Earthquake Geotechnical Engineering for Protection and Development of

Environment and Constructions – Silvestri & Moraci (Eds)

© 2019 Associazione Geotecnica Italiana, Rome, Italy, ISBN 978-0-367-14328-2

Observations on the response of natural fine-grained soils with

respect to particle fabric

D. Wijewickreme, A. Soysa & P. Verma

Department of Civil Engineering – University of British Columbia, Vancouver, BC, Canada

ABSTRACT: Fine-grained silty soils with high levels of saturation are commonly found in
natural river deposits, and experience from recent earthquakes suggests that such soils have
the potential for earthquake-induced softening and strength reduction. In spite of significant
research, only limited work has been undertaken to study the effects of soil particle fabric/
microstructure on the mechanical behavior of natural silts. With this background, a systematic
laboratory program involving monotonic and cyclic shear tests conducted on natural fine-
grained soils was undertaken, and the results from this work are drawn to illustrate the
importance of soil particle fabric/microstructure in governing the response of silt in addition
to the traditionally considered effects from the void ratio (density) and confining stress. The
void ratio (density) and particle fabric seem to compete in governing the mechanical perform-
ance (e.g., shear stiffness and strength), and the lack of accounting for soil particle fabric seem
to impede comprehensive definition of soil behavior

1 INTRODUCTION

Experience from recent earthquakes (e.g., Turkey, Kocaeli 1999; New Zealand, Christchurch
2010-2011) suggests that fine-grained silty soils with high levels of saturation have the poten-
tial for earthquake-induced softening and strength reduction, in turn, leading to geotechnical
hazards: viz., permanent vertical and lateral ground movements, post-cyclic settlements, foun-
dation sliding, tilting and collapsing of structures. While significant research efforts have been
devoted to understand the effect of a wide range of factors such as confining stress, void ratio,
particle size, soil plasticity, stress path and history, etc., on the monotonic/cyclic shear loading
response of silts, only limited work has been undertaken to study the effects arising due to the
differences in soil fabric/microstructure on the mechanical behavior of natural silts.
Soil is a multi-phase particulate material comprising solids, liquids, and gases and consist of
particles that are of different sizes and shapes, yet, it is often treated as a continuum material
in geotechnical engineering. The lack of accounting with respect to shape, size, gradation and
the arrangement of soil particles have limited our ability to comprehensively understand and
predict the mechanical behavior of soils, in turn, causing significant deficiencies between the
observed response of soils and those predicted from well-established soil mechanics theories.
The term soil fabric has been generally used to describe the geometric/spatial arrangement
of soil particles or particle groups and the associated voids in a granular mass (Brewer 1964).
Voids and fissures at larger scale and stratifications account for macrofabric of soil, whereas
microfabric is in the microscopic scale addressing the size, shape, orientation of particles and
pores either in domains or clusters. Collins & McGown (1974) identified three main groups of
fabric elements as: (i) elementary particle arrangements in which particle interactions at indi-
vidual level are considered; (ii) particle assemblies where one or more elementary particle
arrangement forms a definable physical boundaries with distinguishable specific mechanical
function; and (iii) pore spaces in which remaining phases other than soil solids (i.e. liquid and
gas) that fill the void volume is considered.

335
 In assessing the mechanical properties of granular soils, Oda (1972) reported that orienta-
tion of an individual particle and position of the particle and its mutual relationship to other
particles are two important fabric features. The orientation of individual particle, inter-par-
ticle contact orientation and particle shape descriptors alone do not seem to be capable of
effectively describing the stress-strain response of fine-grained soils, as complex force trans-
mission through irregular voids arrangements also contribute significantly to the overall
mechanical response of fine-grained soils.
It has been noted that the discrete nature of the particle and the spatial arrangement of soil
particles are responsible for characterizing the mechanical behavior of soils – in particular,
some characteristics that are not robustly explainable using the commonly used parameters
such as void ratio (density) and confining stress. Geotechnical field and laboratory element-
level investigations, in a similar fashion to the approaches used for the sands, have a key role
to play in advancing the fundamental knowledge on the soil fabric effects with respect to silty
soils. The presence and effects of soil particle fabric is often indirectly observable through sig-
nificant differences or changes in the mechanical behavior of soils that take place with respect
to situations such as: (i) disturbance during soil sampling; (ii) mechanical (laboratory) recon-
stitution; (iii) changes in the magnitudes of shear or confining stresses; (iv) ageing; (v) the
manner in which soils get deposited; etc.
With this background, a systematic laboratory program has been undertaken on this subject -
particularly considering the above potential situations that would allow observing the presence
and effects of soil particle fabric - as a part of a major study on the earthquake response of fine-
grained soils from a number of natural soil deposits in British Columbia, Canada. The factual
findings from both monotonic and cyclic shear tests conducted on natural fine-grained soils
during this work are drawn to illustrate the behavior specifically in relation to the following:
[a] Shear loading response of low-plastic, non-sensitive, normally consolidated silt from two
locations in the Lower Mainland of British Columbia with respect to relatively undisturbed
and reconstituted specimens.
[b] Effects of destructuration due to: (i) consolidating the relatively undisturbed, relatively
high-plastic sensitive silt/clay specimens to significant stress magnitudes beyond the precon-
solidation stress, and (ii) reconstitution of the soil.
[c] Initial and repeated cyclic shear loading response of non-plastic, non-sensitive, normally
consolidated silt observed from reconstituted specimens.
[d] Effects of ageing on the reconstituted low-plastic, non-sensitive, normally consolidated silt
on the cyclic shear resistance.
[e] Effects of disturbance during specimen preparation in the laboratory on the monotonic
shear loading response of low-plastic normally consolidated silt.
The paper initially presents the details related to the soil materials, experimentation meth-
odology, and testing program. This is followed by several sub-sections presenting the experi-
mental results in relation to the above five aspects of the testing program along with
corresponding observations.
From a completeness point of view, the findings that have been reported in previous publi-
cations by the author(s) are also encompassed in the compilation. The important role that soil
particle fabric/microstructure plays in governing the response of silt in addition to the trad-
itionally considered effects from the void ratio and confining stress is highlighted through
these findings. The void ratio (density) and particle fabric seem to compete in governing the
mechanical performance (e.g., shear stiffness and strength), and it appears that the lack of
accounting for soil particle fabric can significantly impede the definition and the applicability
of critical state concepts.

