4.5
Mean roundness of fine sand (2−3 )
Figure 4 The relationship between roundness of very fine sand (3–4 f) and fine sand (2–3 f) for a selection of deserts indicating the variability
in roundness between size fractions and among regions. Data for Kuwait from Khalaf and Gharib (1985), data for UAE from El-Sayed (1999),
data for Simpson Desert author’s own (unpublished), all other data from Goudie and Watson (1981).
246 Examining Particle Shape

characterizing surface roughness), and debris which moves younger deposits. Rounding caused by abrasion can produce
beneath a glacier in the subglacial transport zone that is distinct spatial patterns, reflected, for example, in gradients of
rounded and streamlined. Assuming that these relative differ- increasing roundness with distance alongshore and down-
ences are common to most glacial environments, they can be stream (e.g., Knighton, 1982; Figure 2) and residence time in
used to determine the transport pathways and processes to littoral sediment cells (Clemens and Komar, 1988; Figure 5).
which glacial sedimentary deposits have been subjected. These spatial and temporal gradients are not ubiquitous, they
The shape of a grain may affect how it is entrained, can be disrupted by sediment inputs (e.g., from tributaries)
transported, and deposited. One influence on entrainment is or may not be discernible in particular size fractions, for ex-
the angle through which a grain needs to be rotated to change ample, experimental studies by Kuenen (1959) suggested that
position rather than rock back into its original position. This is sand-sized sediment is not significantly abraded by fluvial
the friction angle and depends on grain size, shape, packing, transport.
and surface texture but the influence of these factors can be If particle rounding increases through time, it may provide an
hard to elucidate. For example, an elongated particle wedged indication of maturity of sediments. For example, Folk (1951)
upright would have a smaller friction angle than a spherical suggested texturally immature sandstone sediments (fluvial and
particle (Li and Komar, 1986) but may be better packed with glacial deposits) are poorly sorted with angular grains; mature
more points of contact and so harder to entrain. For wind-
blown sediments, Rice (1991) found that less spherical par-
ticles were more likely to be entrained than more spherical
particles, but once entrained, the latter attains greater heights
above the surface and so may be transported further.
As particle shape affects how geomorphic processes River
and sediments interact, particles are commonly deposited
with particular packing arrangements or alignment to flow Co
lu m
and these therefore have the potential to yield information bi a R i ve r
about palaeoflow. In water-lain deposits, prolate (elongated)
50
clasts can be oriented both normal and parallel to the dir-
ection of flow. For example, elongated particles form imbri- Beach
%
cated deposits with the long axis aligned normal to the flow
direction under low energy conditions but parallel to the 0
VA A SA SR R WR
direction of flow under high-energy, turbulent conditions
(Leeder, 1982); imbrication may be absent in systems com-
prising spherical grains. In glacial sediments, where particles
show a preferred orientation, they are commonly oriented Beach
parallel to the direction of ice flow.
The spatial variation in grain shape on beaches also reflects
the relative ease with which different-shaped particles are
transported (e.g., Bluck, 1967). Prolate grains, which com- Tillamook
monly occur at the base of a gravel beach, are easily trans- head
ported both up and down a beach, whereas oblate (flattened)
clasts are harder to transport and can be carried to the top of
the beach by the relatively strong wave uprush but cannot be
Beach
removed by the weaker backwash (Bluck, 1967). Spherical
Augite
pebbles are easily transported on and off the beach and tend rounding
to occur further down the beach or offshore. In wave tank and Beach
field experiments, Trenhaile et al. (1996) found that within
5 km
the swash zone on sandy beaches more rounded grains occur
on the upper beach compared with more angular grains. This
is probably because rounded grains are more easily entrained Cape Falcon
and also have a higher settling velocity than the angular
particles.
Roundness is affected by weathering and transport, and in Figure 5 Variation in sand grain (augite) roundness on the north
some circumstances, the relative roundness of particles can be coast of Oregon, USA. Sand south of Tillamook Head is more
used to estimate rates of abrasion and the relative age and rounded than that to the north reflecting exposure to abrasion over
maturity of the sediments. Sediments typically become more thousands of years compared with the more recent, angular, inputs
to the coast from the Columbia River. Roundness based on Powers
rounded with time, although the rate of rounding decreases as
classification (see Figure 10), where VA¼very angular, A ¼angular,
particles become more rounded. Using carefully defined cri- SA¼subangular, SR ¼subrounded, R¼rounded, WR¼well rounded.
teria, Kirkbride and Bell (2010) were able to classify moraines Reproduced from Clemens, K.E., Komar, P.D., 1988. Tracers of sand
of different ages using the relative edge roundness of the movement on the Oregon Costs. Proceedings of the 21st Coastal
constituent boulders; older deposits, exposed to longer peri- Engineering Conference. American Society of Civil Engineers,
ods of weathering, were significantly more rounded than 1338–1351.
 Examining Particle Shape 247

sediments are moderate-to-well sorted with subrounded-to- 1.00

Disc-shaped, platy or oblate Compact or equant
rounded grains and supermature sediments are very well sorted spheroid
with well-rounded grains (desert and beach environments).
However, although this can be a useful indicator, this classifi-
cation is not necessarily reflected in landscapes; most desert ae-
olian sands are subangular to subrounded (Goudie and Watson, 0.67
Bladed or triaxial Rod-like, elongate or
1981) and the Namib ;Sand Sea, for example, is classed as of
prolate
intermediate maturity based on roundness and mineralogy

I/L (or B/A)

(Lancaster, 1989; Muhs, 2004).

Planar
14.20.2 Background

Interest in the significance of, and factors that determine,

particle shape have a long history in sedimentology and geo-
morphology (e.g., Mackie, 1897). Early research recognized 0.10
that grain size and shape are essentially the characteristics that Acicular

distinguish the ways in which different sediments behave and 0 0.10 0.67 1.00
also reflect the processes to which the particles have been S/I (or C/B)
subjected. A variety of approaches to characterizing particle Figure 6 Classification of particle sphericity according to Zingg, T.,
shape have been adopted and although there has been 1935. Beı́träge zur Schotteranalyse: Schweizerische mineralogische
evolution from subjective description to quantification by und Petrologische Mitteilungen 15, 39–140 showing idealized particle
a variety of methods, many of the early methods are still used shapes. L (A), I (B), and S (C) are the long, intermediate, and short
– sometimes in isolation but increasingly in combination with axes measurements, respectively.
newer techniques. It is not possible to do justice here to the
numerous researchers who have developed methods for
particle’s shape. This approach has been successfully used in
assessing particle shape; readers are directed to useful reviews
several studies to discriminate particles of different origin,
by Barrett (1980), Diepenbroek et al. (1992), Blott and Pye
particularly where particles have highly irregular shapes. For
(2008), and Clayton et al. (2009) among others and only a
example, Carey et al. (2000) used fractal analysis of the shape
brief discussion is presented here.
of basaltic grains incorporated into jökulhlaup sediments to
Zingg (1935) provided an early method for distinguishing
differentiate primary and reworked deposits. Technical ad-
particle shape based on measurements of three axes (L, I, and
vances such as the development of terrestrial laser scanners,
S; Figure 1) and plotting I/L against S/I is a simple but useful
which can be used in the field to collect high-resolution
way of distinguishing among bladed, equant, and prolate
topographic data from a range of different surfaces, also
particles. This method has been widely used and adopted but
have the potential to provide information about particle
has undergone some modification, for example, the addition
shape. Laser scanners capable of capturing 1000–10 000 sur-
of planar and acicular (needle shaped) divisions (Figure 6).
face points per m2 can be used to quantify surface roughness
Wadell (1932) clearly demonstrated the difference between
in considerable detail and, for gravel surfaces, have proven
sphericity and roundness and developed procedures for
effective for measuring grain size; surface topography of in-
quantifying these two aspects of particle shape. Wadell’s
dividual clasts; and L, I, and S axes (e.g., Heritage and Milan,
roundness formula was simplified by Krumbein (1941) who
2009; Hodge et al., 2009).
developed a corresponding visual chart that has been widely
Observations of surface texture on sand-sized particles
adopted (Figure 8).
started with Sorby (1880) who used a petrographic micro-
Subsequent innovations have involved digital imaging to
scope to describe the roundness and surface features of sand
provide comprehensive information about particle shape, and
grains. Until the late 1960s, surface microtexture studies were
this is generally combined with mathematical and statistical
limited by microscope technology, but the scanning electron
analyses, such as Fourier, fractal or wavelet transformations, or
microscope (SEM) revolutionized this field of research. A wide
discrete geometry. For example, Schwarcz and Shane (1969)
range of different microtextures can now be identified and are
pioneered the use of Fourier transformations for analyzing the
documented in studies, such as Krinsley and Doornkamp
silhouettes of beach sand grains. This approach has continued
(1973) and Mahaney (2002).
to be developed, for example, to improve methods for locating
the center of mass (centroid) of the form, for determining an
appropriate measure of radius in particles with jagged or
crenellated silhouettes, and to extend its application to a wide 14.20.3 Methodology
variety of particle shapes and sizes (e.g., Diepenbroek et al.,
1992). Orford and Whalley (1983) among others recognized The method for quantitative description of a particle has to be
that using the fractal dimension – a statistical measure of the tailored to the particular research question being addressed.
complexity of the grain silhouette assuming that structural Ideally, methods are intuitive and meaningful and they
complexity at larger scales is equivalent to that at smaller are sensitive enough to detect changes or deviations in particle
scales – could provide a way of quantifying the complexity of a shape. Any descriptive indices, if quantitative, should be
248 Examining Particle Shape

normalized between 0 and 1, where possible (see Table 2 for ratio of the I- and L-axes indicates the degree of grain
examples). Depending on the method, for some particles, elongation and the ratio of the S- and I-axes indicates the
each component of the form (i.e., shape, roundness, and degree of grain flattening. Some of the most common ways of
surface texture) has to be assessed individually; for others, two using combinations of L, I, and S to describe particle form are
or more components may be assessed at the same time. For given in Table 2. Barrett (1980), Blott and Pye (2008), and
example, sphericity and roundness of sand grains may be as- Clayton et al. (2009) provided detailed discussion of the use
sessed using a single index (e.g., Powers, 1982), but assess- of these indices and more comprehensive lists of formulae.
ment of texture of grain surfaces for sand fractions is generally One example is the Corey shape factor (CSF)
conducted separately.
Orford (1991) divided methods of form measurement into S
four main types: visual description, visual comparison, direct CSF ¼ pffiffiffiffiffi
LI
measurement, and response measurement. Of these, visual
comparison and direct measurement are flexible, the most
widely used, and offer a good balance among accuracy, ease of A value of CSF ¼ 0 represents a flat disk, whereas CSF ¼ 1
application, efficiency, replicability, and cost-effectiveness. represents a sphere (Corey, 1949).
For gravel-sized particles and larger ones (42 mm), direct A variety of methods for presenting and assisting in the
measurement of the particle axes and visual assessment of interpretation of particle shape (particularly axis-ratio) data
roundness are the simplest ways to determine aspects of par- has been proposed. One of the most widespread is the use of
ticle shape (Table 1). The three main axes of the particle are triangular (ternary) plots as advocated by Sneed and Folk
measured perpendicular (orthogonal) to each other, but not (1958) (Figure 7), although their use and the most appro-
necessarily intersecting at a common point. These are the long priate values to plot have been the subject of considerable
(L- or A-), intermediate (I- or B-), and short (S- or C-) axes that debate (e.g., Benn and Ballantyne (1992, 1993), Blott and Pye
are also used to quantify particle size. The axial measurements (2008), Illenberger (1991, 1992a, b), Graham and Midgley
are then combined to describe particle shape, for example, the (2000)).

Table 1 Simple assessment of shape for coarse particles

Field Records
1. Measure the three orthogonal axes (L, I, and S)
2. Compare each clast to a visual chart, for example, Krumbein (1941), Powers (1982) and assign a roundness value
3. Record the presence or absence of surface texture on each clast, for example, pits, grooves, striations, flutes, helical forms; estimate the
proportion of the clast with surface texture
4. If possible, record clast lithology
The number of particles that should be measured to obtain a reliable assessment of the particle shape in a particular location will vary but is usually
recommended to be at least 50 and possibly more, particularly if statistical discrimination is required. If a sample is poorly sorted or with mixed
lithology, a larger number of clasts should be examined.
Data analysis
1. Calculate axis ratios using L, I, and S to assign shape values (e.g., Zingg, 1935) and plot data on appropriate plots, for example, ternary plot (e.g.,
Sneed and Folk, 1958)
2. Plot histograms and/or bivariate plots of particle roundness (e.g., Benn and Ballantyne, 1994)
3. Determine relationships between sphericity and roundness
For samples with mixed clast size and lithology, data should be analyzed for individual size fractions and each lithology.

Table 2 Examples of ways of estimating particle form using two or three axis measurements (L¼long axis, I¼intermediate axis, S¼short axis)

Index Formula Range of values Author

Wentworth index of flatness LþI 1–N Wentworth, 1922

2S
Elongation ratio and flatness ratio S
I ; L
I 0–1 Zingg, 1935
Corey shape factor pffiffiffi
S 0–1 Corey, 1949
LI
qffiffiffiffi
Maximum projection sphericity (cP) 3 S2 0–1 Sneed and Folk, 1958
LI
qffiffi
Projection sphericity I 0–1 Pye and Pye, 1943
L
Aschenbrenner shape factor (F) LS 0–N Aschenbrenner, 1956
I 2 ffiffiffiffi
q
Intercept sphericity (c) 3 IS 0–1 Krumbein, 1941; Pye and Pye, 1943
L2
Degree of equancy S
L
0–1 Blott and Pye, 2008
 Examining Particle Shape 249

Compact Although measurements of particle axes give a good indi-

cation of particle shape (and size), they do not reveal infor-

1.0
1.0 mation about roundness. Roundness is generally assessed in two
dimensions based on the maximum projected area (MPA) of the
particle. For example, using the method of Wadell (1932),

0.8
0.8 C roundness is the ratio of the curvature of the corners and edges of
the grain to the average curvature of the whole particle (Table 3).

0.6
S/L I/L This method is complex and time consuming, and consequently
or 0.6 CP CB CE or a more widely used method is visual comparison of the particle
C/A B/A to the roundness chart of Krumbein (1941) (Figure 8). For ex-

0.4
ample, Figure 9 shows clasts from a range of different glacial
0.4 P B E environments in Svalbard. Simple visual comparison of the MPA
of these clasts with Figure 8 demonstrates the variability in

0.2
roundness from 0.2–0.3 for supraglacial debris to 0.8–0.9 for the
0.2 ‘egg gravels.’ Krumbein (1941) based his chart on drawings of
VP VB VE
pebbles 16–32 mm in diameter for which Wadell’s index had
been calculated, but it is widely used for particles outside this size

0
0
range. Some roundness charts combine elements of sphericity as
0.6

0.8

1.0
2

4
0

well and are particularly useful where the three particle axes are
0.

0.

Platy Bladed Elongated harder to measure, for example, the roundness chart by Powers
(L−I)/(L−S) or (A−B)/(A−C) (1982) was designed for sand grains (Figure 10). Diepenbroek
et al. (1992) provides a useful summary and evaluation of 15
Figure 7 Classification of particle shape based on particle axes
ratios. L (A), I (B), and S (C) are the long, intermediate, and short different methods of particle roundness characterization.
axes measurements, respectively. Reproduced from Sneed, E.D., Folk, Roundness data have been presented in a variety of
R.I., 1958. Pebbles in the lower Colorado River, Texas, a study of ways of which the most common is a histogram of qualitative
particle morphogenesis. Journal of Geology 66(2), 114–150. classes (e.g., very angular and angular) and these can

Table 3 Examples of ways of estimating particle roundness (L¼long axis, I¼intermediate axis, Sm ¼shortest axis in maximum projection plane,
Ds ¼diameter of circle fitting sharpest corner, Di ¼diameter of largest inscribed circle, LDS ¼longest diameter measured through sharpest corner,
Dr ¼diameter of curvature of circles fitting any corners)

Index Formula Range of values Author

Wentworth Shape Index Ds 0–1 Wentworth, 1919

LDS
Wentworth Shape Index DS 0–1 Wentworth, 1922
ðLþSm Þ=2
Keunen Roundness Index Ds 0–1 Kuenen, 1956
I
Modified Wentworth Roundness Ds 0–1 Dobkins and Folk, 1970
Di
SðDr =Di Þ
Wadell Roundness
n ¼ ðSDDri=nÞ 0–1 Wadell, 1932

0.1 0.2 0.3 0.4 0.5

0.3 0.4 0.4

0.5
0.5 0.4
0.6 0.7 0.8 0.9 Broken pebbles

Figure 8 The visual comparison chart of Krumbein, W.C., 1941. The measurement and geological signficance of shape and roundness of
sedimentary particles. Journal of Sedimentary Petrology 27, 291–303 for assessing particle roundness.
250 Examining Particle Shape

ML01 ML02 ML03

ML04 ML05 ML06

ML07
Clasts from glacial environments: midtre Lovénbreen, Svalbard
ML01: Distal glaciofluvial, mixed lithology
ML02: Distal glaciofluvial, unilithology (augen gneiss)
ML03: Esker gravels, unilithology (augen gneiss)
ML04: ‘Egg gravels,’ unilithology (augen gneiss)
ML05: ‘Egg gravels,’ mixed lithology
ML06: Supraglacial debris, unilithology (augen gneiss)
ML07: Supraglacial debris, mixed lithology

Figure 9 Clasts from different glacial environments, midtre Lovénbreen, Svalbard, showing different degrees of rounding. Reproduced from
Hubbard, B., Glasser, N., 2005. Field Techniques in Glaciology and Glacial Geomorphology. Wiley, Chichester, UK.

also be assigned numerical values (Figure 10). Even using are taken and examined in the laboratory. Using a microscope,
qualitative techniques, it is unusual for all particles within values of sphericity and roundness can be visually assigned.
a sample to have the same roundness values so the It is also straightforward to obtain digital images of
distribution of values is more useful than a single mean. An individual sand and clay particles, which can be analyzed
alternative approach is to plot mean roundness and the using more sophisticated techniques, such as automated
roundness range for each sample of sediments to be com- scanning electron microscopy (QEMSCANs; e.g., Speirs et al.,
pared. For small particles, such as sands, silts, and clays, where 2008). Some fine-particle sizing methods can also be used
a very low number of particles will be assessed for roundness to provide information about particle shape, for example,
in comparison with the number used to determine the Abbireddy et al. (2009) have developed a method for deter-
particle-size distribution, a single value of roundness for the mining the shape of particles o63 mm diameter using tur-
sample can be calculated by weighting the mean roundness bidity measurements and laser diffraction measurements of
value for each size fraction according to the proportion particle size.
of material in that size fraction (e.g., %volume) (Bullard et al., For fine particles, surface texture is generally assessed using
2007). an SEM, the results of which provide quite detailed information
Some methods of presentation combine aspects of shape about the environmental and transport history of the grains
and roundness, for example, the combination of the RA Index (e.g., Mahaney, 2002). A wide range of microtextures can be
(percentage of angular þ very angular particles) and C40 Index observed, including parallel ridges, depressions, conchoidal
(percentage of particles with an S/L axis ratio of r0.4) as fractures, grooves, chattermarks, abrasion features, dissolution
advocated by Benn and Ballantyne (1994) has proven a useful etching, precipitation features, and clay coatings (Figure 12).
discriminator of the different processes to which sediments in Only a few microtextures are unique to specific environments
glacial environments have been subjected and been widely or processes, for example, microstriations are unique to glacial
adopted (Figure 11). processes and V-shaped percussion scars are unique to fluvial
Particles less than 2 mm diameter cannot generally be processes; however, because most particles will be exposed
measured in the field for roundness and shape, so samples to a variety of different processes, different signatures can
 Examining Particle Shape 251

Roundness
Very angular Angular Subangular Subrounded Rounded Well rounded
0.5 1.5 2.5 3.5 4.5 5.5

Discoidal
0.5
Subdiscoidal
1.5
Prismoidal Subprismoidal Spherical
Sphericity

4.5 3.5 2.5

Figure 10 The visual comparison chart of Powers, M.C., 1982. Comparison chart for estimating roundness and sphericity. AGI Data Sheet 18,
American Geological Institute, Virginia, USA for assessing particle sphericity and roundness.

be overprinted with each change of environment (Figure 13). are still prevalent. In the same that distributions of particle
It is also important to examine a representative population of sizes are affected by the method used to determine that dis-
grains in order to identify the origins of particles or processes tribution (e.g., sieving, settling velocity, laser particle sizer
to which they have been subjected (see Mahaney (2002) for (see Chapter 14.22)) so values of shape descriptors, such
guidance on this). as roundness, depend on the methods used. Consequently,
For all sizes of particles, the minimum number of meas- such values should only directly be compared with those
urements required to be representative from each facies or studies where the same method of assessing roundness
landform should be at least 50 clasts (Hubbard and Glasser, or sphericity has been used. Some researchers offer useful
2005). Given that many researchers have noted how particle comparisons of the performance of different techniques,
shape varies with size and lithology, if the site being sampled for example, Diepenbroek et al. (1992) calculated Fourier
contains sediments of varying size and rock type, then larger roundness and Krumbein’s roundness for gravel particles and
numbers of measurements should be made. The size of the found a good relationship, although variations in particle
particles will, to some extent, determine whether or not sphericity were found to affect this (Figure 14).
measurements have to be made at the field site. For larger Some of the potential errors associated with the assessment
particles and boulders, measurements will be made in situ; but of particle shape include axis selection, operator variance, and
for smaller sediments, it may be possible to collect samples for repeatability. In terms of axes, the L-axis (or A-axis) has
later analysis and will be inevitable for sand-sized fractions traditionally been the longest dimension of the particle
and finer. Increasingly methods are being developed to cap- (Krumbein, 1941) but as Blott and Pye (2008) discussed, in
ture digital images in the field that can then be analyzed later; the case of a cube, the longest axis would be along the diag-
this enables much larger numbers of particles to be charac- onal from corner to the opposite corner. They recommend
terized. Although these image-based techniques are rapidly orientating the particle to present the smallest projected area
improving for particle-size analyses (Graham et al., 2005; to the eye, which will give the I- and S-axes, and then meas-
Warrick et al., 2009), factors such as particle shape can uring the L-axes perpendicular to I and S. Blott and Pye (2008)
introduce bias into measurements and so it remains advisable also found that by using calipers and a ruler to measure gravel
to also conduct field analyses for calibration purposes clasts, the precision of measurements made by a single oper-
(Buscombe et al., 2010). ator varied from 0.5% to 2.9%, whereas precision among
Despite the potentially increased precision, accuracy, and three different operators decreased to 2.5–5.9%. Larger errors
objectivity of digital imaging, and mathematical approaches to were associated with smaller and more irregular particles.
particle shape, characterization based on axis ratios, and Operator variance can also occur in the visual assessment of
qualitative descriptions based on reference shapes or images particle roundness (e.g., Folk 1955, 1972).
252 Examining Particle Shape

100
Input to glaciers
70 80
60 Scree and rockfall
50 60

% RA
Stream dominated
40
30 40
Snow-avalanche dominated
20 20
10
0 0
VA A SA SR R WR VA A SA SR R WR VA A SA SR R WR 0 20 40 60 80 C40
Scree and rockfall Snow-avalanche Stream dominated
(3 sites, n = 150) dominated (2 sites, n = 100)
(a) (2 sites, n = 100)

100
Transport by glaciers
70 80
60 Supraglacial facies
50 60

% RA
Reworked supraglacial facies
40
40 Basal debris facies
30
20 20
10
0 0
VA A SA SR R WR VA A SA SR R WR VA A SA SR R WR 0 20 40 60 80 C40
Supraglacial facies Reworked Basal debris facies
(5 sites, n = 250) supraglacial facies (1 site, n = 50)
(b) (4 sites, n = 200)

100
Direct deposition from ice
70 80
60
End moraine
50 60
% RA

Lateral moraine
40
30 40
20 20
10
0 0
VA A SA SR R WR VA A SA SR R WR 0 20 40 60 80 C40
Lateral moraine End moraine
(c) (10 sites, n = 500) (5 sites, n = 250)

100
Reworking after deposition by ice
70 80
60 Glacial facies
50 60 Alluvial fan
% RA

40 Lake shoreline
30 40
20 20
10
0 0
VA A SA SR R WR VA A SA SR R WR VA A SA SR R WR 0 20 40 60 80 C40
Braided outwash, Alluvial fan Lake shoreline
glaciofluvial facies (1 site, n = 50) (1 site, n = 50)
(d) (9 sites, n = 450)
Figure 11 Particle shape characteristics of different sedimentary environments, Mount Cook area, New Zealand. Roundness based on Powers
(see Figure 10); RA/C40 plots, where RA is the percentage of very angular þ angular clasts and C40 is the percentage of clasts with a S/I ratio of
r0.4. Reproduced from Hambrey, M.J., Ehrmann, W., 2004. Modification of sediment characteristics during glacial transport in high-alpine
catchments: Mount Cook area, New Zealand. Boreas 33, 300–318, with permission from Wiley.
 Examining Particle Shape 253

R64A1

R64A2

1/19/2004 Det HV Mag WD Spot Pressure 100.0µm 1/19/2004 Det HV Mag WD Spot Pressure 200.0µm
3:16:26 PM ETD 20.0 kv 625x 9.7 mm 3.5 --- R35AB 4:16:59 PM ETD 20.0 kv 400x 14.6 mm 3.5 --- R64A
(a) (b)

Figure 12 Scanning electron micrographs of two sand grains from the Simpson Desert, Australia, with contrasting roundness and surface
textures. (a) Spherical, well-rounded quartz grain with limited evidence of pitting and (b) elongated, subrounded, and heavily textured grain with
sulfur and calcium precipitated on the surface.

es

ks
at
s

ac
s re

pl
Lo rp gro ves ure ctu

cr
Ed hap g p fea ces es
es

a g ed

ve ou er le s

n
s in n fa c
o t a

ur

Pr solu fe rin urn

O e r d p rtic ure

te g io
V- her atio sur urfa
Sh igh gro frac r fr

Ad ip ed d s g
M p t tep step s
at

ns
rp n cu s
rin din ss
ec er re hin
Pr ath the etc s
ge
ec ro s s

is on tte pt

g e a t
fe
l a

e
ed lie ul s

ai
St ved ida line

W we tion tur
M re g e

D ear ped ou

D asi ha y u
H iu f ar
C ch lle s

gr
w an ov

Ab tice ica s
Ar sc ief f
on ra ce

re l lie

n a g

t n h
r s ll

d
ur o l

La ha ug
C bpa fa

Li -sh tic
C h re re

c n
re

ig m

ee s
t

e a

t
e

i
Su ctu

ra
a

e
a
Fr
% Frequency of occurrence

60 60
50 50
40 40
30 30
20 20
10 10
(a)
0 0
% Frequency of occurrence

60 60
50 50
40 40
30 30
20 20
10 10
(b)
0 0
% Frequency of occurrence

60 60

50 50

40 40

30 30

20 20
(c)
10 10

0 0
Figure 13 Frequency and occurrence of microtextures visible using SEM on different deposits. (a) Till, (b) glaciofluvial beds, (c) Californian
aeolian Sands.
254 Examining Particle Shape

1.0
0.9
0.8
0.7
P - Fourier
0.6
0.5
0.4
0.3
0.2
0.1
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
P - Krumbein
Figure 14 Relationship between Fourier roundness (P-Fourier) and Krumbein roundness (P-Krumbein) for gravels (25–50 mm) from Calabria,
Italy. Reproduced from Diepenbroek, M., Bartholoma, A., Ibbeken, H., 1992. How round is rounds? A new approach to the topic ‘roundness’ by
Fourier grain shape analysis. Sedimentology 39, 411–422, with permission from Wiley.

14.20.4 Limitations processes can also affect grain roundness and surface texture.
For example, Potter (1978) suggested that sand grains in tro-
Particle shape is arguably an essential property to measure pical rivers are more rounded with embayed corroded surfaces
because it is a reflection of the processes to which a particle than sand grains in temperate rivers, reflecting the action of
has been exposed. Shape can be determined in a variety of chemical weathering in the tropical environment. For sedi-
ways, some of which are very simple, others are more complex, mentary deposits, therefore, some understanding of the cli-
and there is a wide range of indices that can be derived to matic conditions to which they have been exposed in the past
compare and classify particles of different shapes (although can be important.
there is no universally accepted and adopted method). If a study can be constrained to comparison of particles
However, the ease with which form, roundness, and surface in the same land system, relative (rather than absolute) shape
texture can be characterized is in contrast to the difficulties may be an important quality for interpreting their history
that can be faced in interpreting their geomorphological and useful conclusions may be drawn from shape analyses
significance. even where limited contextual data are available. Yingst et al.
One reason why the interpretation of particle shape has (2007) examined the shape of particles in the size range
to be treated with caution is that shape reflects the interaction 16–256 mm on Mars through analysis of Mars Pathfinder
of several different controlling factors that can be difficult Images. The study was reliant on two-dimensional ‘apparent’
to isolate from one another. These factors include the shape and size measurements because the particles could
shape of the original particle; the lithology of the particle not be picked up or rotated. Through analysis of the meas-
(including mineralogy, structure, and crystallography); the urements of sphericity, they concluded that the particles
susceptibility of the particle to weathering and abrasion; and were likely to be of a single lithology but their analysis of
the nature, intensity, and duration of the geomorphological roundness suggested that the particles had been subjected to
processes acting on the particle (e.g., Krumbein and Pettijohn, a variety of different processes such as aeolian transport, im-
1938). pact cratering, and transport by catastrophic floods (Yingst
The initial characteristics of the particle, as determined by et al., 2007).
lithology or crystallography, and the process by which it is A very strong argument exists for combining measurement
released from bedrock have a very significant impact on the of particle shape with other grain characteristics, arguably the
role of the grain in geomorphological processes. Anisotropy most important of which are size and sorting. Particle
can cause particles to be eroded at different rates along dif- size, sorting, and shape are not independent of each other
ferent axes and only isotropic particles are therefore likely to and many studies have found significant relationships
be highly spherical (equant). In single mineral grains, dur- amongst these three variables for a number of reasons.
ability, susceptibility to cleavage, and fracturing as well as As larger particles are more likely to have a composite min-
anisotropy control the overall shape of the grain. If the initial eralogy (anisotropic), they are less likely to be spherical
shape or origin of the particle is unknown, then absolute than smaller particles. In poorly sorted sediments, the larger
particle shape may not be a meaningful descriptor. In addition grains are commonly less spherical but more rounded than
to lithology, spatial variations in the nature and type of smaller grains, but these trends are not universal. The
 Examining Particle Shape 255

processes controlling grain shape can also affect grain size, for Blott, S.J., Pye, K., 2008. Particle shape: a review and new methods of
example, abrasion makes grains rounder and also smaller. characterization and classification. Sedimentology 55, 31–63.
Bluck, B.J., 1967. Sedimentation of beach gravels: examples from South Wales.
The size of a particle may affect the frequency with which it is
Journal of Sedimentary Petrology 37, 128–156.
entrained and transported and hence the period of time to Boulton, G.S., 1978. Boulder shapes and grain-size distributions as indicators
which it is exposed to processes such as abrasion and attrition of transport paths through a glacier and till genesis. Sedimentology 25,
which in turn will affect grain roundness, surface texture, 773–799.
and size. However, shape can also affect susceptibility to en- Bullard, J.E., McTainsh, G.H., Pudmenzky, C., 2007. Factors affecting the nature and
rate of dust production from natural dune sands. Sedimentology 54, 169–182.
trainment. For example, Willetts et al. (1982) found that Buscombe, D., Rubin, D.M., Warrick, J.E., 2010. A universal approximation of grain
rounded sand particles were more easily entrained by size from images of noncohesive sediment. Journal of Geophysical Research –
the wind than angular particles and McGowan and Sturman Earth Surface 115. Art. no. F02015.
(1997) noted that more rounded particles were preferentially Cailleux, A., 1947. L’indice d’émoussé: definition et première application. Comptes
Rendus Somaires de la Société Geologique de France 13, 250–252.
transported higher in the airstream than were less rounded
Carey, S., Maria, A., Sigurdsson, H., 2000. Use of fractal analysis for discrimination
particles of the same size. They suggest that increased round- of particles from primary and reworked jökulhlaup deposits in SE Iceland.
ing with distance from source may be more a reflection Journal of Volcanology and Geothermal Research 104, 65–80.
of preferential entrainment and lifting of rounded Clayton, C.R.I., Abbireddy, C.O.R., Schiebel, R., 2009. A method of estimating the
particles higher in to the airstream than of abrasion during form of coarse particulates. Géotechnique 59, 493–501.
Clemens, K.E., Komar, P.D., 1988. Tracers of sand movement on the Oregon Costs.
transport.
Proceedings of the 21st Coastal Engineering Conference. American Society of
Civil Engineers, Torremolinos, Spain, pp. 1338–1351.
Corey, A.T., 1949. Influence of shape on the fall velocity of sand grains. M.Sc.
thesis, A&M College, Colorado, Unpublished.
14.20.5 Conclusions Cui, B., Komar, P.D., Baba, J., 1983. Settling velocity of natural sand grains in air.
Journal of Sedimentary Petrology 53, 1205–1211.
Diepenbroek, M., Bartholoma, A., Ibbeken, H., 1992. How round is round? A new
Notwithstanding the difficulties in ascribing meaning to approach to the topic ‘roundness’ by Fourier grain shape analysis.
particle shape, it can be a useful variable to measure. This is Sedimentology 39, 411–422.
a strong argument for ensuring that all three measures of Di Stefano, C., Ferro, V., Mirabile, S., 2010. Comparison between grain-size
analyses using laser diffraction and sedimentation methods. Biosystems
particle shape are considered for each particle – commonly
Engineering 106, 205–215.
measurements of only sphericity or only roundness provide Dobkins, J.E., Folk, R.L., 1970. Shape development on Tahiti-Nui. Journal of
poorer discrimination than when two or more measures Sedimentary Petrology 40, 1167–1203.
are used. It is also essential to consider particle size and El-Sayed, M.I., 1999. Sedimentological characteristics and morphology of the
other variables, such as lithology. Quantitative analyses of aeolian sand dunes in the eastern part of the UAE: a case study from Ar Rub’Al
Khali. Sedimentary Geology 123, 219–238.
grain shape potentially reduce subjectivity but there is Fernlund, J.M.R., 1998. The effect of particle form on sieve analysis: a test by
still considerable potential for operator-induced variation image analysis. Engineering Geology 50, 111–124.
(such as identifying the three orthogonal axes or deter- Folk, R.L., 1951. Stages of textural maturity. Journal of Sedimentary Petrology 21,
mining the particle centroid) that limit the accuracy of such 127–130.
Folk, R.L., 1955. Student operator error in determination of roundness, sphericity
measurements. As there are no universally accepted par-
and grain size. Journal of Sedimentary Petrology 25, 297–301.
ameters for describing particle shape and no universally Folk, R.L., 1972. Experimental error in pebble roundness determination
adopted methods, comparison among studies should be by the modified Wentworth method. Journal of Sedimentary Petrology 42,
restricted to those which have used the same approach un- 973–974.
less techniques can be devised to cross-calibrate different Gordon, N.D., McMahon, T.A., Finlayson, B.L., Gippel, C.J., Nathan, R.J., 2004.
Stream Hydrology – An Introduction for Ecologists. Wiley, Chichester, UK.
methods. Goudie, A.S., Watson, A., 1981. The shape of desert sand dune grains. Journal of
Arid Environments 10, 1–12.
Graf, W.L., 1988. Fluvial Processes in Dryland Rivers. Springer-Verlag, Berlin,
Germany.
References Graham, D.J., Midgley, N.G., 2000. Graphical representation of particle shape using
triangular diagrams: an Excel spreadsheet method. Earth Surface Processes and
Abbireddy, C.O.R., Clayton, C.R.I., Huvenne, V.A.I., 2009. A method of estimating
Landforms 25, 1473–1477.
the form of fine particulates. Géotechnique 59, 503–511.
Graham, D.J., Rice, S.P., Reid, I., 2005. A transferable method for the automated
Aschenbrenner, B.C., 1956. A new method of expressing particle sphericity. Journal
grain sizing of river gravels. Water Resources Research 41(7), art. no.
of Sedimentary Petrology 26, 15–31.
W07020.
Barrett, P.J., 1980. The shape of rock particles, a critical review. Sedimentology
Gray, W.A., 1968. The Packing of Solid Particles. Chapman & Hall, London, UK.
27(3), 291–303. Hambrey, M.J., Ehrmann, W., 2004. Modification of sediment characteristics during
Benn, D.I., Ballantyne, C.K., 1992. Pebble shape (and size!) – discussion. Journal
glacial transport in high-alpine catchments: Mount Cook area, New Zealand.
of Sedimentary Petrology 62, 1147–1150. Boreas 33, 300–318.
Benn, D.I., Ballantyne, C.K., 1993. The description and representation of particle Heritage, G.L., Milan, D.J., 2009. Terrestrial laser scanning of grain roughness in a
shape. Earth Surface Processes and Landforms 18, 665–672. gravel-bed river. Geomorphology 113, 4–11.
Benn, D.I., Ballantyne, C.K., 1994. Reconstructing the transport history of glacigenic Hodge, R., Barsington, J., Richards, K., 2009. Analysing laser-scanned digital
sediments: a new approach based on the co-variance of clast form indices. terrain models of gravel bed surfaces: linking morphology to sediment transport
Sedimentary Geology 91, 215–227. processes and hydraulics. Sedimentology 56, 2024–2043.
Beuselinck, L., Govers, G., Poesen, J., Degraer, G., 1998. Grain-size analysis by Hubbard, B., Glasser, N., 2005. Field Techniques in Glaciology and Glacial
laser diffractrometry: comparison with the sieve-pipette method. Catena 32, Geomorphology. Wiley, Chichester, UK.
193–208. Illenberger, W.K., 1991. Pebble shape (and size!). Journal of Sedimentary Petrology
Blott, S.J., Pye, K., 2006. Particle size distribution analysis of sand-sized particles 61, 756–767.
by laser diffraction: an experimental investigation of instrument sensitivity and Illenberger, W.K., 1992a. Pebble shape (and size!) – reply. Journal of Sedimentary
the effects of particle shape. Sedimentology 53, 671–685. Petrology 62, 538–540.
256 Examining Particle Shape

Illenberger, W.K., 1992b. Pebble shape (and size!) – reply. Journal of Sedimentary Orford, J.D., Whalley, W.B., 1983. The use of fractal dimension to quantify the
Petrology 62, 1151–1155. morphology of irregular-shaped particles. Sedimentology 30, 644–668.
Kench, P.S., McLean, R.F., 1997. A comparison of settling and sieve techniques Potter, P.E., 1978. Petrology and composition of modern big-river sands. Journal of
for the analysis of bioclastic sediments. Sedimentary Geology 109, Geology 86, 423–449.
111–119. Powers, M.C., 1982. Comparison chart for estimating roundness and sphericity.
Khalaf, F.I., Gharib, I.M., 1985. Roundness parameters of quartz grains of recent American Geological Institute, Virginia, USA, AGI Data Sheet 18.
aeolian sand deposits in Kuwait. Sedimentary Geology 45, 147–158. Pye, W., Pye, M., 1943. Sphericity determination of pebbles and sand grains.
Kirkbride, M.P., Bell, C.M., 2010. Edge-roundness of boulders of Torridonian Journal of Sedimentary Petrology 13, 28–34.
Sandstone (northwest Scotland): applications for relative dating and implications Rice, M.A., 1991. Grain shape effects on aeolian sediment transport. Acta
for warm and cold climate weathering rates. Boreas 39, 187–198. Mechanica, Supplement 1, 159–166.
Knighton, A.D., 1982. Longitudinal changes in the size and shape of stream bed Rodriguez, J.G., Uriarte, A., 2009. Lase-diffraction and dry-sieving grain size
material: evidence of variable transport conditions. Catena 9, 25–34. analyses undertaken on fine and medium-grained sandy marine sediments: a
Komar, P.D., Reimers, C.E., 1979. Grain shape effects on settling rates. Journal of note. Journal of Coastal Research 25, 257–264.
Geology 86, 193–209. Schwarcz, H.P., Shane, K.C., 1969. Measurement of particle shape by Fourier
Krinsley, D., Doornkamp, J.C., 1973. Atlas of Sand Grain Surface Textures. analysis. Sedimentology 13, 213–231.
Cambridge University Press, Cambridge, UK. Sneed, E.D., Folk, R.I., 1958. Pebbles in the lower Colorado River, Texas, a study
Krumbein, W.C., 1941. The measurement and geological significance of shape and of particle morphogenesis. Journal of Geology 66(2), 114–150.
roundness of sedimentary particles. Journal of Sedimentary Petrology 27, Sorby, H.C., 1880. On the structures and origin of noncalcareous stratified rocks.
291–303. Quaterly Journal of the Geological Society of London 36, 46–92.
Krumbein, W.C., Pettijohn, F.J., 1938. Manual of Sedimentary Petrography. Speirs, J.C., McGowan, H.A., Neil, D.T., 2008. Polar eolian sand transport:
Appleton-Century-Crofts, New York, NY. grain characteristics determined by an automated scanning electron
Kuenen, Ph.H., 1956. Experimental abrasion of pebbles: 2. Rolling by current. microscope (QEMSCANs). Arctic, Antarctic and Alpine Research 40,
Journal of Geology 64, 337–368. 731–743.
Kuenen, Ph.H., 1959. Experimental abrasion. 3. Fluviatile action on sand. American Trenhaile, A.S., Van Der Nol, L.V., La Valle, P.D., 1996. Sand grain roundness
Journal of Science 257(3), 172–190. and transport in the swash zone. Journal of Coastal Research 12(4),
Lancaster, N., 1989. The Namib Sand Sea: Dune Forms, Processes and Sediments. 1017–1023.
Balkema, Rotterdam, The Netherlands. Tucker, M.E., 2001. Sedimentary Petrology, Third ed. Blackwell Science, Oxford, UK.
Landon, R.S., 1930. An analysis of beach pebble abrasion and transportation. Wadell, H., 1932. Volume, shape and roundness of rock particles. Journal of
Journal of Geology 38, 437–446. Geology 40, 443–451.
Leeder, M.R., 1982. Sedimentology: Process and Product. Allen & Unwin, Boston, MA. Warrick, J.A., Rubin, D.M., Ruggiero, P., Harney, J.N., Draut, A.E., Buscombe, D.,
Li, Z., Komar, P.D., 1986. Laboratory measurements of pivoting angles for 2009. Cobble cam: grain-size measurements of sand to boulder from digital
application to selective entrainment of gravel in cement. Sedimentology 33, photographs and autocorrelation analyses. Earth Surface Processes and
413–423. Landforms 34, 1811–1821.
Mackie, W., 1897. On the laws that govern the rounding of particles of sand. Wentworth, C.K., 1919. A laboratory and field study of cobble abrasion. Journal of
Transactions of the Edinburgh Geological Society 7, 298–311. Geology 40, 507–521.
Mahaney, W.C., 2002. Atlas of Sand Grain Surface Textures and Applications. Wentworth, C.K., 1922. The shapes of beach pebbles. USGS Geological Survey
Oxford University Press, Oxford, UK. Professional Paper, 131-C, 75–83.
Masselink, G., Hughes, M., 2003. An Introduction to Coastal Processes and Whalley, W.B., 1972. The description and measurement of sedimentary particles and
Geomorphology. Hodder Arnold, London, UK. the concept of form. Journal of Sedimentary Petrology 42, 961–965.
Mazzullo, J.M., Sims, D., Cunningham, D., 1986. The effects of eolian sorting and Willetts, B.B., Rice, M.A., Swaine, S.E., 1982. Shape effects in aeolian grain
abrasion upon the shapes of fine quartz sand grains. Journal of Sedimentary transport. Sedimentology 29, 409–417.
Petrology 56, 45–56. Wilson, L., Huang, T.C., 1979. The influence of shape on the atmospheric settling
McGowan, H.A., Sturman, A.P., 1997. Characteristics of aeolian grain transport over velocity of volcanic ash particles. Earth and Planetary Science Letters 44,
a fluvio-glacial lacustrine braid delta, Lake Tekapo, New Zealand. Earth Surface 311–324.
Processes and Landforms 22(8), 773–784. Yingst, R.A., Haldemann, A.F.C., Beidermann, K.L., Monhead, A.M., 2007.
Muhs, D.R., 2004. Mineralogical maturity in dunefields of North America, Africa Quantitative morphology of rocks at the Mars Pathfinder landing site. Journal of
and Australia. Geomorphology 59, 247–269. Geophysical Research 112, E06002. http://dx.doi.org/10.1029/2005JE002582.
Orford, J.D., 1991. Particle form. In: Goudie, A.S. (Ed.), Geomorphological Zingg, T., 1935. Beı́träge zur Schotteranalyse. Schweizerische mineralogische und
Techniques, Second ed. Routledge, London, UK, pp. 121–127. Petrologische Mitteilungen 15, 39–140.

Biographical Sketch

Joanna E. Bullard is currently based at the Department of Geography, Loughborough University, UK. Her research
focuses on geomorphology and sedimentology in dryland environments, for which she has established an
international reputation and has recently been awarded the Philip Leverhulme Prize in the field of Earth, Ocean
and Atmospheric Sciences and the Gordon Warwick Award (from the British Geomorphological Research Group).
It contributes to debates concerning, and our understanding of, the response of dryland systems to climate
fluctuations and the potential importance of aeolian processes to global biogeochemical cycles. The majority of
her field research has been conducted in southern Africa and Australia with additional projects in the USA and the
UK.
 14.21 The Scanning Electron Microscope in Geomorphology
PA Bull, University of Oxford, Oxford, UK
RM Morgan, University College London, London, UK
r 2013 Elsevier Inc. All rights reserved.

14.21.1 Introduction 257

14.21.2 Methodology 258
14.21.2.1 Sample Preparation 258
14.21.2.2 Analytical Methodology 258
14.21.3 Case Studies 259
14.21.4 Conclusions 260
References 260

Glossary Geoforensics The study of geological materials and their

Backscattering Backscattered electron (BSE) images application to the law.
provide compositional and topographical contrast within a Ion beam analysis IBA is a technique whereby a beam of
grain (the lighter the grain, the higher the atomic number). charged particles (usually H þ or He þ ) are accelerated to
It is possible to discriminate between minerals of different high energy (MeV) and focused onto the sample to analyze
atomic numbers using BSE images. the elemental composition of the material.
Cathodoluminescence A technique that can be used in Microprobe A chemical analytical tool that is used to
conjunction with an optical or electron microscope in order provide elemental compositional data from a point source
to indicate the provenance or diagenetic textures of a mineral. on a sample.
EDX (EDAX, EDS) Energy Dispersive X-ray analysis is a
means of establishing the spectrum of chemical
constituents (minerals) present in a sample.

Abstract

The scanning electron microscope has developed into one of the most ubiquitous analytical machines available in a
geomorphological laboratory. Its main advantages are that it provides very high magnification (up to 100 000  ) images
and can be associated with chemical, bolt-on, and analytical tools. The widespread application of the electron microscope
includes environmental reconstruction, soil fabric studies, weathering investigations, and more recently it has been used in
geoforensic analysis.

14.21.1 Introduction palaeoenvironmental reconstruction of environments from

The scanning electron microscope (SEM) was first used in the thirty or so mechanically, morphologically, and chemically
geomorphological and geological studies in the 1960s derived quartz-grain, surface texture features. The premise of
(Krinsley and Takahashi, 1962). It evolved from the trans- the approach was that specific environments (e.g., glacial,
mission electron microscope (TEM) which had been com- subaqueous, aeolian, etc.) had specific energy regimes associ-
mercially available since the early 1950s. The SEM ated with that environment and that those energy regimes and
revolutionized microscopic investigations of geological and mechanisms imparted modification to the grain surface. The
biological materials in that it could view the whole specimen grains could portray sequential stages of modification in dif-
rather than a carbon-peel replica that was viewed in a flattened ferent environments and so proved a powerful reconstruction
form by TEM. In the 1960s, the primary use of the SEM in tool for a number of Earth science disciplines including clastic
geomorphology was in clastic sediment studies. This involved sediment studies, archaeology, Quaternary studies, and
quartz-grain surface texture analysis which was pioneered by stratigraphy.
Krinsley and his coworkers (Krinsley and Margolis, 1969, and Utilizing the ability of the machine not only to magnify
see Bull et al., 1985 for review). Their work provided commonly to 10 000  magnifications, but also (subject
specific) more than 100 000  , other disciplines such as
palynology, palaeontology, and various other floral and faunal
Bull, P.A., Morgan, R.M., 2013. The scanning electron microscope in
geomorphology. In: Shroder, J. (Editor in Chief), Switzer, A.D., Kennedy,
studies produced large amounts of published literature. The
D.M. (Eds.), Treatise on Geomorphology. Academic Press, San Diego, CA, earlier SEMs required viewing in a low vacuum chamber with
vol. 14, Methods in Geomorphology, pp. 257–261. energy levels up to 30 kV. Many unfortunate tales exist of

Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00387-0 257

258 The Scanning Electron Microscope in Geomorphology

precious samples vaporizing in front of the eyes of the mi- within a sample in order to determine provenance of the
croscopist due to such intense conditions within the specimen materials (Bull and Morgan, 2006). In this case, it is impera-
chamber. However, for the last 15 years or so, environmental tive that the grains are not washed as they may contain
chambers have been developed which negate the necessity for provenance indicators of extreme importance (soluble salts,
such high voltage and vacuum conditions thus enabling the diatoms, or palynomorphs which themselves can indicate
study of very delicate organisms and structures. The most re- an environmental or biological provenance separate to that of
cent modification and additions to the SEM, particularly for the geological provenance achieved from the quartz-grain
sedimentological investigations, has been the development of analysis).
the SEM–EDX which can be used to determine microstructures Preparation techniques for other minerals will of course
and the chemical spectra of a sample. The SEM is also used to vary dramatically: Acids and alkali solutions may have varied
identify problems related to soil formation (pedology) (Smart effects on the particular mineral being studied. If analysis is by
and Tovey, 1988), mineralogy (crystal morphology, petrology, imaging only then the less intervention required for the better
etc.) weathering processes (Robinson and Williams, 2000), (even washing can remove soluble salts).
diagenetic processes (Marshall, 1987a, b), ground-water Once the sample has been dried, a number of sand grains
movement (hydrology), and even subjects as diverse as mass need to be extracted and placed on an aluminum stub that has
wasting, earthquake investigations (Kanaori, 1985), and extra- been coated with double-sided adhesive tape. It is most cer-
terrestrial studies (Li and Wang, 1980). tainly advantageous for the analyst to put the grains in neat
It is the three-dimensional aspect of sample viewing that lines, preferably with a marker on the stub edge in order that
has made the SEM such a popular tool and its development in specific grains can be revisited under magnification by simple
the last two or three decades has been as an imaging tool, that reference to the lab book description of the position of the
is, a microscope although a large numbers of ‘bolt-on’ ana- grain on the stub. The temptation is to scatter a number of
lytical techniques also occur in SEM catalogues. grains onto the stub before gold-coating rather than the more
laborious method of grain placement. This former method is a
false economy due to the grains being likely to be poorly fixed
to the stub and when viewed under the electron microscope
14.21.2 Methodology
often move and therefore crack the fine gold coating which
results in image flaring.
14.21.2.1 Sample Preparation
The vexed question of how many sand grains should be
The majority of SEMs produced before the year 2000 neces- analyzed for a meaningful representation of the sample and
sitate the samples to be gold-coated by sputter techniques in the consequent confidence in environmental reconstruction
order to facilitate high resolution images. This was a con- and interpretation has long been debated in the literature. The
siderable advancement over the carbon coating techniques early papers of Krinsley and his coworkers, particularly of the
required in the 1950s and 1960s, and commonly still used to TEM carbon replicates era, utilized 10–20 grains for the an-
this day on samples analyzed by TEM. More recent electron alysis (this was of course due to the time consuming sample
microscopes can operate without gold coating in environ- preparation required for each grain). In most environmental
mental chambers, although generally to the detriment of the reconstruction papers published in the 1970s and 1980s using
image resolution when used with Earth science materials. (The the SEM, an order of 30 grains or so was chosen (Marshall,
gold coating is normally only 40 Å thick and can be removed 1987a, b) for analysis, although extremes include Vincent
without harming the specimen if so desired by a number of (1976) who advocated using 200 grains per sample. Recent
chemicals including potassium cyanide.) studies however, have shown that the numbers to be analyzed
In the analysis of quartz-grain surface texture by SEM, the will depend on the number of different provenances from
traditional preparation techniques are to wash the sample in a whence the sample was derived (Bull and Morgan, 2006). In
5% sodium hexametaphosphate solution (Calgon) to dis- most situations (excluding some glacial tills which are some-
aggregate fine materials from the surface of the grains, fol- what of a provenance melange), 30 grains per sample have
lowed by light boiling in dilute hydrochloric acid to remove been shown to be sufficient in geomorphological studies
carbonate cements which commonly cause other particles to whereas in more exacting conditions of the courtroom, 50
adhere to the surface of the grains. Additional washing with a grains are deemed appropriate for geoforensic samples (pro-
dilute Calgon solution, and subsequent rinsing with distilled vided the sample size is large enough to yield 50 grains).
water results in quartz with clean surfaces and textures un- Perhaps the most telling argument of grain numbers can be
affected by the preparation techniques. If an iron coating is seen by employing hindsight. It is particularly noteworthy that
present (as commonly occurring in global sand seas and tro- several of the early papers of the 1960s and 1970s which ne-
pical weathered soils) the grain can be cleansed using a stan- cessitated using 10 or so grains per sample provided conclu-
nous chloride solution to remove the iron. sions which still stand the test of time.
More recently however, electron microscopists have tended
either to view the sample untreated, or to simply use samples
washed by agitation in double distilled water before drying. 14.21.2.2 Analytical Methodology
These two methods are noninterventionist and preserve the
assemblage of materials found on the grain. A more recent The purely qualitative viewing of sand grains using the electron
application of SEM analysis in the field of forensic investi- microscope results in undue operator bias toward the unusual
gation utilizes the textural assemblage of surface features and more selective surface textures (Culver et al., 1983).
 The Scanning Electron Microscope in Geomorphology 259

By convention there are some 30 or so mechanical, morpho- provenance discrimination from quartz sand grains initially
logical, and chemical features identifiable on sand grains classified by SEM analysis (Comini et al., 2010).
(Higgs, 1979), although some analysts, particularly Loic Le The electron microscope also enables soil and rock fabric
Ribault, have identified more 100 features (Le Ribault, 1977). studies to be undertaken. These together with identification of
Obviously therefore, some form of quantitative analysis is de- clay mineralogy under high voltage electron microscopy
sirable in order that grains and indeed samples can be fairly (HVEM) are pertinent in soft sediment studies for soil fabric
compared to each other. Quasi-quantitative analysis was first analysis (Baynes and Dearman, 1978) and for all aspects of
proposed by Margolis (1966, 1968) where the presence or ab- soil and rock engineering (Mahaney, 2002).
sence of the surface texture on a quartz-grain was noted and the
percentage occurrence of these features calculated for the whole
suite of textures on grains within a sample. Such analysis en- 14.21.3 Case Studies
abled statistical interpretation to be undertaken using linear
and multiple discriminant methods (Bull, 1978). Currently, The use of the electron microscope in quartz-grain texture
canonical discriminant function analysis (CDFA) is used to studies is well reported in the literature and the general idea
compare samples and to identify sample dissimilarities and accepted as a viable reconstruction technique following the
facilitates the more visual cluster analysis methodology (Mor- publication of at least 1000 papers on the topic over the past
gan et al., 2010). 50 years. Recent years have seen the development not only of
Krinsley and Wellendorf (1980) attempted to advance environmental reconstruction but also of provenancing quartz
the discrimination of palaeoenvironmental modifications materials, whether by the study of inclusions (Comini et al.,
identifiable on a sand grain by considering palaeoenergy 2010) or by quartz-grain typing (Bull and Morgan, 2006) for
conditions. Semiround/semiangular grains impacting under both geomorphological (Moral-Cardona et al., 1997) and
aeolian transportation conditions were found to result in the forensic purposes (Bull et al., 2006).
formation of distinctive platelets on the surfaces of the grains. Sediment studies in the forensic realm require a subtle
These upturned plates, as they were termed, occur only in philosophical variance in interpretation of results compared
global sand sea conditions and to date remain a sole indicator to that normally exercised in the geomorphological literature.
of aeolian modification. Krinsley and Wellendorf considered In geomorphology and indeed the geological field, samples
the spacing between the platelets to be palaeoenergy indi- are compared and similarities and associations are identified
cators where these features were produced in a high velocity (similar strata, rock types, or bedding). This is in direct con-
(8–25 m s1) wind environment. Subsequent work however, trast to the forensic arena where samples are compared and
has identified the perfect replication of all of the features the only valid conclusion can be that the sample is different to
found in aeolian sand seas and has been able to ascribe energy another given certain geomorphological characteristics, or that
boundaries to these features (Newell et al., 2010). Similar re- a sample cannot be excluded from having derived from the
lationships exist for subaqueous transportation although of same source as another sample. Thus, the idea in forensic
course with different energy regimes given the viscosity of the sedimentology of achieving a ‘match’ between samples is not
fluid flow medium. possible (Morgan and Bull, 2007). It can be seen clearly,
This review has discussed primarily the use of the electron however, that the comparison of samples is in fact a com-
microscope to date with clastic sediments, notably quartz. parison of provenance. To this end, retrospective analysis of
Most other minerals are studied by electron microscopy util- electron microscope data sheets collected from forensic cases
izing the high magnification and high resolution imagery that in specific real life was undertaken in order to identify differ-
can be obtained. Specifically, clays and heavy minerals can be ent types of grains based on specific assemblages present of
viewed for environmental reconstruction purposes as well as quartz-grain surface textures. The database initially contained
for indicators of weathering and chemical alteration processes some 30 000 grains taken predominantly from England in
(Setlow, 1978). Most minerals however are either very sus- forensic locations and most certainly not associated with
ceptible to mechanical breakdown due to cleavage within the geological drift or hard rock lithologies. The results obtained
crystal or to biochemical attack such as in a soil. Perhaps the were extremely interesting and demonstrated that most sam-
most extreme alteration that can affect sand grains viewed by ples analyzed from natural environments (as opposed to in-
electron microscopy are the diagenetic alteration phases pro- cluding the anthropogenic additions from roads, buildings,
duced over geological time as a result of pressure solution (see etc.) contained no more than three grain types. At the first
Marshall, 1987a, b for extensive review). The effect of dia- instance, this would seem very surprising, as many would
genesis on quartz sand grains is normally the production of a imagine that a sample of soil or sand would represent a
smooth silica grain surface which obliterates all previous en- melange of features associated with the great variety of rock
vironmental grain history. If exacting shallow burial con- types and environmental and climatic processes that have
ditions prevail (Flesch and Wilson, 1974), then quartz occurred in the past 2 million years in Britain. What indeed
authigenic overgrowths can form on the grain surfaces pro- this very restricted assemblage of surface textures identified
ducing well-formed euhedral crystals. These can have specific was that soil and sediment, when eroded and transported,
provenance identifiers for sand grains which have been sub- move in temporally and spatially distinct bodies. Indeed the
sequently eroded from these diagenetic sandstones. Recent identification of five or six different grain types in a single
work has investigated the composition of the inclusions sample characteristically suggests to the analyst that they are
within the diagenetic quartz crystals by using ion beam an- dealing with materials of highly anthropogenic origin where
alysis (PIXE and PIGE) indicating yet further potential for mixing and the introduction of materials from many
260 The Scanning Electron Microscope in Geomorphology

provenances has taken place. The classic example of such an Comini, P., Morgan, R.M., Bull, P.A., Bailey, M., 2010. Provenancing of quartz
environment is that of a foot well in a car. Caution must grains using PIXE for forensic applications. International Proceedings of Particle
however be introduced into such interpretation as the pres- Induced X-ray Emission and its Analytical Applications. University of Surrey,
Guildford U.K. 27th June–2nd July 2010.
ence of multiple grain types in a sample is also highly char-
Culver, S.J., Bull, P.A., Campbell, S., Shakesby, R.A., Whalley, W.B., 1983.
acteristic of glacial moraines (Bull, 1984). Environmental discrimination based on quartz grain surface textures: a statistical
investigation. Sedimentology 30, 129–136.
Flesch, G.A., Wilson, M.D., 1974. Petrography of Morrison formation sandstone of
14.21.4 Conclusions the Ojito Spring Quadrangle, New Mexico. New Mexico Geological Society
Guidebook 25th Field Conference, Ghost Ranch, New Mexico, pp. 197–210.
Higgs, R., 1979. Quartz grain surface features on Mesozoic-cenozoic sands from
The SEM has been used in geomorphological investigations the Labrador and Western Greenland continental margins. Journal of
primarily as a high powered microscope. Its potential to pro- Sedimentary Petrology 49(2), 599–610.
vide focused, in-depth, high-magnification photography is Kanaori, A.Y., 1985. Surface textures of intra-fault quartz grains as an indicator of
invaluable and is utilized in almost every branch of geo- fault movement. Catena 12, 271–291.
morphology. The analytical bolt-on procedures that work in Krinsley, D.H., Margolis, S.V., 1969. Scanning electron microscopy: a new
method for studying sand grain surface textures. Trans. NY Acad. Science 31,
conjunction with the electron microscope such as backscatter,
457–477.
cathodoluminescence, edax, and microprobe provide the Krinsley, D.H., Takahashi, T., 1962. The surface textures of sand grains: an
more recent advances in analysis in geomorphology at this application of electron microscopy. Science 135, 923–925.
scale. The use, however, of the electron microscope in con- Krinsley, D.H., Wellendorf, W.G., 1980. Wind velocities determined from the surface
junction with other microanalytical techniques (such as ion textures of sand grains. Nature 283, 372–373.
Le Ribault, L., 1977. L’exoscopie des Quartz. Masson, Paris, 150pp.
beam analysis) indicates that this high powered microscope
Li, W., Wang, D., 1980. Surface morphology and textural studies of Apollo 17 mare
has potential for becoming a prerequisite in the development basalt by SEM. Geochemica 3, 442–446.
analysis of even submicron or proton levels of investigation. Mahaney, W.C., 2002. Atlas of Sand Grain Surface Textures and Applications.
Continual use of the electron microscope has occurred in Oxford University Press, Oxford, 237pp.
both academic research and applied technologies. Even casual Margolis, S.V., 1966. Electron microscopy of modern and ancient quartz sand
investigation of the literature will illustrate the immense grains. MS thesis, Florida State University, unpublished, 157 pp.
Margolis, S.V., 1968. Electron microscopy of chemical solution and mechanical
contribution the SEM has made to geomorphological and, abrasion features on quartz grains. Sedimentary Geology 2, 243–256.
more broadly, geological studies. The problem however, is not Marshall, J.D., 1987a. Diagenesis of sedimentary sequences. Blackwells, Oxford,
in the production of results, for which the electron microscope Geological Society Special Publication No. 36, 360 pp.
is eminently engineered, but in the operator interpretation of Marshall, J.R. (Ed.), 1987b. Clastic Particles: Scanning Electron Microscopy and
the results. Such interpretation can be hampered by the lack of Shape Analysis of Sedimentary and Volcanic Clasts. Van Nostrand Reinhod
Company, New York.
quantitative data, but more importantly, on the lack of ap-
Moral-Cardona, J.P., Gutierrez-Mas, J.M., Sanchez-Bellon, A., Lopez-Aquayo, F.,
preciation of the philosophical strictures in which the tech- Caballero, M.A., 1997. Provenance of multi-cycle quartz arenites of Pliocene age
nique is being employed. at Arcos, South Western Spain. Sedimentary Geology 112, 251–261.
Morgan, R.M., Bull, P.A., 2007. Forensic geoscience and crime detection.
Identification, interpretation and presentation in forensic geoscience. Minerva
References Medicolegale 127(2), 73–90.
Morgan, R.M., Robertson, J., Lennard, C., Hubbard, K., Bull, P.A., 2010. Quartz
grain surface texture analysis of sediments and soils from Canberra, Australia; a
Baynes, F.J., Dearman, W.R., 1978. The microfabric of a chemically weathered forensic reconstruction tool. Australian Journal of Forensic Sciences 42/3,
granite. Bulletin of International Association of Engineering Geology 18, 91–100. 169–179.
Bull, P.A., 1978. A statistical approach to scanning electron microscope analysis of Newell, A.J., Griffin, L.D., Morgan, R.M., Bull, P.A., Graham, G.A., 2010. Texture-
cave sediments. In: Whalley, W.B. (Ed.), SEM analysis in the study of sediments. based Estimation of Physical Characteristics of Sand Grains. Digital Image
Geo Abstracts, Norwich, pp. 212–226.
Computing: Techniques and Applications, 504–509.
Bull, P.A., 1984. Scanning electron microscope studies of sediments from a lower
Robinson, D.A., Williams, R.B.G., 2000. Accelerated weathering of a sandstone in
Palaeolithic hominid site in Wales. In: Green, H.S. (Ed.), Pontnewydd Cave.
the High Atlas Mountains of Morrocco. Zeitschrift fur Geomorpholgie 44,
National Museum of Wales Press, Cardiff, pp. 76–87, 227.
513–528.
Bull, P.A., Morgan, R.M., 2006. Sediment Fingerprints: a forensic technique using
Setlow, L.W., 1978. Age determination of reddened coastal dunes in north-western
quartz sand grains. Science and Justice 46(2), 107–124.
Bull, P.A., Parker, A.G., Morgan, R.M., 2006. The forensic analysis of soils and Florida, USA by the use of the SEM. In: Whalley, W.B. (Ed.), SEM in the Study
sediment taken from the cast of a footprint. Forensic Science International 162, of Sediments. Geo Abstracts, Norwich, pp. 283–286.
6–12. Smart, P., Tovey, N.K., 1988. Electron Microscopy of Soils and Sediments, Volume
Bull, P.A., Whalley, W.B., Magee, A.W., 1985. An annotated bibliography of 1: Techniques. OUP, Oxford.
environmental reconstruction by SEM 1962–1985. Technical Bulletin no. 35 the Vincent, P.J., 1976. Some periglacial deposits near Aberystwyth, Wales, as seen
BGRG. Geobooks Norwich, UK. 94pp. with a scanning electron microscope. Biuletyn Peryglacjalny 25, 59–64.
 The Scanning Electron Microscope in Geomorphology 261

Biographical Sketch

Peter Bull (MSc., PhD, MA(Oxon)) is a Fellow and Tutor at Hertford College and a lecturer at the University of
Oxford Centre for the Environment and has held these positions in Oxford for more than 30 years. Published
research interests include karst studies, sedimentology, environmental reconstruction (both geologic and foren-
sic), and more recently forensic geoscience.

Ruth Morgan (MA (Oxon), DPhil) is the Director of the University College London Jill Dando Institute Centre for
the Forensic Sciences and a lecturer at the UCL department of Security and Crime Science. Her research is focused
on the role of physical evidence in the detection of crime.
She is the corecipient of the P W Allen Award 2007 and a committee member of the London Geological Society
Forensic Geoscience Group, Member of the UCL Institute of Judicial Studies and on the advisory board of Inside
Justice, a not-for-profit organization funded by charitable donations from the Esmee Fairburn Foundation.
 14.22 Determining Organic and Carbonate Content in Sediments
DM Kennedy and JLD Woods, The University of Melbourne, Parkville, VIC, Australia
r 2013 Elsevier Inc. All rights reserved.

14.22.1 Introduction 262

14.22.2 Basic Analytical Principle 263
14.22.3 Measurement Methodologies 264
14.22.3.1 Total Carbon Analysis 264
14.22.3.1.1 Loss on ignition 264
14.22.3.1.2 High-frequency induction furnace 264
14.22.3.1.3 Coulometry 265
14.22.3.2 Inorganic Carbon Analysis 266
14.22.3.2.1 Gravimetric analysis via acid digestion 266
14.22.3.2.2 Pressure calcimeter (vacuum-gasometric technique or carbonate bomb) 267
14.22.3.2.3 Titration 267
14.22.3.2.4 Acid digestion in a laser particle sizer 268
14.22.3.3 Organic Carbon Analysis 268
14.22.3.4 Removing Inorganic Carbon 268
14.22.3.4.1 Dichromate oxidation of organic carbon 269
14.22.3.5 Organic Matter Analysis 269
14.22.3.6 Lightness of Sediment 269
14.22.4 Summary and Conclusions 271
References 271

Abstract

Total carbon within soils and sediments is composed of organic and inorganic forms. Humus and other plant remains
constitute the organic carbon fraction, whereas carbonate minerals, mainly calcite and aragonite, comprise the inorganic
fraction. There are three main approaches (techniques) for measuring these components in geomorphology: physical,
chemical, and relative. The choice of technique depends on the accuracy required for the study. Many of the techniques are
now automated, thus allowing for rapid quantification of both organic and carbonate components of materials in the
landscape.

14.22.1 Introduction entirely of carbonate grains, whereas peat swamps are com-
posed entirely of the organic remains of plants. The degree to
Carbon in soils and sediments is comprised of organic and which these materials are seen as contaminates, or proxies,
inorganic forms. The organic fraction is largely the product of varies widely depending on the scope and time frame of study.
plant decay, whereas the inorganic fraction generally occurs as For example, in deep-ocean paleoenvironmental studies, the
carbonate minerals (Nelson and Sommers, 1996). The an- composition of carbonate material has been used to estimate
alysis of carbonate and organic minerals and elements within deep-sea sedimentation rates over glacial–interglacial cycles
geomorphology is an old, yet fundamental, methodology. The (e.g., Iberian Margin, Atlantic, Thomson et al., 1999; Central
removal of these materials is commonly an initial step in Pacific, Anderson et al., 2008), or as a proxy for the input of
cleaning sediments for basic analyses such as grain size ice-rafted debris (e.g., North Atlantic, Hemming et al., 2002).
measurement, or for more complex isotopic studies using Analysis of the isotopic chemistry of specific carbonate grains
mass spectrometers. Carbon is also of scientific interest in its (namely foraminifera) is now standard for paleotemperature
own right, especially for research on paleoenvironmental studies, as is the measurement of the magnesium (Mg)/cal-
change. In some geomorphic environments, carbonate and cium (Ca) ratios in these grains (e.g., Burdige, 2006). Organic
organic materials may account for all the sediments present in material, on the other hand, is an essential component of soils
a landform. For example, coral atolls are commonly made and as a result significant work has been conducted on the
organic component of soil chemistry (e.g., Fitzpatrick, 1986).
Organic levels in deep-ocean cores have been used as proxies
for past oceanic productivity (e.g., Pacific Ocean, Kawahata
Kennedy, D.M., Woods, J.L.D., 2013. Determining organic and carbonate
content in sediments. In: Shroder, J. (Editor in Chief). Switzer, A.D.,
et al., 2009; Martinez and Robinson, 2010); similar work
Kennedy, D.M. (Eds.), Treatise on Geomorphology. Academic Press, San is also conducted in lacustrine environments to trace both
Diego, CA, vol. 14, Methods in Geomorphology, pp. 262–273. nutrient levels (e.g., Himalayas, Choudhary et al., 2009;

262 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00389-4

 Determining Organic and Carbonate Content in Sediments 263

Argentina, Guilizzoni et al., 2009; Canada, Buhay et al., 2009) 14.22.2 Basic Analytical Principle
and the degradation of the lake’s surrounding catchments
(e.g., Italy, Borgatti et al., 2007; Uganda, Russell et al., 2009; Several methods exist for determining the content of carbon-
Turkey, Dengiz et al., 2009). ates and organic material in sediments. These can be sub-
Minerals containing carbonate occur in a variety of divided into two main methodologies: those which directly
forms. Calcium carbonate (CaCO3) in the form of either measure the amount of material present and those which
aragonite or calcite is generally precipitated in organic form by estimate the composition based on proxy characteristics.
biota in aquatic environments (Lewis, 1984; Tucker and Direct measurement involves either the physical counting
Wright, 1990). Calcium carbonate can also be directly pre- of grains (e.g., Milliman, 1974), X-ray techniques, or the
cipitated from solution either in the open ocean (e.g., ooids, generation of carbon dioxide. The two former methods are
Tucker and Wright, 1990) or as cements (e.g., beach rock, generally time consuming and labor intensive and, as a
Vousdoukas et al., 2007). In soils, calcite is often the dominant result, methods focused on the generation of carbon dioxide
carbonate in active pedogentic environments. Because of cal- are more commonly used in geomorphology. Proxy meas-
cite’s relatively high solubility, it strongly influences a range of urements such as sediment brightness estimate the proportion
soil properties – due to the mineral’s alkalinity and pH buf- of carbonate or organic material present through light
fering properties (Loeppert and Suarez, 1996). Calcite can reflectance from a sediment surface, which is converted into
exist in a variety of forms within soils, from nodule centi- a bulk-percentage composition. This method is becoming
meters in size to submicrometer crystals (Loeppert and Suarez, commonly used, especially with greater availability of light
1996). In marine environments, calcite commonly occurs in scanners; however, the measurements must still be calibrated
foraminifera, echinoids, and coralline algae, and is divided with some direct measures.
into high- and low-Mg types based on the concentration of The fundamental underlying principle of direct measure-
MgCO3 in solid solution (44% is high Mg-calcite, and o4% ment of carbon through carbon dioxide generation is oxi-
is low Mg-calcite) (Bathurst, 1976). Aragonite is a dation with an acid or ignition at high temperatures:
metastable form of calcite and rarely contains up to 1.5 mol.%
magnesium. The mineral occurs in scleractinian corals, bi- CO3 2 þ H1 -CO2 þ OH ½1
valves, and gastropods; it is most often recrystallized to calcite
in Pleistocene age sediments (Bathurst, 1976). In older sedi- The methodologies used to generate carbon dioxide depend
ments, carbonates commonly occur in other mineralogical on whether the analysis is targeting total carbon, inorganic
forms such as dolomite (Ca,Mg(CO3)2), which has a diagenic carbon (calcite), organic carbon, or organic matter. Methods
origin and is composed of alternating layers of calcium and for the analysis of total carbon involve either wet combustion
magnesium carbonate. In older rock units, some carbonates or dry combustion, and the carbon dioxide liberated is de-
occur as economically valuable ore minerals, such as azurite termined through gravimetric, volumetric, spectrophoto-
(Cu3(OH)2(CO3)2), cerussite (PbCO3), and malachite metric, or conductimetric means. Gravimetric and volumetric
(Cu2(OH)2(CO3)) (Lewis, 1984). This chapter focuses on the methods are described here as they are more widely used due
calcium- and magnesium-bearing carbonates (calcite and to their simplicity; alternatively, total carbon can be reduced to
aragonite), as these are the most commonly encountered ones methane gas and analyzed via gas chromatography.
in geomorphology and those which receive the most scientific For inorganic carbon, once carbon dioxide is liberated
attention. from the sample either the volume of carbon dioxide released
Organic carbon consists of the cells of microorganisms; is directly measured or its rate of generation recorded through
plant and animal remains in various stages of decomposi- the increase in pressure measured in a vacuum. This rate of
tion; stable humus; and highly carbonized compounds generation is then correlated to the amount of carbonate
such as charcoal, graphite, and coal (Nelson and Sommers, present in the sample. An alternative and simpler method is
1996). Swift (1996) defined soil organic matter as the nonliv- based on the assumption that on reaction with acid the re-
ing organic material in the soil, whereas the living maining hydroxide ion will form water and the calcium or
organic components are termed ‘soil biota’. Organic matter is magnesium ions (depending on acid used) are soluble and
also composed of the organic fraction of the sediment, as therefore can be removed from the sample by washing. The
outlined above; however, its mass is the combined weight of all change in weight before and after treatment represents the
elements present in the organic component and not carbon amount of carbonate that was present in the original sample.
alone. Organic matter has been estimated by multiplying the For example, on reacting hydrochloric acid with calcium car-
organic carbon concentration by a constant factor; however, bonate, carbon dioxide and soluble calcium chloride is pro-
this factor must be verified by a direct measure (Nelson and duced along with water:
Sommers, 1996). There are many techniques for the charac-
terization of each component of the organic fraction in sedi- CaCO3 þ 2 HCl-CO2 þ CaCl2 þ H2 O ½2
ments (see Swift, 1996); however, for simplicity in this chapter,
the organic carbon is treated as a single entity, as the methods The techniques for the analysis of organic carbon are less
described subsequently do not concisely differentiate between straightforward, as a simple oxidation reaction to dissolve the
different types of organic remains. For in-depth reviews on the material is commonly not feasible. Many methodologies
types of organic material present in soils and sediments, the measure organic carbon indirectly, using the premise that
reader is referred to Hayes and Swift (1978) and Stevenson
(1994). total carbon ¼ inorganic carbon þ organic carbon
264 Determining Organic and Carbonate Content in Sediments

This can be done by either removing the inorganic carbon 14.22.3.1.1 Loss on ignition
component and undertaking a total carbon determination or In the LOI technique (Box 1), the sample is heated incre-
analyzing total carbon and inorganic carbon and solving the mentally in a furnace to allow the organic and carbonate
equation for organic carbon. The alternative method for materials to react with oxygen creating carbon dioxide (CO2).
analyzing organic carbon is oxidation by dichromate (Cr2O7). The weight change resulting from the liberation of CO2 is
The Walkley–Black (Walkley and Black, 1934) method for correlated to the proportion of carbonate, organic matter, and
estimating organic carbon is widely used as it is simple with total carbon in the sediments. The temperatures at which
minimal equipment required, even though it does not pro- carbonates, organics, and water are liberated from the sample
duce quantitative results. The methods used for estimating are selected using a temperature curve (Figure 1). The mois-
organic matter can be used as a proxy for organic carbon, of ture content or the interstitial water of a sample is said to be
which the most common methods are generation of carbon liberated at 105 1C (Pansu and Gautheryrou, 2006). This
dioxide by either oxidation with hydrogen peroxide or loss on temperature has been derived empirically, and although
ignition (LOI). However, organic matter can only be at best heating to a higher temperature would more definitely remove
estimated from these methodologies; hence, a measure of the water content, a higher temperature can structurally alter
organic carbon would more precisely indicate the organic other mineral components of the sediment (Figure 1). The
carbon fraction. sediment is then heated at specific temperatures to remove
Techniques involving the oxidation of a sample to deter- organic matter and carbonates. Specifically a mass of sediment
mine the composition of carbonate and organics are probably is dried at 105 1C to liberate all water and then heated at
the most simple. Oxidation of the elements can be completed 400 1C for 24 h and then at 1000 1C for one hour more (Galle
either by using acid or by heating. The proportion of carbon or and Runnels, 1960; Dean, 1974; Mooers, 1999). Within the
carbonates in the sample is calculated through the change in temperature range 300–500 1C, organic matter is oxidized to
weight in the dried sample before and after processing. While carbon dioxide:
using these techniques, it is important that the entire sample
has been oxidized before the final weight is taken. The re- CH2 O þ O2 -CO2 þ H2 O ½3
action time will be dependent on the amount of sample used
and also the size of the organic and carbonate components. It Although organic matter is fully decomposed at 500 1C, at this
is therefore generally suggested that sediments and soils are temperature some types of minerals are also combusted
completely disaggregated prior to processing and, if too (Figure 1) (Pansu and Gautheryrou, 2006). The trade-off is to
coarse, they are ground to a powder to ensure the entire heat the sample to 400 1C for a longer period to ensure
sample is reacted during processing (Blakemore et al., 1987; combustion of organic matter but not inorganic carbon.
Sparks, 1996). At 1000 1C combustion oxidizes carbonate through

CaMgðCO3 Þ2 þ O2 -2 CaO þ 2 MgO þ 2 CO2 ½4

14.22.3 Measurement Methodologies

The weight loss measures assume that all the CO2 production
is the result of the breakdown of the target minerals at their
The wide range of methodologies that are available for the
respective oxidation temperatures. This assumption ignores
calculation of carbonate and organic contents requires the
the dehydration and dehydroxylation of other mineral con-
researcher to carefully select the accuracy they require as well
stituents within the sediment (Figure 1). Most clay minerals
as the possible forms of carbon present in their sample. Grain
contain 5% or less OH water in their lattice structure that is
size is a major factor that needs to be considered when con-
also liberated during ignition. Experimentally it has been
ducting analyses. Fine-grained microscopic grains will react at
shown that lake sediments with no carbonate, but high clay,
a different rate to coarse sand or gravel-sized clasts. Such a
show a 3–4% weight loss between 550 and 1000 1C (Dean,
variation in reaction times means some forms of chemical
1974).
analyses are inappropriate, as the time of reaction for the
For carbonates, the weight loss resulting from the liber-
larger grains is too long.
ation of CO2 does not account for the entire weight of the
CaCO3. It is therefore necessary to divide the weight lost by
0.44 (the fraction of CO2 in CaCO3) (Dean, 1974). For or-
14.22.3.1 Total Carbon Analysis
ganic matter, conversion to organic carbon can be achieved by
Analysis of total carbon gives the amount of both organic and a correction factor, but only if the organic carbon content for
carbonate carbon forms, by converting all carbon to carbon that sediment or sediment type is known.
dioxide. The simplest and cheapest laboratory technique
involves turning the sample to ash in a muffle furnace 14.22.3.1.2 High-frequency induction furnace
(LOI). This methodology is less accurate due to the possibi- Total carbon can also be measured by igniting the sample and
lity of loss of structural water from minerals during the heating measuring the amount of carbon dioxide that has evolved
process. High-frequency induction furnaces generally have through its absorption in hydroxide. This technique described
greater accuracy, but require a specific apparatus. Many in- by Blakemore et al. (1987) utilizes elements of furnace com-
duction furnaces are commercially available, some of which bustion to liberate CO2 with parts of the titration procedure
are automated. These generally measure total carbon where the combustion products are collected in KOH solu-
gravimetrically. tion. Depending on the amount of pretreatment, this
 Determining Organic and Carbonate Content in Sediments 265

Box 1 Loss on ignition (LOI) or muffle furnace method. Adapted from Galle and Runnels (1960), Dean (1974) and also the
chapter by Nelson and Sommers (1996) in the SSSA BOOK SERIES 5: Methods of soil analysis, and originally described in
Ben-Dor and Banin (1989)

Step 1: Take a pyrex beaker, crucible, or soil moisture can that is lidded. Heat for 2 h at 400 1C. Cool in a dessicator.
Step 2: Weigh crucible.
Weight (a)
Step 3: Add 1–3 g of crushed or powdered sample and dry at 105 1C for 24 h with the lid uncovered from the sample.
Step 4: Remove crucible, replace lid and cool in a dessicator with pressure release valve. When cool weigh crucible and sample.
Weight (b)
Step 5: Place crucible in muffle furnace at 400 1C for 16 h.
Step 6: Remove sample and cool in a desiccator. Once cool weigh crucible and sample.
Weight (c)
Step 7: Return crucible to furnace and heat to 1000 1C for 1 h.
Step 8: Remove sample and cool to room temperature in a desiccator. Once cool, weigh crucible.
Weight (d)
Calculations:
Weight of oven dried sample (W105) ¼ b  a
Weight of ignited sample (W400) ¼ c  a
LOI% ¼ W105WW105
400
 100
Organic matter content is assumed to equal the LOI%.
Weight of incinerated sample (W1000) ¼ d  a
CO3 % ¼ W105 ðWW400
105
W1000 Þ
 100C 0:44

Organic matter
Gypsum (CaSO4.2H2O)
Dolomite (CaMg(CO3)2)
Calcite/aragonite (CaCO3)
Gibbsite(Al(OH3))
Magnetite (Fe2O)
Goethite (FeO-OH)
Allophane
Chlorite
Illite-micas
Smectities
Kaolinite

0 100 200 300 400 500 600 700 800 900 1000
Temperature (°C)

Legend

Minerals which undergo both

dehydrtion and dehydroxylation Dehydration temperature Dehydroxylation temperature
(including decarbonation and
desulfurization reactions)
Dehydration temperature Dehydroxylation temperature
range range

Dehydrxylation temperature range

Minerals which only undergo
(including decarbonatation and
dehydroxylation
desulfurization reactions)

Figure 1 Temperature ranges at which certain common minerals undergo dehydration and dehydroxylation. Data based on Pansu, M.,
Gautheryrou, J., 2006. Handbook of Soils Analysis. Mineralogical, Organic and Inorganic Methods. Springer, Berlin, Heidelberg.

technique can be used to analyze total carbon or either the The remaining gases are then passed into a calibrated burette
inorganic or organic components individually (depending on in which 5% H2SO4 is displaced. The gases are flushed
which is removed prior to heating). A high-frequency in- through concentrated KOH that absorbs the CO2. The differ-
duction furnace heats the sample to 1400 1C and then purifies ence between the original volume in the burette and the
the carbon dioxide produced through several physical and CO2-free gas is equal to the amount of CO2 in the sample
chemical steps: (Blakemore et al., 1987).

