Netherlands Journal of Geosciences — Geologie en Mijnbouw |97 – 4 | 215–227 | 2018 doi:10.1017/njg.2018.

11

Differential subsidence in the urbanised coastal-deltaic plain

of the Netherlands

Kay Koster1,2,∗ , Jan Stafleu1 & Esther Stouthamer2

1 TNO – Geological Survey of the Netherlands, P.O. Box 80015, 3508 TA Utrecht, the Netherlands
2 Department of Physical Geography, Utrecht University, Heidelberglaan 2, Utrecht 3584 CS, the Netherlands
∗ Corresponding author: Email: [email protected]

Manuscript received: 12 January 2018, accepted: 21 August 2018

Abstract

The urbanised peat-rich coastal-deltaic plain of the Netherlands is severely subsiding due to human-induced phreatic groundwater level lowering, as
this causes peat layers to compress and oxidise. To determine the potential susceptibility of this area to future subsidence by peat compression and
oxidation, the effects of lowering present-day phreatic groundwater levels were quantitatively evaluated using a subsidence model. Input were a 3D
geological subsurface voxel-model, modelled phreatic groundwater levels, and functions for peat compression and oxidation. Phreatic groundwater
levels were lowered by 0.25 and 0.5 m, and the resulting peat compression and oxidation over periods of 15 and 30 years were determined. The model
area comprised the major cities Amsterdam and Rotterdam, and their surrounding agricultural lands.
The results revealed that for these scenarios agricultural areas may subside between 0.3 and 0.8 m; potential subsidence in Amsterdam and
Rotterdam is considerably lower, less than 0.4 m. This is due to the presence of several metres thick anthropogenic brought-up soils overlying the
peat below the urban areas, which has already compressed the peat to a depth below groundwater level, and thus minimises further compression and
oxidation. In agricultural areas peat is often situated near the surface, and is therefore highly compressible and prone to oxidation. The averaged
subsidence rates for the scenarios range between 7 and 13 mm a−1 , which is corresponds to present-day rates of subsidence in the peat areas of the
Netherlands. These results contrast with the trend of coastal-deltaic subsidence in other deltas, with cities subsiding faster than agricultural areas.
This difference is explained by the driver of subsidence: in other deltas, subsidence of urban areas is mainly due to deep aquifer extraction, whereas
in the Netherlands subsidence is due to phreatic groundwater level lowering.

Keywords: 3D modelling, compression, Holocene peat, oxidation, water management

Introduction pression and oxidation, as well as past peat mining, caused

c.50% of the surface area of the Holocene coastal-deltaic plain
Many urbanised peat-rich coastal-deltaic plains are subsiding to be presently situated below mean sea level (MSL) (Fig. 1)
due to compression and oxidation of surficial Holocene peat (TNO-GSN, 2016). Large-scale subsidence commenced around AD
layers (Camporese et al., 2006; Gambolati et al., 2006; Serva 1000, when vast wetlands in the western part of the coastal-
& Brunamonte, 2007; Drexler et al., 2009; Deverel et al., 2016). deltaic plain were reclaimed for agriculture (Van Dam, 2001). As
This results in relative sea-level rise, increasing flood risk, salt a consequence of the resulting subsidence, frequent river flood-
water intrusion and damage to infrastructures (e.g. Higgins, ing and sea ingressions affected this area for centuries, caus-
2016; Zoccarato et al., 2018). This form of subsidence is often ing further land loss (Pierik et al., 2017). To date, subsidence
human-induced, and is a consequence of phreatic groundwater by peat compression and oxidation still continues, as phreatic
level lowering and increased surface loading. groundwater levels are progressively lowered to prevent agricul-
The heavily urbanised coastal-deltaic plain of the Nether- tural lands and urbanised areas from inundating (Van den Born
lands is a prime example of an area affected by subsidence as- et al., 2016). Despite the severe volumetric loss, over 15 km3 of
sociated with near-surface peat occurrences. Here, peat com- Holocene peat remains embedded in the subsurface (TNO-GSN,


C Netherlands Journal of Geosciences Foundation 2018 215

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press

 Netherlands Journal of Geosciences — Geologie en Mijnbouw

50000 100000 150000 200000 250000

Elevation of Holocene deposits

600000

>5m

MSL
550000

< -5 m Amsterdam area

500000

A’

North Sea A

The Netherlands
450000

Rotterdam area

B B’
400000

Germany
350000

Belgium

Fig. 1. Surface elevation of the area with Holocene coastal-deltaic surface deposits in the Netherlands. Approximately 50% of this area is situated below
MSL, primarily caused by peat compression, oxidation and past peat mining. The x- and y-axes show coordinates of the national Dutch coordinate system
(Rijksdriehoek) in km.

2016). Consequently, the subsidence potential of the coastal- adapt to subsidence, and eventually to sustain the viability of
deltaic plain of the Netherlands remains high (Hoogland et al., the coastal-deltaic plain of the Netherlands. This requires as-
2012; Van den Born et al., 2016). sessment of the spatial differences in the subsidence potential
Due to differences in the geological build-up of the Holocene of the remaining peat beds in response to near-future phreatic
sequence, past peat mining activities, and drainage history, the groundwater level lowering.
volume of the remaining peat is not equally distributed through- To investigate spatial variations in subsidence potential,
out the subsurface. As a result, local to regional variations ex- detailed subsurface models that describe lateral and vertical
ist in the subsidence potential of the coastal-deltaic plain dur- extensions of peat beds with their physical properties are a
ing future phreatic groundwater level lowering (Nieuwenhuis prerequisite (e.g. Fokker et al., 2015). The Geological Survey
& Schokking, 1997; Van der Meulen et al., 2007; Van Harde- of the Netherlands (TNO-GSN) is developing GeoTOP, a high-
veld et al., 2017; Koster et al., 2018a). Identification of areas resolution (100 × 100 × 0.5 m) 3D geological subsurface voxel-
most vulnerable to subsidence is essential for scientists, stake- model that presently covers most of the Holocene coastal-deltaic
holders and policymakers to design strategies to mitigate and plain of the Netherlands (Stafleu et al., 2011; Van der Meulen

