Landfill Design
N.Dixon, R.V.Jones, R.G.Gregory & D.Hall Hb: 0–415–37049–3
Earthquake Engineering: From Engineering Seismology to Performance-Based Engineering
Ed. V.V.Bertero & Y.Bozorgnia Hb: 0–8493–1439–9
Earthquake Engineering for Structural Design
Ed. W.-F.Chen & E.M.Lui Hb: 0–8493–7234–8
Physics and Mechanics of Soil Liquefaction
Ed. P.V.Lade & J.A.Yamamuro Hb: 90–5809–038–8 Information and ordering details
For price availability and ordering visit our website www.tandf.co.uk/builtenvironment Alternatively our books are available from all good bookshops.
Soil Liquefaction
A critical state approach
Mike Jefferies & Ken Been
4RN Simultaneously published in the USA and Canada by Taylor & Francis
270 Madison Ave, New York, NY 10016, USA
Taylor & Francis is an imprint of the Taylor & Francis Group, an informa business
This edition published in the Taylor & Francis e-Library, 2006. “ To purchase your own copy of this or any of Taylor & Francis or Routledge’s
collection of thousands of eBooks please go to http://www.ebookstore.tandf.co.uk/.” © 2006 Mike Jefferies & Ken Been
Publisher’s Note
This book has been prepared from camera-ready copy supplied by the authors.
All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission
in writing from the publishers.
The publisher makes no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility or liability for any
efforts or omissions that may be made. British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data Jefferies, Mike, 1952– Soil liquefaction : a critical state approach/Mike Jefferies & Ken Been. p. cm. Includes bibliographical references and index. ISBN 0-419-16170-8 (hardback: alk. paper) 1. Soil mechanics. 2. Soil liquefaction. I. Been,
Ken. II. Title. TA710.J365 2006 624.1′5136–dc22 2006002052 ISBN 0-203-30196-X Master e-book ISBN
ISBN10: 0-419-16170-8 (hbk) ISBN10: 0-203-30196-X (Print Edition) (ebk)
ISBN13: 978-0-419-16170-7 (hbk) ISBN13: 978-0-203-30196-8 (Print Edition) (ebk) 2 Park Square, Milton Park, Abingdon, Oxon OX14
Contents
LIST OF TABLES vii
LIST OF FIGURES x
SYMBOLS AND NOTATION xxxii
PREFACE xxxix
DISCLAIMER AND CAUTION
xliv
1INTRODUCTION 1
2DILATANCY AND THE STATE PARAMETER 41 3CONSTITUTIVE MODELLING FOR LIQUEFACTION 102 4DETERMINING STATE PARAMETER IN SITU 161 5SOIL VARIABILITY AND CHARACTERISTIC STATES 229 6STATIC LIQUEFACTION AND POST-LIQUEFACTION
STRENGTH
254 7CYCLIC STRESS INDUCED LIQUEFACTION 340
8CONCLUDING REMARKS 405
APPENDIX ASTRESS AND STRAIN MEASURES 412
APPENDIX BLABORATORY TESTING TO DETERMINE THE CRITICAL STATE OF SANDS
419
APPENDIX CTHE CRITICAL FRICION RATIO M 437
APPENDIX DNORSAND DERIVATIONS 445
FLOW FAILURE
REFERENCES 560
Tables
2.1 Critical state properties for some soils
54
2.2 Clarification of terminology for describing soils
71
2.3 Laboratory test for design parameters in sands and silts
72
2.4 Summary of proposed relationships for M
i89
3.1 Summary of NorSand
136
3.2 NorSand parameters and typical values for sands
137
3.3
Triaxial tests on Erksak 330/0.7 sand to determine CSL and
NorSand parameters (Been et al, 1990)
138
3.4
NorSand parameters for Erksak 330/0.7 drained triaxial
calibration
146
3.5 Paired tests on Brasted sand (data from Cornforth, 1961)
156
4.1 Dimensionless parameter groupings for CPT interpretation
162
4.2 Summary of CPT calibration chamber studies
178
4.3 Approximate expressions for general inversion form ψ=f(Q
p)
198
4.4 Relationship of soil type to soil classification index I
c206
4.5
Summary of near-undisturbed SBP tests in Tarsiut P-45
hydraulically placed sand fill and adjacent CPT data
(Rowe and Craig, 1976)
5.2 Cyclic loading stages in caisson models (Rowe and Craig, 1976) 231
5.3
Model time per cycle and time factors for centrifuge models
(Rowe and Craig, 1976)
232
5.4
Resistance factors for characteristic strength percentiles for an
offshore structure example (Been and Jefferies, 1993)
253
6.1
Observed values of the parameters (su/p′)n, and I
Bfor
consolidated undrained tests on cohesionless soils (Bishop, 1971)
266
6.2
Summary of steady state strength determinations from laboratory
tests for Lower San Fernando Dam (after Marcuson et al., 1990)
304
6.3
Some important case histories giving insight to full-scale
post-liquefaction strength
314
6.4a
Comparison of post-liquefaction residual strength s
r(psf) from
back-analysis of failure as reported by various investigators
316
6.4b
Comparison of corresponding characteristic normalized SPT
blowcount (N
1)
60suggested by the investigators
316
6.5 Fines content adjustment factors for SPT (after Seed, 1987)
317
6.6
Summary of case history data for mobilized post-liquefaction
strength
323
7.1
Proposed factors for difference between cyclic simple shear and
triaxial testing
362
E.1 Boundary Condition Codes, after Parkin et al. (1980)
499
F.1
Summary of strengths and strength ratios determined by Olson et
al. (2000)
524
F.2 Summary of index and critical state properties for Nerlerk Sands 542
F.3
Summary of shear strengths from back analysis of La Marquesa
Dam
1.1 Definition of state parameter ψ
5
1.2
Aerial view of Fort Peck failure (U.S Army Corps of Engineers,
1939)
7
1.3
Nerlerk B-67 berm and foundation cross section (Been et al.,
1987)
8
1.4
Plan of failures that occurred at Nerlerk B-67 and cross section
through Slide3 (Sladen et al., 1985a)
9
1.5
Grain size distribution information for Nerlerk B-67 materials
(Sladen et al., 1985a)
10
1.6
Typical Nerlerk berm CPT (CPTC12 in 1988) including clay
layer between sand fill and the seabed
11
1.7
Summary of CPT distributions in Nerlerk B-67 berm, in Nerlerk
sand and Ukalerk sand
12
1.8
Apartment building at Kawagishi-cho that rotated and settled
because of foundation liquefaction in 1964 Niigata earthquake
(from Karl V Steinbrugge Collection, Earthquake Engineering
Research Center)
14
1.9
Sketch plan of Niigata, showing main area of damage.