CO2 -dust trap-converter furnace of CO to

14.22.3.1.3 Coulometry
Coulometry is an absolute method of carbon measurement that
CO2 -MnO2 ðto remove SO2 Þ is based on Faraday’s law and therefore does not require the use
266 Determining Organic and Carbonate Content in Sediments

of reference standards (Lingane, 1954). In addition, the pre- 2. the sample is agitated until effervescence no longer occurs,
cision of modern electronics means coulometry is a very accurate even after the addition of further acid;
process that is now automated. The automation process means 3. the sample is then washed in freshwater repeatedly to re-
the samples can be measured very accurately, by either chemical move the reacted acid solution, with special care to ensure
reaction or combustion, and processed efficiently. As a result, no material is lost during this process – the reacted fluid
this method has now become standard for many researchers may be filtered during this process;
(e.g., Thomson et al., 1999; Murray et al., 2000; Chang and Lee, 4. the remaining material (and any caught during the filtering
2002; Anderson et al., 2008) and can be used for concurrently process) is dried and weighed; and
analyzing organic carbon, carbonate, and total carbon in sedi- 5. the change in weight between steps (1) and (4) is the as-
ments (Schultz et al., 1999). The range of commercial machines sumed % CaMgCO3 in the sample.
available to conduct coulometric analyses means that the pre-
Often, this method is the first step in the cleaning of
cision and accuracy vary between models. In addition, the rapid
sediments for other laboratory analyses where the carbonates
advances in modern computing means that the cost per sample
will act as contaminates for further analytical procedures
changes as the quality of the machines increases.
(Brewer, 1964) such as for diatom studies of lake sediments
(e.g., Battarbee et al., 2001; Morley et al., 2004). If the con-
14.22.3.2 Inorganic Carbon Analysis centration of the HCl used is too strong then the acid may also
dissolve minerals such as apatite and affect the properties of
Inorganic carbon or carbonates are most often analyzed via others such as biotite (Brewer, 1964). As a result, sodium
acid digestion, with measurements of carbon dioxide gener- acetate has also been used as it has little effect on primary
ation or weight loss. They can also be measured using cou- minerals though is ineffective in dissolving coarse crystals
lometry or LOI (see above). Acid digestion (most commonly (Jackson, 1958). Acetic acid is also a common weaker reagent
using sulfuric acid) can also be used to eliminate inorganic used, such as for investigating Heinrich layers in the North
carbon prior to an analysis of total carbon. Thus, the amount Atlantic (Hemming et al., 2002) and dilute nitric acid has also
of organic carbon can be calculated. been employed for studies of dust accumulation in the Tas-
man Sea (Hesse, 1994). In the case of both reagents, if there is
a high concentration of clays, material may be lost when de-
14.22.3.2.1 Gravimetric analysis via acid digestion
canting the fluid prior to final weighing. Filtering the solution
As summarized in eqn [1], this method involves the quanti-
during washing or flocculating the suspended material with
fication of carbonate content through the production and
sodium chloride can facilitate the removal of all the acid when
measurement of H þ or Ca/Mg or CO2. Methods that involve
the clay size material is to be kept (Brewer, 1964). An alternate
the measurement of CO2 are generally preferred as the
and easier way is to centrifuge the sample prior to decanting
production of CO2 is related to the absolute value of car-
the liquid.
bonate or organics (Loeppert and Suarez, 1996; Nelson and
Method 2: CO2 loss. In this method (U.S. Salinity Soil Lab,
Sommers, 1996). In using the weight loss (gravimetric)
1954; Allison and Moodie, 1965), a set volume of acid and
method for carbonate, it is assumed that the difference in
water is reacted with a specific weight of sample. The entire
weight before and after acid treatment is a function of the
material (flask, acid, and sample) is weighed prior to, and
amount of carbonate present in the sample. Hydrochloric acid
after, reaction. After reaction, the container is opened to allow
is most commonly used (Loeppert and Suarez, 1996) and
the CO2 to escape prior to final weighing. The weight loss
either of the two methods: (1) the total weight loss resulting
recorded is then proportional to the amount of carbonate
from the entire treatment or (2) that directly related to CO2
present in the sample (Box 2).
loss is equated to the proportion of carbonate in the sediment.
In some cases, trichloroacetic acid is used as a reaction
Method 1: Total (dry) weight loss. The details of method 1 is
agent with magnesium perchlorate used in U-tubes which
given in the following:
transfer gases to and from the reaction vessels (Erikson and
1. for total weight loss, the entire sample is initially washed in Gieseking, 1947; Hesse, 1971).
freshwater to remove dissolved salts prior to treatment with Methodological problems. As with all methods of chemical
3 mol (or 10%) HCl; analysis, which do not directly measure the material in

Box 2 Gravimetric method for loss of carbon dioxide. From Loeppert and Suarez, 1996

Step 1: Weigh (to nearest 0.1 mg) a stoppered 50 ml flask containing 10 ml of 3 M HCl (other stoppered containers may be used).
Step 2: Precisely weigh between 1 and 10 g of dried sediment/soil (containing 0.1–0.3 g of CaCO3 equivalent). Add to the flask slowly to prevent
excess frothing.
Step 3: After effervescence has ceased insert stopper and swirl flask occasionally for 15 min.
Step 4: At intervals of 15 min remove stopper and swirl container to allow CO2 to escape (takes 10–20 s).
Step 5: Weigh container at nearest 0.1 g.
Step 6: Repeat (4) and (5) until weight change is no more than 1–2 mg. Total reaction time is generally around 1 h.
Calculations:
 
%CaCO3 ¼ gCO 2 lost
gsoil  2:273
 Determining Organic and Carbonate Content in Sediments 267

Box 3 Vacuum-gasometric technique. From Jones and Kaiteris, 1983; Dunn, 1980

Step 1: Crush samples to 4 j (o63 mm) and dry at 110 1C

Step 2: 350–400 mg of sample is weighed at an accuracy of 70.2 mg and placed in the bottom of the reaction chamber. 3 ml of concentrated
(85%) phosphoric acid is placed in the side arm of the reaction chamber.
Step 3: The reaction chamber cap is sealed and connected to the pump where the air is evacuated. The pressure before reaction is measured.
Step 4: The reaction chamber is separated from the pump and the acid tipped into the sample. The reaction is allowed to proceed for a set time
(e.g., 1.5 h) with the sample agitated regularly (e.g., 20 min).
Step 5: After the desired reaction time is achieved, the reaction chamber is stabilized at the temperature the instrument was calibrated at.
Step 6: The reaction chamber is reconnected to the pressure system (which is still a vacuum) and the pressure recorded in the reaction chamber.
Step 7: The difference between the initial pressure and the final pressure in the CO2 gas pressure.
A standard of analytical grade calcium carbonate should be run for each set of analyses.
Calculations:
CO2 gas pressure ¼ Final pressure  Initial pressure
Pressure
ðsampleÞ
%CaCO3ðsampleÞ ¼ PressureðstandardÞ  100

question but rely on the measurement of reaction products, The method involves the addition of sample and acid to a
several errors can occur when material other than the target reaction vessel where the two materials are initially separated
minerals react during processing. This is particularly import- (Figure 2). The apparatus is connected to a vacuum line where
ant where the concentration of carbonate is quite low (o5%) all the air is removed from the reaction vessel. The acid is then
and fall within the error range of the methodology (e.g., for added to the sample and the change in pressure measured. The
carbonate analyses CO2 may be liberated from organic ma- difference between the initial pump down pressure and the
terial during the reaction (Loeppert and Suarez, 1996)). As a final pressure is the CO2 gas pressure, which is then converted
result, weak acids are recommended for use and errors related to a proportion of CO3 in the sample. Older methodologies
to the decomposition of organic material can be reduced by have utilized water-filled, wide-arm U-tubes as pressure
the addition of 3% FeCl2 (Loeppert and Suarez, 1996). In measures; however, the setup is more complex than the one
addition, other components such as clays are also dissolved in used by Presley (1975), and additional corrections for at-
the acid solution (Dean, 1974). As a result of these errors, mospheric pressure need to be made (Hesse, 1971). The most
other methods such as a pressure calcimeter, coulometric, and important step in the operation of a pressure calcimeter is the
titration are used to provide a more direct measurement of the grain-size analysis (Jones and Kaiteris, 1983). Grains coarser
carbon content. These methods are particularly employed for than 4j (63 mm) commonly do not react completely, which
fine-grained materials, those where little sample is available requires samples to be crushed to o4j.
and for analyses where the precise composition of the sedi- The results obtained from the system of Jones and Kaiteris
ment is critical for the study. (1983) are reproducible to 73% when the system is uncali-
brated, but when calibrated the precision is improved to
70.25%. Calibration is achieved through
14.22.3.2.2 Pressure calcimeter (vacuum-gasometric
1. Correcting for the vessel volume. As each system will have a
technique or carbonate bomb)
different gas volume, it is necessary to know the gas volume
A more direct technique is a pressure calcimeter that directly
for known weights of carbonate. Powered optical grade
measures the pressure of the CO2 released on reaction with
calcite is a good test sample to run.
acid (Box 3). Such techniques react a certain weight of sample
2. Temperature. Just as pressure is normalized to a reference
with acid and record the change in pressure created by the
volume, all pressures are normalized to a specific
production of gas. The change in pressure is converted to
temperature.
the amount of carbonate in the sample by comparison with
3. Gas solubility. This correction relates to Henry’s law, which is
the pressure evolved from the reaction of a known concen-
a function of the different solubility of gases at different
tration of CaCO3 (Dunn, 1980). This method thus assumes
pressures. For example, a small and large amount of pure
that the only inorganic carbon present in the sample is in the
calcite should both result in carbonate measurements of
form of calcium carbonate (not magnesium carbonate).
100%; however, the actual readings may vary by several
A variety of different methods exist and their cost is highly
percent. Gas solubility is particularly important for appar-
variable and methodology complex (e.g., rapid-gasometric
atuses that are run at atmospheric pressure rather than in a
technique of Hulsemann (1966); Karbonate-Bombe of Muller
vacuum.
and Gastner (1971)). The method described here is the vac-
uum-gasometric system of Jones and Kaiteris (1983), which is
the more precise and quickest of these techniques and has
been used widely for marine core studies (e.g., Carter et al., 14.22.3.2.3 Titration
2000, 2002). This technique is based on Presley (1975) but is A further technique that relies on the liberation of CO2 is
redesigned to be run at room temperature without the need titration. In this method, the amount of CO2 evolved from the
for freeze baths for the acid. reaction of the sample with an acid is collected in a
268 Determining Organic and Carbonate Content in Sediments

air filtered through a hydroxide solution to remove carbon

Pressure meter dioxide and then collected in a second solution of hydroxide.
It is this second solution that is used in the titration procedure
(Figure 3). Piper (1942) developed a quick titration method
where the reacted solution of acid and sample is titrated,
which, although less accurate, is less complicated than col-
lecting the carbon dioxide. Rather than directly titrating the
Valves reacted solution, it is possible to determine the carbonate
content using a pH meter. In this method, the sample is
reacted with acetic acid and the change in pH equated to a
proportion of carbonate present (Loeppert et al., 1984; Moore
p
um

et al., 1987).
To p

One problem with titration methodologies is that the

(a) (b) titrant must be standardized regularly, often daily (Huffman,
1977) – this adds significant time to the analytical procedures
as well as the consumption of reagents.

14.22.3.2.4 Acid digestion in a laser particle sizer

Valves A relatively new analytical technique for measuring carbonate
involves the combined use of a laser particle-size machine,
such as Malvern Mastersizer or Coulter Counter and hydro-
chloric acid. In this technique proposed by Trentesaux et al.
Clamp (2001), a sample is continuously run within the machine in
freshwater at first and then in an acid solution. The change in
obscuration as the carbonate is dissolved can be converted
into a composition of carbonate within the sample. Trente-
saux et al. (2001) found that this can be accurate when the
carbonate content of the sample is between 15% and 60%. In
Reaction chamber
their system, they noted that the acid did not appear to
damage the machine and it was also theoretically possible for
the same methodology to be used for organics utilizing per-
oxide to digest the material. It is, however, strongly recom-
(b) mended that the manufacturer of the machine be consulted
before attempting to add acids or oxides to the fluid as
Figure 2 (a) An example of a pressure calcimeter based on the irreparable damage could be caused. It is likely that the po-
design of Jones and Kaiteris (1983). A and B represent valves tential damage to machines means this technique is not
separating the vacuum and reaction chamber (b) with the pressure widely used by researchers.
meter. (b) The reaction chamber in which the sample and acid are
mixed.
14.22.3.3 Organic Carbon Analysis

standardized hydroxide solution (commonly NaOH or KOH), Organic carbon can be determined in three ways. Using the
followed by back-titration of the excess hydroxide (Jackson, equation
1958). The reactions involved are (Loeppert and Suarez, 1996)
total carbon ¼ inorganic carbon þ organic carbon ½9
2Kþ þ 2 OH þ CO2 -2Kþ þ CO3 2 þ H2 O ½5
it is possible to: (1) analyze total carbon and inorganic carbon
An aliquot of the hydroxide solution is first titrated with HCl and solve to find the organic component; (2) remove the in-
to the phenolphthalein endpoint by organic carbon and then analyze for total carbon; or (3)
Kþ þ OH þ Hþ -Kþ þ H2 O ½6 perform wet oxidation with dichromate. Wet oxidation using
the Walkley–Black method is widely used as it is inexpensive
and relatively simple, even though it gives a variable recovery
Kþ þ CO3 2 þ Hþ -Kþ þ HCO3  ½7 of organic carbon (Nelson and Sommers, 1996).

and then to the bromcresol green endpoint where the in- 14.22.3.4 Removing Inorganic Carbon
organic carbon content is proportional to the HCl consumed:
Inorganic carbon can be successfully removed prior to analysis
Kþ þ HCO3  þ Hþ -Kþ þ H2 O þ CO2 ½8 of total carbon via wet combustion, but removal prior to dry
combustion (high-frequency induction furnace) is less satis-
In this process, the carbonate sample is placed in a reaction factory (Nelson and Sommers, 1996). This presents a difficulty
chamber where it interacts with acid. The vessel is flushed with as the most accurate method for inorganic removal is thus
 Determining Organic and Carbonate Content in Sediments 269

HCl

Air
Glass wool
Valves

Vacuum

Reaction vessel
(HCl + sample) KOH

KOH

Figure 3 A schematic of a standard titration setup. Based on Hesse, P.R., 1971. A Textbook of Soil Chemical Analysis. John Murray, London.

coupled with the more time-consuming and technically dif- 14.22.3.5 Organic Matter Analysis
ficult analysis for total carbon. The wet-combustion method
Organic matter can be analyzed either by the LOI method or
for total carbon is not described in this chapter as it is rarely
by oxidation and the generation of carbon dioxide. Reaction is
used in geomorphology for these reasons.
usually with hydrogen peroxide or bleach, and is measured
To remove inorganic carbon prior to analysis in a high-
gravimetrically (as per method 1). Hydrogen peroxide (H2O2)
frequency induction furnace, the sample is reacted with weak
is commonly recommended for the removal of organic ma-
sulfurous acid, and then heated to remove any remaining acid.
terial as a pretreatment process for other analyses (e.g., Lewis,
The process of heating the sample can cause decomposition of
1984). Robinson (1927) originally described an H2O2 method
some organic carbon. Further, dolomite may only be in-
for estimating organic content; however, it has serious limi-
completely removed using this method (Nelson and Som-
tations as the degree of oxidation of organic matter is often
mers, 1996). The lengthy procedures, apparatus, and expertise
incomplete and therefore carbon values will vary between
required for the first two methods have led to the continued
different soil types (Nelson and Sommers, 1996). Incomplete
use of the Walkley–Black method and it is one of the most
oxidation can also occur as a result of sodium chloride oc-
widely used methods for analysis of organic carbon.
curring within the sample, as chloride ions are oxidized to
chlorine, and hence it is not appropriate for saline samples
14.22.3.4.1 Dichromate oxidation of organic carbon (Reitemeier, 1943). Common bleach (NaOH) is also used as a
This technique is based on the principle that organic matter reagent for removing organic material (e.g., Gaffey and
within soils can be oxidized through use of a hot mixture of Bronnimann, 1993; Rapp et al., 2007).
potassium dichromate (K2Cr2O7) and sulfuric acid (H2SO4) In their study of carbonate minerals from modern reef
(Nelson and Sommers, 1996) (Box 4). The technique is environments, Gaffey and Bronnimann (1993) found that
commonly known as the Walkley–Black method, which is both NaOH and H2O2 caused some dissolution and etching
based on Walkley and Black (1934) and Walkley (1947) and of grains and suggested that 5% NaOCl (full-strength com-
is a colorimetric method for the determination of carbon mercial Clorox) was best suited for removing organic material
within sediments (Blakemore et al., 1987). After the reaction, with minimum impact on the carbonate grains.
the solution is titrated with an iron indicator (e.g.,
Fe(NH4)2(SO4)2  8H2O) and the Cr2 O7 2 reduced during
reaction with the soil as equivalent to the amount of organic
14.22.3.6 Lightness of Sediment
carbon in the sample (Nelson and Sommers, 1996). Alter-
natively, the solution can be measured through a spectro- An increasingly popular technique for the measurement of
photometer at 600 nm (Blakemore et al., 1987). It has been carbon, especially in fine-grained deep marine cores, is the
suggested that a mixture of hydrogen peroxide and chromic optical lightness of sediment (e.g., Herbert and Fischer, 1986;
acid can be used to oxidize organic matter (Degtjareff, 1930); Balsam and Deaton, 1991; Boden and Backman, 1996). This is
however, Walkley and Black (1934) conclusively showed that especially the case since the IODP research ship (Integrated
the hydrogen peroxide will also oxidize the dichromate lead- Ocean Drilling Program, IODP) JODIES Resolution has in-
ing to the measurement of artificially high organic levels. The stalled a portable spectrophotometer (Minolta CM-200)
presence of chloride within sediments will interfere with which enables quick assessment of cores while at sea
dichromate methods, as the formation of chromyl chloride (Schneider et al., 1995). Using this technique, the optical
results in the consumption of Cr2 O7 2 . The addition of silver brightness, lightness, or grayscale measures from a core are
sulfate (AgSO4) before analysis will remove the chloride correlated to the carbonate (Balsam et al., 1999) or organic
through the precipitation of AgCl2 (Walkley, 1947). (Kido et al., 2007) content of the sediment. The advantage of
270 Determining Organic and Carbonate Content in Sediments

Box 4 Walkley–Black method. From Nelson and Sommers, 1996; based on Walkley and Black, 1934; Walkley, 1947
NB: This technique uses toxic substances. All work should be conducted in a fume hood and reagents disposed of in an environmentally safe way. Consult your
local authority or hazardous substances manager on appropriate disposal techniques. Always consult the Material Safety Data Sheet (MSDS) prior to commencing
any chemical work.

(1) Titration methodology

Pre-experiment:
Create a solution of ferrous sulfate heptahydrate through dissolving 140 g of reagent grade FeSO4  7H2O in water, add 15 ml of concentrated sulfuric acid. Allow
solution to cool and dilute to a volume of 1000 ml. Standard the reagent daily by titrating it against 10 ml of 0.167 M potassium dichromate.
If Cl is present in the samples add Ag2SO4 to the sulfuric acid at a rate of 15 g/l1.
Step 1: Grind the sample to o0.5 mm avoiding steel or iron mortars.
Step 2: Transfer a weighed sample (containing 10–25 mg of organic C, but o10 g soil) into a 500 ml wide-mouth Erlenmeyer flask.
Step 3: Add 10 ml of 0.167 M K2Cr2O7 and swirl gently in flask to disperse soil.
Step 4: Rapidly add 20 ml of concentrated H2SO4 (496%) directing the stream into the suspension.
Step 5: Immediately swirl the flask gently until soil and reagents are mixed and then vigorously for a total of 1 min.
Step 6: Allow flask to stand on an insulated sheet for about 30 min.
Step 7: Add 200 ml water to the flask. If experience shows the endpoint of titration cannot be clearly observed, filter the suspension through glass fiber
filter paper.
Step 8: Add three to four drops of o-phenanthroline indicator and titrate the solution with 0.5 M FeSO4. As the endpoint is approached, the solution
takes on a greenish cast and then changes to a dark green.
Step 9: As soon as the solution turns dark green add ferrous sulfate heptahydrate solution drop by drop until the color changes sharply from blue to
red.
Step 10: Make a blank determination in the same manner, but without soil, to standardize the K2Cr2O7.

Calculations:
Calculate the results using the following formula and a correction factor (f) of 1.30.
Organic Cð%Þ ¼ ½ðmlblank  mlblank ÞðMFe2þ Þð0:003Þð100Þ=ðweight of dry soil in gÞ  f :

(2) Spectrophotometer Methodology

(Blakemore et al., 1987, pp. 18–19).
Step 1: Grind air-dried soil too0.25 mm.
Step 2: To a 200 ml Taylor-type flask (calibrated to show the 200 ml level) add 1.00 g (if o10% of C is expected) or 0.200 g (if 410% of C is
expected) of ground sediment.
Step 3: Add 12 ml of concentrated H2SO4 (from a 250-ml rapid-flow dispensing burette) and swirl gently. Allow to stand for 10 min.
Step 4: Add 6 ml 3 M chromium trioxide (from a rapid automatic pipette) and mix well by swirling, ensuring no particles are left on the side of the flask.
Step 5: Stand for exactly 10 min and dilute with freshwater nearly to the 200 ml mark. It has been found that that the addition of the chromium reagents
and water dilution can be carried out conveniently at 30 s intervals.
Step 6: Stand diluted solution overnight, preferably at a room temperature of 20 1C.
Step 7: After 18–24 h, add water to fill the flask to the 200-ml mark and stir well.
Step 8: Centrifuge an aliquot for 10–15 min at 2000 rpm.
Step 9: Read adsorption at 600 nm on a spectrophotometer.
To calculate the % C: Prepare a standard curve from the absorbance readings for the standards and read off concentrations of samples as mg C
For a 1 g (mg)/10 ¼ C in soil (%)
sample:
For a 0.2 g (mg)/2 ¼ C in soil (%)
sample
Preparation of standards:
Standards must be run with each sample run.

Weigh out powdered sucrose (o0.25 mm) according to either of the following tables:

Soil carbon (%) Soil carbon (mg) Sucrose carbon equivalent (mg) Sucrose (g)

For soils with less than 10% organic carbon (1 g soil)

0 0 0 0
2 20 19.6 0.0465
4 40 39.0 0.0926
6 60 58.4 0.1386
8 80 77.9 0.1849
10 100 97.4 0.2312

For soils with greater than 10% organic carbon (0.2 g soil)
0 0 0 0
10 20 20.8 0.0494
20 40 41.6 0.0988
30 60 62.4 0.1481
40 80 83.2 0.1974
50 100 104.0 0.2469
 Determining Organic and Carbonate Content in Sediments 271

this technique is that measurements can be made very quickly References

and even from images of cores rather than the core itself. This
technique is not a direct measure of the carbonate content of
Allison, L.E., Moodie, C.D., 1965. Carbonate. In: Black, C.A. (Ed.), Methods of Soil
the sediment and the brightness of the core can be affected by Analysis. Part 2, Second ed. CSSA and SSSA, Madison, MI, pp. 1379–1400.
other components which may darken or lighten the material. Anderson, R.F., Fleisher, M.Q., Lao, Y., Winckler, G., 2008. Modern CaCO3
For example, siliceous oozes can lighten the sediment while preservation in equatorial Pacific sediments in the context of late-Pleistocene
some metal oxides will darken it. In their review of the tech- glacial cycles. Marine Chemistry 111, 30–46.
Balsam, W.L., Deaton, B.C., 1991. Sediment dispersal in the Atlantic Ocean;
nique, Balsam et al. (1999) stated that given how various
evaluation by visible light spectra. Review of Aquatic Science 4, 411–447.
materials can easily change the reflectance of core sediment, Balsam, W.L., Deaton, B.C., Damuth, J.E., 1999. Evaluating optical lightness as a
the technique is surprisingly accurate. As this technique is a proxy for carbonate content in marine sediment cores. Marine Geology 161,
relative measure, it is imperative that absolute measurements 141–153.
of the carbonate content are undertaken to calibrate the Bathurst, R.G.C., 1976. Carbonate Sediments and Their Diagenesis. Elsevier,
Amsterdam.
readings of lightness on the sediment. Battarbee, R.W., Jones, V.J., Flower, R.J., Cameron, N.G., Bennion, H., Carvalho, L.,
Juggins, S., 2001. Diatoms. In: Smol, J.P., Birks, H.J.B., Last, W.M. (Eds.),
Tracking Environmental Change Using Lake Sediments, Volume 3: Terrestrial,
14.22.4 Summary and Conclusions Algal and Siliceous Indicators. Kluwer, Dordrecht, pp. 155–202.
Ben-Dor, E., Banin, A., 1989. Determination of organic matter content in arid-zones
soils using simple loss-on-ignition method. Communications in Soils Science
Methodologies for analyzing the organic and inorganic carbon and Plant Analysis 20, 1675–1695.
components of sediments and soils are highly varied and Blakemore, L.C., Searle, P.L., Daly, B.K., 1987. Methods for Chemical Analysis of
range from direct counting of grains, chemical liberation of Soils. NZ Soil Bureau, Department of Scientific and Industrial Research, Lower
carbon dioxide, to measuring light reflectance. The techniques Hutt.
Boden, P., Backman, J., 1996. A laminated sediment sequence from the northern
that have gained the widest use are those which oxidize the
North Atlantic Ocean and its climatic record. Geology 24, 507–510.
material to liberate carbon dioxide. It is the volume of the Borgatti, L., Ravazzi, C., Donegana, M., Corsini, A., Marchetti, M., Soldati, M.,
carbon dioxide that is then used to calculate the proportion of 2007. A lacustrine record of early Holocene watershed events and vegetation
organic and inorganic carbon in the sediment. There are sev- history, Corvara in Badia, Dolomites (Italy). Journal of Quaternary Science 22,
eral key considerations that need to be accounted for when 173–189.
Brewer, R., 1964. Fabric and Mineral Analysis of Soils. Wiley, Sydney.
choosing a methodology: Buhay, W.M., Simpson, S., Thorleifson, H., et al., 2009. A 1000-year record of dry
conditions in the Eastern Canadian prairies reconstructed from oxygen and
1. Grain size. The size of the sediment is critical in many
carbon isotope measurements on Lake Winnipeg sediment organics. Journal of
techniques, as coarser material takes longer to react. In Quaternary Science 24, 426–436.
some cases, this may preclude the use of some methods, Burdige, D.J., 2006. Geochemistry of Marine Sediments. Princeton University Press,
such as light reflectance techniques; the variation between Princeton, NJ.
grains will distort the results. Most chemical techniques Carter, L., Neil, H.L., McCave, I.N., 2000. Glacial to interglacial changes in non-
carbonate and carbonate accumulation in the SW Pacific Ocean, New Zealand.
will require the grains to be ground to ensure complete Palaeogeography, Palaeoclimatology, Palaeoecology 162, 333–356.
reaction of the sample. Carter, L., Neil, H.L., Northcote, L., 2002. Late Quaternary ice-rafting events in the
2. Accuracy. The accuracy of the techniques is widely variable. SW Pacific Ocean off eastern New Zealand. Marine Geology 191, 19–35.
Titration and coulometric methodologies are not only Chang, S., Lee, Y.I., 2002. Rapid determination of CO2 production rates by a
coulometric method. Geosciences Journal 6, 277–280.
more precise but also time consuming, especially if auto-
Choudhary, P., Routh, J., Chakrapani, G.J., Kumar, B., 2009. Biogeochemical
mated machines are not available for use. Detailed analysis records of paleoenvironmental changes in Nainital Lake, Kumaun Himalayas,
of soil carbon for agriculture or climate change studies India. Journal of Paleolimnology 42, 571–586.
requires greater accuracy than a simple study of carbonate Dean, W.E., 1974. Determination of carbonate and organic matter in calcareous
content in beach sand. Researchers should therefore be sediments and sedimentary rocks by loss on ignition: comparison with other
methods. Journal of Sedimentary Petrology 44, 242–248.
aware of the precision and accuracy of the method in re- Degtjareff, W.T., 1930. Determining soil organic mater by means of hydrogen
lation to the question that aims to be answered. peroxide and chromic acid. Soil Science 29, 233–245.
3. Relative versus absolute measurements. Many techniques do Dengiz, O., Ozaytekin, H.H., Cayci, G., Baran, A., 2009. Characteristics, genesis and
not give an exact measurement of carbonate or organic classification of a basin peat soil under negative human impact in Turkey.
Environmental Geology 56, 1057–1063.
carbon, but assume that the carbon dioxide released during
Dunn, D.A., 1980. Revised techniques for quantitative calcium carbonate analysis
the processing is related to one, or both, of these materials. using the ‘‘carbonate-bombe,’’ and comparisons to other quantitative carbonate
The method of light reflectance is only a proxy of com- analysis methods. Journal of Sedimentary Research 50, 631 636.
position. It is therefore critical, especially for sediments Erikson, A.E., Gieseking, J.E., 1947. A convenient methods for estimating
with a low composition of carbon elements, that precise carbonates in soils and other materials. Soil Science 63, 451–454.
Fitzpatrick, E.A., 1986. An Introduction to Soil Science, Second ed. Longman
measures are taken to calibrate the results. Scientific and Technical, Harlow.
Gaffey, S.J., Bronnimann, C.E., 1993. Effects of bleaching on organic and mineral
Often, it is the available equipment and cost that deter-
phases in biogenic carbonates. Journal of Sedimentary Research 63, 752–754.
mine which method is adopted. In geomorphology, simple Galle, O.K., Runnels, R.T., 1960. Determination of CO2 in carbonate rocks by
gravimetric measures are most commonly used; however, for controlled loss on ignition. Journal of Sedimentary Petrology 30, 613–618.
soil science, dichromic oxidation is used with greater fre- Guilizzoni, P., Massaferro, J., Lami, A., et al., 2009. Palaeolimnology of Lake Hess
quency. As the price of automated machines lessens, it is likely (Patagonia, Argentina): multi-proxy analyses of short sediment cores.
Hydrobiologia 631, 289–302.
that these will become of greater use to the researcher as they Hayes, M.H.B., Swift, R.S., 1978. The chemistry of soil colloids. In: Greenland,
provide greater accuracy of more complex techniques com- D.J., Hayes, M.H.B. (Eds.), The Chemistry of Soil Constituents. Wiley, New York,
bined with automation of the analytical procedure. NY, pp. 179–320.
272 Determining Organic and Carbonate Content in Sediments

Hemming, S.R., Hall, C.M., Biscaye, P.E., et al., 2002. 40Ar/39Ar ages and 40Ar Methods. Soil Science Society of America and American Society of Agronomy,
concentrations of fine-grained sediment fractions from North Atlantic Heinrich Madison, MI, pp. 961–1010.
layers. Chemical Geology 182, 583–603. Pansu, M., Gautheryrou, J., 2006. Handbook of Soils Analysis. Mineralogical,
Herbert, T.D., Fischer, A.G., 1986. Milankovitch climatic origin of mid-Cretaceous Organic and Inorganic Methods. Springer, Berlin, Heidelberg.
black shale rhythms in central Italy. Nature 321, 739–743. Piper, C.S., 1942. Soil and Plant Analysis: A Laboratory Manual of Methods for the
Hesse, P.P., 1994. The record of continental dust from Australia in Tasman Sea Examinations of Soils and the Determination of the Inorganic Constituents of
sediments. Quaternary Science Reviews 13, 257–272. Plants. Hassell Press, Adelaide.
Hesse, P.R., 1971. A Textbook of Soil Chemical Analysis. John Murray, London. Presley, B.J., 1975. A simple method for determining calcium carbonate in
Huffman, E.W.D., 1977. Performance of a new automatic carbon dioxide coulometer. sediment samples. Journal of Sedimentary Petrology 45, 745–746.
Microchemical Journal 22, 567–573. Rapp, P., Ramirez, W.R., Rivera, J.A., Carlo, M., Luciano, R., 2007. Measurement of
Hulsemann, J., 1966. On the routine analysis of carbonates in unconsolidated organic bottom loading under an open-ocean aquaculture cage, using sediment
sediments. Journal of Sedimentary Petrology 36, 622–625. traps on the bottom. Journal of Applied Ichthyology 23, 661–667.
Jackson, M.L., 1958. Soil Chemical Analysis. Constable, London. Reitemeier, R.F., 1943. Semimicroanalysis of saline soil solutions. Industrial and
Jones, G.A., Kaiteris, P., 1983. A vacuum–gasometric technique for rapid and Engineering Chemistry 15, 393–398.
precise analysis of calcium carbonate in sediment and soils. Journal of Robinson, W.O., 1927. The determination of organic matter in soils by means of
Sedimentary Research 53, 655–660. hydrogen peroxide. Journal of Agricultural Research 19, 315–324.
Kawahata, H., Minoshima, K., Ishizaki, Y., Yamaoka, K., Gupta, L.P., Nagao, M., Russell, J.M., McCoy, S.J., Verschuren, D., Bessems, I., Huang, Y., 2009. Human
Kuroyanagi, A., 2009. Comparison of settling particles and sediments at impacts, climate change, and aquatic ecosystem response during the past 2000
IMAGES coring site in the northwestern North Pacific; effect of resuspended yr at Lake Wandakara, Uganda. Quaternary Research 72, 315–324.
particles on paleorecords. Sedimentary Geology 222, 254–262. Schneider, R.R., Cramp, A., Damuth, J.E., et al., 1995. Color-reflectance
Kido, Y., Minami, I., Tada, R., Fujine, K., Irino, T., Ikehara, K., Chun, J.-H., 2007. measurements obtained from Leg 155 cores. In: Flood, R.D., Piper, D.J.W.,
Obital-scale stratigraphy and high-resolution analysis of biogenic components Klaus, A., et al. (Eds.), Proceedings of the Ocean Drilling Programme Initial
and deep-water oxygenation conditions in the Japan Sea during the last 640 kyr.
Report. Ocean Drilling Program, College Station, TX, vol. 155, 697–700.
Palaeogeography, Palaeoclimatology, Palaeoecology 247, 32–49.
Schultz, M.K., Biegalski, S.R., Inn, K.G.W., Yu, L., Burnett, W.C., Thomas, J.L.W.,
Lewis, D.W., 1984. Practical Sedimentology. Hutchinson, Ross, Stroudsburg.
Smith, G.E., 1999. Optimizing the removal of carbon phases in soils and
Lingane, J.J., 1954. Constant current source for coulometric titrations. Analytical
sediments for sequential chemical extractions by coulometry. Journal of
Chemistry 26, 522–626.
Environmental Monitoring 1, 183–190.
Loeppert, R.H., Hallmark, C.T., Koshy, M.M., 1984. Routine procedure for rapid
Sparks, D.L., 1996. Methods of Soil Analysis: Part 3 Chemical Methods. Soil
determination of soil carbonates. Journal of the American Soil Science Society
Science Society of America and American Society of Agronomy, Madison, MI.
48, 1030–1033.
Stevenson, F.J., 1994. Humus Chemistry. Genesis, Composition, Reactions, Second
Loeppert, R.H., Suarez, D.L., 1996. Carbonate and gypsum. In: Sparks, D.L. (Ed.),
ed. Wiley, New York, NY.
Methods of Soil Analysis: Part 3 Chemical Methods. Soil Science Society of
Swift, R.S., 1996. Organic matter characterization. In: Sparks, D.L. (Ed.), Methods of
America. American Society of Agronomy, Madison, MI, pp. 437–474.
Martinez, P., Robinson, R.S., 2010. Increase in water column denitrification during Soil Analysis: Part 3 Chemical Methods. Soil Science Society of America and
the last deglaciation: the influence of oxygen demand in the eastern equatorial American Society of Agronomy, Madison, MI, pp. 1011–1069.
Pacific. Biogeosciences 7, 1–9. Thomson, J., Nixon, S., Summerhayes, C.P., Schonfeld, J., Zahn, R., Grootes, P.,
Milliman, J.D., 1974. Marine Carbonates. Springer, Berlin. 1999. Implications for sedimentation changes on the Iberian margin over the
Mooers, H.D., 1999. Comparison of loss on ignition and coulometric titration for last two glacial/interglacial transitions from (230Thexcess) systematics. Earth and
determining carbonate content of till. Glacial Geology and Geomorphology tn01/ Planetary Science Letters 165, 255–270.
1999, 1–10. Trentesaux, A., Recourt, P., Bout-Roumazeilles, V., Tribovillard, N., 2001. Carbonate
Moore, J.J., Loeppert, R.H., West, L.T., Hallmark, C.T., 1987. A routine method for grain-size distribution in hemipelagic sediments from a laser particle sizer.
calcium carbonate equivalent of soils. Communications in Soil Science and Journal of Sedimentary Research 71, 858–862.
Plant Analysis 18, 265–277. Tucker, M.E., Wright, V.P., 1990. Carbonate Sedimentology. Blackwell, London.
Morley, D.W., Leng, M.J., Mackay, A.W., Sloane, H.J., Rioual, P., Battarbee, R.W., U.S. Salinity Soil Lab, 1954. Diagnosis and Improvement of Saline and Alkali Soils.
2004. Cleaning of lake sediment samples for diatom oxygen isotope studies. US Govt Printing Office, Washington, DC.
Journal of Paleolimnology 31, 391–401. Vousdoukas, M.I., Velegrakis, A.F., Plomaritis, T.A., 2007. Beachrock occurrence,
Muller, G., Gastner, M., 1971. The ‘‘Karbonate Bombe’’, a simple device for the characteristics, formation mechanisms and impacts. Earth Science Reviews 85,
determination of the carbonate content in sediments, soils and other materials. 23–46.
Neues Jahrbuch Fur Mineralogie-Mk 10, 466–469. Walkley, A., 1947. A critical examination of a rapid method for determining organic
Murray, R.W., Knowlton, C., Leinen, M., Mix, A.C., Polsky, C.H., 2000. Export carbon in soils: effect of variations in digestion conditions and of inorganic soil
production and carbonate dissolution in the central equatorial Pacific Ocean over constituents. Soil Science 63, 251–263.
the past 1 Myr. Paleoceanography 15, 570–592. Walkley, A., Black, I.A., 1934. An examination of the Degtjareff method for
Nelson, D.W., Sommers, L.E., 1996. Total carbon, organic carbon and organic determining soil organic matter and a proposed modification if the chromic acid
matter. In: Sparks, D.L. (Ed.), Methods of Soil Analysis: Part 3 Chemical titration method. Soil Science 37, 29–38.
 Determining Organic and Carbonate Content in Sediments 273

Biographical Sketch

David M. Kennedy is a geomorphologist who specializes in the response coastal systems to climatic and en-
vironmental change. He particularly focuses on the Holocene, researching estuarine, coral reef and rocky coast
landforms. He completed his doctorial study at the University of Wollongong, Australia under the guidance of
Professor Colin Woodroffe following undergraduate studies at the University of Sydney, Australia. After a post-
doctoral research position at Wollongong he was based at Victoria University of Wellington, New Zealand. He is
currently a senior lecturer in the Department of Resource Management and Geography at The University of
Melbourne, Australia.

Josephine L.D. Woods is an environmental scientist who specializes in the role of metal and organic pollutants in
estuarine systems. She has field experience on sites in Australia, New Zealand, and the wider Pacific. She is
currently based in the Centre for Aquatic Pollution Identification and Management and the Department of
Zoology at The University of Melbourne.
 14.23 Wet Chemical Methods (pH, Electrical Conductivity, Ion-Selective
Electrodes, Colorimetric Analysis, Ion Chromatography, Flame Atomic
Absorption Spectrometry, Inductively Coupled Plasma-Atomic Emission
Spectroscopy, and Quadrupole Inductively Coupled Plasma-Mass Spectrometry)
C Chagué-Goff, University of New South Wales, Sydney, NSW, Australia, and Australian Nuclear Science and Technology
Organisation, Kirrawee DC, NSW, Australia
HKY Wong, Australian Nuclear Science and Technology Organisation, Kirrawee DC, NSW, Australia
r 2013 Elsevier Inc. All rights reserved.

14.23.1 Introduction 274

14.23.2 Pretreatment of Samples 275
14.23.3 Water for Analytical Methods 275
14.23.4 pH 275
14.23.5 Electrical Conductivity 276
14.23.6 Ion-Selective Electrodes 276
14.23.7 Colorimetric Analysis 277
14.23.8 Ion Chromatography 278
14.23.9 Flame Atomic Absorption Spectrometry 278
14.23.10 Inductively Coupled Plasma Spectrometries 278
14.23.10.1 Interferences 279
14.23.11 Summary 280
References 280

Glossary pH Negative logarithm of the concentration of hydrogen

Anions Negatively charged ions (e.g., Cl and SO42). ions in solution (in normal dilute solutions), pH ¼  log
Cations Positively charged ions (e.g., Na þ and Ca2 þ ). [H þ ].
Electrical conductivity Measure of the concentration of
soluble salts (cations and anions) in solution, expressed as
mS cm1 or mS m1 (1 mS m1 ¼10 mS cm1).

Abstract

A summary of wet chemical methods commonly used in geomorphological studies is presented. Instructions and tips are
provided for the determination of pH and electrical conductivity in the laboratory. General information is presented about
the principles of ion-selective electrodes, colorimetric analysis, ion chromatography, FAAS, ICP-AES, and Q-ICP-MS,
together with some examples. References to specialized textbooks and standard analytical methods are provided for each
analytical technique.

14.23.1 Introduction material, weathering, contamination, and spatial and tem-

poral changes. Wet chemical methods cover a large range of
Analysis of water and soil/sediment extracts is commonly analyses, from pH and electrical conductivity (EC) analysis of
conducted in geomorphology, as it can provide information single analytes to multielemental analysis using standard la-
about many aspects relevant to the discipline, such as the boratory instruments. The aim of this chapter is to provide an
present and past environmental conditions, source of insight in some of the most common analytical techniques,
although acknowledging that the list is not exhaustive. Readers
Chagué-Goff, C., Wong, H.K.Y., 2013. Wet chemical methods (pH, electrical are referred to the (Standard Methods for the Examination
conductivity, ion-selective electrodes, colorimetric analysis, ion of Water and Wastewater, 2005, online edition, http://
chromatography, flame atomic absorption spectrometry, inductively www.standardmethods.org/) for specific analytical techniques
coupled plasma-atomic emission spectroscopy, and quadrupole inductively
and details on methodology. Numerous textbooks on soil
coupled plasma-mass spectrometry). In: Shroder, J. (Editor in Chief),
Switzer, A.D., Kennedy, D.M. (Eds.), Treatise on Geomorphology. Academic chemical analysis also exist, such as Rowell (1994), Sparks
Press, San Diego, CA, vol. 14, Methods in Geomorphology, pp. 274–281. (1996), and Rayment and Lyons (2010).

274 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00391-2

 Wet Chemical Methods 275

Tip: Always refer to the manufacturer’s handbook and distilled water having an EC of less than 10 mS cm1. In
notes for appropriate instrument operating parameters. modern laboratories, distillation units are replaced with de-
ionized water units. Type 1 water, identified as ultrapure water,
with a resistivity reading of at least 18 mO cm1 and a total
14.23.2 Pretreatment of Samples organic carbon (TOC) content of less than 10 mg l1 (Milli-
pore, 2011a), is used for ultra-trace metal analysis. Type 2
Numerous rules must be followed when bringing water or water, identified as pure water, has a resistivity reading 45
soil/sediment samples from the field to the laboratory. First, mO cm1 and a TOC content of less than 30 mg l1 (Millipore,
the batch of samples should be accompanied by detailed in- 2011b), and is used for general laboratory applications.
formation indicating the number of samples and details per-
tinent to their collection, as well as analysis to be carried out.
Characteristically many problems occur as the analyses are 14.23.4 pH
undertaken in commercial laboratories by people who do
not know the environment the samples were collected in, or pH is defined as the negative logarithm of the activity of
the environmental conditions the collected material has ex- hydrogen ions in solution. Note that, except in saline solu-
perienced subsequent to field collection. Transportation time tions, where the activity is smaller than the concentration, in
needs to be kept to a minimum and depending on the analysis normal dilute soil solutions, the activity is equivalent to the
required, samples need to be preserved, either cool on ice or concentration of H þ [H þ ] in mol l1 (Rowell, 1994) and thus
acidified to opH 2. Although they may be frozen in certain pH ¼  log [H þ ]. pH is measured using a glass electrode and a
cases, advice needs to be sought from the analyzing laboratory reference electrode (which are commonly manufactured as
prior to sample collection. a single combination electrode), and a pH meter, which is a
Measurements of pH and conductivity of water have to be millivoltmeter. The pH meter needs to be calibrated using
carried out on unfiltered samples. However, samples need to at least two buffer solutions, usually with pH values of 4 and
be filtered immediately after collection or as soon as practical 7, or 7 and 10, depending on the values to be determined.
for a number of analyte determinations (e.g., dissolved nu- Before and after sample measurements, all standards should
trients and metals). Again, advice is to be sought from the be checked to ensure readings indeed indicate the correct
laboratory regarding the pore size of the filter to be used. pH value.
Where and when possible, pH should be measured in situ (for pH measurement is made either in water, or using a sus-
both water and soil), but for a number of reasons, such as pension made generally of approximately 10 g of dried soil
unavailability of a pH meter, time constraints in the field, or (o2 mm) and 50 ml of deionized water (soil:water ratio of
soil too dry to allow pH determination in the field, it is 1:5), which is stirred for at least 15 min before measurement.
commonly necessary to measure water pH in the laboratory The suspension should also be stirred during measurement.
and for soil to be suspended in liquid to allow measurement The electrodes should be thoroughly washed with deionized
of pH. water before and after calibration and between measurements.
Soil samples should be oven-dried below 35 1C (or The measurement should be carried out in duplicate, of sep-
air-dried), and then ground and sieved to less than 2 mm arate subsamples, and replicate results should be within
(Rayment and Higginson, 1992) before further analysis. 0.1 pH unit. It is best that samples, buffers, and electrodes
are at the same temperature, but a temperature difference of
2–5 1C is acceptable in most cases, as the difference in pH
14.23.3 Water for Analytical Methods reading between 20 and 25 1C, for example, is less than 0.1 pH
unit (Barron et al., 2010). Refer to the manufacturer’s hand-
Analytical methods commonly prescribe the use of distilled book and data on buffer solutions for corrections for each
water for mixing or rinsing. However, distilled water is not specific instrument.
always available in laboratories, where purification of water The measurement of pH is generally seen as one of the
is now commonly obtained through reverse osmosis (RO) and simplest methods in soil chemistry. However, it is not that
deionization. Furthermore, there also appears to be some easy, as the pH theory applies to dilute solutions of simple
misunderstanding about the level of purity of the water (RO, electrolytes (Bates, 1962), whereas the measured value for any
deionized, and distilled water). Thus, definitions are provided soil sample is dependent on procedures used in the laboratory,
below: RO water is water that has undergone reverse osmosis. such as the soil:water ratio, stirring or not during measure-
RO is a process by which water moves across a membrane ment, and position of the electrode during measurement
against the concentration gradient. It results in the removal of if no stirring (e.g., Blakemore et al., 1987; Sumner, 1994). The
ions (e.g., Ca2 þ , Cl, and Na þ ) and large molecules. De- soil:water ratio alone is a matter for debate, and various ratios
ionized water is also called demineralized water. Deionization are provided by different authors (e.g., Rayment and Higgin-
is a physical process that uses specially manufactured ion ex- son, 1992; Rowell, 1994; Sumner, 1994; Kissel et al., 2009).
change resins to remove mineral ions. Because the majority of We suggest a soil:water ratio of 1:5 as it allows the use of a
water impurities are dissolved salts, deionization results in common pH electrode. Researchers might choose a different
high purity water that is generally similar to distilled water. ratio, in particular if comparing with previous results. How-
Distilled water is water that has undergone distillation, ever, one needs to be aware that a soil:water ratio of 1:2.5
whereby water is boiled and the steam is condensed into a results in samples being more like a paste, thus requiring the
clean container. Distillation produces very pure water, with use of a paste pH electrode.
276 Wet Chemical Methods

Tips: The important points to consider when carrying out standard used. Record the units on the EC meter when taking
pH analysis are to make sure that: (1) the pH meter is cali- a reading, as the meter may automatically convert from one to
brated with at least two buffers, (2) the electrodes are washed another (either mS m1 or mS cm1).
before and after calibration, as well as between measurements,
and (3) the difference in temperature between standard buffer
solutions and samples is no more than 5 1C. 14.23.6 Ion-Selective Electrodes
A simple way to check if the electrode is working properly
is to dilute the pH 7 calibration solution 10 times. The reading Ion-selective electrodes (ISEs) allow quick and cost-effective
should be 7.270.05. If the reading is different, refer to the determination of ion activity. Compared to many other ana-
electrode manual for cleaning and maintenance requirement. lytical techniques, they are relatively inexpensive, simple to
Note that the electrode should always have a minor amount of use, and offer a wide range of applications and concentration
saturated KCl crystals as a conditioning requirement. If it is ranges; they are not sensitive to color, viscosity, and suspended
not the case, refill the electrode by adding saturated KCl (with solids, and allow the quick measurement of changes in con-
crystals) into the electrode through the opening on the side. centrations (e.g., Pretsch, 2002; Bedlechowicz et al., 2005).
The working range for the electrode sensitivity should be The ISE, which comprises a special ion-sensitive membrane
97–100%. If it is below this value, the electrode needs to that interacts with the ion of choice, is used in combination
be cleaned. After cleaning and refilling the electrode, the pH with a reference electrode of constant potential and both are
meter should be recalibrated. Also note that there should not immersed into the aqueous solution with ions to be meas-
be any air bubbles between the crystals or at the reference ured. Both electrodes are connected to a millivoltmeter, which
element. Refer to the manufacturer’s handbook for more de- measures the potential difference that develops across the ISE
tails in trouble shooting and cleaning to improve faster re- membrane. The measured voltage is proportional to the
sponse, better accuracy, and reproducibility of the electrode. logarithm of the concentration. The electrodes are first cali-
brated using solutions of known concentrations, and a graph
of mV versus concentration is plotted. The concentration of
14.23.5 Electrical Conductivity the unknown solution is determined by plotting its mV value
on the graph. The temperature of the standard solutions and
EC provides a measure of the concentration of soluble salts sample should be kept within a few degree centigrades, to
(cations and anions) in solution. It is the specific conductance, keep with good laboratory practices.
or ability of a solution to carry electricity, and is the inverse of Although the pH membrane in the pH electrode only
the resistivity. The SI unit for EC is siemens per m (S m1). allows the passage of hydrogen ions, other ISEs are not en-
However, depending on the ion concentration in solution, it is tirely ion specific, thus leading to issues with interfering ions.
generally expressed in microsiemens per cm (mS cm1) or See also the manufacturer’s instructions, which provide in-
millisiemens per m (mS m1), with 1 mS m1 ¼10 mS cm1. formation of the interfering ion to each measured ion. In di-
EC is temperature dependent, and is reported at 25 1C. If lute solutions, it might not be a problem. Otherwise, the
measured at a different temperature, the correction of the interference can be corrected either by removing the inter-
value to 25 1C is made using the following equation: fering ion through precipitation or complexation prior to
analysis, or by the fixed interference method, whereby the
ECðmScm1 Þ ¼ km =ð1 þ 0:0191ðt  25ÞÞ, concentration of the primary ion is changed whereas the
background concentration of the interfering ion is kept con-
where km is the measured conductivity at temperature t (in 1C) stant (e.g., Salt and Vora, 1991).
(method 2510B – Standard Methods for the Examination of ISEs are one of the most popular and widely used method
Water and Wastewater, 2005). However, modern instruments for the determination of fluoride (e.g., Wen et al., 1998;
generally have an automatic temperature compensation. Konieczka et al., 2000; Standard Methods for the Examination
EC is determined with a conductivity cell by measuring the of Water and Wastewater, 2005 – method 4500), and a sum-
electrical resistance of water or a soil suspension made up with mary of the method is described here. The fluoride electrode is
10 g of dried soil (o2 mm) and 50 ml of deionized water (soil: used in combination with a standard calomel reference elec-
water ratio of 1:5), which has been stirred for 1 h (Rayment and trode (sleeve-type) and a pH meter with an extended millivolt
Higginson, 1992). The conductivity cell must be first calibrated scale. A series of fluoride standards is first prepared by diluting
using the KCl reference solution at the same temperature as the a known fluoride standard solution made from a stock
water sample or soil suspension, and rinsed with deionized solution of anhydrous sodium fluoride (NaF) with known
water before use. After allowing the soil suspension to settle for volumes of deionized water.
20–30 min, the EC of the supernatant is measured with the The effects of interfering ions, namely aluminum and iron,
conductivity cell, and the reading is taken once the system has but also hydroxide, can be minimized by adding a known
stabilized. EC of water samples can be measured after the volume of fluoride buffer (e.g., Konieczka et al., 2000;
reading of the electrode immersed in water has stabilized. Standard Methods for the Examination of Water and Waste-
Tips: When calibrating the EC meter, take into consider- water, 2005) to standards and samples, which should be at the
ation the conductivity of the sample. Always calibrate the same temperature. Readers are referred to Standard Methods
meter using standards of similar value to the sample. The for the Examination of Water and Wastewater (2005) for exact
reading needs to be repeated if the sample conductivity is concentration and pH requirements of the buffer solution,
greater than or less than 75% of the conductivity of the which is available commercially. Formation of hydrogen
 Wet Chemical Methods 277

fluoride complex, which can occur in acid solutions, can be colored substance, a characteristic portion of the wavelength is
minimized by maintaining pH above 5. Hydroxide inter- absorbed and the remaining light has the complementary
ference, which can occur in alkaline solutions, can also be color of the absorbed wavelength (e.g., Skoog and Leary,
avoided as long as the pH is maintained between 5.3 and 5.5, 1992).
which is achieved by the buffer (Standard Methods for the A colorimeter is an instrument that determines the con-
Examination of Water and Wastewater, 2005). centration of a solution by measuring its absorbance of
Electrodes are immersed into each standard solution, a specific wavelength of light, as the concentration is pro-
which is then stirred with a magnetic stirrer. Readings of portional to the absorbance (Beer–Lambert law). A spectro-
potential in mV are taken after 3 min in the standard solutions photometer measures the light intensity as a function of
and the samples. Electrodes have to be washed with deionized the color, or more specifically, the light source wavelength.
water and blotted dry between each reading. If a direct-reading Spectrophotometers use a monochromator containing a dif-
measurement instrument is not used, the measurements of fraction grating that splits the spectrum into a large number of
the standard solutions are plotted (potential mV) against bands, and can read each band to define a more detailed
concentration on logarithmic paper. The concentration of spectral graph (Figure 1).
the unknown samples is then read using the potential value Colorimetric methods have long been used for the analysis
reading on the graph. Another commonly used method is of many elements (see Table 1 for examples of analytes).
the standard addition method, whereby known solutions Nitrite, for example, can be determined through formation
of standard are added to each sample. The reader is referred of a reddish purple dye formed through complexation of
to Pungor (1998), Pretsch (2002), and Standard Methods diazotized sulfanilamide with NED dihydrochloride, and
for the Examination of Water and Wastewater (2005), to photometric measurement at 543 nm with a 1-cm light path
name a few, for further details on the principles and uses of (Standard Methods for the Examination of Water and Waste-
ISEs. water, 2005 – method 4500-NO2 B). The concentration is
Although there are many advantages in using ISEs, as their calculated using a standard curve prepared by plotting the
name indicates, they are ion selective, and thus they are ideal if absorbance of standards versus NO2–N concentrations,
only one or a few ions have to be determined, or if it is re- which follows Beer–Lambert law up to 180 mg N l1. Nitrate
quired to measure the change in concentration in the field. If can be determined using a UV spectrophotometric technique,
the determination of more ions are required, then more time- whereby the absorbance of NO3 is measured at 220 nm.
effective, analytical techniques include ion chromatography However, this technique is mainly used for screening samples
(IC) (for anions and cations), inductively coupled plasma- with low organic content (due to the interference of dissolved
atomic emission spectroscopy (ICP-AES) (for cations), or in- organic matter at 220 nm), although a second measurement
ductively coupled plasma-mass spectrometry (ICP-MS) (for may be made at 275 nm to determine the interference and
cations and isotope ratios), which also provide more precise correct the NO3 value (Standard Methods for the Examin-
results and better detection limits. See Table 1 for examples of ation of Water and Wastewater, 2005 – method 4500-NO3 B).
analytes and methods of analysis.

Monochromator
14.23.7 Colorimetric Analysis Detector
Light source Sample in cuvette Recorder
Colorimetric methods are based on the measurement of
electromagnetic spectra absorbed by chemical substances. Al-
though we see white light as uniform in color, it is actually
composed of a broad range of radiation wavelengths from
ultraviolet (UV), visible, and infrared (IR) portions of the
spectrum. The visible spectrum covers a range of approxi-
mately 380–750 nm, with the shortest visible wavelength Figure 1 Schematic diagram of the main elements of a
violet and the longest red. When white light passes through a spectrophotometer.

Table 1 Examples of analytes and possible/preferred methods of analysis, as well as detection limits. Colorimetric¼colorimetric analysis,
ISE¼ion-selective electrodes, IC ¼Ion Chromatography ( subject to type of column used), FAAS¼Flame Atomic Absorption Spectrometry, ICP-
AES¼Inductively Coupled Plasma – Atomic Emission Spectroscopy, Q-ICP-MS¼Quadrupole-Inductively Coupled Plasma – Mass Spectrometry

Method Analytes Detection limit

Colorimetric NO2, 
P , Li , Na , Ca , K , NH4þ , Mg2 þ , Cu2 þ
Br , NO3, þ þ þ 2þ þ
Sub mg l1
ISE Br , Ca , Cl , CN, Cu2 þ , F, I, K þ , Na þ , NH4þ , NO3, Pb2 þ , S2
 2þ 
Sub mg l1
IC F, Cl, NO2, Br, NO3, SO42, PO43, Li þ , Na þ , Ca2 þ , K þ , NH4þ , Mg2 þ , Sr2 þ , Ba2 þ Sub mg l1
FAAS Most metals Sub mg l1
ICP-AES Most metals and some non-metals mg l1 to mg l1
Q-ICP-MS Most metals and some non metals Sub mg l1

Note: parts per million (ppm) ¼ mg l  1 or mg kg  1; parts per billion (ppb) ¼mg l  1 or mg kg  1; 1%¼ 10 000 ppm (mg l  1 or mg kg  1).
278 Wet Chemical Methods

14.23.8 Ion Chromatography 14.23.9 Flame Atomic Absorption Spectrometry

IC is an analytical technique for the determination of the Flame atomic absorption spectrometry (FAAS) is based on a
common cations and anions. It is also known as a liquid–solid sample being aspirated into a flame and atomized. A light
chromatographic method where the eluent (liquid) is passed beam is directed through the flame, then into a mono-
through a column of inorganic or polymeric organic resin chromator, and then onto a detector that measures the
(solid) and then to a flow-through detector (Figure 2). amount of light absorbed by the atomized element in the
A fixed amount of water sample using a sample loop of flame (Standard Methods for the Examination of Water
10–100 ml is injected into the eluent via the switching valve. The and Wastewater, 2005). Each metal has its own characteristic
eluent is continuously pumped through the preconditioned absorption wavelength, thus a source lamp made of that
column at a constant flow rate of approximately 1–2 ml min1. particular element is used for each element to be analyzed,
The column, which is packed with uniform resin, is 2–4 mm which also reduces interferences. The amount of energy at
wide with a typical length of 15 cm. The anions or cations in that characteristic wavelength absorbed in the flame is pro-
the sample are chemically bound to the resin before being re- portional to the concentration of the element, but only over a
placed by the counter ions of the same charge from the eluent. limited concentration range. Many metals can be determined
The elution characteristics are determined by the ion exchange using an air-acetylene flame (e.g., Ca and Mg), with chemical
capacity of the resin and the concentration of competing ions in interferences being reduced or eliminated by adding specific
the eluent. As the sample flows through the column, it begins to compounds to the sample solution. Some elements (e.g., Al, P,
separate into its individual components. The sample com- and Sr) require a hotter flame for complete dissociation, and
ponents that are weakly bound by the resin travel rapidly with nitrous oxide-acetylene flame is thus used. Standard solutions
the flow of the eluent. The components travel through the of known metal concentrations are also prepared in a similar
column at different rate and enter the detector at different matrix of that of the sample.
times. A flow-through membrane device known as the sup- Readers are referred to Skoog and Leary (1992) for details
pressor is added after the column to reduce the background on the principles of FAAS and Standard Methods for the
noise and improve detection limit. This is achieved by changing Examination of Water and Wastewater (2005) (method 3111)
the highly conductive strength of the eluent to a weaker acid. for details on the analytical technique, flame gas required,
The conductivity detector senses the sample components and preparation of standards, and interferences. Although FAAS
produces a chromatogram. By determining the peak area or used to be the preferred method of analysis for most metals,
peak height, one can quantify the concentration by comparing due to its simplicity of use and relatively low cost, it is time
it with known standards. consuming as only one element can be analyzed at one time.
Readers are referred to Haddad and Jackson (1990), Smith Thus, the method has now most often been superseded by
(1990), and Skoog and Leary (1992) for details on the prin- ICP-AES and ICP-MS (see Section 14.23.10), in particular for
ciples and applications of IC. The technique has been used in multielement analysis.
a number of studies assessing the effect of anthropogenic ac-
tivities on soil and water quality, either for the analysis of
cations and anions in water, or the determination of analytes 14.23.10 Inductively Coupled Plasma
(e.g., Cl, S, and F) after extraction from soil samples. For ex- Spectrometries
ample, IC was used in a study investigating the efficiency of a
small wetland treating domestic wastewater (Chagué-Goff Inductively coupled plasma-atomic emission spectrometry
et al., 1999). Data showed that sulfate concentrations had (ICP-AES) and quadrupole-inductively coupled plasma-mass
decreased after passage through the land treatment area, spectrometry (Q-ICP-MS) have become the most popular
whereas phosphate concentrations had not, due to release of techniques used today for multielement analysis, due to the
phosphorus by the sediment. following reasons (e.g., Montaser, 1998):

Sample

Eluent Pump Column Suppressor Detector Data processing

Waste = Switching vial with sample loop

Figure 2 Schematic diagram of ion chromatography.

 Wet Chemical Methods 279

1. Better detection limits than FAAS. The detector measures the light (ICP-AES) or ions (Q-ICP-
2. Improved sample throughput and number of elements MS) and converts them to electrical signal as counts per second
under one set of conditions. (cps) that can be measured. The signal is proportional to
3. Greater linear dynamic range (up to 8 orders of magnitude). the concentration of the element and one can quantify the
4. Isotope analysis is possible with ICP-MS. concentration by comparing the signal intensities to known
5. Lower interferences due to higher temperature than the concentrations.
flame (B2–3 times greater). ICP-AES and ICP-MS are used in a number of studies in-
vestigating natural and anthropogenic changes in marine,
Basically, the sample compartments and plasma of the ICP-
coastal, and freshwater environments, such as that of Pas-
AES and ICP-MS are very similar (Figure 3). The sample, in a
ternack and Brown (2006), who used ICP-AES to reconstruct
liquid form, is pumped or self aspirated via suction into the
the Holocene geochemical history of a floodplain in California.
nebulizer. If the sample to be analyzed is a solid, it is first
dissolved/digested and then analyzed. The liquid forms an
aerosol of fine droplets (1–10 mm) with argon gas using the 14.23.10.1 Interferences
nebulizer. Large droplets are separated from the fine droplets
ICP-AES: spectral – due to the large amount of emission
in the spray chamber. The argon gas carries the fine droplets
spectral lines (greater than 50 000), there are occasions when
into the plasma, where the sample is vaporized, atomized, and
one overlaps another. Interelement correction could reduce
ionized (Figure 4). The plasma is a spherical ball of ionized
the errors but has limited success. Background correction with
argon gas at around 8000–10 000 K.
sophisticated equations could improve the accuracy. The best
Emission Spectroscopy is based on light produced from
option is to find another emission line without the inter-
excited atoms and ions when they return to their ground state
ference. Thus, the resolution of the instrument is important to
or to lower excitation state. An element will emit light with a
resolve the spectral overlap and reduce these interferences.
set of characteristic wavelengths that can be used to identify
Matrix effect – spray chamber effects and viscosity differ-
the element. The spectrum light passes through the mono-
ence between samples and calibration standards could also
chromator containing lens, mirrors, gratings/prisms, slits, and
affect the accuracy of the results. That is, the efficiency of
detectors. The primary function of the monochromator is to
samples carried up to the plasma is different for the standards
isolate light to its specific wavelength of interest, where the
and the samples. One way to overcome this matrix problem is
detector can measure the intensity of the wavelength. There are
to matrix match the standards to the samples or perform a
two basic types of ICP-AES, sequential and simultaneous. The
standard addition instead of the normal external calibration.
sequential instrument as the name implies measures the
Q-ICP-MS – there are less interferences identified than by
wavelengths one by one, whereas the simultaneous instrument
ICP-AES. These interferences occur when species has a similar
measures a large number of wavelengths at the same time.
mass to the analyte. Some of the examples are shown below:
Detection limits have improved by a factor from 2 to 10
through axial viewing (end-on) of the plasma instead of radial 1. Isobaric or elemental interferences – 40Ar on 40Ca,
(side-on) (e.g., Montaser, 1998). Readers are referred to 2. Polyatomic or matrix interferences – 40Ar16O on 56
Fe,
40 35
Montaser and Golightly (1987), Varma (1991), and Skoog Ar Cl or 40Ca35Cl on 75As, and 40Ar37Cl on 77Se,
and Leary (1992) for details on principles and applications of 3. Doubly charged ions – 88Sr2 þ on 44Ca,
ICP-AES. 4. Matrix effects similar to the ICP-AES.
The Q-ICP-MS generally has a much lower detection limit
than the ICP-AES. Excited ions produced in the plasma
are extracted into the mass spectrometer via an ion lens
system. Light is excluded from the quadrupole filter and de-
tector region via a photon stop or through bending ions into
the quadrupole filter to reduce background noise. The quad- Aerosol
rupole mass filter consists of four cylindrical rods of ap-
proximately 1 cm in diameter and 15 cm in length. By varying
the voltage of AC and DC applied to the rods, only ions of a
Ions Atoms Gas Solid Liquid
specific mass to charge ratio are allowed through onto the
detector at any one time. The Q-ICP-MS is a fast sequential
instrument which can separate up to 2400 atomic mass units
(Ionization) (Atomization) (Vaporization) (Desolvation)
(amu) per second. Readers are referred to Skoog and Leary
(1992) and Montaser (1998) for details on principles and Figure 4 Schematic diagram of the inductively coupled
applications of Q-ICP-MS. plasma torch.

Monochromator/mass filter
Sample Nebulizer Spray chamber Plasma
+ detector

Figure 3 Schematic diagram of the principles of inductively coupled plasma – atomic emission spectroscopy and quadrupole-inductively
coupled plasma–mass spectrometry.
280 Wet Chemical Methods

To minimize these interferences, one could: choose another Bedlechowicz, I., Sokalski, T., Lewenstam, A., Maj-zurawska, M., 2005. Calcium
mass/isotope which is free of the interference; optimize the ion-selective electrodes under galvanostatic current control. Sensors and
Actuators B 108, 836–839.
instrument to minimize the oxides and doubly charged ions;
Blakemore, L.C., Searle, P.L., Daly, B.K., 1987. Methods for chemical analysis of
use lower power (cool plasma) or shield torch interface setting soils. NZ Soil Bureau Scientific Report 80. Lower Hutt, New Zealand, p. 103.
to reduce the polyatomic interferences with argide ions; or use Chagué-Goff, C., Rosen, M., Roseleur, M., 1999. Water and sediment chemistry of
internal standards to minimize matrix effects. In the modern a wetland receiving municipal wastewater. New Zealand Journal of Marine and
instruments today, a collision or reaction cell is installed in the Freshwater Research 33, 649–660.
Haddad, P.R., Jackson, P.E., 1990. Ion chromatography principles and applications.
system to further reduce the matrix effects from chloride and Journal of Chromatography Library. Elsevier, Amsterdam, Volume 46, 776 pp.
argide formation. The use of high resolution (HR) ICP-MS Kissel, D., Sonon, L., Vendrell, P., Isaac, R., 2009. Salt concentration and
would also reduce these matrix effects, but the cost is ap- measurement of soil pH. Communications in Soil Science and Plant Analysis
proximately double that of a Q-ICP-MS. 40, 179–187.
Konieczka, P., Zygmunt, B., Namiesnik, J., 2000. Comparison of fluoride ion-
Tips: Always assess if the detection limit by any specific
selective electrode based potentiometric methods of fluoride determination in
method is sufficient for the measurements required. Generally, human urine. Bulletin of Environmental Contamination and Toxicology 64,
instrumental methods are more sensitive than the colorimetric 794–803.
and ISE method. Always include a procedural blank with each Millipore, 2011a. Type 1 (ultrapure) Milli-Q. http://www.millipore.com/lab_water/
set of samples. A procedural blank, which is a reagent water clw4/type1 (accessed 13 July 2011).
Millipore, 2011b. Type 1 (pure) Elix. http://www.millipore.com/lab_water/clw4/type2
blank containing acids, is included in any digestion steps. As
(accessed 13 July 2011).
part of the quality assurance, if possible include a reference Montaser, A., 1998. Inductively Coupled Plasma Mass Spectrometry. Wiley-VCH,
material of a known certified concentration that is similar New York, 964 pp.
in matrix to the sample. This will ensure that digestion and Montaser, A., Golightly, D.W., 1987. Inductively Coupled Plasmas in Analytical
analytical methods are working properly. A good practice is Atomic Spectrometry. VCH Publishers, New York, 660 pp.
Pasternack, G., Brown, K., 2006. Natural and anthropogenic geochemical signatures
to include quality controls checks (i.e., known concentra- of floodplain and deltaic sedimentary strata, Sacramento – San Joaquin Delta,
tion standards) when analyzing samples to check for instru- California, USA. Environmental Pollution 141, 295–309.
ment drifts. Pretsch, E., 2002. The new wave of ion-selective electrodes. Analytical Chemistry
74, 420A–426A.
Pungor, E., 1998. The theory of ion-selective electrodes. Analytical Sciences 14,
249–256.
14.23.11 Summary Rayment, G.E., Higginson, F.R., 1992. Australian Laboratory Handbook of Soil and
Water Chemical Methods. Inkata Press, Melbourne, 330 pp.
Wet chemical methods are widely used in geomorphological Rayment, G.E., Lyons, D.J., 2010. Soil Chemical Methods – Australasia. CSIRO
studies. They range from pH and EC measurements to the Publishing, Collingwood, 520 pp.
Rowell, D.L., 1994. Soil Science: Methods & Applications. Printice-Hall, Singapore,
analysis of elements in water and soil/sediment extracts. 350 pp.
Techniques presented allow for the analysis of single elements Salt, A.N., Vora, A., 1991. Calibration of ion-selective microelectrodes for use with
as well as the simultaneous determination of multianalytes. high levels of interfering ions. Journal of Neuroscience Methods 38, 233–237.
Although advances have been made in the development Skoog, D.A., Leary, J.J., 1992. Principles of Instrumental Analysis, Fourth ed.
Saunders College Publishing, USA, 700 pp. þ 6 App.
of analytical techniques (e.g., Laser Ablation-ICP-MS), wet
Smith, R.E., 1990. Ion Chromatography Applications. CRC Press, Boca Raton, 177 pp.
chemical methods described above are suitable to fulfill the Sparks, D.L., 1996. Methods of Soil Analysis. Part 3. Chemical Methods. Soil
requirements of most geomorphological studies. Science Society of America Book Series No. 5. Soil Science Society of America
and Agronomy Society of America, Madison, WI, 1264 pp.
Standard Methods for the Examination of Water and Wastewater, 2005. American
Public Health Association, Twenty-first ed. American Water Works Association,
References and Water Environment Federation, USA.
Standard Methods for the Examination of Water and Wastewater, online edition.
Barron, J.J., Ashton, C., Geary, L., 2010. The Effects of Temperature on pH http://www.standardmethods.org/
Measurement. Reagecon Diagnostics Ltd., www.reagecon.com/pdf/ Sumner, M.E., 1994. Measurement of soil pH: problems and solutions.
technicalpapers/Effects_of_Temperature_on_pH_v4-_TSP-01-2.pdf [accessed 23 Communications in Soil Sciences and Plant Analysis 25, 859–879.
September 2010]. Varma, A., 1991. Handbook of Inductively Coupled Plasma Atomic Emission
Bates, R.G., 1962. Revised standard values for pH measurements from 0 to 90 1C. Spectroscopy. CRC Press, Boca Raton, 380 pp.
Journal of Research of the National Bureau of Standards – A, Physics and Wen, M., Shi, N., Qin, Y., Wang, C., 1998. Developments in the analysis of fluoride
Chemistry 66A, 179–184. 1995–1997. Fluoride 21, 74–80.
 Wet Chemical Methods 281

Biographical Sketch

Dr Catherine Chagué-Goff is a Senior Research Fellow at the Australia-Pacific Tsunami Research Centre, University
of New South Wales and at the Australian Nuclear Science and Technology Organisation. Her current research
interests include recent and past environmental changes, with particular emphasis on chemical and isotopic
signatures.

Henri Wong is the Manager of the Inorganic Analytical Laboratory in the Institute for Environmental Research, at
the Australian Nuclear Science and Technology Organisation. He is involved in a number of research projects
ranging from studies of groundwater to tracing of contaminant sources. He has 15 years experience working with
ICP-AES and ICP-MS.
 14.24 Use of Sedimentary-Metal Indicators in Assessment of Estuarine
System Health
G Birch, The University of Sydney, Sydney, NSW, Australia
r 2013 Elsevier Inc. All rights reserved.