216

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press

 Netherlands Journal of Geosciences — Geologie en Mijnbouw

human-induced
coastal-barrier maturation
land subsidence
clastic sedimentation peat peat
high stand back-barrier peat drainage reclamation
intercalated peat
basal peat
induced sea-ingressions

eustatic sea-level rise sea-level highstand

isostasy and tectonics
10 cal yr BP (ka) 9 8 7 6 5 4 3 2 1 present

Fig. 2. Schematic overview of Holocene processes in the coastal-deltaic plain of the Netherlands: peat formation, sea-level rise, beach-barrier maturation,
human-induced land subsidence, and associated sea ingressions. The striped lines indicate a decrease in the contribution of the processes. Natural build-up
of the area lasted for c.9000 years, and human-induced degradation for c.1000 years, indicating the relatively short period in which humans affected the
coastal-deltaic plain.

et al., 2013). In GeoTOP, the lateral and vertical distribution in sequences of intercalated peat and overbank clastic layers.
of Holocene peat is incorporated. Furthermore, workflows have In the distal and northern parts of the coastal plain, tidal-clay
been developed to populate the voxels that are assigned as intercalated peat formed until about 5500 cal year BP (Hijma
Holocene peat with void ratio, a property that influences the et al., 2009).
compression potential of a peat bed. Eustatic sea-level rise decreased around 6000 cal year BP and
In this study, we aim to quantitatively evaluate spatial varia- a coastal-barrier complex matured. This limited the amounts of
tions in the subsidence potential of the heavily urbanised peat- clastic sediment that reached the lower coastal plain. As tec-
rich coastal deltaic plain of the Netherlands. The subsidence po- tonic and glacio-isostatic subsidence remained significant com-
tential is quantified as the thickness reduction of GeoTOP voxels ponents of relative sea- and groundwater level rise at rates of
assigned as Holocene peat for four scenarios of phreatic ground- decimetres per thousand years, this kept providing accommo-
water level lowering. We focused on two study areas situated dation space and stimulated widespread peat formation in the
in the western part of the coastal-deltaic plain. They comprise back-barrier area for millennia (Vos, 2015). There, extensive and
the major cities of Amsterdam, Rotterdam and their surrounding thick peat beds began to develop.
agricultural land (Fig. 1). By focusing on both urban and agri-
cultural areas, insights were obtained into the effects of these
Amsterdam and Rotterdam
dominant land-use types on subsidence potential.

The cities of Amsterdam and Rotterdam were founded in the

Geological setting and study area back-barrier wetlands around AD 1250. Amsterdam was founded
at the location where the Amstel peat brook debouched into
In the coastal-deltaic plain of the Netherlands, regional-scale a tidal system, and Rotterdam at the confluence of the Rotte
peat formation began as a result of the drowning of this area peat brook and the meandering Meuse river (Vos, 2015). The
due to relative sea-level rise (Fig. 2) (Van de Plassche, 1982). toponyms of both cities reflect the damming of the peat brooks
The peat formed during this early stage occurs at the base of the to prevent the reclaimed and subsided hinterlands from flooding
Holocene coastal sequence, and in this stratigraphical position (Pons & Van Oosten, 1974). This testifies to the centuries-long
it is referred to as basal peat. Basal peat began to form c.9500 struggle of these areas with inundations, as a consequence of
cal year BP in the lowest areas at c.25 m below mean sea level the subsidence of the former back-barrier wetlands.
(MSL), and by c.6000 cal year BP it had proceeded into the inland The pre-Holocene substrate in the Rotterdam area occurs at
coastal plain at c.−3 m MSL, decimetres to a few metres above a greater depth than in Amsterdam, as it comprises a palaeoval-
contemporary sea level. ley formed by Pleistocene braided rivers (Fig. 3). Consequently,
Between 8500 and 6000 cal year BP a marine transgression the accumulated thickness of Holocene sediments and peat in
succeeded basal peat formation (Fig. 2) (Hijma et al., 2009). the Rotterdam area (15–20 m) exceeds that of Amsterdam (c.12–
Clastic tidal depositional environments covered the drowning 18 m). The Holocene sequence in both areas commences with
peat swamps in the west of the coastal plain. In the inland part basal peat, overlain by tidal basin deposits, and a back-barrier
of the coastal-deltaic plain, peat formation in swamps and fens peat layer. The Holocene subsurface of Rotterdam deviates from
continued. Due to repeated avulsion of river and tidal channels, Amsterdam by the presence of Rhine and Meuse fluvial deposits
sediment supply to the floodbasins shifted over time, resulting interfingering with the tidal basin deposits. Clay overlying the

217

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press

 Netherlands Journal of Geosciences — Geologie en Mijnbouw

A A’
Amsterdam
0
Back-barrier peat
-7.5
Depth (m)

-15
Basal peat

-22.5 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
B B’
Rotterdam
0

Back-barrier peat
-7.5
Depth (m)

Basal peat
-15 Fig. 3. Lithological cross-sections of the
Holocene subsurface of (A) Amsterdam, (B)
-22.5
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Rotterdam, and their surrounding agricultural
Distance (km)
Legend lands, extracted from GeoTOP (TNO-GSN,
Anthropogenic brought-up soil
2016). For locations see Figure 1. The black
Peat
Clay dotted lines indicate the boundary between
Sandy clay the Pleistocene substrate and the Holocene
Sand 63 - 150 micrometer
( - 300 micrometer
Sand 150
coastal-deltaic deposits. Basal peat directly
Sand 300 - 2000 micrometer overlies Pleistocene deposits, and back-barrier
Gravel
peat is situated near the present surface.

Table 1. Overview of the four phreatic groundwater level lowering scenarios.

Phreatic groundwater
Scenario Duration (years) level lowering (m)
1 15 0.25
2 30 0.25
3 15 0.25
4 30 0.25

back-barrier peat resulting from the subsidence-induced inun-

dations derives in the Amsterdam area primarily from sea in-
gressions and the Amstel brook, and in the Rotterdam area from
sea ingressions and river floods.
To enable expansion of the urban areas of both cities, an-
Fig. 4. An exposure of the shallow subsurface of Amsterdam near the Amstel
thropogenic soil was brought up on the existing land surface to
brook: back-barrier peat is overlain by floodbasin clay from the Amstel peat-
increase the bearing capacity of the underlying peat (Kluiving
brook. The floodbasin clay was deposited after the back-barrier peat subsided
et al., 2016; Koster, 2016) (Fig. 4). The burden of this material
by phreatic groundwater level lowering. Anthropogenic soil was subsequently
caused substantial compression of underlying peat at that time.
brought up to increase the bearing capacity of the area. The present-day
phreatic groundwater level equals the bottom of the excavation pit (photo:
Methods Kay Koster).