2 EXPERIMENTAL METHODS AND TEST MATERIALS

The results presented herein arise from a number of experimental research programs under-
taken on a range of fine-grained soils at the University of British Columbia (UBC),

336
Vancouver, Canada. A major part of the testing had been conducted using the UBC direct
simple shear (DSS) device, which is a modified Marshall-Silver-NGI-type DSS device (Bjer-
rum & Landva 1966, Silver & Seed 1971) which uses a cylindrical soil specimen that is placed
inside a steel wire-reinforced rubber membrane. The remaining results arise from element tests
conducted using the conventional triaxial apparatus.
The details about the experimental methods and test materials are summarized in Table 1 to
serve as a key to the investigations undertaken. For the tests on relatively undisturbed soils,
the soils obtained, using specially fabricated stainless steel, thin-walled tube sampler (with no
inside clearance and 5 degree cutting edge), from different locations in the Lower Mainland of
British Columbia, Canada were used. Both monotonic and cyclic shear response of silts were
explored. The reconstituted test specimens were prepared using slurry deposition method. The
slurry deposition method has been also used in many previous research studies at UBC and is
considered to better mimic the in-situ natural deposition of the silt in a river environment
(Sanin 2010, Soysa 2015, Verma & Wijewickreme 2016) compared to other soil specimen
reconstitution techniques for fine-grained soils such as moist-tamping. Keeping the above con-
siderations in mind, the laboratory testing program at UBC was developed. Specific details
related to the specimen preparation and experimental approaches are described as appropriate
in the following section. Further supplementary information can be found in the previous
UBC publications (e.g., Wijewickreme & Sanin 2006, Sanin 2010, Soysa 2015).

3 EXPERIMENTAL RESULTS

3.1 Behavioral differences between undisturbed and reconstituted silts

Laboratory element tests play a key role in understanding and improving the current know-
ledge on the fundamental load-deformation behavior of soil in geotechnical engineering.
Needless to say, that securing good quality – i.e., “undisturbed” – soil specimens for labora-
tory testing is always a key challenge. Due to the difficulties involved in undisturbed field sam-
pling of sands (Yoshimi et al. 1978, Hofmann et al. 2000), most of the studies have been
performed using reconstituted sand specimens with different specimen reconstitution tech-
niques (e.g., Oda 1972, Ladd 1974, Miura & Toki 1982, Vaid et al. 1999). The observed mech-
anical response from the sand specimens prepared using these different reconstitution
techniques were often found to be different. Similarly, different mechanical response has also
been reported in the literature between the undisturbed and reconstituted sand specimens.
Even in the case of fine-grained soils, particularly those that are of relatively low plasticity,
obtaining high quality samples that could meaningfully represent the field soil conditions is a sig-
nificant challenge – due to the disturbance and destructuration that occur during the sampling
process. Studies undertaken on fine-grained soils by many researchers (e.g., Ladd & DeGroot
2003, Santagata & Germaine 2002, Zapata-Medina et al. 2014, Amundsen et al. 2016) serve testi-
mony to these difficulties. As an alternative, testing of reconstituted fine-grained soil specimens is
generally considered to assess the mechanical response of such soils. In this situation, it is import-
ant to select a reconstitution technique that closely mimic the deposition of soil particles in real
life. Leroueil et al. (1979), Oda (1972), Høeg et al. (2000), Long et al. (2001), and Leroueil &
Hight (2003) have noted that the response from specimens prepared from relatively undisturbed
soil samples are significantly different from those observed from soil reconstituted from the same
material. They attributed the difference in the observed response to the difference in soil fabric/
microstructure between the undisturbed and reconstituted soil specimens.
The shear stress-strain and stress path response derived from constant-volume monotonic
DSS tests on relatively undisturbed and reconstituted silt specimens from channel-fill Fraser
River deltaic deposits in British Columbia, Canada are compared in Figure 1. The preconsoli-
dation stresses of the silt specimens derived from one-dimensional consolidation tests were
found to be about 80 ~ 88 kPa, while; the corresponding estimated in situ overburden stresses
based on the field depths were about ~79 kPa and ~84 kPa. Therefore, tests performed on spe-
cimens consolidated to initial vertical effective confining stress (σ′vc) of 100 kPa, 200 kPa, and

337
Table 1. Details of the tested soil and research investigations.
Research
Soil type Material Description Test Type Investigations#

Channel- • Depth range below ground surface: 5.6~8.7 m • TX • [a], [d], and [e]
fill • Water content: 34~39% • DSS
Fraser • LL: ~30%, PL: ~26%, PI: ~4%
River • Sand: ~10%, Silt: ~80%, Clay: ~10%
deltaic • Unified soil classification: ML
deposits* • Specific gravity: 2.69
• Estimated preconsolidation stress: ~85–95 kPa
Nicomekl • Depth range below ground surface: 4.2~5.5 m • DSS • [a]
River • Water content: 38~53%
deposits* • LL: ~41%, PL: ~34%, PI: ~7%
• Sand: ~20%, Silt: ~65%, Clay: ~15%
• Unified soil classification: ML
• Specific gravity: 2.77
• Estimated preconsolidation stress: ~35–45 kPa
Inland • Depth range below ground surface: 4.9~6.2 m • DSS • [b]
Fraser • Water content: 58~69%
River • LL: ~76%, PL: ~42%, PI: ~34%
deposits* • Sand: ~0%, Silt: ~10%, Clay: ~90%
• Unified soil classification: MH
• Specific gravity: 2.75
• Estimated preconsolidation stress: ~75–85 kPa
Fraser • Bulk sampling • DSS • [c]
River • LL: -, PL: -, PI: Non-plastic
deposits** • Sand: ~0%, Silt: ~82%, Clay: ~18%
• Unified soil classification: ML
• Specific gravity: 2.72

* Thin-walled stainless steel tube (“Undisturbed”) soil samples; ** Bulk disturbed soils samples; #As
identified in Section 1 LL: Liquid Limit; PL: Plastic Limit; PI: Plasticity Index; TX: Triaxial test; DSS:
direct simple shear test.

400 kPa were considered to demonstrate the loading response of normally consolidated silt
specimens. As shown in Figure 1, the post-consolidation void ratios (ec) observed for the rela-
tively undisturbed specimens were found to be slightly greater than those derived for the
reconstituted specimens when compared under similar consolidation stress levels. In other
words, reconstituted silt specimens prepared from slurry deposition method indicated a denser
particle structure than that of relatively undisturbed specimens. In terms of stress-strain
response, the undisturbed specimens (as notable from Figure 1) indicates a strain hardening
type response (increasing shear resistance with increasing shear strain) in contrast to those
observed for reconstituted specimens.
In essence, the undisturbed specimens, despite having a slightly higher void ratio (lesser dens-
ity), seem to display a relatively competent stress-strain response than the reconstituted speci-
mens. The behavioral contrast between the relatively undisturbed and reconstituted silt
specimens for constant-volume monotonic DSS tests conducted at σ′vc ≤ 400 kPa suggest that
the original structure of the relatively undisturbed specimens seem to prevail even after consoli-
dating the material to a confining stress level significantly higher than the estimated preconsoli-
dation stress for the natural soil. Høeg et al. (2000) performed an experimental study on
Borlánge silt using triaxial test device and reported weaker shear resistance and lower undrained
shear strength for reconstituted silt specimens when compared to the counterpart undisturbed
silt specimens, although the ec of reconstituted specimens were lesser (if not similar). The obser-
vations made under this study are in accord with the findings by Høeg et al. (2000).

