Western Cape Unit

P.O. Box 572 Bellville 7535 SOUTH AFRICA c/o Oos and Reed Streets Bellville Cape Town

Reception: +27 (0) 21 946 6700 Fax: +27 (0) 21 946 4190

A review on Problem Soils in South Africa

S. Diop, F. Stapelberg, K. Tegegn, S. Ngubelanga, L. Heath

Council for Geoscience Report number: 2011-0062

© Copyright 2011. Council for Geoscience

1
 Contents

Contents.................................................................................................................................................. 2

Figures..................................................................................................................................................... 4

Tables ...................................................................................................................................................... 4

1 Introduction .................................................................................................................................... 5

2 General background and overview of problem soils in South Africa.............................................. 5

2.1 Expansive soil......................................................................................................................... 5

2.2 Collapsible soil ....................................................................................................................... 6

2.3 Soft clays................................................................................................................................ 7

2.4 Dispersive soils ...................................................................................................................... 7

3 Mechanism and conditions leading to hazard events .................................................................... 8

3.1 Expansive soil......................................................................................................................... 8

3.2 Collapsible soil .....................................................................................................................10

3.3 Erodible soils........................................................................................................................11

3.4 Soft clays..............................................................................................................................13

3.5 Dispersive soils ....................................................................................................................14

4 Distribution ...................................................................................................................................15

4.1 Expansive soil.......................................................................................................................15

4.2 Collapsible soil .....................................................................................................................17

4.3 Soft clays..............................................................................................................................17

4.4 Dispersive soils ....................................................................................................................18

5 Effects and cost implications ........................................................................................................19

5.1 General problem statement ................................................................................................19

5.2 Problem soil effects .............................................................................................................21

5.2.1 Erodible soil effects .........................................................................................................21

5.2.2 Active soil effects ............................................................................................................23

5.2.3 Collapsible soil effects.....................................................................................................25

5.2.4 Compressible soil effects.................................................................................................25

5.2.5 Acidic soil effects.............................................................................................................26

5.3 Structural manifestation of problem soil damage ..............................................................27

2
 5.3.1 Erodible soils ...................................................................................................................27

5.3.2 Active soils.......................................................................................................................27

5.3.3 Collapsible soils ...............................................................................................................28

5.3.4 Compressible soils...........................................................................................................29

5.3.5 Acidic soils .......................................................................................................................29

5.4 Mitigation options and costs ...............................................................................................30

5.4.1 Erodible soils ...................................................................................................................30

5.4.2 Active soils.......................................................................................................................30

5.4.3 Collapsible soils ...............................................................................................................34

5.4.4 Compressible soils...........................................................................................................35

5.4.5 Acidic soils .......................................................................................................................36

5.5 Remediation costs associated with problem soil damage: .................................................37

5.5.1 Erodible soils ...................................................................................................................37

5.5.2 Active soils.......................................................................................................................37

5.5.3 Collapsible soils ...............................................................................................................38

5.5.4 Compressible soils...........................................................................................................38

5.5.5 Acidic soils .......................................................................................................................38

5.6 International experience and response to problem soils....................................................38

5.6.1 Erodible soils ...................................................................................................................38

5.6.2 Active soils.......................................................................................................................39

5.6.3 Collapsible soils ...............................................................................................................42

5.6.4 Compressible soils...........................................................................................................42

5.6.5 Acidic soils .......................................................................................................................42

5.7 Conclusions & recommendations........................................................................................43

5.7.1 Erodible soils ...................................................................................................................43

5.7.2 Active soils.......................................................................................................................43

5.7.3 Collapsible soils ...............................................................................................................43

5.7.4 Compressible soils...........................................................................................................44

5.7.5 Acidic soils .......................................................................................................................44

6 References ....................................................................................................................................44

3
 Figures
Figure 1: Schematic image of (a) Swelling clay, (b) Non-swelling clay.................................................... 8

Figure 2: Expansive clays (a) Polygonal pattern of shrinkage cracks observed on the surface of bare
soils; (b) saturated wet expansive clay .................................................................................................10

Figure 3: Schematic image of a collapsing sand structure....................................................................11

Figure 4: Examples of erodible soils: (a) Embankment erosion; (b) Donga erosion.............................12

Figure 5: Damage to structures placed on Problem soils, Moretele and Bela Bela, Nothh-West
Province. ...............................................................................................................................................13

Figure 6: Schematic image of (a) Dispersed clay (b) Deflocculated clay...............................................14

Figure 7: Regional distribution map of swell clay occurrence in South Africa (after http://www//
publicworks.gov.za/PDFs~) ...................................................................................................................16

Figure 8: Regional distribution map of potentially collapsing sand in South Africa (after http://www//
publicworks.gov.za/PDFs~) ...................................................................................................................17

Figure 9: Dispersive Clay occurrences (modified from Elges, 1985).....................................................18

Figure 10: Examples of sheetwash and gully erosion ...........................................................................22

Figure 11: Map of soil susceptibility to water erosion in the RSA

..................................................................................................................................23

Figure 12: Typical crack configurations associated with active soils. ...................................................28

Figure 13: Typical effects of collapse phenomena................................................................................29

Figure 14: Swelling clay map of South Africa (from: AGIS Comprehensive Maps: Online) .................. 33

Figure 15: Swelling clay map of the USA (Olive et al, 1989) .................................................................40

Figure 16: Swell-shrink map of the UK..................................................................................................41

Figure 17: Compressible & collapsible soils of the United Kingdom ................................................... 42

Tables
Table 1: Problem soil ranking................................................................................................................21

Table 2: Expansive soil swelling pressures............................................................................................24

Table 3: Corrosion rates for buried metal elements............................................................................26

Table 4: Heave and construction type costs .........................................................................................31

Table 5: Expansive soils: House foundation mitigation measures and state subsidy variation ...........32

Table 6: Collapsible soils: House foundation mitigation measures and state subsidy variation.......... 34

Table 7: Compressible soils: House foundation mitigation measures and state subsidy variation .....36

4
1 Introduction
Problematic soils can be naturally occurring or man-made soils. This includes natural soils that
have been displaced naturally or by man. Problem soils can give rise to many geotechnical
difficulties including inadequate bearing capacity, the potential for unacceptable settlements
and slope instability (Slocombe, 2001). Damage to structures in South Africa is commonly
related to soil characteristics, with expansive and collapsing soils causing the most problems.
There are, however, many types of problem soils, some of the most noteworthy being expansive
soils, collapsible soils, soft clays and dispersive soils.

It is noteworthy that significant developments occurred in the methods for civil engineering
development on problem soils in South Africa from the mid 1960s until the 1980s (Conference
on Problem soils in South Africa, Geotechnical Division of SAICE, July 1985). However, such
developments generally overshadow more recent advances, so that the State of the Art in
dealing with South African problematic soils did not change significantly over the past two
decades (Jacobsz, 2009).

In addition to the well documented historic concerns dealing with specific problem soils, recent
encounters with significant problem situations have highlighted the need for a comprehensive
documentation on the role of remote sensing and GIS technologies for mapping, characterizing
and monitoring problem soils in South Africa.

2 General background and overview of problem soils in South

Africa
2.1 Expansive soil
Expansive soil is a term generally applied to any soil or rock material that has a potential for
shrinking or swelling under changing moisture conditions. In many parts of South Africa,
expansive soils pose a significant hazard to foundations for light buildings. They owe their
characteristics to the presence of swelling clay minerals. As they get wet, the clay minerals
absorb water molecules and expand; conversely, as they dry they shrink, leaving large voids in
the soil.

Expansive soils are also known under the terms swell clay, active clay, shrinkable clay or heaving
clay. In South Africa with its predominantly dry climate, this soil type is more often referred to
as swelling clay, while in countries such as Great Brittan with its wetter climate; it is more often
referred to as shrinkable clay.

Expansive clays are thus those clays for which variations in the moisture content results in a
volumetric change of the soil skeleton. This is problematic since inter alia seasonal moisture
changes in the foundation and sub-foundation horizons of especially lightly loaded fixed
structures gives rise to volumetric changes. Volumetric change in the soil skeleton in turn
induces stresses in the footings and super-structure, leading to super-structure strain and
cracking. Due to the repetitive nature of the stress variation, conventional crack repair measures
generally are unsuccessful.
At the root of the problem of expansive clays lies the fact that the magnitude of the soil
movement is often not recognized timely (structural damage is in fact possible when as little as
2 to 3% of soil volume expansion-contraction occurs). Furthermore, there is often a lack of
knowledge of the benefits to be obtained by applying proper investigation and design
techniques to counteract the potential soil movement.

Extensive studies have been undertaken on the origin and formation mechanism as well as the
soil & geology types of expansive clay. Their identification, effects on structures as well as
countermeasures and additional construction costs to prevent structural damage are now well
understood. It is thus relatively simple to allow for extra design and construction pre-emptive
measures once the potential problem has been identified and the end user convinced of the
cost-savings in adopting a pro-active approach.

The key to a pro-active approach is identification of the possibility of a swell clay condition in an
area targeted for construction. The array of identification tools which can be utilized for
identification include existing geological and topographical maps and remote sensing imagery,
field investigation and identification, and laboratory testing of soil samples.

2.2 Collapsible soil

Collapsible soils are typically poorly graded with respect to particle size with a porous texture
and generally exhibit low in-situ density. Collapsible soil may undergo a sudden large reduction
in volume when a sufficient triggering mechanism or event occurs. They undergo slight
compression due to imposed stresses at low in-situ moisture content, but exhibit a significant
decrease in volume (large settlements) under the same stress when wetting of the soil occurs.
The mechanism and conditions are further explained in section 3.2.