Kawagishi-cho and South Bank sites marked with X (Ishihara,
1993)
15
1.10
Soil profile and CPT resistance at Kawagisho-cho site (Ishihara
and Koga, 1981)
15
1.11
Soil profile and CPT resistance at South Bank site (Ishihara and
Koga, 1981)
16
1.12 Seed Liquefaction Assessment Chart (Seed et al, 1983)
17
1.13 Liquefaction failure of Lower San Fernando Dam after the 1971
top photograph)
1.14 Possible failure mechanism for Aberfan Tip No 7 (Bishop, 1973) 19
1.15
Aberfan flow slide shortly after the failure. Flowslide distance
added by authors
20
1.16
Aerial view of the Merriespruit tailings dam failure showing the
path of the mudflow that occurred (Fourie et al., 2001)
22
1.17
Sequence of retrogressive failures of Merriespruit containment
postulated by Wagener et al, 1998 (Fourie et al, 2001)
23
1.18
Grain size distribution and Critical State Line of Merriespruit
tailings materials (Fourie and Papageorgiou, 2001)
24
1.19
Distrubution of in situ void ratios obtained during post failure
investigation of Merriespruit tailings dam (Fourie et al., 2001)
25
1.20 Gulf Canada’s Molikpaq structure in the Beaufort Sea
26
1.21
Details of cyclic ice loading and excess pore pressure 12 April
1986
27
1.22
Piezometric response showing accumulating excess pore
pressure to liquefaction (piezometer E1, mid depth in centre of
loaded side)
28
1.23
Failure of embankment on Ackermann Lake triggered by
vibroseis trucks (Hryciw et al., 1990)
31
1.24
Plan and cross section of the Wildlife instrumentation array
(from Youd and Holtzer, 1994 based on Bennett et al., 1984)
32
1.25
Surface accelerometer (N-S) and piezometer P5 (2.9 m) at
Wildlife site during Superstition Hills 1987 Earthquake (from
Youd and Holtzer, 1994)
34
1.26
Shear stress and shear strain history at depth of piezometer P5 at
Wildlife Site, interpreted from accelerometers by Zeghal and
Elgemal (1994)
increments interpreted from NS accelerometers at Wildlife site
(after Youd and Holtzer, 1994)
2.1 Difference between rate and absolute definitions of dilatancy
42
2.2
Early hypothesis of critical void ratio from direct shear tests
(Casagrande, 1975)
44
2.3
Comparison of behaviour of sand as a function of relative
density and state parameter for Kogyuk 350/2 and Kogyuk
350/10 sands
50
2.4
Idealized state path to illustrate relationship of dilatancy to state
parameter
52
2.5
Peak dilatancy of twenty soils in standard drained triaxial
compression
58
2.6
Stress-dilatancy component of peak strength of twenty soils in
standard drained triaxial compression
58
2.7
Volumetric strain at peak stress for drained triaxial compression
tests on 20 sands
59
2.8
Friction angle versus state parameter normalized by range of
accessible void ratios (e
max− e
min). Note the lack of improvement
over Figure 2.6
60
2.9
Maximum dilatancy as a function of ψ/λ. There is no
improvement to the correlation compared to y alone (See Figure
2.5)
60
2.10
Maximum dilation as a function of state parameter normalized
by (1+e). There is a small improvement compared to state
parameter alone (compare with Figure 2.5)
61
2.11
Effect of sample preparation on the behaviour of Kogyuk sand
(Been and Jefferies, 1985)
62
Toyoura sand (after Tatsuoka, 1987)
2.13
Effect of fabric on friction angle of sands reported by Tatsuoka
and by Oda and compared to general correlation of friction angle
to state parameter
64
2.14
Comparison of the effect of void ratio and sample preparation
method on the cyclic strength of two sands in simple shear
(Nemat-Nasser and Tobita, 1982)
65
2.15
Influence of overconsolidation ratio on the friction angle of
Erksak 330/0.7 sand
65
2.16 Effect of sample size on the behaviour of dense Ticino sand
67
2.17
Multiple shear bands evident through membrane in large
(300mm diameter) sample after drained shearing
68
2.18
Schematic illustration of relationship between parameters and
testing methods
70
2.19
Stress controlled CIU triaxial test during which a critical (steady)
state is clearly reached
75
2.20
CIU triaxial test showing dilation at large strains and the “quasi
steady state”. Incorrectly treating the quasi-steady state as the
critical state leads to “non-unique critical states” and other errors
76
2.21
Selection of undrained tests to used to give critical state line in
Figure 2.22
77
2.22
Critical state line for Erksak 330/0.7 sand from undrained tests
that reached a distinct critical (steady) state
78
2.23
Examples of drained triaxial tests on loose samples reaching
critical state
79
2.24
Critical state line for Guindon Tailings B (67% fines) showing
use of drained tests on loose samples to define critical state at
higher stresses.
80
dilatancy for Erksak and Brasted sands under different loading
conditions (Jefferies and Shuttle, 2002)
2.27
Drained triaxial data for Erksak sand reduced to stress dilatancy
form (Been Jefferies, 2004)
87
2.28
Relationship of mobilized friction ratio M
fto ψ for Erksak data.