14.24.1 Introduction 282

14.24.1.1 Use of Sediments 283
14.24.2 Methodology 284
14.24.2.1 Normalization Techniques 284
14.24.3 Magnitude of Human-Induced Change 285
14.24.3.1 Determination of Magnitude of Anthropogenic Change 286
14.24.4 Benthic Risk 286
14.24.5 Use of Sedimentary-Metal Indicators in Estuarine Health Assessment 287
14.24.5.1 Magnitude of Human-Induced Change 287
14.24.5.2 Risk of Possible Biological Stress 287
14.24.6 Lake Macquarie – A Case Study 288
14.24.6.1 Temporal Change 288
14.24.6.2 Sediment Status and Quality by Predictive Modeling 289
14.24.7 Conclusions 289
References 289

Abstract

Normalised-metal data provide key information regarding the ‘‘health‘‘ of an estuarine system, for example, the pristine
condition, the magnitude of human-induced change, history of anthropogenic influence, future trends, and time required
to reach managerial targets in sediment quality. Total-sediment metal concentrations are used to assess increasing risk to
sediment quality, i.e., the ability to support a healthy benthos. Sedimentary metals provide an inexpensive and easily
acquired means of making an initial assessment of estuarine health and provide the framework for more detailed and
expensive investigations. Sediments should be included in a batch of indicators in estuarine evaluation where key indicators
are assessed in a weight-of-evidence matrix.

14.24.1 Introduction 2001). It is therefore critical that contamination of estuarine

environments is assessed accurately and comprehensively to
Estuaries are the most diverse and biologically productive determine the health of these sensitive systems.
ecosystem within the marine environment resulting in highly An assessment of estuarine health requires determination
elevated conservation and ecological values (Boyes and Elliott, of persistence, bioaccumulation, and toxicity (PBT), as well as
2006). Moreover, more than half the world’s population lives the fate and transport of contaminants in bottom sediments to
within 100 km of a shoreline, and models predict that this be known (Simpson et al., 2005). Integrated coastal man-
trend will increase with time (Niemi et al., 2004; Birch and agement is becoming a worldwide practice, however, limited
McCready, 2009). Estuaries also support industrial, com- knowledge of complex relationships regarding the health of
mercial, and recreational activities and have become a focus for sensitive ecosystems continues to confound managers and the
the development of large capital cities around the world with public (Clarke et al., 2008).
large and rapidly growing populations (Birch et al., 2010). As a Definition of appropriate ecosystem indicators to assess
consequence, estuarine environments commonly experience environmental health is a complex issue and one which is
substantial and increasing anthropogenic stress. Pollution of currently gaining considerable attention globally. The Health
estuaries has resulted in adverse affects to the bottom sediment of the Ocean Module of the Global Ocean Observation System
and the water column, which threatens benthic species, pelagic is encouraging the adoption of a common integrated strategy
populations, and migratory species (Chapman and Wang, for determination and quantification of the effects of human
activities on the marine environment (Magni, 2003; Magni
et al., 2004). The UNESCO-sponsored benthic indicator group
of the United Nations Conference on Environment and De-
Birch, G., 2013. Use of sedimentary-metal indicators in assessment of
velopment is currently considering a number of options in the
estuarine system health. In: Shroder, J. (Editor in Chief), Switzer, A.D.,
Kennedy, D.M. (Eds.), Treatise on Geomorphology, Academic Press, San adoption of ecosystem health indicators which may be used
Diego, CA, vol. 14, Methods in Geomorphology, pp. 282–291. globally. These indicators, of necessity, need to be easily and

282 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00392-4

 Use of Sedimentary-Metal Indicators in Assessment of Estuarine System Health 283

inexpensively used to accurately and appropriately measure As shown in the forthcoming section, bottom sediments
many biological, chemical, and physical processes. A weight- are generally not affected by in situ natural temporal and
of-evidence approach is considered most effective for assess- spatial variance, or not affected to the same extent as many
ment of multiple indicators in assessing ecosystem health ecosystem indicators currently being employed would indi-
within a decision framework. cate. The use of sedimentary indicators in assessment of en-
Ecosystem indicators are commonly compromised by sig- vironmental health has been poorly understood and greatly
nificant natural spatial and temporal variability, as well as undervalued and underutilized.
being influenced by anthropogenic stress. It is characteristic-
ally difficult to discriminate between natural- and human-
induced stress under complex conditions and this con-
14.24.1.1 Use of Sediments
founding commonly results in inappropriate and erroneous
ecosystem assessment (Hogg and Norris, 1991). For example, The water column has traditionally been the preferred media
seagrass distributions, which are commonly used as an eco- for indicators of estuarine health assessment. However, the
system health indicator, vary temporally under natural con- water column is dynamic and highly variable under short- and
ditions, and the original distribution and condition of this long-term scales and considerable numbers of samples are
flora are generally unknown and therefore deviation from the required to spatially and temporally assess the characteristics
‘preanthropogenic’ condition cannot be determined. Large of this medium with confidence (Birch and Olmos, 2008). To
natural flood events can cause widespread changes to seagrass determine the concentration of physiochemical and biological
distributions. These floods are arguably mainly natural, and components of the water column is thus problematic. Gen-
adverse changes in seagrass distributions are considerable, but erally, concentrations of these media are very low and analyses
predominantly not anthropogenic. Turbidity is another com- are therefore difficult and expensive, which makes interpret-
monly used indicator of estuarine health and it has been ation of water quality data costly and difficult (Budd and
shown that turbidity significantly varies both spatially and Lester, 1994; Birch and Taylor, 2000a, b). Analysis of floral and
temporally under natural conditions for any one location faunal tissues is variable between animals, genders, and organs
within an estuary (Birch and Olmos, 2008). Turbidity is in- and it can also change with age. Increasingly, therefore, sedi-
fluenced mainly by particulate material within the water col- ments are being used preferentially to assess the health of
umn commonly derived from remobilization of bottom marine environments, rather than traditional media, such as
sediment by wind-induced resuspension. Turbidity can water and biological tissue.
therefore change naturally on a diurnal scale and along with it Sediments have great advantage over other materials in
many other water column indicators, such as nutrients, con- assessment of environmental health in that they faithfully
taminants, and dissolved oxygen (Dennison and Abal, 1999). record and time-integrate environmental events within the
Another common example is algal blooms, which may bloom aquatic system, commonly referred to as ‘the memory of
under natural conditions of upwelling or may be induced by sediments.’ The integration of environmental events over time
anthropogenic contributions of nutrients to the system. Dif- makes sediments useful for acquiring spatial and temporal
ferentiating which of these sources and processes are natural is information (Birch, 2007; Birch et al., 2008a, b). Sediments
extremely difficult. A further example of confounding of eco- provide an extensive habitat and a large storage for con-
system health assessment by natural processes is the complex taminants and therefore have an extensive influence on the
relationship between mangrove and salt marsh distributions. biological health of an estuary. Sediments have a large effect
An increase in mangrove distribution within a tributary of the on the quality of the overlying water column and the chemical
Sydney Harbor estuary, Australia, was originally considered as characteristics of interstitial water through physical (resus-
an improvement in estuarine health until it was shown that pension), biological (bioturbation), and chemical (desorption
mangroves were not present in this location at the time of and benthic diffusion) processes. The majority of toxicants are
European settlement in 1788 (McLoughlin, 1985). An increase adsorbed onto fine-grained materials and thus sediments also
in mangrove stands in Port Hacking, south of Sydney, was also play a major part in the transport of contaminants as sus-
considered to be a sign of an improved environment, until it pended particles in the water column. Contaminants also have
was shown to be at the expense of ambient salt marsh. De- an affinity for finely disseminated organic matter, which be-
ciding whether, and to what extent, a change in estuary con- haves hydraulically in a similar manner to detrital particles.
dition is due to anthropogenic influence is made even more Sediments therefore influence food and feeding habitats of
complex when the estuary is intermittently open and closed benthic and pelagic animals, as well as uptake processes. Thus,
under natural situations (Dye, 2005, 2006). The pristine sediment quality influences, to a large degree, biodiversity and
condition of an estuary is seldom known and therefore to ecological health in estuarine systems. It has been shown that
determine whether present distributions of flora and fauna sediment quality plays a more important part in determining
indicate good or poor health is difficult to judge. The de- the health of the overlying water column than the water itself
termination of estuarine condition therefore, requires baseline and has a larger impact on the survival of the benthic and
information from historic data in order to determine the pelagic organisms (Forstner and Calmano, 1998). Because the
magnitude of human-induced change and to assess the mag- majority of contaminants are transported via particles, sedi-
nitude of adverse effects on an estuarine system. The pristine ment size also plays an important role in source identification
condition on high-population seaboards is generally un- and determination of dispersion pathways of toxicants.
known and control sites for ecosystem assessment are com- Sediment quality is the ability of bottom material to sup-
monly unavailable. port a healthy ecosystem. The health of an ecosystem may be
284 Use of Sedimentary-Metal Indicators in Assessment of Estuarine System Health

assessed by determination of the abundance and structure of 14.24.2.1 Normalization Techniques

biological communities either in the field or in the laboratory
through undertaking bioassays. However, these processes are Chemicals are partitioned primarily to clay-sized particles
time consuming and expensive, which has led to sediment- (o4 mm) in sediments due to an increased surface charge
quality guidelines (SQG) being developed as a surrogate tool (cation exchange capacity) and exponential increase of surface
in estuarine health assessment. Anthropogenic chemicals area with decreasing particle size (Forstner, 1982; Bubb et al.,
comprise a large number of contaminant groups, such as or- 1990). Disseminated organic matter is commonly associated
ganochlorine pesticides, polycyclic aromatic hydrocarbons with fine sediment due to its similar hydraulic behavior and as
(PAH), polychlorinated biphenyls, and a range of metal spe- a result these materials also absorb contaminants to their
cies, such as cadmium (Cd), chromium (Cr), cobalt (Co), surfaces. Hence, contaminants are commonly associated with
copper (Cu), nickel (Ni), lead (Pb), and zinc (Zn). Although a fine-grained organic-rich sediments. It is therefore necessary to
range of chemicals might persist in estuarine environments, remove the confounding information introduced by variable
metals are ubiquitous and of particular concern due to their grain size through some form of normalization process, usu-
environmental persistence. They are also easily recycled in ally physical fractionation or elemental normalization. The
biogeochemical systems and consequentially are of elevated coarse fraction (462.5 mm) of sediments may be coated with
ecological risk (Alkarkhi et al., 2009). Metals are ever present hydrous Mn and Fe oxides, plus organic substances that pro-
in estuarine environments due to high loads derived from vide active surfaces for scavenging organic and metallic sub-
urban expansion and growing road networks. Metals are stances from the water column (Singh and Subramanian,
therefore given emphasized importance in assessment of 1984). However, this mechanism is less important than
ecosystem health for coastal assessment in the present work. adsorption to clay minerals and organic fractions.
The full range of chemical analysis incorporating organic Although many sizes (4, 16, 53, 60, 75, 100, and 300 mm)
and metallic contaminants undertaken by the US Environ- have been used in the past to physically segregate sediment,
mental Protection Authority in its environmental monitoring the most commonly used size for sieving is currently 62.5 mm
and assessment program, as well as in the National Oceano- (Birch et al., 2008a, b). Separation at this size has a number of
graphic and Atmospheric Administration (NOAA) National advantages. It approximates the interface between bed load
Status and Trends Program are beyond the financial capabil- and suspended particulate transport in aquatic systems and
ities of most nations (Long et al., 1995a, b; Hyland et al., therefore provides information regarding the dispersion
2000, 2003). In Australia, as in many other countries, a less mechanism for contaminants. Size fractionation provides
ambitious scheme for assessing the health of estuarine systems enough material for analysis, is rapid, and inexpensive. Al-
is therefore required. In the current work, a strong case is though some diluent associated with the silt-size fraction
made for sediment chemistry as an appropriate indicator of (15–62.5 mm) remains, enough of the nonadsorbing material
ecosystem health. Sedimentary metals are being promoted in is removed to significantly reduce the confounding introduced
the current dialog as an environmental indicator as these by variable grain size. The disadvantage of wet sieving is that it
materials provide an indication of the magnitude of human- is time consuming. Wet sieving is undertaken using milli-Q or
induced change, identify the pristine condition, and deter- ambient water. Consideration is often given to the different
mine risk posed by contaminants to animals in estuarine physiochemical characteristic of milli-Q water to ambient
sediments. water taken at the time and location of sampling, however,
contaminant desorption has been shown not to be significant
in this process (Birch, 2003). Drying and crushing of sediment
14.24.2 Methodology before or after fractionation is not recommended as it result in
disaggregation of coarse particles and artificially increase the
Water and tissue have been used preferentially in environ- mass of what would normally be reactive fine material. Drying
mental monitoring ahead of sediments due to a lack of uni- and sieving the sediment have also been shown to result in
formity in analysis and difficulties in interpretation of increased concentrations of Fe and Mn (Mantei et al., 1993),
sedimentary chemical data. The major problem in interpret- whereas freeze drying results in aggregation of fine sediment
ation of these data is the confounding effects of variable grain that artificially increases the coarse fraction in the sample
size, or more correctly, variable grain surface area. (Barbanti and Bothner, 1993).
Grain size controls contaminant concentrations in bottom The elemental normalization procedure involves normal-
sediments to a greater extent than proximity to source (Loring, ization of anthropogenic metals to a conservative element not
1991; Birch, 2003). In recent times, various forms of nor- associated with human activities and which acts as a proxy for
malization have been used to reduce the confounding effects fine-grained material, particularly clays. In order for this
introduced to chemical data by variable grain size. Variable technique to work successfully, it must be demonstrated that
grain size imposes variations in chemical data such that the the conservative element is strongly correlated with the fine
determination of source, dispersion of the magnitude of an- component of the sediment. Although many elements have
thropogenic change, and the pristine condition cannot be been used in the normalization process, Al (Hirst, 1962;
made without a process of normalization (Forstner and Windom et al., 1989) and Fe (Hornung et al., 1989) are the
Wittmann, 1979; Loring, 1991). Various techniques have been most commonly used in this procedure. Other elements are Li
used to normalize sedimentary chemical data; however, in the (Loring, 1989), Sc and Cs (Ackermann, 1980; Ackermann
present work discussion is confined to size normalization and et al., 1983), Co (Matthai and Birch, 2001), and organic car-
elemental normalization techniques. bon (Luoma and Phillips, 1988). There are problems related
 Use of Sedimentary-Metal Indicators in Assessment of Estuarine System Health 285

to the use of Al and Fe as normalizing elements. Aluminium to an adjacent estuary; and estimation of source material, such
has been shown to be regionally specific (Windom et al., as catchment soils and shales and the use of global mean
1989) as it is associated with Al-feldspars and in rock debris. concentrations. Use of pristine estuarine sediment within the
Iron has also been shown to be a common surface oxide due same ecosystem is generally not possible, considering the
to diagenetic mobilization and is affected by redox potential global nature of atmospheric contamination, and is especially
(Singh and Subramanian, 1984). problematic for highly populated seaboards of many nations.
A post extraction normalization (PEN) procedure under- Pristine fluvial samples taken in the uppermost, unimpacted
takes the size separation phase at the end of the digestion parts of catchment have the added complication of significant
process instead of at the start (Birch, 2003), such as used in the physiochemical changes that take place at the freshwater–
size-normalization methodology. The PEN technique uses saline water interface, whereas the chemistry of catchment
total sediment that is digested in a moderately strong acid soils and shales is considerably divorced from the adjacent
mixture (normally aqua regia). The solute is removed into estuarine environment. Global mean concentrations do not
vials for analysis, whereas the residue is washed, sieved, and consider regional variability and are increasingly regarded as
dried as in the size normalization technique. The residual an illusion (Marschullat et al., 2000).
fraction is the diluent, and the total absorbed fraction is cal- Preanthropogenic metal concentrations determined from
culated from the original weight of the sample. The residue is sedimentary cores are taken below a distinct break in the down-
comprised mostly of nonreactive silt and sand-sized quartz, hole contaminant profile. It is important to determine that no
which acts as a diluent. This procedure has the advantage of physical, biological, or chemical remobilization has taken place
capturing anthropogenic metals associated with Fe and Mn within the sediment column when determining background
oxide coatings on coarse grains and does not require the time concentrations using cores. Two types of data must be obtained
requirement of sieving and undertaking two analyses, i.e., fine- from the use of sedimentary cores in the assessment of an-
grained material and total sediment, for a full environmental thropogenic change in estuarine environments. High-reso-
assessment. lution sampling, especially near the upper part of the core,
provides valuable information on recent changes in anthro-
pogenic activity in the adjacent catchment and how the estuary
14.24.3 Magnitude of Human-Induced Change is responding to these changes. A second requirement of sedi-
mentary cores is determination of the depth of sediment at
Determination of the magnitude of environmental change which the onset of contamination took place and the back-
resulting from human activity requires that the pre- ground concentration itself. This requires that the core inter-
anthropogenic or background condition be known. The sects and penetrates the preanthropogenic section of sediment.
background concentration for chemicals created entirely by For this information, a deep core is generally required and
humans, such as chlorinated biphenyls or organochlorine sampling is emphasized toward the bottom of the core to
pesticides is zero. However, preanthropogenic concentrations provide as many data points for the assessment of background
of chemicals that have both anthropogenic and natural com- concentration as possible. The background concentration is a
ponents, such as metals and PAHs, require that the human population and is characterized using a mean and standard
component of the chemical be determined. A global reference, deviation. Experience has shown that the positive standard
such as materials from the Earth’s crust (crustal abundance, deviation for this population equates to approximately 1.5
global shale), has been used as an artificial background ref- times the mean background concentration.
erence (Bowen, 1979). However, the use of such material as a Samples taken from cores may be dated using 210Pb and
137
background reference does not take into account regional Cs techniques (Goldberg, 1963; Oldfield and Appleby,
variability of parental material and basin-specific soils char- 1984) and are interpreted using basic assumptions of the
acteristics. Background concentrations must be determined for constant rate of supply (CRS) and constant initial supply
individual chemicals and are defined in space and time. Pre- (CIC) models (Oldfield and Appleby, 1984; Appleby, 2001).
anthropogenic concentrations of a contaminant are required The basic assumption of the CRS model is that the rate of
to determine anthropogenic contributions for legal purposes, isotope fallout is constant and reflects a constant flux of 210Pb
to establish the magnitude of environmental impact and to from the atmosphere. The CIC model assumes that the sedi-
determine the loading of anthropogenic chemicals to a site as ments had the same unsupported Pb at the time the materials
well as to estimate total inventories. Background concen- were deposited regardless of the sedimentation rate. The iso-
trations have been determined broadly by two methodologies; tope 137CS is the result of atmospheric nuclear bomb testing,
empirical and statistical (Marschullat et al., 2000). Empirical which achieved a maximum in the year 1963 at which time
techniques use geochemical data from uncontaminated sedi- atmospheric testing was the highest (Appleby, 2001). In-
ment derived from subsurface sediment data or from pristine dependent validation of isotope chronology is an advantage
control areas of the same ecosystem. Statistical techniques and commonly uses surrogates, such as pollen, diatoms, and
separate the data into those which are contaminated and those historical records. As with surficial sediment chemical data,
that are deemed pristine. core contaminant distributions are also confounded by vari-
Empirical methods use five basic approaches in deter- able grain size, and some form of normalization procedure
mining preanthropogenic concentrations of contaminants; needs to be applied to determine preanthropogenic metal
use of cores to penetrate the sediment below levels of human concentrations.
influence; use of pristine estuarine areas within the same Statistical methods used to determine background concen-
ecosystem; use of pristine fluvial samples which are discharged trations aim to separate preanthropogenic from contaminated
286 Use of Sedimentary-Metal Indicators in Assessment of Estuarine System Health

data in a single data set to establish a threshold which sep- (Müller, 1979) and an index of enrichment is (Müller, 1981)
arates these two materials. Most statistical techniques use as follows:
identification of deviation in normal and log normal distri-
butions of metal data. This requires identification of the 0 Igeoo0 Unpolluted
1 0oIgeoo1 Unpolluted to moderately polluted
presence of two subpopulations within the data. Anomalous
2 1oIgeoo2 Moderately polluted
data are assumed to be non-normally distributed with a higher 3 2oIgeoo3 Moderately to strongly polluted
mean concentration to that of natural material. Differen- 4 3oIgeoo4 Strongly polluted
tiation of the populations is made using regression and modal 5 4oIgeoo5 Strongly polluted
analysis. The problem of using statistical techniques is that the 6 5oIgeo Extremely polluted
data may not be normally or log-normally distributed due to
the presence of multiple processes controlling elemental dis-
tribution (Reimann and Filzmoser, 2000). Statistical methods A linear regression technique has also been used to obtain
used for estimating anthropogenic populations include cu- information on enrichment that is invariant to grain size
mulative frequency, regression, modal analysis, and outlier (Loring and Rantala, 1992). For this technique to be valid, the
testing (Marschullat et al., 2000). linear regression must have a 95% confidence level with a high
degree of significance. The enrichment factor is calculated as
the ratio between the real and predicated value of y, where the
14.24.3.1 Determination of Magnitude of Anthropogenic predicated value is within a range of 1 þ 2.
Change
Commonly, the extent of human-induced change is estimated
14.24.4 Benthic Risk
by expressing current contaminant concentrations as enrich-
ment over preanthropogenic or pristine concentrations. Enrich-
Surficial sediments in coastal waterways commonly display an
ment factors are defined as present-day metal concentrations
elevated concentration of a wide range of chemicals. However,
divided by background concentrations. As discussed previously,
chemical concentration per se does not provide an effective
in order for this approach to be valid, it requires that both the
means for determining potential adverse affects on benthic
current data and preanthropogenic concentrations be size-
resources. Persistence, toxicity, and bioaccumulation infor-
normalized. Should total sedimentary chemistry be used for
mation are required to assess the biological significance of
estimating the enrichment factor, a normalizing element must
contaminants bound to sediments. Contaminated sediment
be used to reduce the confounding introduced by variable grain
has an adverse effect on the health and reproduction of
size (Buat-Menard, 1979) (see eqn [1]):
vegetation and bacteria within the benthic environment
(MacDonald et al., 2003). Organisms that rely on a healthy
Meff
EF ¼ ð1Þ benthic environment provide the foundation of a healthy
Mem
food chain. Sediment provides a protective habitat for juvenile
species, which is vital to the breeding of many species. Con-
where EF is the enrichment factor, Meff the metal concen- taminants may bioaccumulate within aquatic animals and be
tration in sediment fine fraction (o62.5 mm), and Mem the bioconcentrated in higher trophic levels and pose a risk to
mean metal concentration in mudstone, or more recently human consumers (Baker and Kravitz, 1992). Testing for the
background concentrations – see below (Szefer and Skwarzec, significance for sediment-bound contaminants is time con-
1988). suming and expensive, and SQG have been developed to make
The technique developed to estimate anthropogenic im- preliminary assessments of sediment toxicity when direct
pact on sediments using a normalizing element (Al) to reduce biological effects information is not available. Although
the effects of variable grain size is calculated using the fol- many approaches are used in the development of SQGs,
lowing equation (Buat-Menard, 1979): empirical methods employing matching sediment chemistry
Xs =Als and biological effects data have been used to determine
EF ¼ ð2Þ the adverse outcomes of contaminants on biological popu-
Xb =Alb
lations. Numerical-effects-based SQGs are now in common
use as a screening management tool for sediment chemical
where X is the metal concentration, s is the sample, and b is the
data that identifies and prioritizes contaminants of concern
background value.
and areas where potential problem may be identified, such as
The level of contamination for a sample has also been
has been undertaken in Sydney Harbor, Australia (Birch et al.,
calculated using the geoaccumulation index (Müller, 1979).
2008a, b).
Geoaccumulation index, where the degree of contamin-
Guidelines for sediment quality are separated into three
ation is calculated using the following formula:
broad groups: effects based; equilibrium partitioning and
Igeo ¼ log2 ðCn=1:5BnÞ use of background concentrations. The use of background
concentrations has been largely superseded, and the empir-
where C is the measured concentration of metal and B is the ical effects-based approach has been shown to be equal or
geochemical background concentration of the average con- slightly superior to other major SQG partitioning-based
tinental shale. A factor of 1.5 is used to include possible models (Simpson et al., 2005). There are a number of
variation of background values due to lithogenic effects earlier effects-based guidelines based on spiked sediments
 Use of Sedimentary-Metal Indicators in Assessment of Estuarine System Health 287

producing no biological responses for benthic organisms, A monitoring and assessment program being conducted on
which have established a cause and effect relationship the central New South Wales (NSW) coast of Australia is
(USEPA, 1989). The apparent-effects threshold (AET) scheme providing a cost-effective, integrated, and regionally consistent
identifies sediment concentration above which a statistically assessment of estuarine health. The scheme provides the
significant biological effect is observed for a given chemical. magnitude of human-induced change for a water body and
Screening level concentrations (SLC) provide the highest assesses the risk of adverse effect to benthic populations posed
concentration of a contaminant that can co-occur with ap- by sedimentary metals. The program uses sedimentary-metal
proximately 95% of a specific infauna. Two levels are used, a indicators as a basis for making judgments on anthropogenic
low level that is the lowest concentration of a toxic effect and stress and deviations from the pristine condition.
a severe level which results in fatality for most benthic or- The methods outlined previously to determine the mag-
ganisms. Perhaps the most commonly used SQG for estuar- nitude of human-induced change in the estuarine environ-
ine and marine environments is that developed by the US ment are utilized in the current investigation, that is the
NOAA (Long and Morgan, 1990; Long et al., 1995a, b, 2000). pristine or background metals concentrations is determined
Concurrent sediment chemical and biological data from the using sedimentary cores, and SQGs are used to assess risk to
laboratory and field for a variety of techniques and benthic benthic health.
end points were combined in a table (Long and Morgan,
1990). Chemical concentrations and biologically observed
14.24.5.1 Magnitude of Human-Induced Change
effects were sorted and the lower tenth and fiftieth percentiles
were used to identify the effects range low (ERL) and the The extent of human activity in the estuarine environment is
effects range median (ERM) concentrations. The former being determined using size-normalized sedimentary-metal
level identifies the concentration below which adverse data obtained from high-resolution short cores and deep cores
biological effects are seldom observed, whereas the latter for determining background concentrations. The magnitude
level distinguishes concentrations above which adverse bio- of anthropogenic change is estimated by expressing surficial
logical effects occur frequently. Concentrations between the metal concentrations as enrichment factors over pre-
two levels identify concentrations of irregular biological anthropogenic values. Low and consistent metals data at the
response. base of the core are interpreted to be pre-European based on
Contaminants commonly do not occur as single chemicals isotope dating in several estuaries in this region. Remobiliza-
within marine sediments, and a scheme of mean ERM quo- tion by chemical, physical, and biological mechanisms has
tients (MERMQ) has been developed to assess the effects of been shown not to be problematic in these NSW estuaries
chemical mixtures in aquatic sediments (Long et al., 2000, (Olmos and Birch, 2010). Once the background concentration
2006). This scheme requires normalizing the concentration of has been established for the three metals, enrichment factors
each chemical with respect to its ERM value, summing the were determined for each sample location. Surficial mapping
quotients for each substance and dividing by the number of of the enrichment factors provided an estimate of the mag-
chemicals for which guidelines are available. The probability nitude of human-induced change for the surface of the estu-
of toxicity for mixtures of chemicals has been determined in ary. A mean enrichment quotient for the three metals was
relation to the MERMQ using amphipod bioassays. This as- determined by summing the enrichment factors for individual
sociation should be used with caution outside of the USA metals and dividing by three. Spatial distributions for metals
where this relationship was originally established due to were determined by kriging interpretation using the Geosta-
differing benthic populations and sedimentary chemical tistical Analyst tool in the program ArcGIS.
conditions.
14.24.5.2 Risk of Possible Biological Stress
Contaminants in estuarine sediments commonly covary spa-
14.24.5 Use of Sedimentary-Metal Indicators in tially in waterways dominated by stormwater contamination.
Estuarine Health Assessment An analysis of metals data from 38 NSW estuaries has shown
that three metals, that is Cu, Pb, and Zn, are closely correlated
In order to make a detailed determination of sediment toxi- to the full suite of nine metals that are routinely analyzed in
city, a full triad assessment would be required. The ‘triad’ ap- these investigations and thus, only these three metals have
proach requires that sediment chemistry for a comprehensive been used in assessment of these estuaries (Olmos and Birch,
range of chemicals, ecotoxicology using a batch of tools and a 2010). A study of the relationship between metals and organic
detailed benthic community assessment be undertaken on the contaminants has shown a surprisingly strong correlation
same sediment. Such schemes are undertaken by the US EMAP between these three metals and other contaminants in estu-
and NS&TP programs, however, these investigations are arine sediment. In the current work, it is not assumed that
extremely expensive and require considerable professional effects of these three metals can compensate for a full chemical
judgment (Long et al., 1995b; Hyland et al., 2000). The investigation, however, for stormwater-dominated systems,
complexity and expense associated with such undertakings are metals, and other contaminants appear to covary. However,
commonly beyond the resources of many institutions. A less such a relationship between metals and other contaminants
expensive approach is needed to determine risk of adverse may not be surprising as this association is the basis of many
biological effects due to sedimentary contaminants for most stormwater prediction models (Birch and Olmos, 2009).
regional surveys. However, estuarine environments dominated by industrial
288 Use of Sedimentary-Metal Indicators in Assessment of Estuarine System Health

discharge, especially of organic contaminants and catchments

of predominantly agricultural and pastoral nature will not
conform to such a relationship due to fertilizer and biocide
applications. Australia and New Zealand adopted the NOAA
approach to SQGs in 2000 and in the current scheme
MERMQs were calculated for Cu, Pb, and Zn to provide the
risk of adverse effects to benthic animals posed by these
chemicals. Although this approach does not reproduce the
same results as a full SQG analysis of metal and organic
contaminants, it does indicate increasing risk of benthic stress
posed by these chemicals in the sediment. NSW

14.24.6 Lake Macquarie – A Case Study

(a)
Lake Macquarie is an extensive (120 km2), shallow (max.
11 m) estuary north of Sydney on the NSW coast of Australia
(Figure 1). The opening to the waterway is narrow, resulting in
poor flushing and minimal oceanic exchange. It has long been
known that sediments in the lake are highly elevated in Cd,
Pb, and Zn due to discharge from a lead–zinc smelter in the
north-east of the estuary (Roy and Peat, 1975; Batley, 1987).
Sediment cores were taken from three locations within the
estuary and analyzed as described previously in this work.
Metal concentrations in the core in the southern part of the
estuary showed low and irregular concentrations throughout
the profile, whereas metal concentrations were significantly
higher in the upper part of a core in the north-east of the
estuary (Olmos and Birch, 2010).
Determination of the magnitude of human-induced
Newcastle
change was made through estimating background concen-
trations from the three cores obtained from the estuary.
Background concentrations were estimated at 0.470.2,
Lake
8.371.7, 10.174.1, and 30.578.2 mg g1 for Cd, Cu, Pb, and Macquarie
Zn, respectively. This provided present-day maximum enrich-
ment factors of 600, 110, 102, and 18 times for Cd, Pb, Zn and
Cu, respectively. Lead and Zn significantly increase over
background concentrations at approximately 0.97 m sediment
depth, which correlated with isotope dating for the initial
start-up of the smelter in 1897. Highest metal concentrations
were located in the north-east of the estuary with a strong
southward gradient in decreasing metal values down the axis
Sydney
of the estuary. A significant decline in metals concentration
0 30 60
was observed toward the upper part of the core indicating that
shutdown of the smelter has resulted in declining metals
Km
discharge to the estuary and significant decrease in metal
concentrations (declining relaxation rates). (b)
Figure 1 Location of Lake Macquarie in relation to Sydney and
Newcastle.
14.24.6.1 Temporal Change
Metal concentrations of surficial sediments were measured in For the majority of the lake, the absolute decrease in Cd,
1975 and again 2003 in Lake Macquarie, using similar field Pb, and Zn over the 28-year period was substantial, especially
and analytical techniques, enabled the spatial change in in the northwest in the vicinity of the shut-down smelter, but a
chemical conditions in the bottom environment for the entire surprising outcome was an increase in Cu in surficial sedi-
estuary to be made using a geographic information system ments in the southern part of the lake, presumably due to
(GIS). Metal distributions were analyzed using the Raster coal-fired power stations in this area of the catchment. These
Calculator tool (Spatial Analyst), which allowed the use of changes in surface sediment metal concentrations were veri-
arithmetic operators for subtraction of two raster layers (ESRI, fied by down-core metal distributions, and ages were con-
2001) for calculation of absolute change for each metal. firmed by isotope dating.
 Use of Sedimentary-Metal Indicators in Assessment of Estuarine System Health 289

14.24.6.2 Sediment Status and Quality by Predictive References

Modeling
Ackermann, F., 1980. A procedure for correcting the grain size effect in heavy metal
Relaxation rates for surficial sediment metal concentrations
analyses of estuarine and coastal sediments. Environmental Technology Letters
may be forward modeled using past rates of temporal change, 1, 518–527.
assuming current metal loadings and dispersion mechanisms Ackermann, F., Bergmann, M., Schleichert, G.U., 1983. Monitoring of heavy
remain constant with time in the lake. Relaxation rates were sediments in coastal and estuarine sediments – a question of grain
also used to predict the time needed for metals to reach par- size: o20 mm versuso60 mm. Environmental Technology Letters 1, 518–
527.
ticular enrichment and target levels of sediment quality (Olmos Alkarkhi, A.F.M., Ismail, N., Ahmed, A., Easa, A.M., 2009. Analysis of heavy metal
and Birch, 2010). For example, relaxation rates for Cu would concentrations in sediments of selected estuaries of Malaysia – a statistical
increase considerably if 1975–2003 conditions remain constant assessment. Environmental Monitoring and Assessment 153, 179–185.
over time, with areas 10 times greater than background levels Appleby, P.G., 2001. Chronostratigraphic techniques in recent sediments. In: Last,
W.M., Smol, J.P. (Eds.), Tracking Environmental Change Using Lake Sediments,
increasing from less than 1 km2 in 2003 to 36 km2 by 2020,
Volume 1: Basin Analysis, Coring and Chronological Techniques. Kluwer
particularly in the southern part of the water body. Relaxation Academic Publishers, Dordrecht, pp. 171–203.
rates for Pb and Zn in the northern lake area were used to Baker, B., Kravitz, M., 1992. Sediment classification method compendium. United
estimate the time required for bottom sediments to reach levels States Environmental Protection Agency 1 (1), 14–18.
acceptable to managers (two times background concentration), Barbanti, A., Bothner, M.H., 1993. A procedure for partitioning bulk sediments into
distinct grain-size fractions for geochemical analysis. Environmental Geology 21,
i.e., 10 and 28 years, respectively. Sediment quality also im-
3–13.
proves for all three metals in the estuary with time, except Cu in Batley, G.E., 1987. Heavy metal speciation in waters, sediments and biota from Lake
the south, with Pb and Zn attaining concentrations below levels Macquarie, NSW. Australian Journal of Marine and Freshwater Research 38,
where adverse effects would be experienced (ERL) in 156 and 591–660.
112 years, respectively (Olmos and Birch, 2010). Birch, G.F., 2003. A test of normalisation methods for marine sediments, including
a new post-extraction normalisation (PEN) technique. Hydrobiologia 492, 5–13.
Birch, G.F., 2007. A short geological and environmental history of the Sydney
estuary, Australia. In: Birch, G.F (Ed.), Water, Wind, Art and Debate. Sydney
14.24.7 Conclusions University Press, The Sydney University, Australia, pp. 217–246, 433 pp.
Birch, G.F., Cruikshank, B., Davis, B., 2010. Modelling nutrient loads to Sydney
estuary (Australia). Environmental Modelling and Assessment 167(1),
Anthropogenic metals are almost ubiquitous in sediments of
333–348.
coastal water bodies on highly populated seaboards of de- Birch, G.F., McCready, S., 2009. Catchment sources of heavy metal contamination
veloped countries. Sedimentary metals provide a measure of and influence on the quality of receiving basin sediments in Port Jackson,
human-induced change and related biological stress un- Australia. Science of the Total Environment 407, 2820–2835.
affected by natural temporal and spatial stress. Sediment- Birch, G.F., McCready, S., Long, E.R., Taylor, S.E., Spyrakis, G., 2008a.
Contaminant chemistry and toxicity of sediments in Sydney Harbour, Australia:
based environmental assessment provides a framework on
spatial extent and chemistry–toxicity relationships. Marine Ecology Progress
which more detailed and focused multidisciplinary investi- Series 363, 71-87.
gations can take place for large regional estuarine monitoring Birch, G.F., Olmos, M., 2009. Sediment-bound heavy metals as indicators of human
and trends programs. The data required for these studies is influence and biological risk in coastal water bodies. Journal of Marine Science
relatively easily and inexpensively acquired. 363, 71–87.
Birch, G.F., Olmos, M.A., 2008. Sediment-bound heavy metals as indicators of
An initial regional assessment of estuarine condition, using human influence and biological risk in coastal water bodies. ICES Journal of
sedimentary metals, requires two sets of data, i.e., normalized Marine Science 65, 1407–1413.
and total sediment chemistry. Normalized data provide in- Birch, G.F., Russell, A.T., Mudge, S.M., 2008b. Normalisation techniques in forensic
formation on: the pristine condition, the human component assessment of contaminated environments. In: Ball, A.S., Mudge, S.M. (Eds.),
of the magnitude of anthropogenic change, source identifi- Methods in Environmental Forensics. Taylor and Francis Group, Boca Raton, FL,
USA.
cation, dispersion, site and regional comparisons, temporal Birch, G.F., Taylor, S.E., 2000a. The distribution and possible sources of
trends, modeling future contaminant distributions, prediction organochlorine residues in sediments of a large urban estuary, Port Jackson,
of time needed to reach target contaminant levels, and stat- Sydney. Australian Journal of Earth Sciences 47, 749–756.
istical assessment. Total sediment metal concentrations are Birch, G.F., Taylor, S.E., 2000b. The use of size-normalisation procedures in the
analysis of organic contaminants in estuarine environments. Hydrobiologia 431,
used for sediment quality assessment, i.e., the ability of the
129–133.
sediment to maintain a health benthic population. Birch, G.F., Taylor, S.E., 2004. The contaminant status of Sydney harbour
A hierarchical approach should be applied to the assess- sediments. A Handbook for the Public and Professionals. Environmental,
ment of estuaries so that limited available resources can be Engineering and Hydrogeology Specialists Group (EEHSG), Geology Society of
most effectively distributed. Sedimentary-metal concen- Australia.
Bowen, H.J.M., 1979. Environmental Chemistry of the Elements. Academic Press,
trations assessed in an initial measure of condition may
London.
identify areas of concern, which may be further investigated Boyes, S., Elliott, M., 2006. Organic matter and nutrient inputs to the Humber
for other metallic and organic contaminants and focus Estuary, England. Marine Pollution Bulletin 53(1-4), 136–143.
additional studies, including persistence, PBT assessment. Buat-Menard, P., 1979. Influence de la retombée atmosphérique sur la chimie des
Sedimentary metals data should be used in a multiple-lines-of- métaux en trace dans la matière en suspension de l’Atlantique Nord. Thèse
Doctorat d’Etat, Paris VII, 434 pp.
evidence approach, where a number of key indicators are Bubb, J.M., Rudd, T., Lester, J.N., 1990. Distribution of heavy metals in the River
assessed in a weight-of-evidence matrix. Clearly, sedimentary- Yare and its associated broads. III. Lead and zinc. Science of the Total
metal indicators provide a suite of powerful tools in the Environment 102, 189–208.
management and governance of coastal water bodies, which Budd, J.M., Lester, J.N., 1994. Anthropogenic heavy metal inputs to low land river
systems, a case study. The River Stour, UK. Water, Air and Soil Pollution 78,
are generally not fully appreciated by environmental scientists.
279–296.
290 Use of Sedimentary-Metal Indicators in Assessment of Estuarine System Health

Chapman, P.M., Wang, F., 2001. Assessing sediment contamination in estuaries. Loring, D.H., 1991. Normalization of heavy-metal data from estuarine and coastal
Environmental Toxicology and Chemistry 20(1), 3–22. sediments. Journal of Marine Science 48, 101–115.
Clarke, C.S.L.M., Roff, J.C., Bard, S.M., 2008. Back to the future: using landscape Loring, D.H., Rantala, R.R.T., 1992. Manual for the geochemical analysis of
ecology to understand changing patterns of land use in Canada, and its effects marine sediments and suspended particulate matter. Earth-Science Reviews 32,
on the sustainability of coastal ecosystems. ICES Journal of Marine Science 65, 325263.
1534–1539. Luoma, S.N., Phillips, D.J.H., 1988. Distribution, variability and impacts of
Dennison, W.C., Abal, E.G., 1999. Moreton Bay Study: a scientific basis for the trace elements in San Francisco Bay. Marine Pollution Bulletin 19(9),
healthy waterways campaign, South East Queensland Water Quality Management 413–425.
Strategy, Brisbane, Australia, 246 pp. MacDonald, D.D., Ingersoll, C.G., Smorong, D.E., Lindskoog, R.A., 2003.
Dye, A.H., 2005. Meiobenthos in intermittently open/closed coastal lakes in New Development and application of sediment quality criteria for managing
South Wales: spatial and temporal patterns in densities of major taxa. Marine contaminated sediment in British Columbia. Submitted to British Columbia
and Freshwater Research 56, 1055–1067. Ministry of Water, Land and Air Protection Environmental Management Branch.
Dye, A.H., 2006. Is geomorphic zonation a useful predictor of patterns of benthic Magni, P., 2003. Biological benthic tools as indicators of coastal ecosystem health.
infauna in intermittent estuaries in New South Wales, Australia? Estuaries and Chemistry and Ecology 19, 363–372.
Coasts 29(3), 455–464. Magni, P., Hyland, J.L., Manzella, H., Rumohr, P., Viaroli, P., Zenetos, A. (Eds.),
ESRI, 2001. ArcGIS v9. Using ArcGIS Geostatistical Analyst. ESRI, Redlands, CA. 8–9 October 2004. Proceedings of the Workshop on Indicators of Stress in the
Forstner, U., 1982. Accumulative phases for heavy metals in limnic sediments. Marine Benthos, Torregrande-Oristano, Italy. UNESCO/IOC, IMC, Paris, 2005, 46
Hydrobiologia 91, 269–284. pp.
Forstner, U., Calmano, W., 1998. Characterization of dredged materials. Water Mantei, E.J., Ernst, R.L., Zhou, Y., 1993. Comparison of metal homogeneity in grab,
Science and Technology 38(11), 149–157. quartered, and crushed-sieved protions of stream sediments and metal content
Forstner, U., Wittmann, G.T.W., 1979. Metal Pollution in the Aquatic Environment. change resulting from crushing–sieving activity. Environmental Geology 22,
Springer-Verlag, New York. 186–190.
Goldberg, E.D., 1963. Geochronology with lead-210. Radioactive Dating. I.A.E.A., Marschullat, J., Ottensten, R., Reimann, C., 2000. Geochemical background – can
Vienna, pp. 121–131. we calculate it? Environmental Geology 39(9), 990–999.
Hirst, D.M., 1962. The geochemistry of modern sediments from the Gulf of Paria I. Matthai, C., Birch, G.F., 2001. Detection of anthropogenic Cu, Pb and Zn in
Geochimica Cosmochimica Acta 26, 309–334. continental shelf sediments off Sydney, Australia – a new approach using
Hogg, I.D., Norris, R.H., 1991. Effects of runoff from land clearing and urban normalization with cobalt. Marine Pollution Bulletin 42(11), 1055–1063.
development on the distribution and abundance of macro-invertebrates in pool McLoughlin, L., 1985. The Middle Lane Cove River: A History and a Future. Centre
areas of a river. Australian Journal of Marine and Freshwater Science 42, for Environmental and Urban Studies, Macquarie University, NSW.
507–518. Müller, G., 1979. Schwermetalle in den sedimenten des Rheins-Veranderungen seitt
Hornung, H., Krom, M.D., Cohen, Y., 1989. Trace metal distribution in sediments 1971. Umschan 79, 778–783.
and Benthic fauna of Haifa Bay, Israel. Estuarine, Coastal and Shelf Science 29, Müller, G., 1981. Die schwermetallbelastung der sedimente des Neckarsund seiner
45–56. nebenflusse: eine bestandssufnahme. Chemical Zwitung 105, 157–164.
Hyland, J.L., Balthis, W.L., Engle, V.D., Long, E.R., Paul, J.F., Summers, J.K., Van Niemi, G., Wardrop, D., Brooks, R., et al., 2004. Rationale for a new generation of
Dolah, R.F., 2003. Incidence of stress in benthic communities along the U.S. indicators for coastal waters. Environmental Health Perspectives 112(9),
Atlantic and Mexico coasts within different ranges of sediment contamination 979–986.
from chemical mixtures. Environmental Monitoring and Assessment 81, Oldfield, F., Appleby, P.G., 1984. Empirical testing of 210Pb-dating models for lake
149–161. sediments. In: Hayworth, E.Y., Lund, J.W.G. (Eds.), Lake Sediments and
Hyland, J.L., Balthis, W.L., Hackney, C.T., Posey, M., 2000. Sediment quality of Environmental History. Leicester University Press, pp. 93–124.
North Carolina estuaries: an integrative assessment of sediment contamination, Olmos, M.A., Birch, G.F., 2010. A novel method for measuring anthropogenic
toxicity, and condition of benthic fauna. Journal of Aquatic Ecosystem Stress change in coastal lake environments. Environmental Science and Pollution
and Recovery 8, 107–124. Research 17(2), 270–287.
Long, E.R., Ingersoll, C.G., MacDonald, D.D., 2006. Calculation and uses of mean Reimann, C., Filzmoser, P., 2000. Normal and lognormal data distribution in
sediment quality guideline quotients: a critical review. Environmental Science geochemistry: death of a myth. Consequences for the statistical treatment
and Technology 40, 1726–1736. of geochemical and environmental data. Environmental Geology 39(9),
Long, E.R., MacDonald, D.D., Severn, C.G., Hong, C.B., 2000. Classifying the 1001–1014.
probabilities of acute toxicity in marine sediments with empirically-derived Roy, P.S., Peat, C., 1975. Bathymetry and bottom sediments of Lake Macquarie.
sediment quality guidelines. Environmental Toxicology and Chemistry 19, Records Geological Survey of New South Wales 17, 53–64.
2598–2601. Simpson, S.L., Batley, G.E., Chariton, A.A., et al., 2005. Handbook for Sediment
Long, E.R., MacDonald, D.D., Smith, S.L., Calder, E.D., 1995a. Incidence of adverse Quality Assessment. CSIRO, Bangor NSW.
biological effects within ranges of chemical concentrations in marine and Singh, S.K., Subramanian, V., 1984. Hydrous Fe and Mn oxides – scavengers of
estuarine sediments. Environment Management 19, 81–97. heavy metals in the aquatic environment. Critical Reviews in Environmental
Long, E.R., Morgan, L.G., 1990. The potential for biological effects of sediment- Control 14, 33–90.
sorbed contaminants tested in the National Status and Trends Program. NOAA Szefer, P., Skwarzec, B., 1988. Distribution and possible sources of some elements
Technical Memorandum, NOS OMA 52. in the sediment cores of the southern Baltic. Marine Chemistry 23, 109–129.
Long, E.R., Wolfe, D.A., Scott, K.J., Thursby, G.B., Stern, E.,Peven, E., Schwartz, T., USEPA, 1989. Guidance manual: bedded sediment bioaccumulation tests. US EPA
1995b. Magnitude and extent of sediment toxicity in the Hudson-Raritan estuary. Pacific Ecosystems Branch, US Environmental Protection Agency Report EPA
NOAA Technical Memorandum NOS ORCA 88. National Oceanic and 600/x-89/3, ER-LN-N111, Washington, DC.
Atmospheric Administration. Silver Spring, MD, 230 pp. Windom, H.L., Schropp, S.J., Calder, F.D., et al., 1989. Natural trace metal
Loring, D.H., 1989. Lithium – a new approach for the granulometric normalization concentrations in estuarine and coastal marine sediments in the southeastern
of trace metal data. Marine Chemistry 29, 155–168. United States. Environmental Science and Technology 23(3), 314–320.
 Use of Sedimentary-Metal Indicators in Assessment of Estuarine System Health 291

Biographical Sketch

Gavin Birch is an Associate Professor (B.Sc. Hons, M.Sc., Ph.D., GDBA) graduated from the University of Cape
Town and worked in the petroleum industry for 12 years. In 1991, he returned to academia and has been at The
University of Sydney for 22 years. His interests are in the source, fate, and effects of contaminants in marine,
estuarine, and fluvial environments. Gavin’s research on catchment-estuary processes has provided novel remedial
options for effective stormwater management, recycling, and water harvesting. Professor Birch has authored more
than 150 peer-reviewed scientific publications, including three books, and has supervised more than 70 post-
graduate and Honours students.
 14.25 Microfossils in Tidal Settings as Indicators of Sea-Level Change,
Paleoearthquakes, Tsunamis, and Tropical Cyclones
BP Horton, University of Pennsylvania, Philadelphia, PA, USA, and Nanyang Technological University, Singapore
SE Engelhart, University of Pennsylvania, Philadelphia, PA, USA
AC Kemp, Yale University, New Haven, CT, USA
Y Sawai, Geological Survey of Japan, National Institute of Advanced Industrial Science and Technology, Tsukuba, Ibaraki, Japan
r 2013 Elsevier Inc. All rights reserved.

14.25.1 Introduction 293

14.25.2 Microfossils and Intertidal Environments 293
14.25.3 Microfossil-Based Reconstructions of Sea-Level Change 294
14.25.3.1 Sea-Level Index Points 297
14.25.3.2 Isolation Basins 297
14.25.3.3 Transfer Functions 299
14.25.4 Microfossils and Land-Level Change 301
14.25.5 Microfossils as Indicators of Paleotsunamis and Storms 302
14.25.5.1 Paleotsunamis 302
14.25.5.2 Paleotempestology 306
14.25.6 Summary 306
Acknowledgments 310
References 310

Glossary in surface mass load. Growth and thickening of an ice sheet

Earthquake deformation cycle For subduction zone results in subsidence of land beneath the ice mass (glacio-
coastlines, a three-part seismic–interseismic earthquake isostasy), which is compensated for by an outward flow of
deformation cycle exists: (1) between earthquakes the mantle material that uplifts a peripheral bulge around the
shallow part of the subduction zone is locked, strain ice margin.
accumulates on the fault and the coastline rises (with Paleotempestology Is the study of past tropical cyclone
subsidence offshore); (2) during the earthquake (the activity by means of geological proxies as well as historical
coseismic phase), the uplifted region quickly subsides, documentary records.
while the offshore region undergoes rapid uplift; this rapid Relative sea level Height of the ocean surface relative to
motion generates a tsunami; and (3) during the interseismic the sea floor at a given location. Relative sea levels are
period the earthquake deformation is recovered at first measured using tide gauges and reconstructed using
rapidly (the postseismic phase), and then more slowly (the information from the geological record. Geological
interseismic phase). This cycle, typically repeating over reconstructions of relative sea level are also relative in time
some hundreds of years, may be superimposed on longer- as they are measured with reference to present-day relative
term patterns of net uplift or subsidence. sea level.
Glacial isostatic adjustment Is the process whereby
Earth’s shape is modified in response to large-scale changes

Abstract

Fine-grained sediments deposited in low-energy, inter-tidal settings are an archive of sea-level change and the occurrence of
paleoearthquakes, tsunamis, and tropical cyclones. Some of the best reconstructions of these coastal processes have been
derived from microfossils, such as pollen, diatoms and foraminifera that accumulate in salt-marsh and estuarine en-
vironments. Early microfossil work in the coastal zone employed pollen as an indicator of vegetation and as a chrono-
stratigraphic marker. Use of diatoms and foraminifera has become increasingly widespread because their distribution is

Horton, B.P., Engelhart, S.E., Kemp, A.C., Sawai, Y., 2013. Microfossils in
tidal settings as indicators of sea-level change, paleoearthquakes, tsunamis,
and tropical cyclones. In: Shroder, J. (Editor in Chief), Switzer, A.D.,
Kennedy, D.M. (Eds.), Treatise on Geomorphology. Academic Press, San
Diego, CA, vol. 14, Methods in Geomorphology, pp. 292–314.

292 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00394-8

Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones 293

closely linked to tidal elevation. In this chapter, we discuss the use of microfossils in estuarine and salt-marsh sediments to
reconstruct sea level along subsiding coastlines in temperate regions. We also describe how microfossils from isolation
basins are used to reconstruct sea level along coastlines undergoing uplift. Microfossils can also estimate land-level changes
along tectonically active coasts associated with paleoearthquakes. We explain the use of transfer functions for calculating
quantitative estimates of past environmental conditions from microfossil data. Finally, we reveal how microfossils are used
to reconstruct the recurrence of tsunamis and tropical cyclones from the sedimentary deposits these high-energy events
leave behind.

14.25.1 Introduction of tropical cyclone activity and provided insight into their
relationship with global climatic changes (e.g., Donnelly and
Relative sea level (RSL) can be accurately reconstructed using Woodruff, 2007).
microfossil assemblages preserved in low-energy coastal sedi- Some of the best reconstructions of RSL change,
mentary deposits. These reconstructions are relevant to varied paleoearthquakes, tsunamis, and tropical cyclones have been
disciplines including studies of coastal evolution and wetland derived using microfossils, such as pollen, diatoms, and for-
development, ice-sheet history, and paleoclimate, and under- aminifera, because these microfossil groups are commonly
standing the environmental context of human activity (e.g., well preserved, easily detectable, and occur in high numbers,
Peltier, 2004; Lawler, 2011). The potential impacts of future sea- providing a robust sample size for statistical paleoenviron-
level rise have led to renewed focus on understanding the re- mental interpretation (Birks, 1995). The utility of microfossils,
lationship between sea level and climate changes (IPCC, 2007). and, in particular, diatoms and foraminifera, as sea-level in-
Microfossil-based reconstructions of RSL reveal how sea level dicators is underpinned by the concept that characteristic
has responded to past climate variations inferred from ice cores, assemblages have strong associations with tidal elevation and
ocean cores, and terrestrial proxies. Evidence from many parts that this modern, observable relationship has existed without
of the world indicates that past climate variability has led to change during the time represented by coastal sedimentary
variations in sea level on timescales ranging from decades to sequences. We provide examples of late Quaternary RSL
millennia (e.g., Kemp et al., 2009b). Accurate and precise RSL reconstructions from salt marshes, estuaries, and isolation
reconstructions permit tide gauge and satellite altimetry-based basins using pollen, diatom, and foraminiferal sea-level
measurements to be evaluated in the context of longer term indicators. We explore the application of microfossil-based
changes (e.g., Engelhart et al., 2009). transfer functions that have the potential to produce high-
Paleoearthquakes have been recognized using microfossils resolution records and identify subtle climate–sea-level rela-
as abrupt RSL changes in temperature regions, such as Alaska tionships and earthquake-induced coastal subsidence and
(e.g., Shennan and Hamilton, 2006), Cascadia (e.g., Atwater uplift. We evaluate the applicability of microfossils as proxies
and Hemphill-Haley, 1997), and Japan (e.g., Sawai et al., in the expanding fields of tsunami and tropical cyclone
2002). Geological indicators of prehistoric analogs to the research.
2004 Indian Ocean earthquake and tsunami have been iden-
tified in Thailand (e.g., Jankaew et al., 2008; Fujino et al.,
2009), Indonesia (Monecke et al., 2008), and Sri Lanka
(Dahanayake and Kulasena, 2008). Indeed, the 2004 Indian 14.25.2 Microfossils and Intertidal Environments
Ocean event provided new, observable insights into the
microfossil characteristics, geomorphological impact, and Pollen assemblages are commonly used to describe paleoen-
sedimentary signature of a large tsunami (e.g., Hawkes et al., vironments and sea-level changes by recognizing salt-marsh
2007; Choowong et al., 2009). It also presented a unique floral communities in buried coastal sediments. All flowering
opportunity to further develop criteria for distinguishing plants produce pollen, whereas ferns and mosses produce
paleotsunami deposits from those deposited by storms or spores; both are identified by their different structures and
other extreme events (e.g., Switzer and Jones, 2008). sculptures that generally allow identification to proceed to the
Recent catastrophic tropical cyclones including Typhoon species level (e.g., Moore et al., 1991). Pollen and spores
Morakot, Hurricane Katrina, and Cyclone Nargis have raised are more readily preserved in buried sediments than other
global awareness of coastal hazards (e.g., Stone and Kerr, plant parts due to their decay-resistant structural chemistry. In
2005). Significant uncertainty surrounds projected changes in tropical environments, they may be the only microfossil
tropical cyclone activity in response to climate changes preserved (e.g., Grindrod, 1985; Ellison, 1989) due to tapho-
(e.g., Knutson et al., 2010). Observational and instrumental nomic loss and degradation of other groups, including for-
records are too short (particularly for less frequent, high- aminifera and diatoms (Berkeley et al., 2009). The first
magnitude events) to provide a reliable basis for predicting microfossil-supported reconstructions of RSL utilized pollen
future activity (e.g., Liu and Fearn, 1993). Therefore, proxy assemblages in the Fenlands of Norfolk, UK (Godwin, 1940).
records of tropical cyclones making landfall and the extent of Godwin (1940) used salt-marsh taxa, such as Chenopodia-
their geological and ecological impacts are a valuable tool to ceae, Aster tripolium, and Plantago maritima to describe
assess future risk. In the field of paleotempestology, micro- paleoenvironmental changes from terrestrial to brackish con-
fossils have helped to identify tropical-cyclone landfalls pre- ditions (a transgressive change as marine influence increased)
served in sedimentary archives (e.g., Parsons, 1998). and also the replacement of brackish environments by ter-
Reconstructions have revealed spatial and temporal variability restrial conditions (regressive change marking decreased
294 Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones

marine influence). Pollen also provides evidence for changes salt marshes, diatoms form distinct assemblages with a strong
in groundwater conditions, which may be related to RSL relationship to tidal elevation, reflecting the varied preferences
changes (Tooley, 1978; Long, 1992; Zong and Tooley, 1996). It and tolerances of diatom species to saline inundation (e.g.,
is deposited following transport by wind, water, and animals. Sherrod, 1999; Zong and Horton, 1999; Sawai et al., 2004;
As such, preserved pollen assemblages include contributions Patterson et al., 2005; Szkornik et al., 2006; Woodroffe and
from local and regional flora, and interpretation is further Long, 2010). For example, Sawai (2001) showed that modern
complicated by variable pollen production rates among spe- diatom assemblages in eastern Hokkaido, Japan, had a clear
cies (e.g., Brush, 1989). Despite these limitations, Hughes transition from marine assemblages associated with tidal-flat
et al. (2002) and Engelhart et al. (2007) used quantitative settings (e.g., Planothidium delicatulum (Achnanthes delicatula)
methods to illustrate that surface pollen assemblages formed and Diploneis decipens var. parallela) through low-marsh vege-
zones parallel to the shoreline in salt marshes (British Col- tated environments where mixed diatom communities oc-
umbia, Canada) and mangroves (Indonesia), respectively. curred, to a group dominated by brackish and freshwater taxa,
Pollen samples from mudflat and low marsh environments in such as Pinnunavis elegans (Pinnularia elegans) in high-marsh
British Columbia had high percentages of arboreal pollen and settings.
spores, but low percentages of herbaceous pollen (Hughes Foraminifera are protozoans consisting of a single cell
et al., 2002). Higher percentages of Poaceae, Triglochin-type, made up of protoplasm with one or more nuclei. The soft
and Chenopodiaceae pollen differentiated the middle marsh. tissue of the foraminiferal cell is largely enclosed within a
Higher-elevation samples had greater amounts of Potentilla- test variously composed of secreted organic matter (tectin),
type, Rosaceae, and Achillea-type pollen. Rhizophora dominated secreted minerals (calcite, aragonite, or silica) or agglutinated
the mangrove-front ecosystem in Indonesia, often coexisting particles. Foraminifera live on sediment surfaces (benthic) and
with Bruguiera or Sonneratia (Engelhart et al., 2007). Mid- and in the water column (planktonic). Brackish and nearshore
rear- mangrove settings were dominated by Avicennia and areas, such as salt marshes, are dominated by benthic species
Ceriops with Heritiera and Lumnitzera (Engelhart et al., 2007). because planktonic species have environmental requirements
Pollen analysis is also employed as a dating technique by best met in oceanic areas. Early studies of foraminifera were
recognizing markers (floral community changes) of known concerned with taxonomy (e.g., Brady, 1884). Although this
age in the sedimentary record, such as the chestnut (Castanea) focus of research continues, ecological and distributional
blight that effected New England in AD 1912 (e.g., Anderson, studies of foraminifera are gaining importance. Scott and
1974). Pollen chronohorizons also provide a method to re- Medioli (1980) demonstrated that modern intertidal, in par-
solve chronological uncertainty associated with the radio- ticular salt-marsh, foraminifera, are vertically zoned with re-
carbon calibration curve plateau of the past 300 years, spect to tide level. Most notably, a near-monospecific
particularly in areas where extensive human activities have assemblage of Trochammina macrescens was recorded just
altered the ecosystem relatively recently: for example, the below the level of highest astronomical tide (HAT), above
emergence of Ambrosia (ragweed) with Plantago and Rumex in which foraminifera shortly disappear in the absence of tidal
eastern North America as a consequence of land clearance by inundation. Scott and Medioli (1978, 1980) suggested that
European settlers to create agricultural plots (e.g., Donnelly this assemblage zone could locate former sea level with a
et al., 2004; Turetsky et al., 2004). precision of 75 cm. Vertical zonation of salt-marsh for-
Although pollen continues to be employed in sea-level aminifera has been demonstrated at sites around the world.
studies, use of diatoms and foraminifera has become increas- Figure 1 shows that while the composition and vertical ranges
ingly important and widespread, because of their potential to of modern assemblage zones vary among sites, two general
more accurately and precisely reconstruct sea level. Diatoms are subdivisions can be made: an agglutinated assemblage (e.g.,
unicellular algae with chrysophyte-like photosynthetic pig- Trochammina and Haplophragmoides species) that is restricted to
ments. The cell wall is silicified to form a frustule, comprising the vegetated salt marsh and mangroves and a calcareous as-
two valves and bands (or copulas). Diatoms have been de- semblage (e.g., Ammonia and Elphidium species) that occupies
scribed and classified for over 200 years. In the late 1890s, mud and sand flats lower in the intertidal zone (Figure 1).
systematic and taxonomic investigations of contemporary and Consequently, the precision with which former sea level can
fossil diatoms began to be supplemented by studies of distri- be reconstructed using foraminiferal assemblage zones varies
butional ecology (e.g., Cleve, 1894, 1895). Since the 1920s, within and among sites, and it remains necessary to document
diatoms have been recognized as a valuable tool in re- modern distributions from the locality and region where fossil
constructing paleoenvironmental changes (e.g., Cleve-Euler, material will be (or has been) recovered (Horton and
1922). Diatoms are generally the dominant microphyte in es- Edwards, 2006). Differences among sites are caused by vari-
tuarine and marine littoral environments (Admiraal, 1984; ations in salinity and tidal regime.
Palmer and Abbott, 1986; Cooper et al., 2010), and taxa have
strong preferences for particular salinity regimes (Hustedt,
1937, 1939; Lowe, 1974; Vos and de Wolf, 1988; Juggins, 1992; 14.25.3 Microfossil-Based Reconstructions of Sea-
Hemphill-Haley, 1995). Diatoms become incorporated in Level Change
coastal sediments through time, resulting in buried assemblages
that constitute an environmental history. Switches among Microfossils are sea-level indicators because they possess a
assemblages of fresh, brackish, and marine diatoms document systematic and quantifiable relationship to elevation in the
the most pronounced changes in marine influence and de- tidal frame. A detailed understanding of the relationship be-
positional environment (Zong and Tooley, 1996). On modern tween sea-level indicators and sea level is established by
 Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones
10
11
7
12
8 13
5 1−4
6

10. Northeast
England
3.0
14 J. macrescens
2. Eastern Maine
15 16. Northern
J. macrescens
B. 16 New Zealand
pseudomacrescens 9
B. J. macrescens
pseudomacrescens T. inflata
T. comprimata T. inflata
T. inflata T. macrescens
M. obliqua

2.0 T. inflata 15. Eastern

M. fusca
J. macrescens Australia
M. fusca M. fusca
8. Oregon
11. Denmark T. inflata
T. macrescens
Elevation (m MSL)

T. irregularis T. inflata
B.
7. British pseudomacrescens
12. Northwest
Columbia J. macrescens
Portugal
H. wilberti A. beccarii
M. fusca T. inflata
A. salsum 4. Connecticut J. macrescens H. manilaensis
1. Nova Scotia 9. Chile P. limnetis
J. macrescens 13. Northern
1.0 T. macrescens 3. Massachusetts Japan
14. Indonesia
T. macrescens J. macrescens J. macrescens M. fusca
T. comprimata T. macrescens T. inflata A. mexicana T. inflata
T. salsa J. macrescens
T. inflata H. wilberti T. inflata A. mexicana
+ H. wilberti M. fusca M. fusca
T. inflata Haplophragmoides T. salsa P. limnetis M. fusca
T. comprimata 5. North Calcareous spp. H. manilaensis
M. fusca spp. M. obliqua
T. macrescens
Carolina H. germanica
M. fusca M. fusca T. inflata
T. comprimata 6. Texas E. williamsoni
T. inflata M. obliqua
T. inflata H. wilberti Quinqueloculina
J. macrescens A. mexicana Calcareous spp.
T. macrescens spp.
T. inflata
A. mexicana M. fusca
M. fusca J. macrescens M. fusca
M. fusca A. mexicana M. fusca
A. salsum A. salsum T. salsa T. macrescens
T. inflata
E. williamsoni P. limnetis T. salsa M. fusca
A. beccarii
0 T. inflata
T. macrescens
Calcareous spp. M. fusca
P. limnetis
Calcareous spp. Haplophragmoides Calcareous spp.
spp. Calcareous spp.
Ammonia spp.
A. salsum M. fusca A. salsum
M. fusca R. nana
Calcareous spp.

−1.0
Figure 1 Summary of foraminiferal assemblages and elevation ranges (m mean sea level, MSL) from Nova Scotia (Scott et al., 1980); Eastern Maine (Gehrels, 1994; Gehrels and van de Plassche,
1999); Massachusetts (Scott and Leckie, 1990); Connecticut (Edwards et al., 2004); North Carolina (Kemp et al., 2009a); Texas (Williams, 1994); British Columbia (Guilbault, 1996); Oregon (Hawkes
et al., 2010); Chile (Jennings et al., 1995); Northeast England (Horton et al., 1999); Denmark (Gehrels and Newman, 2004); Northwest Portugal (Fatela et al., 2009); Northern Japan (Scott et al., 1996);
Indonesia (Horton et al., 2005); Eastern Australia (Horton et al., 2003); and Northern New Zealand (Hayward et al., 1999). Differences in composition and elevational range of assemblages represent local
tidal characteristics and the influence of other environmental variables, such as climate and salinity, and testify to the need to collect appropriate local and regional data sets.

295
296 Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones

Supra tidal Thecamoebians

Highest marsh

Calc. Agglutinated
(e.g., Jadammina macrescens)

Foraminifera
Upland Middle to high marsh
(e.g., Trochammina inflata)
Low marsh and tidal flat
(e.g., Miliammina fusca)
Tidal flat and subtidal
(e.g., Elphidium spp.)
High marsh
Low marsh
Highest tide
Mean tide

IR
level RWL

Lowest tide Tidal flat

Sub tidal
(a) Bivalves

Depth Elevation
(m) (m MTL)
0 0.5 Sample a
500 years old (± 150 yrs)
0.2
supra-tidal wood
Sample b
1 −0.5 1000 years old (± 100 yrs)
−0.7
highest marsh foraminifera
Sample c
2 −1.5 1000 years old (± 100 yrs)
−1.6 low marsh foraminifera
Sample d
1500 years old (± 300 yrs)
(b) sub-tidal foraminifera

Age (yrs BP)

0 500 1000 1500

a Sample elevation
RWL correction
RSL (m MTL)

b c
d

(c)

Figure 2 Schematic representation of the indicative meaning and an illustrative example of its application. (a) Vertical zones of plants (upland,
high marsh, and low marsh), foraminifera, and thecamoebians. The modern relationship between microfossil assemblages and tide levels is
described by the reference water level (RWL) and indicative range (IR). Foraminifera may provide additional elevational zones, making them more
precise sea-level indicators than plants. (b) Example of dated samples recovered from a core. Sample elevation is potentially measured directly.
(c) Production of sea-level index points, and terrestrial (green) and marine (blue) limiting data. The limiting data must have been deposited above
and below mean sea level, respectively.

modern observations. The indicative meaning formalizes this reconstructed from a single type of dated indicator, each
relationship and permits reconstruction of RSL. Figure 2(a) sample is related to its own contemporary tide level (reference
uses the example of intertidal foraminifera to illustrate the water level), such as mean tide level (MTL), mean high water
indicative meaning. Because sea-level histories are seldom (MHW), or HAT. The reference water level is a mathematical
Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones 297

expression of tidal parameters rather than a single elevation 7 Miliammina fusca (Figure 3(d)). Marine diatoms (e.g., Paralia
a constant because of different tidal characteristics among sites sulcata; Plater and Shennan, 1992) were also abundant in this
with micro-, meso-, and macrotidal ranges (Horton et al., clay unit. Pollen assemblages from the overlying peat included
2000). Indicative range is the elevational range occupied by a Gramineae, Chenopodiaceae, P. maritima, and tree and shrub
sea-level indicator. pollen (Betula, Pinus, Quercus, and Corylus), which indicated
RSL is estimated using the following equation (Figure 2(b)): deposition in a high salt-marsh environment. This environ-
mental inference was further supported by abundant Jadammina
RSLi ¼ Ai  RWLi ½1
macrescens (high salt-marsh foraminifera) in the peat. These
changes in microfossil assemblages recorded a decrease in mar-
where Ai and RWLi are the altitude and reference water level of ine influence and thus mark a regressive change that was
sample i, expressed relative to the same tidal datum (e.g., radiocarbon dated to have occurred at 8439–8956 calibrated
MTL). Ai is established by measuring the depth of a sample in years before present (cal. yrs BP).
a core (or outcrop) where the altitude of the core top was The peat was overlain by a second clay unit containing
either established using surveying methods or estimated (less calcareous foraminifera, such as Elphidium williamsoni, Hay-
precisely) from the environment in which the core was col- nesina germanica, and E. earlandi suggesting deposition in a
lected (e.g., high salt marsh). For a modern (surface) sample, tidal-flat environment (Figure 3(e)). This environmental in-
the terms Ai and RWLi are equal; thus, RSL is zero and negative ference was supported by the presence of dinoflagellate cysts
values indicate RSL below present (Shennan and Horton, (e.g., Spiniferites). These microfossil changes showed increased
2002). marine influence, representing a transgressive contact that was
Reconstructions of RSL have vertical errors associated radiocarbon dated to have occurred at 8501–8959 cal. yrs BP.
with Ai and RWLi. Uncertainty in measuring Ai (e.g., surveying The two index points from Warkworth were combined
and angle of core errors) is inherent with collection and with others from Northumberland to produce a regional
processing of samples. The uncertainty associated with RWLi Holocene RSL history (Figure 3(f)). RSL rose from –5 m at
is the indicative range of the sea-level indicator being used. 8500 cal. yrs BP to 0 m at 4300 cal. yrs BP. This trend con-
A total error for each sample (Ei) is estimated from the forms to glacial isostatic adjustment model predictions for an
expression: area near the limits of ice advance at the Last Glacial Max-
imum (e.g., Peltier et al., 2002; Shennan and Horton, 2002).
Ei ¼ ðe21 þ e22 þ e2n Þ1=2 ½2
Regional scatter of index points includes the total influence of
local-scale processes, such as tidal-range change and sediment
where e1yen are individual sources of error for sample i consolidation.
(Shennan and Horton, 2002).

