Engineering properties of stiff Bolders Bank glacial clay till from Cowden

E. R. Ushev
ARUP, London, United Kingdom
T. Liu
University of Bristol, Department of Civil Engineering, Bristol, United Kingdom
R.J. Jardine
Imperial College London, Department of Civil and Environmental Engineering, London, United Kingdom

ABSTRACT: Many Baltic, North, and Norwegian Sea oil, gas and offshore wind structures are founded on
glacial tills, as are some US developments. This paper presents an overview of coordinated programmes of
monotonic and cyclic laboratory testing on natural low-to-medium plasticity, high OCR, stiff Bolders Bank
clay till. Profiles with depth of index properties, undrained shear strength and non-linear stiffness are com-
pared with in-situ measurements and their effective pressure level dependency reported. The till’s stiffness
anisotropy was investigated by high-resolution small-strain stress probing tests and through hollow cylinder
apparatus (HCA) tests which also established its shear strength anisotropy, which is quite different to that of
low OCR clays. The till’s cyclic response was characterised over a wide range of mean and cyclic shear stress
combinations, identifying stable, metastable and unstable styles of strain accumulation and stiffness response
that reflect the cyclic stress paths’ engagement with the till’s Kinematic Yield Surfaces (KYS) and proximity
to its monotonic failure envelope. The experiments highlight new insights, including the till’s relatively stiff
and strong response to horizontal loading, that have important practical implications and provide a basis for
developing and calibrating more representative constitutive models for offshore analyses and designs at till
sites.

into the behaviour of driven steel piles, see Ove Arup

1 Introduction
& Partners Ltd (1986), Lehane and Jardine (1994),
Pleistocene glaciers covered large areas of the North- Karlsrud et al. (2014).
ern Hemisphere and left tills with a wide range of ge- The Cowden till comprises 40 metres of stony, sandy
otechnical properties; Trenter (1999), Clarke (2018). and silty clay units that under-drain into two isolated
While glacial clay tills’ affect offshore foundation de- sand layers. Powell and Butcher (2003) summarised
sign and analysis at multiple, Baltic, North, Norwe- earlier laboratory and in-situ testing on Cowden till
gian Sea and US locations, their geotechnical behav- from the 1970s onwards by the UK Building Re-
iour remains relatively unexplored in the literature. search Establishment (BRE), concentrating on area
This paper presents an overview of in-situ and labor- around 200m west of the PISA site. This paper con-
atory research on natural low-to-medium plasticity, centrates on the clay units found over the top 12m, as
high yield stress ratio (YSR, or apparent OCR) stiff shown in below Figure 1a).
Bolders Bank till. Bolders Bank is one of the most While pore pressures vary seasonally at Cowden they
common glacial sequences encountered at UK North remain broadly hydrostatic below an average water
Sea oil, gas and offshore wind development sites. The table depth of 1.0m down to ≈5m, below which under
study focussed on the monotonic and cyclic behav- drainage into the first sand layer at 12.5m bgl reduces
iour of till sampled at Cowden, near Kingston-upon- the pressures as shown in Figure 1b). Block sampling
Hull, UK under the PISA Joint Industry Project (JIP) from supported excavations and wireline 102mm OD
described by Byrne et al. (2017). Cowden has been Geobore-S rotary sampling to 14m carried out for
employed as a test site for several previous studies PISA revealed highly weathered till to 1.5m, an
Figure 1 Profiles of a) natural water content, liquid and plastic limits and bulk density and b) pore-water pressure, total stress and K0