The occurrence of a collapsible soil fabric is limited to very specific geological conditions which
generally lead to the formation of silty sand or sandy silt with a low percentage of clay sized
particles. A collapsible soil fabric generally occurs in wind deposited sands (loess); old, highly
weathered and leached granite soils; or residually weathered so-called “dirty” sandstones, but
in some instances also in soils which have been deposited by sheetwash, gulley wash, wave
action, or termite activity.

Collapse often do not occur below the entire extent of a structure, but only in localities where
collapse conditions are “favorable”, thus leading to a pattern of localized damage to structures.

If a collapsing soil structure is identified prior to construction, preventative construction

measures are fairly straight foreword, centering on proper densification or compaction of the
founding horizons.
2.3 Soft clays
By definition, soft clays are of low shear strength, high compressibility and give rise to time
related settlement problems. Generally, they are sensitive, have undrained shear strengths of
less than 40 kPa, and consolidate for long time spans (years in many instances) after a structure
has been placed on them. They generally have high moisture contents – at or near saturation
point. In addition, these materials often lead to slumping, slope instability and structural
damage or failure when underlying road embankments, dams or other fixed structures.

2.4 Dispersive soils

Dispersive soils are more common in certain type areas and geological settings than others.
Elges (1985) indicated that dispersion can occur in any soil with high exchangeable sodium
percentage (ESP) values and they generally exhibits pale colour. These soils were once believed
to be confined to arid and semi-arid climates, but erosion problems associated with the
presence of dispersive soils have been found in areas with humid climate in recent years.

Where dispersive clays are used in earth dams and embankments, desiccation cracks may be
deep and large, and also settlement cracks might provide water flow paths from where piping
can start, eventually leading to failure. Failure caused by usage of dispersive soil construction
material is normally associated with rapid filling after construction, or rapid filling after
prolonged drought.

In South Africa, dispersive clays are mainly derived from argillaceous sedimentary rocks of Karoo
Supergroup, Cape Supergroup, Uitenhage group and residual soils of all granites (e.g. Swaziland
Basement Complex). However, the majority of these occur in areas of relative water scarcity.
Typically, their presence is indicated by gully erosion in the field and milky runoff water and they
are widely associated with damage to natural slopes and/ or failed embankments such as earth
dams.

Field identification of these soils is generally easy since they are often associated with areas of
erosion. However, an eroded area does not necessarily indicate the location of a dispersive soil
since this condition can also be induced by factors such as loss of vegetative cover, steep slopes
and high intensity precipitation.

Laboratory testing of soils to determine their dispersivity is complex and a number of tests are
needed to be performed to enable their identification. Particular tests may give varying results,
causing misleading and possibly erroneous deductions. The properties which need to be tested
are the pH, electrical conductivity, cation exchange capacity and exchangeable sodium potential
(ESP) of the soil as well as total dissolved salts and sodium absorption ratio (SAR) of the pore
water. Apart form the chemical testing, physical tests performed to determine soil erosive
potential are the crumb test, pinhole test and double hydrometer test.
3 Mechanism and conditions leading to hazard events
3.1 Expansive soil
Clay minerals are complex aluminum silicates (less than 2 microns) composed of two basic units:
Silica tetrahedron and aluminum octahedron (Das, 2002). Clays belong to a family of minerals
called silicates. Because of electron sharing, the silicon tetrahedrons link together with one
another to form thin tetrahedral sheets. The aluminum octahedrons also link together to form
octahedral sheets. The actual clay crystals are a composite of aluminum and silicon sheets which
are held together by intra-molecular forces (Figure 1).

2:1 clay
1:1 clay
Tetraedra
Tetraedra
Octaedra
Octaedra
Tetraedra Tetraedra
Weak cation / water interlayer bond
Octaedra
Tetraedra
Tetraedra
Octaedra
Octaedra
Tetraedra

(a) clay (b) non-swelling clay

Figure 1: Schematic image of (a) Swelling clay, (b) Non-swelling clay

In the weathering process, hydrolysis, or decomposition by the union with water, plays a
dominant role. Hence the physical drainage conditions together with climate and length of time
during which soil-forming processes act are also important factors.

In regions of high temperature and high rain fall, the bases are removed as soluble compounds
and are carried downward or out of the soil, leaving an insoluble weathered residue of silicates
in which kaolinite is the dominant, non-swelling 1:1 lattice type clay mineral.

With decreasing rain fall or impeded drainage, chemical weathering becomes less intense and
soluble bases released by weathering are not leached from the soil. This leads to the formation
of the 2:1 lattice clay minerals between whose successive sheets in the crystal structure varying
amounts oriented water molecules occur. It is the change in the amount of this water which
causes swelling or shrinking of the sheet structure and hence of the soil mass as a whole.
For a group of prominent and highly expansive clay minerals called smectites, one octahedral
sheet is sandwiched between two tetrahedral sheets to create the mineral structure. In
expansive clay, groupings of the constituent clay crystals will attract and hold water molecules
between their crystalline sheets in a sort of “molecular sandwich”.

The electrical structure of water molecules enable them to interact with other charged particles.
The mechanism by which water molecules become attached to the microscopic clay crystals is
called “adsorption”. Because of their shape, composition and resulting electrical charge, the thin
clay crystals or “sheets” have an electro-chemical attraction for the water dipoles. The clay
mineral “montmorillonite”, which is the most notorious in the smectite family, can adsorb very
large amounts of water molecules between its crystalline sheets and therefore has a large
shrink-swell potential (Arora, 2000).

When potentially expansive soil becomes saturated, more and more water dipoles are gathered
between the crystalline clay sheets, causing the bulk volume of the soil to increase or swell. The
incorporation of the water into the chemical structure of the clay will also cause a reduction in
the capacity or strength of the soil.
Figure 2: Expansive clays (a) Polygonal pattern of shrinkage cracks observed on the
surface of bare soils; (b) saturated wet expansive clay
During periods when the moisture in the expansive soil is being removed, either by gravitational
forces or by evaporation, the water between the clay sheets is released, causing the overall
volume of the soil to decrease or shrink. As the moisture is removed from the soil, the shrinking
soil can develop gross features such as voids or desiccation crack. These shrinkage cracks can be
readily observed on the surface of bare soils (Figure 2) and provide an important indication of
expansive soil activity at the property.

3.2 Collapsible soil

Collapsible soils are typically poorly graded with respect to particle size, have a porous texture,
are derived from residual, aeolian or hillwash soil forming processes and comprise of slightly
clayey fine sand and silt mixtures. The soil texture is open textured (voided), and these particles
are interconnected by clay “bridges” – refer to Figure 3 Upon being subjected to stresses larger
than those imposed by the soil density, and receiving a triggering mechanism such as wetting
under load, stress changes or even dynamic loading due to earth tremors, the “bridges” are
unable to withstand the increased stress, and collapse, leading to a sudden decrease in the soil
volume. The collapse is sudden and non-reversible. In severe instances the percentage collapse
may comprise more than twenty percent of the original soil volume.

Sand grains

Clay
“bridges”

Voids

Figure 3: Schematic image of a collapsing sand structure

3.3 Erodible soils

Erodible soils are soils affected by flowing water to physically remove particles from the exposed
surface (Figure 4). Soil erodability is an estimate of the ability of soils to resist erosion, based on
the physical characteristics of each soil. Generally, soils with faster infiltration rates, higher
levels of organic matter and improved soil structure have a greater resistance to erosion. Sand,
sandy loam and loam textured soils tend to be less erodible than silt, very fine sand, and certain
clay textured soils.
Figure 4: Examples of erodible soils: (a) Embankment erosion; (b) Donga erosion

The impact of raindrops on the soil surface can break down soil aggregates and disperse the
aggregate material. Lighter aggregate materials such as very fine sand, silt, clay can be easily
removed by the raindrop splash and runoff water; greater raindrop energy or runoff amounts
might be required to move the larger sand and gravel particles.

Surface runoff, causing gully formation or the enlarging of existing gullies, is usually the result of
improper outlet design for local surface and subsurface drainage systems. Gully formations can
be difficult to control if remedial measures are not designed and properly constructed. Control
measures have to consider the cause of the increased flow of water across the landscape.
3.4 Soft clays
Clayey soil deposits, especially recent deposits, which have not been buried during their history
by overlying soil or water, are referred to as “normally consolidated”. When an additional load
is placed on these soils, they are temporarily “over-pressurized” and need to re-adjust their soil
skeleton in order to reflect their altered state of loading. This is achieved by expulsion of soil
moisture. However, due to the low permeability of clay materials, many years may pass before
the soil moisture can be sufficiently expelled to complete the consolidation process - particularly
so when the clay horizon is thick. Since clay horizons often contain lenses of sand or gravel, the
rate and magnitude of consolidation may additionally vary somewhat, giving rise to differential
consolidation. The magnitude of damage to structures constructed on the clay gradually
worsens as consolidation progresses.

Figure 5: Damage to structures placed on Problem soils, Moretele and Bela Bela,
Nothh-West Province.
Apart from the consolidation process, the clay may also fail upon load application. This is the
case when the shear strength of the clay is too low to support the applied load, with the result
that the bearing capacity of the soil is superseded and sudden collapse or partial collapse of the
structure takes place.