14.25.3.1 Sea-Level Index Points 14.25.3.2 Isolation Basins

A sea-level index point estimates the unique position of RSL in In formerly glaciated regions characterized by ongoing land
space and time. Where a suite of sea-level index points are uplift from glacio-isostatic adjustment, isolation basins pro-
developed for a locality or region, they describe changes in vide a means to reconstruct former sea level (e.g., Honig
RSL through time and estimate rates of change. The method- and Scott, 1987; Lloyd, 2000; Shennan and Horton, 2002;
ology that is employed to produce valid sea-level index points Berglund, 2005; Long et al., 2006). Isolation basins are natural
is described in many other publications (e.g., Tooley, 1982; rock depressions that were, at different times in their history,
van de Plassche, 1986; Shennan, 1986; Shennan and Horton, isolated from, or connected to, the sea (e.g., Long et al., 2006).
2002; Edwards, 2007; Engelhart et al., 2009, 2011). A multi- When RSL exceeded sill altitude, the basin was inundated by
proxy approach of litho-, bio-, and chronostratigraphic data marine water and became dominated by marine microfossil
is used. Lithostratigraphy at a study site is determined assemblages (Figure 4(a)). Conversely, when RSL was lower
either from sediment cores or in outcrops that are described than the sill, the basin was isolated from the sea and experi-
in the field and further analyzed in the laboratory (e.g., enced freshwater conditions with corresponding microfossil
grain size and organic content). Biostratigraphy (e.g., micro- assemblages (Lloyd, 2000). Isolation of a basin caused by land
fossil assemblages) establishes the nature of the environment uplift exceeding eustatic sea-level rise (falling RSL) is repre-
in which an index point accumulated, based upon under- sented in basin sediments by a switch between marine and
standing the modern distribution and ecology of the sea-level freshwater deposits. The age of basin isolation is estimated by
indicator used. Reconstructions of RSL also require develop- dating the contact between marine and freshwater sediment-
ment of a chronological framework, often from radiocarbon ary units. The height of RSL at the time of isolation is precisely
dating (Edwards, 2007; Kemp et al., 2012 – this publication). constrained by the altitude (with respect to modern sea level)
Figure 3 shows an example of how pollen, diatoms, and of the basin sill (Gehrels, 2007).
foraminifera were used as sea-level indicators to produce two sea- Figure 4(b)–(g) summarizes diatom, radiocarbon, and
level index points from transgressive and regressive changes in altitude data from isolation basin Upernivik T1 in western
core 95/3 from Warkworth, Northumberland, UK (Shennan Greenland (Long et al., 2006). This low-lying, freshwater lake
et al., 2000). At the base of the core, there was a thin layer of clay is currently located just above a modern salt marsh and is
overlying till. In turn, the clay was overlain by peat. Foraminifera isolated from marine inundation (Figure 4(c)). The lake sill is
in the clay included marine and low salt-marsh species, such as comprised of bedrock and is 1.6 m above modern sea level.
298 Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones

10 W 0
NORTH SEA Towns
Warkworth
Other sites
River Aln
Detail in C 500 m

55 N Alnmouth
Warkworth
Detail in B
Maudlin
Amble
Amble Bay
Core River Coquet
Amble
50

Cresswell Ponds
<5 m >5 m Town
(a) (b)
N

16 15 18 8 17 9 10 11 13 7 6 5 4 3
0 5 km NORTH SEA
4 NW 95/3 SE

Peat
Metres OD

0 Clay Sand
Newcastle-upon-Tyne

River Tyne
Till
100 m
−4 (c)
(d)

Foraminifera Test Linings

Warkworth 95/3 Trees Shrubs Herbs Spores Foraminifera
Plantago maritima
Chenopodiaceae

Spiniferites spp.
Umbelliferae
Menyanthes
Cyperaceae

Polypodium
Gramineae
Filipendula

Sphagnum

Salt marsh
Rosaceae
Triglochin

Pteridium

Estuarine
Artemisia
Lithology

Quercus

Filicales
Corylus

Rumex
Betula

Ulmus
Pinus

Age Altitude
Salix
Tilia

(cal yr BP) (m OD) Environment

−2.4

Lower
saltmarsh

8501−8959

Upper
−2.6 saltmarsh

8439−8956
Lower
saltmarsh
0 20 % 0 400 0 300
(e) Relative abundance

5
Cresswell (limiting)
Limiting dates

Basal index points

Amble Bay (basal)
0
RSL (m)

Intercalated index points

Cresswell Ponds
(intercalated)
Warkworth (intercalated)
−5

Tyne (basal)

−10
0 4 8 12
(f)
kyr cal BP
Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones 299

The sediment sequence beneath the lake consisted of a gray basis for developing a transfer function to quantify the rela-
silt-clay basal unit overlain by a 1–4-cm thick laminated black tionship between X and Y (Figure 5(b)). Through cross-val-
silt gyttja, which passed upward into a gyttja extending to the idation, transfer-function performance (estimated accuracy
modern lake bed (Figure 4(e)). Diatoms in core 4 showed an and precision) is assessed. A range of specific methods are
abrupt, up-core replacement of marine and brackish taxa by available to generate the transfer function (Birks, 1995). The
freshwater species (Figure 4(f)). This regressive change in most commonly used method in RSL reconstruction is
microfossil assemblage was caused by (and records) isolation weighted-average partial least squares (WA-PLS) because it has
of the basin from marine influence. A 1-cm thick slice of rigorous underlying statistical theory and works well in spe-
sediment at the lithostratigraphic boundary coincident with cies-rich, noisy data with long environmental gradients (Birks,
the change in diatom assemblages yielded a calibrated radio- 1995). The aim of paleoenvironmental reconstruction is to
carbon age of 4826–4548 cal. yrs BP. In conjunction with provide a proxy-based estimate of the environmental variable
measured sill altitude, a reliable sea-level index point was of interest at some point in the past (Xo). It is not possible to
created (T1; Figure 4(g)). Using a staircase of isolation basins measure this variable directly but microfossil assemblages of
at Upernivik, a RSL history was developed by Long et al. the same age preserved in core material (Yo) can be observed
(2006). RSL fell from þ 41 m at 8800–8000 cal. yrs BP to and enumerated. Using the relationships developed from
reach 0 m at 4500 cal. yrs BP and a minimum (–5 m) between modern samples, transfer functions transform the fossil data
3000 and 2000 cal. yrs BP, after which it rose to present (Yo) into quantitative estimates of the past environmental
(Figure 4(g)). This RSL reconstruction aided investigation of variable (Xo, Figure 5(c)).
Greenland Ice Sheet dynamics during the early Holocene Transfer functions provide a replicable methodology, im-
(Long et al., 2006). proved record comparability and sample-specific, quantified
errors (Horton and Edwards, 2006). Furthermore, transfer
functions employed in RSL reconstructions are able to use all
14.25.3.3 Transfer Functions
salt-marsh sediments rather than relying on contacts between
Since the late 1990s, transfer functions have become increas- sedimentary units. The most precise reconstructions are
ingly prominent in studies reconstructing RSL on seismic and derived from preserved microfossils associated with the
aseismic coasts (e.g., Guilbault et al., 1996; Horton et al., uppermost part of the intertidal zone (Gehrels, 2007). For
1999; Gehrels et al., 2001; Hughes et al., 2002; Sawai et al., example, Kemp et al. (2009b) developed and applied for-
2004; Horton and Edwards, 2006; Shennan and Hamilton, aminiferal-based transfer functions to a core of high-marsh
2006; Nelson et al., 2008; Kemp et al., 2009b). Transfer sediment in North Carolina, USA, to bridge the gap between
functions are empirically derived equations for calculating short-term instrumental records and long-term, geological
quantitative estimates of past environmental conditions from reconstructions of RSL (Figure 6(a)). They used the modern
paleontological data (Imbrie and Kipp, 1971; Sachs et al., elevational distribution of foraminiferal assemblages from 10
1977). The term does not apply to a single specific statistical salt marshes in North Carolina (Kemp et al., 2009a) to de-
technique, but rather describes a broad (although particular) velop transfer functions. The transfer functions were applied
methodological approach (see Birks, 1995 for a fuller discu- to 1-cm thick samples from a core at Sand Point, which
sion). Transfer functions, akin to the techniques they have consisted of salt-marsh peat with abundant remains of the
supplemented, are underpinned by the premise that preserved common salt-marsh plant Juncus roemerianus. The peat sam-
microfossil assemblages can be interpreted on the basis of an ples were dominated by species of foraminifera (J. macrescens
analogy with modern assemblages. Although this was often and Haplophragmoides wilberti) that were indicative of a high
done qualitatively by the judgment of an experienced re- salt-marsh environment (Figure 6(b)). The transfer functions
searcher, transfer functions offer a quantitative, objective estimated that the samples formed between 0.11 and 0.29 m
approach. above mean sea level (MSL) with an average error of 70.05 m.
The basic principle of quantitative environmental re- A composite chronology was developed for the core
construction using transfer functions is to express an en- (Figure 6(b)) using several dating techniques, including a
vironmental variable (X) as a function of biological data (Y) pollen chronohorizon (increased Ambrosia).
consisting of m species in t samples. The first step is the de- Reconstructed RSL was represented by boxes that in-
velopment of a modern training set. In studies aiming to re- corporated elevation (derived from transfer functions) and age
construct RSL using salt-marsh sediments and microfossils, uncertainties (estimated by an age-depth model) for indi-
this involves collecting surface samples along transects from vidual data points. Figure 6(c) shows the sea-level re-
which X and Y are directly measured and enumerated, re- construction from Sand Point and an additional record from
spectively (Figure 5(a)). This modern training set forms the Tump Point (also in North Carolina). Agreement between

Figure 3 Application of microfossil analyses to produce sea-level index points. ((a)–(c)) Location of Warkworth, UK; (d) stratigraphical cross
section showing position of core 95/3; (e) summary of microfossils (pollen and foraminifera) and dating results from Warkworth core 95/3;
pollen frequencies are the percentage of total land pollen with a minimum count of 300 grains per level; summary foraminiferal data are actual
counts; radiocarbon dates are shown as calibrated age ranges (2s); environmental interpretation based on all available microfossil data; and
(f) age–altitude graph of sea-level index points for Northumberland; index points are classified as basal or intercalated; freshwater limiting dates
indicate the maximum altitude of sea level at that time; example index points from Warkworth core 95/3 are highlighted in red. Modified from
Shennan, I., Horton, B., Innes, J., Gehrels, R., Lloyd, J., McArthur, J., Rutherford, M., 2000. Late Quaternary sea-level changes, crustal
movements and coastal evolution in Northumberland, UK. Journal of Quaternary Science 15, 215–237, with permission from Wiley.
300 Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones

Time (nonlinear)
(1) (2) (3) Freshwater

Marine Brackish

(a) Isolation basin Sill

50°40′ W Upernivik T1 (1.61 masl)

Jakobshavns Isfjord
5
4
3
2
1

Area of sill
N
Upernivik T1

0 100 m
(d)
Tasiussaq

Sample core
1 2 3 4 5

2
Greenland
Altitude (m)

Water
c 69°05′ N
70° N
0 Grey gyttja Black gyttja
Lakes Grey silt-clay
60° W
Moraines
−2 0 20 m
(b) 30° W 0 1 km
(e)
(c)
Mesohalobous

Oligohalobous

Total counted
Polyhalobous

Core 4 in e
Lithology

Depth
(cm)
286 50
Marine limit = c.41 m
40
288
30
MSL (m)

290
4150 ± 40 BP 20
(Beta-178167)
4826−4548 10 T1
cal yr BP
0
294
−10
0 2 4 6 8 10
296 Age (ka cal. yr BP)

0 100 0 100 0 300 Lakes below marine limit

(f) Relative abundance (%) (g) Lakes above marine limit
Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones 301

these proxy records and tide-gauge data validated the transfer- Early paleoearthquake research at accessible outcrops used
function approach and justified its application to time periods suddenly buried plant macrofossils to infer instantaneous co-
prior to tide-gauge measurements. Figure 6(d) shows the Sand seismic subsidence (e.g., Atwater et al., 1991), but the broad
Point and Tump Point reconstructions with other Atlantic elevational range of salt-marsh plants limited subsidence es-
coast sea-level records developed using foraminifera. For each timates to at least 1 m (Nelson et al., 1996). Transfer functions
record, rates of glacio-isostatic adjustment were removed to developed using microfossils have provided more precise es-
reveal a c 2 mm yr–1 increase in the rate of sea-level rise that timates of co-seismic subsidence. For example, Hughes et al.
began in the latter part of the nineteenth century (Kemp et al., (2002) used a pollen-based transfer function to estimate
2009b). 0.670.3 m of co-seismic land subsidence (RSL rise) produced
by the AD 1700 Cascadia earthquake at a salt marsh near
Tofino, British Columbia, Canada (Figure 7(b)). Modern
14.25.4 Microfossils and Land-Level Change pollen taxa defined four salt-marsh zones: low, middle, high,
and forest-edge transition (Figure 7(c)). Deposits beneath
Histories of earthquakes, inferred from geological evidence, the marsh included a buried peat capped by tsunami sand.
aid in anticipating the timing and magnitude of future events The sand was abruptly overlain by mud, which graded into
(Satake and Atwater, 2007). This natural warning system now the peaty modern marsh (Hughes et al., 2002). The buried
influences building code and planning in the United States, peat contained pollen indicative of a high to forest-edge
Canada, and Japan, particularly where the RSL record dem- transition marsh (Figure 7(d)), as did the overlying sand,
onstrates past occurrences of earthquakes larger than those which likely reflected entrainment of litter as the tsunami
known from written and instrumental records (Satake and washed over the marsh surface. Sediments above the tsunami
Atwater, 2007). sand yielded pollen characteristic of a low- to middle-salt
Vertical land movements caused by earthquakes are marsh environment, including abundant Cyperaceae, Triglo-
manifest in coastal sedimentary sequences as abrupt RSL chin-type, Poaceae, and Chenopodiaceae.
changes that occurred independently of eustatic changes Transfer functions have produced continuous records of
(Nelson, 2007). The earthquake-deformation cycle describes a RSL change through an earthquake deformation cycle, which
characteristic and repeated pattern of land-level movements is essential information for understanding the mechanics of
associated with great (4magnitude 8) earthquakes along plate-boundary ruptures in subduction zones (e.g., Sawai
much of the Cascadia coast (Pacific coast of North America). et al., 2004). In southern Alaska (Zong et al., 2003; Hamilton
Between earthquakes, interseismic strain accumulates, and and Shennan, 2005; Shennan and Hamilton, 2006), diatoms
land is uplifted due to locking of the plate interface detailed a precise record of coastal land subsidence not
and shortening of the overriding tectonic plate (Figures 7(a) only during great earthquakes but also in the decade or so
and 8(d)). When the rate of land uplift exceeds eustatic sea- prior to such events (Figure 8). A core collected from upper
level rise, this extended interseismic period is one of RSL fall Cook Inlet, Alaska, included two peats separated by mud
associated with a gradual regressive transition from salt marsh (Shennan and Hamilton, 2006). Freshwater (e.g., Eunotia
to organic-rich terrestrial soil, as marine influence decreases exigua) and brackish (e.g., P. lagerstedtii) diatoms dominated
(Leonard et al., 2010). During megathrust earthquakes, strain the peats, whereas marine diatoms (e.g., Delphineis surirella)
is released resulting in abrupt, co-seismic land subsidence were found both in the mud and upper few centimeters of
(e.g., Leonard et al., 2010). High salt marsh or terrestrial soils both peats (Figure 8(c)). Shennan and Hamilton (2006)
are lowered into the intertidal zone to become ‘buried soils’, as suggested it was unlikely that occurrences of marine diatom
they become overlain by muddy, lower tidal sediments rep- assemblages in the peats were a result of biotubation because
resentative of increased marine influence (transgressive brackish species, such as Nitzschia obtusa, Navicula begeri, and
change). These processes have produced geologic evidence of N. brockmanii, did not occur in the overlying mud. Shennan
paleoearthquakes in the form of buried peat-mud couplets et al. (1999) proposed a four-part earthquake deformation
beneath salt marshes in Cascadia (e.g., Atwater et al., 2005; cycle for the AD 1964 Alaska earthquake, adding this subtle,
Nelson et al., 2008), Alaska (e.g., Hamilton and Shennan, short-lived RSL rise as a precursor to abrupt co-seismic RSL
2005), Chile (e.g., Cisternas et al., 2005), Japan (e.g., Sawai rise. Pre-seismic land-level movement immediately prior to
et al., 2004), and New Zealand (e.g., Hayward et al., 2004). the AD 1700 earthquake was also suggested by Hawkes et al.
Environmental changes caused by modified marine influence (2005) using foraminifera from salt marshes in Cascadia.
through an earthquake deformation cycle are particularly well Estimates and interpretations of pre-seismic subsidence are
represented by microfossils in coastal sedimentary archives. complex and subtle; but if this land subsidence and associated

Figure 4 Reconstructing sea level from isolation basin sediments: (a) schematic diagram showing the isolation of a basin by falling relative sea
level: (1) the basin is continuously inundated and marine sediments accumulate, (2) the basin is isolated from the sea at low tides and brackish
conditions are experienced, and (3) sill height is above the highest tides and the isolation basin becomes a freshwater lake; ((b)–(c)) location of
Upernivik, Greenland; (d) Upernivik isolation basin T1, showing location of cores and basin sill; (e) stratigraphy of the Upernivik isolation basin,
showing position of core 4; (f) summary of diatoms in core 4 showing the transition from clastic sediment to organic limus used to produce a
sea-level index point; diatoms are grouped by salinity associations (polyhalobous – marine, mesohalobous – brackish, and oligohalobous –
fresh); calibrated radiocarbon ages and depth (cm) down-core are shown on the left of the lithology column; and (g) sea-level reconstruction
from Greenland isolation basins showing the example index point from Upernivik. Modified from Long, A.J., Roberts, D.H., Dawson, S., 2006.
Early Holocene history of the west Greenland Ice Sheet and the GH-8.2 event. Quaternary Science Reviews 25, 904–922.
302 Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones

Species A

Taxon abundance (Y)

Species B
Assemblage type Gaussian unimodal Species C
o = optimum
High marsh Low marsh Tidal flat c t = tolerance Species D
t c = maximum
Sampling station

o
Species A Partially sampled distribution
(b) Environmental Variable (X; Elevation)

X = U (Y)
Abundance (%) (Y)

Species B
Species Estimation of paleo-elevation
A B C D E High marsh Low marsh Tidal flat
Species C

Species D

Altitude or depth in core

Species E
Tidal elevation (X )

MHHW

MHW

MTL
Upland High marsh Low marsh Tidal flat
X0 = Û (Y0) MHHW MHW MTL

Abundance (%) (Y0) Tidal elevation (X0)

(a) (c)
Estimated paleo-marsh elevation
with sample-specific uncertainty

Figure 5 Schematic of the development and application of a transfer function: (a) distribution of microfossils is described using samples
collected (red circles) from transects across the full environmental (tidal elevation) gradient; modern microfossils (Y) and modern sea level (X)
are observable; (b) elevation (X) is quantitatively expressed as an empirically derived function (U) of microfossil assemblages (Y); (c) fossil
microfossils (Yo) are observed by documenting assemblages preserved in sedimentary archives; the transfer function estimates the elevation with
respect to sea level at which each core sample formed (Xo).

RSL rise do indeed occur, they are warning signs that (Nanayama et al., 2003; Sawai et al., 2009b), Japanese
are detectable for several years before a huge earthquake Trench (Sawai et al., 2008) and Aleutian Islands (Shennan
(Bourgeois, 2006). and Hamilton, 2006), at the Cascadia subduction zone
(Nelson et al., 2008), and in the region of the AD 1960 Chile
(Cisternas et al., 2005) and AD 2004 Indian Ocean (Jankaew
et al., 2008; Monecke et al., 2008) earthquakes.
14.25.5 Microfossils as Indicators of Paleotsunamis
Identification of tsunami deposits is commonly based on
and Storms
recognition of anomalous sand layers in low-energy environ-
ments, such as coastal ponds, lakes, and marshes, which can be
14.25.5.1 Paleotsunamis
supported by microfossil analyses (Figure 9). For example, the
Records of tsunamis developed from the sedimentary deposits AD 1700 Cascadia tsunami can be identified with confidence
they leave behind improve understanding of tsunami pro- from a sheet of sand that tapers landward (generally for several
cesses and frequency by expanding the age range of events kilometers), contains a mixed microfossil assemblage, and co-
available for study (Mamo et al., 2009). Understanding tsu- incides with additional microfossil evidence for abrupt co-
namis is important for assessing and mitigating their effects seimic subsidence (e.g., Hawkes et al., 2011).
on coastal geomorphology, ecology, and rapidly expanding Marine microfossils are likely to dominate tsunami
human communities. Well-preserved sequences of tsunami deposits because tsunamis cause landward transportation
deposits are used to estimate recurrence intervals and are a and deposition of planktonic and benthic forms where they
geological record of their source events, including earth- form anomalous assemblages (Mamo et al., 2009). However,
quakes, landslides, and volcanic eruptions (Bernard and microfossils in tsunami deposits are commonly composed
Robinson, 2009). Reconstructions have shown repeated tsu- of mixed assemblages because tsunamis erode, transport,
namis during the Holocene around the North Sea (Bondevik and deposit marine, brackish, and freshwater sediments
et al., 2005), in New Zealand (Goff et al., 2001), Kamchatka (with associated taxa) as they inundate coastal and inland
(Pinegina et al., 2003), along the southern Kuril Trench areas (e.g., Dawson, 1996). Many paleotsunami studies have
Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones 303

Nova Scotia
(Gehrels et al., 2005) Salt marsh Ben
chm
ark
Maine 64
Sand Point Core e2 Rest
USA
(Gehrels et al., 2002)
R out Area
U.S.

Upland
North Carolina
(Kemp et al., 2009) CROATAN SOUND
200 km
ATLANTIC OCEAN
(a)

J.macrescens H.wilberti Ambrosia PME RSL

0
AMS 14C
High-precision AMS 14C
210
Pb
137Cs

Bombspike AMS 14C

40
Ambrosia Horizon
Depth (cm)

Age-depth model

80

50 100 50 100 4 8 0.1 0.3 −0.6 −0.2 1400 1600 1800 2000
(b) % % % m MSL Year (AD)

0.2

0.0

0.0 −0.1
(m MSL)
Relative sea level (m MSL)

−0.2

−0.2 −0.3
1920 1940 1960 1980

−0.4

−0.6
Tide-gauge records Salt-marsh records
NC average Charleston, SC Sand Point Tump Point
−0.8
1600 1700 1800 1900 2000
(c)
Year (AD)
Figure 6 Estimating the timing and magnitude of recent accelerated sea-level rise: (a) location of Sand Point, North Carolina, USA; (b)
foraminifera preserved in a core of salt-marsh sediment collected at Sand Point; an age-depth model (upper and lower estimates shown by solid
lines) was developed from a composite chronology; transfer functions linking the distribution of modern foraminifera to tidal elevation were used
to estimate paleomarsh elevation (PME) in meters above mean sea level (MSL); only the two most dominant species are shown; relative sea level
(RSL) was estimated by subtracting PME from measured sample altitude; (c) reconstruction of RSL at Sand Point (gray boxes) and Tump Point
(blue boxes); averaged tide-gauge records from North Carolina (NC, green) and Charleston, South Carolina (SC, red), are also shown; (d) RSL
reconstructions from eastern North America following adjustment for GIA and compilations of global-tide gauge records (red, Church and White,
2006; blue, Jevrejeva et al., 2008); timing (blue bars) and estimated magnitude (rate in excess of background) of accelerated sea-level rise are
shown for each study. Modified from Kemp, A.C., Horton, B.P., Culver, S.J., et al., 2009b. Timing and magnitude of recent accelerated sea-level
rise (North Carolina, United States). Geology 37, 1035–1038.
304 Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones

Nova Scotia
Gehrels et al., 2005 +1.6 mm yr−1
0.4

RSL (M)
0.3
0.2
0.1
0.0 Maine
Gehrels et al., 2002 +0.6−1.5 mm yr−1

Sand Point, North Carolina

Kemp et al., 2009

+2.2 mm yr−1
Tump Point, North Carolina
Kemp et al., 2009

Global Tide-Gauge Records

1750 1800 1850 1900 1950

(d) Year (AD)
Figure 6 Continued.

employed diatoms (e.g., Hemphill-Haley, 1995; Nanayama diatoms are clearly contrasted with in situ freshwater taxa
et al., 2003; Sawai et al., 2004, 2009a; Kelsey et al., 2005; (Hemphill-Haley, 1996).
Dawson, 2007). For example, diatom analysis revealed three A number of studies have employed foraminifera to
sandy deposits of marine origin at Suijin-numa, a coastal lake describe tsunami sediment characteristics (e.g., Bondevik
midway along the subduction zone marked by the Japan et al., 1997; Clague et al., 1999; Dawson and Shi, 2000;
Trench (Sawai et al., 2008). The deposits are interbedded with Goff et al., 2001; McMurtry et al., 2004; Smith et al., 2004;
lacustrine peat and mud above a foundation of marine, Cundy et al., 2006; Hawkes et al., 2007; Le Roux et al., 2008).
probably littoral sand. Unlike the lacustrine deposits, all three For example, Hawkes et al. (2007) analyzed foraminiferal
sandy units contain marine and brackish diatoms. The age assemblages in pre-tsunami and tsunami sediment from the
constraints of the middle unit permit correlation with the 2004 Indian Ocean tsunami at five sites along the Malay-
Jogan tsunami in AD 869 (Minora et al., 2001) that reportedly sia–Thailand Peninsula to characterize tsunami sediment
devastated at least 100 km of coast approximately centered on source and deposition (Figure 10). A trench from the modern
Sendai and is the presumed predecessor to the devastating tidal flat at Sungai Burong, Penang, Malaysia, revealed a pre-
2011 Tohoku-Oki earthquake in Japan. Tsunami run-up has tsunami mud dominated by small, tested (o100 mm), marine
also been estimated by mapping the distribution of tsunami- foraminifera (e.g., Ammonia spp.) that was overlain by
transported diatoms. Indeed, diatom distributions suggest tsunami sand with no visible structure (Figure 10(c)). The
that tsunami-inundation area is greater than the distribution sand included a mixed foraminiferal assemblage dominated
of sandy deposits. At Niawaikum River in Washington (USA), by large tests (4200 mm) (Hawkes et al., 2007). Cluster
sand deposited by the AD 1700 Cascadia tsunami reached analyses of foraminiferal assemblages in the tsunami sand
3 km inland from the river mouth, but marine diatoms, revealed at least two separate episodes of deposition
transported by the tsunami, were traced at least 1 km that contained uprush species from the inner shelf (e.g.,
farther inland (Hemphill-Haley, 1995). This approach is E. hispudulum) and backwash species from a mangrove
very effective in freshwater environments because marine environment (e.g., Haplophragmoides spp.).
Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones 305

Time
Inter- and Pre-seismic During earthquake (tsunami)

Uplift Subsidence Subsidence Uplift

Overlying plate

Oceanic plate

(a) Seismogenic zone

130° W 120°
Tidal channel
British Columbia M3 (60 m) Intertidal mudflat
D (90 m) Low marsh
50° N Tofino High marsh
Forest-edge transition
Forest
Small tributaries
Forest edge
49°05′ 45″
Washington
JUAN
N
DE FUCA
PLATE
Oregon
M2
NORTH
Sample locations
AMERICA
1998 transects
PACIFIC PLATE
1999 transects
PLATE
Transects utilized for
pollen analysis
California Transects with gouge
40° core data
100 km M Monolith colection sites
0 40 m

125°50′ 46″ M1
(b)
(c)

Herbs Shrubs and trees

la

M1
yl
an n-ty ae

Pi res oph

og
pe
i e

ae
ge -ty e

R les pe
ch c

An illea -typ
Er lica pe

MHHW
al
n
e

lo ia

Pi a ace
te
ic -ty

es lle
ea

an
ig pod

s e

he
h lla
y

Po go

nu a

or po

n
ac

ce s
og

ea

Al ace

n
tio
Ac nti
no

ta

er
a
er

MLHW MHAT
s
Sp er
ol

ac

na
ug
te

nu

od
he

up
yp

os
th

th
Po

Zo
Ts
Tr

Pl

M
Li

O
C

Middle
20

Low
Depth (m)

Sand

High
30
Middle

0 20
Peaty mud Percent ( less than 0.5 %) 2 3 4 5
Sand Good analog Inferred paleo-elevation (m)

Buried peat No good analog

(d)

Figure 7 Use of pollen to reconstruct earthquake deformation cycles above subduction-zone faults: (a) schematic diagram showing deformation
associated with a subduction-zone thrust fault on a coastline during an earthquake cycle. Modified from Atwater, B.F., Musumi-Rokkaku, S., Satake,
K., Tsuji, Y., Ueda, K., Yamaguchi, D.K., 2005. The orphan tsunami of 1700 – Japanese clues to a parent earthquakes in North America. U.S.
Geological Survey Professional Paper 1707, 133 pp. (b)–(c) location of Tofino, British Columbia; (d) Fossil-pollen diagram for monolith M1; pollen
values are percentages of total pollen (trees þ shrubs þ herbs); spore values are percentages of total pollen plus spores; paleomarsh elevations
were estimated using a transfer function. Modified from Hughes, J.F., Mathewes, R.W., Clague, J.J., 2002. Use of pollen and vascular plants to
estimate coseismic subsidence at a tidal marsh near Tofino, British Columbia. Palaeogeography Palaeoclimatology Palaeoecology 185, 145–161.
306 Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones

14.25.5.2 Paleotempestology research, including recognition of anomalous sand layers (e.g.,

Donnelly and Woodruff, 2007) and microfossils deposited in
The instrumental record of tropical cyclone activity is too
a low-energy environment (e.g., Parsons, 1998).
short to fully capture the occurrence of the rare but most de-
The microfossil signature of tropical cyclones relies on
structive events. Obtaining a record of past landfalling tropical
identification of abrupt switches to allochthonous assem-
cyclones (known as paleotempestology), and their ecological
blages. Parsons (1998) investigated diatoms preserved in a
impacts, is one means to assess future risk, reveal spatial and
salt-marsh pond that was inundated by storm surges during
temporal variability of tropical cyclone activity, and decipher
Hurricane Andrew that made landfall in Louisiana, USA in
their relationship with global climatic changes (e.g., Liu and
1992. Cores of pond sediment identified a mud layer de-
Fearn, 1993). Identification of tropical cyclone deposits uses
posited by Hurricane Andrew. It was characterized by a diverse
many of the techniques and approaches employed in tsunami
assemblage of diatoms from marine, brackish, and freshwater
sources that was clearly contrasted against assemblages dom-
inated by salt-marsh taxa that accumulated in situ prior to
156 150 144 W Hurricane Andrew. Collins et al. (1999), Scott et al. (2003),
62 N Subsided area
in 1964
and Cochran et al. (2005) found unusually high concen-
ALASKA Anchorage trations and increased species diversity of calcareous for-
B
Girdwood aminifera in tropical cyclone deposits compared to underlying
60 sediments. Tropical cyclones are best recognized by a com-
bination of lithological and micropaleontological analysis
tio dge (Hippensteel and Martin, 1999). For example, Hippensteel
ne
zo
and Martin (1999, 2000) and Hippensteel et al. (2005) in-
e

~500 km
bd rd

n
su wa

ferred that sedimentary layers of Oligo-Miocene deepwater

uc

(a)
of Sea

100 km
Pacific benthic foraminifera (e.g., Uvigerina spp. and Siphogenerina
Subsided area Ocean spp.) separated by mud-rich layers containing shallow-
in 1964 earthquake
(b) water foraminifera, such as A. beccarii and Elphidium spp., were
tropical cyclone deposits. By contrast, Horton et al. (2009)
Nitzschia cari var. cincta

Pinnularia subcapitata

described a virtual absence of foraminifera in sediments

Pinnularia lagerstedtii
Navicula brockmanii
Delphineis surirella

deposited by Hurricanes Katrina and Rita along the U.S.

Nitzschia obtusa

Navicula begeri

Eunotia exigua

Gulf coast in AD 2005, whereas abundant agglutinated

Depth (m)

Sea-level
Lithology

relative to present foraminifera were found in underlying salt-marsh deposits

−3 −2 −1 0 1
(Figure 11). Absence of foraminifera may be due to initial size
0
of the standing crop in the source material, the number
of individuals that reproduce, frequency of reproduction
and number of progeny, and/or post-depositional changes
(Murray, 2006).
Preseismic
subsidence

14.25.6 Summary
1
More than 70 years after Godwin presented pollen as an indi-
cator of sea-level change (Godwin, 1940), microfossils remain
at the center of new developments seeking to improve under-
standing of sea-level changes and recurrences of earthquakes,

Figure 8 Diatom analyses during past earthquake cycles suggesting

pre-seismic movement. (a) Location of south-central Alaska, USA; (b)
2 area subsided in the AD 1964 earthquake (Plafker, 1969); (c) RSL
50 % −3 −2 −1 0 1
Peat Mud with rootlets changes reconstructed using a diatom-based transfer function; two
(c)
short periods of pre-seismic submergence immediately prior to the
2
Pre-seismic subsidence/relative sea-level rise substantial co-seismic subsidence were recognized at the top of the
Uplift

Rapid co-seismic subsidence/relative sea-level rise peat unit and are highlighted. Modified from Shennan, I., Hamilton,
S., 2006. Coseismic and pre-seismic subsidence associated with
great earthquakes in Alaska. Quaternary Science Reviews 25, 1–8;
Altitude (m)

and (d) idealized seismic cycle for a coastal area that subsides during
great earthquakes with the small-scale subsidence that is postulated
to precede great earthquakes; the end of the series showing the
Subsidence

Centuries long inter-seismic uplift/relative sea-level fall present elevation of the Alaska coast following the 1964 subduction-
Rapid post-seismic uplift/relative sea-level fall zone earthquake. Modified from Bourgeois J., 2006. Earthquakes: a
0
movement in four parts? Nature, News and Views, with permission
0 1000 2000
(d) Time (year) from Nature.
Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones 307

tsunamis, and tropical cyclones. The utility of microfossils, and with tidal elevation and that this modern, observable rela-
in particular diatoms and foraminifera, as sea-level indicators in tionship has existed without change during the time repre-
low-energy coastal sedimentary deposits is underpinned by the sented by coastal sedimentary sequences. We have provided
concept that characteristic assemblages have strong associations examples from the UK and Greenland that illustrate how

Stratigraphic column

Forams

Forams
Diatom

Diatom
Pollen

Pollen
1 Before tsunami Location A Location B
A B
Freshwater
lake Freshwater

Sedimentation
swamp

Sea Sand Rock

2 During tsunami incursion

A B
Erosion
Soil

Peat

3 Maximum height of tsunami Limit of run-up

A B
Debris

Sand

4 During return flow Tsunami deposit Limit of tsunami deposit

A B

5 Post tsunami
A B
Tsunami deposit

Debris

Diatom Pollen Forams

Freshwater taxa Forest/swamp taxa Marine taxa
Marine taxa Poor or no record Poor or no record

Figure 9 Schematic story of tsunami waves and microfossil assemblages at a coastal freshwater lake (A) and coastal freshwater wetland (B): (1)
before the tsunami, lake sediment consists of a peaty soil with freshwater diatoms; (2) during a tsunami, an incursive wave erodes coastal sand,
which is transported into the lake; (3) at the maximum height of tsunami inundation, eroded fine fractions begin to settle out of suspension; (4)
return flow erodes and then deposits coastal sand due to its high current velocity; and (5) after the tsunami, a return to the prevailing environmental
conditions caps the stratigraphy with freshwater peat. Modified from Bondevik, S., Svendsen, J.I., Johnsen, G., Mangerud, J., Kaland, P.E., 1997.
The Storegga tsunami along the Norwegian coast, its age and run-up. Boreas 26, 29–53, with permission from Wiley.
308 Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones

95° E 100°

INDIAN

Roa
OCEAN

d
THAILAND
15° N

2004 slip 5°20′ 25″ Area of inundation

l
ne
an
SEA Ch

House

Pulau Penang
Wave N
5° direction
M
AL

N
AY

Su Trench
SI

m 50 m
A

at
ra
~500 km 125°11′ 45″

(a) (b)

Foraminifera test size Foraminifera species

Haplophragmoides manilaenesis
Elphidium advenum botaniense

Haplophragmoides wilberti
Quiqueloculina seminulum
Ammobaculites exiguus
Juvenile planktonic sp.
Elphidium hispudulum
Medium (100−200µ)
Stratigraphic unit

Planulia exorna
Small (0−100µ)

Radiolarian sp.
Diversity index
Large (>200µ)

Ammonia sp.
Stratigraphy
Depth (cm)

Species cluster
Zones analysis
0
SB-III
3

SB-II
10
2

1 SB-I
20
0 5 0 1.0
M. sand
C. sand

Total sum of
Silt

squares
0 20 % 0 20 %
(c)

Figure 10 Foraminiferal assemblages deposited by the AD 2004 Indian Ocean tsunami: (a) location map showing rupture area of the 2004
earthquake (Chlieh et al., 2007); (b) location of Sungai Burong, Penang Island, Malaysia and a trench that was dug to sample pre- and post-
tsunami sediment for foraminifera; (c) foraminifera from the Sungai Burong trench; schematic stratigraphic column showing coarse sand
deposited by the 2004 tsunami on top of pre-tsunami silt; size and assemblage composition of foraminifera are shown (small r100 mm,
medium ¼ 100–200 mm, and large Z200 mm); distinct downcore zones of foraminifera were identified using cluster analysis. Modified from
Hawkes, A.D., Bird, M., Cowie, S., et al., 2007. Sediments deposited by the 2004 Indian Ocean Tsunami along the Malaysia–Thailand Peninsula.
Marine Geology 242, 169–190.
Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones 309

microfossils from estuaries and salt marshes have reconstructed the context of scenarios of future sea-level rise. Along the Pacific
RSL along both subsiding and uplifting coastlines. An example coast of the USA and in Alaska, cyclical land-level changes
from the USA showed how high-resolution RSL reconstructions caused by repeated earthquakes have been reconstructed using
are being used to investigate climate–sea-level relationships in microfossils. We also showed how microfossils were used to

(a) (b)

(c)

Figure 11 Foraminiferal assemblages deposited by tropical cyclones: (a) location of study sites at Ocean Springs and St. Andrews, Mississippi;
and Dauphine Island, Alabama, along the Gulf Coast of the United States; the paths of Hurricanes Katrina and Rita are shown; (b) location of St.
Andrews, Mississippi, showing geomorphology and buildings that were impacted by the storm surges; numbers show elevation (m NAVD88) and
maximum water depth (m), respectively, and indicate coastal flooding caused by Hurricanes Katrina and Rita; location of a core used for analysis
of foraminifera is shown; (c) foraminifera from a core at St. Andrews, Mississippi, which included pre- and post-tropical cyclone sediment;
Hurricanes Katrina and Rita deposited sand on top of finer grained clastic and organic sediment; foraminiferal assemblages from samples in the
core are shown; at several intervals, there was an absence of foraminifera. Modified from Horton, B.P., Rossi, V., Hawkes, A.D., 2009. The
sedimentary record of the 2005 hurricane season from the Mississippi and Alabama coastlines. Quaternary International 195, 15–30.
310 Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones

describe sediment deposited by the 2004 Indian Ocean tsunami Cleve-Euler, A., 1922. Om diatomacevegetationen och dess forandringar i Sabysjon,
and Hurricane Katrina and how these modern analogs have Uppland samt negra damda sjoar i Salatrakten. Sveriges Geologiska
Undersokning C309, 1–76.
been used to recognize similar events preserved in coastal
Cochran, U.A., Berryman, K.R., Middlenhall, D.C., Hayward, B.W., Southall, K.,
sedimentary archives. Hollis, C.J., 2005. Towards a record of Holocene tsunami and storms for
northern Hawke’s Bay, New Zealand. New Zealand Journal of Geology and
Geophysics 48, 507–515.
Acknowledgments Collins, E.S., Scott, D.B., Gayes, P.T., 1999. Hurricane records on the South
Carolina coast: can they be detected in the sediment record? Quaternary
This work was supported by funding from National Science International 56, 15–26.
Cooper, S., Gaiser, E.E., Wachnicka, A., 2010. Estuarine paleoecological
Foundation (EAR-0842728; EAR-0951686) awarded to BPH. reconstructions using diatoms. In: Smol, J.P., Stoermer, E.F. (Eds.), The
This paper is a contribution to IGCP Project 588 and PALSEA Diatoms: Applications for the Environmental and Earth Sciences. Cambridge
and is Earth Observatory of Singapore (EOS) publication University Press, Cambridge, pp. 324–345.
number 29. Some symbols in figures were courtesy of the Cundy, A.B., Sprague, D., Hopkinson, L., Maroukian, H., Gaki-Papanastassiou, K.,
Papanastassiou, D., Frogley, M.R., 2006. Geochemical and stratigraphic
Integration and Application Network (ian.umces.edu/sym-
indicators of late Holocene coastal evolution in the Gythio area, southern
bols/), University of Maryland Center for Environmental Peloponnese, Greece. Marine Geology 230, 161–177.
Science. Dahanayake, K., Kulasena, N., 2008. Recognition of diagnostic criteria for recent-
and paleo-tsunami sediments from Sri Lanka. Marine Geology 254, 180–186.
Dawson, A.G., 1996. The geological significance of tsunamis. Zeitschrift fur
References Geomorphologie 102, 199–210.
Dawson, A.G., Shi, S.Z., 2000. Tsunami deposits. Pure and Applied Geophysics
157, 875–897.
Admiraal, W., 1984. The ecology of estuarine sediment-inhabiting diatoms. Progress
Dawson, S., 2007. Diatom biostratigraphy of tsunami deposits: examples from the
in Phycological Research 3, 269–322.
1998 Papua New Guinea tsunami. Sedimentary Geology 200, 328–335.
Anderson, T.W., 1974. The chestnut pollen decline as a time horizon in lake
Donnelly, J.P., Cleary, P., Newby, P., Ettinger, R., 2004. Coupling instrumental and
sediments in eastern North America. Canadian Journal of Earth Sciences 11,
geological records of sea-level change: evidence from southern New England of
678–685.
an increase in the rate of sea-level rise in the late 19th century. Geophysical
Atwater, B.F., Hemphill-Haley, E., 1997. Recurrence intervals for great earthquakes of
Research Letters 31.
the past 3500 years at northeastern Willapa Bay, Washington. U.S. Geological Donnelly, J.P., Woodruff, J.D., 2007. Intense hurricane activity over the past 5,000
Survey Professional Paper 1576, 108 pp. years controlled by El Nino and the West African monsoon. Nature 447,
Atwater, B.F., Musumi-Rokkaku, S., Satake, K., Tsuji, Y., Ueda, K., Yamaguchi, D.K., 465–468.
2005. The orphan tsunami of 1700 – Japanese clues to a parent earthquakes in Edwards, R.J., 2007. Sea levels: resolution and uncertainty. Progress in Physical
North America. U.S. Geological Survey Professional Paper 1707, 133 pp. Geography 31, 621–632.
Atwater, B.F., Stuiver, M., Yamaguchi, D.K., 1991. Radiocarbon test of earthquake Edwards, R.J., Wright, A., Van de Plassche, O., 2004. Surface distributions of salt-
magnitude at the Cascadia subduction zone. Nature 353, 156–158. marsh foraminifera from Connecticut, USA: modem analogues for high-
Berglund, M., 2005. The Holocene shore displacement of Gastrikland, eastern resolution sea level studies. Marine Micropaleontology 51, 1–21.
Sweden: a contribution to the knowledge of Scandinavian glacio-isostatic uplift. Ellison, J.C., 1989. Pollen Analysis of Mangrove Sediments as a Sea-Level
Journal of Quaternary Science 20, 519–531. Indicator – Assessment from Tongatapu, Tonga. Palaeogeography
Berkeley, A., Perry, C.T., Smithers, S.G., Horton, B.P., Cundy, A.B., 2009. Palaeoclimatology Palaeoecology 74, 327–341.
Foraminiferal biofacies across mangrove–mudflat environments at Cocoa Creek, Engelhart, S.E., Horton, B.P., Douglas, B.C., Peltier, W.R., Tornqvist, T.E., 2009.
north Queensland, Australia. Marine Geology 263, 64–86. Spatial variability of late Holocene and 20th century sea-level rise along the
Bernard, E.N., Robinson, A.R., 2009. Tsunamis. Harvard University Press, Cambridge. Atlantic coast of the United States. Geology 37, 1115–1118.
Birks, H.J.B., 1995. Statistical modelling of Quaternary science data. In: Maddy, D., Engelhart, S.E., Horton, B.P., Roberts, D.H., Bryant, C.L., Corbett, D.R., 2007.
Brew, J.S. (Eds.), Quantitative Paleoenvironmental Reconstructions. Quaternary Mangrove pollen of Indonesia and its suitability as a sea-level indicator. Marine
Research Association, Cambridge, Vol. 5, pp. 161–254. Geology 242, 65–81.
Bondevik, S., Mangerud, J., Dawson, S., Dawson, A., Lohne, O., 2005. Evidence for Engelhart, S.E., Peltier, W.R., Horton, B.P., 2011. Holocene relative sea-level
three North Sea tsunamis at the Shetland Islands between 8000 and 1500 years changes and glacial isostatic adjustment of the U.S. Atlantic coast. Geology 39,
ago. Quaternary Science Reviews 24, 1757–1775. X–XX.
Bondevik, S., Svendsen, J.I., Johnsen, G., Mangerud, J., Kaland, P.E., 1997. The Fatela, F., Moreno, J., Moreno, F., et al., 2009. Environmental constraints of
Storegga tsunami along the Norwegian coast, its age and runup. Boreas 26, foraminiferal assemblages distribution across a brackish tidal marsh (Caminha,
29–53. NW Portugal). Marine Micropaleontology 70, 70–88.
Bourgeois, J., 2006. Earthquakes: a movement in four parts? Nature, News and Fujino, S., Naruse, H., Matsumoto, D., Jarupongsakul, T., Sphawajruksakul, A.,
Views. Sakakura, N., 2009. Stratigraphic evidence for pre-2004 tsunamis in
Brady, H.B., 1884. Report on the Foraminifera dredged by HMS Challenger during southwestern Thailand. Marine Geology 262, 25–28.
the years 1873–1876. Gehrels, W.R., 1994. Determining relative sea-level change from salt-marsh
Brush, G.S., 1989. Rates and patterns of estuarine sediment accumulation. foraminifera and plant zones on the Coast of Maine, USA. Journal of Coastal
Limnology and Oceanography 34, 1235–1246. Research 10, 990–1009.
Choowong, M., Phantuwongraj, S., Charoentitirat, T., Chutakositkanon, V., Yumuang, Gehrels, W.R., 2007. Sea level studies: microfossil reconstructions. In: Elias, S.A.
S., Charusiri, P., 2009. Beach recovery after 2004 Indian Ocean tsunami from (Ed.), Encyclopedia of Quaternary Sciences. Elsevier, London, pp. 3015–3023.
Phang-nga, Thailand. Geomorphology 104, 134–142. Gehrels, W.R., Newman, S.W.G., 2004. Salt-marsh foraminifera in Ho Bugt, western
Church, J.A., White, N.J., 2006. A 20th century acceleration in global sea-level rise. Denmark, and their use as sea-level indicators. Danish Journal of Geography
Geophysical Research Letters 33, L01602. 104, 97–106.
Cisternas, M., Atwater, B.F., Torrejon, F., et al., 2005. Predecessors of the giant Gehrels, W.R., van de Plassche, O., 1999. The use of Jadammina macrescens
1960 Chile earthquake. Nature 437, 404–407. (Brady) and Balticammina pseudomacrescens Bronnimann, Lutze and Whittaker
Clague, J.J., Hutchinson, I., Mathewes, R.W., Patterson, R.T., 1999. Evidence for (Protozoa: Foraminiferida) as sea-level indicators. Palaeogeography
late Holocene tsunamis at Catala Lake, British Columbia. Journal of Coastal Palaeoclimatology Palaeoecology 149, 89–101.
Research 15, 45–60. Gehrels, W.R., Roe, H.M., Charman, D.J., 2001. Foraminifera, testate amoebae and
Cleve, P.T., 1894. Synopsis of the naviculoid diatom I and II. Kongliga Svenska diatoms as sea-level indicators in UK saltmarshes: a quantitative multiproxy
Vetenskaps-Akademien 26, 1–194. approach. Journal of Quaternary Science 16, 201–220.
Cleve, P.T., 1895. Synopsis of the naviculoid diatom I and II. Kongliga Svenska Godwin, H., 1940. Studies of the post-glacial history of British vegetation. III.
Vetenskaps-Akademien 27, 1–129. Fenland pollen diagrams. IV. Post-glacial changes of relative land and sea level
Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones 311

in the English Fenland. Philosophical Transactions of the Royal Society of Hughes, J.F., Mathewes, R.W., Clague, J.J., 2002. Use of pollen and vascular plants
London B230, 239–303. to estimate coseismic subsidence at a tidal marsh near Tofino, British Columbia.
Goff, J., Chague-Goff, C., Nichol, S., 2001. Palaeotsunami deposits: a New Zealand Palaeogeography Palaeoclimatology Palaeoecology 185, 145–161.
perspective. Sedimentary Geology 143, 1–6. Hustedt, F., 1937. Systematische und okologische Untersuchungen der Diatomeen
Grindrod, J., 1985. The palynology of mangroves on a prograded shore, Princess floren von Java, Bali, Sumatra. Archiv Hydrobiologie Supplement 15, 131–506.
Charlotte Bay, north Queensland, Australia. Journal of Biogeography 12, Hustedt, F., 1939. Systematische und okologische Untersuchungen der Diatomeen
323–348. floren von Java, Bali, Sumatra. Archiv Hydrobiologie Supplement 16, 1–394.
Guilbault, J.-P., Clague, J.J., Lapointe, M., 1996. Foraminiferal evidence for the Imbrie, J., Kipp, N.G., 1971. A new micropaleontological method for quantitative
amount of coseismic subsidence during a late Holocene earthquake on Vancouver paleoclimatology. Application to a late Pleistocene Caribbean core. In: Turekian,
Island, west coast of Canada. Quaternary Science Reviews 15, 913–937. K.K. (Ed.), The Late Cenozoic glacial ags. Yale University Press, New Haven,
Hamilton, S., Shennan, I., 2005. Late Holocene relative sea-level changes and the pp. 77–131.
earthquake deformation cycle around upper Cook Inlet, Alaska. Quaternary IPCC, 2007. Climate change 2007: the physical science basis. In: Solomon, S.,
Science Reviews 24, 1479–1498. Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K.B., Tignor, M., Miller,
Hawkes, A.D., Bird, M., Cowie, S., et al., 2007. Sediments deposited by the 2004 H.L. (Eds.), Contribution of Working Group I to the Fourth Assessment. Report
Indian Ocean Tsunami along the Malaysia–Thailand Peninsula. Marine Geology of the Intergovernmental Panel on Climate Change. Cambridge University Press,
242, 169–190. Cambridge and New York, NY.
Hawkes, A.D., Horton, B.P., Nelson, A.R., 2011. Coastal subsidence in Oregon, Jankaew, K., Atwater, B.F., Sawai, Y., Choowong, M., Charoentitirat, T., Martin, M.E.,
USA, during the giant Cascadia Earthquake of AD 1700. Quaternary Science Prendergast, A., 2008. Medieval forewarning of the 2004 Indian Ocean tsunami
Reviews 30, 364–376. in Thailand. Nature 455, 1228–1231.
Hawkes, A.D., Horton, B.P., Nelson, A.R., Hill, D.F., 2010. The application of Jennings, A.E., Nelson, A.R., Scott, D.B., Aravena, J.C., 1995. Marsh Foraminiferal
intertidal foraminifera to reconstruct coastal subsidence during the giant Assemblages in the Valdivia Estuary, South-Central Chile, Relative to Vascular
Cascadia earthquake of AD 1700 in Oregon, USA. Quaternary International 221, Plants and Sea-Level. Journal of Coastal Research 11, 107–123.
116–140. Jevrejeva, S., Moore, J.C., Grinsted, A., Woodworth, P.L., 2008. Recent global sea
Hawkes, A.D., Scott, D.B., Lipps, J.H., Combellick, R., 2005. Evidence for possible level acceleration started over 200 years ago? Geophysical Research Letters 35,
precursor events of megathrust earthquakes on the west coast of North America. L08715.
Geological Society of America Bulletin 117, 996–1008. Juggins, S., 1992. Diatoms in the Thames estuary, England: ecology, palaeoecology,
Hayward, B.W., Cochran, U., Southall, K., et al., 2004. Micropalaeontological and salinity transfer function. Bibliotheca Diatomologica 25, 216.
evidence for the Holocene earthquake history of the eastern Bay of Plenty, New Kelsey, H.M., Nelson, A.R., Hemphill-Haley, E., Witter, R.C., 2005. Tsunami history of an
Zealand, and a new index for determining the land elevation record. Quaternary Oregon coastal lake reveals a 4600 yr record of great earthquakes on the Cascadia
Science Reviews 23, 1651–1667. subduction zone. Geological Society of America Bulletin 117, 1009–1032.
Hayward, B.W., Grenfell, H.R., Scott, D.B., 1999. Tidal range of marsh foraminifera Kemp, A.C., Horton, B.P., Culver, S.J., 2009a. Distribution of modern salt-marsh
for determining former sea-level heights in New Zealand. New Zealand Journal foraminifera in the Albemarle–Pamlico estuarine system of North Carolina, USA:
of Geology and Geophysics 42, 395–413. implications for sea-level research. Marine Micropaleontology 72, 222–238.
Hemphill-Haley, E., 1995. Diatom evidence for earthquake-induced subsidence and Kemp, A.C., Horton, B.P., Culver, S.J., et al., 2009b. Timing and magnitude of
tsunami 300 years ago in southern coastal Washington. Geological Society of recent accelerated sea-level rise (North Carolina, United States). Geology 37,
America Bulletin 107, 367–378. 1035–1038.
Hemphill-Haley, E., 1996. Diatoms as an aid in identifying late Holocene tsunami Knutson, T.R., McBride, J.L., Chan, J., et al., 2010. Tropical cyclones and climate
deposits. The Holocene 6, 439–448. change. Nature Geoscience 3, 157–163.
Hippensteel, S.P., Martin, R.E., 1999. Foraminifera as an indicator of overwash Lawler, A., 2011. Did the first cities grow from marshes? Science 331, 141.
deposits, Barrier Island sediment supply, and barrier island evolution: folly Le Roux, J.P., Nielsen, S.N., Kemnitz, H., Henriquez, A., 2008. A Pliocene mega-
island, South Carolina. Palaeogeography Palaeoclimatology Palaeoecology 149, tsunami deposit and associated features in the Ranquil Formation, southern
115–125. Chile. Sedimentary Geology 203, 164–180.
Hippensteel, S.P., Martin, R.E., 2000. Foraminifera as indicators of storm Leonard, L.J., Currie, C.A., Mazzotti, S., Hyndman, R.D., 2010. Rupture area and
deposition: implications for barrier island sediment supply and evolution. displacement of past Cascadia great earthquakes from coastal coseismic
In: Martin, R.E. (Ed.), Environmental Micropaleontology. Kluwer Press, subsidence. Geological Society of America Bulletin 122, 1951–1968.
Dordrecht, pp. 351–369. Liu, K.-b., Fearn, M.L., 1993. Lake-sediment record of late Holocene hurricane
Hippensteel, S.P., Martin, R.E., Harris, M.S., 2005. Discussion: records of activities from coastal Alabama. Geology 21, 793–796.
prehistoric hurricanes on the South Carolina Coast based on Lloyd, J., 2000. Combined foraminiferal and thecamoebian environmental
micropaleontological and sedimentological evidence, with comparison to other reconstruction from an isolation basin in NW Scotland: implications for sea-
Atlantic Coast records. Geological Society of America Bulletin 117, 250–256. level studies. The Journal of Foraminiferal Research 30, 294–305.
Honig, C.A., Scott, D.B., 1987. Postglacial stratigraphy and sea-level change in Long, A.J., 1992. Coastal responses to changes in sea-level in the East Ken Fens
Southwestern New-Brunswick. Canadian Journal of Earth Sciences 24, 354–364. and southeast England, UK over the last 7500 years. Proceedings of the
Horton, B.P., Edwards, R.J., 2006. Quantifying Holocene sea-level change using Geologists’ Association 103, 187–199.
intertidal foraminifera: lessons for the British Isles. Cushman Foundation for Long, A.J., Roberts, D.H., Dawson, S., 2006. Early Holocene history of the west
Foraminiferal Research Special Publication 40, 97. Greenland Ice Sheet and the GH-8.2 event. Quaternary Science Reviews 25,
Horton, B.P., Edwards, R.J., Lloyd, J.M., 1999. UK intertidal foraminiferal 904–922.
distributions: implications for sea-level studies. Marine Micropaleontology 36, Lowe, R. L., 1974. Environmental Requirements and Pollution Tolerance of
205–223. Freshwater Diatoms. EPA-670/4-74-005. Cincinnati, OH. 344 pp.
Horton, B.P., Edwards, R.J., Lloyd, J.M., 2000. Implications of a microfossil transfer Mamo, M., Strotz, L., Dominey-Howes, D., 2009. Tsunami sediments and their
function in Holocene sea-level studies. In: Shennan, I., Andrews, J.E. (Eds.), foraminiferal assemblages. Earth Science Reviews 96, 263–278.
Holocene land–ocean interaction and environmental change around the western McMurtry, G.M., Watts, P., Fryer, G.J., Smith, J.R., Imamura, F., 2004. Giant
North Sea. Geological Society Special Publications, London, pp. 41–54. landslides, mega-tsunamis, and paleo-sea level in the Hawaiian Islands. Marine
Horton, B.P., Larcombe, P., Woodroffe, S.A., Whittaker, J.E., Wright, M.R., Wynn, C., Geology 203, 219–233.
2003. Contemporary foraminiferal distributions of a mangrove environment, Minoura, K., Imamura, F., Sugawara, D., Kono, Y., Iwashita, T., 2001. The 869
Great Barrier Reef coastline, Australia: implications for sea-level reconstructions. Jogan tsunami deposit and recurrence interval of largescale tsunami on the
Marine Geology 198, 225–243. Pacific coast of northeast Japan. Journal of Natural Disaster Science 23, 83–88.
Horton, B.P., Rossi, V., Hawkes, A.D., 2009. The sedimentary record of the 2005 Monecke, K., Finger, W., Klarer, D., Kongko, W., McAdoo, B.G., Moore, A.L.,
hurricane season from the Mississippi and Alabama coastlines. Quaternary Sudrajat, S.U., 2008. A 1,000-year sediment record of tsunami recurrence in
International 195, 15–30. northern Sumatra. Nature 455, 1232–1234.
Horton, B.P., Whittaker, J.E., Thomson, K.H., Hardbattle, M.I.J., Kemp, A., Moore, P.D., Webb, J.A., Collinson, M.E., 1991. Pollen Analysis. Blackwell Science,
Woodroffe, S.A., Wright, M.R., 2005. The development of a modern foraminiferal London, 216 pp.
data set for sea-level reconstructions, Wakatobi Marine National Park, Southeast Murray, J.W., 2006. Ecology and Applications of Benthic Foraminifera. Cambridge
Sulawesi, Indonesia. Journal of Foraminiferal Research 35, 1–14. University Press, Cambridge, 426 pp.
312 Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones

Nanayama, F., Satake, K., Furukawa, R., Shimokawa, K., Atwater, B.F., Shigeno, K., Scott, D.B., Collins, E.S., Gayes, P.T., Wright, E., 2003. Records of prehistoric
Yamaki, S., 2003. Unusually large earthquakes inferred from tsunami deposits hurricanes on the South Carolina coast based on micropaleontological and
along the Kuril trench. Nature 424, 660–663. sedimentological evidence, with comparison to other Atlantic Coast records.
Nelson, A.R., 2007. Tectonics and relative sea-level change. In: Elias, S.A. (Ed.), Geological Society of America Bulletin 115, 1027–1039.
Encyclopedia of Quaternary Science. Elsevier, Amsterdam, pp. 3072–3087. Scott, D.K., Leckie, R.M., 1990. Foraminiferal zonation of Great Sippewissett Salt-
Nelson, A.R., Jennings, A.E., Kashima, K., 1996. An earthquake history derived from Marsh (Falmouth, Massachusetts). Journal of Foraminiferal Research 20,
stratigraphic and microfossil evidence of relative sea-level change at Coos Bay, 248–266.
southern coastal Oregon. Geological Society of America Bulletin 108, Scott, D.B., Medioli, F.S., 1978. Vertical zonations of marsh foraminifera as accurate
141–154. indicators of former sea-levels. Nature 272, 528–531.
Nelson, A.R., Sawai, Y., Jennings, A.E., et al., 2008. Great-earthquake paleogeodesy Scott, D.B., Medioli, F.S, 1980. Quantitative studies of marsh foraminiferal
and tsunamis of the past 2000 years at Alsea Bay, central Oregon coast, USA. distributions in Nova Scotia: implications for sea level studies. Cushman
Quaternary Science Reviews 27, 747–768. Foundation for Foraminiferal Research Special Publication 17, 58.
Palmer, A.J.M., Abbott, W.H., 1986. Diatoms as indicators of sea-level change. Shennan, I., 1986. Flandrian sea-level changes in the Fenland. II: Tendencies of
In: van de Plassche, O. (Ed.), Sea-level Research: A Manual for the Collection sea-level movement, altitudinal changes, and local and regional factors. Journal
and Evaluation of Data. Geobooks, Norwich, pp. 457–488. of Quaternary Science 1, 155–179.
Parsons, M.L., 1998. Salt marsh sedimentary record of the landfall of Hurricane Shennan, I., Hamilton, S., 2006. Coseismic and pre-seismic subsidence associated
Andrew on the Louisiana coast: diatoms and other paleoindicators. Journal of with great earthquakes in Alaska. Quaternary Science Reviews 25, 1–8.
Coastal Research 14, 939–950. Shennan, I., Horton, B., 2002. Holocene land- and sea-level changes in Great
Patterson, R.T., Dalby, A.P., Roe, H.M., Guilbault, J.-P., Hutchinson, I., Clague, J.J., Britain. Journal of Quaternary Science 17, 511–526.
2005. Relative utility of foraminifera, diatoms and macrophytes as high Shennan, I., Horton, B., Innes, J., Gehrels, R., Lloyd, J., McArthur, J.,
resolution indicators of paleo-sea level in coastal British Columbia, Canada. Rutherford, M., 2000. Late Quaternary sea-level changes, crustal movements and
Quaternary Science Reviews 24, 2002–2014. coastal evolution in Northumberland, UK. Journal of Quaternary Science 15,
Peltier, W.R., 2004. Global glacial isostasy and the surface of the ice-age earth: the 215–237.
ice-5G (VM2) model and grace. Annual Review of Earth and Planetary Sciences Shennan, I., Scott, D.B., Rutherford, M.M., Zong, Y., 1999. Microfossil analysis of
32, 111–149. sediments representing the 1964 earthquake, exposed at Girdwood Flats, Alaska.
Peltier, W.R., Shennan, I., Drummond, R., Horton, B., 2002. On the postglacial Quaternary International 60, 55–73.
isostatic adjustment of the British Isles and the shallow viscoelastic structure of Sherrod, B.L., 1999. Gradient analysis of diatom assemblages in a Puget Sound
the Earth. Geophysical Journal International 148, 443–475. salt marsh: can such assemblages be used for quantitative paleoecological
Pinegina, T.K., Bourgeois, J., Bazanova, L.I., Melekestsev, I.V., Braitseva, O.A., 2003. reconstructions? Palaeogeography Palaeoclimatology Palaeoecology 149,
A millennial-scale record of Holocene tsunamis on the Kronotskiy Bay coast, 213–226.
Kamchatka, Russia. Quaternary Research 59, 36–47. Smith, D.E., Shi, S., Brooks, C.L., et al., 2004. The Holocene Storegga Slide
Plafker, G., 1969. Tectonics of the March 27, 1964, Alaska earthquake. U.S. tsunami in the United Kingdom. Quaternary Science Reviews 23/24,
Geological Survey Professional Paper 543, 74 pp. 2295–2325.
Plater, A.J., Shennan, I., 1992. Evidence of Holocene sea-level change from the Stone, A., Kerr, R.A., 2005. Indian Ocean Tsunami – girding for the next killer
Northumberland coast, eastern England. Proceedings of the Geologists’ wave. Science 310, 1602–1605.
Association 103, 201–216. Switzer, A.D., Jones, B.G., 2008. Large-scale washover sedimentation in a
van de Plassche, O., 1986. Sea-Level Research: A Manual for the Collection and freshwater lagoon from the southeast Australian coast: sea-level change, tsunami
Evaluation of Data. Geobooks, Norwich. or exceptionally large storm? Holocene 18, 1014–1016.
Sachs, H.M., Webb, T., Clark, D.R., 1977. Paleoecological Transfer Functions. Szkornik, K., Gehrels, W.R., Kirby, J.R., 2006. Salt-marsh diatom distributions in Ho
Annual Review of Earth and Planetary Sciences 5, 159–178. Bugt (western Denmark) and the development of a transfer function for
Satake, K., Atwater, B.F., 2007. Long-term perspectives on giant earthquakes and reconstructing Holocene sea-level changes. Marine Geology 235, 137–150.
tsunamis at subduction zones. Annual Review of Earth and Planetary Sciences Tooley, M.J., 1978. Sea-Level Changes in North-West England during the Flandrian
35, 349–374. Stage. Oxford University Press, Oxford, 232 pp.
Sawai, Y., 2001. Distribution of living and dead diatoms in tidal wetlands of Tooley, M.J., 1982. Sea-level changes in northern England. Proceedings of the
northern Japan: relations to taphonomy. Palaeogeography Palaeoclimatology Geologists’ Association 93, 43–51.
Palaeoecology 173, 125–141. Turetsky, M.R., Manning, S.W., Wieder, R.K., 2004. Dating recent peat deposits.
Sawai, Y., Fujii, Y., Fujiwara, O., et al., 2008. Marine incursions of the past 1500 Wetlands 24, 324–356.
years and evidence of tsunamis at Suijin-numa, a coastal lake facing the Japan Vos, P.C., de Wolf, H., 1988. Methodological aspects of paleo-ecological
Trench. The Holocene 18, 517. diatom research in coastal areas of the Netherlands. Geologie en Mijnbouw 67,
Sawai, Y., Jankaew, K., Martin, M.E., Prendergast, A., Choowong, M., Charoentitirat, 31–40.
T., 2009a. Diatom assemblages in tsunami deposits associated with the 2004 Williams, H.F.L., 1994. Intertidal benthic foraminiferal biofacies on the Central Gulf
Indian Ocean tsunami at Phra Thong Island, Thailand. Marine Micropaleontology Coast of Texas – modern distribution and application to sea-level
73, 70–79. reconstruction. Micropaleontology 40, 169–183.
Sawai, Y., Kamataki, T., Shishikura, M., et al., 2009b. Aperiodic recurrence of Woodroffe, S.A., Long, A.J., 2010. Reconstructing recent relative sea-level changes
geologically recorded tsunamis during the past 5500 years in eastern Hokkaido, in West Greenland: local diatom-based transfer functions are superior to
Japan. Journal of Geophysical Research 114, B01319. regional models. Quaternary International 211, 91–103.
Sawai, Y., Nasu, H., Yasuda, Y., 2002. Fluctuations in relative sea-levels during the Zong, Y.Q., Horton, B.P., 1999. Diatom-based tidal-level transfer functions as an aid
past 3000 years at the Onnetoh estuary, Hokkaido, northern Japan. Journal of in reconstructing Quaternary history of sea-level movements in the UK. Journal
Quaternary Sciences 17, 607–622. of Quaternary Science 14, 153–167.
Sawai, Y., Satake, K., Kamataki, T., et al., 2004. Transient uplift after a 17th-century Zong, Y.Q., Shennan, I., Combellick, R.A., Hamilton, S.L., Rutherford, M.M., 2003.
earthquake along the Kuril subduction zone. Science 306, 1918–1920. Microfossil evidence for land movements associated with the 1964 AD Alaska
Scott, D.B., Collins, E.S., Duggan, J., Asioli, A., Saito, T., Hasegawa, S., 1996. earthquake. Holocene 13, 7–20.
Pacific rim marsh foraminiferal distributions: implications for sea-level studies. Zong, Y.Q., Tooley, M.J., 1996. Holocene sea-level changes and crustal movements in
Journal of Coastal Research 12, 850–861. Morecambe Bay, northwest England. Journal of Quaternary Science 11, 43–58.
Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones 313

Biographical Sketch

Dr. Ben Horton is an associate professor at the University of Pennsylvania. He received his undergraduate degree
from the University of Liverpool and his PhD from Durham University, UK. His research concerns sea-level
change. He aims to understand and integrate the external and internal mechanisms that have determined sea-level
changes in the past and which will shape such changes in the future. Dr Horton has published over 100 articles in
peer-reviewed journals, including Science, Proceedings of the National Academy of Sciences, and Geology. He is an
author of the Intergovernmental Panel on Climate Change (IPCC) 5th Assessment Report (5AR). He is a com-
mittee member of the National Research Council, USA and PALSEA (PALeo-constraints on SEA-level rise). He is
project leader of International Geoscience Programme (IGCP) 588 and scientific advisor to The Future Ocean
research of the German Research Foundation. Dr Ben Horton has received a Medal for Research Excellence by the
Commanding General of the North Atlantic Division of the United States Army Corps of Engineers (2010);
Geological Society of America’s The W. Storrs Cole Memorial Research Award (2007); The Linnean Society (UK)
Award (2004); the Higher Education Funding Council for England Excellence in Teaching Award (2003); and the
Menzies Australian Bicentennial Award (2001).

Dr Simon Engelhart is a postdoctoral fellow at the University of Pennsylvania. He received his undergraduate and
master’s degrees from Durham University, UK, and his PhD from the University of Pennsylvania, USA. His main
research focus is sea-level change, aiming to understand the mechanisms that have determined sea-level changes
in the past on both passive and active margins and the implications for the future. His PhD research focused on
Holocene changes in relative sea level on the United States’ Atlantic coast (including two publications in Geology),
while his postdoctoral experience includes deciphering the magnitude of co-seismic subsidence associated with
late Holocene megathrust earthquakes in central Oregon.

Dr. Andrew Kemp is a postdoctoral associate at the Yale University Climate and Energy Institute and School of
Forestry and Environmental Studies. He received his undergraduate degree from the University of Durham (UK)
and PhD from the University of Pennsylvania. His research has focused on reconstructing sea-level using salt
marshes, with a particular interest in the late Holocene and Atlantic coast of the United States.
314 Microfossils in Tidal Settings as Indicators of Sea-Level Change, Paleoearthquakes, Tsunamis, and Tropical Cyclones

Dr. Yuki Sawai is a senior research scientist at Geological Survey of Japan, National Institute of Advanced In-
dustrial Science and Technology. He earned BS, MS, and PhD degrees in paleontology at Kyushu University, Japan.
His main research is understanding subduction zone earthquakes by studying relative sea-level changes based on
diatom micropaleontology and tsunami deposits on the Pacific Rim and on shores of the eastern Indian Ocean.
His study has entailed observations of modern diatom assemblages. Sawai built modern diatom data sets in
Japan, the United States, and Chile in support of diatom analysis for quantitative sea-level reconstructions. In this
process, he made findings about the taphonomy and taxonomy of coastal diatoms.
 14.26 Palynology and Its Application to Geomorphology
PT Moss, The University of Queensland, Brisbane, QLD, Australia
r 2013 Elsevier Inc. All rights reserved.

14.26.1 Introduction 315

14.26.2 Palynological Analysis 316
14.26.2.1 Palynological Laboratory Techniques 316
14.26.2.2 Pollen and Charcoal Counting Methods 317
14.26.2.3 Generation of Pollen Diagrams 319
14.26.3 Palynology and Its Applications to Geomorphology 320
14.26.3.1 Age Control 320
14.26.3.2 Environmental Palynology 321
14.26.3.3 Depositional Environments 322
14.26.3.4 Human Impact 323
14.26.4 Conclusion 323
14.26.5 Use of Exotic Markers 324
References 324

Glossary Palynology Analysis of fossil pollen and spores used for

Carbonized particles Burnt organic material (i.e., reconstructing past environments.
charcoal) that occurs in sedimentary rocks or sediment. Palynomorphs Microfossils (generally pollen and spores)
They are commonly associated with palynomorphs as preserved in sedimentary rocks or sediment.
microcharcoal. Sporopollenin Decay-resistant organic compound that
Environmental palynology The use of palynomorphs to forms the surface of pollen and spores.
reconstruct past environmental conditions and ecological Stratigraphic palynology The use of palynomorphs to
communities. correlate sedimentary sequences or provide age control for
Exines Distinctive surface structure and morphology of sedimentary sequences.
pollen and spores.

Abstract

Pollen analysis can provide a number of valuable applications to a geomorphological context, which include providing
both relative and absolute age control; significant environmental data in terms of past climates and depositional
environments; and information on impacts of people on landforms and geomorphological processes. This chapter provides
an overview of these applications, as well as a detailed examination of palynological laboratory techniques, pollen iden-
tification/counting, and the production/interpretation of pollen diagrams.

14.26.1 Introduction related to both taxonomic origin and depositional environ-

ment. Generally, fossil pollen and spores preserve best in
Pollen analysis is the study of fossil pollen and spores and is anaerobic conditions (e.g., bogs, swamps, lakes, and marine
one of the main methods for reconstructing past landscapes. sediments) and can be destroyed if any oxidation occurs
Pollen grains are unicellular, microscopic (15 to 100 mm) re- (Coil et al., 2003; Seppä, 2007). In addition, different plants
productive gametes of seed-producing plants, whereas spores species have different pollen and spore production rates and
are the analogous organs of nonflowering pteridophytes also utilize different transport vectors (i.e., wind, water, or
(mosses, ferns, and fern allies) (Faegri and Iversen, 1991; animal) that can significantly impact pollen representation
Coil et al., 2003). Preservation of pollen and spores (or (Seppä, 2007). For instance, Pinus pollen can be very well
palynomorphs) in sediments depends on characteristics represented in a sediment record, due to high production
and wind dispersal rates (Pohl, 1937; Seppä, 2007), whereas
pollen of the Lauraceae family is rarely preserved in sediments
Moss, P.T., 2013. Palynology and its application to geomorphology. In:
due to low production and dispersal by animal vectors
Shroder, J. (Editor in Chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise
on Geomorphology. Academic Press, San Diego, CA, vol. 14, Methods in (Macphail, 1980). During sedimentation the soft-bodied
Geomorphology, pp. 315–325. reproductive cellular contents are lost, with fossil pollen

Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00395-X 315

316 Palynology and Its Application to Geomorphology

and spores consisting of a decay resistant organic compound, The initial step consists of sampling the sediment. The
that is, sporopollenin (Coil et al., 2003; Seppä, 2007). volume of sample required is dependent on the suspected
The distinctive surface structure and morphology (exines) pollen concentration occurring within the sediment. If the
of fossil pollen and spores can generally be taxonomically sample contains a high amount of organic material (i.e.,
assigned to their plant of origin with a high degree of peats) then generally around 1 cm3 is taken and if the sedi-
precision, sometimes to species but more often to genus or ment is mainly composed of inorganic materials (i.e., deep
family level. marine sediments) then generally around 5 to 10 cm3 of
The general techniques that are used to extract fossil sediment is extracted. In addition, sampling intervals for the
pollen from sediment samples are reviewed here. This review sediment core or rock sample can be selected at this stage. The
includes information on basic laboratory techniques; an out- sampling interval strategy depends on the aims of the pro-
line of pollen identification and counting methods; a review posed research and the total length of the sediment core/rock
of charcoal identification and counting techniques; and pollen exposure. Many palynological studies having sample intervals
data representation, particularly the construction and inter- of approximately 10 to 20 cm along the length of a sediment
pretation of pollen diagrams. Further, an overview of the four core/rock exposure. Furthermore, exotic pollen marker tablets
main areas in which pollen analysis can be used in a geo- can be added in this step, the most commonly used marker
morphological context are provided. These include: (1) age types are Lycopodium clavatum or Eucalyptus sp., and the
control; (2) environmental palynology; (3) depositional en- researcher has to ensure that these marker types are not lo-
vironments; and (4) human impacts/land uses. In each area a cated in their study area. The tablets have a known concen-
case study is discussed that illustrates how palynology has tration of pollen grains, which can then be counted along with
been applied from a geomorphological perspective. the palynomorphs and carbonized particles in the sample to
determine pollen and charcoal concentration values (Maher,
1972; Salgado-Labouriau and Rull, 1986; Wang et al., 1999).
The tablets can be dissolved by heating (to 100 1C) in ap-
14.26.2 Palynological Analysis
proximately 10 ml of distilled water or adding them to 10 ml
of hydrochloric acid (10%). The sediment sample is then
14.26.2.1 Palynological Laboratory Techniques
added to this mixture before the next stage. Plastic micro-
The laboratory techniques used in extracting pollen from spheres can be used in place of exotic marker tablets and are
sediments are outlined in Table 1. These processes have been generally added to the pollen concentrate at the end of the
undertaken since 1916 with Lennart van Post outlining the laboratory analysis procedures (Salgado-Labouriau and Rull,
initial steps in his seminal Kristina lecture (Faegri and Iversen, 1986).
1991). Subsequent research has added a number of key steps, The next stage involves sample disaggregation and defloc-
including the use of hydrofluoric acid (Assarson and Granlund, culation. If the samples are lithified then they may need
1924), the acetylosis process (Erdtman, 1934) and the use of to undergo mechanical disaggregation (i.e., grinding) before
nontoxic heavy liquid solutions (van der Kaars, 1991; Riding deflocculation. Clay dispersal can be achieved by using a
and Kyffin-Hughes, 2004, 2006), that have greatly improved the chemical deflocculant such as tetra-sodium pyrophosphate
success of pollen analysis techniques in terms of the quantity (10%), generally approximately 30 to 40 ml is added to
and quality of palynomorphs extracted from sediment cores or the sample in a 100 ml beaker. The beaker is then heated to
rock samples and are outlined below in more detail. 100 1C for 40 min with occasional stirring to break up larger

Table 1 Summary of pollen analysis techniques

Aims Laboratory techniques

Sample preparation • Generally 1 to 10 cm3 of sediment or rock material is used.

• Addition of exotic pollen marker can also occur at this stage.
Sample deflocculation and disaggregation • Lithified samples may require mechanical disaggregation (i.e., grinding or crushing).
• Clay dispersal can be achieved through the use of a deflocculation chemical.
• Heating and stirring of sample can be used to break up larger aggregates.
• Sieving can also be used to remove both course and fine fractions in the samples.
• Samples with high organic or calcium carbonate content may need to undergo
chemical disaggregation to remove excess material.
Removal of unwanted inorganic material • Acid digestion can be used to remove unwanted inorganic material (i.e., silicates).
• Alternatively, the organic fraction can be separated (floated off) from the inorganic
fraction through the use of a heavy liquid solution.
Removal of unwanted fine organic material and staining of • Acetolysis is used to dissolve fine cellulose material. This process also stains the
pollen grains and spores pollen grains.
• Saffarin stain can be used in the absence of acetolysis if there is any concern over
loss of palynomorphs in low concentration samples.
Mounting of pollen samples on slides • Samples are transferred to a suitable mounting medium and then mounted on slides
for pollen, spore, and microcharcoal counting.
 Palynology and Its Application to Geomorphology 317

aggregates. After cooling, the sample is sieved across a 200 mm The process of acetolysis is undertaken to dissolve excess
mesh to remove material above this size (mainly sand, root- organic (cellulose) material and to darken the palynomorphs to
lets, and other large organic material) and an 8 mm mesh to assist with identification. Prior to acetolysis, it is very important
remove particles (mainly clay) below this size. The samples are that the samples are washed in concentrated glacial acetic acid to
then transferred back to a 100 ml beaker and left overnight to prevent the acetolysis mixture and water contained in the sam-
settle. If the material contains high concentrations of calcium ples reacting, as this will cause a violent and potentially dan-
carbonate then approximately 20 ml of hydrochloric acid gerous reaction. Acetolysis involves adding a 9:1 mixture of
(10%) is added to chemically disaggregate the calcium car- acetic anhydride and concentrated sulfuric acid to the samples
bonate from the sample and the samples are again left over- (generally 9 ml acetic anhydride to 1 ml sulfuric acid), which are
night to settle. The next step involves siphoning the excess then heated for up to 5 min. The samples are then centrifuged at
liquid from the sample and then transferring the material to a 3000 rpm for 5 min and the excess acetolysis liquid is poured
15 ml plastic test-tube. The sample then undergoes a wash to off into a suitable waste container (along with the glacial acetic
remove excess tetra-sodium pyrophosphate. This process in- acid) for neutralizing and disposal at a later date. The samples
volves adding approximately 10 ml of distilled water, then then undergo two water washes before the next step. There is
centrifuging the sample for 3 to 5 min at 3000 rotations per some evidence that prolonged treatment (i.e., beyond 3 min) in
minute (rpm) and pouring off the excess liquid. This step is the acetolysis mixture may damage and corrode pollen grains
repeated again before the next stage is undertaken. If the (Charman, 1992). Therefore, if there are concerns about ob-
sample is suspected of containing high amounts of organic taining suitable pollen yields from samples with low palyno-
materials then potassium hydroxide (10%) can be used to morph concentrations (e.g., marine sediments) or if key taxa
chemically disaggregate the humic acids from the sample. with delicate exines (e.g., members of the Juncaceae) may be
This stage involves adding 10 ml of potassium hydroxide lost from the record (see Nilsson and Praglowski, 1992), then
to the test-tube, heating at 100 1C for 5 min, centrifuging pollen concentrate samples can be stained with safranin in lieu
(at 3000 rpm) and pouring off the excess liquid. Two water of darkening by acetolysis.
washes are then undertaken before the next step. Before slide mounting the samples need to be washed in
The next involves the removal of unwanted inorganic absolute ethanol to remove excess water. This is achieved by
material (mainly silicates) and there are two main methods adding 10 ml of ethanol, then centrifuging at 3000 rpm,
that can be used for this: the acid digestion method (Faegri pouring of the excess liquid and then transferring to a 1 ml
and Iversen, 1991) and the heavy liquid method (van der or 0.5 ml container. The smaller containers are then centri-
Kaars, 1991). The acid digestion method involves digesting fuged at 2500 rpm, with the excess ethanol poured off.
the silicates with 10 ml of hydrofluoric acid (B40%) added to A suitable mounting medium is then added, most commonly
the sample, which is then left overnight. The sample is then glycerol or silicon oil is used but if palynomorphs need to be
centrifuged (at 3000 rpm) and the excess liquid is transferred fixed then they can be mounted in glycerol jelly. Pipettes can
to a suitable waste hydrofluoric acid container for neutralizing then be used to transfer subsamples (generally a few drops) to
and disposal at a later date. The samples then undergo two a slide for counting. The most common sealing agents that are
water washes before the next step. It should be noted that used to fix cover-slips over the pollen concentrate samples
hydrofluoric acid is highly toxic and corrosive and requires include clear nail polish or wax, with wax being used if the
special precautions, including special protective personal samples are required to be stored for a longer time as slides
equipment (face shield, gloves, and apron), exclusive use in a sealed with this method are more permanent than those
fume-hood and the provision of calcium gluconate gel (2.5%) sealed with nail polish.
to treat chemical burns from accidental spillages (Riding and
Kyffin-Hughes, 2004, 2006; Roblin et al., 2006). In addition,
14.26.2.2 Pollen and Charcoal Counting Methods
the acid digestion method may not be particularly effective for
samples with a high component of silicate material (i.e., Pollen, spores, and carbonized particles are generally counted
marine sediments) and the heavy liquid method was initially under a compound light microscope at between  400 and
developed for deep ocean marine sediments but can also be  650 magnification. There are a range of reference collections
applied to wide variety of different sediment types that have that can be used to aid with identification of pollen and spore
traditionally been processed using the acid digestion method, taxa (e.g., USDA pollen reference collection, European pollen
including peats, silts, and clays (van der Kaars, 1991). This reference collection and the Australasian pollen and spore
method (using sodium polytungstate) also has the advantage atlas). The selection of the appropriate reference collection
of being relatively nontoxic and not requiring the special depends on the location of the study site and some examples
precautions that are associated with the use of hydrofluoric of some excellent general collections are shown in Table 2. In
acid. Heavy liquid separation is achieved through the addition addition, a list of the local and regional vegetation can be very
of 6 ml of sodium polytungstate (with a specific density of useful in assisting with taxonomic identification, as well
2.0) into the samples (at the same stage as the acid digestion as a review of any previous palynological studies undertaken
method) the sample is then centrifuged at 2500 rpm for 20 to within the study region. Generally, a minimum pollen count
30 min. The lighter organic fraction floats to the surface, that should be applied to a pollen sample is at least 300
whereas the heavier inorganic fraction is left at the bottom of identifiable dryland pollen grains. The dryland group gener-
the test-tube. The organic fraction is then separated by pouring ally consists of arboreal taxa and dryland herbs, whereas the
off into another 15 ml test-tube and then two water washes are wetland group (consisting of local aquatic taxa and fern
undertaken prior to the next step. spores) is also counted but not generally incorporated within
318 Palynology and Its Application to Geomorphology

Table 2 Examples of pollen and spore reference collection websites

Reference collection URL

United States Department of Agriculture (USDA) reference collection http://pollen.usda.gov/reference_collection/

Reference_Collection.htm
European pollen reference collection http://www.geog.ox.ac.uk/research/biodiversity/lel/refcollections/
pollen-european.html)
African pollen database http://medias3.mediasfrance.org/apd/accueil.htm
The University of Arizona catalogues of internet pollen and spore images http://www.geo.arizona.edu/palynology/polonweb.html
Australasian Pollen and Spore Atlas http://apsa.anu.edu.au

Figure 1 Examples of fossil pollen: (a) Casuarinaceae pollen (equatorial diameter:B40 mm), Quaternary South East Queensland tree. (b)
Poaceae pollen (equatorial diameter:B30 mm), Quaternary South East Queensland grass. (c) Typha pollen (equatorial diameter:B30 mm),
Quaternary North East Queensland bullrush. (d) Polygonum pollen (equatorial diameter:B50 mm), Quaternary North East Queensland aquatic
species. (e) Alnus pollen (equatorial diameter:B20 mm), Eocene Okanagan Highlands, British Columbia alder species. (f) Quercus pollen
(equatorial diameter:B30 mm), Eocene Okanagan Highlands, British Columbia oak species. (g) Sequioa pollen (equatorial diameter:B30 mm),
Eocene Okanagan Highlands, British Columbia redwood species. (h) Picea pollen (total length:B80 mm), Eocene Okanagan Highlands, British
Columbia spruce species, and (i) microcharcoal particles (B10 to 15 mm long), Quaternary South East Queensland.

the pollen sum. The minimum pollen count can vary method involves counting at least 200 points on the slide and
depending of site characteristics and the general aims of the using the following equation to calculate charcoal concen-
pollen analysis. The level of taxonomic identification of tration (pollen concentrations can also be determined by this
the pollen taxa vary between species, genus and family levels. method as well):
Figure 1 provides an example of some pollen grains, as well
Concentration ¼ C  A=AðtÞ  V ðpmÞ=V ðsÞ
as an image of carbonized particles. In addition, taxa can be
divided into a number of groupings based on the occurrence where
in extant plant communities or associations. C (or P) – number of carbonized particles (or palyno-
There are two methods that can be utilized with charcoal morphs) counted,
analysis the point count method devised by Clark (1982) and A – area of a slide,
the spike concentration method of Wang et al. (1999). Both A(t) – area of slide traversed,
methods involve identifying charcoal or carbonized particles V (pm) – volume of prepared material in glycerol,
as black angular particles above 10 mm in size, the point count V (s) – volume of prepared material in one slide.
 Palynology and Its Application to Geomorphology 319

The spike concentration method involves the addition (i.e., vegetation groupings, ash content, and pollen/charcoal
of an exotic marker pollen grain (generally Lycopodium sp. or concentration). The pollen data graphs are converted into
Eucalyptus sp.) or plastic microspheres, then counting the percentage format by the TILIA program, based on the dryland
number of carbonized particles over three transects, along pollen sum, which in this case consist of rainforest, sclerophyll
with the number of grains or microspheres from which char- arboreal taxa and sclerophyll sum and should add up to
coal (and pollen) concentration can be determined by the 100%. Also shown are the wetland taxa components, com-
following equation: prising aquatic taxa and pteridophytes, and these taxa occur
outside of the pollen sum (forming the total pollen sum). The
Concentration ¼ e=E  C y-axis consists of the depth variables, with the shallowest
sample at the top and the deepest sample at the bottom. In
where
addition, ages of the sediment can be shown along the y-axis
e – number of grains or microspheres counted in slide,
(in this case a calibrated AMS 14C age) and the sediment
E – known concentration of grains/microspheres added
stratigraphy can also be shown along the y-axis if appropriate
to sample,
(Table 3).
C (or P) – number of carbonized particles (or palyno-
The pollen diagram can also be zoned by vertical lines and
morphs) counted in slide. Both charcoal and pollen concen-
these lines can be based on chronozones (e.g., oxygen isotope
trations are expressed as number per cubic centimeter.
stages) or by classification of the pollen data. In this example
the pollen data were classified using a stratigraphically con-
strained classification employing the Edward’s and Cavali-
14.26.2.3 Generation of Pollen Diagrams
Sforza’s chord distance measure (Grimm, 1987), which has
A number of specialized computer programs can be used to produced four zones (LSS A to D) and these zones are com-
present pollen data as pollen diagrams. Common programs monly used to discuss the pollen results, generally from bot-
that are used include TILIA (Grimm, 2004) and C2 (Juggins, tom to top. Influx diagrams are the other pollen diagram
2003). Percentage diagrams are generally the most common format and they are generally much rarer, as they are based on
pollen diagram format and Figure 2 presents an example of calculation of pollen concentrations and also require esti-
a percentage pollen diagram produced in the TILIA program mates of pollen production and dispersal from extant plants,
(Mackenzie et al., 2009). The x-axis consists of graphs which can be difficult to determine (Seppä, 2007). The main
of the palynological taxa and other useful information advantage of influx diagrams is that they can discern patterns

)
00
10

)
00
x
m3

10
rc

x
m3
pe

rc
s
ie

pe
tic
ar

ns
(p

i
ii

ra
a
e)
am

ax

(g
Sp kso au eae lora

io
es
gh

lt
a

n
t
ra
or

ea
tic

tio
a
in

f
D the dia buli
al

nt
sp
H no um nta s

hy bs
rc
Po lloc us nn

or

tra
o n a li
us et

ce
M agh nia tra
e

rb

r
ya o tu
y rp cu

er ris m

op e

en
pt ap

ea

ae

s
ic a c

on
la

id ll h
C nop e (

te
s
Eu ade us

yp te er

nc
Ph oca us

ac

ce
ea
s

lc
C tiop te f

yl
m lad

le st
he ea

er y
ca rri

co
d ag

id

na

ph

Pt ph
te e

oa
Sc fore
ac
C rac
As ea

tic
ris

is le
dr

Po ia
Po hof

tio

ro

ro

n
ly

rc
ua
ac

ac
lit

ac

le
n

le

ha
es
ot

al

ai

Zones
Ep

Aq

Po
Ac

Sc

Coniss
N

0
5
LSS D

10
15
3410<61 Ical. yrs BP

20
25
Depth (cm)

LSS C

30
35
40
LSS B

45
50
55
LSS A

60
65
70
0
0.20
0.40
0.60
0.81
0.01
0.21
0.41
0.61
0.82
0.02
0.22
0.42
0.62
0.83
0.0

20 40 60 20 20 40 60 20 20 20 20 40 20 20 40 60 80 100 20 40 20 40 0 0
50 100 150
0
20
0
40
0
60
0 Total sum of squares

Figure 2 Example percentage pollen diagram from Lake Selina Swamp, Tasmania, Australia. Adapted from Mackenzie, L.L., Moss, P.T., Petherick,
L., 2009. Mid to late Holocene vegetation and environments of Lake Selina Swamp, western Tasmania. Quaternary of Australia 26, 2–7.