intermediate section where glacial and/or post-glacial The particle size distribution curves from 5 repre-
processes left desiccated and fissured (fissures were sentative tests are shown in Figure 2 along with upper
vertical to sub-vertical, closed but wet) till and a layer and lower bounds from Powell and Butcher (2003)
from 2.5 to 4.8m depth whose properties appear in- and a PSD curve from the sand encountered at around
distinguishable from the deeper un-weathered till, de- 12.5m bgl.
spite their different colour. Almost identical profiles
Overall, the tests show very broad distributions with
are reported by Powell and Butcher (2003) and Le-
around 30% clay, 40% silt, 22% sand and 8% gravel
hane and Jardine (1994).
fractions. The upper, highly weathered 1.5m till has
The natural water contents, liquid and plastic limit similar clay and gravel contents, although with less
and bulk density profiles are plotted in Figure 1a). silt and more sand. The gravel sized particles found at
Also shown are Powell and Butcher’s (2003) mean all depths were derived from chalk, limestone, sand-
trendlines. The till’s high densities reflect its the gla- stone, igneous and metamorphic rocks.
cial deposition that imparted a full grading distribu-
The K0 profile adopted in 3D-FE analyses of the PISA
tion, high stone contents and liquidity indices less
pile research tests is shown in Figure 1b). This recog-
than zero at shallow depths that rose slightly with
nises the vertical fissures found in the weathered till,
depth. Organic content increases from 0.7 to 1.6% be-
which may have been open in recent times, are incom-
tween 0.7 and 10m bgl, while carbonate contents rise
patible with the high K0 values interpreted by Powell
from 2.7% at 0.7m bgl to 12% at 10m bgl. Grain spe-
and Butcher (2003) and Lehane (1992), which are
cific gravity Gs is 2.73 over the top 2m and decreases
also plotted for reference.
to 2.71 below 4m.
2 CPT profiling reflect the complex glacial depositional and post-dep-
ositional processes. The lodgement till probably ex-
Piezocone Penetration Tests (PCPTs) performed at perienced intense horizontal shear loading during
the PISA research site extended down to the ‘first deposition, cycles of principal stress rotation and pe-
sand layer’ at 12.5m depth. As shown in representa- riods of both periglacial and temperate desiccation.
tive soundings in Figure 3, consistent linear trends ap-
Table 1 Critical state and oedometer parameters from intact 1-D
pear between 3m and 12m with an average corrected oedometer tests
qt ≈ 2MPa; several spikes from hard inclusions are
Depth σ'v0 σ*e Cc Cs Cαe St Ss
evident. A stiffer desiccated and fissured layer is seen
between 1.5m and 3m, where the cone resistance and m bgl kPa kPa - - - -
0.5 15.5 185 0.233 0.069 0.0049 0.822 1.1
the sleeve frictions were at least double the site aver- 0.75* 20 191 0.26 0.064 1.895 1
age. Powell and Butcher (2003) found similar pro- 1.25 27 165 0.249 0.069 0.0050 1.23 0.633
files, although their desiccated maxima were higher, 2.25 34 521 0.177 0.046 0.0022 1.614 1.15
at around 7MPa. CPTu pore-pressures measurements 3.3 49 866 0.154 0.035 0.0025 0.806 -
at the at u2 shoulder positions ranged from -50kPa to 3.5* 52 908 0.18 0.046 0.501 0.75
+300kPa, reflecting the till’s tendency to dilate mark- 4.3 62 511 0.166 0.044 0.0031 1.188 1.368
5.9* 82 1059 0.145 0.03 0.376 1.154
edly when sheared to large strains. Correlations be-
6 84 1616 0.136 0.037 0.0031 0.282 1.375
tween the laboratory CAU triaxial compression tests 7 98 764 0.154 0.044 0.0035 0.391 1.263
and the CPT profiles suggests a best fitting site-spe- 7.3 103 931 0.15 0.041 0.0035 0.786 1.333
cific average Nkt=18 (Ushev and Jardine, 2022b). 7.3* 103 619 0.145 0.021 0.562 2.222
10.15* 158 406 0.157 0.044 1.079 1

Figure 3 CPT profiles for qt, fs and u2 pore pressures at Cowden

PISA site (after Ushev and Jardine, 2022)