Dense sand data at initial D
p= 0 shown as filled squares; loose
sand data shown as traces for complete strain path. Also shown
are several proposed constitutive model relationships (Been and
Jefferies, 2004)
88
2.29
Effect of sample preparation on undrained behaviour of Erksak
330/0.7 sand
91
2.30
Comparison of critical states from pluviated and moist
compacted samples of Erksak 330/0.7 sand (data from Been et
al, 1991). Note that pluviated samples cannot be prepared at high
void ratios
92
2.31
Peak dilation rate in drained triaxial compression tests as a
function of distance from critical state line determined from
undrained tests. The trend line passes close to zero, indicating
that drained and undrained behaviour relate to the same CSL
92
2.32 Stress conditions in the simple shear test
94
2.33
Undrained simple shear tests on Fraser River sand (Vaid and
Sivathayalan, 1996)
95
2.34
Comparison of triaxial compression, extension and simple shear
behaviour of Fraser River Sand (Vaid and Sivathayalan, 1996)
96
2.35
Critical state loci for several sands whose properties are given in
Table 2.1
98
2.36
Relationship between slope of the critical state line and fines
content; uniformly graded soils
99
(Γ
1) and maximum void ratio (e
max); uniformly graded soils
2.38
Comparison of critical state lines for uniformly graded and well
graded silty sands
100
3.1 Illustration of normality through hockey puck analogy
106
3.2 Definition of normality (associated plastic flow)
106
3.3 Dilation implied by normality to Mohr Coulomb surface
107
3.4
Correct association of yield surface with soil strength, from
Drucker, Gibson and Henkel (1957)
108
3.5 Comparison of isotropic compression idealizations
109
3.6
Separation of state parameter from overconsolidation ratio
(Jefferies and Shuttle, 2002)
110
3.7
Example of variation of critical friction ratio M with Lode angle
θ (Jefferies and Shuttle, 2002)
114
3.8 Illustration of the consistency condition
122
3.9
Implied overconsolidation for a given state ψ in Cambridge
models
124
3.10 Illustration of the Hvorslev surface idealization
125
3.11
Distribution of fill density in normally consolidated hydraulic
sand fill (Stewart et al., 1983)
126
3.12
Experimental evidence for an infinity of NCL (isotropic
consolidation of Erksak 330/0.7 sand)
128
3.13
Illustration of NorSand yield surface, limiting stress ratios and
image condition
132
3.14 Dilatancy as a function of state parameter at image condition
134
3.15
Measured bulk modulus of Erksak sand in isotropic
unload-reload tests (Jefferies and Been, 2000)
3.17
Examples of calibrated fit of NorSand to Erksak 330/0.7 sand in
drained triaxial compression
145
3.18
Plastic hardening modulus versus state parameter ψ
0for Erksak
sand (Ticino and Brasted sand shown for comparison)
147
3.19
Effect of NorSand model parameters on drained triaxial
compression behaviour
148
3.20
Example of experimentally determined yield surfaces in Fuji
River sand (Tatsuoka and Ishihara, 1974)
151
3.21
NorSand yield surfaces for comparison with experimental results
on Fuji River sand
151
3.22 Isotropic plastic compression behaviour of NorSand
154
3.23 Failure of sample in plane strain test carried out by Cornforth
156
3.24
Peak dilatancy of Brasted sand in triaxial compression versus
state (from Jefferies and Shuttle, 2002)
158
3.25 Calibration of NorSand to Brasted sand in triaxial compression 159
3.26
Validation of NorSand in plane strain by comparison of
predictions versus data for Brasted sand (Jefferies and Shuttle,
2002)
159
4.1 Comparison of SPT and CPT repeatability
165
4.2
Illustration of soil type classification chart using CPT data
(Robertson, 1990)
166
4.3
Relation between q
c/N and soil type (Burland and Burbidge,
1985)
167
4.4
Example stress-strain behaviour of NAMC material in triaxial
compression (properties for medium dense sand)
171
NAMC material with Bolton’s s approximation of
stress-dilatancy
4.6
Spherical cavity limit pressure ratio versus state parameter
(broken lines indicate linear approximation of equation [4.7])
175
4.7
Comparison of experimental spherical cavity limit pressure with
penetration resistance of blunt indenter (after Ladanyi and Roy,
1987)
175
4.8 Example of CPT calibration chamber (Been et al., 1987b)
176
4.9
Example of CPT chamber test data (Erksak sand, from Been et
al, 1987b)
177
4.10
Grain size distribution curves for sands tested in calibration
chambers
179
4.11
CPT resistance versus relative density for three sands (after
Robertson and Campanella, 1983)
180
4.12
CPT resistance calibration for Monterey N° 0 sand (test data
from Villet, 1981; graph from Been et al, 1986)
181
4.13
Effect of stress on penetration resistance in normally
consolidated sand (a) vertical stress; (b) horizontal stress.
(Clayton et al., 1985)
182
4.14
Experimentally determined C
Nfunctions for Reid Bedford and
Ottawa sand by Marcuson and Bieganowski (1977) and
recommended C
Nfunction by Liao and Whitman (1986)
185
4.15
Dimensionless CPT resistance versus state parameter for
Monterey sand (data from Fig 4.12, after Been et al., 1986)
187
4.16
Normalized Qp−ψ relationships from calibration chamber
studies (NC=normally consolidated; OC=over-consolidated)
188
4.17
Normalized CPT resistance of normally consolidated and
overcon-solidated Ticino Sand
190
4.20
Summary of stress level bias in Q
p–ψ relationship for Ticino
sand as suggested by Sladen (1989a,b)
193
4.21
Numerical calculation of Q
p–ψ relationship for Ticino sand,
showing linearity and effect of elastic modulus as cause of stress
level bias (Shuttle and Jefferies, 1998)
194
4.22
Computed effect of I
ron k,m coefficients for Ticino sand
(Shuttle and Jefferies, 1998)
195
4.23
Shear modulus of Ticino sand versus confining stress: p
ris a
reference stress level, here taken as 100 kPa (Shuttle and
Jefferies, 1998)
196
4.24
Computed Q
p–ψ relationship for Ticino sand, shown as
trendlines, compared to individual calibration chamber tests
197
4.25
Effect of soil properties on spherical cavity expansion pressure
ratio (Shuttle and Jefferies, 1998)
199
4.26
Performance of approximate general inversion on 10 sands with
randomly chosen properties (Shuttle and Jefferies, 1998)
200
4.27
Trends in effective inversion parameters k′, m′ with soil
compressibility λ
10203
4.28
Relationship between λ
10and F suggested by Plewes, Davies and
Jefferies (1992)
204
4.29
Suggested relationship between λ and I
c(adapted from Been and
Jefferies, 1992)
205
4.30 Soil type classification chart showing constant I
ccontours
206
4.31
Shear modulus determined from VSP tests in hydraulically
placed sandfill (Molikpaq core at Tarsiut P-45)
213
4.32 Comparison of I
rbetween silts and sands
214
4.33 Results of SBP tests in hydraulically placed Erksak sand
217
4.34
Horizontal geostatic stress in hydraulic fills (Graham and
Jefferies, 1986)
219
4.35
CPT horizontal stress amplification factor versus state (Jefferies
Jönsson and Been, 1987)
221
4.36
Comparison of geostatic stress from SBP and CPT in
hydraulically placed sandfill (Jefferies, Jönsson and Been, 1987)
222
4.37
Effect of uncertainty in horizontal stress on uncertainty in
estimated in situ state parameter from CPT data (Jefferies,
Jönsson and Been, 1987)
224
5.1
Measured response of caissons subject to increasing stages of