Table 3 Pollen diagram programs

Pollen diagram programs Reference and web link

TILIA Grimm (2004); http://www.ncdc.noaa.gov/paleo/tiliafaq.html

C2 Juggins (2003); http://www.staff.ncl.ac.uk/staff/stephen.juggins/software/C2Home.htm
320 Palynology and Its Application to Geomorphology

that are not obvious in percentage diagrams by highlighting (Traverse, 2007). Stratigraphic palynology is an important
poorly produced/dispersed pollen taxa and smoothing pollen component of biostratigraphy, which is a discipline that
taxa with high pollen production/good dispersal character- applies evolutionary and geological principles to under-
istics (Davis et al., 1973). standing sedimentary sequences and the geological record.
Generally, stratigraphic palynology is applied to pre-Quater-
nary sediments, particularly associated with the petroleum
industry, but the basic stratigraphic principles also apply to
14.26.3 Palynology and Its Applications to
Quaternary sediments. Figure 3 shows a stratigraphic profile
Geomorphology
for a number of Eocene sites from the Okanagan Highlands in
British Columbia/Washington State (Greenwood et al., 2005;
Four main areas occur, either individually or in combination,
Moss et al., 2005). The solid black lines reflect radiometric
in which pollen analysis can be applied in a geomorpho-
ages, whereas the gray lines indicate ages based on biostrati-
logical context. As noted at the outset, these include: (1) age
graphy, predominately determined from stratigraphic paly-
control; (2) environmental palynology; (3) depositional en-
nology. That is diagnostic taxa, such as Tilia, Quercus, Ilex,
vironments; and (4) human impacts/land uses.
and Platycarya in these samples denotes a post Late Paleocene
age (Manchester, 1987; Harrison et al., 1999), whereas the
presence of the diagnostic taxa Pistillipollenites macgregorii
14.26.3.1 Age Control
reflects a Upper Palaeocene to Middle Eocene age (Rouse,
The identification, distribution, and abundance of palyno- 1977; Harrison et al., 1999) and provides further chrono-
morphs have long been used to correlate sedimentary se- logical resolution. The absence of particular palynomorphs
quences or provide age control for these sedimentary can also be used to provide age control for sediments.
sequences, and are generally defined as stratigraphic palynology For instance the absence of Gothanipollis pollen from the

Geological British columbia (canada)

time scale & washington state (USA)

20.4
Oligocene

1. Republic
30
2. Allenby Fm. (princeton)
Early

Rupelian
3. Falkland
4. McAbee
33.9
35 5. Hal creek
Late

Priabonian
6. Horsefly
37.2 7. Driftwood canyon
Bartcnian
Cenozoic (in part) / plaeogene (in part)

40
40.4
Middle
Eocene

45 Lutetian Okanagan highlands

1 2 3 4 5 6 7
la

48.6
50 ? ?
A
Early

Ypresian B

55
55.8
Thanetian
58.7
Late
Paleocene

60 Selandian
61.7
Early

Danian
65
Figure 3 Stratigraphic chart showing the age and relationship of the Okanagan Highlands fossil sites. Taken from Moss, P.T., Greenwood, D.R.,
Archibald, S.B., 2005. Regional and local vegetation community dynamics of the Eocene Okanagan Highlands (British Columbia – Washington
State) from palynology. Canadian Journal of Earth Science 42, 187–204 (page 190).
 Palynology and Its Application to Geomorphology 321

Okanagan samples is also indicative of a Middle Eocene age, reconstructions are generally dependant on one or more of the
as this pollen taxon occurs in Late Eocene and Oligocene following: (1) reconstruction of ecological communities or
samples (Rouse, 1977; Long and Sweet, 1994; Ridgway et al., plant associations with their present distribution dependant
1995). on climatic parameters; (2) identification of climatic indicator
Palynological analysis can also be utilized to provide direct species; and (3) establishment of numerical transfer functions
age control through the use of accelerated mass spectrometry linking the fossil plant assemblage to climatic parameters that
(AMS) radiocarbon dating of pollen concentrate, particularly reflect modern conditions. Generally stratigraphic and en-
with samples with low organic content for bulk radiocarbon vironmental palynology can be combined to provide a very
dating, lacking large fragments of charcoal, plant macrofossils, detailed and well dated reconstruction that examines key
shell or bone for AMS dating, or providing comparison dates climatic and ecological processes.
with bulk samples or other material dated by AMS techniques For instance, Figure 4 presents the environmental re-
(Brown et al., 1989). The laboratory analysis for the prepar- construction for the Middle Eocene Okanagan region, with
ation of pollen concentrate samples for AMS radiocarbon three major communities or plant associations reflecting dif-
dating is very similar to the heavy liquid technique outlined ferent climatic and geomorphological settings (Moss et al.,
above, except that exotic marker spores are not added, acet- 2005). The Fir-Spruce association dominates higher ridges and
olysis is not undertaken (as it would remove the humic acids cooler north aspect slopes; the Birch-Golden Larch association
that are crucial for radiocarbon dating) and the samples are dominates local lakeshore, streamside and adjoining com-
sent to the radiocarbon laboratories in distilled water. At least munities; whereas the Palm-Swamp Cypress association oc-
10 mg of pollen concentrate is required for successful and cupies specialized habitats, such as low elevation areas of the
reliable radiocarbon dates (Regnéll, 1992). landscape, upland swamps, or warmer south-facing slopes.
Furthermore, a high degree of local variability on a local
scale occurs, which may reflect the influence of subregional
environmental factors (e.g., aspect, moisture availability,
14.26.3.2 Environmental Palynology
volcanic activity, or fire). In addition, different deposition
Pollen data can be used to reconstruct past climatic conditions environments may also impact pollen preservation and rep-
and ecological communities that are pertinent to the explan- resentation. For instance, there is significantly higher repre-
ation of geomorphological features or processes. These sentation of the Palm-Swamp Cypress association in coal

N
C

Salix Carpinus Corylus Itea Rhus Ulmus Sassafras

Fagus Torreya
Thuje Tsuga Picea Abies Pinus
Platanus Castanea Juglans Nyssa Populus Betula llex Zelkova Taxus

Plam Liquidambar Cornus Alnus Fraxinus Quercus Acer Ginkgo Metasequoia Taxodium Pseudolarix
Figure 4 Reconstruction of a generalized Okanagan Highland forest structure and depositional setting: (a) Palm-Swamp Cypress association
dominant in low wet areas; (b) Birch-Golden Larch association mixed with Fir-Spruce association on better drained ground and hillsides; (c)
Fir-Spruce association dominant on higher ridges and north aspects. (d) Birch-Golden Larch association dominant streamsides; (e) lacustrine
shale formations; and (f) coal formation. Taken from Moss, P.T., Greenwood, D.R., Archibald, S.B., 2005. Regional and local vegetation
community dynamics of the Eocene Okanagan Highlands (British Columbia – Washington State) from palynology. Canadian Journal of Earth
Science 42, 187–204, (page 199).
322 Palynology and Its Application to Geomorphology

swamp deposits than in similarly aged shale sediments re- qualitative and narrative estimates and should also be com-
flecting lacustrine conditions, which have a more significant pared with other palaeoclimatic proxies (e.g., Leaf Margin
representation of the Birch-Golden Larch and Fir-Spruce Analysis and transfer functions based on other fossil types,
associations. e.g., beetles or chrinomids).
Palaoeclimatic conditions for the Okanagan Highlands
have also been estimated using bioclimatic analysis or the
coexistence approach (Kershaw and Nix, 1988; Kershaw, 1996;
14.26.3.3 Depositional Environments
Mosbrugger and Utescher, 1997; Greenwood et al., 2005).
This technique requires the development of ‘climatic profiles’ Pollen data along with sedimentological results can provide
(parameters such as mean annual temperature or mean an- significant information on depositional environments, par-
nual precipitation) from the climate associated with the dis- ticularly related to sea-level alteration or other significant
tribution of modern plant genera, in this case a study of North palaeoclimatic events (e.g., the late Pleistocene/Holocene
American tree genera by Thompson et al. (1999) and the deglaciation). Figure 5 illustrates the changes in depositional
PALAEOFLORA database (PALAEOFLORA database, 2003; environments during the sea-level rise associated with the late
Mosbrugger and Utescher, 1997). The next step is to identify as Pleistocene/Holocene deglaciation for the continental shelf of
many ‘nearest living analogs’ (NLA) as possible in the fossil the northern Australian tropics (Moss, 1999). Five lithological
record (which can include both pollen and spore taxa, as well units occur in this record, ranging from a sand–clay rich re-
as plant macrofossils). A library of climatic profiles is then gressive unit, a sand rich lowstand unit, a mangrove mud early
generated for several key taxa based on key climatic values and transgressive unit, a carbonate/mangrove mud late transgres-
a zone of overlap for the set of NLAs is then developed, which sive unit to a carbonate rich highstand unit. A clear link exists
reflects the most likely climate space occupied for the indi- between this lithology and alterations in mangrove abun-
vidual fossil taxa (Kershaw and Nix, 1988; Kershaw, 1996; dances, with mangrove values increasing in the lowstand
Mosbrugger and Utescher, 1997; Greenwood et al., 2003). period (B15 000 years ago), peaking during the early trans-
This approach is generally associated with large errors but gressive period (B9000 years ago), and declining in the late
does provide a more quantitative estimate than more transgressive sediments (B7500 years ago). This sequence of

a
er
lo a n
Xy zer mo

gi
ru

a
nn us

es
m ste

io a

or
am nia

/b
i
at
So rp

ov
ph
Lu to

ps
er
ca
n

gr
p

zo
it
ce

an
er

hi
i
Av

Lithology

M
C

R
0

2500 1
Highstand

2
5000
Late transgressive
3
OIS 1
Depth (m)
Age (years BP)

Early transgressive 4
7500

5
Lowstand

6
10 000
12 500
Regression 15 000 7 OIS 2
17 500
20 000
22 500
8
20 20 20 40
Figure 5 Pollen (%) and sediment diagram of the late Pleistocene-Holocene transition from the Ocean Drilling Program core 820, humid tropics
of northeastern Australia. Adapted from Moss, P.T., 1999. Late Quaternary environments of the humid tropics of northeastern Australia. Ph.D.
thesis, Monash University, Australia, unpublished.
 Palynology and Its Application to Geomorphology 323

ii

e)

bs ed )
am

ru uc ae
ea

)
ye rris et ei

er d sh trod ore
hr nia sp h

0)
or
Be de ap enc
Po car p. tica

C te f is s ore
At so us ing

ee s (l ta
t o a m
ca
Eu ria ell la

10
sp ata

r u s e if l
.

l
sp

e s

Tr ex ae ola
an (in lif
As tag iat eru
a
ca sp. ipti

C ea ea bul
ick ad nn

ea ore
sp
m rri wr
do is s arc

s p. igu
.

e ea

x
xa
ca gin

no sp um

m3
yp er p.
.

d c
D locl cu

Po ade s la

ic ac tu
sp

on e e

Tr iopt rn

ce es
um ce c
ac p
Po tax ant

an ra ro
ea ac

ta
eu ar

M cea ea

R era lan
es ae .

er n s
R rid ae(
C rre erm

R ce sp

fe
Pl ac ae
yl us

a s

Pl us n p

du hyt
us

(c
ac odi
m pu

d
Fa ife p.
al m

Po p .
M ksia p.

c
Ap t a e

ile er
H lete
b re
Ph fag

ac ce

ia o

da a
an ea
pt

ch tics
n s

al
tro p
n i
C osp

Pi och
g

Iri tion

In ido
he a
ly

Ba cia

co
Ep era
ho

Pt s
o
o

ua
b
c
h
pt

ar
l

er
a

ig
el
ot

is
Zone

Aq
As
Ac

Le

Tr
N

H
0
10
20
30
DP2B
40
Depth (cm)

50
60
70
80
90 DP2A

100
110
120
20 20 20 40 20 20 40 20 40 60 80 100 200 400 600

Figure 6 Den Plain pollen diagram. The þ symbol reflects taxa that occur in low values (less than 2.0%) and the dotted line marks the
presence of exotic pollen. Adapted from Moss, P.T., Thomas, I., Macphail, M., 2007. Late Holocene vegetation and environments of the Mersey
Valley, Tasmania. Australian Journal of Botany 55, 74–82.

alterations in mangrove pollen and sedimentology correlates associated with European settlers in Australia. Further, the
well with the late Pleistocene/Holocene sea level rise, but also sedimentological results for the record also reflect profound
illustrates the effects of regional geomorphology. For instance, impacts associated with European settlement. A significant
there is a limited scope for mangrove growth during the last increase in inorganic content from 85 cm occurs in the sedi-
glacial maximum due to lower sea-levels (  120 m) being ment record that best reflects land-clearance and pastoral ac-
located at the continental shelf and forming a very steep tivities within the Mersey River catchment (Den Plain is
coastline; mangrove growth peaking during the early trans- located on the banks of this river). Below 85 cm the sediment
gressive, when the coastline is protected by the outer Great records consist primarily of organic-rich peats, with little in-
Barrier Reef, gently sloping (i.e., flat drowning continental organic material and the base of the record (at B120 cm) has
shelf) and increased mud sediment (due to increased pre- been dated to B3000 years before from now, suggesting that
cipitation); and limited mangrove growth during the late there has been a 14.7 times increase in sedimentation rates
transgressive/highstand period due to steep coastline associ- within the region since European settlement. Although evi-
ated with coastal mountain ranges, coastal progradation and dence of agricultural activities is relatively clear in pollen data
the distance of the site from mainland mangrove communities and associated sedimentary records, the evidence for the ef-
(Grindrod et al., 1999, 2002). fects of relatively low-impact hunter gatherers is not as dis-
cernable. Generally increases in carbonized particles in pollen
records have been used as an indicator for human arrival into
14.26.3.4 Human Impact a region, however, these increases may also be associated with
climate alterations and it can be somewhat difficult to dis-
Pollen records can also provide detailed information on al- entangle the relative influences of human impact from natural
terations in plant communities and land-use changes associ- climatic alterations.
ated with human activities. Changes in fire management,
development of agriculture, introduction of exotic taxa, for-
estry practices, urbanization, and land degradation can be
observed in pollen records and associated sedimentary records 14.26.4 Conclusion
across the globe. Figure 6 shows an example of the impacts of
European settlement on a small plain in north-western Tas- Palynological analysis can be applied to a wide range of geo-
mania, Australia (Moss et al., 2007). The occurrence of exotic morphological contexts and as outlined above it can provide a
pollen grains at 60 cm has been used to date the record to the range of data and information that greatly improves under-
1860s, when European agricultural activity was first practiced standing of geomorphological processes or of a particular
in the region. The key impacts associated with this activity landform. This includes the provision of a relative chrono-
was significant land clearance associated with a decline in logy through stratigraphic palynology, absolute age control
forest taxa and an increase in grasses, reflecting grazing as through AMS dating of pollen concentrate, palaeoclimatolo-
the dominant land-use within the plain. In addition, a sig- gical, or palaeoecological data, information on depositional
nificant alteration in carbonized particle representation environments and the impacts of people/human land use.
occurs that may reflect the onset of fire suppression as a sig- Therefore, it is crucial for geomorphologists to have an
nificant management technique. Something that is generally understanding of pollen analysis techniques, whether it is
324 Palynology and Its Application to Geomorphology

basic knowledge of how to interpret a pollen diagram of their Grindrod, J., Moss, P., van der Kaars, S., 1999. Late Quaternary cycles of
study area to a more advanced understanding of pollen mangrove growth and decline on the north Australian continental shelf. Journal
of Quaternary Science 14, 465–470.
identification, counting, and laboratory techniques.
Grindrod, J., Moss, P.T., van der Kaars, S., 2002. Late Quaternary mangrove pollen
records from the continental shelf and deep ocean cores in the north Australian
region. Bridging Wallace’s Line – The Environmental and Cultural History and
Dynamics of the SE-Asian-Australian Region. In: Kershaw, A.P., David, B.,
14.26.5 Use of Exotic Markers
Tapper, N., Penny, D., Brown, J. (Eds.), Advances in Geoecology 34. Cantena
Verl, Reiskirchen, Germany, pp. 119–148.
Exotic markers are used to calculate pollen and charcoal Harrison, J.C., Mayr, U., McNeil, D.H., et al., 1999. Correlation of Cenozoic
concentrations, that is, the number of pollen grains (or char- sequences of the Canadian Arctic region and Greenland: implications for the
coal particles) per cm3 of sediment. Concentration values tectonic history of northern North America. Bulletin of Canadian Petroleum
Geology 47, 224–254.
may change through the course of a pollen or charcoal Juggins, S., 2003. C2 Data Analysis. University of Newcastle, Newcastle, U.K..
record and may provide information on environmental Kershaw, A.P., 1996. A biolclimatic analysis of Early to Middle Miocene brown coal
changes (e.g., higher pollen concentrations during warmer floras, Latrobe Valley, southeastern Australia. Australian Journal of Botany 45,
and wetter climates) or sedimentological alterations (e.g., 373–383.
Kershaw, A.P., Nix, H.A., 1988. Quantitative palaeoclimatic estimates from pollen
changes in depositional rates altering pollen concentrations).
data using bioclimatic profiles of extant taxa. Journal of Biogeography 15,
In addition, alterations in charcoal concentrations generally 589–602.
reflect changes in fire regimes that may reflect broader alter- Long, D.G.F., Sweet, A.R., 1994. Age and depositional history of the Rock River
ations in the environment (i.e., climate change or anthropo- coal basin, Yukon Territory, Canada. Canadian Journal of Earth Science 31,
genic activity). Exotic markers can be pollen grains that do not 865–880.
Mackenzie, L.L., Moss, P.T., Petherick, L., 2009. Mid to late Holocene vegetation
occur in the study region (commonly Lycopodium sp. or Eu-
and environments of Lake Selina Swamp, western Tasmania. Quaternary of
calyptus sp.) or plastic microspheres that are of similar size to Australia 26, 2–7.
pollen grains (B15 to 20 mm). The markers (with a known Macphail, M.K., 1980. Fossil and modern pollen of Beilschmiedia (Lauraceae) in
concentration) are added during the laboratory analysis and New Zealand. New Zealand Journal of Botany 18, 149–152.
then counted on the slide as the microscope analysis occurs to Maher, Jr. L.J., 1972. Absolute pollen diagram of Redrock Lake, Boulder County,
Colorado. Quaternary Research 2, 531–553.
determine the pollen or charcoal concentration of the sedi- Manchester, S.R., 1987. The fossil history of Jungandaceae. Monograph
ment sample. in Systematic Botany. Missouri Botanical Gardens, St Louis, MO,
1–137.
Mosbrugger, V., Utescher, T., 1997. The coexistence approach – a method
for quantitative reconstructions of Tertiary terrestrial palaeoclimate data
References using plant fossils. Palaeogeography, Palaeoclimatology, Palaeoecology 134,
61–86.
Moss, P.T., 1999. Late Quaternary environments of the humid tropics of
Assarson, Goch E., Granlund, E., 1924. En Metod for Pollenanalys av northeastern Australia. Ph.D. thesis, Monash University, Australia,
Minerogena Jordarter. Geologiska Föreningens i Stockholm Förhandlingar 46, unpublished.
76–82. Moss, P.T., Greenwood, D.R., Archibald, S.B., 2005. Regional and local vegetation
Brown, T.A., Nelson, D.E., Mathewes, R.W., Vogel, J.S., Southon, J.R., 1989. community dynamics of the Eocene Okanagan Highlands (British Columbia –
Radiocarbon dating of pollen by accelerator mass spectrometry. Quaternary Washington State) from palynology. Canadian Journal of Earth Science 42,
Research 32, 205–212. 187–204.
Charman, D.J., 1992. The effects of acetylation on fossil Pinus pollen and Moss, P.T., Thomas, I., Macphail, M., 2007. Late Holocene vegetation and
Sphagnum spores discovered during routine pollen analysis. Review of environments of the Mersey Valley, Tasmania. Australian Journal of Botany 55,
Palaeobotany and Palynology 72, 159–164. 74–82.
Clark, R.L., 1982. Point count estimation of charcoal in pollen preparations and Nilsson, S., Praglowski, J., 1992. Erdtman’s Handbook of Palynology, Second ed.
thin sections of sediments. Pollen et Spores XXIV, 523–535. Munksgaard, Copenhagen.
Coil, J., Korstanje, M.A., Archer, S., Hastorf, C.A., 2003. Laboratory goals and PALAEOFLORA database. 2003. Paleoclimate reconstructions using tertiary plant
considerations for multiple microfossil extraction in archaeology. Journal of taxa. The Palaeobotanical Workgroup of Volker Mosbrugger at Tübingen
Archaeological Science 30, 991–1008. University. Available from http://www.geologie.uni-bonn.de/Palaeoflora/
Davis, M.B., Brubaker, L.B., Webb, T., 1973. Calibration of absolute pollen influx. Palaeoflora_home.htm
In: Birks, H.J.B., West, R.G. (Eds.), Quaternary Plant Ecology. Blackwell Pohl, F., 1937. Die pollenerzeugung der windblütler. Beiheftezum Botanischen
Centralblatt 56, 365–470.
Scientific Publications, Oxford, pp. 9–25.
Regnéll, J., 1992. Preparing pollen concentrates for AMS dating – a
Erdtman, G., 1934. Über die verwendung von essigsaureanhydrid bei
methodological study from a hard-water lake in southern Sweden. Boreas 21,
pollenuntersuchungen. Svensk Botanisk Tidskrift 41, 104–114.
373–377.
Faegri, K., Iversen, J., 1991. Textbook of Pollen Analysis. Wiley, Chichester.
Ridgway, K.D., Sweet, A.R., Cameron, A.R., 1995. Climatically induced floristic
Greenwood, D.R., Archibald, S.B., Mathewes, R.W., Moss, P.T., 2005. Fossil biotas
changes across the Eocene-Oligocene transition in the northern high l
from the Okanagan Highlands, southern British Columbia and northeastern
atitudes, Yukon Territory, Canada. Geological Society of America Bulletin 107,
Washington State: climates and ecosystems across an Eocene landscape.
676–696.
Canadian Journal of Earth Science 42, 167–185. Riding, J.B., Kyffin-Hughes, J.E., 2004. A review of laboratory preparation of
Greenwood, D.R., Moss, P.T., Rowett, A.I., Vadala, A.J., Keefe, R.L., 2003. Plant palynomorphs with a description of an effective nonacid technique. Review of
communities and climate change in southeastern Australian during the early Brasilian Paleontology 7, 13–44.
Paleogene. Causes and Consequences of Globally Warm Climates in the Early Riding, J.B., Kyffin-Hughes, J.E., 2006. Further testing of a non-acid palynological
Palaeogene. In: Wing, S.L., Gingerich, P.D., Schmitz, B., Thomas, E. (Eds.), preparation technique. Palynology 30, 69–87.
Geological Society of America Special, Paper 369, pp. 365–380. Roblin, I., Urban, M., Flicoteau, , Martin, C., Pradeau, D., 2006. Topical treatment
Grimm, E.C., 1987. CONISS: A FORTRAN 77 program for stratigraphically of experimental hydrofluoric acid skin burns by 2.5% calcium gluconate.
constrained cluster analysis by the method of incremental sum of squares. Journal of Burn Care & Research 27, 889–894.
Computational Geology 13, 13–35. Rouse, G.E., 1977. Paleogene palynomorph ranges in western and northern Canada.
Grimm, E.C. (2004). TGView Version 2.0.2. Illinois State Museum, Springfield, IL, Contributions of stratigraphic palynology. (Ed.) W.D. Elsik. Cenozoic Palynology
USA. 1, 48–65.
 Palynology and Its Application to Geomorphology 325

Salgado-Labouriau, M.L., Rull, V., 1986. A method of introducing exotic pollen for Traverse, A.T., 2007. Paleopalynology (Topics in Geobiology 28), second ed.
paleoecological analysis of sediments. Review of Palaeobotany and Palynology Springer, London.
47, 97–103. van der Kaars, W.A., 1991. Palynology of eastern Indonesian marine piston-cores: a
Seppä, H., 2007. Pollen analysis, principles. In: Elias, S. (Ed.), Encyclopedia late quaternary vegetational and climatic record for Australasia. Palaeogeography,
of Quaternary Science. Elsevier Academic, New York, vol. 3, pp. 2486–2497. Palaeoclimatology, Palaeoecology 85, 239–302.
Thompson, R.S., Anderson, K.H. and Bartlein, P.J. 1999. Atlas of relations between Wang, X., van der Kaars, S., Kershaw, A.P., Bird, M., Jansen, F., 1999. A record of
climatic parameters and distribuitions of important trees and chrubs in North fire, vegetation and climate through the last three glacial cycles from Lombok
America. US Geol. Survey, Prof. Paper 1650 A&B. Online version 1.0, 14 Ridge core G6-4, eastern Indian Ocean, Indonesia. Palaeogeography,
December 1999. Palaeoclimatology, Palaeoecology 147, 241–256.

Biographical Sketch
Patrick Moss is a Senior Lecturer in Physical Geography at The University of Queensland, Brisbane, Australia. His
research interests are in the areas of Quaternary environmental change (i.e., last 2 million years) in eastern
Australia and the Eocene greenhouse environments (i.e., 50 million years ago) of British Columbia, Canada. He
gained his PhD from Monash University and has a BA (Hons.) and BSc from The University of Melbourne. He
teaches in the areas of physical geography, palaeoecology, biogeography, and landscape ecology.
 14.27 Investigating the Strength of Materials: Introduction
DM Kennedy, The University of Melbourne, Parkville, VIC, Australia
r 2013 Elsevier Inc. All rights reserved.

The strength of a material (such as soil or a rock face) can be jointing of the rock surface which is the major driver of
described as its maximum resistance to an applied stress landform evolution (e.g., Shag Point, New Zealand (Kennedy
(Whittow, 1984). When a landform responds to stress, such as and Dickson, 2006) and Sydney Harbor, Australia (Kennedy,
a landslide on a hillslope, its point of failure, generally termed 2010)). In the case of cliffs, it is the combination of a rock’s
a shear plane, is related to the combination of vertical compressive strength and internal discontinuities, which de-
stress (due to gravity) and normal stress, which acts at right termine their maximum height (e.g., Sunamura, 1992).
angles to the shear plane (Crozier, 1986; Summerfield, 1991). In this section, two chapters are presented that explore
Counteracting this force are two main processes of resistance, specific aspects of measuring material strength. Petley and
cohesion and friction (Whittow, 1984). It is the balance of Clarke start with an investigation of the concept of stress and
these forces that will determine the stability of a landscape and its place in geomorphology (see Chapter 14.28). Their focus is
its resistance to perturbations in the natural environment, on direct shear testing methodologies and the behaviors that
such as a storm event. Measurement of material strength is materials may exhibit when stress is applied, such as brittle,
therefore a fundamental aspect of geomorphology and it is in ductile, and strain hardening behavior. A new type of machine
this section of the Methods Volume that some common is described, the back-pressured shear box, which improves
methods of measuring strength are investigated. the design of traditional shear boxes and importantly explores
Geomorphic systems typically respond to change in a how pore water pressure can influence shear box readings.
nonlinear fashion, most commonly as a stepped threshold The second chapter in this section focuses on a technique
response (Phillips, 2006). These thresholds for change are that is used to directly measure the hardness of a rock surface
commonly of two types, the ratio of force to resistance and the (see Chapter 14.29). This technique is the Schmidt Hammer,
relative rates of linked processes (e.g., glacial accumulation vs. which measures the rebound of a piston on the rock surface,
ablation). Shear strength versus shear stress in hillslopes is a which is directly related to the rock’s compressive strength and
classic example of the relation between force and resistance hence its hardness. The measurement of hardness is a critical
(Phillips, 2006). parameter when calculating the response of a landform to
The measurement of strength in the landscape applies weathering and erosion processes, but it must be interpreted
equally to hard rock materials as it does to unconsolidated in regard to the discontinuities within the particular landform.
sediments. Naylor and Stephenson (2010) used the term geo- Although the first chapter is focused on unconsolidated
logical contingency to define the inherent geological properties sediments and the second on hard rock, the principles of both
of a landscape that control the nature of landform evolution. measurements are applicable to all landscapes. That is, the
In particular, they focused on the role of discontinuities in rock reaction and relaxation time of a landform will depend on its
platform surfaces as key in determining rock resistance to ability to resist an external driving force and the time it takes
erosion and therefore the evolutionary pathways of platform to adjust back to its original condition or to adjust to a new
development. This is an important consideration in geo- equilibrium.
morphology as the strength of a material, in this case bedrock,
will generally relate to many variables such as the jointing,
foliation, and bedding angle, as well as the hardness of the References
rock material. In hillslope studies and engineering geology,
the impact of these variables on rock strength is critical when Crozier, M.J., 1986. Landslides: Causes, Consequences & Environment. Croom
Helm, London.
developing hazard indices for mountainous areas or road Hoek, E., Brown, E.T., 1997. Practical estimates of rock mass strength. International
cuttings. The combined effect of these geological discontinu- Journal of Rock Mechanics and Mining Sciences 34, 1165–1186.
ities is commonly assessed using the rock mass strength index Kennedy, D.M., 2010. Geological control on the morphology of estuarine shore
(e.g., Hoek and Brown, 1997). The coast of Wellington, platforms: Middle Harbour, Sydney, Australia. Geomorphology 114, 71–77.
Kennedy, D.M., Beban, J.G., 2005. Shore platform morphology on a rapidly
New Zealand, is also an example on how the strength of ma-
uplifting coast, Wellington, New Zealand. Earth Surface Processes and
terial is not a straightforward variable to measure. The plat- Landforms 30, 823–832.
forms are eroding at 150 mm yr 1, yet are formed in highly Kennedy, D.M., Dickson, M.E., 2006. Lithological control on the elevation of shore
resistant graywacke sandstones and mudstones. This bedrock platforms in a microtidal setting. Earth Surface Processes and Landforms 31,
is, however, highly fractured, and it is the intersections of these 1575–1584.
Naylor, L.A., Stephenson, W.J., 2010. On the role of discontinuities in mediating
fractures that allow the waves to easily pluck blocks from the shore platform erosion. Geomorphology 114, 89–100.
surface and therefore erode the platforms (Kennedy and Phillips, J.D., 2006. Evolutionary geomorphology: thresholds and nonlinearity in
Beban, 2005). In fact, in many shore platform areas it is the landform response to environmental change. Hydrology and Earth System
Science 10, 731–742.
Summerfield, M.A., 1991. Global Geomorphology. Longman, Harlow.
Kennedy, D.M., 2013. Investigating the strength of materials: introduction. Sunamura, T., 1992. Geomorphology of Rocky Coasts. John Wiley & Sons,
In: Shroder, J. (Editor in Cheif), Switzer, A.D., Kennedy, D.M. (Eds.), Chichester.
Treatise on Geomorphology. Academic Press, San Diego, CA, vol. 14, Whittow, J., 1984. Dictionary of Physical Geography. Penguin, London.
Methods in Geomorphology, pp. 326–327.

326 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00396-1

 Investigating the Strength of Materials: Introduction 327

Biographical Sketch

David M. Kennedy is a geomorphologist who specializes in the response coastal systems to climatic and en-
vironmental change. He particularly focuses on the Holocene, researching estuarine, coral reef, and rocky coast
landforms. He completed his doctorial study at the University of Wollongong, Australia under the guidance of
Professor Colin Woodroffe following undergraduate studies at the University of Sydney, Australia. After a post-
doctoral research position at Wollongong he was based at Victoria University of Wellington, New Zealand. He is
currently a senior lecturer in the Department of Resource Management and Geography at The University of
Melbourne, Australia.
 14.28 Direct Shear Testing in Geomorphology
DN Petley and AL Clark, University of Durham, Durham, UK
r 2013 Elsevier Inc. All rights reserved.

14.28.1 Introduction 328

14.28.2 The Importance of Shear Strength in Geomorphology 328
14.28.3 Direct Shear Testing in Geomorphology 328
14.28.4 Data Analysis 330
14.28.5 Strengths and Weaknesses of Direct Shear Tests in Geomorphology 331
14.28.6 The Principles of the Back-Pressured Shearbox 332
14.28.7 Direct Shear Testing of Fine Sand 333
14.28.8 Discussion 335
14.28.9 Conclusions 336
References 337

Abstract

Although being a well-established technique within geotechnical engineering, and being admirably simple to use,
laboratory-based direct shear testing has found surprisingly limited application within geomorphology. The reasons for this
are not simple, revolving mainly around the lack of insight that the technique provides into mechanisms of deformation,
beyond simple classification (i.e., brittle or ductile). However, recent advances in control systems and in software provide
the potential to create a new generation of direct shear devices that genuinely recreate the conditions within, for example,
deforming slopes, allowing a proper understanding of process to be established. Such devices potentially represent an
important advance.

14.28.1 Introduction at the bed of a river, where the water flow exerts a force on
the bed; within a glacier, where deformation of the ice
Concepts of stress and strain are greatly undervalued in the and the basal till allows movement; or at the bed of a land-
consideration of geomorphic systems. In consequence, there is slide, where movement occurs because of the application of a
also a dearth of work undertaken using direct shear testing, shear force. It is the shear strength of the materials that form
which has been reflected in a remarkable lack of development the system that resist the deformation, and thus allow or
of this technique until recently. This chapter starts with a brief prevent movement or change. In this context, the lack of
explanation as to why an understanding of the shear strength consideration of shear strength within studies of geomorphic
properties of materials is important for geomorphology. This systems is remarkable. Indeed, a Web of Science topic search
is then followed by an examination of the use of laboratory- for the phrase ‘‘shear strength’’ in the titles and abstracts of all
based direct shear testing for understanding material behavior, papers in two key geomorphic journals (Earth Surface Processes
and hence the ways in which geomorphic systems operate, and Landforms and Geomorphology) yielded only 39 papers,
followed finally by a consideration of new developments in with a maximum of six in a single year mentioning this
direct shear testing approaches, and in particular the recent phrase (Figure 1(a)), in the period between 1991 and October
innovation of the back-pressured shear box. 2010. During this time a total of 5787 papers were published
(Figure 1(b)).
Even where material deformation is considered in geo-
14.28.2 The Importance of Shear Strength in
morphic papers, the concept tends to be treated in a some-
Geomorphology
what simplistic manner. In general, shear strength is
considered to be a static parameter (or pair of parameters),
Material deformation is a fundamental control on the be-
essentially a simple threshold value that determines whether
havior of all geomorphic systems, no matter where they occur.
permanent strain (deformation) can occur. Only a few studies
Change in a geomorphic system occurs because a force is
consider the rather more complex nature of shear strength and
imposed on some element of the system that is sufficiently
of material deformation (e.g., Petley and Allison, 1997;
large that the capacity for the system to prevent permanent
Wilson et al., 2003; Cristescu et al., 2002; Corominas et al.,
deformation is exceeded. This applies in all cases, whether it is
2005). Thus, although the first part of this chapter considers
the conventional approach to shear strength measurement,
Petley, D.N., Clark, A.L., 2013. Direct shear testing in geomorphology. In:
the second part introduces the new generation of equipment
Shroder, J. (Editor in Chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise
on Geomorphology. Academic Press, San Diego, CA, vol. 14, Methods in that is allowing the role of material deformation in geo-
Geomorphology, pp. 328–337. morphic systems to be considered properly.

328 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00397-3

 Direct Shear Testing in Geomorphology 329

Number of papers
4

0
90
91
92
93
94
95
96
97
98
99
00
01
02
03
04
05
06
07
08
09
19
19
19
19
19
19
19
19
19
19
20
20
20
20
20
20
20
20
20
20
(a) Year

600

500
Number of papers

400

300

200

100

0
90
91
92
93
94
95
96
97
98
99
00
01
02
03
04
05
06
07
08
09
19
19
19
19
19
19
19
19
19
19
20
20
20
20
20
20
20
20
20
20

(b) Year
Figure 1 Bar graphs showing the number of mentions of the phrase ‘‘shear strength’’ in the title and abstract of two key geomorphic journals
(Earth Surface Processes and Landforms and Geomorphology) through time. (a) The number of mentions of the phrase each year; (b) the total
number of papers published each year.

14.28.3 Direct Shear Testing in Geomorphology via a calibrated linear proving-ring, allowing the strength of the
material to be determined.
Shear strength testing of earth materials is usually undertaken This simple approach for measuring shear strength
using the direct shear method, using the shear box apparatus was developed by the pioneering geotechnical engineer
(BSI, 2005). In essence this equipment is very simple (Figure 2) Arthur Casagrande during the period 1929–1932, but is
and comparatively inexpensive. A small sample (typically generally considered to have reached full maturity with the
100 100 mm, and 25 mm thick), is placed in a frame that is work of Skempton and Petley (1967). The basic Casagrande
split along the horizontal axis. One half of the frame is pre- design continues to form the basis of the instrument used in
vented from moving by being locked to the machine frame, soils labs around the world. In most instruments the normal
whereas the other can be displaced on low friction bearings via load is applied via weights, which permit a very reliable ap-
the application of a shear force applied by a simple electric plication of a constant load, usually via a lever system to
motor, or indeed through manual means, operating through a produce the desired stress level (Figure 2). The shear force is
gearbox. The sample is loaded with a pre-determined ‘‘normal’’ applied using a simple electric motor, generally driven via a
(vertical) load, whereon the sample is allowed to compact gearbox to allow the requisite rate of displacement to be
(consolidate). When this process is complete the mobile part of generated. The shear force is measured via a simple
the spilt frame is displaced at a constant rate. The force required linear proving ring located between the motor and the sample
to cause this constant rate of movement is measured, generally frame.
330 Direct Shear Testing in Geomorphology

Measurement
of vertical
displacement

Measurement
of horizontal
displacement

Shear force

Water

Sample
Figure 2 A schematic cross-section of a conventional direct shear (shear box) apparatus.

A key decision in undertaking a shear box text focuses on stress state of the system of interest. For example, if a landslide
whether to undertake the test in a drained or an undrained was 10 m thick, the typical total normal stress would be ap-
state. Unlike the more sophisticated triaxial cell, which pro- proximately 200 kPa, and the normal effective stress in the
vides for the application and monitoring of back pressure, range of 200 kPa (for a completely dry slope)–100 kPa (if
conventional direct shear experiments do not provide a means completely saturated). Thus, the normal effective stresses se-
to measure or control pore water pressures. Thus, the experi- lected might typically be in the range of 50–300 kPa to bracket
mentalist has no knowledge of whether pore water pressures the ground conditions appropriately.
have developed within the sample. As pore water pressures
serve to reduce the normal effective stress (Craig, 2004), and
thus the frictional resistance of the sample, the shear strength
14.28.4 Data Analysis
measured when excess pore water pressures are allowed to
develop will generally be lower than when their development
For each direct shear test, two sets of data are generated. First,
is prevented (Craig, 2004). Thus, in most cases research will be
the application of the normal load before the shear phase
undertaken using tests that are designed to prevent excess pore
induces consolidation (time-dependent compaction) of the
water pressure development. In practice this means applying a
sample, resulting from the generation of excess pore water
displacement rate that is sufficiently slow that any developed
pressure in the sample, which then drives fluid flow and hence
excess pore water pressures can dissipate; although this state
volume loss. In terms of measurements, a consolidation curve
has to be assumed given the lack of measurement. This has the
is generated, which is generally expressed as a graph of vertical
disadvantage of a longer experiment length, but the advantage
displacement or vertical strain against time (Figure 3).
of more reliable results. In commercial testing, where the re-
Although consolidation is very important in many engineering
sults are generally used for design purposes, rapid (undrained)
settings (for example, determining foundation behavior where
tests are commonly used (Head, 1986). This will generate a
loaded by a building), its relevance for geomorphological
lower strength value for the design, effectively providing an
settings is generally limited. However, consolidation data can
additional margin of error in the design calculations. How-
be used to determine the permeability of the material, given
ever, for research purposes such a test introduces a margin of
that the length of the drainage path is known. This provides
uncertainty that should generally be considered to be un-
information that can be useful for examining potential flow
acceptable. Thus, research-led experiments using the standard
paths of water through landslide systems, for example.
direct shear apparatus should be undertaken at sufficiently
Of greater importance is the direct shear dataset, which
slow strain rates to avoid pore water pressure development.
can be used to determine the strength of the material under
This will depend on the material in question – typically for
test. For each test the direct shear test results are generally
loose sand, which has a high permeability, a displacement rate
presented in two key graphs. The first is a graph of shear
of 0.1 mm/minute will be acceptable, whereas a low per-
stress against shear strain (Figure 4), which provides a
meability pure clay may need a rate as low as 0.001 mm/
mechanism for interpreting the style of behavior that the
minute, or even less.
sample displays. Generally speaking three styles of behavior
At a minimum, three tests need to be undertaken, each at a
occur (Figure 4):
different normal stress value. However, given the variability of
natural materials, four tests should be considered to be the 1. Brittle behavior, in which the shear stress reaches a peak
minimum, and ideally five or more will need to be performed and then rapidly declines, usually to a constant ‘‘residual’’
for any material in its natural (i.e., original and intact, and value. This style of behavior characteristically occurs in first
thus not remolded) state. The normal loads chosen should be time failures in landslides that are formed from a material
designed to provide a set of normal stresses that bracket the that displays cohesion, such as cemented sandstone.
 Direct Shear Testing in Geomorphology 331

Vertical displacement or strain

Shear stress
Friction angle

Time Cohesion
Figure 3 The normal shape of a conventional consolidation curve.
Normal stress
Figure 5 Plotting the derived shear stress against normal stress for
three or more tests allows the failure envelope to be determined. The
cohesion is the intercept of the best fit line on the x-axis and the
Brittle behavior angle of internal friction is given by the slope of the best fit line.

two sets of values can be determined – the peak value and that
Shear stress

in the residual strength state (points A and B on Figure 6). For

a strain hardening material, it is more difficult to determine
the correct value, but generally the shear stress at a set (large)
Strain hardening shear strain is used – typically, for example, 10% strain.
behavior

Ductile behavior
14.28.5 Strengths and Weaknesses of Direct Shear
Tests in Geomorphology

Shear strain The direct shear test allows determination of the shear strength
(cohesion and angle of internal friction) for the material
Figure 4 Typical stress strain curves for brittle, ductile and strain
in question, and in addition can provide an indication of the
hardening materials.
style of deformation that the material displays. As such it is a
very useful test, which has the advantage of shear stress appli-
2. Ductile behavior, in which the shear stress increases to a cation in a manner that, superficially at least, replicates
peak value that is then maintained through time. This style the conditions in the basal region of a landslide. However,
of behavior generally occurs in materials that have no co- the conventional direct shear test suffers from a number of
hesion, such as an uncemented sand. weaknesses:
3. Strain hardening behavior, in which shear stress continues to
increase as displacement increases. Such behavior occurs in a. The application of shear stress is undertaken by applying a
some geological materials, such as rocksalt and greywacke. constant rate of displacement to the sample carriage. This is not
representative of the ways in which the real systems operate.
In addition, a plot is commonly generated of the vertical strain b. The surface area of the shear zone reduces through time.
(sometimes expressed as a volume strain) against shear stress, Although the surface area can be determined at any point
as this allows interpretation of the material behavior in terms during the test using the displacement data, in a system in
of dilation or compaction. This provides further information which the normal stress is applied via a dead load (i.e. hanging
about the mechanism of failure in the material, although this weights) the normal stress in effect increases with time.
information is surprisingly underused in most analyses. c. No mechanism exists for determining the pore pressure
However, in a conventional direct shear experiment the condition of the sample.
critical information is obtained from a plot for each test of d. The direct shear test predetermines the location of the shear
maximum shear stress against normal stress. For a ductile plane, which is unlikely to be in the same location as would
material, this is generally achieved by plotting the peak shear occur under natural conditions. Thus, the test might be ex-
stress against the normal stress, with the resulting points being pected to slightly over-estimate the true strength of material.
joined with a straight line (Figure 5). The intercept with the e. From a more technical perspective, the direct shear test also
y-axis provides the cohesion value, whereas the slope of the generates a non-uniform shear stress distribution, which
line gives the angle of internal friction. For a brittle material, might change the material behavior.
332 Direct Shear Testing in Geomorphology

2004 for a discussion of these problems). The basic design of

the apparatus is as per the conventional shearbox wherein the
A sample is located in horizontally split holder, to which a
horizontal shear force and a normal load can be applied.
However, the additional capabilities of the machine are as
follows (Figure 7):
Shear stress

1. The normal load is applied by way of a ram equipped

with a load cell, with the force being applied via a
piston connected to a pressure controller, which in turn is
B
controlled by advanced software. This allows normal loads
to be applied rapidly, as in a conventional direct shear
machine, or to be ramped at any desired rate. It is even
possible to apply a dynamic normal load if required, al-
though a high frequency cycle would need further modi-
fication of the existing equipment. As second generation
machine, which facilitates dynamic loading, is currently
Normal stress
under construction at the University of Durham, U.K.
2. The shear stress is applied via a stepper motor through
a load cell and monitored by a displacement transducer
(LDVT), controlled by the same software as the one
above. This permits a constant displacement rate to be
A
Peak strength applied, as in a conventional machine, or for the shear
envelope stress to be ramped at any desired rate. It also allows
Shear stress

shear stress to be held at a constant value or even to be

cycled.
3. The sample is located in a water bath located within a
pressure vessel. The water bath has two outlet pipes via
B valves, one of which is connected to a pore pressure
transducer. This permits the pore pressure within the water
Residual strength
envelope bath, and thus the sample, to be monitored. The second is
connected to a pressure controller, also connected to the
software system, which allows pore pressure to be con-
trolled if required.
Normal stress
Figure 6 Where a material shows both peak (A on the graph) and This machine represents a substantial advance over con-
residual strength (B), two failure envelopes can be defined. ventional machines by permitting control of shear stress and
normal stress, and by allowing monitoring and control of
f. And finally, during the test, rotation of the principal sample pore water pressure. The software can automatically
stresses occurs, which means that the test set up does not calculate shear displacement, and thus the changing sample
simulate the conditions in the ground as well as is ideal. area, allowing the normal load to be altered in order to
Although most of these problems do not inherently change maintain a constant normal stress on the sample. Perhaps
the observed behavior of the material, a clear need exists to most importantly, as well as being usable as a conventional
improve on the apparatus in order to overcome at least some direct shear machine, the BPSB can be used to replicate the
of these issues. In particular, the inability to control or stresses acting on a landslide shear surface by applying a
measure pore pressures is a serious shortcoming, and the in- representative total normal stress, a representative pore water
ability to apply shear stress in a realistic manner is also pressure and a representative shear stress to replicate the
problematic. Thus, there has been a need for research in conditions within the slope induced by the inclination. Thus,
geomorphology to develop a new, more sophisticated version the shear stresses acting on a slope can be determined from
of the shear box. This instrument is the back-pressured shear the geometry of the system (i.e., the depth and inclination of
box; the remainder of this chapter describes this instrument. the shear surface) and the properties of the slope materials,
and can be replicated in the BPSB. The pore water pressure
conditions can then be varied, perhaps even replicating
measurements collected from piezometers located within the
14.28.6 The Principles of the Back-Pressured landslide. The machine can then be used to examine behavior
Shearbox under a range of scenarios, such as, for example, unusually
high pore water pressure conditions. Thus the BPSB can be
The back-pressured shearbox (BPSB) is a new type of direct used to examine the likely future behavior of slopes in a way
shear machine designed to overcome the major problems as- that other apparatus, and existing techniques (see Craig, 2004)
sociated with conventional direct shear apparatus (see Craig, cannot.
 Direct Shear Testing in Geomorphology 333

Load cell
Normal load
ram

Axial
Load cell load ram
Sample

To pore
pressure
controller

Load frame

To
Sample
axial load
carriage
contoller

Figure 7 A schematic cross-section of a conventional beck-pressured shear box apparatus.

Table 1 Experimental details of the seven experiments described in this chapter

Experiment Type of test Initial Consolidation stress (kPa) Shear stress (kPa) PPR rate (kPa min1)

SSC1 Conventional 100

SSC2 Conventional 200
SSC4 Conventional 400
SSC8 Conventional 800
PPR2 Pore pressure reinflation 400 470 0.0015
PPR3 Pore pressure reinflation 400 470 0.015
PPR4 Pore pressure reinflation 400 470 0.15

Of course the BPSB does not overcome all of the short- of direct shear test results from this new testing system, a total
comings of the conventional direct shear machine, most not- of seven experiments have been undertaken, consisting of four
ably the problems of rotation of the principal stresses and the direct shear and three landslide simulation experiments
forcing of the development of the shear surface at a particular (Table 1). The latter represent the so-called pore-pressure
location in the sample, but it represents a substantial advance reinflation tests in which a constant total normal and shear
over conventional machines. In the next section, an example stress are applied to the sample, and the pore pressure is in-
of the use of the BPSB is provided. creased, reducing the normal effective stress to induce hori-
zontal strain and hence failure (Figure 8).
As the consolidation data are beyond the scope of this
14.28.7 Direct Shear Testing of Fine Sand chapter, and are of limited geomorphic interest, they are not
considered here. The conventional shear experiments generate
To provide an example of the application of the BPSB, testing stress–strain curves that are as expected (Figure 9), showing an
has been undertaken on a uniformly graded fine sand. The increase in shear stress with shear strain, eventually settling to
BPSB has been used to undertake both conventional and a constant residual strength value. The vertical strain indicates
landslide simulation experiments. Thus, in the experimental that the three samples consolidated to 200 kPa or greater
program reported here, which represents the first presentation displayed compaction during the initial phases of shear
334 Direct Shear Testing in Geomorphology

deformation, and dilated thereafter (Figure 10). The most In Figure 12 the plot of shear stress against shear strain for
compacted sample (i.e., sample subject to the highest mean the conventional shear component of the three experiments,
normal effective stress) shows the lowest level of compaction and the 400 kPa conventional experiment, is shown. The
at the start of initial shear (Figure 10). However, the 100 kPa stress-strain curves are identical within error for all four tests.
sample shows a notably different style of behavior, displaying In each of the landslide simulation tests, the material has been
dilation throughout the shear phase, possibly suggesting a taken to its peak strength and then held at constant shear
behavior akin to static liquefaction. The four samples together stress. In each case the displacement rate was then allowed to
generate a linear failure envelope (Figure 11) with a cohesion decline (Figure 13), showing an exponential reduction in
of 29 kPa and an angle of internal friction of 45.81. These are displacement with time. Note, however, that displacement did
typical values for a fine, unbonded sand (Bell, 2000). not cease completely, with creep continuing as expected at the
The pore pressure reinflation tests can be used to examine peak strength condition – in effect the sample is at a factor of
the patterns of displacement seen for specific effective stress safety of close to one – i.e., it is on the verge of failing - in this
conditions. In these tests, identical samples have been con- condition.
solidated to a normal effective stress of 400 kPa, and a shear Thereafter the samples were subject to a steady increase in
stress of 470 kPa has been applied. Thereafter the pore pres- pore pressure, at three different rates, inducing a reduction in
sure has been increased, reducing the normal effective stress mean effective stress. The development of the first 1% strain
and inducing failure. In the three experiments reported here (in this case, normalized to the start of the pore pressure
the pore pressure was increased at differing rates. reinflation phase) is shown in Figure 14. The three curves

5 200 kPa
Total normal stress
400 kPa
4
Stress and pore pressure (kPa)

800 kPa
Nor
Vertical strain (%)
mal 3 100 kPa
effe Horizontal strain
ctiv
e st
ress 2
Strain (%)

0
Shear stress
–1

e –2
s sur 0 1 2 3 4 5 6 7 8 9 10
e pre
Por Shear strain (%)

Time Figure 10 Vertical strain against shear strain for the BPSB tests
shown in Figure 9.
Figure 8 Graphical representation of the pore pressure reinflation
test. The total normal and the shear stresses are both held at a
constant value, whereas the pore pressure is increased. This reduces
the normal effective stress state, reducing shear resistance and thus
triggering the development of shear strain. 900
y = 1.0278x + 28.998
800

900 700
Peak shear stress (kPa)

800 600
100 kPa
700 200 kPa
Shear stress (kPa)

500
600 400 kPa
800 kPa
500 400
400
300
300
200 200
100
100
0
0 1 2 3 4 5 6 7 8 9 10 0
Shear strain (%) 0 100 200 300 400 500 600 700 800 900
Normal effective stress (kPa)
Figure 9 Conventional shear stress against shear strain graphs
derived from the BPSB on tests of fine sand. Figure 11 The failure envelope derived from the tests in Figure 9.
 Direct Shear Testing in Geomorphology 335

500 1.0

Shear strain during PPR (%)

450 0.9
400 0.8
Shear stress (kPa)

350 0.7
0.6
300
0.5
250
0.4
200
400 kPa conventional 0.3 PPR2
150 0.2
PPR2 PPR3
100 PPR3 0.1 PPR4
50 PPR4 0.0
0 340 350 360 370 380 390 400
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 Effective normal stress (kPa)
Shear strain (%) Figure 14 Shear strain against normal effective stress plot for the
Figure 12 Shear stress against shear strain curves for the reinflation phase of the PPR tests. In each case a substantial increase
conventional phase of the PPR tests, plotted with the same curve for in strain rate occurred at 365–380 kPa.
a conventional test as shown in Figure 9.

with the sample with the lowest reinflation rate showing the
lowest strain rate. This suggests that in a real landslide system
6.5
the rate of movement might be controlled not only by the
6.4
effective stress state (i.e. the factor of safety), but also by the
6.3 rate at which the pore pressure changes.
Displacement (mm)

6.2 Also in Figure 15, note that the PPR2 and PPR3 both
6.1 showed substantial fluctuations in strain rate, even in this
6.0 comparatively simple system. Changes in strain rate of this
5.9 type have been observed in natural landslides (e.g., Allison
5.8 and Brunsden, 1990; Petley et al., 2005a), and are generally
PPR2 thought to be associated with local pore pressure build up and
5.7
PPR3 dissipation.
5.6
PPR4 Finally, it has frequently been suggested that the easiest way
5.5 to forecast the point at which the rate of strain will rapidly
0 5000 10 000 15 000 20 000 25 000 30 000
increase (i.e. rapid landslide movement will begin) is to plot
Time (min) the inverse of strain rate (l) against time (t) (e.g., Petley et al.,
Figure 13 Plot of shear strain against time for the end of the 2002). This plot has been generated, based on the data that are
conventional phase of the PPR tests. Under constant stress smoothed to remove scatter, for experiment PPR2 in
conditions the samples showed an exponential decline in strain rate, Figure 16. In this case, normal effective stress (p0 ) is used as a
but this did not reduce to zero. proxy for time; this is reasonable as the normal effective stress
is reduced at a constant rate. In this plot, a y-axis value of zero
represents infinite strain rate – i.e., catastrophic failure. Due to
differ slightly, with the sample for which pore pressures were its non-brittle nature, loose sand does not show this style of
being increased at the lowest rate demonstrating the greatest deformation. This plot shows that the early data can be used
accumulation of strain for any given effective stress state. This to forecast rapid movement at a normal effective stress of
is probably a result of the time taken for pore pressure approximately 376 kPa. The non-linear behavior displayed in
equilibration in the sample (as per Head, 1986). Note, how- l – p0 space is as predicted for non-brittle materials in Petley
ever, that initially strain increases slowly but that at approxi- et al. (2002, 2005b).
mately 380 kPa (or 370 kPa for PPR4) there is a dramatic
increase in strain accumulation. This illustrates the sensitivity
of landslides to higher pore pressure conditions, when small
changes in effective stress can lead to a large change in be- 14.28.8 Discussion
havior. Thereafter the three samples behave generally in the
same way, although at the very highest pore pressure con- Although the conventional direct shear apparatus is a proven
ditions PPR4 accumulated notably less strain than the other and highly trusted piece of equipment in the geotechnical
samples. engineering domain, within geomorphology its application
Of course a key parameter of landslide movement is the has been surprisingly limited. In part this is probably because
rate of displacement (Petley et al., 2002). Although the be- of the lack of need for standard geotechnical parameters (co-
havior of the three samples is the same in strain – shear stress hesion and angle of internal friction) in applications beyond
(or time) space, the strain rate (movement rate) behavior is conventional engineering design, and in part it may be
rather different (Figure 15). It is clear that the rate of strain for associated with a resistance to quantification of material
the samples is controlled by the rate of pore pressure increase, deformation within geomorphology. However, the key issue is
336 Direct Shear Testing in Geomorphology

1.0 E + 00

Strain rate during PPR (%/min)

1.0 E – 01

1.0 E – 02

1.0 E – 03

1.0 E – 04

PPR2
1.0 E – 05
PPR3
PPR4
1.0 E – 06
280 290 300 310 320 330 340 350 360 370 380 390 400
Effective normal stress (kPa)
Figure 15 Plot of strain rate (plotted on a log scale) against normal effective stress for the three PPR tests. Note that behavior differed
substantially between the three samples.

In this chapter, the application of the BPSB to a simple

1/strain rate during PPR (min/%)

6.0 E + 05
material has been demonstrated. An advantage of the appar-
PPR2
5.0 E + 05 atus is that it allows conventional direct shear testing to be
undertaken, as well as the PPR tests. The results of the con-
4.0 E + 05 ventional tests demonstrate that the apparatus generates the
3.0 E + 05 types of results that would be expected, but with an improved
level of precision and reliability. The PPR tests provide a great
2.0 E + 05 deal, however, providing a basis for the interpretation and
1.0 E + 05
analysis of movement patterns, and in particular for the
forecasting of behavior. Thus, in a translational landslide the
0.0 E + 00 stress conditions can be replicated in the BPSB and a range of
360 370 380 390 400 pore pressure scenarios recreated. If monitoring of pore pres-
Effective normal stress (kPa) sures in the landslide system has been undertaken, then a
Figure 16 The Saito curve (here plotted as the inverse of strain rate frequency (probability) of occurrence can be determined for
against normal effective stress, showing the asymptotic trend particular high pore pressure conditions, permitting for the
identified by Petley et al. (2002). first time a probability of movement of particular magnitudes
to be deduced. Potentially, this is a very powerful management
probably associated with the interest within geomorphology tool for landslide systems.
in process rather than in designs for mitigation. Unfortunately, Finally, these new developments in direct shear test provide
the conventional direct shear machine does not provide sig- the potential for the investigation of other aspects of landslide
nificant insight into process due to its limitations. As a result, behavior with suitable modification to the apparatus. For ex-
geomorphologists have tended to focus on field-based in- ample, the normal stress and shear stress actuators could be
vestigations (e.g., Gonzalez-Diez et al., 2009; Crozier, 2010), modified to provide a dynamic capability, permitting simu-
sometimes backed up with direct shear testing for the geo- lation, for the first time, of the stress conditions on a shear
technical parameters. surface during an earthquake event. Given our poor under-
The BPSB represents a new generation of device that starts standing of the processes acting within a slope during seismic
to work around some of the key issues, providing the potential excitation, this is a very interesting potential development.
for insight into landslide mechanisms. The ability to recreate Furthermore, fitted with a suitable system to maintain low
the stress conditions within a slope is important, allowing the temperatures, the device could also be used to explore the
development of failure and post-failure movement patterns to behavior of ice-sediment boundaries at the beds of glaciers.
be properly represented for the first time. Thus, the researcher
can begin to understand both how failure develops (for ex-
ample the ways in which crack development controls the 14.28.9 Conclusions
movement pattern – see Petley et al., 2002) and the ways in
which post-failure strains accumulate (for example, the rela- This chapter has reviewed the use of the conventional direct
tionship between rate of change of pore pressure and rate of shear testing, demonstrating that although simple and well-
strain). This provides the very real potential for behavior established, the technique is under-used in geomorphology.
forecasting for real landslide systems, currently one of the holy The primary reasons for this are associated with the lack of
grails of landslide research. information that the technique generates on process. The
 Direct Shear Testing in Geomorphology 337

BPSB apparatus, which represents a long overdue evolution of Crozier, M.J., 2010. Landslide geomorphology: An argument for recognition, with
the direct shear machine, making use of recent advances in examples from New Zealand. Geomorphology 120, 3–15.
Cristescu, N.D., Cazacu, O., Cristescu, C., 2002. A. model for slow motion of
hardware and software systems, represents a substantial im-
natural slopes. Canadian Geotechnical Journal 39(4), 924–937.
provement, permitting mechanism to be explored for the first Gonzalez-Diez, A., Schrott, L., Brunsden, D., 2009. Theories and methods in
time. As such, it represents an important advance for geomorphology: Introduction. Geomorphology 110, 1–2.
goemorphological studies of slope and other deformation- Head, K.H., 1986. Manual of Soil Laboratory Testing. Effective Stress Tests. John
controlled systems. Wiley and Sons, New York, Volume 3, 1053 pp.
Petley, D.N., Allison, R.J., 1997. The mechanics of deep-seated landslides. Earth
Surface Processes and Landforms 22, 747–758.
Petley, D.N., Bulmer, M.H.K., Murphy, W., 2002. Patterns of movement in rotational
References and translational landslides. Geology 30(no. 8), 719–722.
Petley, D.N., Higuchi, T., Petley, D.J., Bulmer, M.H., Carey, J., 2005b. The
development of progressive landslide failure in cohesive materials. Geology
Allison, R.J., Brunsden, D., 1990. Some mudslide movement patterns. Earth Surface
33(3), 201–204.
Processes and Landforms 15(4), 297–311.
Petley, D.N., Mantovani, F., Bulmer, M.H.K., Zannoni, F., 2005a. The interpretation
Bell, F.G., 2000. Engineering Properties of Rocks and Soils. Wiley-Blackwell.
of landslide monitoring data for movement forecasting. Geomorphology 66(1-4),
Chichester, 496 pp.
133–147.
BSI, 2005. Geotechnical investigation and testing. Field testing. Standard
Skempton, A.W., Petley, D.J., 1967. The strength along structural discontinuities in
penetration test. British Standards Institute, London, BS EN ISO 22476-3.
stiff clays. Proceedings of the Geotechnical Conference, Oslo volume 2,
Corominas, J., Moya, J., Ledesma, A., Lloret, A., Gili, J.A., 2005. Prediction of
pp. 29–46.
ground displacements and velocities from groundwater level changes at the
Wilson, A., Petley, D.N., Murphy, W., 2003. Geotechnical and pore fluid controls on
Vallcebre landslide (Eastern Pyrenees, Spain). Landslides 2(2), 83–96.
large-scale landslide movement in the Alps. Geomorphology 54(1-2), 49–62.
Craig, R.F., 2004. Soil mechanics, 7th edition Taylor & Francis, London, 464 pp.

Biographical Sketch

David Petley is the Wilson Professor of Hazard and Risk in the Department of Geography, and the Executive
Director of the Institute of Hazard, Risk and Resilience at Durham University in the UK. While his PhD was on
laboratory testing of oil reservoir overburden rocks, his subsequent research has focused on the behavior of
landslides using approaches that combine laboratory testing with field monitoring, in both cases using novel
approaches. His research is undertaken mostly in mountainous or coastal environments, especially in the UK,
Taiwan, New Zealand, and China.

Alison Clark is a Laboratory Technician in the Department of Geography at Durham University in the UK. Her
background is in material testing in industry, before moving into geotechnical testing within the university sector.
As well as playing a key role in the generation of datasets using a range of standard geotechnical tests, she has
played an important part in the development of novel lab-based experimental devices for the simulation of
landslides and in the establishment of protocols for their implementation.
 14.29 The Schmidt Hammer and Related Devices in Geomorphological
Research
AS Goudie, St. Cross College, Oxford, UK
r 2013 Elsevier Inc. All rights reserved.

14.29.1 Introduction 338

14.29.2 Operation of the SH 339
14.29.3 The Equotip and Piccolo 340
14.29.4 The Uses of the SH and Equotip 341
14.29.5 Conclusions 342
References 343

Abstract

The Schmidt Hammer and Equotip are two methods that are available to measure rock surface hardness. Being quick,
simple, cheap, and portable, they are commonly used by geomorphologists to relate rock hardness to landform devel-
opment, and to undertake relative dating of geomorphological phenomena. This chapter describes the advantages and
limitations of the instruments, their usage, and their applications.

14.29.1 Introduction reduced impact energy form (BL) (0.735 N m), which are re-
spectively the same as the impact energies of the N and L
In 1948, E. Schmidt devised the classic Schmidt Hammer (SH) versions of the classic SH.
(Figure 1) for carrying out in situ, nondestructive tests on
concrete to determine its hardness (Schmidt, 1951; Day and
Goudie, 1977; Day, 1980; Goudie, 2006). Geomorphologists
have now adopted it for a variety of roles, which include
relative dating, the study of weathering, and the effects of rock
strength on landforms (e.g., Ericson, 2004).
The instrument measures the distance of rebound of a
controlled impact on a rock surface. There are various versions
of the SH. The one that has been most used by geomorph-
ologists is the N type, which provides data on a range of rock
types from weak to very strong with compressive strengths
ranging from c. 20 to 250 MPa. The L-type hammer has an
impact three times lower than the N type (0.735 compared to
2.207 N m). It is suitable for testing of weak rocks and those
with thin (i.e., 0.1–2 mm) weathering crusts, which would be
fractured by the higher-impact N type. The P type is a pen-
dulum hammer for testing materials of very low hardness,
with compressive strengths of less than 70 kPa. Recently, a new
type of SH, the Silver Schmidt (Figure 2), has been introduced
(Viles et al., 2011). It is lighter in weight than the classic SH
(weighing 600 g). The readings are presented in a digital form
and these can be stored electronically and downloaded later.
Measurements with the Silver Schmidt are said to be in-
dependent of impact direction. In addition, it provides an
automatic conversion to the required measurement units
(kg cm 2, N mm 2, and psi). There are two different versions:
the standard impact energy form (BN) (2.207 N m) and the

Goudie, A.S., 2013. The Schmidt hammer and related devices in geo-
morphological research. In: Shroder, J. (Editor in Chief), Switzer, A.D.,
Kennedy, D.M. (Eds.), Treatise on Geomorphology. Academic Press, San
Diego, CA, vol. 14, pp. 338–345. Figure 1 The classic Schmidt Hammer being used in the field.

338 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00398-5

 The Schmidt Hammer and Related Devices in Geomorphological Research 339

before the impact was applied. Yavuz et al. (2006) collected 20

values and only used the top 10; and Gupta (2009) collected
50 samples per site and discarded the upper 10 and the lower
10. Aydin (2009) proposed a new International Society for
Rock mechanics (ISRM) method for the SH that recommends
20 R-values from single impacts separated by at least a plunger
width and further suggested that all values should be used to
calculate summary statistics and no values (high or low)
should be discarded. Niedzielski et al. (2009) undertook a
survey to determine the appropriate minimum sample size, in
terms of individual impact readings, for different rock types,
and recommended 15 for sandstone and shales, 25 for weak
and moderately strong rocks, and 30 for strong and coarse
rocks.
The SH has many advantages for geomorphological re-
search: portability (weighing between 1.7 and 2.3 kg in the
case of the Classic Schmidt and only 0.6 kg in the case of the
Silver Schmidt); simplicity; low cost (US$670 for a type N);
ability to make many readings in situ in the field; lack of op-
erator variance (Day and Goudie, 1977); empirical evidence
that R-values correlate well with other rock properties such as
compressive strength; easy calibration with a test anvil in the
laboratory (US$525); and lack of temperature effects on
readings (Day and Goudie, 1977: 28). No explicit studies
appear to have been done to see when and how often cali-
bration needs to be undertaken, although even with long-term
use (i.e., over some years) no change has been detected in the
hammers used by the author.
Figure 2 The Silver Schmidt. Conversely, the SH does have some limitations. For ex-
ample, it is sensitive to discontinuities in a rock, and fissile,
closely foliated and laminated rocks cannot easily be investi-
14.29.2 Operation of the SH gated. The test is sensitive to moisture contents, especially in
weak rocks (Sumner and Nel, 2002). Results may be influ-
When the SH is pressed against a surface, its piston rebounds. enced by surface texture, with smooth planar surfaces giving
The distance of this rebound is called the rebound value (R). higher readings than rough or irregular surfaces (Williams and
Harder rocks have higher R-values (Aydin and Basu, 2005) Robinson, 1983). Surface irregularities are often crushed be-
and this is shown by a pointer on a scale on the side of the fore the plunger tip reaches the main rock surface, resulting in
instrument (range 10–100). R-values are influenced by gravi- some loss of impact energy. Both the magnitude and repeat-
tational forces so that nonhorizontal R-values must be nor- ability of hammer readings increase with the degree of surface
malized with reference to the horizontal direction (see Day polishing. There may be between-hammer variation and de-
and Goudie, 1977, table 2; Kolaiti and Papadopoulos, 1993; terioration of the hammer with age. The block mass of the
Aydin and Basu, 2005). Aydin (2009) also reported that unless rock to be tested is significant, and the test cannot be used on
the hammer impact direction remains roughly perpendicular small, light blocks, either in the field (e.g., pebbles or talus
to the tested surface, there is a danger of frictional sliding of clasts) or in the laboratory (e.g., blocks used in weathering
the plunger tip, material removal by chipping, and a partial simulations). Sumner and Nel (2002) suggested that block
transfer of energy to and from the hammer. He suggested that weight should exceed 25 kg for accurate and consistent re-
if 20 readings are taken at each site, and are spaced 2 cm apart, bound determinations. Experimental laboratory-based studies
then the area over which readings should be taken would be by Demirdag et al. (2009) have shown that block size is critical
268 cm2. to SH results, with volumes of around 1000 cm3 considered as
In spite of six decades of use, there remains a wide vari- the minimum required. The SH cannot be used on soft ma-
ation in the recommended testing procedures (Goktan and terials and is destructive in that context (Aoki and Matsukura,
Gunes, 2005). This is particularly true as regards the number 2007). The N type, for example, is generally not suitable for
of impacts used to obtain R-value. Matthews and Shakesby use on the weaker varieties of shale or chalk or materials with
(1984) recommended 15 R-values for each sample, with R-values o10.
5 values that deviate most from the mean being discarded. It is important that the SH is used with care (McCarroll,
Katz et al. (2000) performed 32–40 individual impacts and 1987) and is properly calibrated. Because of the sensitivity of
averaged the upper 50%. Abnormally low values are omitted R-values to rock surface conditions, many workers recom-
for a variety of reasons: they may relate to the fact that the rock mend rubbing the surface to be tested with a carborundum
was weakened by the actual impact of the hammer on the rock block to remove surface irregularities or any organic cover.
surface, or to small rock flaws that were not spotted visually Whether or not, however, extensive pretreatment with a
340 The Schmidt Hammer and Related Devices in Geomorphological Research

Box 1 How to use the Classic SH

1. To reduce the risk of operator variance, readings should be made by one individual.
2. Before taking readings, if the aim is to assess the hardness of intact rock, remove any surface protuberances, lichen cover, or weathering crust with the
carborundum provided.
3. Do not use the SH like a hammer! The device applies a measured impact.
4. As readings (shown as R-values on a scale on the side of the instrument) may be affected by moisture content, record whether or not the surface being tested is
wet or dry, and try to ensure comparability between sites.
5. Take readings on blocks that have a weight of more than 25 kg or have edges that are at least 11 cm long. Do not take readings if the block moves in response
to the impact of the hammer.
6. Take readings at least 10 cm away from cracks or joints in the rock mass.
7. At any one site, undertake a pilot study of the variability of readings and adjust the sample size accordingly. The methodology is described by Viles et al.
(2010). Suggested sample sizes for different rock types are provided by Niedzielski et al. (2009).
8. As readings are affected by the angle between the SH and the surface being tested, this angle should be recorded. Nonhorizontal R-values should be normalized
with reference to the horizontal direction (as suggested by Aydin and Basu, 2005).

carborundum is undertaken depends on whether or not the

investigation is concerned with the hardness of the virgin rock
or with the degree of weathering it has undergone.
Box 1 makes suggestions as to how to use the classic SH.