Also shown in Figure 4 is the average intrinsic normal

Figure 2 Particle size distribution curves compression line (NCL*) derived from 12 tests on
un-weathered reconstituted till samples from below
1.5m depth. The void ratio-mean effective stress rela-
3 One dimensional compression
tionship suggests a specific volume-equivalent
Void ratio (e) – vertical effective stress (σ′v) curves ‘Hvorslev’ mean effective stress pˊe relationship
from incremental and CRS oedometer tests on 50 and v=2.02-0.093ln(p′e), calculated assuming K0=1–sin
100mm diameter natural specimens are presented in φˊ=0.58 at OCR=1. All natural samples’ curves
Figure 4. Also plotted in the figure are three unweath- crossed the NCL* without converging with its gradi-
ered 1-D compression curves reported by Lehane ent, showing that the till possesses a dense and insen-
(1992). Vertical compression stresses up to sitive fabric which could not be erased by simple oe-
σ′v=10MPa did not lead to sharply defined yield dometer compression. The natural compression (Cc),
points. Casagrande construction was adopted to inter- swelling (Cs) and creep (Cαe) parameters applying un-
pret the yield stresses, σ′vy and compute the YSRs (or der 1-D consolidation conditions are summarised in
apparent OCRs=σ′vy/σ′v0) plotted in Figure 5, which Error! Reference source not found. along with the
samples’ oedometer stress (St=σ'vy/σ*e) and swell q/p′=MTXC=1.09, corresponding to Mohr-Coulomb
(Ss=C*s/Cs) sensitivities. φ′cs=27.5° with c′=0 while the un-weathered samples
gave MTXC=1.0 or φ′cs=25.4°. Undrained compres-
The uppermost 1.5m of till shows steeper curves that
sion tests on reconstituted samples gave similar out-
plot above the NCL* and reflect its open weathered
comes, see Ushev et al (2019).
structure, higher liquid and plastic limits, higher wa-
ter content and lower silt and sand content. A transi-
tion zone extends between 1.5 and 2.5m in Figure 4.
The average coefficient of consolidation, cv, applying
to the upper part of the profile is 1.8x10-8m2/s while
the deeper specimens show lower cv values that con-
verge towards 1x10-7m2/s at σ'v=250-500kPa. The
highly weathered, high clay content samples show oe-
dometer permeability of around 3x10-11m/s at stresses
up to 250kPa, which decreases gradually to around
1x10-11m/s at σ'v=8MPa. The un-weathered till sam-
ples exhibit a range of higher permeabilities at low
effective stresses, which converge at σ'v=250kPa and
decreases gradually to around 1x10-11m/s at
σ'v>10MPa.