338
Figure 1. Shear stress-strain (I) and stress path (II) response from constant-volume monotonic DSS
tests on relatively undisturbed and reconstituted silt specimens from channel-fill Fraser River deltaic
deposits for σ′vc = 100, 200 and 400 kPa (after Wijewickreme & Sanin 2008 and Sanin 2010).

The accumulation of shear strain and development of excess pore water pressure for the
reconstituted specimens (see Figure 2) during constant-volume cyclic DSS testing seem to
occur at higher rates when compared with those observed for relatively undisturbed specimens
when tested at similar cyclic stress ratio (CSR). When cyclic shear resistance is assessed based
on the number of loading cycles to attain a selected threshold shear strain value of 3.75%, the
reconstituted specimens, despite having higher density (lower ec), consistently showed signifi-
cantly weaker cyclic shear resistance than those derived for relatively undisturbed specimens
as shown in Figure 2. Considering that the comparisons are made at essentially identical
effective confining pressures, the results shown in both Figure 1 and Figure 2 reinforce the
significant dominance of natural fabric of the silt in competing with the effects arising from
density alone on the mechanical loading response.
In contrast to the results shown in Figure 1, for relatively undisturbed silt (PI = 7), retrieved
from Nicomekl River deposits in Lower Mainland, British Columbia Canada, showed a simi-
lar or weaker stress-strain response when compared with the counterpart reconstituted speci-
mens (see Figure 3) in constant-volume monotonic DSS tests when tested under normally

Figure 2. Comparison of shear strain accumulation (I) and pore water pressure development (II) during
constant-volume cyclic DSS test and cyclic stress ratio versus number of loading cycles for shear strain of
3.75% (III) for relatively undisturbed and reconstituted specimen from channel-fill Fraser River deltaic
deposits (after Wijewickreme & Sanin 2008 and Sanin 2010).

339
Figure 3. Shear stress-strain (I) and stress path (II) response from constant-volume monotonic DSS
tests on relatively undisturbed and reconstituted silt specimens from Nicomekl River deposits for σ′vc =
100, 200, and 400 kPa (after Soysa & Wijewickreme 2015a).

consolidated conditions (Soysa & Wijewickreme 2015a). Initial stiffness and shear stress-strain
characteristics up to about 0.5% shear strain of reconstituted and relatively undisturbed speci-
mens (for a given confining stress) are almost identical. However, at relatively large shear
strain values (~15%), the reconstituted specimens displayed a slightly higher shear resistance
than that observed for undisturbed specimens. At all tested consolidation stress levels, recon-
stituted specimens indicate slightly higher peak shear strength than those of relatively undis-
turbed specimens. It is important to note that the reconstituted silt specimens prepared from
Nicomekl River deposits have a lower ec value than those of relatively undisturbed specimens
under similar confining stresses. Similar responses have been observed by Fleming & Duncan
(1990) for Alaskan silt, when undrained shear strength of undisturbed specimens was com-
pared with that of the reconstituted specimens. From the results of isotropically consolidated
undrained triaxial tests on Alaskan silt, Fleming & Duncan (1990) noted that remolded and
reconsolidated specimens showed an increased undrained shear strength of 25% to 40% com-
pared to undisturbed specimens.
Typically, decrease in void ratio (denser particle arrangement) should result in an increasing
shear stiffness and strength, whereas destructuration of the particle fabric should causes an
opposing effect. It appears that in the observed monotonic shear loading response of reconsti-
tuted specimens from channel-fill Fraser River deltaic deposits, the effect from change in
fabric and micro-structure has over-shadowed the influence due to the change in void ratio,
while it is the opposite for the reconstituted specimens from Nicomekl River deposits.
However, a significantly opposing result see to emerge when the constant-volume cyclic
DSS loading response between reconstituted and undisturbed silt specimens from Nicomekl
River deposits are compared. The reconstituted specimens of this material indicated higher
rate of shear strain accumulation and excess pore water pressure development than those
observed for relatively undisturbed specimens [see Figure 4(I), (II)]. As such, the reconstituted
Nicomekl River silt yielded significantly lower cyclic shear resistance [shown in Figure 4 (III)]
compared to those from undisturbed specimens under cyclic loading, despite the relatively
lesser void ratio of reconstituted specimens. It is possible that the evolution of particle struc-
ture rearrangement during monotonic shear loading might be different to that during the
cyclic shear loading (which involves cyclic rotation of principal stresses). This might be the
source for the observed difference in behavior of reconstituted and undisturbed specimens
during cyclic and monotonic shear loading where either effect from the change in fabric or the
void ratio could be prominent for a given case.
Figure 5 presents the void ratio (e) versus effective confining stress (σ′v) relationships
derived from the constant-volume monotonic DSS testing of relatively undisturbed and recon-
stituted specimens at initial consolidation (and prior to shearing) and after attaining a shear
strain of about 15%. The data derived for undisturbed and reconstituted specimens can be
expressed by a linear relationship in the e-log σ′v space for case 1 - specimen state prior to

340
Figure 4. Comparison of shear strain accumulation (I) and pore water pressure development (II) during
constant-volume cyclic DSS test and cyclic stress ratio versus number of loading cycles for shear strain of
3.75% (III) for relatively undisturbed and reconstituted specimen from Nicomekl River deposits (after
Soysa & Wijewickreme 2015a).

shearing - with the two lines paralleling each other. Similar relationship (another two lines
parallel to each other) can be made for case 2 – specimen state after reaching a shear strain of
about 15%. It is also important to note that, although the soil type/material is identical, and
the mode of shear loading is similar, the straight lines depicting the state in case 1 for the rela-
tively undisturbed and reconstituted specimens are at distinctly different locations in the e-log
σ′v space. This is much in accord with the observations on an intrinsic consolidation line made
by Burland (1990). The observation of distinctly different locations in the e-log σ′v space
between undisturbed and reconstituted specimens is also notable for the straight lines

Figure 5. e-log σ‘v relationships for relatively undisturbed and reconstituted silt specimens of channel-
fill Fraser River deltaic deposits (I) and Nicomekl River deposits (II) after initial consolidation (prior to
shearing) compared with those after reaching ~15% shear strain in monotonic constant-volume DSS
tests.

341
depicting the e-log σ′v after achieving a ~15% shear strain state (case 2). In an overall sense,
the experimental observations presented herein, with respect to the e-log σ′v domain, suggest
that the reconstituted and undisturbed silt specimens have exhibited significantly different
monotonic loading characteristics, in spite of their identical mineralogical origin and grain
size. These differences also suggest that the commonly used variables e and σ′v alone is not
sufficient to define/determine the shear behavior of low plasticity silt.