3.5 Dispersive soils

Clays in a dispersed state are clays of which the platelets are largely orientated parallel while for
deflocculated clays the orientation is largely random – refer to Figure 6.

Dispersive erosion depends on variables such as the clay mineralogy and chemistry as well as
the dissolved salts in the soil water and eroding water.

Smectitic clays (e.g. montmorillonite) and illite potentially have high (ESP) values. The repulsive
electrical surface forces between individual clay particles exceed the attractive (Van der Waal’s)
forces, causing the progressive detachment of particles from the surface when in contact with
water.

(a) (b)

Figure 6: Schematic image of (a) Dispersed clay (b) Deflocculated clay

The main clay property causing repulsion is the percentage absorbed sodium cations on clay
surfaces relative to other exchangeable cations (calcium, magnesium and potassium). Due to the
high repulsive forces between particles, once in suspension, they take do not settle out over the
normal time span to be expected for that particular particle size of material and thus lend a high
total dissolved solid content to the erosive water.

The second most important factor governing dispersion is the total content of dissolved salts in
the water - the lower the content the greater the susceptibility of the clay to dispersion.
Since montmorillonitic clays absorb water and thus swell at a slower rate than kaolinitic clay,
they are less capable of ‘plugging holes’ increasing their susceptibility to dispersion.

4 Distribution
4.1 Expansive soil
Expansive clay is widely distributed throughout South Africa, its distribution largely being
dictated by geology, soil type and by local climatic conditions and land form.

The importance of their origin cannot be over-emphasized as it provides an informed observer

with essential information regarding their engineering/ geotechnical characteristics. Generally,
these soils originate from either basic igneous rocks or argillaceous sedimentary rocks.

Basic igneous rock units associated with expansive clays in South Africa include the norite from
the Bushveld Igneous Complex, dolerite of the Karoo Supergroup, diabase and andesite from the
Pretoria Group and also andesite from the Ventersdorp Supergroup.

Furthermore, argillaceous rock units particularly those of Karoo Supergroup, are the most
important source of expansive soils in Southern Africa. The shales and mudrocks of Dwyka, Ecca
and Beaufort Group weathers to heaving clays which are characterized by their slope stability
problems in some parts of South Africa, i.e. KwaZulu-Natal Province. The following is a summary
of known occurrences of expansive soils in South Africa:

- Soils derived from lava occurring in a number of areas of the Limpopo Province [area west
of Lepalale (Ellisras); area between Makhado (Louis Trichardt), Alldays and Musina
(Messina); a strip all along the eastern borders of the Limpopo and Mpumalanga
Provinces; the Mookhophong (Naboomspruit) area and the Bela Bela (Warmbaths)
area], the northern part of the Eastern Cape Province (north of Dortrecht).

- Black “turf” in the Onderstepoort to Rustenburg area and northwards towards Thabazimbi
(residual norite soils)

- Andesite and diabase soils in the Pretoria and Lydenburg areas

- Soils derived from lava occurring in the south eastern parts of the Northwest Province and
in some areas south of Johannesburg

- Soils derived from mudstone/shale covering the western parts of the Northern Cape,
northern (largest) parts of the Free State, eastern parts of Mpumalanga, western
(largest) parts of Kwa-zulu Natal, northern (largest) parts of the Eastern Cape and north-
eastern pars of the Western Cape provinces (so-called Karoo mudrock and tillite)
 - Soil derived from mudstone/shale in the Port Elizabeth and Uitenhage area (Eastern Cape
Province)

- Soil derived from clayey sandstones and shale in the Malmesbury area (Western Cape
Province)

- Soil derived from dolerite rock (intruded). These occur as small irregular bodies all over the
interior of the county in particularly Karoo sedimentary rock.

Figure 7 is also of interest in that it shows a comprehensive regional distribution map of swelling
clays in South Africa.

Figure 7: Regional distribution map of swell clay occurrence in South Africa (after
http://www// publicworks.gov.za/PDFs~)
Figure 8: Regional distribution map of potentially collapsing sand in South Africa (after
http://www// publicworks.gov.za/PDFs~)

4.2 Collapsible soil

Due to the fact that quite a number of deposition mediums can lead to the formation of a
collapsible soil structure, the phenomenon is very widespread, however generally on a localized
scale. Areas of more continuous distribution however include the granite soils in the area
between Pretoria and Johannesburg, in the Mpumalanga Lowveld and Limpopo; windblown
sand covering the northern parts of the Northern Cape, and large parts of Northwest Province,
Limpopo, the Free State, Gauteng, the interiors of Kwa-Zulu Natal and the Eastern Cape. Refer
to Figure 8 for a distribution map.

4.3 Soft clays

In South Africa soft clay occurrences are generally restricted to the eastern and southern coastal
areas and occur particularly in the immediate vicinity of existing or old river channels, but can
also occur locally in the interior in poorly drained areas.

These clays occur primarily as transported soils and are predominantly found in a number of
depositional environments along the coastal areas like Durban, Richards Bay, Knysna and
Langebaan. However, they also occur in some areas further inland as fairly shallow deposits:
 - Durban, Richards Bay, KwaZulu-Natal North and South Coasts
- The Eastern Cape Coast at a number of estuaries – notably in the Knysna area
- The Western Cape south coast at a number of estuaries and at localised estuaries in
the Cape Town area
- Localized pans in the interior

Figure 9: Dispersive Clay occurrences (modified from Elges, 1985)

4.4 Dispersive soils

In South Africa most dispersive clays have in the past largely been encountered in soils derived
from the following geological formations and groups – the distribution of which is indicated in
Figure 9:

Karoo Supergroup: Beaufort Group, Ecca Group, Molteno

Formation and the Dwyka Formation.

Cape Supergroup: Witteberg Group, Bokkeveld Group, Table

Mountain Group and Malmesbury Group

Uitenhage Group: Cretaceous Enon Formation, Kirkwood

Formation and Sundays River Formation.

Swaziland Basement All granites and granodiorites.

Complex:
Dispersive clays can also develop under the following circumstances:

- Low-lying areas where the rainfall is such that seepage water has a high SAR value,
especially in dry regions (so-called ‘Weinert N lines’1 displaying high values). Soils
developed on granite are especially prone to the development of high ESP values in
low-lying areas.

- Areas where the original sediments contain large quantities of illite smectitic clays
(montmorillonite, vermiculite) with high ESP values. This is particularly the case with
the mudstones and siltstones of the Beaufort Group and the Molteno Formation in
regions where the Weinert N-value is higher than 2. Soils in low-lying areas of these
formations are dispersive.

- In the more arid areas, where the Weinert N-value exceeds 10, the development of
dispersive soil is generally inhibited by the presence of free salts despite high SAR
values. Highly dispersive soil can develop if the free salts with high SAR values are
leached out.

5 Effects and cost implications

5.1 General problem statement
The SAICE State of the Art Conference on Problem Soils held in Stellenbosch, South Africa, from
11th to 12th September 1985 considered:

-swelling clays

-collapsible soils

-dispersive soils

-soft clays

-dolomite (incl. wad & ferroan soils)

These same topics were addressed again in 2008, 23 years later, at the Problems Soils
Conference held at the University of Pretoria. However the NHBRC via its Home Builder Manual
site geotechnical classification system (incorporated into GFSH2-2002), only consider three (3)
types of problem soils, namely:

expansive : (H, H1, H2, H3 classes) – 35% of South Africa

1
The Weinert’s N-value is calculated from climatic data as follows:

12 ⋅ E j
N= , where Ej = evaporation during January; Pa = annual precipitation
Pa
 residual mafics/ultra mafics, alluvial clays

collapsible : (C, C1, C2) - % of South Africa

residual granite, colluvium

compressible : (S, S1, S2) – no data

estuarine clays, inland peats, residual & colluvial clays, loose sands

The Council for Geoscience (CGS), by contrast, via its 1:50 000 scale regional geotechnical
mapping program incorporates the identification of 13 geotech parameters including five (5)
problem soil types, namely: Active (Act), Collapsing/settling (Col), Dispersive (Dis), Acidic (Aci)
and Erodible (Ero) soils.

For example the Vereeniging 2627DB sheet (boundary between Gauteng & Free State
provinces), lies within a region broadly defined as being underlain by both moderately expansive
and potentially collapsible soils (Anon, 1991). Mapping showed the following actual spatial
distribution with respect to the total sheet area:

-Active soils (expansive/shrinking) = 87%

-Collapsing or settling soils = 18%

-Erodible soils = 6%

-Acidic soils = 1%

Other geotechnical conditions (slope instability, karst, outcrop, inundation areas etc), occupy
remaining areas. It must be borne in mind that both heave and collapse condition can occur at
one locality (viz: collapsible aeolian soils over active residuum). Thematic maps are thus
prepared for each of the 13 geotechnical parameters, when present and spatial extent over the
approximately 620km2 sheet area, calculated in GIS.

Examination of a second 1:50k sheet example, the White River 2531ac sheet (Mpumalanga
province), which falls in a region dominated by potentially collapsible soils, indicates problem
soil distribution to be:

-Active soils: (expansive/shrinking) = zero %

-Collapsing or settling soils = 53%

-Erodible soils = 64%

-Acidic soils = no data.

As engineering geological maps for: (i) South Africa, (ii) each province, (iii) each district municipal
area or (iv) each local municipal area; have yet to be prepared and only 22 of the two thousand
(2000) 1:50 0000 sheet areas, or conversely only three of the five metro regions, have been
geotechnically mapped; it is difficult to accurately rank the overall effect and cost implications of
problem soils, in South Africa. However, a provisional ranking is shown in Table 1.