Figure 5 Profiles for YSR and σ'vy derived from incremental oe-
dometer tests

The peak compression and extension triaxial test Su

values are plotted against depth with solid symbols in
Figure 6, along with best fit trend lines. The compres-
sion and (lower) extension Su profiles are similar in
shape and reflect, as with the CPT and σ′vy profiles,
Figure 4 1-D incremental compression curves from weathered the till’s glacial deposition and differ considerably
and un-weathered till
from the patterns expected for gravitationally consol-
idated waterborne sediments. The 1.0m depth sam-
4 Profiles of undrained shear strength ples exhibited the lowest Su values, which are likely
to vary seasonally. The compression shear strength
Monotonic undrained triaxial compression and exten- profile exhibited steep dSu/dz gradients and maxi-
sion stress path tests were undertaken on natural ro- mum Su in the desiccated zone between 2 and 3m bgl,
tary cored specimens consolidated to the estimated in- as shown by the qt and fs CPT profiles in Figure 3.
situ K0 stresses, shown in Figure 1b). Ushev and Compression and extension Su show local minima at
Jardine (2022b) show that the Cowden till’s response 4.8m level and increase linearly below this level.
is ductile and strain-hardening in compression and Ushev (2018) undertook parallel triaxial compression
only approaches critical state failures at large strain,
and extension tests on reconstituted specimens that
with ε>25%. Necking led to premature failures in ex- had been K0 consolidated to a range of OCRs and led
tension tests which masked the soil’s true response. to the power law relationship between Su/σ′v0 and
Weathered specimens exhibited a critical state failure OCR given by Equation 1.
 𝑆𝑢
′ = 𝐴(OCR)𝑏 .…………………………….(1) which falls only slightly below the triaxial φ′cs=25.4o
𝜎𝑣0
found with un-weathered samples. Lupini et al.
The best fits for parameters A and b were 0.28 and (1981) reported φ'residual=25o for Cowden till, while
0.75 respectively for triaxial compression conditions, Lemos (1986) found similar values of 24.5o-26.5o that
and 0.18 and 0.92 for extension tests. agreed also with Bromhead ring shear apparatus and
reversal shear box tests by Powell and Butcher
Also plotted with open symbols in Figure 6 are iso- (2003).
tropically consolidated, but relatively rapid, un-
drained triaxial tests from pushed, hammered and ro- Bishop interface shear tests that employed mild steel
tary samples reported by Powell and Butcher (2003) interfaces prepared with the ≈10μm CLA roughness
and rapid UU tests presented by Lehane (1992). of industrial piles and followed the Jardine et al
(2005) procedures gave similar average δpeak and ulti-
mate δresidual angles of 19.1o. However, Lehane and
Jardine (1994) reported slightly higher (20o to 24o) in-
terface δ values from highly instrumented field ICP
compression pile tests and lower δ=18o angles in cor-
responding tension pile tests.

6 Profiles of undrained triaxial stiffness

Ushev and Jardine (2022b) report on the till’s steeply

non-linear stiffness behaviour. High-resolution, lo-
cally instrumented triaxial tests, provided reliable
Young’s moduli Eu measurements under triaxial com-
pression and extension at axial strains of 0.001, 0.01,
0.1 and 1%, which are also equal to invariant shear
strain εs under undrained conditions. Their profiles
with depth for secant Eusec indicate nearly parabolic
trends with mean effective stress p′ for any given
strain level. Log-log plots of undrained triaxial
Young’s moduli against p′ at strain levels between
0.001 and 1% (Ushev 2018), indicate that Eu scales as
shown by Equation 2 (where pref is atmospheric pres-
Figure 6 Profiles of undrained shear strength from K 0 consoli- sure = 101.3kPa) and plotted in Figure 7. The secant
dated compression and extension tests (solid symbols) and his-
A values decrease consistently with invariant shear
toric laboratory tests (open symbols).
strain from around 1700 at 0.001% to 40 at 1.0%
The historic laboratory data plot below the trends strain. Ushev and Jardine (2022b) show how the scal-
from the Authors’ CAU tests on Geobore-S and block ing parameter A varies with shearing mode and pre-
samples, showing greater scatter and sensitivity to sent the equivalent relationships between tangent
their sampling methods. Powell and Butcher (2003) stiffness and invariant shear strain.
also report Su values from in-situ pressuremeter tests
0.5
which agree well with the mean TXC Figure 6 trend. 𝐸 𝑢 = 𝐴𝑝𝑟𝑒𝑓 (𝑝
𝑝′
) ……………………….(2)
𝑟𝑒𝑓