3.2 Effect of changes in the magnitudes of confining stress

Ladd (1964) noted that the undrained shear strength of fine-grained soils can be characterized
by normalizing with respect to the vertical effective consolidation stress. This approach is
known as the SHANSEP (Stress History and Normalized Soil Engineering Properties), and it
is widely used for characterizing the fine-grained soils in engineering practice. For example,
for normally consolidated, non-sensitive clays, identical stress-path characteristics can be
observed when shear stresses are normalized with respect to the initial vertical effective con-
solidation stresses. The behavior of natural undisturbed silt from both channel-fill Fraser
River deltaic deposits and Nicomekl River deposits, which were presented and discussed in
above section, are found to be stress history-normalizable (Sanin 2010, Soysa 2015). This indi-
cates that the response of normally consolidated silt is similar to that typically observed for
normally consolidated clays (Atkinson & Bransby 1978).
In contrast, the relatively high plastic (PI = 34) silt/clay retrieved from inland Fraser River
deposits in British Columbia, Canada exhibited a different monotonic direct simple shear
loading response. As shown in Figure 6, the shear resistance for relatively high plastic (PI =
34) silt was found to increase with increasing shear strain until a peak is reached; this behavior
is then followed by a decrease in shear resistance with further increase in shear strain. It is also
visible that the degree of post-peak drop in shear resistance decreased as the level of effective
consolidation stress level increased above the preconsolidation stress (~75-85 kPa) for the rela-
tively undisturbed specimens. For instance, significant post peak drop is visible in the stress-
strain curve obtained for the specimen at σ′vc of 100 kPa; while post peak drop observed for
the specimen tested at σ′vc of 400 kPa was negligible. Moreover, it is important to note that
the observed shear resistance also decreases, when the specimens were tested at greater σ′vc. As
indicated in Soysa & Wijewickreme (2015c), when the consolidation stress for these soils are
increased beyond the estimated preconsolidation stress (~75-85 kPa), a significant amount of
vertical strain with a substantial reduction in the void ratio was noted which suggests a pos-
sible ‘destructuration’ of the initial particle fabric/microstructure in the soil specimen as the
confining stress increases. This collapse of natural fabric that the relatively undisturbed speci-
men previously possessed seems to result in comparatively weaker normalized shear resistance
at σ′vc of 400 kPa, despite it has denser particle arrangement (lower void ratio) than that for
the case at σ′vc of 100 kPa.
The cyclic shear resistance characteristics in terms of cyclic stress ratio versus number of
loading cycles to reach shear strain of 3.75% shown in Figure 7 further corroborated the
above attributed possible destructuration with increasing effective consolidation stress level
for the soils from inland Fraser River deposits. The cyclic shear resistance clearly seems to
reduce with increasing σ′vc.
It should also be noted that the comparison of relative undisturbed and reconstituted speci-
mens prepared from inland Fraser river deposits for monotonic DSS loading response at σ′vc
of 100 kPa, shown in Figure 6, clearly indicated weaker shear resistance of reconstituted speci-
men despite its’ lower void ratio. Further, reconstituted specimen at σ′vc of 100 kPa did not
indicate any post peak drop as opposed to the clearly visible significant post peak drop in
counterpart relatively undisturbed specimen. Moreover, it is also interesting to note that the
reconstituted specimen at σ′vc of 100 kPa (ec = 1.3) did indicate a comparatively stronger nor-
malized shear resistance when compared with relatively undisturbed specimen at σ′vc of 400
kPa (ec = 1.14). A lower void ratio (denser particle arrangement) as noted for relatively undis-
turbed specimen at σ′vc of 400 kPa would generally result in a higher shear resistance,

342
Figure 6. Normalized shear stress-strain (I) and normalized stress-path (II) responses for constant-
volume monotonic DSS test on relatively undisturbed high plastic silt specimens from inland Fraser
River deposits at σ′vc = 100, 150, 200 and 400 kPa and reconstituted specimen at σ′vc = 100 kPa (after
Soysa & Wijewickreme 2015b).

however, the destructuration effects observed in the relatively undisturbed specimen at higher
confining stresses (σ′vc = 400 kPa) would cause a weaker shear resistance. At this stage, it is
important to highlight the importance of soil fabric in governing the seismic response of sensi-
tive fine-grained soils. The reconstituted specimen prepared using a slurry deposition tech-
nique has a completely different soil fabric when compared with relatively undisturbed
specimen. The noted observation provides good evidence to indicate the complex uncertainty
of dominance in void ratio and soil fabric in governing the monotonic shear loading response
of fine-grained soils.
The cyclic shear resistance derived for relatively undisturbed and reconstituted specimen at
σ′vc of 75 kPa is compared in Figure 7. A higher cyclic shear resistance could be noted for
relatively undisturbed specimens, indicating that the natural soil fabric from relatively undis-
turbed specimens (ec = 1.84 ~ 1.67) has a significant dominating influence over the reconsti-
tuted specimens with lower void ratios (ec = 1.41 ~ 1.21). It is also interesting to note that
cyclic shear resistance derived for reconstituted specimens at σ′vc of 75 kPa almost coincided
with that derived for relatively undisturbed specimens at σ′vc of 200 kPa. This observation
from Figure 7 implies that even after the occurrence of destructuration at σ′vc = 200 kPa for
relatively undisturbed specimens with comparatively greater void ratios (ec = 1.61~1.32), the
natural (but destructured) soil fabric seems to have the ability to produce similar cyclic shear

Figure 7. Cyclic stress ratio versus number of loading cycles for shear strain of 3.75% from constant-
volume cyclic DSS tests on relatively undisturbed high plastic silt specimens from inland Fraser River
deposits at σ′vc = 75, 150 and 200 kPa and reconstituted specimen at σ′vc = 75 kPa (after Soysa & Wije-
wickreme 2015c).

343
resistance of the reconstituted specimens which had completely different soil fabric due to
slurry deposition with comparatively lower void ratios (ec = 1.41~1.21).
In an overall sense, the above-observed differences in monotonic and cyclic loading charac-
teristics for the soils from inland Fraser River deposits when tested at higher confining stress
levels beyond the preconsolidation stress can be justifiably attributed to the breakdown of the
natural fabric/microstructure and loss of inter-particle bonds. Hence, it would be relevant to
highlight the importance of selecting the appropriate consolidation procedures/recompression
techniques, as developed by NGI (Bjerrum 1973), in the testing of such structured/sensitive
soils.