cyclic loading in centrifuge test (from Rowe and Craig, 1976)
230
5.2 Layout of loose pockets below caissons (Rowe and Craig, 1976) 233
5.3
Scaled displacements and pore pressures observed in model with
4% loose zones in fill (Rowe and Craig, 1976)
235
5.4
Scaled displacement and piezometric data for centrifuge model
with 10% loose zones in fill (Rowe and Craig, 1976)
236
5.5
Schematic cross section of the Molikpaq at Tarsiut P-45 showing
locations of CPTs to determine fill properties (adapted from
Jefferies et al., 1985 by Popescu et al., 1997)
238
5.6
Examples of CPTs in Tarsiut P-45 fill These CPTs are spaced
about 9m apart. (see Figure 5.5) MAC 05 & 32 and MAC 08 &
33 are spaced 1m apart to demonstrate repeatability of
measurements. (adapted from Jefferies et al., 1985 and Popescu
et al, 1997)
239
5.7
Selected Tarsiut P-45 CPTs plotted against depth with average
trends in core and berm fill shown. Inset histograms are
distributions of q
cvalues in 1m depth intervals at depths of 5 m,
15 m and 25 m
239
5.8
Statistical profile of normalized penetration resistance Q and
state parameter at Tarsiut P-45
5.10
Distribution of fines content measured in Tarsiut P-45 fill
(Jefferies et al, 1988)
243
5.11 Liquefaction of variable fill computed by Popescu 1995
244
5.12
Comparison of uniform and variable fill results in Popescu et al.,
1997
245
5.13
Summary of CPT statistics in the area of Slide 3 of Nerlerk
berm. State parameter interpretation using variable shear
modulus methodology of Chapter 4
246
5.14
Distribution of random ψ field mapped onto Nerlerk berm
geometry, computed by Onisiphorou (2000)
248
5.15
Results of Nerlerk Berm analysis with uniform fill states
(Onisiphorou, 2000)
249
5.16
Results of analysis of Nerlerk berm with variable field ψ:
µ=−0.08, σ=0.05 (Onisiphorou, 2000)
250
6.1 Undrained triaxial compression of Erksak 330/0.7 sand
257
6.2 Loose Ticino sand in undrained triaxial compression
258
6.3 Particle size distribution curve for four liquefying soils
259
6.4
Loose silty sand (Bennett Dam) and sandy silt (Guindon
Tailings) in undrained triaxial compression
259
6.5 Triaxial extension test data for Erksak 300/0.7 sand
260
6.6
Comparison of extension and compression tests on Erksak sand
(normalized)
261
6.7 Effect of stress path on the critical state locus (at expanded scale) 262
6.8
Comparison of Bonnie Silt in simple shear, triaxial compression
and triaxial extension (all tests at initial confining stress of
80kPa, 0.683<e
0<0.753)
6.9 Normalized undrained strength of loose liquefiable sands
267
6.10
Undrained strength ratio of normally consolidated clay (equation
in Wroth, 1984)
267
6.11 Pore pressure ratio A
fof loose sands at peak strength
269
6.12 Comparison of normalized peak and residual undrained strengths 270
6.13
Brittleness index of sands with a range of λ (Filled symbols
represent low λ and open symbols higher λ values)
271
6.14
Comparison of collapse surface and instability or flow
liquefaction line representations for onset of liquefaction (after
Yang, 2002)
272
6.15
Erksak sand test G609 illustrating the nature of static
liquefaction and collapse surface at η
I.
273
6.16
Very loose Erksak drained test D684 showing intersection of
stress path with “collapse surface”
275
6.17
Initial state diagram for the series of triaxial tests on Erksak
330/0.7 sand
279
6.18
Triaxial compression static liquefaction—NorSand compared to
Erksak sand data
280
6.19
NorSand simulations showing effect of elastic modulus on
undrained behaviour
282
6.20
NorSand simulations showing effect of plastic modulus on
undrained behaviour
283
6.21
Peak undrained triaxial compression strength for liquefying
sands with trends from NorSand
284
6.22
S and F lines from NorSand simulations of triaxial compression
tests (Jefferies and Been, 1992)
285
6.23
Simulations showing modeling of sample preparation effects (for
both samples Γ=0.816, λ
10=0.031, v=0.2)
Impoundment
6.25 Simulation of triaxial extension, Erksak 330 test CIUE-G642
290
6.26
Failure in declining mean stress at constant shear experiment of
Sasistharan et al., (1993)
291
6.27
NorSand simulation for declining mean stress drained brittle
failure in experiments by Sasitharan et al., (1993)
292
6.28 NorSand simulations of pseudo steady state condition
293
6.29
Comparison of NorSand with measured undrained silt behaviour
in simple shear
295
6.30
Example of effect of initial geostatic stress ratio K
0on undrained
strength in simple shear
295
6.31
Effect of initial state on peak undrained strength in simple shear
versus triaxial conditions
297
6.32
Computed peak undrained strength in simple shear versus
triaxial conditions when normalized by the initial vertical
effective stress
298
6.33
Peak undrained strength of normally consolidated clay in
different strain conditions (based on Figs 7 and 27 of Wroth,
1984)
299
6.34
Steady State School method to determine steady state strength of
soil at in situ void ratio (after Poulos et al., 1985)
302
6.35
Cross sections of Lower San Fernando dam (from Seed et al.,
1988)
304
6.36
Adjustments of measured undrained steady state strengths to in
situ conditions at Lower San Fernando dam (from Seed et al.,
1988)
305
6.37 Comparison of in situ void ratios and remoulded SSL for Lower
Sand Fernando dam determined by Vasquez-Herrera and Dobry
305
(1988)
6.38
X-ray images of shear bands in triaxial samples (Oda and
Kazama, 1998)
308
6.39
Proposed but incorrect correlation between steady state
(residual) strength and adjusted SPT penetration resistance (after
Seed and Harder, 1990)
313
6.40
Alternative correlation between steady state (residual) strength
and adjusted SPT penetration resistance proposed by Stark and
Mesri (1992)
318
6.41
Illustration of the importance to practical engineering of the
difference between the two correlations proposed for s
rfrom
SPT data
319
6.42
Residual undrained strength ratio versus penetration resistance
from liquefaction case histories
324
6.43
Relationship between initial in situ state parameter and
mobilized steady state strength from case history data
326
6.44
Lower San Fernando dam showing as-constructed section
(above) and section during the 1985 investigation (below) after
Castro et al. (1989)
328
6.45
Plan of Lower San Fernando dam in 1985 and showing location
of investigation borings/soundings (after Castro et al., 1989)
329
6.46
Cross section through Lower San Fenando dam at St. 09+35
(approximately centerline of sliding mass) showing inferred
zonation of dam from 1985 study
330
6.47
Comparison of CPT and SPT resistances at Lower San Fernando
Dam (from Castro et al., 1989)
331
6.48 CPT C103 from Lower San Fernando dam investigation in 1985 331
6.49
Particle size distribution of soils within Zone 5 of Lower San
Fernando Dam (“Batch Mix No 7” was used for steady state
triaxial tests, after Castro et al., 1989)
different depths because of different collar elevations)
6.51
CPT C104 from Lower San Fernando dam investigation in 1985,
including screening level interpretation of ψ
334
6.52
Computed residual strength ratio S
r/σ′v
0in hydraulic fill at
Lower San Fernando Dam
335
6.53
Boundary between satisfactory and unsatisfactory undrained
performance of sands in terms of CPT penetration resistance.