The 0.5 power law exponent given in Equation 2 for

5 Residual shear strength
the 12m deep profile of Bolders Bank till encountered
‘Bishop’ ring-shear apparatus soil-soil and soil-steel at Cowden is lower than the 0.85 and 0.75 values re-
interface tests explored the till’s ultimate residual ported by Jardine (1995) from triaxial Eu maxima
shearing behaviour at different vertical effective sensed with lower resolution electro-level local strain
stresses. Ushev (2018) reports residual soil-soil an- transducers and from Gmax measurements reported by
gles that increase with vertical effective stress and Viggiani and Atkinson (1995) with bender element
reached a plateau for φ′residual=24.5o when σ՛ >200kPa tests on other glacial tills.
7 Shear stiffness profiles opposing extreme respective b values of 1 and 0. In
addition, their tendency towards premature necking
Profiles are plotted in Figure 8 of Gvh, Ghh and Ghv failure renders extension tests unstable, unless the
from laboratory bender element measurements made samples fail at small strains. With strain hardening
on 100mm diameter samples after reconsolidation to soils, like Cowden till, the extension Su values are
in-situ K0 stresses. Also shown are two seismic CPT misleadingly low and give an inaccurate indication of
Gvh profiles reported for the PISA site by Zdravković shear strength anisotropy. Hollow Cylinder (HCA)
et al. (2020). While the in-situ Gvh values fell below experiments can avoid such problems by allowing α
the laboratory shear stiffness measurements at shal- and b to be varied independently to any combination
low depth and about 10% above them below 2.5m of values between 0 and 90o and 0 and 1 respectively.
bgl, there is a good agreement between laboratory
(bender element) and seismic field measurements. Liu et al. (2020) present HCA tests on Cowden till
which measured its undrained shear strength anisot-
ropy rigorously. While the triaxial tests indicate
SuTXC/SuTXE = 1.25, the HCA tests show that the till is
in fact ≈35% stronger when σ1 acts horizontally than
vertically under plane strain conditions.

Figure 7 Profiles with depth of normalised secant stiffness [Eu-

sec/pref]/[p'/pref]0.5 defined at axial strain levels of 0.001, 0.01,
0.1 and 1% in triaxial compression tests

8 Shear strength and stiffness anisotropy

Figure 8 Gmax profile with depth from laboratory and field geo-
Anisotropy in shear strength is often characterised, physical measurements
although inaccurately, in terms of ratios between un- The HCA experiments also show that the non-linear
drained triaxial compression and extension shear horizontal Young’s moduli are greater than those ap-
strengths. This does not account for the important ef- plying under vertical loading. Multi-axis (drained and
fects of different intermediate principal stress param- undrained) triaxial probing tests (as interpreted by Liu
eters b = (σ2 – σ3)/(σ1 – σ3), or Lode angle (θ) that et al. 2020) confirmed that the till has higher horizon-
apply in the two test types. While extension failure tal-than vertical elastic stiffness. The ratio of horizon-
involves the σ1 axis acting horizontally (at an angle α tal-to-vertical Youngs moduli (EUH/ EUV) ≈2.3 at 2m
= 90o to the vertical) and compression tests fail with and gradually reduces to ≈1.5 at 11m. The horizontal
σ1 acting vertically (with α = 0o) the tests also have laboratory EUH/3 profile deduced from triaxial tests is
close to those from dynamically measured field or la- loading variables imposed relative to the till’s ulti-
boratory shear stiffness trends, even though their mate failure envelope.
strain rates were far higher than imposed in the HCA
or triaxial probing tests.
10 Summary and conclusions