14.29.3 The Equotip and Piccolo

In addition to the SH, various other instruments have recently

been developed to measure rock hardness. The Equotip
(Figure 3) is an electronic rebound hardness tester that was
developed in the 1970s (Kompatscher, 2004). Originally de-
signed for testing metals, it has now been used for testing both
rock hardness (e.g., Kawasaki et al., 2002; Aoki and Matsu-
kura, 2007) and the effects of weathering on rock hardness
(Kawasaki and Kaneko, 2004; Mol and Viles, 2010). Aye et al.
(2010) used the Equotip to assess resistance of cement mortars
to sulfate attack.
The device fires by spring force an impact body containing
a permanent magnet and a very hard indenter sphere (a
tungsten carbide ball) toward the surface of the material to be
tested. The velocity of the impact and the rebound phase is
measured by the induction voltage generated by the moving
magnet through a defined induction coil. The hardness value
is expressed as the Leeb number (L-value) or Leeb hardness
(HL) after the name of the inventor. This is the ratio of the
rebound velocity to the impact velocity multiplied by 1000.
Low hardness values are expressed by low L-values on a liquid
crystal display (LCD), stored electronically, and then can be Figure 3 The Equotip being used in the field.
downloaded later.
The instrument uses automatic compensation for impact
direction and is light in weight (780 g plus a 120-g battery rough surfaces, such as those with protruding quartz grains or
pack). The impact energy of the standard type (D) is low with surfaces that have micro-pits caused by chemical wea-
(11 N m), although versions with an impact value of 3 N m thering. Aoki and Matsukura (2008a) found that good cor-
(type C) and 90 N m (type G) are also available (Verwaal and relations could be achieved between L-values and unconfined
Mulder, 1993). The impact energy of the D type is approxi- compressive strength. Various methods for collecting L-values
mately 1/200 that of the SH N type, and 1/66 that of the SH L and calculating descriptive statistics have been suggested; for
type, so that less damage is caused to the surface being tested. example, Aoki and Matsukura (2007) used the repeat impacts
It is therefore possible to test softer rocks than is possible with method (RIM) to calculate Lmax from the mean of the three
the SH (Aoki and Matsukura, 2007). On the debit side, al- largest of 20 impacts on exactly the same spot, whereas the
though the device can be used on relatively small samples and single impacts method (SIM) was used to calculate Ls from 20
on those of limited thickness, it cannot be used successfully on individual impacts in one small area.
 The Schmidt Hammer and Related Devices in Geomorphological Research 341

modulus, attempts have been made to determine the correl-

ation between R-values and other measures of rock physical
properties, including the point load index (Aggistalis et al.,
1996), the Shore scleroscope (Yas- ar and Erdoğan, 2004; Sha-
labi et al., 2007), and the slake durability index and P-wave
velocity (Sharma et al., 2010).
The SH has been used to obtain relative dates of geo-
morphological phenomena, for intuitively there should be a
relationship between the degree of weathering and the length
that the rock surface has been exposed to weathering attack
(Matthews and Shakesby, 1984). This is the basis upon which
the SH has been used to estimate relative ages of, inter alia,
glacial moraine, rock glaciers, mass movements, talus, raised
shorelines and platforms, and anthropogenic features. The
relative dates obtained by this method can be compared with
other relative dating techniques such as weathering rind de-
Figure 4 The Piccolo being used in the field. velopment, clast angularity, soil development, lichenometry,
etc., or compared with radiocarbon dating and archival data
(Matthews et al., 1986). They have also recently been com-
A compact version of the Equotip has been introduced into pared with cosmogenic nuclide absolute exposure ages (Sán-
the market and is known as the Piccolo (Figure 4). This chez et al., 2009).
weighs only 110 g and has a low impact energy (11 N m). Like SH dates often show considerable variability, partly be-
the standard Equotip, it uses automatic compensation for cause of local microclimatic differences, including the influ-
impact direction. ence of late-lying snow patches in certain topographic
A detailed comparison of the Equotip and the SHs is pro- situations. For this reason, some workers only take readings
vided by Viles et al. (2010). In particular, they point out that on ridge crests (e.g., Aoyama, 2005) and over small altitudinal
whereas the SHs have a large impact force and volume, and ranges (White et al., 1998). It is also important to be aware of
thus record hardness within a broad near-surface zone, the the limitations of the SH technique itself because of factors
Piccolo and Equotip, with smaller impact forces and volumes, such as surface roughness of boulders (McCarroll, 1989) and
record hardness within a narrow surface zone. Equotip and variability in weathering rates between different lithologies
Piccolo readings also tend to be more variable than those from and mineralogies. In any one area, only the same rock types
the SH and this may be the result of their smaller impact should be tested because different original surface hardness
picking up microscale variations in surface conditions on and weathering resistance will affect the SH measurements
natural rock outcrops that are not detected by the coarser SHs. (Winkler, 2005). Care also needs to be taken with surface
Correlations of hardness values between the two types of SHs preparation if it is the degree of weathering that is the phe-
were very strong, and also between the Equotip and the Pic- nomenon being measured. Excess use of the carborundum
colo, but correlations between mean hardness values collected (e.g., to remove lichen cover) will remove the weathering
by the two SHs and the Equotip and Piccolo mean values were crust.
less strong, though still statistically significant. In some locations, SH dating has not been found to be very
effective at differentiating among depositional episodes so that
other measures, such as weathering rind development, have
14.29.4 The Uses of the SH and Equotip been found to be more useful (Nicholas and Butler, 1996).
Elsewhere, the SH has been shown to be a successful techni-
Because of its many virtues, the SH has been used as a means que in comparison with some others, but, plainly, the com-
of estimating other rock properties, such as compressive bined use of various techniques is likely to be superior to the
strength (Sendir, 2002). Various researchers have studied the use of one technique in isolation (Boelhouwers et al., 1999).
relationship between rock compressive strength and R-values Evans et al. (1999) found that the success of the SH technique
(Yas- ar and Erdoğan, 2004; Yagiz, 2009; Aydin, 2009). How- in dating Icelandic moraines was influenced by geomorpho-
ever, the regressions (listed in Goudie, 2006, table 3) vary logical setting, with areas subject to surging glaciers and re-
greatly between different rock types (Dinc- er et al., 2004), and working of moraines by debris flows being problematical. The
so should be used only for the individual lithologies to which resolution of SH dating is variable. On the one hand, there are
they apply (Sachpazis, 1990). Nonetheless, as Hack and those that believe that the maximum resolution can be c.
Huisman (2002) pointed out, a large number of simple SH 200–300 years (e.g., Winkler, 2005), whereas, on the other
tests in the field will tend to give a better estimate of the intact hand, there are those who have found it to be only useful for
rock strength at various locations than a much more limited distinguishing sites deglaciated during the Little Ice Age from
number of more complex tests. In addition, studies have in- those deglaciated during the Late Glacial and early Holocene
dicated strong empirical relations between SH R-values and (McCarroll and Nesje, 1993).
Young’s modulus (Katz et al., 2000), with R2 values as high as One area where rock hardness determinations have proved
0.99 (see also Sachpazis, 1990; Aggistalis et al., 1996; Yagiz, to be important in geomorphology is in determining of rock
2009). In addition to compressive strength and Young’s mass strength (RMS) (Del Potro and Hürlimann, 2008), and
342 The Schmidt Hammer and Related Devices in Geomorphological Research

in relating this to issues such as slope instability (Borrelli et al., rock hardness on the development of inselbergs has been
2007), slope form (Selby, 1980; Synowiec, 1999), and coastal studied by Pye et al. (1986) in the Kora district of Kenya. Here,
morphology, especially of shore platforms (Trenhaile et al., SH measurements showed no link between inselberg devel-
1998; Dickson et al., 2004; Kennedy, 2010; Naylor and opment and rock hardness, whereas geochemical analyses
Stephenson, 2010; Chelli et al., 2010). An RMS classification indicated potassium feldspar content to be the discriminating
involving the SH was used in the Napier Range of Australia to factor. With the aid of the SH, Stephenson and Kirk (2000)
identify seven main slope forms associated with different fa- found that weathering had reduced rock strength on some
cies of a Devonian reef, from which it was possible to draw New Zealand shore platforms by up to 50%, and so played a
associations between slope profile morphology and RMS major role in their development. Thornton and Stephenson
(Allison and Goudie, 1990). Similarly, Placek and Migoń (2006) identified a relationship between rock hardness and
(2007) investigated the relationship between SH values and shore platform elevation in Australia, whereas Kennedy and
gross relief in the Polish Sudetes. In central California, some Dickson (2006) used the SH to assess the importance of case
relationship was found between the yield of sediment from hardening on shore platforms at Shag Point in southern New
landslides and the SH values of the catchments from which Zealand, and found it less important than structural controls,
they were derived (Hapke, 2005). Attempts have also been notably jointing. Similarly, the Equotip has been used for as-
made to assess glacial trough morphology in relation to SH sessing the importance of case hardening in the development
values (Augustinus, 1992a, 1992b; Brook et al., 2004). of raised rims on intertidal shore platforms in Japan (Aoki and
Form of river channels with rock beds is a topic of in- Matsukura, 2008b).
creasing geomorphic interest. Mitchell et al. (2005) found that Rock hardness measurements can be used to create quan-
in the Colorado River SH rock hardness values correlated titative weathering indices (Karpuz and Pas- amehmetoğlu,
significantly with channel width and gradient. In Idaho, Lifton 2004). For example, La Pera and Sorriso-Valvo (2000) related
et al. (2009) found a strong negative correlation between SH SH R-values and a weathering classification to the biotite
values and valley width, with wide valley floors corresponding content of granites, and Arikan et al. (2007) used SH values in
to weak bedrock, and narrow valley floors to strong bedrock. a weathering classification system of acidic volcanic rocks.
They also found a statistically significant difference (po0.001) Quantitative assessment of the degree of weathering is helpful
in SH values between N- and S-facing slopes, indicating that in the identification of nunataks, trimlines, and glaciation
aspect affects weathering intensity and bedrock strength. In extent, for which the SH is a useful tool (Ballantyne et al.,
Japan, Hayakawa and Matsukura (2003) investigated the re- 1997; Anderson et al., 1998; Rae et al., 2004). Findings suggest
lationship between recession rates of waterfalls and various that R-values tend to be lower above the glacial limit because
rock properties, including R-values. Contrasts in channel and of prolonged periglacial weathering. Nicholson (2008, 2009)
floodplain geomorphology have been ascribed to differences estimated the relative degree of weathering of roches mou-
in the erosional resistance (i.e., surface hardness) between tonnées in a periglacial environment by comparing R-values
dolerite and sandstone sections of the Klip River, South Africa, on unprepared surfaces by taking two readings at a single
for which the SH was used to illustrate and quantify such point (i.e., without moving the hammer). The first reading
contrasts (Tooth et al., 2002). (R1) represented the hardness of the weathered surface,
Rock hardness measurements have also been used to ad- whereas the second R-value (R2) was believed to provide an
dress other questions relating to weathering, such as the approximation of intact rock strength. She believed that
controversy as to whether rates are linear or nonlinear. Sjöberg comparison of R1 and R2 values provides an index of surface
and Broadbent (1991) were able to obtain a measure of how weakening due to weathering.
weathering developed through time by examining the SH
values of raised beaches at different elevations in Sweden.
Spatial variations in hardness can also be used to investigate 14.29.5 Conclusions
environmental influences on weathering, such as aspect (Hall,
1993; Waragai, 1999) and the presence of seasonal snow Since it was first developed over five decades ago as a means of
patches (Ballantyne et al., 1989; Benedict, 1993; Grab et al., assessing the hardness of concrete, the SH has been found to
2005). have an increasing range of applications in geomorphology. It
Rock hardness measurements have been used to investigate is a convenient means of establishing rock hardness in the
weathering process–form relationships, with many studies field, provided certain precautions are taken in the light of its
of tafoni development (Matsukura and Matsuoka, 1996; known limitations. Being portable, cheap, free from operator
Matsukura and Tanaka, 2000; Mellor et al., 1997; Goudie variance, simple, easily calibrated, and free from any notice-
et al., 2002). Aoki and Matsukura (2007) used the Equotip to able temperature effects, it can, with due care, produce rock
assess the role of rock strength in determining the rate and size hardness values that correlate well with such parameters as
of tafoni growth in Japan. uniaxial compressive strength or Young’s modulus of elasticity.
Several studies have related different landforms to hardness By providing, quite rapidly, a large number of tests in the field,
as measured with the SH and Equotip. For instance, various the SH may in the end give a better estimate of the intact rock
karst features have been investigated in relation to limestone strength at various locations than a limited number of more
hardness (Day, 1981; Tang, 1998; Haryono and Day, 2004; complex and laboratory-based techniques. It can, of course, be
Goudie et al., 1989), whereas case hardening of limestone has used alongside other techniques to provide a more rigorous
also been identified with SH measurements (Yaalon and and robust characterization of rock properties and, to that
Singer, 1974, Day, 1980; Lyew-Ayee, 2004). The influence of end, R-values are one component in RMS characterizations. If
 The Schmidt Hammer and Related Devices in Geomorphological Research 343

geomorphologists are going to understand how rock type in- Boelhouwers, J.C., Jager, D.F., De Joode, A., 1999. Application of relative-age
fluences relief in a more quantitative manner than is usual, the dating methods to openwork debris flow deposits in the Cederberg
Mountains, Western Cape, South Africa. South African Geographical Journal 81,
SH can play a role. It is also becoming evident that RMS
135–142.
studies involving the SH can throw light on such diverse Borrelli, L., Greco, R, Gullà, G., 2007. Weathering grade of rock masses as a
phenomena as glacial troughs, river channel profiles, slope predisposing factor to slope instabilities: reconnaissance and control procedures.
forms, and coastal morphology. Geomorphology 87, 158–175.
The SH has now been extensively used for relative dating of Brook, M.S., Kirkbride, M.P., Brock, B.W., 2004. Rock strength and development of
glacial valley morphology in the Scottish Highlands and northwest Iceland.
mainly Holocene landforms, including moraines, rock gla- Geografiska Annaler 86A, 225–234.
ciers, and mass movement phenomena. It has also been Chelli, A., Pappalardo, M., Llopis, I.A., Federici, P.R., 2010. The relative influence
proved to be a valuable tool for assessing the rates at which of lithology and weathering in shaping shore platforms along the coastline of
weathering takes place through time, the extent to which the Gulf of La Spezia (NW Italy) as revealed by rock strength. Geomorphology
118, 93–104.
weathering is affected by aspect, the nature of tafoni, the form
Day, M.J., 1980. Rock hardness: field assessment and geomorphic importance.
of karstic landforms, the role of case hardening on natural Professional Geographer 32, 72–81.
rock outcrops and on buildings, the lithological control on Day, M.J., 1981. Rock hardness and landform development in the Gunong Mulu
inselberg formation, the classification of weathering residues, National Park, Sarawak, E. Malaysia. Earth Surface Processes and Landforms 6,
the role of weathering in shore platform development, and in 165–172.
Day, M.J., Goudie, A.S., 1977. Field assessment of rock hardness using the
the delimitation of glacial limits.
Schmidt test hammer. BGRG Technical Bulletin 18, 19–29.
Del Potro, R., Hürlimann, M., 2008. Geotechnical classification and characterization
of materials for stability analyses of large volcanic slopes. Engineering Geology
98, 1–17.
References Demirdag, S., Yavuz, H., Altindag, R., 2009. The effect of sample size on Schmidt
rebound hardness value of rocks. International Journal of Rock Mechanics and
Aggistalis, G., Alivizatos, S., Stamoulis, D., Stournaras, G., 1996. Correlating Mining Sciences 46, 725–730.
uniaxial compressive strength with Schmidt hardness, point load index, Young’s Dickson, M.E., Kennedy, D.M., Woodroffe, C.D., 2004. The influence of rock
Modulus, and mineralogy of gabbros and basalts (northern Greece). Bulletin of resistance on coastal morphology around Lord Howe Island, southwest Pacific.
the International Association of Engineering Geology 54, 3–11. Earth Surface Processes and Landforms 29, 629–643.
Allison, R.J., Goudie, A.S., 1990. The form of rock slopes in tropical limestone and Dinc- er, I., Acar, A., Cobangulu, I., Uras, Y., 2004. Correlation between Schmidt
their associations with rock mass strength. Zeitshrift für Geomorphologie 34, hardness, uniaxial compressive strength and Young’s modulus for andesites,
129–148. basalts and tuffs. Bulletin of Engineering Geology and the Environment 63,
Anderson, E., Harrison, S., Passmore, D.G., Mighall, T.M., 1998. Geomorphic 141–148.
evidence of Younger Dryas glaciation in the Macgillycuddy’s Reeks, south west Ericson, K., 2004. Geomorphological surfaces of different age and origin in granite
Ireland. Journal of Quaternary Science 13, 75–90. landscapes: an evaluation of the Schmidt test hammer. Earth Surface Processes
Aoki, H., Matsukura, Y., 2007. A new technique for non-destructive field and Landforms 29, 495–509.
measurement of rock-surface strength: an application of the Equotip hardness Evans, D.J.A., Archer, S., Wilson, D.J.H., 1999. A comparison of the lichenometric
tester to weathering studies. Earth Surface Processes and Landforms 32, and Schmidt hammer dating techniques based on data from the proglacial areas
1759–1769. of some Icelandic glaciers. Quaternary Science Reviews 18, 13–41.
Aoki, H., Matsukura, Y., 2008a. A formative process of raised rims along joints on Goktan, R.M., Gunes, N., 2005. A comparative study of Schmidt hammer testing
sandstone layers forming intertidal shore platforms. Transactions, Japanese procedures with reference to rock cutting machine performance prediction.
Geomorphological Union 29, 387–397. International Journal of Rock Mechanics and Mining Sciences 42, 466–472.
Aoki, H., Matsukura, Y., 2008b. Estimating the unconfined compressive strength of Goudie, A.S., 2006. The Schmidt hammer in geomorphological research. Progress
intact rocks from Equotip hardness. Bulletin of Engineering Geology and the in Physical Geography 30, 703–718.
Environment 67, 23–29. Goudie, A.S., Bull, P.A., Magee, A.W., 1989. Lithological control of rillenkarren
Aoyama, M., 2005. Rock glaciers in the northern Japanese Alps: development in the Napier Range, Western Australia. Zeitschrift für
palaeoenvironmental implications since the Late Glacial. Journal of Quaternary Geomorphologie Supplementband 75, 95–114.
Science 20, 471–484. Goudie, A.S., Migoń, P., Allison, R.J., Rosser, N., 2002. Sandstone geomorphology
Arikan, F., Ulusay, R., Aydin, N., 2007. Characterization of weathered acidic volcanic of the Al-Quwayra area of south Jordan. Zeitschrift für Geomorphologie 46,
rocks and a weathering classification based on a rating system. Bulletin of 365–390.
Engineering Geology and Environment 66, 415–430. Grab, S., Van Zyl, C., Mulder, N., 2005. Controls on basalt terrace formation in the
Augustinus, P.C., 1992a. Rock resistance to erosion: some further considerations. eastern Lesotho highlands. Geomorphology 67, 473–485.
Earth Surface Processes and Landforms 16, 563–569. Gupta, V., 2009. Non-destructive testing of some Higher Himalayan rocks
Augustinus, P.C., 1992b. The influence of rock mass strength on glacial valley in the Satluj Valley. Bulletin of Engineering Geology and Environment 68,
cross-profile morphometry: a case study from the Southern Alps, New Zealand. 409–416.
Earth Surface Processes and Landforms 17, 39–51. Hack, R., Huisman, M., 2002. Estimating the intact rock strength of a rock mass by
Aydin, A., 2009. ISRM Suggested method for determination of the Schmidt hammer simple means. Proceedings 9th Congress of the International Association for
rebound hardness: revised version. International Journal of Rock Mechanics and Engineering Geology and the Environment. Durban, South Africa, 16–20
Mining Sciences 46, 627–634. September 2002, pp. 1971–1977.
Aydin, A., Basu, A., 2005. The Schmidt hammer in rock material characterization. Hall, K., 1993. Enhanced bedrock weathering in association with late-lying snow
Engineering Geology 41, 1211–1214. patches – evidence from Livingston Island, Antarctica. Earth Surface Processes
Aye, T., Oguchi, C.T., Takaya, Y., 2010. Evaluation of sulfate resistance of Portland and Landforms 18, 121–129.
and high alumina cement mortars using hardness test. Construction and Hapke, C.J., 2005. Estimation of regional material yield from coastal landslides
Building Materials 24, 1020–1026. based on historical digital terrain modelling. Earth Surface Processes and
Ballantyne, C.K., Black, N.M., Finlay, D.P., 1989. Enhanced boulder weathering Landforms 30, 679–697.
under late-lying snow patches. Earth Surface Processes and Landforms 14, Haryono, E., Day, M., 2004. Landform differentiation within the Gunung Kidul
745–750. Kegelkarst, Java, Indonesia. Journal of Cave and Karst Studies 66, 62–69.
Ballantyne, C.K., McCarroll, D., Asje, A., Dahl, S.O., 1997. Periglacial trimlines, Hayakawa, Y., Matsukura, Y., 2003. Recession rates of waterfalls in Boso Peninsula,
former nunataks and the altitude of the last ice sheet in Wester Ross, northwest Japan, and a predictive equation. Earth Surface Processes and Landforms 28,
Scotland. Journal of Quaternary Science 12, 225–238. 675–684.
Benedict, J.B., 1993. Influence of snow upon rates of granodiorite weathering, Karpuz, C., Pas- amehmetoğlu, A.G., 2004. Field characterisation of weathered Ankara
Colorado Front Range, USA. Boreas 22, 87–92. andesites. Engineering Geology 46, 1–17.
344 The Schmidt Hammer and Related Devices in Geomorphological Research

Katz, O., Reches, Z., Roegiers, J.-C., 2000. Evaluation of mechanical rock properties Placek, A, Migoń, P., 2007. Rock–landform relationships in the Sudetes in the light
using a Schmidt Hammer. International Journal of Rock Mechanics and Mining of rock strength assessment using the Schmidt hammer. In: Goudie, A.S.,
Sciences 37, 723–728. Kalvoda, J. (Eds.), Geomorphological Variations. P3K, pp. 287–311.
Kawasaki, S., Kaneko, K., 2004. Estimation method for weathering thickness of Pye, K., Goudie, A.S., Watson, A., 1986. Petrological influence on differential
man-made weathering rocks by using the Equotip hardness Tester. Proceedings weathering and inselberg development in the Kora area of central Kenya. Earth
of the ISRM Regional Symposium EUROROCK 2004 and 53rd Geomechanics Surface Processes and Landforms 11, 41–52.
Colloquy. Salzburg, Austria, 7–9 October 2004, pp. 491–494. Rae, A.C., Harrison, S., Mighall, T., Dawson, A.G., 2004. Periglacial trimlines and
Kawasaki, S., Tanimoto, C., Koizumi, K., Ishikawa, M., 2002. An attempt to estimate nunataks of the Last Glacial Maximum: the Gap of Dunloe, southwest Ireland.
mechanical properties of rocks using the Equotip Hardness tester. Journal of Journal of Quaternary Science 19, 87–97.
Japan Society of Engineering Geology 43, 244–248, (in Japanese with English Sachpazis, C.I., 1990. Correlating Schmidt hardness with compressive strength and
abstract). Young’s modulus of carbonate rocks. Bulletin of the International Association of
Kennedy, D.M., 2010. Geological control on the morphology of estuarine shore Engineering Geology 42, 75–83.
platforms: Middle Harbour, Sydney, Australia. Geomorphology 114, 71–77. Sánchez, J.S., Mosquera, D.F., Romani, J.R.V., 2009. Assessing the age-
Kennedy, D.M., Dickson, M.E., 2006. Lithological control on the elevation of shore weathering correspondence of cosmogenic 21Ne dates of Pleistocene surfaces
platforms in a microtidal setting. Earth Surface Processes and Landforms 31, by the Schmidt hammer. Earth Surface Processes and Landforms 34,
1575–1584. 1121–1125.
Kolaiti, E., Papadopoulos, Z., 1993. Evaluation of Schmidt rebound hammer testing: Schmidt, E., 1951. A non-destructive concrete tester. Concrete 59, 34–35.
a critical approach. Bulletin International Association of Engineering Geology 48, Selby, M.J., 1980. A rock mass strength classification for geomorphic purposes:
69–76. with test from Antarctica and New Zealand. Zeitschrift für Geomorphologie 24,
Kompatscher, M., 2004. Equotip-rebound Hardness Testing after D. Leeb. 31–51.
Conference on Hardness Measurements Theory and Application in Laboratories Sendir, H., 2002. Correlation of Schmidt hardness with unconfined compressive
and Industries. Washington, DC, 11–12 November 2004, 7 pp. strength and Young’s modulus in gypsum from Sivas (Turkey). Engineering
La Pera, E., Sorriso-Valvo, M., 2000. Weathering and morphogenesis in a Geology 66, 211–219.
Mediterranean climate, Calabria, Italy. Geomorphology 34, 251–270. Shalabi, F.I., Cording, E.J., Al-Hattamleh, O.H., 2007. Estimation of rock
Lifton, Z.M., Thackray, G.D., Glenn, N.F., Van Kirk, R., Glenn, N.F., 2009. Influence engineering properties using hardness tests. Engineering Geology 90,
of rock strength on the valley morphometry of Big Creek, central Idaho, USA. 138–147.
Geomorphology 111, 173–181. Sharma, P.K., Khandelwal, M., Singh, T.N., 2010. A correlation between Schmidt
Lyew-Ayee, P., 2004. Digital topographic analysis of cockpit karst: a morpho- hammer rebound numbers with impact strength index, slake durability index and
geological study of the Cockpit Country Region, Jamaica. D Phil thesis, P-wave velocity. International Journal of Earth Science, doi:10.1007/s00531-
University of Oxford, UK, unpublished. 009-0506-5.
Matsukura, Y., Matsuoka, N., 1996. The effect of rock properties on rates of tafoni Sjöberg, R., Broadbent, N., 1991. Measurement and calibration of weathering using
growth in coastal environments. Zeitschrift für Geomorphologie Supplementband the Schmidt Hammer on wave washed moraines in the upper Norrland Coast,
106, 57–72. Sweden. Earth Surface Processes and Landforms 16, 57–64.
Matsukura, Y., Tanaka, Y., 2000. Effect of rock hardness and moisture content on Stephenson, W.J., Kirk, R.M., 2000. Development of shore platforms on Kaikoura
tafoni weathering in the granite of Mount Doeg-Sung. Korea. Geografiska Peninsula, South Island, New Zealand II: the role of subaerial weathering.
Annaler 82A, 59–67. Geomorphology 32, 43–56.
Matthews, J.A., Dawson, A.G., Shakesby, R.A., 1986. Lake shoreline development, Sumner, P., Nel, W., 2002. The effect of rock moisture on Schmidt hammer
frost weathering and rock platform erosion in an alpine periglacial environment, rebound: tests on rock samples from Marion Island and South Africa. Earth
Jotunheimen, southern Norway. Boreas 15, 33–50. Surface Processes and Landforms 27, 1137–1142.
Matthews, J.A., Shakesby, R.A., 1984. The status of the ‘Little Ice Age’ in southern Synowiec, G., 1999. Rock mass strength assessment for geomorphological
Norway: relative-age dating of Neoglacial moraines with Schmidt hammer and purposes and its application to sandstone slopes in the Stolowe Mountains.
lichenometry. Boreas 13, 333–346. Czasopismo Geograficzne 70, 1–361.
McCarroll, D., 1987. The Schmidt Hammer in geomorphology: five sources of Tang, T., 1998. Field testing of rock hardness and its relationship to limestone
instrument error. BGRG Technical Bulletin 36, 16–27. dissolution in Guilin, southern China. Middle States Geographer 31, 15–22.
McCarroll, D., 1989. Schmidt hammer relative-age evaluation of a possible pre- Thornton, L.E., Stephenson, W.J., 2006. Rock strength: a control of shore platform
‘‘Little Ice Age’’ Neoglacial moraine. Leirbreen, southern Norway. Norsk evolution. Journal of Coastal Research 22, 224–231.
Geologisk Tiddskrift 69, 125–130. Tooth, S., McCarthy, T.S., Brandt, D., Hancox, P.J., Morris, R., 2002. Geological
McCarroll, D., Nesje, A., 1993. The vertical extent of ice sheets in Nordfjord, controls on the formation of alluvial meanders and floodplain wetlands: the
Western Norway – measuring degree of rock surface weathering. Boreas 22, example of the Klip River, eastern Free State, South Africa. Earth Surface
255–265. Processes and Landforms 27, 797–815.
Mellor, A., Short, J., Kirkby, S.J., 1997. Tafoni in the El Chorro area, Andalucia, Trenhaile, A.S., Pepper, D.A., Trenhaile, R.W., Dalimonte, M., 1998. Stacks and
southern Spain. Earth Surface Processes and Landforms 22, 817–833. notches at Hopewell Rocks, New Brunswick, Canada. Earth Surface Processes
Mitchell, K.J., Mackley, R.D., Pederson, J.L., 2005. Quantifying bedrock strength and Landforms 23, 975–988.
with respect to fluvial erodibility along the Colorado River: comparing in situ Verwaal, W., Mulder, A., 1993. Estimating rock strength with the Equotip hardness
and laboratory methods. Geological Society of America Abstracts with Programs tester. International Journal of Rock Mechanics and Mining Sciences 30,
37, 295. 659–662.
Mol, L., Viles, H.A., 2010. Geoelectric investigations into sandstone moisture Viles, H.A., Goudie, A.S., Grab, S., Lalley, J., 2011. The use of the Schmidt
regimes: implications for rock weathering and the deterioration of San Hammer and Equotip for rock hardness assessment in geomorphology and
Rock Art in the Golden Gate Reserve, South Africa. Geomorphology 118, heritage science: a comparative analysis. Earth Surface Processes and
280–287. Landforms 36(3), 320–333.
Naylor, L.A., Stephenson, W.J., 2010. On the role of discontinuities in mediating Waragai, T., 1999. Weathering processes on rock surfaces in the Hunza Valley,
shore platform erosion. Geomorphology 114, 89–100. Karakoram, North Pakistan. Zeitschrift für Geomorphologie Supplementband 119,
Nicholas, J.W., Butler, D.R., 1996. Application of relative-age dating techniques on 119–136.
rock glaciers of the La Sal Mountains, Utah: an interpretation of Holocene White, K., Bryant, R., Drake, N., 1998. Techniques for measuring rock weathering:
paleoclimates. Geografiska Annaler 78A, 1–18. application to a dated fan segment sequence in southern Tunisia. Earth Surface
Nicholson, D.T., 2008. Rock control on microweathering of bedrock surfaces in a Processes and Landforms 23, 1031–1043.
periglacial environment. Geomorphology 101, 655–665. Williams, R.B.G., Robinson, D.A., 1983. The effect of surface texture on the
Nicholson, D.T., 2009. Holocene microweathering rates and processes on ice- determination of the surface hardness of rock using the Schmidt Hammer. Earth
eroded bedrock, Rǿldal area, Hardangervidda, southern Norway. Geological Surface Processes and Landforms 8, 289–292.
Society of London Special Publication 320, 29–49. Winkler, S., 2005. The Schmidt hammer as a relative-age dating technique: potential
Niedzielski, T., Migoń, P., Placek, A., 2009. A minimum sample size required from and limitations of its application on Holocene moraines in Mt Cook National
Scmidt hammer measurements. Earth Surface Processes and Landforms 34, Park, Southern Alps, New Zealand. New Zealand Journal of Geology and
1713–1725. Geophysics 48, 105–116.
 The Schmidt Hammer and Related Devices in Geomorphological Research 345

Yaalon, D.H., Singer, S., 1974. Vertical variation in strength and porosity of calcrete Yas- ar, E., Erdoğan, Y., 2004. Estimation of rock physicomechanical properties using
(nari) on chalk, Shefela, Israel and interpretation of its origin. Journal of hardness methods. Engineering Geology 71, 281–288.
Sedimentary Petrology 44, 1016–1023. Yavuz, H., Altindag, R., Sarac, S., Ugur, I., Sengun, N., 2006. Estimating the index
Yagiz, S., 2009. Predicting uniaxial compressive strength, modulus of elasticity and properties of deteriorated carbonate rocks due to freeze–thaw and thermal shock
index properties of rocks using the Schmidt hammer. Bulletin of Engineering weathering. International Journal of Rock Mechanics and Mining Sciences 43,
Geology and Environment 68, 55–63. 520–530.

Biographical Sketch

Andrew Goudie is a master of St. Cross College, University of Oxford, a professor of geography, and a former
president of the International Association of Geomorphologists. He holds the Founders Medal of the Royal
Geographical Society, the David Linton Award of the British Society for Geomorphology, and the Farouk El Baz
Award for Desert Research of the Geological Society of America.
 14.30 An Introduction to Dating Techniques: A Guide for Geomorphologists
CR Sloss, Queensland University of Technology, Brisbane, QLD, Australia
KE Westaway, Macquarie University, NSW, Australia
Q Hua, Australian Nuclear Science & Technology Organisation, Kirrawee DC, NSW, Australia
CV Murray-Wallace, University of Wollongong, Wollongong, NSW, Australia
r 2013 Elsevier Inc. All rights reserved.

14.30.1 Introduction 347

14.30.2 Dating Issues 347
14.30.2.1 Depositional Context 348
14.30.2.2 Taphonomy of Fossil Material 348
14.30.3 Dating Methods 349
14.30.4 Sidereal or Incremental Dating 349
14.30.5 Isotopic: Change in Isotopic Composition 350
14.30.6 Radiocarbon Dating 350
14.30.6.1 Principles of Radiocarbon Dating 350
14.30.6.2 Sample Preparation and Radiocarbon Measurements 351
14.30.6.3 Radiocarbon Calibration 351
14.30.6.4 Reservoir Effects 351
14.30.6.5 Bomb-pulse 14C Dating 352
14.30.6.6 Applications of Radiocarbon Dating to Geomorphology 353
14.30.7 Radiogenic: Luminescence Dating 356
14.30.7.1 Resetting of the OSL Signal in Different Sedimentary Environments 356
14.30.7.2 Technological Developments that have made Luminescence Techniques More Applicable to Geo-
morphological Research 357
14.30.7.3 Advances in Methods of Analysis – Single Aliquot and Single Grain 357
14.30.7.4 Emission Spectra – Use of Different Luminescence Wavelengths 358
14.30.7.5 Application of OSL to Key Geomorphological Issues 358
14.30.7.6 Practical Issues for Dating Strategies and Sample Collection 359
14.30.8 Time Dependent Chemical Reactions 361
14.30.9 Amino Acid Racemization 361
14.30.9.1 Advantages of Amino Acid Racemization Dating 361
14.30.9.2 Principles of Amino Acid Racemization Dating 362
14.30.9.3 The Racemization Reaction 362
14.30.9.4 Factors Influencing the Racemization Rate in Carbonate Fossils 362
14.30.9.5 Diagenetic Temperature History 363
14.30.9.6 Moisture Regime 364
14.30.9.7 Taxonomic and Genus Effect 364
14.30.9.8 Numerical Age Assessments 364
14.30.10 Conclusion 364
References 365

Glossary Dose rate The total amount of radiation that the

Allochthonous Occurring in a place other than where sample has been exposed to during the burial period,
they or their constituents were formed. expressed as radiation dose per thousand years
Biocoenosis The interacting organisms living together in (Gy/ka).
a habitat, a diverse community inhabiting a single biotope. OSL Optically stimulated luminescence – a light-sensitive
Bleaching Exposing the mineral (quartz or feldspars) to signal that builds up in nature during a burial period and is
light so that all (complete bleaching) or part (partial measured under laboratory conditions using a stimulating
bleaching) of the luminescence signal is removed. light source.

Sloss, C.R., Westaway, K.E., Hua, Q., Murray-Wallace, C.V., 2013. An

introduction to dating techniques: a guide for geomorphologists. In:
Shroder, J. (Editor in Chief), Switzer, A.D., Kennedy, D.M. (Eds.), Treatise
on Geomorphology. Academic Press, San Diego, CA, vol. 14, Methods in
Geomorphology, pp. 346–369.

346 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00399-7

 An Introduction to Dating Techniques: A Guide for Geomorphologists 347

Paleodose A laboratory-measured estimate of the grains contribute to the single-aliquot signal while single
radiation dose that the sample received during a burial grains are measured individually.
period. Taphonomy The study of decaying organisms over time
Racimization Transformation of one-half of the and how they become fossilized.
molecules of an optically active compound into molecules Thanatocoenosis The assemblage of dead organisms or
that possess exactly the opposite (mirror image) fossils that occurred together in a given area at a given
configuration. moment of geologic time, also known as a death
Sidereal Sidereal time is time kept with respect to the assemblage.
distant stars. TL Thermoluminescence – a heat (and light)-sensitive
Single-aliquot/single-grain A portion of sample that is signal that builds up in nature during a burial period and is
placed on a stainless-steel disc for measurement – multiple measured under laboratory conditions using heat.

Abstract

This chapter provides researchers with a guide to some of the types of dating techniques that can be used in geo-
morphological investigations and issues that need to be addressed when using geochronological data, specifically issues
relating to accuracy and precision. This chapter also introduces the ‘types’ of dating methods that are commonly used in
geomorphological studies. This includes sidereal, isotopic, radiogenic, and chemical dating methods.

14.30.1 Introduction before sample collection often results in a confusing errone-

ous age that is difficult to interpret.
To fully grasp the intricate nature of the landscapes that we The second question is ‘which method is most appropriate
investigate as geomorphologists, an accurate understanding of for dating the chosen event, landform or depositional en-
geomorphological change through time is essential. This in- vironment in my specific geomorphological situation?’ The
volves establishing the long-term rates of geomorphological answer to this question is complicated by numerous factors
processes and the age of specific events, sedimentary units, and including the type of material that is available for dating (e.g.,
depositional environments. Developing an accurate chrono- bone, sediment, charcoal, flowstone, shell, etc.), the age of the
logical framework for landscape development and processes material being dated, the integrity of the material as an ac-
provides a context for how landforms and depositional en- curate indication of the age of the landform, event, or deposit
vironments have responded to variations in external factors that is being dated, the resolution required, and the geological
such as sediment supply, geological and tectonic processes, and setting. As no one dating method can be universally applied to
variability in climate. Armed with detailed landscape evolution provide an age estimate of landscape evolution, careful con-
models, geomorphologists are also able to make informed as- sideration of the most appropriate dating method for the
sessments of how specific landforms will change in the future. specific situations is needed. Figure 1 provides a summary of
the more commonly used dating methods and their age ranges.
After selecting the most appropriate dating method, the
next question that must be addressed is ‘what does the age
14.30.2 Dating Issues determination represent?’ As each dating method has its own
distinctive age range, limitations, and inherent uncertainty,
Establishing an age of specific landforms, deposit, or event, or there are always questions of ‘precision’ and ‘accuracy’ that
providing a chronology for processes involved in the devel- need to be addressed (Figure 2). Precision refers to the stat-
opment of a specific landform is a complicated issue. Often, istical uncertainty that is associated with either the physical or
field researchers submit samples for dating without fully the chemical measurements related to the analytical process of
grasping the issues associated with establishing a dating the dating technique (Lowe and Walker, 1997; Williams et al.,
strategy before sample collection, the dating method being 1998; Noller et al., 2000). This is, in other words, the meas-
used, and the integrity of the age determination in relation to ured uncertainties associated with the analytical technique.
the actual age of the landforms or deposits that are being This is usually represented as 7 uncertainties or an age range.
dated. The first question that needs to be addressed is ‘what is It is a fundamental requirement that the precision of age as-
being dated?’ For example, the timing of deposition of a sessment be reported.
sedimentary unit or the age of specific landform or even the Once the numeric age has been derived (precision), the re-
death of an organism, the formation of flowstone, etc. De- searcher needs to assess how ‘accurate’ that age determination is
fining the desired landform or specific event for dating is the in relation to the landscape, event, or depositional environment
first step toward establishing a dating strategy and choosing that is being dated. In other words, the degree of confidence
an appropriate dating technique. This may sound overly that any given age determination has is in relation to the actual
simplistic but not clearly establishing this consideration age (or true age) of the material being dated or the relationship
348 An Introduction to Dating Techniques: A Guide for Geomorphologists

* * * * * * Volcanic Dendrochronology
* Glass Radiocarbon
* * * Obsidian
Unburnt sediment Potassium-Argon
* *
* * Burnt flint and stone Uranium-Series
* * * Slag Fission Track
* * * * Pottery, baked earth
Precipitated calcite Luminescence
* * * *
* * * * Shells ESR
* * * Tooth enamel
Amino Acid Racemisation
* * * Bone antler, ivory, teeth
* * Wood, plant, seeds Obsidian Hydration
Magnetic Polarity
M
H net
Am rat m
ES ino on
Lu acid
Fi in
U io ce
Po niu tra ce
R ass se s
D io m ie s
yd is

ra n n

a d iu r
en ca -a
ag

ss es
m

t m ck
R

d r rb rg o
oc on n
i

hr 0.01 0.1 1 10 100 1 10 100

on
ol
og Time a Time ka Time Ma
y

Figure 1 Materials and age ranges for the main geochronological techniques. (Left panel) Materials used for dating. A star denotes a
suitable material for that dating technique. (Right panel) Age range of the chosen dating techniques. The x-axis represents time and is a
logarithmic scale (in years (a), thousands of years (ka) and millions of years (Ma)).

if attempting to obtain an age of a sedimentary deposit, it is

critical to assess the material being dated from that deposit
and whether it represents the actual age of the deposit, event,
(c)
or landform. Older sediment, wood, or carbonate material can
be reworked into younger deposits, giving an older apparent
age. In contrast, infaunal organisms can burrow into older
deposits, disturbing the sedimentary layers, giving an appar-
ently younger age (see Chapters 14.9, 14.11, and 14.12).

(a) (b)
14.30.2.2 Taphonomy of Fossil Material
12 11 10 9 ka Organic remains are commonly utilized to provide a chron-
ology for sedimentary succession and landscape evolution.
"TRUE Making an accurate assessment of the processes involved
AGE" once an organism dies is fundamental to understanding age
Figure 2 An example of the difference between precision and biases that may be represented in the fossil record. There are
accuracy. (a) Age estimates are precise but inaccurate as they are 2 ka numerous factors that need to be considered:
too old. (b) Age estimates are accurate as they are close to the true
age but imprecise. (c) Age estimates are accurate and precise. • Orientation: After death, skeletal remains and fossil ma-
Redrawn from Lowe, J.J., Walker, M.J.C., 1997. Reconstructing terial can be moved by transporting medium and their
Quaternary Environments. Second edn., Longman, London, pp. 446, resting orientation is determined by their hydrodynamic
with permission from SAGE. properties. The organism’s orientation will give an indi-
cation of whether the organism is in life position (in situ)
of that dated material with the landscape or depositional en- or has been transported and represents a death assemblage
vironment being studied (Lowe and Walker, 1997; Williams (thanatocoenosis).
et al., 1998; Colman et al., 1987; Noller et al., 2000; Walker, • Articulation: Upon death, multi-element skeletons (e.g.,
2005). Factors that need to be taken into account when as- vertebrates and bivalves) soon become disarticulated
sessing the accuracy of a specific age determination include the (broken apart). Disarticulated fossils generally represent
following (Sections 14.30.2.1 and 14.30.2.2). postmortem transportation and can therefore be non-
synchronous with the time of deposition. In contrast,
if articulated skeletons or bivalves are recovered from the
14.30.2.1 Depositional Context
sedimentary record, then this may indicate rapid burial or
This involves careful assessment of the material being dated in that the organism(s) were preserved as a life assemblage
relation to the landform or the deposit. In other words, is the (life position) otherwise known as biocoenosis (interacting
material being dated synchronous with the land surface or organisms living together in a habitat) and thus represent
deposit? Or, is the material allochthonous (i.e., transported synchronous deposition and preservation with the sedi-
from where the material was originally formed)? For example, ments that they are preserved in.
 An Introduction to Dating Techniques: A Guide for Geomorphologists 349

• Erosion: Modification due to transport by agents of erosion age from 1950 AD. Due to past variations in atmospheric
also provides evidence of postmortem transport. This in- radiocarbon concentration, radiocarbon ages that have to be
cludes abrasion, rounding, and fragmentation. The degree calibrated to sidereal years are reported as ‘cal. yr BP’ or ‘cal. yr
of these modifications can also be used as an indication of BC/AD’ (see Section 14.30.6). The age determinations must be
energy levels experienced during transportation or the reported with their associated uncertainty term so that the re-
duration of postmortem transport. sults reflect an age range rather than just the median or the
• Weathering: The degree of weathering or dissolution of mean value. Their associated uncertainties are reported as plus
fossil material can also give an indication of how appro- and minus values (7) for example, 4572 ka.
priate the material is for dating. Strongly weathered ma-
terial generally represents exposure before incorporation
into a sedimentary unit. Thus, weathered and degraded
14.30.3 Dating Methods
material that has undergone postmortem transport or
altered the chemical composition can significantly influ-
Relative dating refers to the relative order of past events,
ence the results of age determinations that are reliant on
stratigraphic units, or landform development without deter-
chemical reactions (e.g., amino acid racemization).
mining their numeric (absolute) age. There are several prin-
• Postdepositional diagenesis and chemical alteration:
ciples geologists and geomorphologists can use to determine
Following deposition, and as part of the process of lithifi-
relative age. These methods of establishing relative age are
cation, sediments undergo compaction, dewatering, ce-
based on the fundamental laws of geology such as the law of
mentation, and changes in temperature and chemistry.
horizontality, superposition, lateral continuity, faunal suc-
Fossil material appropriate for dating will also undergo
cession, cross-cutting relationships, and inclusion. However,
these physical and chemical processes and can alter the
relative dating can only determine the sequential order in
accuracy of any age determination (e.g., radiocarbon dating
which a series of events occurred, not when they occur.
and amino acid racemization).
Establishing a specific time of landscape development, an
event, or depositional environment requires numerical age
Any age determination comes with an inherited bias depending
determinations. Numeric dating is the process of producing
on an array of physical and chemical conditions and in as-
a quantifiable age and uncertainty for a specific sample that
sessing the reliability of the date. It is fundamental that both
is being dated (narrow numerical age range). These dating
precision and accuracy be addressed when discussing the rele-
methods are also known as ‘absolute’ dating methods. How-
vance of any age determinations, for example, a particular
ever, the word ‘absolute’ implies a certainty and accuracy that
dating method can provide a very precise age determination
is rarely possible in geochronological studies, and following
(i.e., a numerical age with a known uncertainty term). However,
the recommendations of Colman et al. (1987) and Noller
how accurate that age is to the true age of the sample or the
et al. (2000), we suggest that the term ‘numeric age’ be used.
deposit that the sample is recovered from may not correspond
Similarly, the use of ‘date’ should be minimized in favor of ‘age’
to that age. If such factors outlined above are not taken into
or ‘age estimate’ or ‘age determination’ (Noller et al., 2000).
account, the reliability (accuracy) will be poor (Figure 2).
There are numerous dating methods that can be applied to
The next question is ‘how can the age estimate be used?’. In
provide an age assessment for a specific event or deposit, and
other words, how are the results interpreted to estimate the age
to accurately determine landscape change over a variety of
of the landform, event, or depositional environment being
time scales. It is not within the scope of this chapter to provide
dated and how are the age determinations to be used to infer
a detailed explanation of all the dating methods. This chapter
the age of the associated material? For example, age estimates
introduces the ‘types’ of dating methods that are commonly
obtained for a specific depositional environment, landform, or
used in geomorphological studies. This includes sidereal,
event can be interpreted as equivalent in age if the depositional
isotopic, radiogenic, and chemical dating methods. Selected
environment, event, or landform being investigated is clearly
examples have been included to illustrate the issues that need
defined, if samples for dating have been strategically collected
to be considered when undertaking geochronological studies.
to constrain the age range, and there are no obvious systematic
For a more compressive review on the specific examples
errors such as partial bleaching, diagenesis, or postmortem re-
of different dating methods in these categories, see Rutter et al.
working. In addition, the resulting age estimates of fossils or
(1985), Noller et al. (2000), and Walker (2005).
sediments can also be used to infer the age of sedimentary
layers or fossils in the sediment by direct association.
Finally, the researcher must report the age determinations
in a consistent and accurate manner. Reporting ages should 14.30.4 Sidereal or Incremental Dating
be based on the North American Commission on Stratigraphic
Nomenclature (1983) and the IUPAC–IUGS Task Group This method of age determination is based on regular
2006-016-1-200 summarized in Colman et al. (1987), accumulation of sedimentary deposits (varve chronology),
Noller et al. (2000), and Rose (2007). They recommend SI addition of organic material (dendrochronology), or a min-
abbreviations of ‘a’ or ‘yr’ for years, ‘ka’ for thousands of years, eral phase onto organic material (sclerochronology). The
and ‘Ma’ for millions of years of age from present. With this regular accumulations are controlled by seasonal variation
notation there is no need for the addition of ‘before present’ or and can be used to count annual events. In this sense, this
‘ago.’ However, in the case of radiocarbon dating, age de- approach to dating provides an actual ‘date’ or a specific point
terminations are conventionally expressed as yr BP to indicate in time (i.e., calendar event). However, as this dating method
350 An Introduction to Dating Techniques: A Guide for Geomorphologists

is based on contemporary processes, the limit of this dating may have a history of exposure, burial, and subsequent reex-
method is usually within the Holocene. posure. For a more detailed review of cosmogenic dating and
application geomorphic problems, see Cockburn and Sum-
• Varve chronology is based on the cyclic accumulation of
merfield (2004).
fine sands alternating with silts, clays, or organic material.
These alternating layers (also known as rythmites) are the
result of seasonal accumulation of fine (winter months) 14.30.6 Radiocarbon Dating
and coarser (summer months) sedimentary successions
deposited mostly within alpine and glacial lakes. The se- Radiocarbon dating is one of the most reliable and well-
quence of varve layering can then be related to the calendar established methods for dating the past B50 ka. WF Libby
time scale (Lowe and Walker, 1997; Palmer et al., 2010) demonstrated the potential of radiocarbon dating in the mid-
and also provide calibration for numeric dating methods 1940s, and later received the Nobel Prize in Chemistry for this
(Prasad et al., 2009). discovery. To date, over 150 radiocarbon laboratories world-
• Dendrochronology is based on the tendency of most trees wide provide tens of thousands of radiocarbon age assess-
in nontropical environments to produce annual growth ments every year (Geyh, 2005; Hua, 2009).
rings (tree rings or annual rings) that result from the
change in the growth speed of the tree through the seasons 14.30.6.1 Principles of Radiocarbon Dating
of the year; thus, one ring generally marks the passage of
1 year in the life of the tree (Lowe and Walker, 1997; Radiocarbon or 14C is one of the three natural isotopes of
Friedrich et al., 2004). carbon. The other two isotopes are 12C and 13C, which are
stable with relative abundances of B98.9% and B1.1%, re-
• Sclerochronology is based on the number of annual
growth rings occurring in corals. These growth rings are the spectively. Meanwhile, 14C is radioactive and occurs only in
result of physical and chemical variations that result in the very small amounts (e.g., B1.2 1010% in the atmosphere).
accretion of a mineral phase (generally calcium carbonate) Radiocarbon is produced continuously in the upper atmos-
to protect the soft-bodied organisms. Other organisms that phere as a result of a nuclear reaction between neutrons
secrete calcium carbonate exoskeletons can also be used in produced by cosmic rays and atmospheric 14N. Following
the style of sidereal dating (Buddemeier et al., 1974; production, 14C is oxidized to form 14CO2, which is quickly
Knutson et al., 1972; Halfar et al., 2008). dispersed throughout the atmosphere. The 14C is then trans-
ferred to the oceans through air–sea exchange of CO2 and
enters living organisms (e.g., plants and animals) via photo-
14.30.5 Isotopic: Change in Isotopic Composition synthesis, food chain, and metabolic processes. Once the 14C
uptake ceases (e.g., at the death of an animal or when wood
Isotopic dating methods are those that measure the isotopic cellulose is formed), the 14C concentration of the organism
composition of a material as a result of radioactive decay. starts to decrease by radioactive decay at a rate, which is only
These are isotopes produced as a result of the bombardment controlled by the radiocarbon half-life. The radiocarbon age t
of the Earth by high-energy gamma or cosmic rays. However, of the organism is determined by:
unstable isotopes undergo changes in the atomic organization  
to become a more stable atomic form. Thus, the isotopic T1=2 NðtÞ
t¼  ln ½1
composition of a material and the concentration of radio- ln2 N0
isotopes change through time. If the rate of this change to a
more stable atomic form is known, then the age of the ma- where T1/2 is the radiocarbon half-life, N0 is the initial 14C
terial being dated can be established (Lowe and Walker, 1997; concentration in the organism, and N(t) is its residual 14C
Noller et al., 2000). Numerous elements undergo radioactive concentration at time t.
decay over a variety of time scales including U-series. Radio- In practice, atmospheric 14C concentration is assumed to
carbon dating (14C) is probably the most commonly used be constant through time and the initial 14C concentration of
numeric dating method for geomorphological research and is living terrestrial materials is conventionally set at one fraction
outlined in more detail below. modern carbon (F14C) in order to calculate radiocarbon
The isotopic dating method can also be based on certain conventional ages. From eqn [1], the conventional radio-
cosmogenic nuclides (e.g., 10Be, 21Ne, 26Al, 36Cl) that form carbon age of a sample S is defined as:
in situ in the upper few meters of the Earth’s surface since
T1=2
the cosmic ray flux is attenuated at greater depths. However, t¼  lnðF 14 CS Þ ½2
ln2
the accumulation of cosmogenic nuclides at the surface
and concentration in the upper few meters is controlled by the where F14CS is the 14C concentration in sample S after nor-
balance between the production rate of the isotope, its decay, malization to d13 C¼  25% PDB (an average value of
and the rate of erosion (Zreda and Phillips, 2000; Duller, C3 plants) to correct for isotopic fractionation. This value is
2000). This dating method enables age determinations to be determined by measuring sample S against 14C standard ref-
made on the time that a particular surface has been exposed erence materials such as NIST (National Institute of Standards
to cosmic rays. This provides the time frame in which and Technology) oxalic acid I or II, or ANU (Australian
the timing and rate of geomorphic processes can be assessed. National University) sucrose (Stuiver and Polach, 1977;
However, interpretation of cosmogenic nuclide results be- Donahue et al., 1990). T1/2 is the Libby half-life of radio-
comes very complex in environments where a surface carbon of 5568 years.
 An Introduction to Dating Techniques: A Guide for Geomorphologists 351

14.30.6.2 Sample Preparation and Radiocarbon 0–12 550 cal yr BP and on independently dated marine sam-
Measurements ples such as foraminifera and corals for the remaining period.
There is a small difference in the natural atmospheric 14C
Fossil skeletons recovered from sediments can be contamin-
concentration between the NH and the Southern Hemisphere
ated with other carbon-containing materials in the sur-
(SH). The SH has a larger surface ocean area than the NH
rounding environment, such as carbonates from groundwater,
(B60% compared with B40%), with greater wind velocities.
humic acids derived from decomposed organic materials, and
As a result, more 14C-depleted CO2 from the oceans (see
roots and rootlets that may have penetrated from upper
discussion later) is transported to the SH via air–sea exchange
horizons. For example, dissolved carbonate from groundwater
of CO2, causing an interhemispheric 14C offset, which is
can deposit on shell samples as calcite or recrystallization of
B40 years (with SH being older) but varies with time. The
shell carbonate can occur possibly due to chemical exchange
current internationally ratified radiocarbon calibration curve
between the sample and the surrounding environment. Re-
for terrestrial samples from the SH is SHCal04 (McCormac
crystallization and diagenetic alteration of carbonate samples
et al., 2004). This curve covers the past 11 000 cal yr, which is
can be detected using X-ray diffraction (Klug and Alexander,
based on the dendrochronologically dated tree rings for the
1954), thin section examination (McGregor and Gagan,
last millennium and on model ages for the remaining period.
2003), or scanning electron microscopy (Webb et al., 2007;
Calibration of 14C ages is usually undertaken using a com-
Nothdurft and Webb, 2009). Before samples are processed for
puter program such as CALIB, CalPal, and OxCal. More in-
dating, all contaminants must be removed using physical and/
formation on these and other calibration programs can be
or chemical pretreatments; otherwise, incorrect ages may be
found on the Radiocarbon journal website at http://www.ra-
determined (Hedges, 1992).
diocarbon.org/Info/index.html.
The 14C concentration in a sample was originally deter-
mined by detecting b-particles resulting from 14C radioactive
decay. For this b counting or radiometric method, 14C is 14.30.6.4 Reservoir Effects
measured using either gas proportional or liquid scintillation
Carbon reservoirs, which are different from the atmosphere,
counters (Taylor, 1987; Walker, 2005). These techniques usu-
experience reservoir effects. The surface ocean is in contact
ally use CO2 and benzene prepared from pretreated samples,
with the atmosphere and the 14C-depleted intermediate/deep
respectively. Another advanced method for measuring 14C is
ocean and has a 14C level intermediate between these two
accelerator mass spectrometry (AMS), which was developed in
reservoirs. This results in an apparent 14C age of surface ocean
the mid-1970s. The AMS method counts 14C atoms directly
samples (e.g., marine shells, planktonic foraminifera) several
(relative to those of the stable carbon isotopes 13C and 12C in
hundred years older than contemporaneous terrestrial sam-
the samples). Compared with the radiometric method, the
ples. Such apparent age is known as the marine radiocarbon
AMS method requires a much smaller sample mass (0.1–2 mg
reservoir age (R), which is B400 years on average. To calibrate
of carbon instead of 0.5–2 g of carbon for the former method)
a radiocarbon age for a surface ocean sample, one can use
and a much shorter measurement time (from tens of minutes
either a terrestrial calibration curve with a known value of R or
to a few hours rather than a few days for the former method)
the current internationally ratified marine calibration curve
(Tuniz et al., 1998; Jull and Burr, 2006; Walker, 2005). The
Marine09 (Reimer et al., 2009) with a known value of regional
AMS method usually uses graphite prepared from pretreated
offset from the ‘global’ marine model age for that sample,
samples and currently is able to reliably analyze samples as
defined as DR. The latter method is generally preferred and an
small as 10–20 mg of carbon (Hua et al., 2004; Santos et al.,
online database of DR for different regions is available (Reimer
2007; Smith et al., 2010).
and Reimer, 2001). For age calibration, DR and R of a location
are usually assumed to be constant through time (Stuiver
et al., 1986). However, recent studies have reported variations
14.30.6.3 Radiocarbon Calibration
in these values of several hundreds to a couple of thousands of
It was recognized that the 14C concentration of the atmos- years for several regions during Late Glacial and the Holocene
phere in the past had not been constant. Variability in at- (Bondevik et al., 2006; Franke et al., 2008; McGregor et al.,
mospheric 14C concentrations is mainly due to variations in 2008; Yu et al., 2010; Ortlieb et al., 2011). These variations are
the rate of radiocarbon production in the atmosphere, caused due to changes in ocean circulation and the carbon cycles
by changes in the Earth’s magnetic field and/or variability in associated with climatic change. Temporal variations in DR
solar activity, and changes in the carbon cycle (Taylor, 1987; and R values may therefore need considering when calibrating
14
Damon and Sonett, 1991). This implies that radiocarbon and C ages of marine samples.
calendar ages are not the same, and a conversion is required of Lake sediments and freshwater materials (e.g., aquatic plants
radiocarbon ages to the calendar ages using a calibration and freshwater shells) are also influenced by reservoir effects
curve, which describes the atmospheric 14C concentration in because a portion of the carbon in lakes may have come from
the past, measured in precisely and independently dated ma- depleted 14C–C sources, such as dissolved inorganic carbon
terials. The current internationally ratified calibration curve for from groundwater and carbonates from limestone (Deevey
terrestrial samples (e.g., wood, charcoal, and macrofossils) et al., 1954). As a result, these materials can appear older than
from the Northern Hemisphere (NH) is IntCal09 (Figures contemporaneous terrestrial samples. This reservoir age is of
3(a) and 3(b)), which covers the past 50 000 calendar years several hundred to more than a thousand years (Colman et al.,
(cal yr) (Reimer et al., 2009). This curve is based on 2000; Zoppi et al., 2001) and can change significantly with time
dendrochronologically dated tree rings for the period (Geyh et al., 1998). Radiocarbon ages from these materials
352 An Introduction to Dating Techniques: A Guide for Geomorphologists

50 000
1:1 line

40 000

Radiocarbon age (BP)

30 000

20 000 IntCa109 calibration curve

10 000

0
50 000 40 000 30 000 20 000 10 000 0
(a) cal yr BP

R_date (3000,40)
95.4% probability
Conventional radiocarbon age (BP)

3200 3335 (95.4%) 3070 caI BP

3000

2800

2600

3400 3200 3000 2800

(b) Calibrated date (caI BP)
Figure 3 (a) IntCal09 radiocarbon calibration curve (1s range) showing differences between radiocarbon ages and calendar ages for the past
50 000 cal yr BP. (b) An example of radiocarbon calibration. The calibrated age range of a terrestrial sample from the NH having a radiocarbon
age of 3000740 year BP is 3335–3070 at 2s (95.4% confidence level). The calibration was performed using the IntCal09 data set (Reimer et al.,
2009) and the OxCal program version 4.1.3 (Bronk Ramsey, 2009). Blue lines depict the IntCal09 calibration curve. The red curve indicates the
Gaussian distribution of the radiocarbon age determination. The gray histogram represents the probability distribution of the calibrated age.

must therefore be corrected for any reservoir effect before being slightly higher than its prebomb level. The main feature of
calibrated using a terrestrial calibration curve. atmospheric bomb 14C is that there are significantly different
atmospheric 14C levels between consecutive years during
the bomb period, offering the possibility of dating terrestrial
14
14.30.6.5 Bomb-pulse C Dating samples formed after 1955 by 14C with a resolution of one to a
few years (Figure 4). The most reliable materials for this
Nuclear bomb blasts, from hundreds of above-ground nuclear dating method are short-lived materials such as small twigs
weapon tests mostly in the NH in the late 1950s and the early and grass stems (Hua, 2009). This dating method is usually
1960s, produced intense fluxes of thermal neutrons, which in called bomb-pulse dating (for the interval from 1950 onward)
turn interacted with atmospheric 14N to form 14C. As a result, to differentiate from the traditional radiocarbon dating (for
the atmospheric 14C level reached a maximum in the NH in the period from 1950 backward) as mentioned previously.
1963–64, at almost double its prebomb level. Since then, the Four zonal data sets of tropospheric bomb 14C data at
atmospheric 14C concentration has been decreasing due to (mostly) monthly resolution (three in the NH and one in the
rapid exchange between the atmosphere and other carbon SH) are available as ‘calibration’ curves for use in bomb-pulse
14
reservoirs such as the biosphere and oceans, and now is C dating (Hua and Barbetti, 2004; Hua, 2009).
 An Introduction to Dating Techniques: A Guide for Geomorphologists 353

R_F14C(1.1555,0.0008)
1.6 95.4% probability
1959 (9.2%) 1959 caI AD

Radiocarbon determination (F14C)

1988 (1.2%) 1988 caI AD
1.5
1989 (85.0%) 1993 caI AD

1.4

1.3

1.2

1.1

1
1960 1970 1980 1990
Calibrated date (caI AD)
Figure 4 An example of bomb-pulse 14C dating. Radiocarbon measurement is reported in F14C instead of radiocarbon age for the traditional
radiocarbon dating (pre-1950 AD). For an F14C value of 1.1555 7 0.0008 measured in a terrestrial sample from the SH, bomb 14C delivers two
possible calendar dates, 1959 AD and 1988–93 AD, during the rising and falling arms of the bomb 14C curve, respectively. The calibration was
performed using the SH Zone data set (Hua and Barbetti, 2004) and the OxCal program version 4.1.3 (Bronk Ramsey, 2009). Blue lines depict
the SH atmospheric bomb curve. The red curve indicates the Gaussian distribution of the radiocarbon determination. Gray histograms represent
the probability distribution of the calibrated ages.

14.30.6.6 Applications of Radiocarbon Dating to radiocarbon dating, reported that short-term intervals of flu-
Geomorphology vial activity matched well with periods of North Atlantic
cooling (Figure 5). In addition, Zielhofer et al. (2008)
Radiocarbon dating has widely been used as an effective investigated influences of landscape changes on human be-
chronological tool in different scientific disciplines including havior and reported a correlation between periods of lower
geomorphology. Typical material for radiocarbon dating used human activity (based on the regional archaeological data of
in geomorphologic studies includes charcoal, wood, peat, and Mediterranean North Africa) and phases of increased flooding.
shell. For wood and charcoal samples, short-lived rather than Radiocarbon dating has been used to study the history of
long-lived species should be used for dating in order to avoid reef island evolution. Woodroffe et al. (2007) used 14C dates
the ‘old wood’ problem, an apparent age of wood when the on specific skeletal components of carbonate sediments (ra-
tree stopped receiving 14CO2 from the atmosphere at the time ther than those on bulk carbonate samples) to reconstruct a
of death. This apparent age may be up to several hundreds of robust depositional chronology of Warraber Island in Torres
years and become significant when the time span of a study is Strait, Australia (Figure 6), and to estimate rates of island
short (hundreds to a thousand of years or less). progradation. Radiocarbon dating has also been used to
Radiocarbon chronologies have been used for the study constrain the timing of significant events such as tsunamis
of landscape evolution of various landscape settings such (Bondevik et al., 2005; Scheffers et al., 2009), and ice retreat
as floodplains (Tipping et al., 2008; Zielhofer et al., 2008), and advance (Derouin et al., 2007). For example, Scheffers
stream and river valleys (Kidder et al., 2008; Mandel, 2008), et al. (2009) used 14C dates on boring bivalves occurring in
hillslopes (Clarke et al., 2003), coasts (Long et al., 2006; dislocated boulders and on mollusc shells and peat associated
Clemmensen et al., 2009), alpine environments (Favilli et al., with gravel deposits to estimate the timing of tsunamis along
2010), and loess environments (Rommens et al., 2007). Long the west coast of Ireland. Chronology of storminess variation
et al. (2006) used radiocarbon dating to document the gradual has been reconstructed by radiocarbon. De Jong et al. (2006)
inundation of a coastal wetland in southeast England after used mineral grain content in a well-radiocarbon-dated peat
B700–850 AD and the establishment of a saltmarsh. Clem- bog from a coastal zone in southwest Sweden as a proxy of
mensen et al. (2009) built a chronology of the evolution of aeolian sand influx to derive the storminess magnitude and
coastal dunefields in Denmark for the past 5000 years based frequency during the last 6500 years. Timings and scales of
on radiocarbon dating of peaty palaeosols and optically gully erosion have also been established using radiocarbon
stimulated luminescence dating of aeolian sand deposits. They chronologies (Schmitt et al., 2006; Panin et al., 2009). Re-
found that increased storminess during the wet/cool summer cently, Shellberg et al. (in prep.) studied recent alluvial gully
climate caused aeolian sand movement and dune formation. erosion in the woodlands of the lower Mitchell River in
The climate influence on landscape evolution was also re- northern Queensland, Australia. The authors determined the
ported by Zielhofer et al. (2008). The authors, while estab- ages of living Eucalyptus trees growing in and around alluvial
lishing the Holocene evolution of Tunisian floodplains using gullies using bomb-pulse 14C dating to define the timings
 354
Chronology Series Process Synthetic Age Sedimentation Bond et al.

An Introduction to Dating Techniques: A Guide for Geomorphologists

profile [ka cal BP] [mm/year] Haematite
[%]
[m]
0 0 1 2 3 0 10 20
Post medieval determination
5
(Maunder)
1
0.5
Late medieval climatic alternation
Medieval cooling (Oort) 4 2 North Atlantic
Early medieval warming force
Post-Roman crisis 1.8 3
Roman humic optimum 2.8
3.0 ka aridity 3.0 4
3
Mid- to late Holocene stability 3.2 5
5 4.6
4.7 ka collapse 6
4.7
Mid-Holocene stability 2d
6.0 7
6.6 ka aridity 2c 6.1
8
6.7
9
7.5
Early to mid-Holocene 2b 10
climatic optimum
10 11
11.8
12
Younger Dryas 2a 12.3
1 13
Late Pleistocene warming 13.4
14

Fluvial dynamics Sedimentation

River incisison (stable) Clay/silt/loam Soils Dating (subset)
Still water environment (stable) Sand A horizon Charcoal/humic acid
Flooding / aggradation (active) Gravel Gleysols Charcoal (fire-site)
Figure 5 Composite profile and average sedimentation rates of the mid-Medjerda floodplain in Tunisia showing a good correlation between short-term periods of fluvial activity and North Atlantic
coolings. Reproduced from Zielhofer, C., Faust, D., Linstädter, J., 2008. Late Pleistocene and Holocene alluvial archives in the Southwestern Mediterranean: Changes in fluvial dynamics and past human
response. Quaternary International 181, 39–54.
 An Introduction to Dating Techniques: A Guide for Geomorphologists 355

P10B P8A P5A P3B P1B RT2

Bulk Bulk Bulk Bulk Bulk Bulk

3627−4013 3698−4088 3426−3794 2493−2861 2321−2678 2441−2770

4335−4713 2723−2995 3553−4066 3354−3641 1082−1299 1159−1365

5291−5908 4617−5651 4513−5451 5592−6639

2463−2744 1272−1491 632−839 314−509 Modern

256−450
2322−2661 1614−1917 926−1162 287−493
490−637

P10B P8A P7B P6B P5A P4B P3B P2B P1B RT2

Bulk Q39
3177−3473 Modern
4180−4583
WS9

N
C2
Modern
Elevetion (m)
6.0−8.0 B2
5.6−6.0 0−236
5.3−5.6
5.0−5.3
1.2−5.0
A2
0 100 200 m 474−632

BT3 BT2 Q1A1

42−286 Modern Modern
736−1012 1414−1732 2064−2321
Figure 6 Calibrated radiocarbon ages on different specific skeletal components of carbonate sediments and bulk carbonate samples from
Warraber Island, indicating that the most reliable depositional chronology for the island was based on 14C dates of mollusc shells, dominant
species of the Island. All ages are reported with range of 2s in cal yr BP. Dates are in plain text for corals, in italic for foraminifera, and in bold
for shells. Reproduced from Woodroffe, C.D., Samosorn, B., Hua, Q., Hart, D.E., 2007. Incremental accretion of a sandy reef island over the past
3000 years indicated by component-specific radiocarbon dating. Geophysical Research Letters 34, L03602. doi: 10.1029/ 2006GL028875, with
permission from American Geophysical Union.
356 An Introduction to Dating Techniques: A Guide for Geomorphologists

and estimate rates of gully erosion. Collectively, these studies 1998; Stokes, 1999a, b). In addition, they can provide a
illustrate the diversity of applications of the radiocarbon timescale for key paleoenvironmental proxies such as river
method in geomorphological investigations. terraces, an estimation of the rate of long-term geomorphic
processes such as soil erosion and river incision, and
the timing of rapid geomorphic events such as storm surges
14.30.7 Radiogenic: Luminescence Dating and tsunamis. In this respect, the techniques have mostly
impacted on sediment-based disciplines such as archeology
Luminescence techniques (both thermoluminescence (TL) and and paleontology, and have become very powerful tools in
optically stimulated luminescence (OSL)) measure when sedi- geomorphological research (Stokes, 1999a, b). Depositional
ments were either last heated (e.g., burnt or fired, Aitken, 1985) chronologies have been established for up to 800–1.5 Ma al-
or exposed to sunlight during transport or deposition (e.g., by though B200 ka is more typical and for samples as young as a
wind or water; Huntley et al., 1985) via the stimulation of few years (Aitken, 1998). Ages are obtained in calendar years
minerals using heat (TL) or light (OSL). These methods are also with typical precisions of 5–10%.
termed ‘electron capture methods,’ whereby a light-sensitive It is not within the scope of this section to provide a com-
signal is ‘trapped’ in defects in the crystal lattice of minerals prehensive overview of luminescence and all its applications;
such as quartz and feldspar (Figure 7(1), Irradiation) and is researchers are directed to more comprehensive reviews in
evicted during laboratory measurement (Figure 7(3), Eviction). Wintle (1997), Aitken (1998), Stokes (1999a, b), Duller (2004),
This signal builds up slowly and is stored in the crystal during a Lian and Roberts (2006), and Lian (2007a, b). Instead, this
period of burial in the sediment column (Figure 7(2), Storage). section will provide a starting point for potential users wanting
The amount of trapped energy is proportional to the amount of to establish chronologies for geomorphological research. For the
naturally occurring ionizing radiation (Figure 7(1), Irradiation) nonspecialist this jargon-rich discipline can often prove in-
occurring in the surrounding sediment and the amount of accessible; hence, we aim to direct further reading to four key
exposure time during the burial period (Aitken, 1985). areas: (1) resetting of the luminescence signal in different sedi-
As the signal is reset to zero by sunlight, it represents a clock mentary environments; (2) the main technological develop-
to measure the time since last exposure (Duller, 2004). This ments that have increased its usefulness in geomorphological
makes these techniques unique among dating methodologies research; (3) the impact on key geomorphological issues; and
as they can be used to estimate the time elapsed since sediments (4) practical issues for nonspecialist sample collection.
were buried (Stokes, 1999a, b), and as most processes or events
are based on the deposition of sediment, depositional ages are
14.30.7.1 Resetting of the OSL Signal in Different
critical in geomorphological research. See Aitken (1998) for a
Sedimentary Environments
more detailed explanation of the luminescence process.
Luminescence techniques are versatile and widely applic- There are four processes that adequately bleach or reset the
able and useful techniques for establishing the depositional luminescence clock to zero and are known as the resetting
age of noncarbon-bearing sediments. This versatility stems event: (1) formation by crystallization (e.g., precipitation,
from the use of both quartz and feldspar minerals, which are magma); (2) extreme pressure; (3) heating above 200–300 1C;
derived in a wide range in most landscapes worldwide and and (4) exposing the crystal to light (known as ‘bleaching’)
enable ages to be produced in a wide range of sedimentary (Duller, 2004). The latter is the basis for OSL dating.
environments, for example, aeolian (dune and loess), fluvial, A major assumption is that the sediment has been well
marine, coastal, lacustrine, cave, glacial, and colluvial (Aitken, bleached before deposition. According to Godfrey-Smith et al.

1 2 3
on
si
ffu
di

Electron Conduction band Light

E
T T T

ionisation L
Energy

L L
Light
Hole
no

Valence band
si
ffu
di

Irradiation Storage Eviction

Figure 7 Basic principles of optically stimulated luminescence dating, redrawn from Aitken (1998). (1) Irradiation – Ionization due to natural
radiation causes trapping of electrons (closed circles) and holes (open circles) at defects in the crystal lattice, known as traps (T) and
luminescence centers (L) that are stored over the burial period. (2) Storage. (3) Eviction – The electrons are evicted by stimulation with light or
heat and reach luminescence centers where luminescence is emitted in the form of light.
 An Introduction to Dating Techniques: A Guide for Geomorphologists 357

(1988) this can be as little as 1 min of direct sunlight more logical in its application to sedimentary deposits. In
to reduce OSL levels to below 5% of the initial values. This addition, OSL contains more light-sensitive traps; thus, there
is easily achieved in aeolian environments, where the grains is more opportunity for the signal to be reset during geo-
are exposed to sunlight for long periods during transport morphological processes, and does not leave a residual
and deposition (Aitken, 1998; Singarayer et al., 2006) unbleachable component (Figure 9). Furthermore, the meas-
(Figure 8(a)). However, bleaching opportunities vary widely urement of the OSL signal in the laboratory creates a greater
for different sedimentary environments, for example, bleach- signal to noise ratio with minimal background interference,
ing is more difficult in colluvial environments, where sedi- which makes it possible to date much younger samples
ments are deposited rapidly (Wu et al., 2010), or in fluvial (Stokes, 1999a, b).
environments, where the light is diffracted by the water,
for example, 3 h for 12–14 m of water depth (Berger,
14.30.7.3 Advances in Methods of Analysis – Single
1990; Rendell et al., 1994; Rhodes and Pownall, 1994)
Aliquot and Single Grain
(Figure 8(b)). Bleaching occurs during geomorphic processes
such as transport, erosion, and deposition (Aitken, 1998); Traditional OSL and TL techniques used multiple grains to
therefore, the degree of bleaching is dependent on the dom- measure trapped signal during burial (e.g., Huntley et al.,
inant processes occurring in each sedimentary environment. 1985) (Figure 10(A4)). In sedimentary environments that are
The potential for signal resetting in these environments will exposed to sunlight for long periods before burial, such as
need to be considered when collecting samples and inter- deserts, the grains contain a similar amount of burial signal
preting the resulting age estimates. (Stokes, 1999a, b). However, in other sedimentary environ-
ments, such as fluvial, marine, cave, coastal, estuarine, and
lacustrine, the amount of sun exposure before burial is either
14.30.7.2 Technological Developments that have made
insufficient or nonuniform, resulting in grains with varying
Luminescence Techniques More Applicable to
amounts of residual plus burial signal. Standard multiple-
Geomorphological Research
grain OSL techniques are unable to differentiate between these
In the 1970s and the early 1980s TL was originally applied to grains (Duller, 1995; Adamiec, 2005) (Figure 10(A4)).
flint, burnt stones, and pottery (e.g., Valladas, 1982; Valladas However, methodological and technological developments
et al., 1987) with limited geomorphological application. now enable the routine measurement of individual sand-sized
But in the mid 1980s and 1990s, luminescence techniques grains; thus, a statistically significant number of independent
underwent rapid revolution with the development of OSL estimates of age can be obtained, thereby increasing precision
(Stokes, 1999a, b). This methodological advance meant (Roberts et al., 2000) (Figure 10(A4)). This has allowed issues
that the light-sensitive signal being measured was reset by such as insufficient sunlight exposure (Singarayer et al., 2006)
sunlight rather than heat (Huntley et al., 1985), which is and postdepositional reworking and mixing (e.g., Roberts

Well-bleached Partially-bleached
environments environments
Bleaching as subglacial
material brought to surface and
Bleaching during during glacio-fluvial transport
transport and
deposition
in shallow water
Glacial
environments
Bleaching during
transport, erosion
Lake and deposition
environments in shallow water
Fluvial
environments Cave
environments
Rapid transport
and deposition - limited
opportunity for bleaching Bleaching during
deposition in fan

Aeolian environments Colluvial

environments Transport occurs underground
- bleaching during deposition
and reworking
Well bleached during
transport and deposition

Bleached through the Continental

water column in shelf
shallow water
Marine
environments
Limited bleaching in
deep water Deep sea

(a) (b)

Figure 8 The potential for well-bleached and partially bleached sediments in different sedimentary environments. (a) Aeolian environments are
well bleached (red dots) as bleaching occurs during wind-blown transport and deposition with little residual luminescence signal. (b) In
comparison, lacustrine, colluvial, fluvial, cave, and glacial environments contain a mixture of partially bleached (blue dots) and well-bleached
grains as deposition occurs either rapidly (e.g., colluvial), subaerially (e.g., in caves or under glaciers), or underwater (e.g., fluvial, lacustrine and
marine), producing large residual luminescence signals that result in an overestimation of age.
358 An Introduction to Dating Techniques: A Guide for Geomorphologists

Thermoluminescence geomorphological situations. For example, the use of red

luminescence (600–650 nm) rather than the more tradition-
Level before transport
ally used blue/UV luminescence (400–500 nm) has permitted

Bleaching during
application to volcanic deposits such as ignimbrites, lava

transport
flows, and landforms within volcanic provinces such as
river terraces. These volcanic deposits were previously difficult
TL signal The 'natural' to date due to their weak blue emissions, but instead, the
signal red TL methodology utilized their strong volcanic red emis-
sions. Westaway et al. (2009, 2010) used red TL techniques
Stimulated
Level of residual TL to provide a chronology for a series of stacked river terraces
by heat in
laboratory in Flores, Indonesia, enabling the rate of river incision to be
calculated and the rate of landscape evolution to be estimated
(Figure 11). Using red TL, Fattahi and Stokes (2000) were
able to establish a chronology for volcanic products in
Years of burial
New Zealand that extended back B1.6 Ma, which is far be-
yond previous luminescence ages and has created new op-
Laboratory portunities for establishing long chronologies for volcanic
Deposition
(a) measurement sequences.