Table 1: Problem soil ranking

! !

" "

Subsequent paragraphs follow the above ‘effects’ ranking scheme for the five noted problem
soil types, described in this report. The effects, manifestation, mitigation options & costs,
remediation activities & costs, plus some insights into international experiences; are briefly
examined for each of these problem soils.

5.2 Problem soil effects

5.2.1 Erodible soil effects
It was stressed at the Annual Land Degradation Conference of the International Erosion Control
Association (IECA) South Africa Branch, in Port Elizabeth (Lindique, 2010), that South Africa is
heading for a national disaster, through the loss of productive farmland due to poor farming
practices, low levels of government funding and soil erosion control.

One measure to determine soil erosion is the Universal Soil Loss Equation (USLE), which is
commonly expressed as:

#$ %& & ' '(

#) * * &

# * & * * &(

# +&

# &

# , + &* &

# & & & & %

In areas of moderate to highly dispersive soils, where poor land management and cropping
practices occur, sheet and rill erosion soon develops into badlands type terrain and the
development of dongas (Figure 10).

$( - + 1- 2% % . & 3
&- & & ( & . &
4 - 3 12. 3 55" /
& + % / &
& 0
0

6 + & & + 3
6 + % & 9 :% +3 12. / & 3 55"0
* $ 7 /8 & (3 - & 0

Figure 10: Examples of sheetwash and gully erosion

The AGIS online Comprehensive Maps (Anon, 2011), include a national erodibility map (Figure
11), which highlights the widespread extent of areas susceptible to water erosion.
Figure 11: Map of soil susceptibility to water erosion in the RSA ! " #
"$ % & $ '

The impacts and effects of erodible soils on land degradation in South Africa, are more fully
described by Le Roux (2007), Meadows & Hofmann (2002) and Garland et al (2000). Maharaj
and Gilli (2008) refer to three (3) men as having fallen to their deaths in the 10-30m deep
Dumbuza erosion gulley near Pietermaritzburg, KZN, in 2005.

The clayey fine grained colluvial soils derived from the Karoo Super Group in particular, normally
have a high exchangeable sodium content (>10%), making them dispersive and highly erodible
(Brink, 1985). Deep and pervasive donga development is thus typical across Kwa Zulu Natal and
the former Transkei, along the footslopes of the Drakensberg Escarpment.

5.2.2 Active soil effects

The effect of vertical heave on structures is evaluated via a consideration of differential heave.
Structures are subjected to varying insitu heave conditions as soil moisture content varies
spatially i.e the wet outer edges versus dry inner areas under a building. The deflection
resistance of walls and foundations to these movements is directly dependent upon their design
and any additional reinforcement present.

Swell pressures exerted can be determined directly via a one dimensional oedometer test,
under constant volume conditions (Jennings & Knight, 1957, Burland, 1975); or through
empirical relationships with other easily measured soil parameters (Vijayvergiya & Ghassahy,
1973; Chen , 1988). It is important to note that about six metres of clay soil overburden stress
(kPa), balances out vertical heave pressures, while the depth limit for housing site geotechnical
investigation heave calculations, is invariably set by the excavator (TLB) maximum reach of
3,50m.

For the ubiquitous single storey housing developments now being seen in many parts of South
Africa; the mandatory NHBRC Home Builders Manual site geotechnical classification system
applied, is based on a maximum foundation loading of 50kPa (i.e.: single storey masonry
buildings). Furthermore the class ranges of expected vertical soil movements on expansive soils
(viz: H, H1, H2, H3 as per NHBRC standard) and their corresponding appropriate foundation
designs to mitigate these deflections; correlate to a differential heave movement of just 50% of
total heave.

Clay soils can exhibit swell pressures well in excess of 1000kPa (Table 2), easily lifting and
damaging both single and double storey housing. The value at risk varies from low cost state
subsidized housing that used to be constructed for <R25000 in 1999, but now costs
approximately R60000; while up-market new single and double storey housing ranges from
R500 000 to R2million.

There is thus a clear risk to home owners of significant financial losses should there be: (i)
incorrect identification of this problem soil and (ii) the application of inappropriate construction
methods (ie: incorrect mitigation measures).

Table 2: Expansive soil swelling pressures

& 9$ = >% = & 6 + * ( )
< $ *
+ ;" - ' 55
!+
<

? " ? 5 ?@5 ( + ? 555

@5,A" 5, 5 !5,@5 + "5, 555

5,@5 5, 5 5,!5 % "5, "5

B 5 B 5 B 5 - B"5

*& 3 ACC % !C19 * + # + & & % &%

5.2.3 Collapsible soil effects
The widespread occurrence of potentially collapsible soils in South Africa, estimated at 67% of
the total area of the country (Anon, 1990), will result in an impact on most infrastructural
development.

Some of the earliest evidence of this problem in South Africa, related to the sudden settlement
of a large steel framed building at Witbank in 1955. Subsequently documented case histories of:
(i) a number of buildings (Knight K 1961, in Schwartz 1985) and (ii) a tilting water tower near
Nelspruit (Brink 1979, in Schwartz 1985); highlighted firstly how collapse can occur many years
after construction, secondly how large settlement can occur for even lightly loaded foundations,
and thirdly how very localized the problem can be at a site.

The latter scenario is well illustrated by the case of the three storey Industrial Development
Corporation (IDC) building founded on deeply weathered residual granite, where an individual
column settled up to 300mm (Terblanche, 1989). Inadequate site investigation, poor site
characterization, inappropriate foundation design and inflexible underfloor, wet service
connections, were found to have lead to collapse damage.

5.2.4 Compressible soil effects

All soils will exhibit a degree of consolidation related settlement when and as subjected to loads
exceeding their normal consolidation or pre-consolidation pressures. Resulting re-arrangement
of soil particles is progressive and the expulsion of air or water results in surface settlement
movements, as localized volume reduction occurs. Collapsible soils are a special category, where
this occurs suddenly under specific insitu soil conditions and changing moisture regime.

Loosely consolidated sands for example, will exhibit relatively rapid compressibility concurrent
with loading rates, until inter-granular packing and a reduction in void ratio mobilizes sufficient
resisting strength; while clays can manifest long term settlements over many years as pore
water is slowly expelled and thus loading induced excess pore water stresses, are dissipated.

In South Africa past sea level rise has submerged previously deeply incised river mouths such
that present estuaries and lagoons, are now underlain by inter alia thick, soft saturated clay
deposits of high compressibility and low shear strength. Development of ‘ideal’ flat lands
adjacent Durban bay, the Tugela river estuary and Richards Bay, for harbour works, commodity
handling and/or factories, all encountered extreme settlement and low bearing capacity
problems, on these problem soils.

Mival (in Brink 1985) describes 2 to 3metres of measured consolidation settlement due to
emplacement of 13m of imported fill, over a 4 year period, at the Richards Bay coal terminal, on
nearly 80m of soft clay. In another case, the 7,5m high approach embankments of the
Connaught Road bridge over the Tugela river, settled 300mm (excludes fill), due to underlying
soft estuarine clay horizons totaling 13m thickness (Francis in Brink 1985). Such movements,
apart from the risk of bearing capacity failure, present severe challenges in foundation and
superstructure designs.

Far lower orders of consolidation settlement can be expected to occur on the less extensive and
thinner, inland alluvial clay deposits, flanking major rivers in South Africa. Virtually all fall below
the 1:50year flood-line and are thus excluded from development. Older paleo-deposits at higher
levels, along the Vaal and Orange rivers for example, will either be dry and stiff to very stiff or
firm, at low insitu moisture contents, resulting in low compressibility.

Saturated and partially saturated residual clays developed on Ventersdorp lavas will present
differential consolidation settlement problems due to an irregular weathering profile (Brink,
1979); while residual silty clays derived from Karoo Supergroup mudrocks can also manifest
consolidation settlement. Brink (1983) describes the foundation investigations for the Duhva
Power station, where unforeseen foundation stresses of 2100Kpa, discovered during
construction monitoring, instead of a safe 200kPa design load, would have induced
consolidation settlements of 50-150mm. Additional investigations and design changes thus had
to be carried out during construction. Final investigation costs were calculated to be 0,015% of
the facility construction cost, whereas foundation investigations at the Kriel Power station 50km
away, were only 0,005%.

5.2.5 Acidic soil effects

Three types of acid soils are recognized. (i)Naturally occurring acidic soils in high rainfall areas,
(ii) acid sulphate soils of coastal swamps and (iii)polluted soils that are derived from contact
with sulphide rich mine waste or its acidic leachate, such as from gold mine slimes, in which the
pH can often be less than 3.

The Council for Geoscience uses a cut of <pH 5 to delineate any acid soils during regional 1:50k
geotechnical mapping (Joubert, 2004). These are invariably areas adjacent to and including
existing slimes dumps, along outwash streams, sites of slimes dam failures, or surface residues
left after dump removal for re-processing.

Apart from containing a high dissolved toxic metal load, vegetation growth in acid soils is poor,
exacerbating wind and water erosion; while contact with local streams causes loss of aquatic
life. Polymer geo-textiles (incl. reinforced earth bands, filter membranes, erosion protection
mats) and pipe or cable coatings are more susceptible to acid attack when the pH is <3 (Hoare,
1987)

The Corrosion Control Institute of South Africa estimate the direct cost of corrosion (mitigation
& remediation), to be R150bn per annum (Holman, 2008). This includes corrosion effects
directly attributable to acid soils, but also many other forms of air, soil and water corrosion.
Corrosion within the soil environment is dependent on various factors, including: soil resistivity,
pH, chloride content, sulphate content, sulphide ion content, soil moisture and oxygen content.
The effects on buried metalliferous objects are indicated in Table 3.