9 Cyclic loading response This mechanical behaviour of glacial clay tills has an
important impact on the design and analysis of off-
Ushev and Jardine (2022a) show that the Bolders shore foundations at multiple locations. While the
Bank till’s effects of the cyclic loading parameters on mechanical properties of tills are not reported widely
the number of cycles to failure and mean effective in the literature, recent characterisation research con-
stress drifts response can be summarised in contoured ducted in connection with the PISA JIP on the Bolders
interaction diagrams. An example is shown in Figure Bank till encountered at Cowden, Humberside, UK
9. The accumulated mean and cyclic strains, as well adds to current knowledge, as described in detail by
as the degradation of stiffness with strain and loading Ushev (2018), Ushev et al (2019), Liu et al (2020)
amplitude, can be expressed as empirical relation- Zdravkovic et al (2020) and Ushev and Jardine
ships. Ushev (2018) shows how the till’s cyclic re- (2022a, b). This paper summarises their findings; its
sponse differs in multiple respects from that shown by seven most important conclusions drawn are:
classical benchmark test cyclic sets, such as those on
1. The Bolders Bank till encountered at Cowden
low YSR Drammen marine clay; Andersen (2015).
has profiles of index properties, PSD compo-
sitions, vertical yield stresses and one-dimen-
sional compression strengths that differ from
those of waterborne sediments due to its gla-
cial deposition and post-depositional history.
2. The till’s Su, σ′vy and CPT’s qt and fs profiles
reveal a possibly glacially desiccated zone of
maximum shear strength and yield stress at
relatively shallow depth.
3. The till shows a generally ductile, strain hard-
ening, response to shearing. Its residual shear
Figure 9 Contours of number of cycles to failure as a function of strength is similar to its critical state angle, as
qcyc/2Su and qm/2Su as well as qcyc/p'in-situ vs qm/p'in-situ
identified in triaxial compression tests.
The Cowden till’s cyclic interaction diagrams can be
4. The till shows only a very limited elastic
divided into ‘safe’ and ‘unsafe’ zones as well as areas
range of stress-strain behaviour, beyond
showing ‘abrupt’ and ‘creeping’, failure modes in the
which its non-linear stiffnesses vary steeply
unsafe zone. The till’s cyclic behaviour can also be
with strain level. Stiffness also varies with
interpreted within a framework that recognises the
mean effective stress (pꞌ)0.5.
onset of significant changes of behaviour once care-
fully defined kinematic yield surfaces are engaged. 5. Field and bender element laboratory geophys-
Ushev and Jardine (2022a) show that cycling loading ical elastic Gvh shear moduli are closely com-
leads to completely stable outcomes with negligible parable, while the laboratory Ghh trends are
p′ drifts and stiffness degradation, if the cyclic paths marginally softer than those found in Gvh
remained within the till’s Y2 kinematic yield surface. shearing modes.
Cycling under such conditions allows the till to grow,
sustain or recover its secant stiffness and undrained 6. Drained and undrained multi-axis triaxial
shear strength. However, cyclic loading paths that en- probing and undrained Hollow Cylinder Ap-
gage the Y2 yield surface but remain within its outer paratus confirm that the till is highly aniso-
Hvorslev surface develop more significant p′ drifts, tropic with higher horizontal than vertical
permanent strains and stiffness losses that vary sys- stiffness and undrained shear strength. Triax-
tematically with the number of cycles and cyclic ial and HCA tests show 1.5 ≤ EUH/EUV ≤2.3
 and Suα=0/Suα =90 =0.73 under plane strain con- Lehane, B.M. and Jardine, R.J. 1994. Displacement pile behav-
ditions. iour in glacial clay. Canadian Geotechnical Journal. 31 (1), 79–
90
7. The till’s cyclic behaviour, which differs in Lemos, L.J.L. 1986. The effect of rate on residual strength of
multiple respects from that shown by low soil. PhD Thesis, Imperial College London
YSR marine clays can also be interpreted
Liu, T., E. Ushev, R. J. Jardine. 2020. Anisotropic stiffness and
within an effective-stress based critical state
shear strength characteristics of a stiff glacial till. ASCE J. of
and kinematic yielding framework. Geotechnical and Geoenvironmental Engineering, 146(12),
2020. https://doi.org/10.1061/(ASCE)GT.1943-5606.0002387