These curves are illustrative and must be computed for any
specific soil
338
7.1 Void ratio reduction induced by cyclic shear (after Youd, 1972) 342
7.2 Schematic illustration of the different forms of cyclic loading
343
7.3
Stress conditions in triaxial, simple shear and hollow cylinder
tests
346
7.4
The hollow cylinder test apparatus at University of British
Columbia (Vaid et al., 1990)
348
7.5
Example of sand behaviour in undrained cyclic triaxial test
(Nevada sand, Arulmoli et al, 1992).
350
7.6
Cyclic strength of Toyoura Sand in triaxial compression (data
from Toki et al 1986)
352
7.7
Cyclic triaxial test data on 13 sands for which CSL is known,
with ranges obtained by Garga and McKay (1984) for sands and
tailings sands shown
353
7.8
Cyclic triaxial test data from 7.7 normalized to cyclic resistance
ratio for 15 cycles, CRR15
354
7.9
Cyclic triaxial data normalized to consolidation stress ratio, K
c,
illustrating absence of trend as a function of K
c355
parameter for 13 sands
7.11
Example of effect of fabric (sample preparation method) on
cyclic strength of sands (Ladd, 1978)
356
7.12
Cyclic simple shear test on Nevada sand (Velacs project data,
Arulmoli et al., 1992)
357
7.13 Cyclic simple shear data from several sands
358
7.14
Cyclic simple shear tests on Ottawa sand, illustrating how the
effect of static bias varies depending on the selected “failure”
criterion (Vaid and Finn, 1979)
360
7.15
Shaking table tests “corrected” for compliance effects (De Alba
et al, 1976)
361
7.16
Comparison between cyclic triaxial and shaking table tests on
Monterey sand, with cyclic simple shear tests on Oosterschelde
sand
361
7.17
Demonstration of the importance of principal stress rotation on
behaviour of dense sand by Arthur et al. (1980)
363
7.18
Cumulative volumetric strain in lightly dilatant Leighton
Buzzard sand caused by principal stress rotation (Wong and
Arthur, 1986)
364
7.19
Result of a pure principal stress rotation test on loose Toyoura
sand (Ishihara and Towhata, 1983)
365
7.20
Behaviour of Erksak sand in hollow cylinder test simulating
principal stress history for Molikpaq piezometer E1 during 12
April 1986 ice loading event
367
7.21
Cyclic triaxial test on Bonnie Silt (Velacs project, Arulmoli et
al., 1992)
368
7.22
Cyclic simple shear test on Bonnie Silt (Velacs project, Arulmoli
et al., 1992)
369
Molikpaq core (for comparison with measured dissipation on
Figure 1.22)
7.25
Earthquake magnitude scaling factors tabulated and
superimposed on cyclic strength curve in Seed and Idriss (1982),
with power law trend added
377
7.26
Most recent version of Seed 1 iquefaction diagram (Youd et al,
2001) Data points numbers correspond to the case history
reference assigned by Fear (1996) based on Ambraseys (1988)
379
7.27
CPT version of Seed liquefaction diagram (Robertson and
Wride, 1998)
380
7.28
CPT resistance adjustment factor to provide equivalent clean
sand value from that measured in the actual soil at the site
(Robertson and Wride, 1998)
381
7.29
Recommended magnitude scaling factors from NCEER
Workshop (Youd and Noble, 1997)
383
7.30 K
σvalues (after Seed and Harder, 1990)
384
7.31 K
σvalues recommended by Hynes and Olson (1999)
385
7.32
Illustration redrawn from Lee and Seed (1967) showing apparent
effect of consolidation stress ratio K
con liquefaction resistance
387
7.33
Summary of recommended values for K
α(Harder and
Boulanger, 1997)
388
7.34 Illustration of alternative hardening laws
390
7.35
Biaxial compression test on an assembly of photo-elastic discs
(after de Josselin de Jong and Verruijt, 1969)
391
7.36
Schematic of yield surface softening induced by principal stress
rotation (from Been et al. 1993)
392
Nevada sand, from Velacs project
7.38
Comparison of Seed liquefaction diagram with state parameter
presentation of cyclic triaxial test data. Note that both diagrams
are based on 15 cycles of loading, as M=7.5 corresponds to 15
cycles (Figure 7.23)
396
7.39 Field case history data on liquefaction expressed in terms of ψ
397
7.40
Comparison of K
σrecommended by NCEER with K
σcomputed
from ψ changing with stress level at constant soil density
398
7.41
Illustration of the critical level of repeated loading (from
Sangrey et al., 1978)
401
7.42 Critical state model for liquefaction of sands, silts and clays
402
B.1
Effect of sample preparation on the cyclic resistance of sand
samples (from Ladd, 1977)
420
B.2
Illustration of sample preparation methods for clean sands
(Ishihara, 1993)
421
B.3
Illustration of vacuum saturation apparatus for triaxial sample
preparation (Shen and Lee, 1995)
427
B.4
Volume changes during triaxial sample lifetime (for a drained
test on a dilatant sample)
428
B.5
Potential error in void ratio if volume changes during saturation
are not considered (from Sladen and Handford, 1987)
429
B.6
Normalized membrane penetration coefficient as a function of
median grain size
431
B.7
Comparison of CSL determined from load controlled and strain
rate controlled triaxial compression tests
433
B.8
Triaxial cell with axial load cell located underneath cell to
minimize dynamic effects. Pore pressure transducer is also close
to cell to reduce system compliance
C.1
Erksak sand stress dilatancy in triaxial compression and
extension (after Jefferies and Shuttle, 2002)
438
C.2
Brasted sand stress dilatancy in plane strain and triaxial
conditions (after Jefferies and Shuttle, 2002)
439
C.3
Lode angle at peak strength in plane strain (after Jefferies and
Shuttle, 2002)
440
C.4
Comparison of functions for M(θ) with Brasted sand data (after