Table 3: Corrosion rates for buried metal elements

* * * , -- -
"

& 55 " '( % & ;"( * & 3 55

+ *
* % + @+ %+ %+ & & % & : & D & &*% A
& + * &
-& B 5( & - (

. & E % (3 % + &

Costly economic disruption and environmental damage can occur in respect of high pressure
pipeline wall weakening and perforation. For example water jetting over karst terrain can result
in sinkholes, while oil or gas leaks may ignite resulting in hazardous fires.

5.3 Structural manifestation of problem soil damage

5.3.1 Erodible soils
Deep donga erosion and concurrent sidewall undercutting can initiate small slope failures, which
if left unchecked can threaten adjacent rural housing or even up-market urban complexes,
nearby roads and other infrastructure. There is thus no immediate indication of structural
damage, only a consequent effect, such as donga sidewall collapse undercutting the foundations
of nearby buildings.

5.3.2 Active soils

Typical crack configurations that indicate the presence of heaving soil conditions are illustrated
in Figure 12 and include angled cracking at windows and above doorways. Internal floors may
also be upwardly domed and irregularly cracked. Cracks are wider at the top for ‘doming’
conditions (drought periods), but narrower at the top for structure ‘dishing’ (wet periods). Wide
concrete aprons around houses minimize or abate the development of such potentially
damaging soil moisture regimes.

% + & ) & - + &( - 7 )% & ( -

% & % - +3 % + ( + 7 + % & & * &
& AA & 55C
)% ( % & - + +
7 + - - & . & 7 &% * 7 + % & + ( & = 7 3
* + % 9 & + * &( % & % , 7 &
& 55A

& & * 9 & 3 - + ( & * % 7 + &

( + - & & % & &
55C 55C

Figure 12: Typical crack configurations associated with active soils.

5.3.3 Collapsible soils
The onset of the collapse phenomenon can occur many years after construction, provided there
has been no partial saturation event, a rise in water table, or an increase in foundation loading
in excess of the overburden pressure on the potentially collapsible layer.

For poorly drained housing sites it is usually the first heavy downpour event that triggers
localized collapse of one corner or the end of the building; or it is the addition of a second floor,
that initiates collapse at the first mild wetting event. Some examples of the effects of colla[se
phenomena is dispayed in Figure 13.
 & & * % + + ( ** &
& - 7 + / 0
9 3 F & /
0

Figure 13: Typical effects of collapse phenomena

Increased loading of a road can also cause collapse. This was noted by Knight & Dehlen (1963) in
Schwartz (1985), for a road between Springs and Witbank where introduced unexpected heavy
coal truck traffic, after a period of good service under normal traffic, suddenly resulted in up to
150mm collapse settlement. The road had been unwittingly constructed over collapsible soils.

In the IDC building case (Terblanche, 1989), mentioned above, four phases of costly
(unspecified) remediation entailing: compaction grouting around columns, additional
geotechnical site & laboratory investigation, underpinning of some internal columns, repairs to a
cracked slab, further compaction grouting and finally repairs to internal walls and finishes, were
necessary.

5.3.4 Compressible soils

Inadequately founded structures on soft estuarine clays will manifest severe differential
settlement cracking within months, whereas firm to stiff inland residual clays can result in
damage only after some years. In the latter instance these soils will also exhibit swell or
shrinkage settlement, necessitating additional geotechnical site investigation and soil testing in
order to determine the precise origin of foundation deflections.

5.3.5 Acidic soils

Concrete structures in contact with acid sulphate soils will deteriorate due to the expansive
forces generated by the precipitation of sulphate. Contact with buried metal elements on the
other hand, will lead to various forms and rates of corrosion. For a pipe it is only through poor
performance (low flow rate, line and pressures losses), some possible surface evidence of
leakage (raw water, sewage, gas, oil) and eventual rupture and blockage, that the problem is
manifested. Other buried metal components at risk include fuel tanks and cables, where soil
corrosion (includes the role soil acidity) can result in environmental pollution or critical power or
communication shutdowns.

5.4 Mitigation options and costs

5.4.1 Erodible soils
Improved water runoff conservation and farming practices, such as planned grassland burning,
waterway construction and grassing, correct tilling methods and early remediation of rill and
donga erosion; are baseline mitigation efforts. Government incentives via subsidies and tax
relief are other possible strategies in areas of highly erodible soils.

For low cost state subsidized housing an additional amount of R411,09/site for medium
erodibility risk sites and R1431,48/site in high erodibility risk areas, is payable to developers.

Traditional soil erosion physical prevention measures entail barriers constructed of wood,
dump-rock, dry stone walling, old tyres or cars, small concrete dams, stone filled mattresses and
gabion sediment trapping and flow retention structures. The latter are labour intensive in
construction and current designs offer a cost effective solution. These are accompanied by the
concurrent planting of hardy deep rooted grass species or willow slips. In areas of highly
dispersive soils susceptible to piping erosion, attention must be paid to improve drainage to
prevent pore water build up, grassing and inclusion of filters on the upstream face of flow
retention structures (gabions) (Beckedahl & Gilli, 2008).

At another scale of intervention, modification of the design of earth dams (all sizes), constructed
of highly erodible dispersive clay soils, entails reduction of the permeability of filters to at least
10-6cm/s to limit initiation of soil piping, surface moisture barriers such as grassing and geo-
textiles to prevent the formation of deep desiccation cracking, stabilization of compacted soil
layers by the addition of lime or aluminium sulphate and concurrent correct moisture and
density control. Desiccation cracking should also not be allowed to occur during long
interruptions in embankment construction.

No cost data on these additional precautions was located. However Harmse (1985) cites the
case of the Budalphs Berg dam failure near Senekal in 1969, due to piping erosion in dispersive
clays. Elges (1985) however, mentions the successful construction of several large multi million
rand embankment dams (Mnjoli [40m], Elandsjacht [70m] and Goedertrouw [88m]), on and with
dispersive materials, by the department of Water Affairs.

5.4.2 Active soils

Three considerations determine an appropriate housing construction method on
expansive/shrinking clay sites (Williams et al, 1985).

-nature of the soil profile and predicted heave (fully understanding the geo-hazard)

-the type of foundation loading and stiffness of the intended structure (vulnerability)

-the intended use of the structure (appropriate mitigation level)

Appropriate foundation design solutions corresponding to heave levels, are set out in Table 4
(Williams et al, 1985).

Table 4: Heave and construction type costs

$( * G & & + "
& % & %
& &
* & &

. 5,@ !555 5<

* @, 555 , <

& ,"5 !C5 ", 5<

9 "5, 55 , 5<

H , ? 55 , 5< I

& ** *& , , ;, "<

Of interest is their estimation of additional mitigation costs to counter this slow acting geo-
hazard. For a new conventional brick house of 50m2 with a median construction cost of
R60000-150000 on a stand of 900 - 1200m2; the potential problem soil hazard imposed costs,
can range from R600 to R30000. This assessment excludes additional stand servicing
(connection, pipe bedding) and new township infrastructure road layer construction costs, on
expansive soils. Market prices of R400000 upwards for conventional housing of this nature,
include these developmental expenses.

Williams & Pellissier (1991) examined the South African situation with respect to residential
building programs and concluded that almost R1bn would be spent on additional design and
precautionary measures as a result of inter alia expansive soils. They estimated that half of the
annual construction investment is on residential buildings (R6 430 million in 1990) and 15% of
this would be needed to accommodate all problem soils.

The National Department of Housing, now Human Settlements (DHS) spent a minimum of
R18bn, to deliver 239533 low cost housing units in the 2008/09 financial year, assuming a unit
cost of R77868 (serviced stand included) (Anon, 2011: DHS web site, housing delivery archives).
Utilizing a similar conservative approach as Williams & Pellissier, one sees that at least R2bn,
could have been spent mitigating problem soil hazards, of which expansive soils constitute the
bulk in South Africa.

A calculator (Excel spreadsheet) available from DHS facilitates the determination of variations to
the basic state subsidy of R55706,00 for a 40m2 single storey masonry house (as used in Table 5)
(Anon, 2009). Current experience by DHS is that all problem soils add approximately 20% to the
overall cost of housing development (vdWalt, 2011).

Responsible planning decisions are thus called for from town planners and municipal and
provincial housing authorities, when considering and allowing new green-field developments.
Firstly from an affordability perspective and secondly in respect of the utilization of applicable
government subsidies. This has the potential to be less than optimal if expended on mitigating
possibly avoidable site geotechnical problems (incl. expansive problem soils). The scheme was
obviously implemented to facilitate necessary development where limited or zero alternative
geotechnical options are within economic distance.

The GFSH2-2002 document and more recently the National Housing Code 2009, sets out the
allowable additional state funding for inter alia, expansive soils. The allowed extra amounts as a
percentage of construction cost are shown in Table 5 (Anon, 2011: DHS subsidy calculator).