Lupini, J.F., Skinner, A.E. & Vaughan, P.R. 1981. The drained
Acknowledgements residual strength of cohesive soils. Géotechnique. 31 (2), 181–
213.
The Authors acknowledge gratefully the funding pro-
vided for their research by the PISA project, Ørsted Ove Arup & Partners Ltd 1986. Research on the Behaviour of
and Imperial College London They also thank the Piles as Anchors for Buoyant Structures. HMSO Books, Lon-
don.
PISA project’s Academic Working Group’s for their
support and the Imperial College Geotechnics tech- Powell, J., and Butcher., A. 2003. Characterisation of a glacial
nical team who supported the experimental work de- clay till at Cowden, Humberside. In Proceedings of the Int.
Workshop on Characterisation and Engineering Properties of
scribed. The Cowden rotary core and block sampling
Natural Soils. CRC Press, London, pp. 983-1020, 2003.
was undertaken by Concept Drilling Service Ltd and
SOCOTEC, respectively, as part of the PISA JIP pro- Trenter N.A. 1999. Engineering in glacial tills. Construction In-
dustry Research Information Association (CIRIA) Report C504,
ject managed by Ørsted.
London, UK, p 260.
References Ushev, E. 2018. Laboratory investigation of the mechanical
properties of Cowden till under static and cyclic conditions”,
Andersen, K.H. (2015). Cyclic soil parameters for offshore
PhD Thesis, Imperial College London.
foundation design. The 3rd McClelland Lecture. Proc Int. Symp
on Frontiers in Offshore Geotechnics. Oslo, Pub. Taylor and Ushev, E., Schutt, D. and Jardine, R. 2019. Stress path testing on
Francis, London, pp 5-82. intact and reconstituted Bolders Bank till. In: Proceedings of the
XVII ECSMGE. 2019
Byrne, B. W., R. A. McAdam, H. J. Burd, G. T. Houlsby, C. M.
Martin, W. Beuckelaers, L. Zdravkovic, D. M. G. Taborda, D. Ushev, E., R. J. Jardine. 2022a. The behaviour of Bolders Bank
M. Potts, R. J. Jardine, E. Ushev, T. Liu, D. Abadias, K. Gavin, glacial till under undrained cyclic loading. Géotechnique, 72(1),
D. Igoe, P. Doherty, J. S. Gretlund, M. P. Andrade, A. Muir 1-19, 2022a. https://doi.org/10.1680/jgeot.18.P.236
Wood, F. Schroeder, S. Turner, M. Plummer. (2017). PISA: New
design methods for offshore wind turbine monopiles”. Society Ushev, E., R. J. Jardine. 2022b. The mechanical behaviour of
for Underwater Technology: 8th International Conference on Bolders Bank till. Canadian Geotechnical Journal, 59, 2163-
Offshore Site Investigation and Geotechnics, Smarter Solutions 2183, 2022b. https://doi.org/10.1139/cgj-2021-0436
for Offshore Developments, London, UK, vol. 1, pp. 142-161, Viggiani, G. and Atkinson, J.H. 1995. Stiffness of fine-grained
Byrne, B. W. 2021. Lateral pile design for offshore wind tur- soil at very small strains. Géotechnique. 45 (2), 249–265.
bines. In Piling 2020: Proceedings of the Piling 2020 Confer- Zdravković, L., B. W. Byrne, D. M. G. Taborda, H. J. Burd, K.
ence, pp. 13-33. ICE Publishing. Gavin, G. T. Houlsby, D. Igoe, R. J. Jardine, T. Liu, C. M. Mar-
Clarke, B. G. 2018. The engineering properties of glacial tills. tin, R. A. McAdam, A. Muir Wood, D. M. Potts, J. Skov
Geotechnical Research 5(4), 262-277, 2018. Gretlund, and E. Ushev. 2020. Ground characterisation for PISA
https://doi.org/10.1680/jgere.18.00020 pile testing and analysis. Géotechnique, 70(11), 945-960.
https://doi.org/10.1680/jgeot.18.PISA.001
Jardine, R.J. 1995. One Perspective on the Pre-Failure Defor-
mation Characteristics of some Geomaterials. In: International
Symposium on Pre-Failure Deformation Characteristics Of Ge-
omaterials. 1995 Hokkaido, Japan. pp. 885–886

Karlsrud, K., Jensen, T.G., Wensaas Lied, E.K., Nowacki, F., et

al. 2014. Significant ageing effects for axially loaded piles in
sand and clay verified by new field load tests. In: Proceedings of
the Offshore Technology Conference. 2014 Houston.

Lehane, B. 1992 Experimental investigations of pile behaviour

using instrumented field piles. PhD Thesis, Imperial College
London