3.3 Effect of repeated cyclic loading (re-liquefaction)

The effects of repeated cyclic loading has received increased attention after several instances
of liquefaction at the same sites in Japan, Greece, United States of America and New Zealand
during successive earthquakes (Sims & Garvin 1995, Wakamatsu 2012, Quigley et al. 2013,
Papathanassiou et al. 2016). To illustrate the effect of repeated cyclic shear loading, reconsti-
tuted specimens prepared from non-plastic Fraser River silt obtained from an area east of the
Port Mann Bridge in British Columbia, Canada were subjected to multiple cyclic shear load-
ing phases.
The normalized shear stress-strain, normalized stress-path, pore water pressure develop-
ment, and shear strain accumulation observed by Soysa & Wijewickreme (2019) in initial and
repeated loading stages at σ′vc of 100 kPa during constant-volume cyclic DSS tests for recon-
stituted non plastic Fraser River silts are presented in Figure 8. The constant-volume cyclic
DSS loading was terminated when the shear strain of the specimen has reached a selected
threshold value of 3.75%. In Figure 9, the shear loading response of a similar reconstituted silt
specimen, where the constant-volume cyclic DSS loading was continued till the shear strain of
the specimen reached a selected threshold value of 10%, is presented. The intention herein is
to present the observed repeated shear loading response of reconstituted silt with respect to
initial shear loading, while focusing the effects of the magnitudes of the strain developed
during initial loading stages. Both these tests comprise several repeated cyclic shearing stages,
therefore associated number of loading cycles to reach certain shear strain thresholds (i.e.,
cyclic shear strain, γ = 3.75%, 5% and 10%) and excess pore water pressure thresholds (i.e.,
excess pore water pressure ratio, ru = 0.85 and 0.9) are listed in Table 2 - only for those stages
which have comparable cyclic stress ratios are presented.
From Figure 8, it can be noted that the reconstituted silt specimens under CSR of 0.11
underwent 21 loading cycles to reach shear strain of 3.75% during the initial cyclic loading
phase (Stage 1). In contrast, for repeated cyclic loading (Stage 2) under similar CSR of 0.11,
the specimen underwent 43 loading cycles before reaching a shear strain of 3.75%. The reduc-
tion of void ratio from 0.68 in shearing Stage 1 to 0.63 in the shearing Stage 2 due to post-
cyclic reconsolidation seems to be the plausible candidate for the above observed increase in
cyclic shear resistance in Stage 2 when compared with Stage 1. Similar increase in cyclic shear
resistance can be identified in Table 2 for shearing Stage 7 with respect to shearing Stage 6
under similar CSR of 0.23. In terms of excess pore water pressure ratio-based comparisons,
the increase in cyclic shear resistance for repeated loading stages is again notable when com-
pared with initial loading stages (see Table 2). It should be reminded that these observations
are for the specimens which have been sheared till the specimen reached a cyclic shear strain
of 3.75%, and none of these loading stages caused the specimen to reach pore water pressure
ratio beyond 0.95.
A: Cyclic shear loading is terminated when γ = 3.75%; B: Cyclic shear loading is terminated
when γ = 10%
Figure 9 compares the responses observed for reconstituted silt specimen during initial
(Stage 1) and repeated (Stage 2) cyclic loading phases that were terminated when the shear
strain reached up to 10%. The shearing Stage 2 (repeated) of the specimen, withstood 12 and
15 loading cycles to reach 3.75% and 10% shear strain respectively, (shown in Figure 9 and

344
Figure 8. Normalized shear stress-strain and normalized stress path, pore water pressure development
and shear strain accumulation of reconstituted Fraser River silt during initial (first – S1) and repeated
(second – S2) stages of constant-volume cyclic DSS tests with a CSR of 0.11 at σ′vc = 100 kPa (cyclic shear
loading termination criteria is achievement of 3.75% shear strain) (after Soysa & Wijewickreme 2019).

Figure 9. Normalized shear stress-strain and normalized stress path, pore water pressure development
and shear strain accumulation of reconstituted Fraser River silt during initial (first – S1) and repeated
(second – S2) stages of constant-volume cyclic DSS tests with a CSR of 0.11 at σ′vc = 100 kPa (cyclic
shear loading termination criteria is achievement of 10% shear strain

Table 2) while during initial loading stage, both 3.75% and 10% shear strain magnitudes were
attained at 18 loading cycles with a CSR of 0.11. A reduction in cyclic shear resistance in the
repeated (Stage 2) cyclic loading in comparison to that in the initial (Stage 1) cyclic loading
also can be seen from the lesser number of loading cycles for selected excess pore water pres-
sure ratios in repeated cyclic loading. This observed reduction of cyclic shear resistance in

345
Table 2. Summary of the repeated cyclic shear loading tests on reconstituted Fraser River silt.
Cyclic Ncyc Ncyc Ncyc Ncyc Ncyc
Test ID Stage CSR ec [γ=3.75%] [γ=5%] [γ=10%] [ru=0.85] [ru=0.9]

A S1-68-11 1 0.11 0.68 21 - - 21 21

S2-63-11 2 0.11 0.63 43 - - 41 43
S6-54-23 6 0.23 0.54 44 - - 26 36
S7-52-23 7 0.23 0.52 84 - - 45 60
B S1-76-11 1 0.11 0.76 18 18 18 17 18
S2-69-11 2 0.11 0.69 12 13 15 12 13
S4-59-18 4 0.18 0.59 3 5 13 5 7
S5-56-18 5 0.18 0.56 7 12 32 11 14

repeated (Stage 2) loading in comparison to that in initial (Stage 1) loading for the specimens
sheared up to 10% shear strain (Figure 9) is in contrast to the observed increase in cyclic shear
resistance during the repeated (Stage 2) loading of the specimen which was sheared up to only
3.75% shear strain shown in Figure 8. Contrary to the observed reduction of cyclic shear
resistance in shearing Stages 1 and 2 (with a CSR of 0.11) as in Figure 9, an increase in cyclic
shear resistance could be observed for the Stages 4 and 5 (with a CSR of 0.18).
These observations imply that the reductions of void ratios during post cyclic reconsolida-
tion in the specimens, that had been sheared up to terminated shear strain of 3.75%, effectively
displayed a consistent increase in cyclic shear resistance in repeated loading stages. Con-
versely, the competing influences arising from the changes in fabric and reduction of void
ratios, when the specimen was sheared up to terminated shear strain of 10% does not seem to
produce a consistent trend. A careful observation on the accumulation of shear strains in
repeated shearing stages revealed that shearing up to 3.75% shear strain generally result in
symmetric/unbiased shear strains whereas shearing up to 10% shear strain caused significantly
unsymmetrical/biased shear strains (as shown in Figure 10).

Figure 10. Comparison of normalized shear stress-strain and normalized stress path, pore water pres-
sure development and shear strain accumulation of reconstituted Fraser River silt during fourth stage of
repeated shearing in constant-volume cyclic DSS tests observed for specimens with cyclic shear loading
termination criteria of achievement of shear strain 3.75% versus 10%.