Table 5: Expansive soils: House foundation mitigation measures and state subsidy
variation
.J:) $ & & % & *

& & &(

/ !5 % &3 % "5 % &

& & 0

J B; "5 , 4

J ; "5, " , * );!"35 & * * *

);!3"5
-soil raft

J ", 5 , + && % 3 ) !@3 & ) 35

* & *& & * * *

-stiffened or cellular raft ) !3@ & ) 3 5

-piled construction

-split construction & soil raft

J ? 5 , (& 3 )C55C3! & ) C 3C! &

* * *
* & *&
)C553C! & ) C 3 C
-stiffened or cellular raft

-piled construction

-soil raft

* * .J:) J :% % 3 3 AAA 9 & 3 & 3$ "

. & 6J % & >% & * >% & &3 % + * % & &

% &

& & & % & &( & (& % & 3 & & 3 &(
The availability of expansive soil distribution maps at the GFSH2 Phase 1 level of housing site
investigation (reconnaissance, land acquisition decisions), is currently limited to indirect
information off the 1:250 000 Land Type maps of South Africa, inferences in respect of
underlying lithologies on available published 1:250 000 to 1:50 000 geological sheets, a limited
number of 1:50 000 published geotechnical sheets, the AGIS swelling clay map of RSA (Paige-
Green P & Turner D, 2008) (Figure 14) and possibly previous or adjacent site scale Phase 2 level
investigation reports held in CGS archives.

Very generalized maps showing the distribution of expansive soils in South Africa were also
published in a guideline document by the Department of Local Government & Housing (Anon,
1990). GIS capture of this map reveals the distribution of expansive soils to be:

-non clay or low expansive areas = 64,4%

-medium expansive areas = 17,7%

-highly expansive areas = 17,7%

This shows as a first estimate, that 35.4% of South Africa, or 677955 km2 may be expected to
present damaging expansive problem soils conditions. The total land area of South Africa is
1214470km2 (Anon, 2011: Fact book).

A more accurate determination using the recently compiled swelling clay map of South Africa
(AGIS online, seeFigure 14), confirms the widespread extent and thus potential impact, of this
problem soils geo-hazard.

Figure 14: Swelling clay map of South Africa (from: AGIS Comprehensive Maps:
Online)
Even canal construction across potentially heaving soils, requires costly mitigation measures.
The 4m deep and 12,70m wide Zuikerbosch canal between Vaal Dam and Rand Waters
Vereeniging purification works, for example, traverses thick highly active residual clays derived
from Ventersdorp lavas and Karoo Supergroup mudrocks (Watermeyer, 1984).

Measured heaves of up to 95mm were recorded. Design measures to prevent damage and
maintenance disruption of this crucial water supply conduit to Johannesburg and elsewhere;
entailed over excavation, surface pre-wetting followed by compaction of a 600-1500mm inert
layer over the 40degree side slopes and narrow floor, before the casting of a continuous steel
reinforced concrete lining of 100mm.

These mitigation measures were also complimented by a buried 3m wide polyethylene layer
either side to minimize sub-soil desiccation. Nearby trees along the canal route were also
removed and the soil pre-wetted to restore the natural insitu moisture regime, prior to canal
construction.

5.4.3 Collapsible soils

Mitigation can vary from efforts to minimize the occurrence of a wetting event caused by
localized surface ponding, via good site drainage, flexible wet service connections and increasing
levels of foundation treatment and design as indicated in below in Table 6.

Table 6: Collapsible soils: House foundation mitigation measures and state subsidy
variation
.J:) G && & & % & *

& && & &( (

/ !5 % &3

% "5 % & & & 0

B" , 4

", 5 , * );!"35 & ) "3@C

, D & & * * *

-* & + );!3"5 & ) 3";

-deep strip footings

-soil raft

? 5 , & -* & + ) "3@C & ) C 3C!

, + && 3 * & * * *
 & *& ) 3"; & ) C 3 C

-stiffened strip or stiffened

cellular raft

-deep strip with

compaction below footings

-piled or piers

-soil raft

* * .J:) J :% % 3 3 AAA 9 & 3 & 3$ @

. & 6J % & >% & * >% & &3 % + * % & &

% &

& & & % & &( & (& % & 3 & & 3 &(

For heavier structures such as cement kilns, milling plants, smelter plants, reservoirs, roadways
and railways, pre-site treatment using (i)impact rollers (IR) over the entire site and (ii) dynamic
compaction (DC)using large drop weight plates (eg: 3m2 , 13t ) on pre-wetted areas. At a site
covered by 2.50m collapsible aeolian soils, trials indicated that a piling option would cost 2.50
times more than insitu compaction (Rees, 1987).

In response to the widespread collapse problem in South Africa, notably the Cape Flats and on
Kalahari aeolian sand deposits, the CSIR redesigned the impact roller such that densification of
soil layers up to 4,0m deep was feasible (Clifford, 1978). The technique is particularly useful in
road and airfield construction, in semi desert areas of low in-situ moisture content, to mitigate
the effects of soil collapse.

Other foundation treatment methods include: vibro-flotation, vibro-replacement (stone

columns), injection of lime, silicate grouts, stabilization of insutu soils with phosphoric acid or
cement or bitumen emulsions, and even deep explosions. A minimum additional cost of 20%
could be safely assumed for this problem soil.

5.4.4 Compressible soils

Drainage of slow consolidating clay soils for airports and factory sites, can be accelerated via
auger emplaced stone columns, or their shear strength increased via electro-osmosis. High costs
are incurred through these necessary foundation treatments, while loads are invariable
transferred via piles (compacted stone columns, driven reinforced concrete, cast insutu
reinforced concrete), through the problem soil horizons.

Appropriate housing foundation designs to mitigate against ranges of soil compressibility, are
less costly and are as stipulated in Table 7.

Consolidation settlements on soft clays or loose to medium dense soils can easily exceed the
worst category above (S2 as applied to light housing structures of 50kPa), by many orders of
magnitude, as foundation loading increases. Mitigation costs will escalate similarly, as tall
buildings, heavy plants, grain silos, tall chimney stacks and elevated water towers for example,
are normally piled on competent horizons (dense sands, gravels, hardpan {calcrete, ferricrete}),
or bedrock.

Table 7: Compressible soils: House foundation mitigation measures and state subsidy
variation
.J:) G && & & % & *

& && & ( / !5 % &3

&(
% "5 % & & & 0

B 5 , 4

5, 5 , * );!"35 & ) "3@C

, D & -* & + & * * *

-deep strip footings );!3"5 & ) 3";

-soil raft

? 5 , & -* & + ) !@3 & ) C 3C!

, + &( * & & & * * *

*& ) !3@ & ) C 3 C

-stiffened strip or stiffened cellular raft

-deep strip with compaction below footings

-piled or piers or soil raft

* * .J:) J :% % 3 3 AAA 9 & 3 & 3$ ;

. & 6J % & >% & * >% & &3 % + * % & &

% &

& & & % & &( & (& % & 3 & & 3 &(

5.4.5 Acidic soils

The prevention or mitigation of corrosion damage to buried metal piping in acid soil
environments, can entail a number of options:

i) pipe laying in concrete lined ducts, to avoid contact with the soil.

ii) protective coatings such as epoxy or bitumen paints, galvanizing or plastic sheathing.

iii) use of sacrificial cathodic protection measures via magnesium anodes.

iv) use of corrosion resistant or stainless steel piping.

Recent research by the US Army Corps of Engineers (Lux, 2009) has examined the use of low
strength, alkaline, cementitous slurry pipeline backfilling (Lux, 2009).
All of the above mitigation measures are most cost effective when done at the design stage.
Post construction remediation is the most disruptive and costly, as a lengthy investigation into
the causes of corrosion linked failure, facility disruption, updated design and pipe relaying costs
for example, are incurred.

5.5 Remediation costs associated with problem soil damage:

5.5.1 Erodible soils
Short term erosion protection measures may entail sidewall stabilization with dump rock, or
repairs to damaged gabion barriers. These activities are normally carried out by farm workers,
using materials at hand, at relatively low overall input cost to landowners.

The topic is dealt with more extensively by pedologists, soil agronomists conservationists and
agricultural engineers. Stabilization of eroded areas has been achieved via a number of methods
such as: hydro-seeding, planting of hardy vetiver grass hedge rows to trap sediment, achieving
changes in grazing loadings, applying longer fallow periods and in reduced de-forestation.

The deep 1.4km long Dumboza Gulley at Edendale immediately west of Pietermaritzburg, KZN,
was repaired over a three year period at a cost of R4,50 million (Maharaj & Gilli, 2008). Apart
from some initial excavator work, subsequent laying of Reno mattresses and gabions was
achieved at a rate of 40-200m2/day by sixty (60) local labourers (mostly women). Side slope
protection entailed vegetated cylinders, turf reinforcement mats (polypropylene, coir) and
hedge rows of deep rooted Vetiver and Eragrostis tef grasses.

5.5.2 Active soils

Responses to heave or shrinkage damage can range from an application of flexible crack filler
and re-painting, to bored or jacked piling and the installation of reinforced concrete ring beams,
to isolate structures from contact with the soil.

Specific data on associated costs is difficult to locate for South Africa, but Williams (1985)
provides the following site specific data:

-Queenswood, Pretoria: R24 000 repairs to a R60 000 house (40% of original outlay)

-Roodepan, Kimberley: R490 000 repairs to a R1,2m housing scheme (41%)

-An unnamed housing scheme in which up to 50% of the structures cracked.