Optically-stimulated luminescence

Level before transport 14.30.7.5 Application of OSL to Key Geomorphological

Bleaching during transport

Issues
Luminescence techniques have revolutionized many long-
standing geomorphological issues (Duller, 2004) such as
OSL signal The 'natural' the link between landscapes and climates (Figure 11(a)), the
signal source area of sediments, the rate of processes (Figure 11(c)),
and the timing of major events (Figure 11(b)). For example,
Stimulated OSL chronologies for aeolian activity in northern Australia
by light in
have enhanced our understanding of desert environments and
laboratory
their response to climate change (e.g., Nanson et al., 1992).
No OSL residual Hardy and Lamothe (1997) used the intensity of lumi-
nescence to determine the provenance of sediment, whereas
Years of burial
Hashimoto et al. (1986) used the different luminescence
wavelengths of individual quartz grains to establish the
Deposition Laboratory source area of samples. OSL chronologies have been used to
(b) measurement establish the rate of geomorphic processes such as soil erosion
Figure 9 Basic principles of thermoluminescence (a) versus optically (e.g., Fuchs et al., 2010), river incision (e.g., Westaway et al.,
stimulated luminescence dating (b), the difference being the amount of 2009; Wang et al., 2010; Martins et al., 2010), denudation (e.g.,
residual luminescence after bleaching of the TL signal compared with Thamo-Bozso et al., 2010), desertification (e.g., Zhao et al.,
the minimal residual OSL signal. This makes the OSL technique more 2007; Lomax et al., 2003), neotectonics (e.g., Porat et al.,
useful for establishing ages estimates for sediment deposition. 1996; Zilbermann et al., 2000; Choi et al., 2009), and land-
Redrawn from Stokes, S., 1999b. Optical dating of young (modern) scape evolution (e.g., Westaway et al., 2010). For example,
sediments using quartz: results from a selection of depositional Porat et al. (1996) applied OSL dating to colluvial and alluvial
environments. Quaternary Science Reviews 11, 153–159. sediments to constrain tectonic activity in Arava valley in Israel
to between 14 and 37 ka and used this chronology to establish
et al., 1998) to be addressed and enabled ages to be derived the rate of Quaternary displacement. In addition, OSL
for complex sedimentary environments such as rockshelters chronologies have also been used to establish the timing of
(Jacobs et al., 2003), colluvium (Wu et al., 2010), and glacial major events such as stormsurges (e.g., Reimann et al., 2010),
deposits (Narama et al., 2007; Kondo et al., 2007; Ou et al., floods (e.g., Huang et al., 2007), tsunamis (e.g., Banerjee et al.,
2010). Single-grain techniques can be efficiently conducted 2001; Cunha et al., 2010), earthquakes (Porat et al., 1996;
with a laser attached to the luminescence reader (Duller and Fattahi et al., 2010), hurricanes (e.g., Davids et al., 2010), and
Murray, 2000). volcanic eruptions (e.g., Banerjee, 2010). For example, Fattahi
et al. (2010) were able to estimate the timing of the last
earthquake along the Dehshir fault, Iran, to B2 ka using al-
luvial and colluvial sediments that filled fissures after the
14.30.7.4 Emission Spectra – Use of Different
rupturing of the ground surface, whereas Cunha et al. (2010)
Luminescence Wavelengths
favorably compared the timing of a known tsunami event on
The use of different luminescence spectra or wavelengths has the Algarve coast, southern Portugal, with an OSL chronology
increased the potential application of luminescence dating to for tsunami-laid sand and gravel deposits (Figure 11(b)).
 An Introduction to Dating Techniques: A Guide for Geomorphologists 359

Estimating palaeodose Estimating the dose rate

A1 Careful site selection A2 Collect sample for dating, Estimate Take an in situ
either tube, block or bulk changes to B1
Establish dating strategy measurement of
Log section B2 hydrology dose rate
and
Careful selection of sample
overburden
In the field locations Collect a dosimetry
over time
sample from around
the sample area
B3

A3 Laboratory treatment of the sample A4 Luminescence

to isolate material for luminescence measurements to Use sample for
measurements: most commonly quartz calculate the palaeodose B4 laboratory dose rate
measurements

Multiple aliquot Single-aliquot Single-grain Alpha/ beta High resolution NAA

.......
.......
....... counting gamma spectrometry
.......

Standardised Estimate
6 145
6 Calculate dose rate
113

Palaeodose (Gy)
In the 5 appropriate for the material
5
laboratory 4 2 81
B5 used taking into
4
0 consideration site
3 3 65 conditions
-2
2 2 49
1 1 Relative Error (%) 41
12 6 4 3 B6 Dose rate
0
-200 -100 0 100 200 300 0 100 200 300 400 0 9 18 27 36
Dose (Gy) Dose (Gy) Precision
Dose response curve
Dose response curve Radial plot AB Calculation of luminescence ages
- interpolation
- extrapolation
Age (yrs) = Palaeodose
A5 Palaeodose Dose rate
(a) (b)

Figure 10 Flow chart for OSL sampling, processing, and analysis procedures. The chart has been vertically divided into procedures to estimate the
paleodose and dose rate, and horizontally divided into procedures for the field and laboratory. The A/B numbers correspond to similar notations in
Figure 6. The measurements to calculate the paleodose have been divided into multiple- and single-aliquot and single-grain with examples of the
most common types of graphic presentation; dose–response curves and radial plots, respectively. To determine the paleodose, the multiple aliquot
results are extrapolated onto the minus dose axis, the single-aliquot results are interpolated onto the dose axis, whereas the single-grain results are
projected from the origin of the left axis (standardized estimate) through the data point to the radial dose axis on the right.

14.30.7.6 Practical Issues for Dating Strategies and 1. Tube sampling – conducted in soft sediments sections and
Sample Collection in daylight using light-proof plastic (or metal) tubing. This
is the easiest of all sampling methods and is recommended
The procedures involved in luminescence analysis are outlined for the inexperienced. The tube diameter is usually B5 cm,
in Figure 10. Sites for collecting OSL samples need to be cut into lengths of 20–30 cm with a cap on the end, and
carefully selected, considering the integrity of the stratigraphy is hammered horizontally into a cleaned vertical exposure
– evidence of reworking, bioturbation – from roots, burrowing using a wooden board. Newspaper is used as a reverse
animals, etc., the amount of quartz/feldspar in the deposit (as plunger to prevent mixing of the light exposed and
it will determine the amount of sediment required) – esti- unexposed sediments and to prevent the grains from
mated using geological maps, and whether the site is bouncing out of the tube during sampling. When the
suitable to constrain the dating event. To create a dating outside end of the tube is flush with the cleaned face, the
strategy, it is imperative to ensure that the stratigraphy is well tube is dug out using a sharp-ended trowel. The dug-out
established through detailed section drawing, sediment log- material is kept for radioactivity (dosimetry) measurements
ging, pictures, and stratigraphic interpretation. For the opti- and stored in a zip-lock plastic bag to retain soil moisture.
mum sampling locations, the following should be kept in Newspaper is placed in each end to ensure that the tube is
mind: (1) processing is time consuming and costly, and so, a packed solid to prevent movement and mixing of sedi-
‘whack a tube in and see’ mentality should be avoided and ments. The tube is then sealed using thick black plastic and
instead careful sampling should be performed; (2) if sampled PVC duct tape and clearly labeled. At least two tubes are
strategically, one age estimate can be used to constrain more filled (side by side) at each sampling location, three if the
than one event; and (3) large amounts of samples are rarely all sediment is known to contain low amounts of quartz
processed – it is better to collect fewer but more strategically (Figure 10(A2)).
placed samples – think conservative and strategically. 2. Block sampling – conducted in hard nonfriable cemented
Two types of samples are collected: for paleodose de- sediments and sampled in light conditions. This technique
termination and dosimetry measurements (Figure 10(A2, can be difficult without previous experience. A block of
B3)). For the former, samples are collected in one of three intact sediment is cut from a cleaned vertical exposure
ways depending on the composition of the sediment and the using a hammer and chisel, ensuring it does not crack.
circumstances at the chosen site: Hand saws have also been used in the case of mixed
360 An Introduction to Dating Techniques: A Guide for Geomorphologists

0
Palaeoenvironmental proxy Rate of processes
0.06+/-0.01
Merty Merty 1
2.3+/-0.2
Upper trench 4.4+/-0.4
2
2.1+/- 0.2 44+/-4
Merty Merty 46+/-8
3
lower trench 58+/-6
0 511 ka
35+/-3 75+/-9 61+/-6 4 620 m
1 T1
420 ka 420 ka
0 1 2 3 4 5m 585 m 589 m
2m 0 T2 373 ka 567 m
0 1 2
22+/-2 Lomax et al. (2003)
Clayey sand (fluvial-lacustrine) 1
48+/-5
Clayey sand (alluvial) 43+/-4 Wae Racang 250 ka
2 510 m
Coarse grained sand River Valley
(fluvial lacustrine) 50+/-5
62+/-6 3 T3
Aeolian sand with roots T3
69+/-9
Aeolian sand
4m T4
(a) Westaway et al. (2009) T5

Timing of events
Boca Do Rio (southern trench) 510 m
0
Silt
Very fine sand
100 Medium sand
492 m
Silt
Depth (cm)

Sandy Mud Tsunami Layer (AD1755)

200 T3
OSL
Radicarbon 487 m 18 ka −1
230 mm/ka
300 Medium sand MAAD OSL T4
480 m 5 ka −1
538 mm/ka Alluvial terrace
T5 downcutting rate
400 0 ka −1 −1
1200 mm/ka = 305 mm/ka
474 m
Cunha et al. (2010)
0 500 1000 1500 2000
1755
Years AD
(b) Tsunami (c)

Figure 11 Examples of luminescence applications and impacts on geomorphological issues. (a) Luminescence techniques applied to
paleoenvironmental proxies in South Australia. Lomax et al. (2003) used OSL dating to establish the timing of dune building and stability related
to climatic changes in the Strezlecki Desert. (b) Luminescence techniques used to establish the timing of significant events. Cunha et al. (2010)
used OSL dating to establish the timing of a known tsunami event at Boca do Rio on the Algarve coast (southern Portugal) by dating tsunami-
laid sands. (c) Luminescence techniques used to establish the rate of geomorphic processes. Westaway et al. (2009, 2010) applied TL
techniques to river terraces in Indonesia to estimate down-cutting rates for the WaeRacang River.

quartz-carbonate sands such as indurated aeolianites. The B30 cm from the sampling area, which represents the range
block is wrapped in black plastic in case of breakage during of gamma rays. Other on-site dose rate considerations include
transport. The outer light-exposed layers are removed in the estimations of moisture content changes over time (which
laboratory, while the inner light-tight core is used for requires an understanding of the local hydrology) and changes
processing (Figure 10(A2)). in the thickness of the rock/sediment overburden.
3. Bulk sampling – conducted in hard cemented sections For external processing, the tube/block/bulk samples
and sampled in the dark using red light. This technique retain their light-proof wrappings for transport, but the dos-
is not recommended for inexperienced practitioners. Using imetry samples need to be dried and wet weight/dry weight
a cleaned exposure, the sampling area is prepared in white measurements are required. Some techniques, such as high-
light and defined using metal pins. Under red light con- resolution gamma spectrometry, require that the samples are
ditions, the sampling area is then cleared of the light- ground down to fine silt before analysis.
exposed outer 2 cm of sediment, and then the unexposed A basic knowledge of the OSL technique is required for
inner layers are sampled for dating. The bulk sample is efficient sampling, and is essential for an understanding and
collected in a sample bag, covered in two layers of black interpretation of the results. Comprehensive reviews of the
plastic, and taped (Figure 10(A2)). physics and methodologies of luminescence can be found in
Lian (2007a, b) and Duller (2008). Luminescence results are
For dosimetry estimation – two sets of radioactivity generally presented in a tabular form (dosimetry data, paleo-
measurements are usually conducted – one in situ (in the doses, and resulting age estimates) along with the associated
field) (Figure 10(B1)) using a gamma spectrometer and the shine-down curves, growth curve, and radial plots; examples
other in the laboratory using either a beta/alpha counter of each of these data are provided in (Figure 12). Radial plots
combination, high-resolution gamma spectrometry, or neu- are the most commonly used plots as they provide a graphic
tron activation analysis (NAA; Figure 10(B4)). The in situ representation of the range of paleodoses determined for in-
measurements are conducted by inserting the probe into the dividual aliquot (single aliquot or single-grain) and therefore
hole created by the sampling procedure (using tubes, bulk, or are a useful means of distinguishing between different dose
block), while the laboratory measurements are conducted on populations to overcome the problems associated with partial
the dugout sediment (minimum 300 g) within a radius of bleaching.
 An Introduction to Dating Techniques: A Guide for Geomorphologists 361

TABLE X Examples of OSL dating of sediments: dose rate data, equivalent doses, and OSL ages
___________________________________________________________________________________________________________________________________________
–1 b
Sample Radionuclide activities (Bq kg ) Field gamma Total OSL
Cosmic-ray Water
Sample depth dose rate dose rate content dose rate Palaeodose age
–1 –1 (%) –1 (Gy) (ka)
code (m) 238
U 226
Ra 210
Pb 228
Ra 228
Th 40
K (Gy ka ) (Gy ka ) (Gy ka )
___________________________________________________________________________________________________________________________________________
Sector I
LBS1-12 8.40–45 - - - - - - 0.302 ± 0.009 0.075 20.8 / 22 ± 5 1.09 ± 0.07 479 ± 25 440 ± 36
60 ± 53 55 ± 49
LBS1-16 5.80–85 19.7 ± 2.6 21.8 ± 0.6 23.3 ± 3.1 18.4 ± 1.1 19.1 ± 0.8 109 ± 8 0.227 ± 0.007 0.086 5.6 / 5 ± 2 1.17 ± 0.06 406 ± 21 348 ± 26
61 ± 23 52 ± 20
Sector III
LBS3-2 7.40–53 12.7 ± 0.8 12.4 ± 0.2 14.4 ± 1.0 9.7 ± 0.3 9.4 ± 0.2 101 ± 3 0.286 ± 0.005 0.090 7.7 / 8 ± 3 0.81 ± 0.04 349 ± 13 430 ± 25
86 ± 25 106 ± 31
LBS3-4 4.91–96 18.8 ± 0.9 28.9 ± 0.2 34.7 ± 1.3 19.8 ± 0.4 19.3 ± 0.3 128 ± 3 0.272 ± 0.003 0.089 3.0 / 5 ± 2 1.40 ± 0.05 372 ± 19 266 ± 16
96 ± 17 69 ± 12
Sector IV
LBS4-30 6.45–50 9.9 ± 1.4 32.4 ± 0.4 35.6 ± 2.1 12.3 ± 0.6 11.7 ± 0.3 55.9 ± 3.5 0.205 ± 0.007 0.078 4.6 / 4 ± 1 1.04 ± 0.04 60 ± 4 58 ± 4
15 ± 3 14 ± 3
LBS4-26 6.49–54 14.3 ± 1.2 18.2 ± 0.2 32.7 ± 1.6 17.3 ± 0.4 17.0 ± 0.3 159 ± 4 0.257 ± 0.000 0.083 3.8 / 5 ± 2 1.40 ± 0.05 344 ± 9 247 ± 11
90 ± 35 65 ± 25

(a) A1 B3/4 B1 B2 B2 B5/6 A3/4 AB

245
205
165
125 12

85 10
Standardised estimate

65
8
Frequency of grains
Palaeodose (Gy)

45 6
2
0 35
−2 4
25
2
20

15 0
10 20 30 40 50 60 70 80 90 100 More
Dose (Gy)
Relative error (%) 10

24 12 8 6
5
0 7 14 21 28
(b) Precision (c)

Figure 12 Examples of the types and format of OSL results. (a) A typical OSL results table including dose rates, paleodose, and age
estimates. All results represent median values and are presented with a 1s uncertainty term in the form of 7 values. The letters and numbers
correspond to the same notations in Figure 5. (b) An example of single-grain data in the form of a radial plot from a river terrace in Indonesia
(Westaway et al., 2009), which displays a wide distribution of paleodoses. (c) The same single-grain data in the form of an OSL histogram. The
histogram illustrates the frequency of grains with a similar paleodose and suggests that at least two dose populations are present within this sample.

14.30.8 Time Dependent Chemical Reactions dependent amino acid racemization reaction as a geo-
chronological tool was first reported by Hare and Abelson
Numerous time-dependent chemical reactions can take place in
(1968). Since then, numerous studies have utilized the extent
sediments, rock, and fossil material. These include the diagenesis
of amino acid racemization observed in fossil carbonate ma-
of organic compounds (amino acid racemization), absorption
terial as a geochronological tool, with recent research focusing
of water (obsidian hydration), and weathering rates (Noller
on applying the dating method to solve local and regional
et al., 2000). In these processes, there is a chemical alteration
stratigraphical and geomorphological problems. Reviews by
that is based on time-dependent reactions and can provide a
Miller and Brigham-Grette (1989), Murray-Wallace (1993,
relative age assessment and, with effective calibration, these
1995, 2000), and Wehmiller and Miller (2000) provide
dating methods can also be utilized to provide numeric age
detailed assessments of the amino acid racemization dating
determinations (see amino acid racemization below). However,
method, and the problems associated with its application
the chemical reactions are also significantly influenced by the
in geochronology. The following section provides a summary
presence of water and changes in diagenetic temperature. This
of the principles of amino acid racemization as a geochrono-
complicates the application of these dating methods, which are
logical tool.
generally restricted to relative age assessments.

14.30.9 Amino Acid Racemization 14.30.9.1 Advantages of Amino Acid Racemization Dating
Protein residues in fossil carbonate were first recognized by The advantages of amino acid racemization as a geochrono-
Abelson (1954, 1963) and the potential of the time- logical tool are that the method permits relative and
362 An Introduction to Dating Techniques: A Guide for Geomorphologists

numerical age assessments, stratigraphical correlation, identi- of each of the amino acids can allow for an internal check on
fication of reworked fossils, and paleo-temperature studies the validity of results (Miller and Brigham-Grette, 1989).
(Murray-Wallace, 1993; Wehmiller et al., 1995; Wehmiller and The amino acids most commonly used for geochrono-
Miller, 2000; Kaufman, 2003). The method also extends be- logical applications include valine, leucine isoleucine, and
yond the limits of radiocarbon dating (450 ka), as well as aspartic acid (Miller and Brigham-Grette, 1989; Mitterer and
providing Kriausakul, 1989; Manley et al., 2000; Sloss et al., 2004). The
a dating technique for younger material (o300 years) rate of the racemization reaction observed in the commonly
where the traditional radiocarbon dating may be imprecise utilized amino acids for geochronological problems is gener-
due to problems associated with calibrating conventional ally defined by the relationship where valine o leucine and
radiocarbon ages to sidereal years (Miller and Hare, 1980; alloisoleucine r isoleucine o proline{alainine{ aspartic
Goodfriend, 1991, 1992; Goodfriend et al., 1992; Sloss et al., (Murray-Wallace and Kimber, 1987; Rutter and Cato, 1995).
2004, 2006). Valine and leucine racemize relatively slowly and are also
Amino acid racemization analysis using either high- commonly used as a dating method for Quaternary fossil
performance liquid chromatography (HPLC) or gas chroma- samples due to the relative stability of these acids over lengthy
tography (GC) has the advantage of requiring only a geological intervals (Miller and Hare, 1980; Miller and Brig-
small mass of sample (mg to 1 g CaCO3) compared with ham-Grette, 1989). In contrast, aspartic acid rapidly racemize
conventional radiocarbon analysis, which requires 10–20 g and is thus appropriate for establishing a chronology for
(Miller and Brigham-Grette, 1989; Mitterer and Kriausakul, geologically young (o10 000 to recent) material (Goodfriend,
1989). Recent advances in the analysis of amino acid race- 1991, 1992; Sloss et al., 2004, 2006, 2011).
mization using reverse-phase, high-performance, liquid chro-
matography have made it possible to use a much smaller
sample mass (milligram-size samples; Kaufman and Manley, 14.30.9.3 The Racemization Reaction
1998; Kaufman, 2003).
With the exception of glycine, all enantiomeric amino acids
Amino acid racemization as a geochronological method
form an asymmetric tetrahedral structure around a central
is also applicable to a wide variety of organic remains such
(chiral) carbon atom, which allows the development of two
as carbonate fossils, wood, bone, teeth, eye lenses and,
optical isomers (Rutter et al., 1985; Miller and Brigham-Grette,
with advances in analytical techniques using reverse-phase li-
1989; Rutter and Cato, 1995; Figure 13). As a result, amino
quid chromatography, individual specimens of microfossils,
acids with one chiral carbon atom retained in proteins may
making the technique applicable to a broad range of appli-
potentially exist in two molecular forms:
cations (Miller and Brigham-Grette, 1989; Murray-Wallace,
1993, 1995, 2000; Wehmiller and Miller, 2000). These bene- 1. L (levorotatory) enantiomers that rotate plane-polarized
fits make amino acid racemization a viable technique for age light to the left and
resolution and paleotemperature studies that span the late 2. D (dextrorotatory) enantiomers rotating plane-polarized
Quaternary. light to the right.

Although the biochemical properties of the two forms of

amino acids are similar, they rotate plane-polarized light in
14.30.9.2 Principles of Amino Acid Racemization Dating opposite directions (Wehmiller and Miller, 2000; Murray-
Wallace, 1993, 1995; Figure 13). While alive, organisms ex-
Amino acids combine to form structural proteins, which
clusively form L-configured amino acids. With the death of
are the building blocks of organic life. With the cessation of
an organism and the cessation of protein formation, the
protein formation, which essentially corresponds with the
high-molecular-weight proteins are broken into smaller
death of short-lived organisms, protein within carbonate
polypeptide chains, eventually releasing free amino acids
shells begins to degrade through a complex series of chemical
(hydrolysis), and the enzymatic reaction that maintained the
reactions, whereby structural proteins undergo a number of
disequilibrium condition during life ceases. This results in a
molecular changes. Some of the chemical changes are time-
gradual reequilibration (the racemization reaction), where
dependent, including the hydrolysis of peptide bonds to re-
L-configured amino acids reversibly change to a correspon-
lease smaller polypeptides and free amino acids, and can
ding D-configuration until equilibrium is reached (i.e., D/L
provide the basis for chronological control (Miller and Brig-
ratio ¼ 1). The conversion from L-enantiomer into the corres-
ham-Grette, 1989; Wehmiller and Miller, 2000).
ponding D-enantiomer is a time-dependent reaction. Thus,
Of the twenty amino acids, most can be utilized for geo-
calculation of the rate of the diagenetic racemization of amino
chronological applications. However, each amino acid has a
acids within fossils allows the time since death to be assessed
characteristic rate of racemization depending on the position
(Murray-Wallace, 1993; Rutter and Cato, 1995; Wehmiller and
of the amino acid in the protein chain. Amino acids bound
Miller, 2000; Sloss et al., 2004, 2006).
within the protein chain (internal sites) and amino acids not
bound to the protein chain (free) racemize relatively slowly. In
contrast, amino acids at terminal sites racemize rapidly. As a
14.30.9.4 Factors Influencing the Racemization Rate in
result, direct comparisons of the different amino acids cannot
Carbonate Fossils
be made and individual amino acids are used to determine
relative and numeric ages for fossil material. However, a direct There are a number of parameters that affect the rate of
comparison of the individual characteristic racemization rates racemization observed in fossil molluscs that need to be
 An Introduction to Dating Techniques: A Guide for Geomorphologists 363

CO2H HO2C A. trapezia R 2 = 0.90

1
kL
H2N C C NH2 0.8

Aspartic acid D/L radio

kD
H R R H N. trigonella R 2 = 0.95
0.6
L-configured aspartic acid D-configured aspartic acid
0.4
Figure 13 Asymmetrical carbon atoms in aspartic acid showing two
possible stereoisomers where kL and kD are constants for the
inversion of the carbon to L- and D-isomers. Redrawn from Miller, 0.2
G.H., Brigham-Grette, J., 1989. Amino acid geochronology: resolution
and precision in carbonate fossils. Quaternary International 1, 0
111–128. 0 20 40 60 80 100 120
Heating Days
considered when using amino acid racemization as a geo- 0.5
Notospisula trigonella
chronological technique. The following section provides a
summary of the main parameters affecting racemization rates 0.4

Aspartic acid D/L radio

in fossil molluscs such as diagenetic temperature history,
moisture regime, and the genus-effect. A more comprehensive 0.3
review of the factors that influence the racemization reaction is
provided by Miller and Brigham-Grette (1989), Rutter and
0.2
Cato (1995), and Wehmiller and Miller (2000).
y = 0.0046x + 0.049
0.1 R 2= 0.92
14.30.9.5 Diagenetic Temperature History
0
The rate of racemization is primarily controlled by tempera-
0 20 40 60 80 100
ture in an exponential relationship (Miller and Hare, 1980;
Sloss et al., 2004, 2006; Figure 14). As a result, the rate of Square root of radiocarbon age (cal. yr BP)
racemization will increase by 18% for every 1 1C increase in 0.6
temperature. Accordingly, racemization rates will double with Anadara trapezia
a 5 1C increase in temperature (Miller and Brigham-Grette, 0.5
Aspartic acid D/L ratio

1989; Murray-Wallace, 1993, 1995). Examples of the influence

of temperature on the rates of racemization have been ob- 0.4
served in Arctic regions, where the racemization reaction is
severely impeded. Within the temperature range from  7 to 0.3
 12 1C equilibrium of the slower racemizing amino acids is
generally reached within 10 Ma from the onset of racemiza- 0.2
y = 0.0051x + 0.06
tion (Wehmiller, 1982, 1984). Conversely, racemization re- R 2= 0.94
0.1
actions taking place in tropical and temperate environments
will reach equilibrium much quicker than in colder environ-
0
ments. For example, Hearty and Aharon (1988) noted that 0 20 40 60 80 100
near-equilibrium conditions were obtained from molluscs of
Last Interglacial age (125 ka) from the Huon Peninsula, Papua Square root of radiocarbon age (cal. yr BP)
New Guinea. Therefore, the age range based on amino acid Figure 14 Apparent parabolic kinetic trend of the total aspartic acid
D/L values will vary from region to region depending on the racemization in A. trapezia and N. trigonella based on laboratory-
diagenetic temperature histories of those specific regions. In induced (heating) experiments (top panel). Square-root of radiocarbon
the absence of an independent source of age calibration, the age plotted against the Asp D/L ratio of an equivalent fossil sample for
temperature history of the sample must be firmly established (bottom panel) A. trapezia sample and (middle panel) N. trigonella.
Adapted from Sloss, C.R., Murray-Wallace, C.V., Jones, B.G.,
if numeric ages are to be delineated (Murray-Wallace and
Wallin, T., 2004. Aspartic acid racemization dating of mid-Holocene to
Kimber, 1987; Rutter and Cato, 1995).
recent estuarine sedimentation in New South Wales, Australia: a pilot
Another consideration in relation to the influence of tem- study. Marine Geology 212, 45–59, and Sloss, C.R., Murray-Wallace,
perature on the rate of racemization is whether the sample has C.V., Jones, B.G., 2006. Aminostratigraphy of two Holocene wave-
been subaerially exposed for a lengthy part of its diagenetic dominated barrier estuaries in southeastern Australia. Journal of
history. For example, marine shells uplifted or exposed shortly Coastal Research 22, 113–136.
after death experience mean terrestrial temperature and
dramatic temperature oscillations. In such situations, summer samples that have remained submerged and deeply buried
temperatures will have a disproportionate influence on the since death will have experienced a relatively constant diage-
rate of racemization, resulting in higher D/L values, and netic temperature (Miller and Brigham-Grette, 1989; Murray-
therefore an older apparent age determination. In contrast, Wallace, 1993, 1995, 2000). For example, the effective
364 An Introduction to Dating Techniques: A Guide for Geomorphologists

diagenetic temperature for fossil molluscs buried at depths of occur at a genus level in fossil molluscs (Miller and Brigham-
1 m or greater in sedimentary successions will closely ap- Grette, 1989; Murray-Wallace, 1993; Murray-Wallace and
proximate the current mean annual temperature (Miller and Kimber, 1987; Sloss et al., 2004; Figure 14). The difference in
Brigham-Grette, 1989; Murray-Wallace, 2000). Deeply buried the racemization reaction observed at the taxonomic level is
samples (41 m) are less likely to have experienced large caused by the original makeup of proteins and polypeptides
temperature variations associated with daily and seasonal and by morphological and structural differences of the host
temperature variations and, on a regional scale, such samples matrix of the protein residue (Miller and Brigham-Grette,
would have had the same diagenetic temperature histories. 1989; Murray-Wallace, 1993; Wehmiller and Miller, 2000;
The problems associated with variable diagenetic temperature Kaufman, 2003).
histories can be minimized by careful sample selection, col-
lection, and preparation.
14.30.9.8 Numerical Age Assessments
Assessment of numerical ages requires a secure reconstruction
14.30.9.6 Moisture Regime of the integrated thermal history of the site (Miller and Hare,
Another parameter that needs to be considered in relation to 1980; Rutter et al., 1985) or an accurate model of racemiza-
the rate of racemization reactions is the moisture regime in tion kinetics and a direct calibration with an independent
which the fossil material is preserved (Miller and Brigham- dating technique (Miller and Brigham-Grette, 1989; Murray-
Grette, 1989; Murray-Wallace, 1993; Rutter and Cato, 1995). Wallace, 1993; Murray-Wallace, 2000; Sloss et al., 2004, 2005).
Water is required for hydrolysis of peptide bonds; thus, fluc- One approach to construct an accurate model of kinetic trends
tuations in sediment water content may potentially result in involves the determination of a racemization constant (k)
periods of increased rates of protein degradation and also by laboratory pyrolysis experiments (Murray-Wallace, 1993;
result in the mobilization of the highly racemized lower mo- Rutter and Cato, 1995; Sloss et al., 2004, 2006). Using this
lecular weight peptides in weathered and degraded fossil method, numeric ages are calculated on the degree of amino
material (Wehmiller, 1982; Rutter and Cato, 1995). Accord- acid racemization using an apparent parabolic kinetic model
ingly, fossil molluscs that are preserved in sedimentary suc- (Mittererand Kriausakul, 1989; Goodfriend et al., 1992;
cessions above a permanent water table potentially experience Murray-Wallace and Kimber, 1987; Sloss et al., 2004, 2006).
significant fluctuations in moisture regime associated with Using this age assessment, the square-root of radiocarbon age
percolating ground water and changes in soil moisture on a (in sidereal years) is used to calibrate the degree of racemi-
daily or a seasonal basis. In such environmental conditions, zation with respect to time. This results in a near-linear rela-
fossil molluscs become highly weathered and degraded, with tionship between measured D/L values and the square-root of
a chalky and visibly porous appearance (Miller and Brigham- time within the reaction range and can then be used to de-
Grette, 1989; Murray-Wallace, 1993). Such environmental termine numeric age from additional D/L values (Figure 14).
conditions can potentially result in the leaching of the Other mathematical approaches used to provide a best-fit
lower molecular weight peptides and lower amino acid con- linearization have been used based on power-law functions. For
centrations within the carbonate matrix of fossil mollusc. example, Goodfriend et al. (1996) produced linear regressions
Conversely, during periods of excess moisture, a higher race- in heating experiments using the relationship between time and
mization rate of the remaining amino acids potentially results Asp D/L values raised to the power of values ranging from 1.7 to
in an older age for fossil molluscs. Thus, fluctuating moisture 4 (depending on the mollusc genus). Manley et al. (2000)
regimes potentially affect the rate of racemization as well as developed a more complex version of this power function
the measured extent of amino acid racemization. However, model to produce linearization of trends in Asp D/L values that
samples that are permanently saturated within a sub-aqueous are not adequately linearized by the simple power function. A
environment, or below the water table, do not appear to pose more comprehensive review of mathematical expressions used
a problem (Murray-Wallace, 1993, 1995, 2000). in amino acid racemization geochronology is presented in
Clarke and Murray-Wallace (2006).

14.30.9.7 Taxonomic and Genus Effect

14.30.10 Conclusion
Rates of racemization can potentially vary between taxa due to
differences in variable bonding strengths between adjacent The question of time is a fundamental component of virtually
amino acids retained in carbonate fossils (Lajoie et al., 1980; all geomorphological research. Thus, dating techniques such
Miller and Hare, 1980; Miller and Brigham-Grette, 1989). as isotopic, radiogenic, and biochemical dating have become
For example, Kaufman (2003) reported that amino acids essential tools to provide a chronological framework for geo-
preserved in fossil molluscs racemize up to 35% faster than morphological problems. It has been demonstrated that each
amino acids preserved in ostracods. Different rates of race- dating method has particular applications and limitations. For
mization observed at the genus level were first recognized in example, radiocarbon dating (an isotopic dating technique)
fossil foraminifera collected from deep sea sediments (King, and amino acid racemization (a biochemical dating technique)
1980; Lajoie et al., 1980). These studies showed that the approximate the time of death of an organism. If the dated
racemization rate for each foraminiferal species investigated material is placed in its proper geomorphological context, then
differs to a greater or a lesser extent. Subsequent research has both dating techniques can provide an accurate chronological
shown that different rates of amino acid racemization also framework. OSL dating (radiogenic dating) provides the burial
 An Introduction to Dating Techniques: A Guide for Geomorphologists 365

age of sediments and therefore plays a key role in determining Bronk Ramsey, C., 2009. Bayesian analysis of radiocarbon dates. Radiocarbon 51,
the timing of sediment deposition in complex sedimentary 337–360.
Buddemeier, R.W., Maragos, J.E., Knutson, D.W., 1974. Radiographic studies of reef
environments, while recent methodological advances such as
coral exoskeletons: rates and patterns of coral growth. Journal of Experimental
single-grain analysis allow the processes of transport and de- Marine Biology and Ecology 14, 179–199.
position to be explored. When used separately, these techniques Clarke, M.L., Vogel, J.C., Botha, G.A., Wintle, A.G., 2003. Late Quaternary hillslope
can provide important chronological information but when evolution recorded in eastern South African colluvial badlands. Palaeogeography,
used in combination they can provide a more accurate Palaeoclimatology, Palaeoecology 197, 199–212.
Clarke, S.J., Murray-Wallace, C.V., 2006. Mathematical expressions used in amino acid
chronological framework to assess the rate of geomorphologic racemization geochronology - a review. Quaternary Geochronology 1, 261–278.
processes, age of depositional landforms, specific events, or Choi, J.H., Kim, J.W., Murray, A.S., Hong, D.G., Chang, H.W., Cheong, C.S., 2009.
sedimentary environments. OSL dating of marine terrace sediments on the southeastern coast of Korea with
However, it has been demonstrated that establishing the implications for Quaternary tectonics. Quaternary International 199, 3–14.
Clemmensen, L.B., Murray, A., Heinemeier, J., de Jong, R., 2009. The evolution of
age of a specific landform, event, or depositional environment
Holocene coastal dunefields, Jutland, Denmark: a record of climate change over
can be a complex task unless a strategy is developed before the past 5000years. Geomorphology 105, 303–313.
sample collection and fundamental questions are addressed Cockburn, H.A.P., Summerfield, M.A., 2004. Geomorphological applications of
before the final conclusions are drawn. These include: cosmogenic isotope analysis. Progress in Physical Geography 28(1), 1–42.
Colman, S.M., King, J.W., Jones, G.A., Reynolds, R.L., Bothner, M.H., 2000.
• What is the material being dated and what is the most Holocene and recent sediment accumulation rates in southern Lake Michigan.
appropriate dating technique for the samples being col- Quaternary Science Reviews 19, 1563–1580.
lected? This includes an understanding of the composition Colman, S.M., Pierce, K.L., Birkeland, P.W., 1987. Suggested terminology for
Quaternary dating methods. Quaternary Research 28, 314–319.
and expected age range of the material being dated and Cunha, P.P., Buylaert, J.-P., Murray, A.S., et al., 2010. Optical dating of clastic
then choosing the most appropriate dating method. deposits generated by an extreme coastal flood: the 1755 tsunami deposits in
• What is the precision of the ‘type’ of dating method being the Algarve (Portugal). Quaternary Geochronology 5, 329–335.
used and how does that numeric age determination relate to Davids, F., Duller, G.A.T., Roberts, H.M., 2010. Testing the use of feldspars for
optical dating of hurricane overwash deposits. Quaternary Geochronology 5,
the landform, event, or depositional environment being
125–130.
dated (i.e., the accuracy of the date)? This includes an ac- Damon, P.E., Sonett, C.P., 1991. Solar and terrestrial components of the atmospheric
14
curate assessment of the relationship of the material being C variation spectrum. In: Sonett, C.P., Giampapa, M.S., Matthews, M.S. (Eds.),
dated with the specific landform, event, or depositional The Sun in Time. University of Arizona Press, Tucson, pp. 360–388.
environment. Deevey, E.S., Gross, M.S., Hutchinson, G.E., Kraybill, H.L., 1954. The natural C14
contents of materials from hard-water lakes. Proceedings of the National
By addressing these issues, a researcher will ensure that they Academy of Sciences of the United States of America 40, 285–288.
Derouin, S.A., Lowell, T.V., Hajdas, I., 2007. Landscape evolution and deglaciation
develop an accurate chronological framework for landscape
of the Upper Peninsula, Michigan: an examination of chronology and
development and geomorphological problems. Armed with stratigraphy in Kettle Lake Cores. Journal of Great Lakes Research 33, 875–886.
geochronologically well-constrained landscape evolution Donahue, D.J., Linick, T.W., Jull, A.J.T., 1990. Isotope-ratio and background
models, geomorphologists will then be able to make informed corrections for accelerator mass spectrometry radiocarbon measurement.
assessments of how specific landforms have changed in the Radiocarbon 32, 135–142.
Duller, G.A.T., 1995. Luminescence dating using single-aliquots: methods and
past as well as potentially change in the future. applications. Radiation Measurements 24, 217–226.
Duller, G.A.T., 2000. Dating methods: geochronology and landscape evolution.
Progress in Physical Geography 24, 111–116.
Duller, G.A.T., 2004. Luminescence dating of quaternary sediments: recent advances.
References Journal of Quaternary Science 19, 183–192.
Duller, G.A.T., 2008. Luminescence Dating: Guidelines on Using Luminescence
Abelson, P.H., 1954. Organic constituents of fossils. Carnegie Institution of Dating in Archaeology. English Heritage, Swindon.
Washington Yearbook 53, 97–101. Duller, G.A.T., Murray, A., 2000. Luminescence dating of sediments using individual
Abelson, P.H., 1963. Geochemistry of Amino Acids. In: Berger, I.G. (Ed.), Organic mineral grains. Geologos 5, 88–106.
Geochemistry. Macmillan, New York, pp. 431–455. Fattahi, M., Nazari, H., Bateman, M.D., et al., 2010. Refining the OSL age of the
Adamiec, G., 2005. OSL decay curves - relationship between single- and multiple- last earthquake on the Dheshir fault, Central Iran. Quaternary Geochronology 5,
grain aliquots. Radiation Measurements 39, 63–75. 286–292.
Aitken, M.J., 1985. Thermoluminescence Dating. Academic Press, London. Fattahi, M., Stokes, S., 2000. Extending the time range of luminescence dating
Aitken, M.J., 1998. An Introduction to Optical Dating: The Dating of Quaternary using red TL (RTL) from volcanic quartz. Radiation Measurements 32,
Sediments by the Use of Photon-stimulated Luminescence. Oxford University 479–485.
Favilli, F., Cherubini, P., Collenberg, M., Egli, M., Sartori, G., Schoch, W., Haeberli, W.,
Press, Oxford.
2010. Charcoal fragments of Alpine soils as an indicator of landscape evolution
Banerjee, D., 2010. Thermoluminescence and optically stimulated luminescence
during the Holocene in Val di Sole (Trentino, Italy). The Holocene 20, 67–79.
signals from volcanic ash: history of volcanism in Barren Island, Andaman Sea.
Franke, J., Paul, A., Schulz, M., 2008. Modeling variations of marine reservoir ages
Quaternary Geochronology 5, 279–282.
during the last 45000years. Climate of the Past 4, 125–136.
Banerjee, D., Murray, A.S., Foster, I.D.L., 2001. Optical dating of a possible tsunami
Friedrich, M., Remmele, S., Kromer, B., et al., 2004. The 12,460-year Hohenheim
deposit from the 1755 Lisbon earthquake. Quaternary Science Reviews 20, oak and pine tree-ring chronology from central Europe – a unique annual
715–718. record for radiocarbon calibration and paleoenvironment reconstructions.
Berger, G.W., 1990. Effectiveness of natural zeroing of the thermoluminescence in Radiocarbon 46, 1111–1122.
sediments. Journal of Geophysical Research 95, 12375–12397. Fuchs, M., Fisher, M., Reverman, R., 2010. Colluvial and alluvial sediment archives
Bondevik, S., Mangerud, J., Birks, H.H., Gulliksen, S., Reimer, P., 2006. Changes in temporally resolved by OSL dating: implications for reconstructing soil erosion.
North Atlantic radiocarbon reservoir ages during the Allerød and Younger Dryas. Quaternary Geochronology 5, 269–273.
Science 312, 1514–1517. Geyh, M.A., 2005. 14C dating – still a challenge for users? Zeitschriftfür
Bondevik, S., Mangerud, J., Dawson, S., Dawson, A., Lohn, Ø., 2005. Evidence for Geomorphologie Supplement 139, 63–86.
three North Sea tsunamis at the Shetland Islands between 8000 and 1500 years Geyh, M.A., Schotterer, U., Grosjean, M., 1998. Temporal changes of the 14C
ago. Quaternary Science Reviews 24, 1757–1775. reservoir effect in lakes. Radiocarbon 40, 921–931.
366 An Introduction to Dating Techniques: A Guide for Geomorphologists

Godfrey-Smith, D.I., Huntley, D.J., Chen, W.H., 1988. Optical dating studies P.E., Hoering, T.C., King, K. (Eds.), Biogeochemistry of Amino Acids. Wiley, New
of quartz and feldspar sediment extracts. Quaternary Science Reviews 7, York, pp. 305–340.
373–380. Lian, O.B., Roberts, R.G., 2006. Dating the Quaternary: progress in luminescence
Goodfriend, G.A., 1991. Patterns of racemization and epimerization of amino acids dating of sediments. Quaternary Science Reviews 25, 2449–2468.
in land snail shells over the course of the Holocene. Geochimica et Lian, O.B., 2007a. Thermoluminescence. In: Elias, S.A. (Ed.), Encyclopaedia of
Cosmochimica Acta 55, 293–302. Quaternary Science. Elsevier, Oxford, pp. 1480–1491.
Goodfriend, G.A., 1992. Rapid racemization of aspartic acid in mollusc shells and Lian, O.B., 2007b. Optically-stimulated luminescence. In: Elias, S.A. (Ed.),
potential for dating over recent centuries. Nature 357, 399–401. Encyclopaedia of Quaternary Science. Elsevier, Oxford, pp. 1491–1505.
Goodfriend, G.A., Brigham-Grette, J.B., Miller, G.H., 1996. Enhanced age resolution Long, A.J., Waller, M.P., Stupples, P., 2006. Driving mechanisms of coastal change:
of the marine quaternary record in the artic using aspartic acid racemization peat compaction and the destruction of late Holocene coastal wetlands. Marine
dating on bivalve shells. Quaternary Research 45, 176–187. Geology 225, 63–84.
Goodfriend, G.A., Hare, P.E., Druffel, E.R.M., 1992. Aspartic acid racemization and Lowe, J.J., Walker, M.J.C., 1997. Reconstructing Quaternary Environments, Second
protein diganesis in corals over the last 350years. Geochimica et Cosmochimica edn. Longman, London, pp. 446.
Acta 56, 3847–3850. Lomax, J., Hilgers, A., Wopfner, H., Grun, R., Twidale, C.R., Radtke, U., 2003. The
Halfar, J., Steneck, R.S., Joachimski, M., Kronz, A., Wanamaker, Jr. A.D., 2008. onset of dune formation in the Strzelecki Desert, South Australia. Quaternary
Coralline red algae as high-resolution climate recorders. Geology 36, 463–466. Science Reviews 22, 1067–1076.
Hardy, F., Lamothe, M., 1997. Quaternary Basin analysis using infrared stimulated Mandel, R.D., 2008. Buried paleoindian-age landscapes in stream valleys of the
luminescence on bore hole cores and cuttings. Quaternary Science Reviews 16, central plains, USA. Geomorphology 101, 342–361.
417–426. Manley, W.F., Gifford, H.M., Czywczynski, J., 2000. Kinetics of aspartic acid
Hare, P.E., Abelson, P.H., 1968. Racemization of amino acids in fossil shells. racemization in Mya and Hiatella: modelling age and palaeo-temperature of
Carnegie Institution of Washington Yearbook 66, 526–528. high-latitude Quaternary mollusks. In: Goodfriend, G.A., Collins, M.J.,
Hashimoto, T., Hayashi, Y., Koyanagi, A., Yokosaka, K., Kimura, K., 1986. Red and Fogel, M.L., Macko, S.A., Wehmiller, J.F. (Eds.), Perspectives in
blue colouration of thermoluminescence from natural quartz sands. Nuclear Amino Acids and Protein Geochemistry. Oxford University Press, Melbourne,
Tracks and Radiation Measurements 11, 229–235. pp. 202–218.
Hearty, P.J., Aharon, P., 1988. Amino acid chronostratigraphy of Late Quaternary Martins, A.A., Hobo, N., Cunningham, A.C., Versendaal, A.J., Makaske, B.,
coral reefs: Huon Peninsula, New Guinea, and the Great Barrier Reef, Australia. Middelkoop, H., 2010. K-feldpspar IRSL dating of Pleistocene river terrace
Geology 16, 579–583. staircase sequence of the LowerTejo River (Portugal, western Iberia). Quaternary
Hedges, R.E.M., 1992. Sample treatment strategies in radiocarbon dating. In: Taylor, Geochronology 2010, 176–180.
R.E., Long, A., Kra, R.S. (Eds.), Radiocarbon After Four Decades: An McCormac, F.G., Hogg, A.G., Blackwell, P.G., Buck, C.E., Higham, T.F.G., Reimer,
Interdisciplinary Perspective. Springer, New York, pp. 165–183. P.J., 2004. SHCal04 Southern Hemisphere calibration 0–11.0 cal kyr BP.
Hua, Q., 2009. Radiocarbon: a chronological tool for the recent past. Quaternary Radiocarbon 46, 1087–1092.
Geochronology 4, 378–390. McGregor, H.V., Gagan, M.K., 2003. Diagenesis and geochemistry of Porites corals
Hua, Q., Barbetti, M., 2004. Review of tropospheric bomb 14C data for carbon cycle from Papua New Guinea: implications for paleoclimate reconstruction.
modeling and age calibration purposes. Radiocarbon 46, 1273–1298. Geochimica et Cosmochimica Acta 67, 2147–2156.
Hua, Q., Zoppi, U., Williams, A.A., Smith, A.M., 2004. Small-mass AMS McGregor, H.V., Gagan, M.K., McCulloch, M.T., Hodge, E., Mortimer, G., 2008.
radiocarbon analysis at ANTARES. Nuclear Instruments and Methods in Physics Mid-Holocene variability in the marine 14C reservoir age for northern coastal
Research B 223–224, 284–292. Papua New Guinea. Quaternary Geochronology 3, 213–225.
Huang, C.C., Huang, C.C., Pang, J., Zha, X., Su, H., Jia, Y., Zhu, Y., 2007. Impact Miller, G.H., Hare, P.E., 1980. Amino acid geochronology: integrity of the carbonate
of monsoonal climatic change on Holocene overbank flooding along Sushi matrix and potential of molluscan fossils. In: Hare, P.E., Hoeing, T.C., King, K.
River, middle reach of the Yellow River, China. Quaternary Science Reviews 26, (Eds.), Biochemistry of Amino Acids: Selected Papers. Wiley, New York, pp.
2247–2264. 415–451.
Huntley, D.J., Godfrey-Smith, D.I., Thewalt, M.L.W., 1985. Optical dating of Miller, G.H., Brigham-Grette, J., 1989. Amino acid geochronology: resolution and
sediments. Nature 313, 105–107. precision in carbonate fossils. Quaternary International 1, 111–128.
Jacobs, Z., Duller, G.A.T., Wintle, A.G., 2003. Optical dating of dune sand from Mitterer, R.M., Kriausakul, N., 1989. Calculation of amino acid racemization
Blombos Cave, South Africa: II- single grain data. Journal of Human Evolution ages based on apparent parabolic kinetics. Quaternary Science Reviews 88,
44, 613–625. 353–357.
de Jong, R., Björck, S., Bjokman, S., Clemmensen, L.B., 2006. Storminess Murray-Wallace, C.V., 1993. A review of the application of the amino acid
variation during the last 6500 years as reconstructed from an ombrotrophic racemization reaction to archaeological dating. The Artefact 16, 19–26.
peat bog in Halland, southwest Sweden. Journal of Quaternary Science 21, Murray-Wallace, C.V., 1995. Aminostratigraphy of Quaternary coastal sequences in
905–919. southern Australia - an overview. Quaternary International 26, 69–86.
Jull, A.J.T., Burr, G.S., 2006. Accelerator mass spectrometry: is the future bigger or Murray-Wallace, C.V., 2000. Quaternary coastal aminostratigraphy: Australian data in
smaller? Earth and Planetary Science Letters 243, 305–325. a global context. In: Goodfriend, G.A., Collins, M.J., Fogel, M.L., Macko, S.A.,
Kaufman, D.S., 2003. Amino acids paleothermometry of Quaternary Ostracodes from Wehmiller, J.F. (Eds.), Perspectives in Amino Acids and Protein Geochemistry.
the Bonneville Basin, Utah. Quaternary Sciences Reviews 22, 899–914. Oxford University Press, Melbourne, pp. 279–300.
Kaufman, D.S., Manley, W.F., 1998. A new procedure for determining D/L amino Murray-Wallace, C.V., Kimber, R.W.L., 1987. Evaluation of the amino acid
acid ratios in fossils using reverse phase liquid chromatography. Quaternary racemization reaction in studies of quaternary marine sediment in South
Geochronology 17, 987–1000. Australia. Australian Journal of Earth Sciences 34, 279–292.
Kidder, T.R., Adelsberger, K.A., Arco, L.J., Schilling, T.M., 2008. Basin-scale Nanson, G.C., Price, D.M., Short, S.A., 1992. Wetting and drying of Australia over
reconstruction of the geological context of human settlement: an example the past 300 ka. Geology 20, 791–794.
from the lower Mississippi Valley, USA. Quaternary Science Reviews 27, Narama, C., Kondo, R., Tsukamoto, S., Kajiura, T., Ormukov, C., Abdrakhmatov, K.,
1255–1270. 2007. OSL dating of glacial deposits during the last glacial in the
King, K., 1980. Applications of amino acid biogeochemistry from marine sediments. Terskey-Alatoo range, Kyrgyz Republic. Quaternary Geochronology 2, 249–254.
In: Hare, P.E., Hoering, T.C., King, K. (Eds.), Biogeochemistry of Amino Acids. Noller, J.S., Sowers, J.M., Lettis, W.R. (Eds.), 2000. Quaternary Geochonology:
Wiley, New York, pp. 377–391. Methods and Applications. American Geophysical Union, Washington, DC.
Klug, H.P., Alexander, L.E., 1954. X-Ray Diffraction Procedures: For Polycrystalline Nothdurft, L.D., Webb, G.E., 2009. Earliest diagenesis in scleractinian coral
and Amorphous Materials. Wiley, New York, 716 pp. skeletons: implications for palaeoclimate-sensitive geochemical archives. Facies
Knutson, D.W., Buddemeier, R.W., Smith, S.V., 1972. Coral chronometers: seasonal 55, 161–201.
growth bands in reef corals. Science 177, 270–272. Ortlieb, C., Vargas, G., Saliège, J.-F., 2011. Marine radiocarbon reservoir effect
Kondo, R., Tsukamoto, S., Tachibana, H., Miyairi, Y., Yokoyama, Y., 2007. Age of along the northern Chile – southern Peru coast (14–241S) throughout the
glacial and periglacial landforms in northern Hokkaido, Japan, using OSL dating Holocene. Quaternary Research 75, 91–103.
of fine grain quartz. Quaternary Geochronology 2, 260–265. Ou, X.J., Xu, L.B., Lai, Z.P., Long, H., He, Z., Fan, Q.S., Zou, S.Z., 2010. Potential
Lajoie, K.R., Wehmiller, J.F., Kennedy, G.I., 1980. Inter- and intrageneric trends in of quartz OSL dating on moraine deposits from eastern Tibetan Plateau using
apparent racemization kinetics of amino acids in quaternary mollusks. In: Hare, SAR protocol. Quaternary Geochronology 5, 257–262.
 An Introduction to Dating Techniques: A Guide for Geomorphologists 367

Palmer, A.P., Rode, J., John, K., 2010. Annually resolved events of Younger Dryas Stokes, S., 1999a. Luminescence dating applications in geomorphological research.
glaciation in Lochaber (Glen Roy and Glen Spean), western Scottish Highlands. Geomorphology 29, 153–171.
Journal of Quaternary Science 25, 581–596. Stokes, S., 1999b. Optical dating of young (modern) sediments using quartz:
Panin, A.V., Fuzeina, J.N., Belyaev, V.R., 2009. Long-term development of Holocene results from a selection of depositional environments. Quaternary Science
and Pleistocene gullies in the Protva River basin, central Russia. Reviews 11, 153–159.
Geomorphology 108, 71–91. Stuiver, M., Polach, H.A., 1977. Discussion: reporting of 14C data. Radiocarbon 19,
Porat, N., Wintle, A.G., Amit, R., Enzel, Y., 1996. Late Quaternary earthquake 355–363.
chronology from luminescence dating of colluvial and alluvial deposits of the Stuiver, M., Pearson, G.W., Braziunas, T., 1986. Radiocarbon age calibration of
Arava Valley, Israel. Quaternary Research 46, 107–117. marine samples back to 9000 calyr BP. Radiocarbon 28, 980–1021.
Prasad, S.P., Negendank, J.F.W., Stein, M., 2009. Varve counting reveals high Taylor, R.E., 1987. Radiocarbon Dating an archaeological perspective. Academic
resolution radiocarbon reservoir age variations in palaeolake Lisan. Journal of Press, London, 212 pp.
Quaternary Science 24, 690–696. Thamo-Bozso, E., Csillag, G., Fodor, L.I., Muller, P.M., Nagy, A., 2010. OSL dating
Reimann, T., Naumann, M., Tsukamoto, S., Frechen, M., 2010. Luminescence dating the Quaternary landscape evolution in the Vertes Hills forelands (Hungary).
of coastal sediments from the Baltic Sea coastal barrier-spit Darss-Zingst, NE Quaternary Geochronology 5, 120–124.
Germany. Geomorphology 122, 264–273. Tipping, R., Jones, A.P., Carter, S., Holden, T., Cressey, M., 2008. The chronology
Reimer, P.J., Reimer, R.W., 2001. A marine reservoir correction database and online and long term dynamics of a low energy river system: the Kelvin Valley, central
interface. Radiocarbon 43, 461–463. http://radiocarbon.pa.qub.ac.uk/marine/ Scotland. Earth Surface Processes and Landforms 33, 910–922.
Reimer, P.J., Baillie, M.G.L., Bard, E., et al., 2009. IntCal09 and Marine09 Tuniz, C., Bird, J.R., Fink, D., Herzog, G.F., 1998. Accelerator Mass Spectrometry:
radiocarbon age calibration curves, 0–50,000 years cal BP. Radiocarbon 51, Ultrasensitive Analysis for Global Science. CRC Press, Boca Raton, 371 pp.
1111–1150. Valladas, H., 1982. Some TL properties of burnt prehistoric flints. Nuclear Tracks
Rendell, H.M., Webster, S.E., Sheffer, N.L., 1994. Underwater bleaching of signals and Radiation Measurements 10, 785–788.
from sediment grains: new experimental data. Quaternary Geochronology 13, Valladas, H., Joron, J.L., Valladas, G., Arensburg, B., 1987. Thermoluminescence
433–435. dates for the Neanderthal burial site at Kebara in Israel. Nature 330,
Rhodes, E.J., Pownall, N., 1994. Zeroing of the OSL in quartz from young 159–160.
glaciofluvial sediments. Radiation Measurements 23, 581–586. Walker, M.J.C., 2005. Quaternary Dating Methods. Wiley, Chichester, 286 pp.
Roberts, R.G., Galbraith, R.F., Yoshida, H., Laslett, G.M., Olley, J.M., 2000. Wang, P., Jiang, H., Yuan, D., Liu, X., Zhang, B., 2010. Optically stimulated
Distinguishing dose populations in sediment mixtures: a test of single-grain luminescence dating of sediments from the Yellow River terraces in Lanzhou:
optical dating procedures using mixtures of laboratory-dosed quartz. Radiation tectonic and climatic implications. Quaternary Geochronology 5, 181–186.
Measurements 32, 459–465. Webb, G.E., Price, G.J., Nothdurft, L.D., Deer, L., Rintoul, L., 2007. Cryptic meteoric
Roberts, R.G., Bird, M., Olley, J.M., et al., 1998. Optical and radiocarbon dating at diagenesis in freshwater bivalves: implications for radiocarbon dating. Geology
Jinmium rock shelter in northern Australia. Nature 393, 358–362. 35, 803–806.
Rommens, T., Verstraeten, G., Peeters, I., et al., 2007. Reconstruction of late- Wehmiller, J.F., 1982. A review of amino acid racemization studies in Quaternary
Holocene slope and dry valley sediment dynamics in a Belgian loess mollusks: stratigraphic and chronological applications in coastal and interglacial
environment. The Holocene 17, 777–788. sites, Pacific and Atlantic coasts, United States, United Kingdom, Baffin Island
Rose, J., 2007. The use of time units in Quaternary Science Reviews. Quaternary and tropical islands. Quaternary Science Reviews 1, 83–120.
Science Reviews 26, 1193. Wehmiller, J.F., 1984. Relative and absolute dating of Quaternary molluscs with
Rutter, N.W., Cato, E.R. (Eds.), 1995. Dating Methods for Quaternary Deposits amino acid racemisation: evaluation, applications and questions. In: Mahaney,
Geotext 2. Geological Association of Canada, St Johns. W.C. (Ed.), Quaternary Dating Methods. Elsevier, Amsterdam, pp. 171–193.
Rutter, N.W., Crawford, R.J., Hamilton, R.D., 1985. Amino acid racemization Wehmiller, J.F., Miller, G.H., 2000. Aminostratigraphic dating methods in Quaternary
dating. In: Rutter, N.W. (Ed.), Dating methods of Pleistocene Deposits geology. In: Noller, J.S., Sowers, J.M., Lettis, W.R. (Eds.), Quaternary
and their Problems. Geological Association of Canada, Ontario, pp. 23–30. Geochronology: Methods and Applications. AGU Reference Shelf 4. American
Santos, G.M., Southon, J.R., Griffin, S., Beaupre, S.R., Druffel, E.R.M., 2007. Ultra Geophysical Union, Washington DC, pp. 187–222.
small-mass AMS 14C sample preparation and analyses at KCCAMS/UCI Wehmiller, J.F., York, L.L., Bart, M.L., 1995. Amino acid racemisation
facility. Nuclear Instruments and Methods in Physics Research B 259, 293–302. geochronology of reworked Quaternary mollusks on U.S. Atlantic coast beaches:
Scheffers, A., Scheffers, S., Kelletat, D., Browne, T., 2009. Wave-emplaced coarse implication for chronostratigraphy, taphonomy, and coastal sediment transport.
bebris and megaclasts in Ireland and Scotland: boulder transport in a high- Marine Geology 124, 303–337.
energy littoral environment. Journal of Geology 117, 553–573. Westaway, K.E., Roberts, R.G., Sutikna, T., Morwood, M.J., Drysdale, R., Zhao, J.-x.,
Shellberg, J.G., Hua, Q., Zawadski, A., Brooks, A.P., (pers com). Quantification of Chivas, A.R., 2009. The evolving landscape and climate of western Flores: an
alluvial gully erosion rates using radium-226/228 and carbon-14 dating of environmental context for the archaeological site of Liang Bua. Journal of
Eucalyptus tree age, Mitchell River, northern Australia, in preparation. Human Evolution 57, 450–464.
Singarayer, J.S., Bailey, R.M., Stokes, S., 2006. Assessing the completeness of Westaway, K.E., Sutikna, T., Morwood, M.J., Zhao, J.-x., (2010). Estimating rates of
optical resetting of quartz OSL in the natural environment. Radiation karst landscape evolution in the tropics: a context for archaeological sites in
Measurements 40, 13–25. western Flores, Indonesia. Journal of Quaternary Science. DOI: 10.1002/
Schmitt, A., Rodzik, J., Zgaobicki, W., Russok, C., Dotterweich, M., Bork, H., 2006. jqs.1390.
Time and scale of gully erosion in the JedlicznyDol gully system, south-east Williams, M., Dunkerley, D., De Decker, P., Kershaw, P., Chappell, J., 1998.
Poland. Catena 68, 124–132. Quaternary Environments. Appendix: Dating Methods in Quaternary Research.
Sloss, C.R., Jones, B.G., Murray-Wallace, C.V., McClennen, C.E., 2005. Holocene Oxford University Press, New York, pp. 270–284.
sea level fluctuations and the sedimentary evolution of a barrier estuary: Wintle, A.G., 1997. Luminescence dating: laboratory procedures and protocols.
Lake Illawarra, New South Wales, Australia. Journal of Coastal Research Radiation Measurements 27(5–6), 769–817.
21, 943–959, 974–975. Woodroffe, C.D., Samosorn, B., Hua, Q., Hart, D.E., 2007. Incremental accretion of
Sloss, C.R., Jones, B.J., Brooke, B.P., Heijnis, H., Murray-Wallace, C.V., 2011. a sandy reef island over the past 3000 years indicated by component-specific
Contrasting sedimentation rates in Lake Illawarra and St Georges Basin, two radiocarbon dating. Geophysical Research Letters 34, L03602. http://dx.doi.org/
large barrier estuaries on the southeast coast of Australia. Journal of 10.1029/2006GL028875.
Paleolimnology 46, 561–577. Wu, T.-S., Jaiswal, M.K., Lin, Y.N., Chen, Y.-W., Chen, Y.-G., 2010. Residual
Sloss, C.R., Murray-Wallace, C.V., Jones, B.G., Wallin, T., 2004. Aspartic acid luminescence in modern debris flow deposits from western Taiwan: a single-
racemization dating of mid Holocene to recent estuarine sedimentation in New grain approach. Journal of Asian Earth Sciences 38, 274–282.
South Wales, Australia: a pilot study. Marine Geology 212, 45–59. Yu, K., Hua, Q., Zhao, J-x., Hodge, E., Fink, D., Barbetti, M., 2010. Holocene
Sloss, C.R., Murray-Wallace, C.V., Jones, B.G., 2006. Aminostratigraphy of two marine 14C reservoir age variability: evidence from 230Th-dated corals in the
Holocene wave-dominated barrier estuaries in southeastern Australia. Journal of South China Sea. Paleoceanography 25, PA3205. http://dx.doi.org/10.1029/
Coastal Research 22, 113–136. 2009PA001831.
Smith, A.M., Hua, Q., Williams, A., Levchenko, V., Yang, B., 2010. Developments in Zhao, H., Lu, Y., Yin, J., 2007. Optical dating of Holocene sand dune activities in
micro-sample 14C AMS at the ANTARES AMS facility. Nuclear Instruments and the Horqin sand-fields in inner Mongolia, China, using the SAR protocol.
Methods in Physics Research B 268, 919–923. Quaternary Geochronology 2, 29–33.
368 An Introduction to Dating Techniques: A Guide for Geomorphologists

Zielhofer, C., Faust, D., Linstädter, J., 2008. Late Pleistocene and Holocene Zoppi, U., Albani, A., Ammerman, A.J., Hua, Q., Lawson, E.M., Serandrei, B.R.,
alluvial archives in the southwestern Mediterranean: changes in 2001. Preliminary estimate of the reservoir age in the Lagoon of Venice.
fluvial dynamics and past human response. Quaternary International 181, Radiocarbon 43, 489–494.
39–54. Zreda, M.G., Phillips, F.M., 2000. Cosmogenic nuclide buildup in surficial
Zilbermann, E., Amit, R., Heimann, A., Porat, N., 2000. Changes in Holocene materials. In: Noller, J.S., Sowers, J.M., Lettis, W.R. (Eds.), Quaternary
palaeoseismic activity in the Hula pull-apart basin, Dead Sea Rift, northern Geochronology: Methods and Applications. AGU Reference Shelf 4, American
Israel. Tectonophysics 321, 237–252. Geophysical Union, pp. 61–76.

Biographical Sketch
Dr. Craig Sloss completed his PhD in Science at the University of Wollongong in 2005. He took up an academic
position as a coastal geomorphologist at Massy University, Palmerston North, from 2006 to 2008. He is currently
employed as a sedimentologist at the Queensland University of Technology. Dr. Sloss research interests include
sea-level and climate change over the last glacial cycle, coastal geomorphology, and sedimentology and extreme
events (storm surge or tsunami). He is also interested in the anthropogenic impacts on coastal and marine
environments and the role that humans have played in the historic modification of coastal geomorphology.

Dr. Kira Westaway came to Australia in 2003 to complete a PhD at the University of Wollongong by 2006. She
then took up a tenured lecturing position at Macquarie University, Sydney, in the Department of Environment
and Geography, where she has established a new luminescence dating facility for the Sydney region. Her research
interests include reconstructing the environments of archeological sites and establishing chronological frame-
works for archeological evidence in the southeast Asian region. As part of a larger research team, she provided
chronological control for the ‘Hobbit’ discovery at Liang Bua cave, western Flores, Indonesia, and is currently
establishing chronologies for modern human arrival from cave sites in southeast Asia.

Dr. Quan Hua is a Senior Research Scientist at the ANTARES AMS Facility in the Australian Nuclear Science and
Technology Organization (ANSTO). He obtained his PhD in Science at the University of Sydney in 2005. He has
worked at the AMS Facility since 1994. His research interests include radiocarbon calibration, and applications of
radiocarbon in geology, archeology, Quaternary research, and climate change studies. Dr. Hua is the author of 70
scientific publications.
 An Introduction to Dating Techniques: A Guide for Geomorphologists 369

Prof. Colin Murray-Wallace completed his PhD and DSc degrees in geology at the University of Adelaide. He is
currently the Head of the School of Earth & Environmental Sciences at the University of Wollongong and Editor-
in-Chief of Quaternary Science Reviews. His research interests include Quaternary sea-level changes, neotectonics,
coastal evolution, and amino acid racemization dating.
 14.31 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment
AC Kemp, Yale University, New Haven, CT, USA
AR Nelson, United States Geological Survey, Golden, CO, USA
BP Horton, University of Pennsylvania, Philadelphia, PA, USA, and Nanyang Technological University, Singapore
r 2013 Elsevier Inc. All rights reserved.

14.31.1 Introduction 370

14.31.2 Growth, Deposition, and Decay of Tidal-Marsh Plants 371
14.31.3 Radiocarbon Dating of Plant Macrofossils 372
14.31.4 Building Chronologies by Interpreting Ages 374
14.31.5 Examples of Radiocarbon Dating of Plant Macrofossils in Coastal Sequences 375
14.31.5.1 Recent Sea-Level History in the North Atlantic 376
14.31.5.2 Great Earthquake Frequency and Size at the Cascadia Subduction Zone of Western North America 376
14.31.5.3 Tsunami History beneath Tidal Marshes of Northern Japan 380
14.31.5.4 Hurricane History in Eastern North America – a Record of Deposition and Erosion 380
14.31.6 Recommendations for Selection of Plant Macrofossil Samples 381
14.31.7 Recommendations for Sample Preparation 383
Acknowledgments 385
References 385

Abstract

Tidal-marsh sediment is an archive of Holocene environmental changes, including movements of sea and land levels, and
extreme events such as hurricanes, earthquakes, and tsunamis. Accurate and precise radiocarbon dating of environmental
changes is necessary to estimate rates of change and the recurrence interval (frequency) of events. Plant macrofossils
preserved in growth position (or deposited soon after death) in tidal-marsh sediment are ideal samples for dating such
changes. In this chapter, we focus on the selection of plant macrofossils for radiocarbon dating and the application of ages
from different types of macrofossils to varied research projects, and make recommendations for selection and preparation
of tidal-marsh samples for dating.

14.31.1 Introduction et al., 1991, 2005; Nelson et al., 1995; Atwater and Hemphill-
Haley, 1997; Cisternas et al., 2005), and tsunamis (Nanayama
Organic and clastic sediment continuously deposited on tidal et al., 2003; Kelsey et al., 2005; Bourgeois et al., 2006; Sawai
marshes for decades to centuries can form sedimentary se- et al., 2009). Such events need to be accurately and precisely
quences that are meters thick along low-energy coasts. Burial dated to estimate variability of event recurrence and magni-
in anaerobic tidal environments promotes preservation of tude through correlation of evidence among multiple coastal
identifiable plant parts, termed ‘macrofossils’ (Warner, 1988; sites (Bourgeois et al., 2006; Nelson et al., 2006; Hawkes et al.,
Birks, 2002). The precision and accuracy of chronologies de- 2010). Improved coastal histories help in understanding pro-
veloped for coastal sediment sequences have improved since cesses and rates of coastal evolution, predict future changes,
the late 1980s through accelerator mass spectrometer (AMS) and implement management strategies that will reduce risk to
radiocarbon (14C) dating of individual macrofossils from coastal populations (Nicholls et al., 2007).
tidal-marsh deposits (Atwater et al., 1991; Törnqvist et al., Since the advent of AMS dating more than two decades
1992; Marshall et al., 2007). Sedimentary sequences beneath ago, plant macrofossils have been a preferred sample type for
tidal marshes record sea-level changes associated with degla- radiocarbon dating (Nelson, 1992; Törnqvist et al., 1992;
ciation (Shennan, 1986; Törnqvist et al., 2004) and allow Oldfield et al., 1997; Wohlfarth et al., 1998). AMS can meas-
high-resolution reconstructions of sea-level rise during recent ure 14C in milligram-sized samples and therefore allows dat-
climate changes (Donnelly et al., 2004; Gehrels et al., 2005; ing of identifiable remains (such as rhizomes, fruits, stems,
Kemp et al., 2009). Radiocarbon-based chronologies have and leaves) from individual plants preserved in tidal-marsh
improved estimates of the timing of major coastal events, sediment (Törnqvist et al., 1990; Gehrels et al., 1996; van de
such as hurricane landfalls (Donnelly et al., 2001a, 2001b; Plassche et al., 1998). Dating individual macrofossils from a
van de Plassche et al., 2004, 2006), great earthquakes (Atwater single bed potentially increases the resolution of stratigraphic
studies to the subcentimeter level. AMS ages from in situ
(found in growth position) plant macrofossils are inherently
Kemp, A.C., Nelson, A.R., Horton, B.P., 2013. Radiocarbon dating of plant
macrofossils from tidal-marsh sediment. In: Shroder, J. (Editor in Chief),
more accurate than ages from bulk (undifferentiated) organic
Switzer, A.D., Kennedy, D.M. (Eds.), Treatise on Geomorphology. Academic sediment such as peat, which commonly consists of in situ as
Press, San Diego, CA, vol. 14, Methods in Geomorphology, pp. 370–388. well as transported macrofossils and unidentifiable carbon in

370 Treatise on Geomorphology, Volume 14 http://dx.doi.org/10.1016/B978-0-12-374739-6.00400-0

 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment 371

the surrounding sediment matrix from varied sources. Even sediment. The rate of vertical accretion must at least equal the
following extensive laboratory treatments prior to dating, bulk rate of sea-level rise for a tidal marsh to maintain its elevation in
samples may contain carbon of significantly different ages the intertidal zone (Reed, 1995; Orson et al., 1998). Stems and
(van de Plassche, 1982; Mook and van de Plassche, 1986), leaves of tidal-marsh vascular plants dampen waves and slow
whereas carefully cleaned macrofossils usually contain carbon down currents to promote deposition of fine-grained clastic
from a few years or less. AMS-dated plant macrofossils yield sediment carried in suspension. Organic material is added to
ages as much as 2000 years younger than undifferentiated tidal-marsh sediment through accumulation and decay of
sediment from the same level in lake and bog sediment se- aboveground and belowground plant parts (Allen, 2000),
quences (Turney et al., 2000; Nilsson et al., 2001; Grimm which, through the binding action of roots, makes tidal-marsh
et al., 2009). In tidal environments, paired ages from macro- sediment resistant to erosion. The varying organic and clastic
fossils and the peat in which they were located can be similar proportions of tidal-marsh sediment accumulated either in situ
(Törnqvist et al., 1990) or differ by more than 1000 years or after being transported (Figure 1).
(van de Plassche, 1982; Nelson, 1992). Tidal-marsh plants are adapted to anaerobic conditions
In this chapter, we focus on the collection, selection, and (during submergence) and tolerate brackish or marine water.
preparation of plant macrofossils from tidal-marsh sediment Grasses (e.g., Spartina spp. and Distichlis spicata), rushes (e.g.,
for radiocarbon dating. We also summarize aspects of radio- Juncus spp.), and sedges (e.g., Carex spp. and Scirpus spp.) are
carbon dating that are important for interpreting ages from dominant forms of vegetation, although shrubs (e.g., Iva
tidal-marsh macrofossils and make recommendations for se- fructescens) and forbs (herbaceous flowering plants that are not
lection and preparation of typical samples. To illustrate how grasses, rushes, or sedges, e.g., Salicornia spp.) are also com-
selection of macrofossil samples varies with study objectives, mon, particularly at the transition between upland and high-
we summarize the sample selection and preparation pro- tidal marsh. The most common species of tidal-marsh plants
cedures used in six coastal dating studies. are perennial (with seasonal growth) and grow from elongate
rhizomes that extend parallel to, and within a few centimeters
of, the tidal-marsh surface (Eleuterius, 1976, 2000; van de
14.31.2 Growth, Deposition, and Decay of Tidal- Plassche et al., 1998). Rhizomes may support hundreds of
Marsh Plants vertical stems. Finer roots extend downward from the rhizome
and can reach depths of 2 m or more (van de Plassche, 1982).
On low-energy, temperate coasts, tidal marshes mark the tran- In the case of Juncus roemerianus, rhizomes live for approxi-
sition between terrestrial and marine environments. Differing mately 3 years (Eleuterius, 1975).
preferences and tolerances of vascular plant species to the fre- Following death, rhizomes and the subsurface parts of stems
quency and duration of inundation by saltwater (a function of remain in growth position and decompose primarily through
elevation in the intertidal zone) produces the distinctive zon- microbial activity (Benner et al., 1991). Aboveground parts
ation of tidal-marsh floral communities (Figure 1; Chapman, (mainly leaves and aboveground stems) of tidal-marsh plants
1960; Redfield, 1972). These communities keep pace with accumulate on the marsh surface where their high cellulose
moderate rates of sea-level rise through vertical accumulation of content promotes decomposition (Benner et al., 1987;

Transported carbon (dissolved or particulate)

delivered from upland. Decay-resistant
macrofossils (e.g. seeds, needles) deposited
on marsh surface are good for dating
(more common in high marsh)
Carbon (dissolved and
Upland
particulate) transported
among tidal marsh environments Leaves and stems
deposited on
marsh surface Transported carbon (dissolved or
High marsh particulate) delivered by tides,
Low marsh most common in low marsh
Rhizomes High tide

Fine roots
Low tide
Rhizomes, stalks, and roots
in growth position Sub-tidal

Figure 1 Sources of carbon in tidal-marsh environments. In-place carbon, ranging from fine particles to large macrofossils, may come from
decay of plant parts above (e.g., leaves and stems) or below (e.g., rhizomes, stalks, and roots) the marsh surface. Transported carbon (gray
arrows) may be washed in from the soils of surrounding uplands or be carried by tides from nearby parts of the tidal marsh or distant sites.
This carbon may consist of small particles or be dissolved in water. Decay-resistant plant macrofossils, especially those from upland
environments bordering the marsh, are commonly preserved well enough to be used for dating marsh surfaces. Symbols courtesy of the
Integration and Application Network, University of Maryland Center for Environmental Science.
372 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment

Ember et al., 1987; Fogel et al., 1989). Aboveground plant However, analysis using AMS takes only minutes compared to
detritus is susceptible to transport, unless burial occurs rapidly days for decay counting. In a practical sense, decay counting is
after deposition. As the duration of subaerial exposure increases often constrained by the mass of material available (Grimm
toward the upper edge of tidal marshes, oxidation of organic et al., 2009). Because AMS directly counts 14C atoms, it requires
material is more rapid and complete than in the lower parts of a sample with only 103–104 as much carbon as is needed for
marshes. However, in the high-tidal marsh, macrofossils decay counting (Fairbanks et al., 2005). In decay counting,
suitable for dating are still more abundant than in the low grams of sample material are required, whereas, for AMS dat-
marsh because the more rapid rate of decay is compensated for ing, 2–25 mg of dry plant tissue is sufficient (although AMS
by a greater density and diversity of plants. dating of samples containing as little as 25 mg of carbon is
In addition to carbon derived primarily from plants possible). The dramatic reduction in sample size made possible
growing at a site (Chmura and Aharon, 1995), tidal marshes by AMS has greatly expanded the types of datable sample ma-
receive carbon transported as particulate or dissolved material terials (e.g., foraminifera, single seeds, or pollen grains). Both
from fresh, brackish, and marine environments (Figure 1; decay counting and AMS yield conventional 14C ages; that is,
Lamb et al., 2006; Gebrehiwet et al., 2008). Plant macro- ages following the conventions of the radiocarbon community
fossils, including leaves and seeds dispersed by wind or water recommended by Stuiver and Polach (1977) and discussed by
(especially during storms or floods), are transported to tidal- van der Plicht and Hogg (2006) and Burr et al. (2007).
marsh surfaces from distances of a few meters to tens of The degree to which an individual sample meets the geo-
kilometers. Detrital macrofossils deposited within 1 or 2 years physical and geochemical assumptions of radiocarbon dating
after death may accurately date tidal-marsh surfaces. Where is assessed by statistical and experimental factors, sample
dense forest borders a tidal marsh, decay-resistant fruits, composition factors, and systemic factors (Taylor, 1987).
leaves, and needles of trees may be common in high-marsh Analytical precision (statistical and experimental factors) of
sediment and are characteristically a preferred type of macro- radiocarbon ages, compared with the uncertainties of sample
fossil sample for radiocarbon dating (Figure 1). selection and interpretation, is now insignificant for most ages
(Hatte and Jull, 2007; Bronk Ramsey, 2008b). Precisions of
70.2–0.5% are typical of modern AMS measurements
14.31.3 Radiocarbon Dating of Plant Macrofossils (equating to reported errors of approximately 730 14C years
on late Holocene samples) and ongoing inter-laboratory
A series of reviews (Mook and van de Plassche, 1986; Taylor, comparisons provide a transparent assessment of analytical
1987; Taylor et al., 1996; Cook and van der Plicht, 2007; Hatte accuracy and reliability (Burr et al., 2007; Scott et al., 2007).
and Jull, 2007; Jull, 2007a, 2007b; Bronk Ramsey, 2008b; Hua, Many laboratories commonly report (as 1s, one standard
2009; Trumbore, 2009) on the development of radiocarbon deviation) the larger of: (1) Poisson analytical error, plus any
dating aimed at archeologists, geologists, and other Quaternary additional laboratory-determined random error; or (2) the
scientists provides a wealth of information on production and error derived from the standard deviation of individual AMS
cycling of atmospheric 14C, the analytical process of estimating measurements of the 14C sample target (typically six to nine
sample age, and calibration of 14C ages. We focus this review on measurements; Jull, 2007a). Debate continues as to whether
aspects of radiocarbon dating especially relevant to plant or not these procedures fully account for all uncertainty (Burr
macrofossils from tidal marshes. Current information on ana- et al., 2007; Scott et al., 2007). Samples dated prior to the mid-
lytical techniques and sample requirements is available at la- 1990s may have underestimated age errors (Taylor, 1996).
boratory websites (e.g., National Oceanic Sciences Accelerator Depending on study objectives (e.g., along-coast correlation of
Mass Spectrometer Facility; Oxford University – Research Lab tsunami or hurricane deposits), however, overestimating age
for Archaeology and Radiocarbon Accelerator Unit; Rafter errors may create as many interpretive problems as under-
Radiocarbon Laboratory) and in the journal Radiocarbon. estimating errors (Burr et al., 2007).
Development of chronological frameworks using radio- Sample composition factors include fractionation of carbon
carbon dating (Figure 2) involves two main components: (1) isotopes through biochemical processes and contamination of
estimating the degree to which geochemical and geophysical a sample with carbon whose age differs from that of the in-
assumptions of radiocarbon dating methods are true for a tended sample. In calculating a radiocarbon age, it is assumed
particular sample (this section); and (2) interpreting the that the dated material was in quasi-equilibrium with the at-
stratigraphic or geomorphic context of dated samples and their mospheric ratio of carbon isotopes and that the 14C/12C ratio
relationship to events of interest (Figure 2; Sections 14.31.4 and in a sample has changed since death only through radioactive
14.31.5; Waterbolk, 1971; Taylor, 1987; Bronk Ramsey, 2008b). decay. Fractionation of carbon during photosynthesis causes
Radiocarbon laboratories measure the amount of 14C in a plant tissues to have a reduced 13C/12C ratio (expressed as
sample by decay counting or AMS. Decay counting measures d13C, parts per mil, %) compared to the atmosphere in which
the number of beta particles produced by 14C decay in a sample they grew. Values of d13C are expressed relative to the PeeDee
through gas-proportional counting of CO2 or through liquid- Belemnite (PDB) standard, which is from the Cretaceous
scintillation counting of benzene (Cook and van der Plicht, limestone formation of the same name in South Carolina,
2007). AMS separates 12C, 13C, and 14C on the basis of their USA. In tidal marshes, plants using the C4 photosynthetic
respective masses and directly measures the ratio of 12C and 13C pathway (e.g., Spartina spp.) create plant tissues with d13C from
isotopes to 14C (Jull, 2007a). Both methods provide similar  17% to  9% (Chmura and Aharon, 1995). By contrast,
accuracy and precision, given sufficient sample size and plants utilizing the C3 photosynthetic pathway (e.g., J. roe-
counting time (Walker et al., 2001; Bronk Ramsey, 2008b). merianus) have a d13C signature of  34% to  23% (Chmura
 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment 373

Life cycle of Sample collection, selection and

Data analysis
plant macrofossil preparation of plant macrofossils

Reported data Sample analysis at radiocarbon laboratory

Sample preparation(14.31.7).
Sample selection. Chosen to Physical cleaning under Contamination
meet study objectives with a magnification to isolate the (e.g. dust in sample vial,

Submission guidelines
balance between the best macrofossil, remove younger handling sample)

(e.g. sample weight)

identifiable macrofossils and rootlets and older sediment
the need to date specific particles. Dry at <50 °C to
stratigraphic intervals. avoid fractionation. Weigh.

by younger or older carbon

Effect of contamination
Radiocarbon analysis.
Storage. Cool, dry and dark (14.31.3)
Contamination
conditions inhibit bacterial and
(e.g. bacterial and fungal growth, fungal growth which adds
storage medium) younger carbon to fossil.
Radiocarbon age with
lab error.
Reported in 14C yrs following
Stuiver and Pollach (1977).
Collection (14.31.6). Material
from cores or sections,
Contamination following stratigraphic
(e.g. sediment smearing) investigation and selection of Calibration. Convert

Stratigraphic and geomorphic context

interval for detailed analysis. ‘True’ age (difference between time of radiocarbon age to calibrated
Detailed description of fossil. age. Choice of calibration
Evaluate geomorphic setting.
curves and software.
Growth position?