346
 The results of constant-volume cyclic DSS tests for two repeated shearing stages on rela-
tively undisturbed Fraser River silt specimens tested at different confining stresses by Sanin
(2005) reported that second cyclic shearing stages showed consistently lesser cyclic shear resist-
ance than those observed during the first cyclic shearing stages. The observed decrease in
cyclic shear resistance in the repeated cyclic shearing stages may be due to the destructuration
of soil fabric of relatively undisturbed specimen because of first shearing stages, despite the
void ratio reduction in second stage of shearing. The observations on the repeated shear load-
ing response of relatively undisturbed specimens reported by Sanin (2005) and of reconstituted
specimens presented above clearly indicate the significant influence of the soil fabric on the
cyclic shear resistance in comparison to void ratio. It seems that different magnitudes of the
shear strain accumulated during initial loading stage likely caused different soil fabric to
emerge, and in turn, the shear response during repeated loading stages were found to be differ-
ent. The observations noted from this study are in accord with the observations made by Finn
et al. (1970) from repeated cyclic loading of sand. They reported that if the cyclic shear strains
exceed a threshold value, it would further weaken the cyclic shear resistance of test specimens
against future cyclic loading; on the other hand, if the cyclic shear strains were lesser than that
threshold value, it would result in a better interlocked particle structure (which would elimin-
ate minor local instabilities) thereby causing an increase in cyclic shear resistance.

3.4 Effect of ageing

In soil mechanics and geotechnical engineering, the term ‘ageing’ has been widely used to
describe the changes in soil properties and states exclusively with respect to time. Both
improvements and degradation of soil properties have been noted with ageing (Troncoso et al.
1988, Schmertmann 1991), perhaps it may due to the combined and simultaneously associated
effects from chemical, mechanical and other aspects, other than the effects arising from time
alone. However, distinguishing and quantifying the isolated effects from time alone - pure
ageing- with respect to other simultaneously occurring processes is a tough challenge
(Schmertmann 1993); hence, quantifying the ageing process alone, remains a complex
phenomenon.
The effects of time of consolidation (‘ageing’ effects), on the cyclic shear resistance of recon-
stituted Fraser River silt specimen reported by Sanin (2010) are presented in Figure 11 and
Figure 12. Constant-volume cyclic DSS tests were performed on three reconstituted silt speci-
mens that had been left to consolidate for 3 hours, 24 hours and 7 days, respectively, prior to
the application of cyclic loading with a CSR of 0.12; the development of pore water pressure
and accumulation of shear strains (with increase number of loading cycles) are compared in
Figure 11.
As notable, the specimen that was consolidated under similar stress conditions (~100 kPa)
for a period of 7 days developed a lower rate of pore water pressure generation and shear
strain accumulation when compared with other specimens which were tested under similar
stress conditions and void ratios with different time of consolidation. In Figure 12, the cyclic
stress ratio versus number of loading cycles to attain 3.75% shear strain relationships derived
for the above-mentioned three differently aged reconstituted Fraser River silt specimens with
CSR of 0.12, are presented and compared with those response observed for reconstituted spe-
cimens (which were kept under consolidation stress for 3 hours in the DSS device) with differ-
ent CSR values. From a completeness viewpoint, similar relationship derived for relatively
undisturbed Fraser River specimens are also included in Figure 12. A clear increase in number
of loading cycles to attain shear strain of 3.75% (for a CSR of 0.12) with increasing time of
consolidation (ageing) implies that ageing process of reconstituted specimens assists in
improving the cyclic shear resistance. Using a simple linear extrapolation (based on the
derived three test results for 3 hours, 24 hours and 7 days), the number of cycles to reach
shear strain of 3.75% with a CSR of 0.12 for 10,000 years (estimated maximum age of the
deposit of Fraser River silt specimen), is approximately 30 cycles, which is still low in compari-
son to the curve obtained for the relatively undisturbed specimens.

347
Figure 11. Accumulation of shear strain (I) and development of pore water pressure (II) during con-
stant-volume cyclic DSS tests on reconstituted specimens having different times of ageing tested at σ′vc =
100, with CSR of 0.12 (after Sanin 2010).

The variation in the curves shown in Figure 12 for relatively undisturbed and reconstituted
Fraser River silts highlights the importance of soil fabric and aging in controlling the seismic
response of fine-grained soils.

3.5 Effect of disturbance during soil specimen preparation

Laboratory element testing assists in determining the mechanical loading response of soil by
mimicking the field loading conditions. However, the observed/recorded measurements
derived from a soil specimen during a laboratory testing when subjected to field loading con-
dition could be affected by the disturbances caused due to: (i) soil sampling; (ii) transportation
and handling of soil samples; (iii) changes in stress state, moisture content, temperature, and

Figure 12. Comparison of cyclic shear resistance of relatively undisturbed and reconstituted specimens
with different time of consolidation (after Sanin 2010).

348
soil composition by intrusion of foreign material; (iv) specimen preparation and placement of
specimens in the testing device; etc. The disturbances caused in the soil specimen due to above
mentioned factors could alter the soil particle structure/fabric and should be carefully acknow-
ledged when interpreting the soil response to the field conditions (Hight & Leroueil 2003,
Ladd & DeGroot 2003).
Significant advancements have been made to improve the quality of retrieved soil samples.
For example, the Sherbrooke sampler (Lefebvre & Poulin 1979) and Laval sampler (Rochelle
et al. 1981) are some of the notable advancements in improving the quality of the retrieved
fine-grained soil samples by minimizing the potential disturbance during sampling process as
opposed to the conventional Shelby tube sampling. Modifications to the tube sampler geom-
etry to reduce the cutting edge to be 5 degrees or less as suggested by Tanaka & Tanaka
(1999) has also contributed significantly in reducing the disturbances during soil sampling
process.
To minimize the soil disturbance during specimen preparation stage, many previous
researchers (Kimura & Saitoh 1982, Atkinson et al. 1992, Lo Presti et al. 1999) have under-
taken experimental studies and contributed towards developing the laboratory specimen prep-
aration procedures and techniques to minimize the disturbance. The disturbance due to the
potential deformations of relatively soft soil sample that takes place during the specimen
assembly – specifically immediately after the extrusion from the tube and just prior to the
placement and securing on the test device – where the soil specimen remains upright in a free-
standing position was investigated by Wijewickreme & Sanin (2006). They suggested the use
of an Undisturbed Soil Sample Holder (USSH) in order to address the observed tilting and
vertical deformation associated with lateral budging of the specimens during upright free-
standing period due to self-weight of the specimen. The experimental observations noted for
the responses of undisturbed soft to firm silt (PI = 4) specimens from channel-fill Fraser River
deltaic deposits in British Columbia, Canada prepared with and without the use of USSH is
compared in Figure 13 to illustrate the effectiveness of the USSH in minimizing the specimen
disturbance during the triaxial specimen preparation process.
With respect to the observed stress-strain response, specimen prepared without the use of
USSH exhibited a significantly higher mobilized shear strength than that observed for the spe-
cimen prepared with the use of USSH. Although, both specimens (with similar post consolida-
tion void ratios) exhibited contractive tendency in the initial stage of shearing, a notable
dilative tendency could be observed for specimen prepared without the use of USSH after
reaching the point of maximum contraction. The notable differences in the stress-strain, pore
water pressure, and stress-path responses shown in Figure 13, could be attributed to the differ-
ences in the particle structure/fabric during specimen preparation stage.