Recurring remediation costs will arise if measures implemented are not accompanied by other
actions such as correct tree placement away from buildings (1,5x height normally) and improved
site drainage to prevent any ponding.
5.5.3 Collapsible soils
Description of a remediation case history by Tanner & Craig (1991), explains how twenty (20)
single storey dwellings, on a site underlain by thick silty to clayey sand colluvium, were repaired
by means of the addition of 1,0m wide concrete aprons to prevent ponding against footings and
6,0 to 8,50m long mini-piles jacked down to a competent pebble marker layer, using the
underside of the strip footings as kentlidge. External and internal wall cracks and finishings were
then repaired. No costs for the work carried out by Esor, are provided unfortunately, but similar
jacking elsewhere, cost in the order of R30-40k for four (4) jacked piles.

5.5.4 Compressible soils

Similar repair measures and costs as described for collapsible soils apply.

Mini or larger pre or insitu cast piles, depending on post construction site access, installed by
screw or jacking techniques; taken down to higher bearing capacity materials, is the normal
solution. Alternatives can include electro osmosis to increase insitu shear strength, insitu
injection of expandable resin cylinders to increase soil modulus, or piled reinforced concrete
ring beams that are then tied into the foundations of the damaged structure. The latter would
also require removal of floors to facilitate the casting of a network of supporting beams.

5.5.5 Acidic soils

Provided the thickness of polluted acid soils is relatively thin, agricultural approaches to
controlling soil acidity via liming could be applied to the affected area. Between 2400-2500kg/ha
would need to be applied and ploughed into the soil. Ongoing applications would be required as
the acidification process in acid soils is ongoing. This just as site flooding which has been shown
to increase soils pH, will be impractical in most civil engineering applications. Removal prior to
construction and replacement with inert material, or the use of costly, high corrosion resistant
materials would usually be considered.

5.6 International experience and response to problem soils

5.6.1 Erodible soils
Numerous studies have examined the negative effects and/or associated costs of soil erosion
within general land degradation and area specific studies. In one example, Berry et al (2003)
concluded that the impact can vary between 3-7% of agricultural GDP (AGDP). Statistics from
Scherr (1999), used by Berry et al (2003) show that South Africa experienced <1,00% of AGDP
land degradation (mostly soil erosion linked); while countries dominated by rural agriculture,
like China, Uganda, Zimbabwe and Ethiopia, were far worse (ie: 2,00-4,00% gross annual
immediate losses). (Berry et al, 2003).

They also noted in the seven countries considered, that remediation and mitigation efforts
focused on physical barriers and were several orders of magnitude less than the losses (ie: China
1-2 US $ billion, Ethiopia 0,20 – 0,50% AGDP). Berry et al (2003) were also adamant that policy
changes could either cause or reduce land degradation and should thus focus more on targeted
mitigation efforts such as terracing, re-afforestation and rural poverty.

5.6.2 Active soils

Jones & Holtz (1973) estimated that 60% of houses built on expansive soils in the United States
were severely damaged necessitating repair costs of US $ 30m/annum, at that time.

According to Krohn & Slosson (1980) some 20% of the USA is mantled by moderate to highly
expansive clay soils. They stated that direct and indirect damages amounted to US $ 7bn
annually, worse than other more noticeable geo-hazards such as floods, earthquakes and
landslides.

An expansive soils map of the conterminous USA, first published in 1979 (Figure 15), that
establishes a synoptic overview of the actual extent of this geo-hazard, supports this position.

Shrink - swell maps are now prepared by the USGS in many states, to assist homeowner, builder,
engineering consultant and mortgage institution awareness of the spatial distribution and
severity of this potentially costly geo-hazard and their likely risk exposure. These maps better
inform legislators, litigators and defendants; as almost 80% of the engineering consultants
canvassed in 1979, were embroiled in law suits. Soil engineers who are always soft targets and
successfully litigated against, were found to be increasingly avoiding residential work (Chen,
1980).
Figure 15: Swelling clay map of the USA (Olive et al, 1989)

The severity of the problem in the USA is illustrated in the example of a damage survey of
houses in Texas in 1974. It found that 98 % of 4870 house foundations had failed as a result of
swelling clays (Anon, 2011). A low level of awareness by both authorities, builders and the public
at the time, is indicated.

The British Geological Survey geo-hazards program, included the preparation of a shrink-swell
map of the United Kingdom (Figure 16).

Shrinkage desiccation by sidewalk tree roots beneath roadways and housing, in London, for
example, lead to tree removal to reduce costly insurance claims against the local municipality.
Figure 16: Swell-shrink map of the UK
In Italy a regional (1:100000) and local scale (1:10000) swelling/ shrinkage hazard assessment
study in the Apennines region, found that even the larger mapping scale could not account for
discrepancies between the incidence of damaged houses and hazard map prediction (Meisina,
2002) .

The most common method of remediation in the Italian case, entailed underpinning, but this
added 20% to the initial building costs. Structural damage was also found to be primarily due to
dessication shrinkage. Susceptibility assessment at the indicated map scales considered
lithology, insitu geometry, structure, mineralogy, type of cover soils and thickness of cover.
5.6.3 Collapsible soils
The collapse phenomenon in Brazil and the United States has been well documented. Hunt
(1984) describes deliberate collapse inducement along the California aqueduct (canal) route, in
the San Joaquin valley of California. Fine grained silty sand to clayey, valley floor deposits of 75m
thickness were saturated over a period of 484 days and exhibited an average collapse
settlement of 4,10m.

5.6.4 Compressible soils

In response to urban development problems with these problem soils, the BGS have as part of
their geo-hazards program, prepared a national map indicating areas susceptible to both
collapse and compressibility (Figure 17).

Figure 17: Compressible & collapsible soils of the United Kingdom.

5.6.5 Acidic soils

Worldwide estimates of all corrosion costs is US $ 1.2 trillion covering prevention (mitigation),
repair and replacement (remediation) (Lieser & Xu, 2010). The American National Bureau of
Standards has also reported that the annual cost of all corrosion is US $70bn per annum. This
includes loss of useful life of equipment, corrosion mitigation efforts, litigation and down time
(Paul, 2003).
The impact of corrosion on the oil and gas transmission pipeline industry for instance, was
estimated at US $5.4bn to US $8.6bn annually (Thompson, 2000). Once again a number of
corrosion scenarios including those initiated or directly attributable to acid soils, are included.

The magnitude of the cost is well illustrated by the example of a rehabilitation project on a
1.60km (1.0mile) section of the Trans Canada 864mm diameter oil pipeline in 1996, which was
found to cost 60% of the replacement option (ie: US $804 000) (Thompson, 2000).

Statistics on the international costs associated with acid soils per se are not easily located or
determined. An indication for the Canadian oil and gas industry, attributes 12% of pipeline
failures to external corrosion sources (Lieser & Xu, 2010). Soil burial corrosion cost estimates,
based on the figures cited by Thompson (2000), may have been in the order of US$0.6bn to
US$1bn then.

5.7 Conclusions & recommendations

5.7.1 Erodible soils
Soil erosion is a worldwide phenomenon, exacerbated by the presence of erodible soils, but
more so by anthropogenic influences, including poor planning, de-forestation, rudimentary
agricultural practices and relatively low abatement efforts in some countries. South Africa is
predisposed to soil erosion, due to poor farming practices in slope, rainfall and vegetation cover
susceptible areas, together with a wide occurrence of erodible soils.

One estimate of the cost of soil erosion per se, is <1% of the agricultural GDP of South Africa
(RSA). The RSA state housing subsidy allows for an extra amount of R411,09/site to
R1431,46/site, for medium and high risk areas respectively.

5.7.2 Active soils

Expansive clay soils constitute a real financial risk and disruption to the quality of life for
societies world-wide. Careful analysis of the problem in the USA resulted in an upward revision
from US $2.5bn (1973) to US $7bn (1979), for all categories of infrastructural development
negatively affected by expansive soils. A rough estimation for South African government funded
low cost housing for 2008/09 puts the figure at about R2bn additional costs for all problem soils.
Further research is necessary to attribute just active soils costs.

Additional input costs per low cost housing unit can vary from about R800 to as much as R14000
per subsidized house.

5.7.3 Collapsible soils

Collapsible problem soils are a worldwide problem, more so in countries with extensive loess
deposits, such as China and parts of Europe. Their effect on infrastructural development in
South Africa, is widespread and negatively impacts on housing, large buildings, roads, power line
pylons, and occasionally dam, canal and reservoir construction sites. Mitigation can entail pre-
wetting, dynamic compaction, chemical grouting, piling or removal and re-compaction for lightly
loaded structures. The total annual cost to the economy is unknown.
Allowable additional housing subsidies in South Africa, to deal with this problem soil condition,
can vary from R800 to R14000 per individual low cost housing site, depending on its severity.

5.7.4 Compressible soils

Soft compressible clays impact on the choice of foundation designs for many coastal cities in the
world. Bangkok, Amsterdam and New Orleans for example, are all located on soft saturated
alluvium. In South Africa such deposits underlie Durban and Richards Bay harbours. Adjacent
industrial areas have to accommodate large orders of settlement ranging from 300mm up to
3000mm.

Lower orders of consolidation settlement also occur on fine colluvium and clayey residual soils
in South Africa. The financial impact for heavy structures can be significant however, if site
founding characterization is incorrect. The annual cost of mitigation in all construction sectors in
the RSA is unknown.

For low cost housing the additional subsidy to mitigate this problem varies from approximately
R800 to R14000 per unit.

5.7.5 Acidic soils

Extremely acidic soils can severely corrode buried metal and some plastic elements. The
problem of corrosion per se severely impacts on and affects the cost of infrastructural
development throughout the world, in the order of US$45billion. However specific figures for
the additional mitigation measures and repair costs, related specifically to acidic soils are
difficult to locate.