Estimate proportion of carbon in

sample from different reservoirs
Deposited on marsh surface?
death and collection)

Relationship to other samples.

Post-depositional processes. Calibrated age.
Macrofossil decays and may Estimated time of death
Contamination by particulate
be transported out of growth incorporating lab and
and dissolved C from varied
position or vertically reworked. calibration errors. Reported in
sources. Possible C fractionation
(e.g. burning; figure 1) years before present (AD 1950)

Death of organism. Interpretation of ages (14.31.4).

Reference collection (14.31.6). Relationship between this time Use stratigraphic and
Description of plant parts and dating target (e.g. tsunami, geomorphic context, other
likely to be found as hurricane, sea-level change). ages, and knowledge of plant
macrofossils and their macrofossils derived from
modern environment. Additional ages.
relationship to marsh surface.
Radiocarbon and others.
Living organism (14.31.2). Life cycle
Carbon Reservoirs Development of
Assumed equilibrium with C Examples.
(e.g. atmosphere, ocean) chronological framework.
correction?

in atmopshere (C3 vs C4).

Reservoir

Use ages and all available Other studies, all

Geomorphic position of plant pertinent information.
parts, lifecycle (e.g. seed dist.). information with specialist
chronologic software to
answer research questions.

Change in 14C production over time, storage and exchange between reservoirs

Figure 2 Conceptual radiocarbon dating flowchart showing stages in the growth, selection, collection, pretreatment, laboratory analysis, and age
interpretation of plant macrofossils from tidal-marsh sediments. The section numbers referenced in some boxes refer to sections of this chapter.
Handling of samples by the radiocarbon dating laboratory is beyond the scope of this paper. Descriptions of submission requirements, pre-
treatment, preparation and measurement are available on laboratory websites (e.g., NOSAMS) and in specialist literature. Most tidal-marsh plant
macrofossils undergo standard acid-base–acid treatment. Based on Figure 2 in Bronk Ramsey, C., 2008b. Radiocarbon dating: revolutions in
understanding. Archaeometry 50, 249.

and Aharon, 1995). If unaccounted for, this fractionation Contamination of a sample with carbon younger or older
could yield a difference of as much as several hundred years in than the sample is less important in dating plant macrofossils
a radiocarbon age. However, almost all radiocarbon labora- than in many other types of 14C samples because con-
tories minimize the error associated with isotopic fractionation taminants in macrofossils are often easily identified and re-
by correcting radiocarbon measurements using an assumed or moved (van de Plassche, 1982; Mook and van de Plassche,
measured d13C value (either from an aliquot of CO2 recovered 1986; Törnqvist et al., 1992; Wohlfarth et al., 1998). Through
during sample combustion or in the accelerator). Fraction- dissection and physical cleaning of the sample in water at
ation-corrected radiocarbon ages are termed ‘normalized’, magnifications of 10–50  all sample parts that might be
meaning that the reported activity has been adjusted to a d13C contaminants, such as rootlets or their decayed remains, are
of  25% with respect to the PDB standard. removed followed by low-temperature (o50 1C) drying
374 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment

(Section 14.31.7). The acid-base–acid pretreatment routinely temporal changes in 14C production. These advances are com-
applied by radiocarbon laboratories removes residual car- parable in importance with the development of AMS 14C
bonate and soluble organic material, such as fulvic and humic measurement (Reimer and Reimer, 2007). Calibration curves
acids. Most identifiable macrofossils are large enough to retain based on exhaustive studies of tree rings provide continuous
sufficient carbon for analysis, even in special cases where annual to decadal calibration back to 12.4 ka BP and this re-
additional or stronger chemical treatments are desirable. cord extends with less detail to 26 ka BP using corals and for-
However, moist samples that have been stored for long peri- aminifera (Reimer et al., 2004; Fairbanks et al., 2005). Data sets
ods, even if kept cool, may be contaminated with modern specifically for calibration of Southern Hemisphere radiocarbon
carbon from bacteria or fungal spores growing in sample ages (McCormac et al., 2004) and post AD-1950 samples
material (Wohlfarth et al., 1998). In selecting samples for (Levin and Kromer, 2004) are available. Regional data sets
radiocarbon dating, the term ‘contamination’ is relative. For provide a means to estimate the marine reservoir correction
example, roots extending down into a tidal-marsh peat from a required for calibration of radiocarbon ages on materials that
higher stratigraphic level are often removed from the peat as formed in the ocean (Reimer et al., 2004).
younger contaminants (van de Plassche, 1982). However, if Terrestrial plant macrofossils are strongly preferred as 14C
the dating objective was to determine a minimum age for peat sample material because they have exchanged carbon with only
deposition and no other identifiable macrofossils were avail- the atmospheric reservoir. Dating them avoids the difficulties
able, the roots might be dated and peat fragments adhering to associated with estimating the degree of 14C depletion in
the roots would be removed as contaminants. aquatic or marine samples, which typically incorporate carbon
Systemic factors in developing chronological frameworks derived from a mixture of nonatmospheric reservoirs, such as
include variations in 14C production through geological time, the ocean, estuaries, and lakes (Törnqvist et al., 1992; Wohlfarth
variability of 14C concentrations in carbon reservoirs (e.g., et al., 1998; Nilsson et al., 2001). For example, Hatte and Jull
atmosphere and ocean), and the rates at which 14C is ex- (2007) recommended against dating parts of floating plants,
changed among reservoirs (Taylor, 1987). 14C (six protons and such as the Potamogeten spp. seeds common in some tidal-
eight neutrons) is produced continually in the upper atmos- marsh deposits, because floating plants absorb older carbon
phere by collision of cosmic rays with 14N (seven protons and from CO2 dissolved in the water in which they float. A similar
seven neutrons) atoms. Small fluctuations (wiggles) in at- depletion in measured 14C concentration might occur in
mospheric 14C production over decades to a few centuries ground-hugging plants on tidal marshes if CO2 derived from
(termed De Vries effects) are superimposed on major changes much older decaying peat beneath the marsh mixed with the
over thousands of years. These changes are primarily driven by atmospheric carbon photosynthesized by plants (Nilsson et al.,
changes in cosmic-ray flux and variability in Earth’s magnetic 2001; Reimer and Reimer, 2007). We discuss the possibility of
field that serves to shield the atmosphere (Burr, 2007). Burn- this process affecting ages in Section 14.31.5.2.
ing of fossil fuels and aboveground testing of nuclear weapons Calibrated dates are expressed as a time interval (usually
have strongly influenced atmospheric 14C concentrations over 2s) moving forward in calibrated years, and are also termed
the past three centuries (Hua, 2009). Atmospheric 14C is ‘calendar years’ (e.g., 2340–2170 cal. yr BP). Variability in at-
mixed and dispersed rapidly within hemispheres, but at a mospheric 14C concentrations on timescales of decades to
slower rate between hemispheres because the intertropical centuries adds significant uncertainty to calibration of
14
convergence zone acts as a barrier to mixing. C ages, particularly for time periods where the calibration
When ocean surface water, which exchanges carbon with curve slopes gently (plateaus). For most ages, these un-
the atmosphere, sinks into the deep ocean, its 14C is isolated certainties are sufficiently large that improved analytical pre-
from the atmosphere for hundreds to thousands of years cision in measurements at 14C laboratories would not yield a
(average age of deep water is 800 years). When this water corresponding increase in the precision of calibrated dates.
returns to the surface through upwelling, radioactive decay has However, in some cases, closely spaced dates are able to ex-
depleted its 14C concentration relative to the atmosphere. This ploit these small fluctuations of the calibration curve to obtain
marine reservoir effect gives the upwelled water, and organisms ages of unusual precision – a technique termed ‘wiggle
that source their carbon from it, an older radiocarbon age than matching’. In an example discussed further in Section 14.31.5
organisms of the same true age who took their carbon from (Figure 3), Marshall et al. (2007) eliminated age ambiguities
the atmosphere. The average age difference is 400 years, but caused by the broad plateau in the calibration curve for the
varies by at least many hundreds of years at different locations past 300 years by matching seven ages on detrital fragments of
and over thousands of years (Reimer and Reimer, 2007). tidal-marsh grass with wiggles on the curve. This matching
Southern Hemisphere tree rings from the last 1000 years were reduced 2s errors on individual ages from 200–300 years to
up to 100 years older than those from the Northern Hemi- about 10–90 years.
sphere because of hemispheric differences in the atmospheric
reservoir of radiocarbon (McCormac et al., 2004). Supply of 14.31.4 Building Chronologies by Interpreting Ages
older, oceanic, CO2 to the atmosphere is likely larger in the
Southern Hemisphere because the area of ocean is larger than The issues at stake here have been well understood for decades;
in the Northern Hemisphere (Braziunas et al., 1995). however, there are many instances in which they are ignored. In
Over the past 40 years improved records of atmospheric 14C some ways, the much easier access to radiocarbon dates now can
concentration through geological time, coupled with develop- tempt people not to think in detail about each date and simply
treat large numbers of dates as data to be summarized (Bronk
ment of specialist calibration software, have enabled Quater-
Ramsey, 2008b).
nary scientists to make corrections for reservoir effects and
 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment 375

0 300 2.2
Manually selected calibrated age range Bomb-spike samples
AMS radiocarbon age
(all intervals shown in panel d) (errors are smaller than symbol)

Radiocarbon age (14C years BP)

High-precision AMS radiocarbon age (1 error for 14C age, 2 error for 2.0

Fraction modern radiocarbon

(1 error bars) 250 calibrated age) Northern hemisphere
20 Calibration curve zone 1 calibration curve
Depth in core (cm)

1.8
200
40 1.6
150 26 cm 25 cm
1.4 16 cm
60 11 cm
100 8 cm
1.2
4 cm

80 50 1.0
0 50 100 150 200 250 300 1600 1700 1800 1900 2000 1950 1960 1970 1980 1990 2000 2010
(a) Radiocarbon age (14C years BP) (b) Calibrated age (AD) (c) Calibrated age (AD)
0

Spheroid carbonaceous particles Rejected 2 calibrated age range

Pollen age estimate Accepted 2 calibrated age range (panel b)

20 Calibrated age for bomb-spike sample (panel c)
Depth in core (cm)

40

60

80
1600 1700 1800 1900 2000
(d) Calibrated age (AD)

Figure 3 Dating of a 0.8-m section of tidal-marsh sediment at Poole Harbor (southern England). (a) Depth distribution of AMS 14C ages (open
circles) and high-precision (replicate targets analyzed with 2% instrumental precision) ages (closed squares). All ages were measured on detrital
fragments of fragile parts of above-ground parts of tidal-marsh plants. (b) Radiocarbon ages from (a) manually adjusted along the x-axis to fit
the radiocarbon calibration curve; stratigraphic relations among samples helped in the adjustment. Calibrated age errors show only the accepted
age range for each calibrated age (see panel (d)). This technique of matching a group of ages in sequence to the calibration curve is termed
‘wiggle matching’. (c) Calibration of bomb-spike 14C ages using post-AD 1950 14C concentration values for the Northern Hemisphere. Values of
14
C concentration (measured as fraction of modern 14C, where 1 ¼ AD 1950) were matched by horizontally adjusting fraction modern values to
match the shape of the calibration curve (Northern Hemisphere data set). A group of samples whose stratigraphic relations are known is needed
to determine which side of the peak (AD 1963) a sample lies on. Labels indicate sample depths. This technique is termed ‘bomb-spike wiggle
matching’. (d) An age–depth model for the 0.8-m section. Gray bars show the rejected ranges of the pre-AD 1950 ages’ intercepts with the
calibration curve; accepted ranges are black. Post-AD 1950 bomb-spike ages are shown by filled circles. The age model marked with the dotted
line (which joins midpoints of accepted age ranges) is supported by independent age estimates based on pollen assemblages (star;
corresponding to an eighteenth-century rise in Pinus and rapid expansion of Spartina at AD 189575 years) and the appearance of spheroid
carbonaceous particles (open squares) caused by high-temperature combustion of fossil fuels. Reproduced from Marshall, W.A., Gehrels, W.R.,
Garnett, M.H., Freeman, S.P.H.T., Maden, C., Xu, S., 2007. The use of ’bomb spike’ calibration and high-precision AMS C-14 analyses to date
salt-marsh sediments deposited during the past three centuries. Quaternary Research 68, 325–337.

Developing an accurate history for a site or coastal region methods, error distributions, nonuniform age distributions,
from a series of 14C ages requires evaluation of all available and varied models of sediment deposition (Heegaard et al.,
information about 14C samples and their stratigraphic or geo- 2005; Yeloff et al., 2006; Bronk Ramsey, 2008a; Parnell et al.,
morphic context (Waterbolk, 1971; Taylor, 1987; Bronk Ram- 2008; Blaauw, 2010). As research objectives, 14C sample com-
sey, 2008b). The development and wide distribution of positions, and sample context vary widely, modeling strategies
software (e.g., CALIB, OxCal, BCal, DateLab, and CalPal) to are best developed through scrutiny of examples, such as those
handle age calibration, account for some reservoir effects, and cited previously and in Section 14.31.5.
provide objective methods for evaluating ages (Stuiver and
Reimer, 1986; Bronk Ramsey, 2001; Reimer et al., 2004) is a
major advance in the application of radiocarbon dating. Soft- 14.31.5 Examples of Radiocarbon Dating of Plant
ware with robust underlying statistics helps users build inte- Macrofossils in Coastal Sequences
grated chronologies by evaluating ages (and their associated
uncertainties) within the context of additional information The examples in this section show a breadth of research topics
about dated samples, including results from other dating that have been investigated using radiocarbon-dated plant
376 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment

macrofossils from tidal marshes. They demonstrate the im- This bomb-spike type of 14C wiggle-matching calibration
portance of sample selection in answering research questions provides very precise ages, but requires multiple samples with
and illustrate some of the techniques that have been used to known stratigraphic ordering to determine on which side of
interpret radiocarbon results. the AD 1963 14C peak a sample lies. Instead of a radiocarbon
age, bomb-spike calibration uses values of 14C concentration
compared with fraction modern carbon (Fm), where a value of
14.31.5.1 Recent Sea-Level History in the North Atlantic
1 is equal to atmospheric concentration in AD 1950 and 1.99
Studies reconstructing late Holocene sea-level changes com- is the peak value (AD 1963). This study compared 14C ages
monly rely on radiocarbon ages from continuous sequences of with the results of other dating methods from the same core
high tidal-marsh peat in a single core to generate high- (137Cs, 210Pb, and changes in pollen assemblages and spheroid
resolution records of sea-level change. Recent methods for carbonaceous particle concentrations with depth) to select the
assimilating chronological information (such as age estimates most likely of several possible calibrated age ranges for each
from other dating methods or historical markers) produce sample (Figure 3(d)).
age–depth models that statistically predict the age (with un- At Hammock River Marsh (Connecticut, USA), van de
certainty) of sediment from any level (Bronk Ramsey, 2001; Plassche et al. (1998) reconstructed sea-level changes over
Heegaard et al., 2005; Parnell et al., 2008; Blaauw, 2010). the past 600 years from a 1.1-m long core of tidal-marsh peat
Foraminifera, diatoms, or plant macrofossils frequently pro- with 14 radiocarbon ages (Figure 4). The dated material
vide estimates of elevation relative to past sea level (Thomas included identifiable subsurface stalks and rhizomes of com-
and Varekamp, 1991; Varekamp et al., 1992; Nydick et al., mon tidal-marsh plants with a known relationship to modern
1995; van de Plassche et al., 1998; Gehrels et al., 2005, 2008; sea level. The study assumed that subsurface stalks of all spe-
Kemp et al., 2009). cies formed 0–2 cm below the marsh surface, D. spicata rhi-
Marshall et al. (2007) used two different types of radio- zomes within 4–12 cm of the surface, and stem-rhizome
carbon calibration in their investigation of sediment accu- transitions of Spartina alterniflora within 3–6 cm (4–8 cm for
mulation during the past three centuries at Poole Harbour in larger individuals). Other types of macrofossils (e.g., fine
southern England (Figure 3). In addition, they obtained age roots) were not dated because their depths relative to the
estimates from independent dating techniques and used all surface when living were unknown. The original sample
available chronological and stratigraphical information to depths (Figure 4(a)) were corrected for each type of macro-
propose an age–depth model for deposition of tidal-marsh fossil to estimate the depth of the paleomarsh surface
sediment. (Figure 4(b); van de Plassche et al., 1998). For each sample,
For radiocarbon dating, Marshall et al. (2007) selected the 14C laboratory prepared two separate targets and a
fragile, aboveground parts of tidal-marsh grass lying hori- weighted average of the AMS results was made. Following
zontal in muddy sediment on former tidal-marsh surfaces to calibration, van de Plassche et al. (1998) developed an
minimize the chance of analyzing samples reworked from age–depth model to identify the most plausible accretion
older sediment. Plant macrofossils of this type were probably history (Figure 4(c); van de Plassche et al., 1998). When
deposited on the marsh surface soon after death. Materials combined with foraminiferal data and other radiocarbon
such as these, with a known relationship to the surface of a dates, the resulting reconstruction of sea-level variability
former tidal marsh, are the most useful type of sample in spanned the past 1400 years and was correlated with climate
reconstructing a history of sediment accumulation and sea- changes including a sharp sea-level rise at about AD 1000 and
level rise. From one sampling interval in which a grass frag- falling sea level during the Little Ice Age.
ment could not be found, a fibrous woody plant stem was
selected. Although the stem would constitute an excellent
sample in many studies, its older-than-expected age indicated
14.31.5.2 Great Earthquake Frequency and Size at the
that it was reworked (by at least a decade) and that it
Cascadia Subduction Zone of Western North
should not be used in an age–depth model with this intended
America
degree of resolution. Marshall et al. (2007) obtained high-
precision AMS ages for five samples by preparing repeat (2–3) Radiocarbon-dated plant macrofossils were pivotal in showing
graphite targets, for which measurement of 14C activity ex- that the Cascadia subduction zone (western North America)
ceeded laboratory standards to reduce analytical uncertainty. ruptured multiple times in the past 5000 years during great
Individually, these samples yielded 1s errors of 715 years, but (magnitude 8–9) earthquakes. Organic-rich soil horizons of
when averaged gave errors of 79–11 years for each sample. tidal marshes and swamps, now buried beneath modern tidal
They then manually matched these, and three other ages of marshes, testify to the sudden subsidence of hundreds of
standard precision (725 years), to fit fluctuations (wiggles) of kilometers of Cascadia’s coast during great earthquakes
the 14C calibration curve (Figure 3(b)). Like all radiocarbon (Figure 5; Atwater, 1987; Nelson et al., 1995; Atwater and
ages from the past 300 years, samples in this study had mul- Hemphill-Haley, 1997; Clague et al., 2000; Atwater et al.,
tiple (up to eight) intercepts with the calibration curve span- 2004). Initial dating of these soils focused on decay counting
ning hundreds of years (Figure 3(d)). of large samples, including bulk peat, tree roots, and detrital
Marshall et al. (2007) exploited the addition of 14C to the sticks and cones. Inconsistent and uninterpretable ages on
atmosphere from aboveground testing of nuclear weapons, these materials, coupled with increasing availability of AMS
which peaked in AD 1963, to calibrate six post-1950 samples dating, pushed investigators toward careful selection and
(Figure 3(c); Shotyk et al., 2003; Hua and Barbetti, 2004). dating of identified plant macrofossils rooted in the tops
 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment 377

Core 0.8
lithology

0.6

0.4
Elevation (m NGVD)

0.2

0.0

Age on: Age on:

Spartina alterniflora Spartina alterniflora
−0.2
Distichlis spicata Distichlis spicata
Age-depth model
Scirpus robustus Scirpus robustus
0 100 200 300 400 500 0 100 200 300 400 500 2000 1800 1600 1400
14 14
Radiocarbon age ( C years BP) Radiocarbon age ( C years BP) Calibrated age (AD)
(1 error bars) (1 error bars)

Upper-marsh peat High-marsh peat Clay with plant fragments Erosive boundary

(a) (b) (c)

Figure 4 Sea-level reconstruction for the past 600 years at Hammock River marsh (Connecticut, USA; van de Plassche et al., 1998).
(a) Average of two laboratory-reported (uncalibrated) 14C ages for each sample. Depth of samples and species dated are shown. (b) Ages from
(a) following correction for macrofossil type and its depth below tidal-marsh surface. Vertical error bars show uncertainty in depth of macrofossil
type. (c) Age–depth model for core. Horizontal dimensions of rectangles are calibrated age ranges (2s) for dated samples; multiple boxes for the
same sample show multiple, discontinuous age ranges for that sample. Vertical dimensions show uncertainty in depth of different types of
macrofossils.

of suddenly subsided soils (Atwater and Yamaguchi, 1991; to o60 calibrated years (using OxCal software). Where w2
Nelson, 1992; Nelson et al., 1995). These ages showed the tests of nontree ring ages showed that ages were not signifi-
recurrence of great earthquakes and tested the synchroneity of cantly different (2s), they were averaged to reduce error
earthquake subsidence among coastal sites to infer earthquake prior to calibration. Although only maximum limiting
rupture extent and magnitude. ages were obtained for four of the seven soils, wiggle matching
Atwater et al. (2004) summarized earthquake history of outer and inner ring ages greatly reduced uncertainties i
for southwest Washington (USA) by selecting 40 radiocarbon n the timing of subsidence for five of the soils (Figure 5).
ages to date earthquake-induced subsidence of seven buried For the youngest soil, such wiggle matching dated its subsid-
soils that were reliably correlated in outcrops along 100 km ence to AD 1680–1720. Later dating (through matching of
of coast (Figure 5; Atwater and Hemphill-Haley, 1997). Dated tree-ring chronologies) showed that subsidence-induced tides
materials included rooted herbs, especially the decay-resistant, probably flooded trees at the same sites within months of
low-marsh plant Triglochin maritimum, branches of rooted January AD 1700. Although environmentally stressed trees
shrubs, and conifer cones. The majority of samples were may live for decades without growing new rings (Bigler and
groups of about 10 rings from spruce and cedar trees rooted Bugmann, 2004), trees suddenly submerged by coastal sub-
in the buried soils that died slowly following sudden subsid- sidence during the great AD 1964 earthquake in Alaska died
ence into the intertidal zone. Extended gas-proportional within 2 years.
counting of high-molecular-weight cellulose in these 60–90-g Nelson et al. (1995) used radiocarbon-dated herbs rooted
tree-ring-wood samples yielded ages with errors of 711–23 near the top of the youngest subsided soil to test for non-
14
C years (1s). Minimum ages from Triglochin rhizomes synchroneity of subsidence at seven sites along 500 km
in mud above two soils narrowed the times of subsidence of the Cascadia subduction zone (Figure 6). Radiocarbon ages
 Soil J Copalis L N
N S U W A,s,F Y
River 124°
Sites A
A 47° A A

378
Age in b B
Grays Harbor B F
Buried soil
exposed
but B
A C

Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment

D B C s

100 km
numerical
B Willapa Bay B C C
age lacking F
or Tidal
imprecise E marsh or C D D
C swamp
WA s
Columbia
River A s
x

OR x

F x

D x
E x
E,F x x

1227±30 (E) rhizomes Anomalous rings during

(a) Dated material relative to earthquake
Ages 1247±22 (B) rhizomes first decade after A.D.
1300±25 (C) rhizomes 1700, in trunk of
Numerical age in 14C yr B.P.
Younger 1598±23 (C) rhizomes 1302±21 (D) rhizomes surviving tree (s)

Older 3165±17 (B) cones 2846±17 (A) 16−24 2475±23 (B) cones 1720±25 (D) 1−15 1260±14 (C) herb 1080±40**(A) shrub Final ring A.D. 1699***
3166±18 (B) stick 2793±16 (C) 24−26 2540±25 (C) shrub 1710±17 (B) 18−22 1431±23 (F) 59−63 (F)
3180±30 (B) stick 2791±11 (B) 21−27 2472±23* (D) 62−71 1698±15 (E) 18−25 112±10 (A) 1−10
3177±16 (A) 48−54 2814±17 (E) 23−31 2550±60**(A) >71−>82 1696±15 (D) 22−24 152±30 (C) 1−20
3200±17 (C) 51−60 2811±16 (D) 27−31 2508±16 (A) >118−>124 1716±16 (A) 60−69 184±14 (B) 30 −35
2886±23* (F) 74−82 1740±15 (C) 60−69 189±20 (B) 30 −49
Matching of wiggles in calibration curve (c) limits 207±14 (A) 30 −49
Site mapped in a Numerical age of dated rings in tree root, in
age of event Y to 1680−1720 (d). rings averaging 5−40 211±14 (B) 30 −49
and keyed to d tree-ring years before death. Used in d to convert
years before tree death rule out event before A.D. 219±13 (C) 30 −49
radiocarbon age of the rings into the time of tree 199±14 (A) 35−44
(b) death. 1680. Earlier rings from site C (75−86 years before
death) exclude age later than A.D. 1720.

Event
J J L N S U W Y

3300 A.D. 1700

3000 2600 1500 1200
Calibration 1800 300 tsunami in
data Japan
3200 2900 Radiocarbon ages in b
2500 1400 1100
converted to calendar 1700 200
ages in d with the
3100
Age 2800 calibration data of 1300 1000
2400 1600 100
in 14C yr B.P. Stuiver et al. (1998)
3000
2700 2300 1200 900
0
Age, in calendar years before A.D. 2000

3500 3000 2500 2000 1500 1000 500 0

(c) A A A A A
B B A B B
Calendar A A
B C C C
ages B D B C C
C E C D D
F D E E B
Event age (>95% c.i.) F F
N Site in a
from averaged ages Maximum age for earthquake Minimum age for earthquake, measured on C
measured on detritus in buried soil remains of plants that colonized tidal flats
Time of death (T ) for trees (black) or shrubs (gray) killed by submergence soon *** Exact date from matching ring width patterns (event Y only; Yamaguchi et al. (1997)
after coseismic subsidence. T = C-A, where
C = Calibrated age of sample in cal yr before A.D. 2000 ± Error listed in b lacks lab multiplier, which is included in d. Multipliers: 1.6 (ages
A = Mean age of dated rings, in ring years before death, as counted (in trees) or from Seattle lab, no asterisk in b), 1.25 (Belfast*), and 2.0 (Beta**).
guessed (in shrubs). Ring counts are in italics in b.
(d)

Figure 5 History of great earthquakes during the last 3700 years along the Cascadia subduction zone inferred from tidal-marsh sediments. (a) Location of 14C ages pertaining to seven great
earthquakes. (b) Radiocarbon ages used minimum (above gray line) and maximum (below gray line) estimates of the timing of the seven earthquakes. Material dated is also described. (c, d) Calibrated
ages (note year 0 in the timescale used on this diagram is AD 2000) and estimation of the timing of each earthquake. Reproduced with permission from Figure 11 in Gillespie, A.R., Porter, S.C., Atwater,
B.F., (Eds.), 2003. The Quaternary Period in the United States. Elsevier, Amsterdam, ISBN: 978-0444514714.
 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment 379

130° 125° 120°

7
British Columbia Control 138±17 (177±46/91±41)

50° Va NORTH
nc 7
ou AMERICA
ve
rI
sla
PLATE Copalis 139±17 (165±44/67±42)
CA nd
SC
AD Puget
IA Sound
1
SUBDU

Washington
Copalis112±11 (125±18/103±19)
a
Columbi
CTION

45° (c)
JUAN DE FUCA 8 Bench exposing top of marsh buried in AD 1700
PLATE
Oregon Niawiakum 161±15 (272±47/120±41)
ZONE

PACIFIC
PLATE
8
0 100
Naselle 162±16 (232±45/121±44)
Km
California
40° Mendocino
Fracture Zone

Copalis River Nehalem 179±15 (275±44/102±41)

Willapa
Bay Niawiakum River
Naselle River (d)
3
Modern and fossil stem bases of Potentilla pacifica
Nehalem River
Nehalem 128±09 (132±11/124±14)
Netarts Bay
Seaward edge of Cascadia subduct

45°
Salmon River
8

Netarts 188±18 (212±41/133±63)

7 161±12
Coos Bay Salmon 157±17 (199±49/114±43)
Coquille River

Cape
ion zone

Blanco 8 181±11
Coquille 192±17 (242±45/69±71)

4
Mad River
Humboldt Bay Mad 124±14 (137±14/98±14)
0 100
(e)
1650

1700

1750

1800

1850

1900

1950

40° km
125° Automated pretreatment of Potentilla
Calibrated years AD pacifica stem base with KOH
(a) (b)

Figure 6 Means and calibrated probability distributions of AMS-dated plant macrofossils and trees dated by high-precision gas-proportional
counting at eight sites inferred to have been killed by sudden subsidence of the coast of the Cascadia subduction zone during the great
earthquake of AD 1700. (a) Location of the Cascadia subduction zone and the eight sites where plant macrofossils and trees were dated.
(b) Nelson et al. (1995) dated 7–8 rooted stem and leaf bases of P. pacifica (photo in (d)), J. balticus, or Carex sp. at seven sites in Oregon and
Washington. Cleaned bases were pretreated 3–6 times with automated equipment to remove contaminates (photo in (e)). Ages on bases,
collected within centimeters of each other at the upper contact of the marsh soil buried in AD 1700 (red line in B, photo in (c)), were averaged
to give a mean age with a reduced standard deviation. The range of the 7–8 ages (9 ages listed in 14C years BP with 1s errors) are shown in
parentheses to right of the mean age. Dating of the outer rings of spruce trees killed by subsidence at Copalis River and Nehalem are described
by Atwater et al. (1991) and the same methods were used to date rings on a tree from the Mad River site in California. A control sample of
outer rings from Copalis River was dated by the same AMS methods. Probability distributions for the calibrated means (methods of Nelson et al.,
1995) are shown for herbs (in green) and brown (for tree rings). Dashed vertical line and lighter colors on distributions to the right shows that
dated plants are unlikely to date from the historical period during which no great earthquakes have been reported. Herb AMS ages, reported in
1992–93, had standard deviations of þ 41–71 years; AMS analyses on similar samples done today would likely have errors of þ 25–35 years.
To show that increased precision would not result in narrower calibrated age ranges for samples about 300 years old, we assumed errors of
þ 30 years for all ages from the Salmon and Coquille rivers and recalibrated the mean ages. The probability distributions for the newly calibrated
means using ages with assumed much lower errors (shown in blue), in fact, span slightly wider time intervals than the original probability
distributions calculated in 1995.
380 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment

on tree-ring wood were compared with herb ages from the window, but OxCal narrowed the 2s range to 290 calibrated
same soil at the Copalis (Atwater et al., 1991) and Nehalem years (Figure 7).
sites. Wiggle-matched, ring-wood ages at an eighth site
(Mad River; Figure 6) did not differ significantly from other
ring and herb ages. Thus, this study failed to set limits on the 14.31.5.4 Hurricane History in Eastern North America – a
rupture extent and size of the most recent great earthquake Record of Deposition and Erosion
at Cascadia.
Nelson et al. (1995) used special care in selecting herbs for The study of past hurricanes using sedimentary records sup-
AMS radiocarbon dating. They assumed that herbs with intact plements observational and instrumental records of hurri-
stem or leaf bases rooted in the tops of the soil had died canes, which provide a limited perspective on the frequency of
within a few years of sudden subsidence (Atwater and the largest hurricanes. In New Jersey (USA), Donnelly et al.
Yamaguchi, 1991). Woody stem bases of Potentilla pacifica (2001a) interpreted buried sand layers within sequences of
(Figure 6(d)), distinctive rhizomes with vertical attached stem organic tidal-marsh sediment as hurricane overwash deposits.
bases of Juncus balticus, or the leaf bases of Carex lyngbyei At Whale Beach, abrupt contacts between three sand layers
were collected from horizontal benches, 0.5–2 m2 in area, and underlying tidal-marsh sediment recorded overwash
cut to expose the top of the soil at each site (Figure 6(d)). during hurricanes (Figure 8; Donnelly et al., 2001a). Eleven
At four sites, stems and leaves protruded into sand deposited samples of AMS radiocarbon-dated S. alterniflora from the base
by the tsunami accompanying the earthquake. After dissection of the sand units constrained the timing of overwashing dur-
and physical cleaning in distilled water under a microscope ing hurricanes. At three of the 11 dated levels, a replicate
at 10–50  magnification to remove all possible rootlets macrofossil of the same species was also dated. The authors
and decayed tissue, Nelson et al. (1995) dried seven to eight assumed that the death of dated plants was within a few years
samples from each site and used automated equipment of sand deposition. Rhizomes in the uppermost tidal-marsh
to pretreat each sample 3 times with HCl–KOH–HCl sediment whose stems extended into overlying overwash de-
(Figure 6(e)). If the final acid rinse of each sample showed posits were likely to have been killed during the hurricane.
any yellow discoloration (suggesting that all humic acids Tidal-marsh sediment beneath porous sand sheets deposited
may not have been removed), they ran another sequence of by hurricanes and tsunamis may be contaminated by down-
three pretreatments on the sample. Averaging the 7–8 ward percolation of groundwater containing younger carbon
ages from each site (on the assumption that all herbs (Switzer et al., 2006; Switzer and Jones, 2008). This source of
died within a few years of each other) reduced laboratory re- contamination is avoided by dating of plant macrofossils,
ported errors of 741–84 14C years on ages (1s) to 716–18 which ceased to assimilate carbon prior to the contamination.
14
C years. Cleaning and laboratory procedures are intended to
remove contaminating carbon adhered to plant macrofossil
samples.
Donnelly et al. (2001a) used inconsistencies with 137Cs
14.31.5.3 Tsunami History beneath Tidal Marshes of
dating and pollen chronohorizons to eliminate portions of
Northern Japan
calibrated age ranges (Figure 8). Using their selected age
Along the Pacific coast of northern Japan, layers of sand in ranges they identified the Ash Wednesday storm of AD 1962,
tidal-marsh peat at Kiritappu, Hokkaido, are evidence of the AD 1821 hurricane, and a prehistoric hurricane that oc-
15 tsunamis that struck between 200 and 6000 years ago curred between AD 1278 and 1438. These ages were consistent
(Sawai et al., 2009). This study extended the historical record with ages for overwash deposits in a Rhode Island tidal marsh,
of tsunamis and showed that many prehistoric tsunamis were where hurricane landfalls were recognized at AD 1295–1407
larger than any documented in the last 200 years. Sawai et al. and AD 1404–46 (Donnelly et al., 2001b).
(2009) used radiocarbon ages to more tightly constrain the An alternative to reconstructing hurricane history with
timing of tsunamis and show variability in their recurrence overwash deposits is to correlate distinct erosive contacts over
interval in this region, which averaged 400 years and ranged large areas of a tidal marsh or among marshes. During a
from 100 to 800 years. hurricane, even high-marsh peat near the upper edge of a
Radiocarbon ages from plant macrofossils, charcoal, and marsh may be eroded. When this occurs, the eroded areas are
insect remains immediately beneath and above tsunami- subsequently filled by a regressive sequence of tidal mud, low-
deposited sand layers provided bracketing maximum and marsh peat, and high-marsh peat (van de Plassche et al.,
minimum ages for each tsunami (Figure 7). Although plants 2006). In Connecticut, van de Plassche et al. (2004, 2006)
growing on the tidal marshes that were buried by sand may inferred that two widespread erosive contacts represented two
not have died immediately, the time lag between burial and hurricane landfalls during the last 600 years.
plant death was at most a few years. Younger tidal-marsh As with the earlier example of constraining the time of
plants then recolonized the new tidal-marsh surface above the tsunami deposition, van de Plassche et al. (2004, 2006) used
tsunami layer. Sawai et al. (2009) averaged multiple ages and minimum and maximum limiting ages to estimate the timing
then used stratigraphic ordering and Bayesian statistics (OxCal of hurricane erosion. An in situ rhizome and stalk of D. spicata,
calibration software) to narrow the estimated age for when recovered from immediately below the erosive contact, yielded
each sand layer was deposited. For example, two minimum a maximum age for erosion of 1120 14C years730 years
and four maximum ages placed the 14th tsunami identified at (Figure 9). In situ rhizomes and stalks of S. alterniflora from
Kiritappu Marsh (K14; Figure 7) within a 1000-year-long time the base of the low-marsh peat overlying the tidal mud
 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment 381

Salt-marsh peat Tsunami sand Mud

0

K14

20
K2

K3
min
40
max

K4
K5 OxCal calculated age
60
5.2 5.0 4.8 4.6 4.4
K6

80 K7

K8
Depth (cm)

100 K9

K10

120
K11

140 K12
Radiocarbon-dated sample

K13

160
K14
K14 limiting ages
(inset )
180
K15
K16

200
5 4 3 2 1
Calibrated age (1000 yrs BP)
Figure 7 In a core from Kiritappu marsh in northern Japan, Sawai et al. (2009) interpreted 16 sand layers (K1 through K16) as deposited by
tsunamis. Maximum and minimum ages for each sand layer were estimated using AMS dated samples below and above the sand layers,
respectively. Using multiple maximum (blue) and minimum (red) ages for the K14 layer, OxCal software narrowed the age range for tsunami
deposition to 290 calibrated years (inset, gray bar). Reproduced from Sawai, Y., Kamataki, T., Shishikura, M., et al., 2009. A periodic recurrence
of geologically recorded tsunamis during the past 5500 years in eastern Hokkaido, Japan. Journal of Geophysical Research 114, B01319.

deposited following erosion (Figure 9) likely grew within plants had recolonized the marsh between AD 1640 and
3–8 cm of the contemporary tidal-marsh surface (van de 1670, which was in agreement with hurricane landfalls in
Plassche et al., 1998). These macrofossils gave minimum southern New England in AD 1635 and 1638 (Donnelly et al.,
ages for erosion of 220725 and 241735 14C years BP. 2001b).
Van de Plassche et al. (2004) inferred that the difference be-
tween the maximum age from high-marsh peat and minimum
ages from low-marsh peat confirmed that the contact recorded 14.31.6 Recommendations for Selection of Plant
significant erosion and shows the time span of the sediment Macrofossil Samples
eroded from the sequence. A radiocarbon age on a S. alterni-
flora rhizome and stalk at the transition from low-marsh This section includes general recommendations for field
to high-marsh peat provided a constraint on the time taken sampling and selecting plant macrofossils for AMS dating
for tidal-marsh floral communities to recover from disturb- (Figure 2). For guidance on specialized dating of macrofossils,
ance. Calibration of these ages showed that tidal-marsh such as wiggle matching of samples from tidal swamp trees or
382 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment

WB9 WB1 WB7 WB8 are sampled with plastic or metal sleeves (monoliths) pushed
0 into, and then cut out of, the section or with a series of hand-
excavated blocks. Sections must be thoroughly cleaned prior
to sampling to ensure that stratigraphy is not affected by
50 weathering, slumping, or postdepositional draping. Photo-
graphs of the position of the sample before and after col-
Depth (cm)

lection are helpful. At unusual sites with sufficient personnel,

100 3,4 5,6 vertical outcrops that are meters high can be described and
7 sampled in the field (Atwater and Hemphill-Haley, 1997).
2 Modern leveling equipment and high-resolution digital cam-
150 eras make documenting such outcrops fast and accurate. All
sections or cores chosen for study should be photographed
(with scales and labels in the photos) and the lithologies and
stratigraphic features described in enough detail to provide
Overwash sand Mud adequate context for subsequent sampling for radiocarbon
dating (Long et al., 1999; Nelson et al., 2008). Wrap cores,
Salt-marsh peat monoliths, and other samples from sections in multiple layers
Mud with Spartina alterniflora of plastic wrap followed by aluminum foil, carefully label
them (e.g., site and sample names, depth intervals, and
Spartina alterniflora macrofossil orientation), transport them to the laboratory, and refrigerate
2 them as quickly as is practical to inhibit bacterial growth and
3 sample degradation.
4 As part of plant macrofossil dating studies, multiple,
5 modern samples of above- and below-ground parts of com-
mon plants collected in the study area provide a reference
6
set for identifying preserved macrofossils (Warner, 1988; Birks,
7
2002). Field visits should coincide with times when seeds
1500 1600 1700 1800 1900 2000 and fruits are attached to plants, with collection focused on
Calibrated age (AD) the parts most likely to be preserved in studied sections.
For roots, rhizomes, and stem bases, their average depth
Rejected 2 calibrated age range
(and depth range) below the modern marsh surface should
Accepted 2 calibrated age range
be documented (van de Plassche et al., 1998, 2006).
Mutual overlap of accepted age ranges
(timing of Spartina alterniflora death and sand deposition) Wash reference specimens thoroughly, dissect some to expose
interior structures and store them in labeled containers
Figure 8 Dating of a depositional event associated with hurricane with ethanol for preservation. The use of glycerin may cause
landfalls in New Jersey, USA. Radiocarbon dated samples were
plant parts to swell beyond their normal size. Publications
placed in stratigraphic context by a series of cores. Identifiable
with photographs, drawings, and descriptions supplement
remains of Spartina alterniflora at the boundary between overwash
sand and the underlying sediment were used to constrain the timing the reference collection (e.g., Martin and Barkley, 1961; Eleu-
of the hurricane. In the lower panel gray horizontal bars show terius, 1976, 2000; Niering et al., 1977). Photographs are
calibrated age ranges that were narrowed (black bars) using other available from an increasing number of online resources such
dating techniques and stratigraphical relationships among samples. as the United States Department of Agriculture PLANTS
Vertical green-shaded error shows mutual overlap of accepted database.
calibrated age ranges and is an estimate of when dated Spartina In the laboratory, unwrap cores and other section samples
alterniflora plants were killed (i.e., when a hurricane deposited the on a clean and clear work space with good lighting. We
overwash sand). Sample numbers are those reported in the original recommend working with one sample at a time. Use field notes
publication. Reproduced from Donnelly, J.P., Roll, S., Wengren, M.,
and photographs to place vertical (and horizontal) scales at the
Butler, J., Lederer, R., Webb, T., 2001a. Sedimentary evidence of
correct depth next to sample material. Take both overview and
intense hurricane strikes from New Jersey. Geology 29, 615–618.
close-up photographs of samples and scales (Figure 10(a)). To
minimize contamination of samples through smearing, trim
post-AD 1950 (bomb-spike) samples, readers should consult the edges of monoliths and blocks before sampling.
specialist literature. Selecting macrofossils for dating from material collected in
In reconstructing coastal history, a specific core or strati- the field depends on the objectives of the study as well as the
graphic section is usually selected from a series of previously characteristics of the sampled sediment. Where large macro-
studied sections as being the most suitable for the goals of the fossils (41–2 cm long) are present and the sediment is peaty,
investigation. Commonly, personnel and funding restrict de- tease the sediment apart along the sample’s vertical axis while
tailed laboratory analyses to a single core or section. Cores are maintaining its correct orientation. Macrofossils can then
collected using a range of equipment including simple push be removed from the matrix by hand, being careful to
corers (gouge, Russian, or piston) or larger, motor-driven note where they were positioned and how they were oriented.
equipment (vibracores of various sizes, geoslicers, and geo- Next, photograph and carefully describe each macrofossil
probes). Vertical sections of small outcrops or hand-dug pits (Figure 10(b)). Larger plant macrofossils are identified with
 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment 383

210
4
0
Elevation (m NVGD)

241
2
3
220
1
1120
High-marsh peat
Low-marsh peat
−1
Tidal mud
Basal unit
1 Distichlis spicata stem in high-marsh peat (maximum age of erosion) Erosive contact
2 3 Spartina alterniflora stem at base of low-marsh peat (minimum age of erosion) 1 Radiocarbon age
4 Spartina alterniflora rhizome (maximum age for tidal marsh recovery) (14C yrs BP)

0 10 20 30 40
Distance (m)

Figure 9 An example of dating hurricane-induced erosion from Connecticut, USA. The erosional contact (red line) is recognized by a regressive
sequence of tidal mud grading into low-marsh and then high-marsh peat abruptly overlying eroded high-marsh peat. The age below the contact
is a maximum age and dated plants that grew in low-marsh mud soon after erosion provide minimum ages. Reproduced from van de Plassche,
O., Erkens, G., van Vliet, F., Brandsma, J., van der Borg, K., de Jong, A.F.M., 2006. Salt-marsh erosion associated with hurricane landfall in
southern New England in the fifteenth and seventeenth centuries. Geology 34, 829–832.

the help of the reference collection (Figures 10(b) and 10(c)). sieved samples should be removed, if there is any chance that
By working systematically through a section of sediment in outer sediment was smeared from above or below during
this fashion, catalog a vertical suite of potential dating sam- collection.
ples. This approach is well suited to thick sections of peaty
sediment where the emphasis is on dating the highest-quality
macrofossils to produce the most accurate age–depth model 14.31.7 Recommendations for Sample Preparation
(van de Plassche et al., 1998).
Commonly, only small (o1 cm long) macrofossils are Primary cleaning of macrofossils isolated from the sediment
available, especially where the sediment is less peaty and more matrix involves removing adhering material that may contain
compact. In this situation, maintaining the orientation and carbon older or younger than the intended sample. Due to
stratigraphic context of macrofossils remains a primary goal, fungal and bacterial growth, even on samples stored in cool,
but magnification is needed to identify potential samples. We dark conditions (Wohlfarth et al., 1998), it is best to prepare
recommend cutting sections of sample sediment less than samples for dating within weeks of collection. Gloves help
10 cm thick into vertical slices 3–8 mm thick with a serrated minimize contamination. Tools (e.g., picks and trays) used in
knife (Figures 10(f) and 10(g)). Two to five slices are laid on a this process should be cleaned thoroughly between samples.
0.5–1.0-mm sieve maintaining their orientation and are pro- Wash larger macrofossil samples gently with water (ideally
gressively washed with water (ideally Millipore) from a Millipore) on a sieve (usually 0.25-mm mesh); chemical
squeeze bottle and teased apart with tweezers or a spatula treatments should be avoided and samples should not be
under a microscope at 5–20  magnification (Figure 10(e)). allowed to dry until cleaning and identification is complete
As the slices break apart, select macrofossils and describe because dry plant parts are often more difficult to identify
their orientation and relation to other stratigraphic features. (Warner, 1988). Soaking samples overnight or using a sonic
Selected fossils are placed in small dishes with water, bath can disaggregate sediment adhering to the sample, but
photographed, and identified using reference materials may damage delicate samples. Small macrofossils are cleaned
(Figure 10(f)). In many cases, when identifiable fossils cannot by scraping them under a microscope with nylon brushes and
be found it is tempting to select an undecayed but unidenti- a metal spatula and transferring them from one dish of water
fiable plant structure. This approach is reasonable when the to another as the water becomes dirty (Figure 10(h)).
fossil is clearly an aboveground plant part, but any structure Photographs and notes of the condition of the sample
that could be a younger root or rhizome should be avoided throughout its preparation provide a valuable record for in-
(van de Plassche, 1982). Picking through sediment slices is terpreting ages.
also effective when seeking samples only from a narrow ver- A secondary cleaning of large and small samples under a
tical range of section, for example, the 1–2-cm-thick interval of binocular microscope is necessary to remove any remaining
sediment above and below a tsunami or hurricane deposit. potential contaminates, in particular the fine rootlets that
Sieving horizontal slices of sediment is a third, usually commonly penetrate plant macrofossils (van de Plassche,
faster, approach to recovering macrofossils. However, in siev- 1982; Figures 10(j) and 10(k)). The rhizomes of some species
ing bulk sediment information about macrofossil orientation (e.g., J. roemerianus) appear more susceptible than others
and relation to stratigraphy may be lost. The outer edges of (e.g., D. spicata) to contamination by younger rootlets.
384 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment

(e)

Cleaning
Triglochin maritimum tools
stem and leaf bases

(a)

Sample
vial

Glass
dishes

(b) Unwashed stems of

Juncusroemerianus (f)

Tidal mud
High-marsh peat
(c)
Washed stems of
Distichlis spicata (g)

(d) Seeds of Scirpus spp. Younger rootlet in

Moss stems and
deciduous leaf fragment
deciduous leaf fragments
(h) (i)

Figure 10 Examples of selecting and preparing macrofossils for radiocarbon dating. (a) Two halves of a vibracore from a tidal marsh in
northern Oregon showing stems and leaves of Triglochin maritimum in growth position. This large macrofossil can be removed from the core
and washed directly on a sieve, scale in centimeters. (b) Unwashed stems of Juncus roemerianus from a core of tidal-marsh sediment in North
Carolina, scale in centimeters. (c) Washed stems of Distichlis spicata from North Carolina, scale in centimeters. (d) Seeds of Scirpus spp. picked
during washing of slices of a high-marsh peat from central Oregon, scale in millimeters. (e) Tools used for secondary cleaning of macrofossil
samples. (f) Binocular microscope arranged so that a sieve containing slices from a 5-cm-thick section of core can be washed under the scope
at 10–50  . Potential macrofossil dating samples found by washing the slices are placed in the glass dishes for later cleaning (every dish has a
cover to keep out dust). Cleaned samples are placed in glass vials and dried at 50 1C. (g) Close-up of vertical slices on sieve in (f). The lower
third of the slices consists of high-marsh peat, which is abruptly overlain by lower intertidal mud. (h) The stems of an aquatic moss and
fragments of a deciduous leaf picked during washing of slices of a detrital peat bed within a sequence of intertidal peat and mud from northern
Oregon, scale in millimeters. (i) A rootlet penetrating a fragment of a deciduous leaf picked during washing of slices of a high-marsh peat from
central Oregon.
 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment 385

Macrofossils are broken open, and tweezers, dissecting nee- tidal-marsh stratigraphy and selection of plant macrofossils
dles, and spatulas are used to remove sediment or unwanted for radiocarbon dating. Nelson’s work on this topic since 1986
material from pieces of macrofossil tissue with periodic rins- has been supported by the Earthquake Hazards Program of the
ing and transferring, if necessary, to clean glass dishes. U.S. Geological Survey. Horton’s work on U.S. tidal marshes
Examining specimens in the reference collection may help has been supported by NSF (EAR-0842728; EAR-0951686),
distinguish contaminants from target macrofossil tissue. Many NOAA (NA05NOS4781182), USGS (02ERAG0044) and
partially decayed plant structures appear similar to invasive NICCR (DE-FC02-06ER64298). We thank considered reviews
roots and rootlets, although the latter are generally lighter in from Mark Roberts (NOSAMS), Brian Sherrod and Debra
color. If in doubt about the origin of rootlet-like material, it is Willard (USGS) and Adam Switzer. This work is a contribution
prudent to remove it. Invasive roots can extend to depths of to IGCP Project 588 and PALSEA.
several meters and may be up to 1000 years younger than
surrounding peaty sediment, especially if the rate of sediment
accumulation is low (van de Plassche, 1982). Tidal-marsh peat References
containing young roots yielded ages 400 years younger than
samples of the same peat with the roots removed (van de Allen, J.R.L., 2000. Morphodynamics of Holocene salt marshes: a review sketch
Plassche, 1982; Törnqvist et al., 1990, 1992). from the Atlantic and Southern North Sea coasts of Europe. Quaternary Science
Reviews 19, 1155–1231.
Careful selection and thorough cleaning of samples in this
Atwater, B.F., 1987. Evidence for great Holocene earthquakes along the outer coast
fashion, followed by the standard acid-base–acid treatment of of Washington state. Science 236, 942–944.
samples by radiocarbon laboratories, minimize the possibility Atwater, B.F., Hemphill-Haley, E., 1997. Recurrence intervals for great earthquakes of
of sample contamination by humic and fulvic acids dissolved the past 3500 years at northeastern Willipa Bay, Washington. U.S. Geological
in groundwater, fine particulate organic material, petroleum- Survey Professional Paper 1576, 108 pp.
Atwater, B.F., Musumi-Rokkaki, S., Satake, K., Tsuji, Y., Ueda, K., Yamaguchi, D.K.,
based lubricants on sampling equipment, oil on human 2005. The Orphan Tsunami of 1700: Japanese Clues to a Parent Earthquake in
skin from handling samples, or by including fragments of North America. University of Washington Press, Seattle, WA, United States
tape and other sampling materials in samples submitted to Geological Survey Professional Paper 1707.
laboratories. Atwater, B.F., Stuiver, M., Yamaguchi, D.K., 1991. Radiocarbon test of earthquake
magnitude at the Cascadia subduction zone. Nature 353, 156–158.
Cleaned plant macrofossil samples are dried at o50 1C to
Atwater, B.F., Tuttle, M.P., Schweig, E.S., Rubin, C.M., Yamaguchi, D.K., Hemphill-
avoid bacterial or fungal growth promoted by damp samples. Haley, E., 2004. Earthquake recurrence inferred from paleoseismology.
Higher temperatures may cause further fractionation of carbon In: Gillespie, A.R., Porter, S.C., Atwater, B.F. (Eds.), The Quaternary Period in the
in the sample. Weigh dry samples to 0.1 mg. Using the as- United States. Elsevier, Amsterdam, pp. 331–350.
sumption that 10–40% of dry plant tissue is carbon, aim for a Atwater, B.F., Yamaguchi, D.K., 1991. Sudden, probably coseismic, submergence
of Holocene trees and grass in coastal Washington State. Geology 19, 706–709.
sample weight of 1–2 mg of carbon. Charcoal is generally Benner, R., Fogel, M.L., Sprague, E.K., 1991. Diagenesis of belowground biomass
about 50% carbon, but may be less. Samples that are too large of Spartina alterniflora in salt-marsh sediments. Limnology and Oceanography
may be as great a problem for AMS laboratories as samples that 36, 1358–1374.
are too small. For large samples, the responsibility of defining Benner, R., Fogel, M.L., Sprague, E.K., Hodson, R.E., 1987. Depletion of 13C in
lignin and its implications for stable carbon isotope studies. Nature 329,
exactly what portion of the sample will be dated is the re-
708–710.
sponsibility of the sample submitter. Clear borosilicate glass Bigler, C., Bugmann, H., 2004. Predicting the time of tree death using
vials with caps are often preferred for sample submission. dendrochronological data. Ecological Applications 14, 902–914.
Vials should be inspected for contamination under the Birks, H.H., 2002. Plant macrofossils. Developments in Paleoenvironmental
microscope and cleaned, if necessary. Autoclaving will kill all Research 3, 49–74.
Blaauw, M., 2010. Methods and code for ‘classical’ age-modelling of radiocarbon
organisms, but will not remove their modern carbon. Plastic sequences. Quaternary Geochronology 5, 512–518.
vials occasionally contain tiny fragments of plastic, whereas Bourgeois, J., Pinegina, T.K., Ponomareva, V., Zaretskaia, N., 2006. Holocene
aluminum foil (commonly used to wrap submitted samples) tsunamis in the southwestern Bering Sea, Russian Far East, and their tectonic
can be contaminated with a carbon-containing residue during implications. Geological Society of America Bulletin 118, 449–463.
Braziunas, T.F., Fung, I.Y., Stuiver, M., 1995. The preindustrial atmospheric 14CO2
production, both of which could add infinitely old carbon to
latitudinal gradient as related to exchanges among atmospheric, oceanic, and
the sample. Keep the cap on cleaned vials to prevent intro- terrestrial reservoirs. Global Biogeochemical Cycles 9, 565–584.
duction of dust. Vials should be clearly labeled (etching labels Bronk Ramsey, C., 2001. Development of the radiocarbon calibration program.
on glass vials eliminates the possibility that ink labels will be Radiocarbon 43, 355–363.
smeared) and placed inside individual, clear, and sealable Bronk Ramsey, C., 2008a. Deposition models for chronological records. Quaternary
Science Reviews 27, 42–60.
plastic bags that are also labeled. Guidelines for sample weight Bronk Ramsey, C., 2008b. Radiocarbon dating: revolutions in understanding.
and preferred sample containers are provided by radiocarbon Archaeometry 50, 249.
laboratories (laboratory web sites are listed by the journal Burr, G.S., 2007. Radiocarbon dating; causes of temporal variations. In: Elias, S.A.
Radiocarbon). (Ed.), Encyclopedia of Quaternary Science. Elsevier, Oxford, pp. 2931–2941.
Burr, G.S., Donahue, D.J., Tang, Y., et al., 2007. Error analysis at the NSF-Arizona
AMS facility. Nuclear Instruments and Methods in Physics Research Section B:
Beam Interactions with Materials and Atoms 259, 149–153.
Acknowledgments Chapman, V.J., 1960. Salt Marshes and Salt Deserts of the World. Interscience
Publishers, New York, NY.
Kemp thanks a graduate internship at the National Ocean Chmura, G.L., Aharon, P., 1995. Stable carbon isotope signatures of sedimentary
carbon in coastal wetlands as indicators of salinity regime. Journal of Coastal
Sciences Accelerators Mass Spectrometry laboratory at Woods Research 11, 124–135.
Hole supervised by Mark Roberts and the guidance of Orson Cisternas, M., Atwater, B.F., Torrejon, F., et al., 2005. Predecessors of the giant
van de Plassche in sharing his careful and detailed approach to 1960 Chile earthquake. Nature 437, 404–407.
386 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment

Clague, J.J., Bobrowsky, P.T., Hutchinson, I., 2000. A review of geological records (Eds.), Quaternary Research Association Techincal Guide 7. Quaternary Research
of large tsunamis at Vancouver Island, British Columbia, and implications for Association, London, pp. 267–286.
hazard. Quaternary Science Reviews 19, 849–863. Marshall, W.A., Gehrels, W.R., Garnett, M.H., Freeman, S.P.H.T., Maden, C., Xu, S.,
Cook, G.T., van der Plicht, J., 2007. Radiocarbon dating: conventional method. 2007. The use of ‘bomb spike’ calibration and high-precision AMS C-14
In: Elias, R.W. (Ed.), Encyclopedia of Quaternary Science. Elsevier, Amsterdam, analyses to date salt-marsh sediments deposited during the past three centuries.
pp. 2899–2911. Quaternary Research 68, 325–337.
Donnelly, J.P., Bryant, S.S., Butler, J., et al., 2001b. 700 yr sedimentary record of Martin, A.C., Barkley, W.D., 1961. Seed Identification Manual. University of
intense hurricane landfalls in southern New England. Geological Society of California, Berkeley, CA.
America Bulletin 113, 714–727. McCormac, F.G., Hogg, A.G., Blackwell, P.G., Buck, C.E., Higham, T.F.G., Reimer,
Donnelly, J.P., Cleary, P., Newby, P., Ettinger, R., 2004. Coupling instrumental and P.J., 2004. SHCal04 Southern Hemisphere Calibration, 0–11.0 cal kyr BP.
geological records of sea-level change: evidence from southern New England of Radiocarbon 46, 1087–1092.
an increase in the rate of sea-level rise in the late 19th century. Geophysical Mook, W.G., van de Plassche, O., 1986. Radiocarbon dating. In: van de Plassche,
Research Letters 31, L05203. O. (Ed.), Sea-Level Research – A Manual for the Collection and Evaluation of
Donnelly, J.P., Roll, S., Wengren, M., Butler, J., Lederer, R., Webb, T., 2001a. Data. Geo Books, Norwich, pp. 525–560.
Sedimentary evidence of intense hurricane strikes from New Jersey. Geology 29, Nanayama, F., Satake, K., Furukawa, R., Shimokawa, K., Atwater, B.F., Shigeno, K.,
615–618. Yamaki, S., 2003. Unusually large earthquakes inferred from tsunami deposits
Eleuterius, L., 1975. The life history of the salt marsh rush, Juncus roemerianus. along the Kuril trench. Nature 424, 660–663.
Bulletin of the Torrey Botanical Club 102, 135–140. Nelson, A.R., 1992. Discordant 14C ages from buried tidal-marsh soils in the
Eleuterius, L., 1976. Vegetative morphology and anatomy of the salt-marsh rush Cascadia subduction zone, southern Oregon coast. Quaternary Research 38,
Juncus roemerianus. Gulf Research Reports 5, 1–10. 74–90.
Eleuterius, L., 2000. Tidal Marsh Plants. Pelican Publishing, Gretna, LA. Nelson, A.R., Atwater, B.F., Bobrowsky, P.T., et al., 1995. Radiocarbon evidence for
Ember, L.M., Williams, D.F., Morris, J.T., 1987. Processes that influence carbon extensive plate-boundary rupture about 300 years ago at the Cascadia
isotope variations in salt marsh sediments. Marine Ecology Progress Series 36, subduction zone. Nature 378, 371–374.
33–42. Nelson, A.R., Kelsey, H.M., Witter, R.C., 2006. Great earthquakes of variable
Fairbanks, R.G., Mortlock, R.A., Chiu, T.-C., et al., 2005. Radiocarbon calibration magnitude at the Cascadia subduction zone. Quaternary Research 65, 354–365.
curve spanning 0 to 50,000 years BP based on paired 230Th/234U/238U and 14C Nelson, A.R., Sawai, Y., Jennings, A.E., et al., 2008. Great-earthquake paleogeodesy
dates on pristine corals. Quaternary Science Reviews 24, 1781–1796. and tsunamis of the past 2000 years at Alsea Bay, central Oregon coast, USA.
Fogel, M.L., Kent Sprague, E., Gize, A.P., Frey, R.W., 1989. Diagenesis of organic Quaternary Science Reviews 27, 747–768.
matter in Georgia salt marshes. Estuarine, Coastal and Shelf Science 28, Nicholls, R.J., Wong, P.P., Burkett, V., et al., 2007. Coastal systems and low-lying
211–230. areas. In: Solomon, S., Qin, D., Manning, M., et al. (Eds.), Climate Change
Gebrehiwet, T., Koretsky, C.M., Krishnamurthy, R.V., 2008. Influence of Spartina and 2007: The Physical Science Basis. Contribution of Working Group I to the
Juncus on saltmarsh sediments III; organic geochemistry. Chemical Geology Fourth Assessment Report of the Intergovernmental Panel on Climate Change.
255, 114–119. Cambridge University Press, Cambridge.
Gehrels, W.R., Belknap, D.F., Kelley, J.T., 1996. Integrated high-precision analyses Niering, W.A., Warren, R.S., Weymouth, C.G., 1977. Our dynamic tidal marshes:
of Holocene relative sea-level changes: lessons from the coast of Maine. vegetation changes as revealed by peat analysis. Connecticut Arboretum Bulletin
Geological Society of America Bulletin 108, 1073–1088. 22, 12.
Gehrels, W.R., Hayward, B., Newnham, R.M., Southall, K.E., 2008. A 20th century Nilsson, M., Klarqvist, M., Bohlin, E., Possnert, G.R., 2001. Variation in 14C age of
acceleration of sea-level rise in New Zealand. Geophysical Research Letters 35, macrofossils and different fractions of minute peat samples dated by AMS.
L02717. Holocene 11, 579–586.
Gehrels, W.R., Kirby, J.R., Prokoph, A., et al., 2005. Onset of recent rapid sea-level Nydick, K.R., Bidwell, A.B., Thomas, E., Varekamp, J.C., 1995. A sea-level rise
rise in the western Atlantic Ocean. Quaternary Science Reviews 24, 2083–2100. curve from Guilford, Connecticut, USA. Marine Geology 124, 137–159.
Grimm, E.C., Maher, Jr. L.J., Nelson, D.M., 2009. The magnitude of error in Oldfield, F., Crooks, P.R.J., Harkness, D.D., Petterson, G., 1997. AMS radiocarbon
conventional bulk-sediment radiocarbon dates from central North America. dating of organic fractions from varved lake sediments: an empirical test of
Quaternary Research 72, 301–308. reliability. Journal of Paleolimnology 18, 87–91.
Hatte, C., Jull, A.J.T., 2007. Radiocarbon dating: plant macrofossils. In: Elias, R.W. Orson, R.A., Warren, R.S., Niering, W.A., 1998. Interpreting sea-level rise and rates
(Ed.), Encyclopedia of Quaternary Science. Elsevier, Amsterdam, pp. 2958–2965. of vertical marsh accretion in a southern New England tidal salt marsh.
Hawkes, A.D., Horton, B.P., Nelson, A.R., Hill, D.F., 2010. The application of Estuarine, Coastal and Shelf Science 47, 419–429.
intertidal foraminifera to reconstruct coastal subsidence during the giant Parnell, A.C., Haslett, J., Allen, J.R.M., Buck, C.E., Huntley, B., 2008. A flexible
Cascadia earthquake of AD 1700 in Oregon, USA. Quaternary International 221, approach to assessing synchroneity of past events using Bayesian
116–140. reconstructions of sedimentation history. Quaternary Science Reviews 27,
Heegaard, E., Birks, H.J.B., Telford, R.J., 2005. Relationships between calibrated 1872–1885.
ages and depth in stratigraphical sequences: an estimation procedure by mixed- Redfield, A.C., 1972. Development of a New England salt marsh. Ecological
effect regression. Holocene 15, 612–618. Monographs 42, 201–237.
Hua, Q., 2009. Radiocarbon: a chronological tool for the recent past. Quaternary Reed, D.J., 1995. The response of coastal marshes to sea-level rise: survival or
Geochronology 4, 378. submergence? Earth Surface Processes and Landforms 20, 39–48.
Hua, Q., Barbetti, M., 2004. Review of tropospheric bomb (super 14) C data for Reimer, P.J., Baillie, M.G.L., Bard, E., et al., 2004. IntCal04 terrestrial radiocarbon
carbon cycle modeling and age calibration purposes. Radiocarbon 46, 1273–1294. age calibration, 0-26 cal kyr BP. Radiocarbon 46, 1029–1058.
Jull, A.J.T., 2007a. Radiocarbon dating: AMS method. In: Elias, S.A. (Ed.), Reimer, P.J., Reimer, R.W., 2007. Radiocarbon dating: calibration. In: Elias, R.W.
Encyclopedia of Quaternary Science. Elsevier, Amsterdam, pp. 2911–2918. (Ed.), Encyclopedia of Quaternary Science. Elsevier, Amsterdam, pp. 2941–2950.
Jull, A.J.T., 2007b. Dating techniques. In: Elias, S.A. (Ed.), Encyclopedia of Sawai, Y., Kamataki, T., Shishikura, M., et al., 2009. A periodic recurrence of
Quaternary Science. Elsevier, Oxford, pp. 453–459. geologically recorded tsunamis during the past 5500 years in eastern Hokkaido,
Kelsey, H.M., Nelson, A.R., Hemphill-Haley, E., Witter, R.C., 2005. Tsunami history of Japan. Journal of Geophysical Research 114, B01319.
an Oregon coastal lake reveals a 4600 yr record of great earthquakes on the Scott, E.M., Cook, G.T., Naysmith, P., Bryant, C.L., O’Donnell, D., 2007. A Report
Cascadia subduction zone. Geological Society of America Bulletin 117, 1009–1032. on Phase 1 of the 5th International Radiocarbon Intercomparison (VIRI).
Kemp, A.C., Horton, B.P., Culver, S.J., et al., 2009. Timing and magnitude of recent Radiocarbon 49, 409–426.
accelerated sea-level rise (North Carolina, United States). Geology 37, 1035–1038. Shennan, I., 1986. Flandrian sea-level changes in the Fenland. II: tendencies of
Lamb, A.L., Wilson, G.P., Leng, M.J., 2006. A review of coastal palaeoclimate and sea-level movement, altitudinal changes, and local and regional factors. Journal
relative sea-level reconstructions using d13C and C/N ratios in organic material. of Quaternary Science 1, 155–179.
Earth-Science Reviews 75, 29–57. Shotyk, W., Goodsite, M.E., Roos-Barraclough, F., et al., 2003. Anthropogenic
Levin, I., Kromer, B., 2004. The tropospheric 14CO2 level in mid-latitudes of the contributions to atmospheric Hg, Pb and As accumulation recorded by peat
northern hemisphere (1959–2003). Radiocarbon 46, 1261–1272. cores from southern Greenland and Denmark dated using the 14C ‘‘bomb pulse
Long, A.J., Innes, J.B., Shennan, I., Tooley, M.J., 1999. Coastal stratigraphy: a case curve’’. Geochimica et Cosmochimica Acta 67, 3991–4011.
study from Johns River, Washington. In: Jones, A.P., Tucker, M.E., Hart, J.K. Stuiver, M., Polach, H.A., 1977. Reporting of 14C data. Radiocarbon 19, 355–363.
 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment 387

Stuiver, M., Reimer, P.J., 1986. A computer program for radiocarbon age calibration. van de Plassche, O., Erkens, G., van Vliet, F., Brandsma, J., van der Borg, K., de Jong,
Radiocarbon 28, 1022–1030. A.F.M., 2006. Salt-marsh erosion associated with hurricane landfall in southern
Switzer, A.D., Bristow, C.S., Jones, B.G., 2006. Investigation of large-scale washover New England in the fifteenth and seventeenth centuries. Geology 34, 829–832.
of a small barrier system on the southeast Australian coast using ground van de Plassche, O., van der Borg, K., de Jong, A.F.M., 1998. Sea level-climate
penetrating radar. Sedimentary Geology 183, 145–156. correlation during the past 1400 yr. Geology 26, 319–322.
Switzer, A.D., Jones, B.G., 2008. Large-scale washover sedimentation in a van de Plassche, O., Wright, A.J., van der Borg, K., de Jong, A.F.M., 2004. On the
freshwater lagoon from the southeast Australian coast: sea-level change, tsunami erosive trail of a 14th and 15th century hurricane in Connecticut (USA) salt
or exceptionally large storm? Holocene 18, 787–803. marshes. Radiocarbon 46, 775–784.
Taylor, R.E., 1987. Radiocarbon Dating: An Archaeological Perspective. Academic van der Plicht, J., Hogg, A., 2006. A note on reporting radiocarbon. Quaternary
Press, Orlando, FL. Geochronology 1, 237–240.
Taylor, R.E., 1996. Radiocarbon dating: the continuing revolution. Evolutionary Varekamp, J., Thomas, E., van de Plassche, O., 1992. Relative sea-level rise and
Anthropology 4, 169–181. climate change over the last 1500 years. Terra Nova 4, 293–304.
Taylor, R.E., Stuiver, M., Reimer, P.J., 1996. Development and extension of the Walker, M.J.C., Bryant, C.L., Coope, G.R., Harkness, D.D., Lowe, J., Scott, E.M.,
calibration of the radiocarbon time scale: archaeological applications. Quaternary
2001. Towards a radiocarbon chronology of the late-glacial: sample selection
Science Reviews 15, 655–668.
strategies. Radiocarbon 42, 1007–1020.
Thomas, E., Varekamp, J., 1991. Paleo-environmental analysis of marsh sequences
Warner, B.G., 1988. Methods in Quaternary Ecology 3: plant macrofossils.
(Clinton, CT); evidence for punctuated sea-level rise during the latest Holocene.
Geoscience Canada 15, 121–129.
Journal of Coastal Research 11, 125–158.
Waterbolk, H.T., 1971. Working with radiocarbon dates. Proceedings of the
Törnqvist, T.E., De Jong, A.F.M., Oosterbaan, W.A., Van der Borg, K., 1992.
Accurate dating of organic deposits by AMS 14C measurement of macrofossils. Prehistorical Society 37, 15–33.
Radiocarbon 34, 566–577. Wohlfarth, B., Skog, G., Possnert, G., Holmquist, B., 1998. Pitfalls in the AMS
Törnqvist, T.E., De Jong, A.F.M., Van der Borg, K., 1990. Comparison of AMS 14C radiocarbon-dating of terrestrial macrofossils. Journal of Quaternary Science 13,
ages of organic deposits and macrofossils: a progress report. Nuclear 137–145.
Instruments and Methods in Physics Research B 52, 442–445. Yeloff, D., Bennett, K.D., Blaauw, M., et al., 2006. High precision 14C dating of
Törnqvist, T.E., Gonzalez, J.L., Newsom, L.A., van der Borg, K., de Jong, A.F.M., Holocene peat deposits: a comparison of Bayesian calibration and wiggle-
Kurnik, C.W., 2004. Deciphering Holocene sea-level history on the US Gulf matching approaches. Quaternary Geochronology 1, 222–235.
Coast: a high-resolution record from the Mississippi Delta. Geological Society
of America Bulletin 116, 1026–1039.
Trumbore, S., 2009. Radiocarbon and soil carbon dynamics. Annual Review of Earth
and Planetary Sciences 37, 47–66.
Relevant Websites
Turney, C.S.M., Coope, G.R., Harkness, D.D., Lowe, J.J., Walker, M.J.C., 2000.
Implications for the dating of Wisconsinan (Weichselian) late-glacial events of http://www.radiocarbon.org
systematic radiocarbon age differences between terrestrial plant macrofossils Radiocarbon: An International Journal of Cosmogenic Isotope Research.
from a site in SW Ireland. Quaternary Research 53, 114–121. http://plants.usda.gov
van de Plassche, O., 1982. Sea-level change and water-level movements in the United States Department of Agriculture (USDA): Natural Resources Conservation
Netherlands during the Holocene. Mededelingen rijks geologische dienst 36, 1. Service.

Biographical Sketch

Andrew Kemp is a postdoctoral associate at the Yale University Climate and Energy Institute and School of
Forestry and Environmental Studies. He received his undergraduate degree from the University of Durham (UK)
and PhD from the University of Pennsylvania. His research has focused on reconstructing sea level using salt
marshes, with a particular interest in the late Holocene period along Atlantic coast of the United States.

Alan R Nelson is a paleoseismologist and United States Geological Survey (USGS) Project Chief whose research
focuses on earthquake and tsunami hazards, especially in the US Pacific Northwest, Alaska, and Chile. He received
his BS and MS degrees from the University of Wisconsin-Madison and a PhD from the University of Colorado-
Boulder. Following early work on glacial chronology using relative soil development and amino acid dating in the
Arctic and Colorado, his research in the 1980s involved reconstructing earthquake histories on faults in Colorado
and Utah, particularly the Wasatch fault. He began investigating prehistoric earthquakes and tsunamis along the
Oregon coast in 1986 and has studied Holocene faults in Puget Sound, Washington, for the past 13 years. He was
a two-term Secretary for the Quaternary Geology and Geomorphology Division of The Geological Society of
America and designed and maintained the first web page devoted to paleoseismology (1996–2000). He has
authored or coauthored about 100 peer-reviewed publications and has been selecting samples for radiocarbon
dating since 1975.
388 Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment

Dr. Ben Horton is an Associate Professor at the University of Pennsylvania. He received his undergraduate degree
from the University of Liverpool and his PhD from the University of Durham (UK). His research concerns sea-
level change. He aims to understand and integrate the internal and external mechanisms that have determined
sea-level changes in the past, and which will shape such changes in the future. Dr. Horton has published over 100
articles in peer-reviewed journals, including Science, Proceedings of the National Academy of Sciences, and Geology. He
is a committee member of the National Research Council, USA and PALSEA (PALeo-constraints on SEA-level rise).
He is project leader of International Geoscience Programme (IGCP) 588 and scientific advisor to The Future Ocean
research of the German Research Foundation. Dr. Horton received a Medal for Research Excellence by the
Commanding General of the North Atlantic Division of the United States Army Corps of Engineers (2010); The
Geological Society of America’s W. Storrs Cole Memorial Research Award (2007); the Linnean Society (UK) Award
(2004); the Higher Education Funding Council for England Excellence in Teaching Award (2003); and the
Menzies Australian Bicentennial Award (2001).