The amount of subsidence by peat compression and oxidation

was determined by quantifying the thickness reduction of vox- of recent (post-World War II) periodic groundwater level lower-
els in GeoTOP that are classified as Holocene peat (Fig. 5). The ing recorded in polder systems in the Netherlands (Schothorst,
thickness reduction was determined as the amount of com- 1977; TNO-GSN, 2016), whereas a period of 30 years is regarded
pression and oxidation for four scenarios (Table 1). The upper as the maximum time necessary to complete primary com-
limit of 0.5 m was selected as it represents a maximum value pression (CUR, 1992). Thus, the selected scenarios represent a

218

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press

 Netherlands Journal of Geosciences — Geologie en Mijnbouw

0.5 m σ' e 0.00

0.00
(kPa) (-) U(t) = 1
Oxidation/ End of primary compression after 0.2 yr,
compression 9.5 10.6 further subsidence by creep and oxidation
-0.05
Compression 10.6 10.2
-0.05

cα =
-0.10 0.00
70
Vox = 0

(m)
Undefined sediment .0048

Subsidence (m)
-0.10
σ =18 kPa

subsidence
-0.15 Vox
=0
7m

.02
Peat 38
σ =12 kPa -0.15 cα =
0.0
-0.20 341

Initial phreatic GW level

-0.20
New phreatic GW level -0.25

-0.30
-0.25
00 55 10
10 15
15 20
20 25
25 30
30
Time (yr)
52.7 6.4
me (yr)
Compression

Fig. 5. Schematic representation of the maximum subsidence of a hypothetical vertical voxel-stack after phreatic groundwater level lowering (scenario 1;
phreatic groundwater level lowering of 0.5 m during 30 years). The peat-voxels are attributed with vertical effective stress, determining vertical variation in
void ratio and compression behaviour. The top peat-voxel is situated above the lowered phreatic groundwater level and is therefore subjected to oxidation.
The black line in the right-hand panel shows: (1) subsidence over time, and (2) the outcome of the analysis with minimum and maximum values of Vox
(red lines) and Cα (green lines).

plausible maximum, and three intermediate subsidence poten- by the vertical permeability (kv ) of peat, which is derived from
tials for near-future phreatic groundwater level lowering. its void ratio (e) (cf. Mesri & Ajlouni, 2007). For Holocene peat
This study focused on 3D subsections of GeoTOP compris- embedded in the subsurface of the Netherlands, vertical perme-
ing the Holocene subsoil of the cities of Amsterdam, Rotter- ability was empirically derived from a void ratio – permeability
dam and their surrounding agricultural areas. A python-scripted dataset previously acquired from a database of TNO-GSN (Koster,
workflow was established to (1) select each voxel assigned as 2017) (eqn 5).
Holocene peat within the subsections, (2) populate them with Primary and secondary compression occur simultaneously,
voxel-specific vertical effective stress and void ratio, which are but, after primary compression has ended, only creep processes
parameters necessary to calculate compression, (3) implement of secondary compression will cause further land subsidence
elevations of present and lowered phreatic groundwater levels (Zhang et al., 2018). The rate of creep is determined by the
within the selected subsections, (4) calculate the thickness re- coefficient of secondary compression (Cα ). In this study, a stan-
duction of each selected voxel after the phreatic groundwater dard coefficient of secondary compression value was used for
lowering, by using widely applied compression and oxidation Holocene peat in the Netherlands (Blok, 2014), as no empiri-
functions, and (5) sum the total thickness reduction in the voxel cal relations between peat field conditions and creep intensities
stack per x, y coordinate to produce 2D subsidence maps. have been established yet (CUR, 1992).
   
hp Cc σ v + σ v
εcom = = · U(t ) · log
Peat compression hp (1 + e0 ) σ v
 
t0 + t
+ Cα log (1)
Peat compression was calculated using the internationally t0
widely applied ‘Bjerrum function’ for soft soil compression (CUR,
1992). The function (eqn 1) calculates a compression εcom 
from variables that describe primary and secondary compression T3
U(t ) = 6
(2)
(CUR, 1992). Primary compression is the thickness reduction of (T 3 + 0.5)
peat due to increasing vertical effective stress (σ v’ ), whereas
secondary compression comprises time-dependent creep. cv · t
T=  2 (3)
The rate at which pore water expulses from peat during verti- β · hp
cal effective stress increase determines the duration of primary
compression. This duration is indicated by the degree of consol- kv
cv = (4)
idation (U(t) ) (eqn 2). The degree of consolidation is determined mv · γw

219

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press

 Netherlands Journal of Geosciences — Geologie en Mijnbouw

Holocene deposits per m sediment thickness in the Netherlands:

peat, 11 kPa; clay, 14 kPa; sandy clay, 16 kPa; sand, 20 kPa; and
kv = 1 · 10−11 · e0 2.98 (5)
brought-up soil, 18 kPa (cf. Kruiver et al., 2017). The hydro-
static pressure was determined per voxel by subtracting the z-
e coordinate of the centre of the voxel from that of the phreatic
mv = (6)
(1 + e0 ) · σ v groundwater level (eqn 9). The phreatic groundwater levels were
εcom = compression (–) obtained from an online portal for hydrological data of the
hp = peat thickness (m) Netherlands (NHI, 2016).
hp = decrease in peat thickness (m) Changes in total vertical stress and vertical effective stress af-
Cc = compression index; 5.8 (–) (Koster et al., 2018b) ter phreatic groundwater level lowering depend on the phreatic
U(t) = degree of consolidation (–) storage of surficial deposits (e.g. Zanello et al., 2011). Within
σ ʹv = increase in vertical effective stress (kPa) a peat layer, the phreatic storage coefficient is very variable
Cα = coefficient of secondary compression; 0.0153 (–) in both lateral and vertical directions, as it depends on the
t0 = reference time (s) degree of decomposition and void ratio. It can range between
t = increase in time (s) 0.8 and 0.5 for freshly formed peat, to values between >0
T = time factor (–) and 0.1 for buried peat layers (Bot, 2016). Here, we used a
cv = coefficient of consolidation (m2 s−1 ) widely applied averaged phreatic storage coefficient for near-
kv = vertical permeability (m s−1 ) (Koster, 2017) surface peat layers in the Netherlands of 0.1, to determine the
β = drainage constant; 1 (–) changes in total vertical stress and vertical effective stress af-
mv = coefficient of volume compressibility (kPa−1 ) ter phreatic groundwater level lowering (eqns 11 and 12) (Bot,
γ w = unit weight of water; 9.81 kN m−3 2016).