4 CONCLUDING REMARKS

The evidence from laboratory element tests studying the effects of: (i) reconstitution; (ii)
reconsolidation beyond the preconsolidation stress; (iii) repeated cyclic loading; (iv) ageing;
and (v) disturbance during specimen preparation show that the shear loading response behav-
ior of silts cannot be explained only with respect to the variations in commonly used variables
such as void ratio (e) and effective confining stress (σ'v). The difference in particle structure
(soil fabric) and microstructure (e.g., age of soil), which is not necessarily reflected by the
above two variables can be reasonably attributed to the observed deviations of shear loading
response. Cleary, the natural fabric and aging effects that would likely be prevalent in undis-
turbed silt deposits would not be present in the reconstituted specimens of the same silt pre-
pared from a slurry state.
The important role that soil particle fabric/microstructure plays in governing the response
of silt in addition to the traditionally considered effects from the void ratio and confining
stress is highlighted. The void ratio (density) and particle fabric seem to compete in governing
the mechanical performance (e.g., shear stiffness and strength), and it appears that the lack of

349
Figure 13. Mechanical response during consolidated-undrained triaxial compression testing–with and
without the use of USSH. (I) stress-strain response, (II) excess pore water pressure response, and (III)
stress path response (after Wijewickreme & Sanin 2006).

accounting for soil particle fabric can significantly impede the definition and the applicability
of critical state concepts.

ACKNOWLEDGEMENTS

The laboratory research program herein was funded by the Natural Sciences and Engineering
Research Council of Canada (NSERC) Discovery Grants Program (RGPIN 249603-12) and
NSERC – Accelerator Supplements Grant (RGPAS 429675-12). The undisturbed soil samples
were obtained as a result of a field-testing program supported by the British Columbia Geo-
logical Survey, the City of Richmond, the Joint Emergency Preparedness Program, ConeTec
Investigations Ltd. of Richmond, BC, and Mudbay Drilling Co Ltd. of Surrey, BC. The
authors would like to gratefully acknowledge Dr. Patrick Monahan for facilitating access to
the field sampling program as well as the support and discussions during this research. Major
contributions to the research in the form of meticulous laboratory research by previous gradu-
ate student, Maria Sanin, and the technical assistance of Messrs. Harald Schremp, Bill Leung,
Scott Jackson, and John Wong of the Department of Civil Engineering Workshop are deeply
appreciated.

REFERENCES

Amundsen, H.A., Thakur, V., & Emdal, A. 2016. Sample disturbances in block samples of low plastic
soft clays. In Proceedings of the 17th Nordic Geotechnical Meeting, Reykjavik, Iceland, 25-28 May.
Atkinson, J.H. & Bransby, P.L. 1978. The mechanics of soils: An introduction to critical state soil mechan-
ics. London and New York: McGraw-Hill Book Co.

350
Atkinson, J.H., Allman, M.A., & Böese, R.J. 1992. Influence of laboratory sample preparation proced-
ures on the strength and stiffness of intact Bothkennar soil recovered using the Laval sampler. Géo-
technique, 42(2): 349–354.
Bjerrum, L. & Landva, A. 1966. Direct simple-shear tests on a Norwegian Quick clay. Géotechnique, 16
(21): 1–20.
Bjerrum, L. 1973. Problems of soil mechanics and construction on soft clays. In Proceedings of the 8th
International Conference on Soil Mechanics and Foundation Engineering, Moscow, 3: 111–159.
Brewer, R. 1964. Fabric and mineral analysis of soils. John Wiley and Sons Inc, 129-158.
Burland, J.B. 1990. On the compressibility and shear strength of natural clays. Géotechnique, 40(3):
329–378.
Collins, K. & McGown, A. 1974. The form and function of microfabric features in a variety of natural
soils, Géotechnique, 24(2): 223–254.
Finn, W.D.L., Bransby, P.L. & Pickering, D.J. 1970. Effect of strain history on liquefaction of sand.
Journal of the Soil Mechanics and Foundations Division, ASCE, 96(6): 1917–1934.
Fleming, L.N. & Duncan, J.M. 1990. Stress-deformation characteristics of Alaskan silt. Journal of Geo-
technical Engineering, 116(3): 377–393.
Høeg, K., Dyvik, R., & Sandbækken, G. 2000. Strength of undisturbed versus reconstituted silt and silty
sand specimens. Journal of Geotechnical and Geoenvironmental Engineering, 126(7): 606–617.
Hight, D.W. & Leroueil, S. 2003. Characterisation of soils for engineering purposes. In Proceedings of the
Characterization and Engineering Properties of Natural Soils: International Workshop, Singapore, 2-4
December 2002, Singapore: 255–360.
Hofmann, B.A., Sego, D.C., & Robertson, P.K. 2000. In situ ground freezing to obtain undisturbed sam-
ples of loose sand. Journal of Geotechnical and Geoenvironmental Engineering, 126(11): 979–989.
Kimura, T. & Saitoh, K. 1982. The influence of disturbance due to sample preparation on the undrained
strength of saturated cohesive soil. Soils and Foundations, Japanese Society of Soil Mechanics and
Foundation Engineering, 22(4): 109–120.
Ladd, C.C. 1964. Stress-strain behavior of saturated clay and basic strength principles: Research in Earth
Physics - Research Report R64-17, Massachusetts Institute of Technology.
Ladd, R.S. 1974. Specimen preparation and liquefaction of sands. Journal of Geotechnical Engineering
Division, 100(10): 1180–1184.
Ladd, C.C. & DeGroot, D.J. 2003. Recommended practice for soft ground site characterization: Arthur
Casagrande Lecture Práctica. In 12th Panamerican Conference on Soil Mechanics and Geotechnical
Engineering, Boston, MA: 3–57.
Lefebvre, G. & Poulin, C. 1979. A new method of sampling in sensitive clay. Canadian Geotechnical Jour-
nal, 16(1): 226–233.
Leroueil, S., Roy, M., La Rochelle, P., Brucy, F. & Tavenas, F. 1979. Behaviour of destructured natural
clays. Journal of Geotechnical Engineering Division, 105(6): 759–778.
Leroueil, S. & Hight, D. 2003. Behaviour and properties of natural soils and soft rocks. In Proceedings of
the Characterisation and Engineering Properties of Natural Soils: International workshop, Singapore, 2-
4 December 2002: 29–254.
Long, M. 2003, Sampling disturbance effects in soft laminated clays, Proceedings of the Institute of Civil
Engineers - Geotechnical Engineering, 156(4): 213–224.
Long, M., Vaid, Y.P. & Sivathayalan, S. 2001. Strength of undisturbed versus reconstituted silt and silty
sand specimens. discussion and closure. Journal of Geotechnical and Geoenvironmental Engineering,
127(11): 991–994.
Lo Presti, D.C., Jamiolkowski, M., Pallara, O. & Tordella, M.L. 1999. Assessment of sample disturbance
in the laboratory. In Pre-failure Deformation Characteristics of Geomaterials: Proceedings of the
Second International Symposium on Pre-Failure Deformation Characteristics of Geomaterials, Torino,
Italy: 11–18.
Miura, S. & Toki, S. 1982. A sample preparation method and its effect on static and cyclic deformation-
strength properties of sand. Soils and foundations, Japanese Society of Soil Mechanics and Foundation
Engineering, 22(1): 61–77.
Oda, M. 1972. The Mechanism of fabric changes during compressional deformation of sand. Soils and
Foundations, Japanese Society of Soil Mechanics and Foundation Engineering, 12(2):1–8.
Papathanassiou, G., Ganas, A. & Valkaniotis, S. 2016. Recurrent liquefaction-induced failures triggered
by 2014 Cephalonia, Greece earthquakes: Spatial distribution and quantitative analysis of liquefaction
potential. Engineering Geology, 200: 18–30.
Quigley, M. C., Bastin, S. & Bradley, B.A. 2013. Recurrent liquefaction in Christchurch, New Zealand,
during the Canterbury earthquake sequence. Geology, 41(4): 419–422.