Jefferies and Shuttle, 2002)
442
D.1
Schematic illustration of self-consistency requirement for
internal cap to yield surface (Jefferies, 1997)
453
D.2
Yield surface softening induced by principal stress rotation
(Been et al., 1993)
459
D.3
Fit of NorSand to data with modified stiffness and dilatancy for
reloading (R>1, P<1) (from Jefferies, 1997)
467
D.4 Comparison of undrained liquefaction alternative softening rules 479
D.5
Dense sand after failure in Imperial College plane strain
apparatus, tested by Cornforth, 1964
482
D.6 Simple shear conditions (from Potts et al, 1987)
486
E.1 Chamber size standardization factors (Been et al., 1986)
499
F.1
Location of flowslide on the coast of Zeeland from 1881–1946
(Koppejan et al., 1948)
519
F.2 Vlietpolder flowslide geometry (Koppejan et al., 1948)
520
F.3
Typical CPT soundings in flowslide material (Koppejan et al.,
1948)
521
F.4
Longitudinal and transverse sections of North Dyke of
Wachusett Dam (Olson et al., 2000)
F.5
Cross section of Wachusett dam failure with 1991 investigation
results (Olson et al., 2000)
523
F.6
Typical section of hydraulic fill dam during construction (Hazen,
1918)
525
F.7
Sketch of Calaveras Dam failure showing surface before and
after slip (Hazen, 1918).
252
F.8
Sheffield dam, based on Seed et al., (1969), modified to show
liquefying layer
528
F.9 Aerial photographs of Fort Peck Dam failure
531
F.10
Critical state summary for Fort Peck Dam shell material
(Middlebrooks, 1940)
532
F.
11
Section through Fort Peck Dam failure (Casagrande, 1975)
532
F.12 Plan and section of Hokkaido tailings dam (Ishihara et al, 1990) 534
F.
13
CPTs from Hokkaido tailings dam (Ishihara et al, 1990)
535
F.14
Cross section of Mochikoshi tailings dams (Dam N°1 is top,
Dam N°2 is the bottom) (Ishihara et al, 1990)
536
F.
15
Double-tube cone penetration test at Mochikoshi Tailings Dams
(Ishihara et al 1990)
537
F.16
Nerlerk B-67 berm and foundation cross section (Been et al,
1987)
538
F.
17
Plan of failures that occurred at Nerlerk B-67 as reported by
Sladen et al. (1985a)
539
F.18
Example of bathymetric survey data at Nerlerk showing
interpolation of berm contours
540
F.19
Summary of state and stress paths in triaxial tests of
reconstituted Nerlerk 270/1 samples
sand and Ukalerk sand
F.21
Reconstructed cross section through failed portion of La
Marquesa Dam (De Alba et al., 1988)
546
F.22
Cross section through failure zone of La Marquesa Embankment
(De Alba et al 1988)
546
F.23
Reconstructed cross section through failed portion of La Palma
Embankment (De Alba et al., 1988)
549
F.24
Cross section through failure zone of La Palma embankment (De
Alba et al., 1988)
549
F.25
Illustration and cross-section through failure zone of Sullivan
Mine tailings dyke slide (Davies, Dawson and Chin, 1998)
551
F.26
CPT soundings through Sullivan Dyke failure (see Figure F.25b
for location)
553
F.27 Estimated in situ state from CP91–29 data by screening method 554
F.28
Example of flowslide geometry at Jamuna (Yoshimine et al.,
1999)
556
F.29
Plan view of west guide bund of Jamuna Bridge showing CPT
locations (Yoshimine et al, 2001)
556
F.30
Statistical summary of Jamuna west bund CPT results (based on
data provided by Prof Yoshimine)
Subscripts
c Critical state (some workers also use “cv” denoting constant volume) cyc Cyclic
L Limit value i
Image condition (occurs when Dp≡0 for all ) k Characteristic (in the sense of limit state Codes) st Static
tc Triaxial compression condition (θ=π/6) te Triaxial extension condition (θ=−π/6) h Horizontal
v Vertical; volume
0 Initial condition
1 Denotes value at reference stress level of 100 kPa (≈1 tsf) 1, 2, 3 Principal directions of stress or strain
Superscipts
e Elastic p Plastic
Stress Variables (bar over or ‘denotes effective)
σ1, 2, 3
[FL−2] Principal stresses [FL−2]
Mean effective stress [FL−2] Deviatoric stress invariant
p′ [FL−2]
Mean effective stress q [FL−2]
Triaxial deviator stress η [-]
Dimensionless shear measure as ratio of stress invariants θ [Rad]
Lode angle,
α [Rad] Included angle between direction of major principal stress (the “1” direction) and coordinate frame of reference
u [FL−2] Pore pressure
Strain Variables (dot superscript denotes rate)
ε1, 2, 3 [-] Principal strains (assumed coaxial with principal stresses)
[-]
Volumetric strain rate [-]
Shear strain rate measure work conjugate with
Dp [-]
e [-] Void ratio K0 [-]
Geostatic stress ratio, γd [FL-3] Dry unit weight
γt [FL−3] Total (and generally saturated) unit weight
γ′ [FL−3
] Submerged unit weight ψ [-] State parameter, ψ= e−ec
R [-]
Overconsolidation ratio
Geometry Variables
H [L] Slope height; layer thickness
θ [deg] Slope angle
Laboratory Testing Parameters and Variables
A [-] Skempton’s triaxial excess pore pressure parameter A=∆u/∆σq
B [-] Skempton’s excess pore pressure parameter B=∆u/∆σ3
Gs [-] Specific gravity
[deg] Mohr Coulomb friction angle (effective implied by context) su [FL−2] Undrained shear strength
sr [FL−2] Residual (post-liquefaction) undrained shear strength Elasticity E [FL−2] Young’s modulus G [FL−2] Shear modulus K [FL−2] Bulk modulus Ir [-]
Soil shear rigidity κ [-]
Slope of elastic line in space v [-] Poisson’s ratio
Critical State
Γ [-]
Reference void ratio on CSL, conventionally defined at p′=1 kPa λ
[-] Slope of CSL in space for semi-log idealization λ10
[-]
Slope of CSL, but defined on base 10 logarithms M
[-]
Critical friction ratio, equals ηc at the critical state. Varies with Lode angle, value at triaxial compression (Mtc) taken as reference.