Prime input variable for this function is vertical effective stress σv = hph gw · S · γw (10)
and increases herein. Changes in vertical effective stress change
σ v = (−S + 1) hph gw · γw (11)
the void ratio of peat, which is the ratio between volumes of
the non-solid and solid components. A reduction in void ratio of
σ v = changes in total vertical stress
subsurface peat layers is reflected at the surface as subsidence.
S = phreatic storage coefficient; 0.1 (–)
For peat embedded in the Holocene sequence of the Netherlands,
an empirical relation between vertical effective stress and void with S = ϕ − ϕw , where ϕ is the saturated porous medium
ratio was used, based on a vertical effective stress – void ra- porosity and ϕ w is the moisture content (in % of the total porous
tio dataset acquired from a database of TNO-GSN (eqn 7; Koster medium volume) above the phreatic surface.
et al., 2018b).
−0.413 Peat oxidation
e = 25.1 σ v (7)

σ v = σv − μ (8) The thickness reduction of the peat voxels by oxidation was only
 
μ = hph gw − hvoxel · γw (9) calculated for those parts of the voxels that were situated above
the lowered phreatic groundwater level. A generally tested and
e = void ratio (–) applied function was used to quantify peat thickness reduction
σ ʹv = vertical effective stress (kPa) due to oxidation εox over time (eqn 12) (Van der Meulen et al.,
σ v = total stress (kPa) 2007; Van Hardeveld et al., 2017). The rate of oxidation (Vox ) is
μ = hydrostatic pressure (kPa) a value relative to the thickness of the layer above the ground-
hph gw = z-coordinate of phreatic groundwater level (m) water level, and was set at 0.015 m m−1 a−1 , which is a typical
hvoxel = z-coordinate of the centre of a peat voxel (m) value for Holocene peat in the Netherlands (cf. Van der Meulen
Eqns 7–9 were used to populate the GeoTOP voxels contain- et al., 2007).
ing peat with estimated current void ratio, depending on ef-
hp dry · t )
fective stress experienced by the overburden. The void ratio εox = = 1 − exp(−Vox (12)
hp dry
of the selected voxels was calculated using eqn 7. Vertical ef-
fective stress was quantified following Terzhagi’s principles of εox = relative height reduction due to oxidation (–)
soil mechanics as the difference between total vertical stress hp dry = peat organic matter thickness above phreatic water
(σ v ) and hydrostatic pressure (μ) (eqn 8). The total vertical level (m)
stress per selected voxel was determined by summing lithology- hp dry = decrease in peat organic matter thickness above
specific total stress values of the overlying voxels. These total phreatic water level (m)
stress values comprised previously determined values typical for Vox = rate of oxidation; 0.015 (m m−1 a−1 )

220

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press

 Netherlands Journal of Geosciences — Geologie en Mijnbouw

Table 2. Summarised results of the four phreatic groundwater-level lowering and 0.4 m (Fig. 6C). The peat layers in these areas have relatively
scenarios. high void ratios due to their near-surface position, and hence
high compressibility. The difference in subsidence by compres-
Compression (m) Oxidation (m) sion for the four scenarios is only a few centimetres (Table 2),
despite differences in vertical effective stress increases and du-
Scenario Area mean st. dev. mean st. dev.
ration of creep. In the former peat reclamation areas, subsidence
1 Amsterdam 0.09 0.07 0.10 0.09
by compression is limited to less than 0.05 m. These low values
2 Amsterdam 0.11 0.07 0.18 0.16
are due to the absence of compressible near-surface back-barrier
3 Amsterdam 0.10 0.07 0.15 0.09
peat, and subsidence is here primarily caused by compression of
4 Amsterdam 0.11 0.08 0.26 0.16
tidal basin- and basal peat layers at several metres depth in the
1 Rotterdam 0.15 0.09 0.08 0.09
subsurface. Because of their deeper position, these peat layers
2 Rotterdam 0.17 0.10 0.14 0.16
have a lower void ratio and are therefore less compressible than
3 Rotterdam 0.16 0.10 0.12 0.09
the surficial peat. In spite of the large thickness of the peat
4 Rotterdam 0.18 0.10 0.21 0.16
beds underlying Amsterdam, subsidence by compression is less
than 0.3 m, except for an area in the eastern part of the city.
In the eastern part compression is up to 0.4 m, similar to the
Subsidence was calculated for time steps of 0.01 year, during agricultural areas, because in that area overburden is relatively
which the thickness reduction of each selected peat voxel was thin (0.5–1 m).
calculated (Fig. 5). For oxidation, the thickness reduction dur- The results of subsidence by oxidation show that this pro-
ing each time step was used to adapt hp dry in eqn 11 to calcu- cess is limited to areas where peat is situated near the sur-
late oxidation in the next time step. Thickness reduction due to face (Fig. 6B and D). These areas primarily comprise agricultural
compression was determined for an instantaneous lowering of lands and patches in the south and east of Amsterdam. The mean
the phreatic groundwater level. Per time step, the decrease in subsidence by oxidation deviates more than 0.1 m between the
the peat thickness situated above the new phreatic groundwa- scenarios (Table 2). This indicates that oxidation contributes
ter level by compression of deeper-situated selected voxels was more to subsidence than creep (both are time-dependent).
determined, and subtracted from the amount of organic matter The total subsidence for this area reveals that the agricultural
subjected to oxidation in the consecutive time step. regions north and southeast of Amsterdam have the highest sub-
sidence potential when phreatic groundwater levels are lowered,
Results with subsidence amounts between 0.5 and 0.8 m (Fig. 6E). Nev-
ertheless, the highest subsidence potential occurs in a narrow
Future subsidence estimates local peat rim in the southern part of Amsterdam, which primar-
ily subsides due to oxidation. The difference between subsidence
Figures 6 and 7 show the spatial patterns in subsurface lithol- in the urban and agricultural areas is caused by the presence
ogy, peat thickness and subsidence resulting from peat com- of several metres of anthropogenic brought-up soil overlying
pression and oxidation for the four scenarios in the Amsterdam the back-barrier peat in Amsterdam. This has already reduced
and Rotterdam areas. Compression, oxidation and cumulative the void ratio and hence compressibility of the peat underlying
subsidence are given separately. The modelling results are sum- the city in the past centuries. Furthermore, the brought-up soil
marised in Table 2. The general trend reveals that agricultural ar- pushed the peat below phreatic groundwater levels, minimising
eas are more prone to subsidence by phreatic groundwater level oxidisation.
lowering than urban areas.
Rotterdam In the Rotterdam area, the total thickness of
Amsterdam The Amsterdam area comprises thick (3–6 m) and Holocene peat ranges between 1 and 6 m (Fig. 7A). In the agri-
thin (>1 m) peat beds (Fig. 6A). The areas with thickest peat cultural area east of Rotterdam, peat is the thickest, with val-
reflect relatively well-preserved surficial back-barrier peat lay- ues up to 6 m. Here, peat is back-barrier peat, and is largely
ers, whereas thin peat beds occur in former mining areas. Below directly situated at the surface. In the agricultural area north
the urbanised zone of Amsterdam and in the agricultural area of Rotterdam, this back-barrier peat is covered by flood deposits
north of the city, the back-barrier peat is the thickest and most (Fig. 7B). The summarised results in Table 2 show that in Rot-
abundant. In the agricultural areas, north and south of Ams- terdam subsidence by compression is higher, and subsidence by
terdam, the back-barrier peat is directly situated at the surface oxidation is lower, than in the Amsterdam area.
(Fig. 6B). The agricultural area east of Rotterdam is most affected by
Subsidence by compression in the four scenarios is most compression. There, subsidence ranges between 0.2 and 0.4 m
severe in the agricultural areas directly north-northeast and (Fig. 7C). Compression also affects peat underlying Rotterdam.
southeast of Amsterdam, with subsidence ranging between 0.2 However, because of anthropogenic overburden the void ratio