No additional housing subsidy is claimable for acidic soil sites in South Africa. An avoidance
policy with respect to prior mine dump areas, is assumed due to the high cost of site
remediation. Other housing sites on naturally acidic soils have a very limited occurrence in South
Africa. Further research on this aspect is required.

6 References
Anon (1994) Reinforced raft foundations for houses where they are needed. Civil Engineering,
vol 2, No 10, pp9-10.

Anon (1999) Home Building Manual. National Home Builders Registration Council, Revision 1,
part 1&2, Feb1999, pp17.

Anon (2003) Corrosion Guidelines V1.0. California Dept. of Transportation, Materials

Engineering & Testing Services (CALTRANS, METS), 51pp.

Anon (2011) Fact Book web site: South Africa. [ https://www.cia.gov/library/publications/the-

world- factbook/geos/sf.html ]
Anon (2011) AGIS Comprehensive Atlas: Online at:
http://www.agis.agric.za/agismap_atlas/AtlasViewer.jsp

Anon (2011) US GS Swelling clays map: Texas

http://www.foundationwatering.com/soilmap.aspx

Arora, K.R. (2000) Soil mechanics & foundation engineering. Standard publishers distributers,
Delhi

Baynes, F.J. (2008) Problem soils in South Africa. Collapsible soils. The civil engineer in South
Africa, November 2008, pp9-21.

Beckedahl H & Gilli A (2008) A classification of problem soils for the rehabilitation of erosion.
Proceedings of the Conference on Problem Soils in South Africa, Midrand, 3-4Nov2008,
pp67-73.

Bell, F.G. (1983) Engineering properties of soils and rock (second edition), Butterworths, pp. 26-
28 & 33-35.

Bell, F.G. (1999) Problem Soils in Geological Hazards. Their assessment, avoidance and
mitigation, Spon Press, pp.181-244.

Bell F.G. & de Bruyn I. (1994) Sensitive, expansive, dispersive and collapsive soils. Bulletin of the
IAEG, No 56, October 1997, pp19-38.

Bojie, F.; Xilin, W.; Gulinck, H. (1995) Soil erosion types in the Loess Hill and Gully area of China.
J. Environ. Sci. Eng., 7, 266-272.

Brink, A.B.A. (1979) Engineering Geology of Southern Africa, Volume 1. Building Publications,
Pretoria. pp. pp. 57-72, 161-177.

Brink, A.B.A. (1983) Engineering Geology of Southern Africa, Volume 3. Building Publications,
Pretoria. pp. 73-115.

Brink, A.B.A. (1985) Engineering Geology of Southern Africa, Volume 4. Building Publications,
Pretoria. pp. 23-26 & 30-33.

Brink, A.B.A.; Partridge, T.C.; and Williams, A.A.B. (1982) Soil survey for engineering. Claredon
Press, Oxford, pp. 108-118, 118-124 & 132-136.

Byrne, G.; Everett, J.P.; and Schwartz, K. (1995) Problem soils and their foundation solutions in A
guide to practical geotechnical engineering in Southern Africa (third edition), Franki, pp.
228-240.

Carter M & Bentley SP (1991) Correlations of soil properties. Pentech press, London, pp107-115.
Chen F H (1980) Legal Aspect of expansive soils. Proceedings of the 14th International
Conference on: Expansive Soils. Colorado, 16-18June1980, pp639-645.

Clifford J (1978) The impact roller – problems solved. Civil Engineer in South Africa, vol 20, No
12, pp321-324.

Das, B.M. (2002) Principles of geotechnical engineering, Thomson Learnings California state
University, Sacramento fifth edition.

Elges, H.F.W.K. (1985) Problem soils in South Africa–State of the art. Dispersive soils. The Civil
Engineer in South Africa, July, 1975 pp. 347-353.

Garland G.G., Hoffman T. and Todd S. (2000). Soil degradation. In A national review of land
degradation in South Africa, unpublished report, eds T. Hoffman, S. Todd, Z. Ntshona, and S.
Turner, pp. 69–107. South African National Biodiversity Institute, Pretoria. Online at:
http://www.nbi.ac.za/landdeg

Hoare DJ (1987) Geotextiles, In Grooun d Engineers reference Book, Ed. by FG Bell,

Butterworths, pp34/3 – 34/18.

Holland JE (1982) Economical housing design on expansive soils. Ground Profile, No 31, July
1982, pp5-28.

Holman J (2008) Conference emphasizes the high costs of corrosion. Engineering news, vol 28,
No 20, 30May-June5 issue, pp66.

Jacobsz, S.W. (2009) Problem soils in South Africa. Preface,. The civil engineer in South Africa.

Joubert R (2004) Explanation of the engineering and geotechnical conditions for the
Vereeniging 2627DB 1:50000 scale map sheet. Council for Geoscience.

Kehew, A.E. (2006) Expansive soil in Geology for engineers and environmental scientists (third
edition), Spon Press, pp. 375-379.

Krohn JP & Slosson JE (1980) Assessment of expansive soils in the United States. Proceedings of
the 14th International Conference on: Expansive Soils. Colorado, 16-18June 1980, pp596-
608.

Le Roux et al (2007) Monitoring soil erosion in South Africa at a regional scale: review and
recommendations. S. Afr. j. sci. vol.103 no.7-8 Pretoria July/Aug. 2007.

Lieser MJ & Xu J (2010) Preventing a legacy of costly corrosion with modern materials.Owens
Corning, Marketing report, Sept 2010.

Lindique L (2010) Soil erosion in South Africa: Annual Land Degradation Conference of the
International Erosion Control Association (IECA) South Africa Branch, in Port Elizabeth.
 [http://www.engineeringnews.co.za/article/national-land-degradation-disaster-looms-for-
sa-2010-10-29]

Lux S (2009) Use of dilute flowable backfill for corrosion mitigation of buried pipe. [Powerpoint
presentation] US Corps of Engineers, US Army Engineer Research and Development Centre,
Construction Engineering Research Laboratory,

Maharaj R & Gilli A (2008) Innovative use of hard and soft solutions to rehabilitate a 1.4km long
gulley.Proceedings of the Conference on Problem Soils in South Africa, Midrand, 3-
4Nov2008, pp95-101.

Meadows ME & Hoffman MT (2002) The nature, extent and causes of land degradation in South
Africa: legacy of the past, lessons for the future? Jnl., Royal Geographical Society, vol 34,
issue 4, pp428-437.

Olive W et al (1989) Swelling clays map of the conterminous United States. Map 1-1940:
Miscellaneous investigation series, USGS.[ http://geology.com/articles/expansive-soil.shtml]

Pack, R.T. (2002) Engineering geologic mapping using 3-D imaging technology, 37th Annual
Symposium on Engineering Geology and Geotechnical Engineering, Boise, Idaho.

Paige-Green, P. (2008) Dispersive and erodible soils – fundamental differences in Problem soils
in South Africa, Proceedings of a Seminar of the South African Institute for Engineering and
Environmental Geologists and the Geotechnical Division of the South African Institution of
Civil engineering. Eskom Conference Centre, Midrand Gauteng, pp. 59 - 65.

Paul S (2003) Minimizing infrastructure corrosion. Proceedings of the International Conference

on Pipeline Engineering and Construction, pp. 494-503.

[http://ascelibrary.org/proceedings/resource/2/ascecp/130/40690/32_1]

Rust, E.; Heyman, G.; and Jones, G.A. (2008) Collapsible Soils: An overview in Problem soils in
South Africa, Proceedings of a Seminar of the South African Institute for Engineering and
Environmental Geologists and the Geotechnical Division of the South African Institution of
Civil engineering. Eskom Conference Centre, Midrand Gauteng, pp. 47-56.

Schwartz, K. (1985) Problem soils in South Africa – sate of the art. Collapsible soils. The Civil
Engineer in South Africa, July, 1975 pp. 379-393.

Slocombe, B.C. (2001) Deep compaction of problematic soils, Problematic Soils, Thomas Telford
(London), pp 163-181.

Tanner A & Craig HJ (1991) A case study of the effects of unsuitable foundations for residential
dwellings constructed on collapsing sands and the remedial measures required. Geotechnics
in the African Environment, Balkema, Rotterdam, p367-373.
Terblanche EH (1989) Settlement and underpinning of a building founded on residual granite: A
case history. The Civil Engineer in South Africa, vol 31, No3, pp77-80, 99.

Von Harmse HJ (1985) Dispersive soils: Proceedings of the Conference on Problem Soils, 11-
12Sept 1985, Stellenbosch, South Africa. 5pp.

Waltham, T. (2002) Subsidence on clays in Foundations of engineering geology (second edition),

Spon Press, pp. 56-57.

Watermeyer CF (1984) Canal construction in expansive slickensided soils. The Civil Engineer in
South Africa, vol 26, No10, pp475-481.

Williams, A.A.B.; Pidgeon, J.T.; and Day, P.W. (1985) Problem soils in South Africa – sate of the
art. Expansive soils. The Civil Engineer in South Africa, July, 1975 pp. 367-407.

Williams, A.A.B. & Pellisier, J.P. (1991 New options for foundations on heaving clay. Geotechnics
in the African environment ,Blight et al (eds) Balkema, Rotterdam,pp243-247

http://www.publicworks.gov.za/PDFs/consultants_docs/GUIDELINE_FOR_PROBLEM_SOILS_IN_
SA_JUNE_2007.doc, visited 2011.01.14