χ [-] Dilatancy constant H [-] Plastic hardening modulus
Hr [-] Plastic softening modulus under principal stress rotation
Mi [-] Current value of η at Dp=0 (used in the flow rule)
CPT Parameters and Variables
qc [FL−2] CPT tip resistance, as measured
qt [FL−2] CPT tip resistance after correction for unequal area effect
fs [FL−2] CPT friction sleeve stress measurement
uc [FL
−2
] Pore pressure measured by CPT during sounding at shoulder location (sometimes denoted as u2 location in the literature).
Q [-] Dimensionless CPT resistance based on vertical stress.
Corresponds to standard usage within in situ testing community,
Qp [-]
Dimensionless CPT resistance based on mean stress, Bq [-]
CPTu excess pore pressure ratio,
F [-]
Stress normalised CPT friction ratio,
Ic [-] Soil classification index from Been and Jefferies (1992)
Ic,RW
Nk [-]
CPT undrained strength factor,
Preface
Soil liquefaction is a phenomenon in which soil loses much of its strength or stiffness for a generally short time but nevertheless long enough for liquefaction to be the cause of many failures, with both deaths and some very large financial losses. Unsurprisingly, there is a vast literature on liquefaction. Apart from a sustained set of contributions to the usual journals over the past thirty years or so there have been publications by university departments specializing in the subject, specialty conferences, theme sessions at geomechanics conferences, state-of-the-art papers, two Rankine lectures, the Canadian Liquefaction Experiment (Canlex), a book by the National Research Council (USA) and research competitions sponsored by the National Science Foundation to test theories and modeling techniques (VELACS in particular).
So where does this book fit in? This book assumes some background in soil mechanics, and an interest in understanding liquefaction. But, this book is most certainly not an overview of the literature nor does it seek to strike a balanced view of the various theories and experience about liquefaction (if indeed a balanced view is possible). Rather the aim is demystifying liquefaction, both in terms of the physics of soil behaviour and how it can be avoided in practice by good engineering. To achieve these dual goals, a number of viewpoints are adopted:
• Full scale experience of liquefaction, while immensely important as calibration
information, can be erroneously extrapolated without an underlying constitutive model to guide the interpretation of observations. An “understanding” of liquefaction is not enough, and a mere geological classification approach is unacceptable for engineering as such classification approaches are likely to mislead in some circumstances. • Liquefaction is simply another facet of soil behaviour which proper constitutive models
capture, including the effect of density and stress level on that behaviour. There are now several such models, and there is no need to resort to the witchcraft or dogma about liquefaction (e.g. “metastable soil particle arrangements” and similar “explanations” of the phenomenon).
• Constitutive models are idealizations, but idealization is not the sole domain of constitutive models. Both laboratory test data and physical model tests must be used equally cautiously. There are issues of scale.
• A proper appreciation of liquefaction thus requires a mixture of constitutive theory, soil properties and their determination (the in situ state in particular), and detailed
evaluation of the case history record.
In exploring these views, the book does not follow the mainstream US and Japanese research on liquefaction. The reason is that the subject is much broader than earthquake engineering. There is also much valuable experience outside the Pacific Rim. More importantly, there are aspects of the Pacific Rim approach, which is based on a reference stress concept, that are questionable mechanics. In some instances, the usual reference
proper dimensionless approach based on mechanics.
Mainstream experience around the Pacific Rim has been concerned with earthquake induced liquefaction through cyclic changes of shear stress, which is entirely reasonable given the economic base in California and Japan and the earthquake hazard of both these areas. Earthquakes are by no means necessary for liquefaction, however, and in seeking a full understanding it is important to use full scale experience. There is useful experience to be assimilated from coastal flowslides in The Netherlands which have been triggered quasi-statically. Equally, earthquakes usually last less than twenty or so significant load cycles and there is additional understanding to be gained from the offshore and coastal facilities exposed to several hundreds or thousands of load cycles in storm loading.
This book aims to help, and to do this we have made the files used to create plots downloadable from the internet (including the supporting raw data). Downloads can be found at the Golder Associates website (use the link www.golder.com/liq or search under the “library” menu) and which has a readme file with the latest information. Much of the interesting data on liquefaction is hidden in hard to access files at universities, consultants and research organizations. So we have made available for downloading all sorts of source data that we have found useful, as well as raw and processed data for many of the tests carried out by Golder Associates. The source code for numerical routines is entirely open code, and hope that you might be prepared to load an Excel file and try something. Look for the following on the website:
• Files with actual test data for the various plots in the book, These include triaxial test results, CPTs, and summaries of CPT calibration chamber studies.
• Source code for numerical routines discussed in the book, provided as Visual Basic within Excel spreadsheets. These are readily translated to other languages as needed. We refer to plots or routines with italics in the text and also to the name of the file used to create a particular figure with the figure number. Source code for the numerical routines is in commented open-code, and we hope the commenting is sufficiently detailed that the way the code works is clear. Files can be downloaded and the various routines used within Excel while reading the book so the development of ideas can be seen. We encourage this, as seeing the simulations come alive on a screen makes the effect of changing properties way more interesting than looking at the plots only on paper. Running the simulations also gives a feel for things, Plus, plots scales can be changed and so forth if interested in some detail. Please contact us if you discover any bugs as that will help everyone. You will have guessed by now that we are strong supporters of open software and freely shared data, and we hope for the same from you the reader. Although the downloads are “freeware”, they are copyrighted; but, they can be redistributed and/or modified under the terms of the GNU General Public License Version 2 as published by the Free Software Foundation (see page xxviii). Broadly, we are asking for developments built on this book to be as open as provided here—that will help everyone. Some of the articles on the Free Software Foundation website (http://https://www.fsf.org) show why this will be helpful to the profession in general. If it is desired to incorporate code in a proprietary product, contact us.
We also draw your attention to the Appendices, which contain detailed derivations that you will not find in published papers (and which textbooks do not usually explain either).