221

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press

 Netherlands Journal of Geosciences — Geologie en Mijnbouw

Fig. 6. Subsurface lithology and subsidence by peat compression and oxidation for four different scenarios of phreatic groundwater level lowering (GWL
lowering) and duration of subsidence in the Amsterdam area. The scenarios are: (1) 0.25 m GWL lowering during 15 years, (2) 0.25 m GWL lowering during
30 years, (3) 0.50 m GWL lowering during 15 years, and (4) 0.50 m GWL lowering during 30 years. The hatched areas indicate urban zones. For location
within the Netherlands, see Figure 1.

222

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press

 Netherlands Journal of Geosciences — Geologie en Mijnbouw

Fig. 7. Subsurface lithology and subsidence for the four different subsidence scenarios in the Rotterdam area. The hatched areas indicate urban zones. For
location within the Netherlands, see Figure 1.

223

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press

 Netherlands Journal of Geosciences — Geologie en Mijnbouw

of peat is relatively low, and therefore compression is moderate completely oxidises and its thickness approaches zero. This will,
(<0.1 to 0.4 m). The compression of the peat underlying Gouda however, occur beyond 30 years. Subsidence by a minimum oxi-
is more severe, with values between 0.3 and 0.4 m. dation rate after 30 years is c.13 cm, and for maximum oxidation
Because of the surficial position of peat east of Rotterdam, it c.23 cm, or −35% and +15% relative to the subsidence scenario
is most affected by oxidation. Similar to Amsterdam, the peat used in this study. The lines of the different Cα values diverge di-
underlying the urbanised area of Rotterdam is less subjected to rectly after onset of consolidation. Contrary to the different Vox
oxidation. Oxidation only occurs along the eastern and northern lines, these lines will continue to diverge over time, as creep con-
boundaries of the city (Fig. 7D). This is also due to the protec- tinues infinitely. After 30 years, subsidence with minimum Cα is
tion of peat by anthropogenic brought-up soil. This protection c.16 cm, and with a Cα maximum c.24 cm, or −20% and +20%
does not account for the smaller city of Gouda. There, approxi- relative to subsidence calculated with the parameters used in
mately half of the surface area of the city is subjected to oxida- this study. Within the time period of 30 years, both parameters
tion. are important for subsidence after primary consolidation ends.
The total subsidence (Fig. 7E) shows large differences be- However, for longer subsidence periods, the Cα parameter is of
tween the agricultural area east of Rotterdam, the city of Rot- most influence, as peat oxidation eventually ends.
terdam, and the agricultural area north of the city. In the area
east of Rotterdam, including the city of Gouda, total subsidence
is between 0.4 and 0.8 m. In the agricultural area north of Rot-
terdam, subsidence is low, with values <0.4 m. In Rotterdam, Discussion
total subsidence ranges between 0.1 and 0.4 m.
Subsidence patterns
Model sensitivity analysis
In most low-lying coastal-deltaic plains that are subjected to
The rate of oxidation (Vox ) of 0.015 (m m−1 a−1 ), and the coef- human-induced subsidence, the highest amount of subsidence
ficient of secondary compression (Cα ) of 0.0153 (–) used here, is measured in urban areas such as cities and industrial com-
derive from previously published values of Holocene peat in the plexes (Erkens et al., 2015). Here, concentrations of human pop-
Netherlands (Van der Meulen et al., 2007; De Lange et al., 2012; ulations and industry demand large quantities of fresh water,
Blok, 2014). They are two time-dependent parameters that de- which is extracted from deep aquifers. As a result, coastal cities
termine subsidence when primary compression is finished, and such as Tokyo (Japan) (Furuno et al., 2015), Shanghai (China)
therefore their importance increases with time. To assess the in- (Chai et al., 2004), Jakarta (Indonesia) (Abidin, et al., 2001) and
fluence of the selected parameters on the output of this study, Ho Chi Minh (Vietnam) (Thoang & Giao, 2015) have irreversibly
Vox and Cα were substituted for minimum and maximum val- subsided during the 20th and 21st centuries. These cities are
ues encountered in published research on Holocene peat in the generally subsiding at higher rates than their surrounding agri-
Netherlands. For Vox , values provided by Schothorst (1977) were cultural lands, because the groundwater extraction only affects
used (0.0048 and 0.0238 m m−1 a−1 ), and for Cα a minimum value the direct vicinity of exploitation wells.
of 0.007 (–) is provided by Fokker et al. (2015) and a maxi- In the coastal-deltaic plain of the Netherlands, subsidence
mum value of 0.0341 (–) by Visschedijk (2010). Again the hy- occurs due to a different mechanism; as a result of centuries
pothetical vertical voxel-stack of Figure 5 was used. The phreatic of human-induced phreatic groundwater lowering in reclaimed
groundwater level lowering and duration of scenario 4 was used polder areas, agricultural areas have experienced major subsi-
to determine the subsidence (0.5 m phreatic groundwater level dence due to compression and oxidation of shallow Holocene
lowering during 30 years), as this scenario provides the high- peat layers. These are still the most vulnerable areas for future
est subsidence values caused by oxidation and creep (longest subsidence when phreatic groundwater levels are progressively
duration, largest vertical interval of peat exposed to the atmo- lowered. In urban areas, thick anthropogenic brought-up soils
sphere). covering peat have caused it to compress in the past. Still, fu-
The black line in Figure 5 indicates subsidence for the hypo- ture subsidence is expected to be much lower here, since the
thetical voxel-stack with parameter values used in this study. underlying peat cannot oxidise, and the overburden made the
Total subsidence after 30 years is 20 cm. The red lines represent peat layers much less compressible. Consequently, surface ele-
differences in Vox , and green lines Cα . The graph shows that de- vations resulting from subsidence in the coastal-deltaic plain of
viations in subsidence by differential selection of Vox commence the Netherlands are reversed from those in many other urbanised
directly after the period of primary consolidation. After that coastal zones. If the process of periodical phreatic groundwater
period, the Vox lines show a divergent trend. This confirms that level lowering continues in the near future, the urban areas will
with increasing time, the influence of the selected Vox parameter become regional heights in a low-lying subsided coastal-deltaic
values increases. However, after a certain period, the minimum plain, whereas cities in many other coastal-deltaic plains will be
and maximum lines will show a convergent behaviour as peat situated in depressions.