351
Rochelle, P.L., Sarrailh, J., Tavenas, F., Roy, M. & Leroueil, S. 1981. Causes of sampling disturbance
and design of a new sampler for sensitive soils. Canadian Geotechnical Journal, 18(1): 52–66.
Santagata, M.C. & Germaine, J.T. 2002. Sampling disturbance effects in normally consolidated clays.
Journal of Geotechnical and Geoenvironmental Engineering, 128(12): 997–1006.
Sanin, M.V. 2005. Cyclic Shear Loading Response of Fraser River Delta Silt. M.ASc. Thesis, Department
of Civil Engineering, The University of British Columbia, Canada.
Sanin, M.V. 2010. Cyclic Shear Loading Response of Fraser River Delta Silt. PhD. Thesis, Department of
Civil Engineering, The University of British Columbia, Canada.
Schmertmann, J.H. 1991. The mechanical aging of soils. Journal of Geotechnical Engineering, 117
(9):1288–1330.
Schmertmann, J.H. 1993. Update on the mechanical aging of soils (25th Terzaghi Lecture). In symposium
“Sobre Envejecimiento de Suelos” The Mexican Society of Soil Mechanics, Mexico City.
Silver, M.L. & Seed, H.B. 1971. Deformation characteristics of sands under cyclic loading. Journal of the
Soil Mechanics and Foundation Division ASCE, 97(8): 1081–1098.
Sims, J.D. & Garvin, C.D. 1995. Recurrent liquefaction induced by the 1989 Loma Prieta earthquake
and 1990 and 1991 aftershocks: Implications for paleoseismicity studies. Bulletin of the Seismological
Society of America, 85(1): 51–65.
Soysa, A.N. 2015. Monotonic and cyclic shear loading response of natural silts. M.ASc. Thesis, Depart-
ment of Civil Engineering, University of British Columbia, Canada.
Soysa, A. & Wijewickreme, D. 2015a. Simple shear loading response of undisturbed and reconstituted
silt. In proceedings of the 68th Canadian Geotechnical Conference, Québec City, Québec, Canada: 8pp.
Soysa, A. & Wijewickreme, D. 2015b. Some observations on the mechanical behavior of natural fine-
grained soil from laboratory direct simple shear testing. In Proceedings of International Conference on
Geotechnical Engineering, Colombo, Sri Lanka: 173–180.
Soysa, A. & Wijewickreme, D. 2015c. Cyclic shear loading response of relatively high-plastic natural fine-
grained soil from the Fraser River Delta. In Proceedings of 11th Canadian Conference on Earthquake
Engineering. Victoria, BC, Canada: 9pp.
Soysa, A. & Wijewickreme, D. 2019. Repeated cyclic shear loading response of reconstituted Fraser
River silt. In Proceedings of the 7th International Conference on Earthquake Geotechnical Engineering,
Rome, Italy: 8pp.
Tanaka, H. & Tanaka, M. 1999. Key factors governing sample quality. In Characterization of soft marine
clays - Proceedings of International Symposium on Coastal Engineering in Practice: 55–82.
Troncoso, J., Ishihara, K. & Verdugo, R. 1988. Aging effects on cyclic shear strength of tailings mater-
ials. In Proceedings of the 9th World Conference on Earthquake Engineering, Kyoto, Tokyo: 121–126.
Vaid, Y.P., Sivathayalan, S. & Stedman, D. 1999. Influence of specimen-reconstituting method on the
undrained response of sand. Geotechnical Testing Journal, 22(3),187–195.
Verma, P. & Wijewickreme, D. 2016. Some observations on the effect of cyclic shear loading polarity on
reconstituted natural silt with initial static shear bias. In Proceedings of the 69th Canadian Geotechnical
Conference. Vancouver, BC, Canada: 8pp.
Wakamatsu, K. 2012. Recurrence of liquefaction at the same site induced by the 2011 Great East Japan
Earthquake compared with previous earthquakes. In Proceedings of the 15th World Conference on
Earthquake Engineering. Lisbon, Portugal: 10pp.
Wijewickreme, D. & Sanin, M.V. 2006. New sample holder for the preparation of undisturbed fine-
grained soil specimens for laboratory element testing. Geotechnical Testing Journal, 29(3),242–249.
Wijewickreme, D. & Sanin, M.V. 2008. Cyclic shear response of undisturbed and reconstituted low-plastic
Fraser River silt. In proceedings of the Geotechnical Earthquake Engineering and Soil Dynamics IV,
American Society of Civil Engineers, Sacramento, California: 10pp.
Yoshimi, Y., Hatanaka, M. & Oh-Oka, H. 1978. Undisturbed sampling of saturated sands by freezing.
Soils and Foundations, Japanese Society of Soil Mechanics and Foundation Engineering,18(3): 59–73.
Zapata-Medina, D.G., Finno, R.J. & Vega-Posada, C.A. 2014. Stress history and sampling disturbance
effects on monotonic and cyclic responses of overconsolidated Bootlegger Cove clays. Canadian Geo-
technical Journal, 51(6): 599–609.

352