We trust that there is no equation in this book that appears without an appropriate derivation, as having a detailed derivation removes any black-magic aspects. The Appendices also document testing procedures, give the entire database of cone chamber calibration tests, and have details on interesting case histories as well as records of difficult to find information. These appendices are a substantive part of the book.
Those of you who have read some of our papers will recognize that we owe a great deal to the Canadian offshore oil industry and Golder Associates. Much of what is presented in this book was developed on construction projects for Gulf Canada Resources, Esso Resources Canada, and Dome Petroleum. We thank in particular John Hnatiuk, Bill Livingstone, Brian Wright, Brian Rogers, Chris Graham, David James, and Sanjay Shinde. Without the support from these gentlemen, much of what you will find in this book would not exist.
We also owe an enormous debt to our colleagues in Golder Associates who have both encouraged and humoured us. Our co-authors are acknowledged by reference to the various joint papers, and we thank in particular Jack Crooks, Brian Conlin, John Cunning, Dave Horsfield, Joe Hachey, and Dawn Shuttle. Production of this book has also been supported by Golder Associates. We thank all people in the company for this.
As we pulled the final version for this book together, we realized that we also had worked rather closely with the University of British Columbia. Although we have disagreed with Yogi Vaid over many issues, his testing has been a tremendous stimulus and he directed some of the more interesting testing of Erksak sand. This relationship has continued today in the context of both laboratory (Dharma Wijewickreme) and in situ testing (John Howie). The most recent developments of NorSand, and its implementation in computable models (FLAC and Finite Element codes) have actually been carried out by Dawn Shuttle and her students. Some their routines and programs can be downloaded from the UBC Geotechnical Engineering website (http://www.civil.ubc.ca/).
Many friends and acquaintances not already mentioned also provided useful case history and laboratory data over the years. At the risk of missing a few names we acknowledge data provided by Mike Jamiolkowski (calibration chamber data), Ramon Verdugo (Toyoura sand test data), Mike Hicks and Radu Popescu (probabilistic analyses), Gonazalo Castro (Lower San Fernando data), Scott Olson (various CPT data), Prof Yoshimine (Jamuna Bridge data) and Howard Plewes (Sullivan Dam data).
It is also appropriate that we acknowledge intellectual mentoring over the years, Professor Bob Gibson introduced one of us (MJ) to theoretical soil mechanics. Other guidance and encouragement was provided by Professor Andrew Palmer, and the late Professors Peter Wroth and Dan Drucker. We are very appreciative of their time and interest. Hopefully, this book will do justice to their influence.
Finally, both of us must thank our parents for providing our initial engineering education and our families for their forbearance subsequently, This book, and the papers that lead to it, have taken many hours of our lives. Our families have humoured us by not complaining as we disappeared into our studies for many nights over the best part of twenty years, Thank you, Hilary and Fiona,
Mike Jefferies, C.Eng. Associate, Golder Associates (mjefferies@golder.com)
Ken Been, P.E. Principal, Golder Associates (kbeen@golder.com)
As a of the subject, this book and may be hazardous. Substantial and is to work safely. This book not cover work for Appropriate training should be received attempting any of the in this book. Neither the authors nor the publisher any liability for your working
This book provides and equations. Neither the nor the publishers warrant the or suitability for a particular of any of the information or and you must carry out your own quality relying on trends, equations, or numerical routines. This is a normal requirement for a professional engineer, and this book is to be only by professional geotechnical engineers.
Software can be downloaded for use the of this book is provided on the Golder Associates website http://www.golder.com/. Downloaded. is but is copyrighted and provided as “Freeware”; you can redistribute it and/or modify it under the of the GNU General Public License Version 2 as published by the Free Foundation. A copy of this is on the Golder website, and will be automatically included with every download. Programs and are downloadable from Golder in the that they will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for details.
CHAPTER ONE
Introduction
1.1 WHAT IS THIS BOOK ABOUT?
Soil liquefaction is a phenomenon in which soil loses much of its strength or stiffness for a generally short time but nevertheless long enough for liquefaction to be the cause of many failures, deaths and major financial losses. For example, the 1964 Niigata (Japan) earthquake caused more than $1 billion in damage and most of this damage was related to soil liquefaction. The Aberfan (Wales) colliery spoil slide was caused by liquefaction and killed 144 people (116 of whom were children) when it inundated a school. Liquefaction was involved in the abandonment of the Nerlerk (Canada) artificial island after more than $100 million had been spent on its construction. Liquefaction at Lower San Fernando dam (California) required the immediate evacuation of 80,000 people living downstream of the dam. Liquefaction is an aspect of soil behaviour that occurs worldwide and is of considerable importance from both public safety and financial standpoints.
In terms of age of the subject, Ishihara in his Rankine Lecture (Ishihara, 1993) suggests that the term spontaneous liquefaction was coined by Terzaghi and Peck (1948). The subject is much older than that, however. Dutch engineers have been engineering against liquefaction for centuries in their efforts to protect their country from the sea. Koppejan et al. (1948) brought the problem of coastal flow slides to the soil mechanics fraternity at the 2nd International Conference in Rotterdam. In the last paragraph of their much-cited paper, they mention flow slides in the approach to a railway bridge near Weesp in 1918 triggered by vibrations from a passing train. They claim that this accident, with heavy casualties, was the immediate cause of the start of practical soil mechanics in the Netherlands.
At about the same time, Hazen (1918) reporting on the Calaveras Dam failure clearly recognized the phenomenon of liquefaction and the importance of pore pressures and effective stresses. If the files of the US Corp of Engineers are consulted one finds that Colonel Lyman densified fill for the Franklin Falls dam (part of the Merrimack Valley Flood Control scheme) in the late 1930s specifically to ensure stability of the dam from liquefaction based on the concept of critical void ratio given in Casagrande (1936). Reading these files and reports is an enlightening experience as Casagrande discussed many topics relevant to the subject today, and Lyman’s report (1938) of the Corps’ engineering at Franklin Falls is a delight that historians of the subject will enjoy.
We will not attempt our own definition of liquefaction, or adopt anyone else’s, beyond the first sentence of this book. Once it is accepted that liquefaction is a constitutive behaviour subject to the laws of physics, it becomes necessary to describe the mechanics mathematically and the polemic is irrelevant.
Liquefaction evaluation is only considered worthwhile if it may change an engineering decision. Testing and analysis of liquefaction potential is only undertaken in practice in the context of a particular project. As will be seen later, liquefaction is an intrinsically