224

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press

 Netherlands Journal of Geosciences — Geologie en Mijnbouw

Subsidence rates tion of the OCR of Holocene peat in GeoTOP could be conducted
using large available datasets of Cone Penetration Testing that
Subsidence rates in the Netherlands by peat oxidation and are maintained by TNO-GSN (Mayne & Kemper, 1988; Robertson,
compression have been relatively stable since the 1970s, with 2009).
regional subsidence rates ranging between 1 and 12 mm a−1 Another important physical peat property that reduces subsi-
(Schothorst, 1977; Nieuwenhuis & Schokking, 1997; Van den dence by compression and oxidation is the presence of admixed
Akker et al., 2008; Hoogland et al., 2012; Van de Born et al., sediments in peat (Koster et al., 2018a). Koster et al. (2018a)
2016). Predicted future subsidence rates in the peat areas of presented a method to assign peat voxels in GeoTOP with or-
the Netherlands are slightly lower, although they are within the ganic matter and sediment content, yielding 3D grids of dry
range of recent and present-day rates. Hoogland et al. (2012) mass and volumes of peat organic matter and sediments. It is
predicted future subsidence rates between <1 and 5.3 mm a−1 widely accepted that admixed sediments in peat reduce its com-
for near-surface peat layers in agricultural areas of the central pressibility as it occupies pore spaces. Furthermore, as oxidation
Netherlands for a period of 15 years, with water levels main- only affects peat organic matter, admixed sediments decrease
tained at fixed elevations relative to the subsiding surface. Sim- the volume lost during phreatic groundwater level lowering. The
ilar results were obtained by Van Hardeveld et al. (2017) for organic matter and sediment content of peat are implementable
subsidence of near-surface peat layers in agricultural areas of in the Bjerrum function (Den Haan, 1992); however, this does
the central and western part of the Netherlands. They predicted not account for peat oxidation functions. Consequently, in the
subsidence rates between 0.6 and 4.5 mm a−1 for the entire 21st presented subsidence modelling, 3D grids of peat organic matter
century, with water levels maintained between 30 and 90 cm be- and sediments were not used. It will therefore be a critical step
low the surface. forward to develop peat oxidation functions that comprise the
Subsidence rates predicted in this study range from 7 to amount of admixed sediments.
8 mm a−1 for scenarios 3 and 4 (30 year interval), and 11 to
13 mm a−1 for scenarios 1 and 2 (15 year interval). Since primary
Conclusions
compression takes place in the first few years after groundwa-
ter level lowering, the subsidence rates decrease with time. Al-
The subsidence potential of the urbanised and agricultural peat-
though our predicted future subsidence rates are twice as high
rich coastal-deltaic plain of the Netherlands was estimated for
as those of Hoogland et al. (2012) and Van Hardeveld et al.
periods of 15 and 30 years after a 0.25 and 0.5 m phreatic
(2017), they are in good correspondence with documented re-
groundwater level lowering. Agricultural areas were found to
cent and present-day subsidence rates.
have a higher subsidence potential than urbanised areas. Peat in
agricultural areas is often situated near the surface, and conse-
Model limitations quently very prone to oxidation, whereas peat beds in urbanised
areas are systematically overlain by anthropogenic brought-up
In this study, the Bjerrum function was applied to predict com-
soil and thereby protected from aeration. Furthermore, the an-
pression of peat (cf. CUR, 1992). The formulation used assumes
thropogenic brought-up soil causes peat underlying cities to
that the peat experiences virgin compression when phreatic
have low initial void ratios, and therefore the peat is less com-
groundwater levels are lowered. In practice, this will not be the
pressible than peat in agricultural lands. Subsidence by com-
case, as creep reduces the compressibility of peat by increasing
pression was more widespread than subsidence by oxidation, as
its vertical effective stress. This process is expressed by the over-
compression reduced the thickness of all peat layers whereas
consolidation ratio (OCR). The OCR of a peat layer determines
oxidation was restricted to those above the lowered phreatic
the threshold between reversible and irreversible compression
groundwater level. Future phreatic groundwater level lowering
and influences creep rates. Although the OCR was initially not
will result in increased elevation differences between the heavily
implemented in the Bjerrum function (CUR, 1992), at present
urbanised zones and the agricultural areas of the coastal plain,
it is considered a prime input parameter (Visschedijk, 2010).
with significant implications for maintaining infrastructure and
It is for future updates of the subsidence modelling to use the
land use planning.
state-of-the-art Bjerrum-function formulation, and to parame-
terise GeoTOP peat voxels with OCR. Implementation of lateral
and vertical differential OCR values in GeoTOP will for instance Acknowledgements
cause older and deeper-situated peat layers to experience mi-
nor elastic compression when phreatic groundwater levels are This study is part of the PhD research of K.K., funded by
lowered, whereas surficial peat layers will irreversibly compress TNO – Geological Survey of the Netherlands. Hans Middelkoop
and furthermore experience substantial creep. Mapping OCR over is acknowledged for his input and critical review of an earlier
large spatial scales, however, is challenging, as this needs to draft of the manuscript. Kim Cohen, Freek Busschers, Willem
be determined on local scales per subsurface unit. Implementa- Jan Zaadnoordijk and Aris Lourens are thanked for their input

225

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press

 Netherlands Journal of Geosciences — Geologie en Mijnbouw

at different stages of the research. We also thank G. Gambolati Hoogland, T., Van den Akker, J.J.H. & Brus, D.J., 2012. Modelling the subsidence
and an anonymous reviewer for their helpful comments. of peat soils in the Dutch coastal area. Geoderma 171–172: 92–97.
Kluiving, S., De Ridder, T., Van Dasselaar, M., Roozen, S. & Prins, M., 2016.
Soil archives of a Fluvisol: subsurface analysis and soil history of the medieval
city centre of Vlaardingen, the Netherlands – an integral approach. Soil 2:
References 271–285.
Koster, K., 2016. Cone Penetration Testing: a sound method for urban archaeolog-
Abidin, H.Z., Djaja, R., Darmawan, D., Hadi, S., Akbar, A., Rajiyowiryono, H., ical prospection. Archaeological Prospection 23(1): 55–69.
Sudibyo, Y., Meilano, I., Kasuma, M.A., Kahar, J. & Subarya, C., 2001. Land Koster, K., 2017. 3D characterization of Holocene peat in the Netherlands.
subsidence of Jakarta (Indonesia) and its geodetic monitoring system. Natural Implications for coastal-deltaic subsidence. PhD Thesis. Utrecht University
Hazards 23: 365–387. (Utrecht):185 pp. Utrecht Studies in Earth Sciences 140.
Blok, R., 2014. Tabellen voor bouw- en waterbouwkundigen. ThiemeMeulenhoff Koster, K., Stafleu, J., Cohen, K.M., Stouthamer, E., Busschers, F.S. &
(Amersfoort): 592 pp. Middelkoop, H., 2018a. Three-dimensional distribution of organic matter in
Bot, B., 2016. Grondwaterzakboekje. Bot Raadgevend Ingenieur (Rotterdam): coastal-deltaic peat: implications for subsidence and carbon dioxide emissions
456 pp. by human-induced peat oxidation. Anthropocene 22: 1–9.
Camporese, M., Gambolati, G., Putti, M. & Teatini, P., 2006. Peatland subsi- Koster, K., De Lange, G., Harting, R., De Heer, E. & Middelkoop, H., 2018b.
dence in the Venice watershed. In: Martini, I.P. et al. (eds): Peatlands: basin Characterizing void ratio and compressibility of Holocene peat with CPT for
evolution and records on global environmental and climatic changes. Elsevier assessing coastal-deltaic subsidence. Quarterly Journal of Engineering Geology
(Amsterdam): 529–550. and Hydrogeology, in press: 1–20. doi: 10.1144/qjegh2017-120.
Chai, J., Shen, S.-L., Zhu, H. & Zha, X.-L., 2004. Land subsidence due to ground- Kruiver, P., Van Dedem, E., Romin, R., De Lange, G., Korff, M., Stafleu, J.,
water drawdown in Shanghai. Géotechnique 54: 143–147. Gunnink, J.L., Rodriguez-Marek, A., Bommer, J.J., Van Elk, J. & Doornhof,
CUR (Civil Engineering Centre for Implementation Research and Regulation), D., 2017. An integrated shear-wave velocity model for the Groningen gas field,
1992. Construeren met grond. Report 162. CUR (Gouda): 411 pp. The Netherlands. Bulletin of Earthquake Engineering 15(9): 3555–3580.
De Lange, G., Gunnink, J., Houthuessen, Y. & Muntjewerff, R., 2012. Bodem- Mayne, P.W. & Kemper, J.B., 1988. Profiling OCR in stiff clays by CPT and SPT.
dalingskaart Flevoland. Rapport GM-0042778. Houten: Grontmij en Deltares: Geotechnical Testing Journal 11: 139–147.
58 pp. Mesri, G. & Ajlouni, M., 2007. Engineering properties of fibrous peats. Journal of
Den Haan, E.J., 1992. The formulation of virgin compression of soils. Géotech- Geotechnical and Geoenvironmental Engineering 133: 850–866.
nique 42(3): 465–483. NHI, 2016. Online portal of hydrological data of the Netherlands, www.nhi.nu.
Deverel, S.J., Ingrum, T. & Leighton, D., 2016. Present-day oxidative subsidence Nieuwenhuis, H.S. & Schokking, F., 1997. Land subsidence in drained peat areas
of organic soils and mitigation in the Sacramento-San Joaquin Delta, Califor- of the Province of Friesland. Quarterly Journal of Engineering Geology and
nia, USA. Hydrogeology Journal 24: 569–586. Hydrogeology 30: 37–48.
Drexler, J.Z., De Fontain, C.S. & Deverel, S.J., 2009. The legacy of wetland Pierik, H.J., Cohen, K.M., Vos, P.C., Van der Spek, A.J.F. & Stouthamer, E.,
drainage on the remaining peat in the Sacramento-San Joaquin Delta, Cali- 2017. Late Holocene coastal-plain evolution of the Netherlands: the role of
fornia, USA. Wetlands 29(1): 372–386. natural preconditions in human-induced sea ingressions. Proceedings of the
Erkens, G., Bucx, T., Dam, R., De Lange, G. & Lambert, J., 2015. Sinking coastal Geologists’ Association 128: 180–197.
cities. Proceedings of the IAHS 372: 189–198. Pons, L.J. & Van Oosten, H.F., 1974. De bodem van Noord-Holland. Toelichting bij
Fokker, P.A., Gunnink, J., De Lange, G., Leeuwenburgh, O. & Van der Veer, blad 5 van de bodemkaart van Nederland 1:200.000. Stichting voor Bodemkar-
E.F., 2015. Compaction parameter estimation using surface movement data in tering (Wageningen): 193 pp.
Southern Flevoland. Proceedings of the IAHS 372: 183–187. Robertson, P.K., 2009. Interpretation of cone penetration tests: a unified ap-
Furuno, K., Kagawa, A., Kazaoka, O., Kusuda, T. & Nirei, H., 2015. Groundwater proach. Canadian Geotechnical Journal 46: 1337–1355.
management based on monitoring of land subsidence and groundwater levels Schothorst, C.J., 1977. Subsidence of low moor peat soils in the western Nether-
in the Kanto Groundwater Basin, central Japan. Proceedings of the IAHS 372: lands. Geoderma 17: 265–291.
53–57. Serva, L. & Brunamonte, F., 2007. Subsidence in the Pontina plain, Italy. Bulletin
Gambolati, G., Putti, M., Teatini, P. & Gasparetto Stori, G., 2006. Subsidence of Engineering Geology and the Environment 66: 125–134.
due to peat oxidation and impact on drainage infrastructures in a farmland Stafleu, J., Maljers, D., Gunnink, J.L., Menkovic, A. & Busschers, F.S., 2011. 3D
catchment south of the Venice Lagoon. Environmental Geology 46(6): 814– modelling of the shallow subsurface of Zeeland, the Netherlands. Netherlands
820. Journal of Geosciences / Geologie en Mijnbouw 90: 293–310.
Higgins, S.A., 2016. Review: advances in delta-subsidence research using satellite Thoang, T.T. & Giao, P.H., 2015. Subsurface characterization and prediction of
methods. Hydrogeology Journal 24: 587–600. land subsidence for HCM City, Vietnam. Engineering Geology 199: 107–124.
Hijma, M.P., Cohen, K.M., Hoffmann, G., Van der Spek, A.J.F. & Stouthamer, E., TNO-GSN, 2016. Online portal for digital geo-information. Geological Survey of
2009. From river valley to estuary: the evolution of the Rhine mouth in the the Netherlands, www.dinoloket.nl/en.
early to middle Holocene (western Netherlands, Rhine-Meuse delta). Nether- Van Dam, P.J.E.M., 2001. Sinking peat bogs. Environmental change in Holland
lands Journal of Geosciences / Geologie en Mijnbouw 88(1): 13–53. 1350–1550. Environmental History 1: 32–46.

226

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press

 Netherlands Journal of Geosciences — Geologie en Mijnbouw

Van de Plassche, O., 1982. Sea-level change and water-level movements in the lands. Netherlands Journal of Geosciences / Geologie en Mijnbouw 92: 217–
Netherlands during the Holocene. PhD Thesis. VU University Amsterdam (Am- 241.
sterdam). Mededelingen Rijks Geologische Dienst 36: 1–93. Van Hardeveld, H.A., Driessen, P.P.J., Schot, P.P. & Wassen, M.J., 2017. An
Van den Akker, J.J.H., Kuikman, P.J., De Vries, F., Hoving, I., Pleijter, M., integrated modelling framework to assess management strategies steering
Hendriks, R.F.A., Wolleswinkel, R.J., Simoes, R.T.L. & Kwakernaak, C., soil subsidence in peatlands. Environmental Impact Assessment Review 66:
2008. Emissions of CO2 from agricultural peat soils in the Netherlands and 66–77.
ways to limit this emission. In: Farrell, C. and Feehan, J. (eds): After wise use Visschedijk, M., 2010. Isotachen berekening op een sigarendoosje. Geotechniek
– the future of peatlands. Proceedings of the 13th International Peat Congress, Juli: 30–3.
8–13 June 2008, Tullemore, Ireland: 1–6. Conference proceedings. Vos, P.C., 2015. Origin of the Dutch coastal landscape. Long-term landscape evo-
Van den Born, G.J., Kragt, F., Henkens, D., Rijken, B., Van Bemmel, B. & Van lution of the Netherlands during the Holocene, described and visualized in na-
der Sluis, S., 2016. Dalende bodem, stijgende kosten. PBL (The Hague): 94 pp. tional, regional and local palaeogeographical map series. PhD Thesis. Utrecht
Van der Meulen, M.J., Van der Spek, A.J.F., De Lange, G., Gruijters, S.H.L.L., University (Utrecht): 369 pp.
Van Gessel, S.F., Nguyen, B-L., Maljers, D., Schokker, J., Mulder, J.P.M. & Zanello, F., Teatini, P., Putti, M. & Gambolati, G., 2011. Long term peatland sub-
Van der Krogt, R.A.A., 2007. Regional sediment deficits in the Dutch low- sidence: experimental study and modelling scenarios in the Venice coastland.
lands: implications for long-term land-use options. Journal of Soils and Sedi- Journal of Geophysical Research 116: F04002. doi: 10.1029/2011F002010.
ments 7: 9–16. Zhang, Y., Huang, H., Yilin, L., Liu, Y. & Haibo, B., 2018. Spatial and temporal
Van der Meulen, M.J., Doornebal, J.C., Gunnink, J.L., Stafleu, J., Schokker, variations in subsidence due to the natural consolidation and compaction of
J., Vernes, R.W., Van Geer, F.C., Van Gessel, S.F., Van Heteren, S., Van sediment in the Yellow River delta, China. Marine Georesources & Geotechnol-
Leeuwen, R.J.W., Bakker, M.A.J., Bogaard, P.J.F., Busschers, F.S., Griffioen, ogy, in press: 1–11. doi: 10.1080/1064119X.2017.1414903.
J., Gruijters, S.H.L.L., Kiden, P., Schroot, B.M., Simmelink, H.J., Van Berkel, Zoccarato, C., Minderhoud, P.S.J. & Teatini, P., 2018. The role of sedimentation
W.O., Van der Krogt, R.A.A., Westerhoff, W.E. & Van Daalen, T.M., 2013. 3D and natural compaction in a prograding delta: insights from the mega Mekong
geology in a 2D country: perspectives for geological surveying in the Nether- delta, Vietnam. Scientific Reports 8: 1–12. doi: 10.1038/s41598-018-29734-7.

227

https://doi.org/10.1017/njg.2018.11 Published online by Cambridge University Press