See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/274427988

Analysis of Factors Influencing Soil Classification Using Normalized Piezocone

Tip Resistance and Pore Pressure Parameters

Article  in  Journal of Geotechnical and Geoenvironmental Engineering · November 2008

DOI: 10.1061/(ASCE)1090-0241(2008)134:11(1569)

CITATIONS READS

81 1,182

4 authors, including:

James Schneider Mark Randolph

Consulting Engineer University of Western Australia
52 PUBLICATIONS   953 CITATIONS    355 PUBLICATIONS   9,920 CITATIONS   

SEE PROFILE SEE PROFILE

Paul W. Mayne
Georgia Institute of Technology
239 PUBLICATIONS   4,186 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Efficient export riser solution for gas field developments in deep-water sites. View project

Evaluation of effective stress strength parameters from CPTu View project

All content following this page was uploaded by James Schneider on 07 November 2015.

The user has requested enhancement of the downloaded file.

 Analysis of Factors Influencing Soil Classification
Using Normalized Piezocone Tip Resistance
and Pore Pressure Parameters
James A. Schneider1; Mark F. Randolph2; Paul W. Mayne3; and Nicholas R. Ramsey4

Abstract: This paper discusses the development of a framework for classifying soil using normalized piezocone test 共CPTU兲 data from
the corrected tip resistance 共qt兲 and penetration pore-water pressure at the shoulder 共u2兲. Parametric studies for normalized cone tip
resistance 共Q = qcnet / ␴⬘v0兲 and normalized excess pressures 共⌬u2 / ␴⬘v0兲 as a function of overconsolidation ratio 共OCR= ␴⬘vy / ␴⬘v0兲 during
undrained penetration are combined with piezocone data from clay sites, as well as results from relatively uniform thick deposits of sands,
silts, and varietal clays from around the globe. The study focuses on separating the influence of yield stress ratio from that of partial
consolidation on normalized CPTU parameters, which both tend to increase Q and decrease the pore pressure parameter 共Bq
= ⌬u2 / qcnet兲. The resulting recommended classification chart is significantly different from existing charts, and implies that assessment of
data in Q – ⌬u2 / ␴⬘v0 space is superior to Q – Bq space when evaluating piezocone data for a range of soil types. Still, there are zones of
overlap for silty soils and heavily overconsolidated clays, thus requiring that supplementary information to Q and ⌬u2 / ␴⬘v0 be obtained in
unfamiliar geologies, including variable rate penetration tests, dissipation tests, CPT friction ratio, or soil sampling.
DOI: 10.1061/共ASCE兲1090-0241共2008兲134:11共1569兲
CE Database subject headings: Soil classification; Cone penetration tests; Silts; Sand; Clays; Overconsolidated soils.

Introduction soil properties, but also in identification of when penetration oc-

Parameters measured during a piezocone penetration test 共CPTU兲
are commonly used for soil profiling and geostratigraphy, as well
as parameter assessment in geotechnical design. Correlations to
engineering design parameters can be relatively reliable for pen-
etration in sands and clays, in which both penetration and the
design application typically occur under fully drained or un-
drained conditions, respectively. In “transitional” soils, such as
clayey sands and silts, silty clays, silts, and many residual soils,
CPTUs are often conducted under conditions of partial consoli-
dation, i.e., where some 共but not full兲 dissipation of excess pore-
water pressure occurs locally around the advancing cone. In such
cases, there is significant uncertainty not only in the assessment of
1
Ph.D. Student, School of Civil and Resource Engineering, The Univ.
of Western Australia, 35 Stirling Highway, Crawley, Western Australia
6009, Australia 共corresponding author兲. E-mail: schneider@civil.uwa.
edu.au
2
Federation Fellow, Centre for Offshore Foundation Systems, The
Univ. of Western Australia, 35 Stirling Highway, Crawley, Western
Australia 6009, Australia. E-mail: randolph@civil.uwa.edu.au
3 water pressure-based soil classification charts. This framework for
Professor, Geosystems Engineering Group, School of Civil and
Environmental Engineering, Georgia Institute of Technology, 790
Atlantic Dr. N.W., Atlanta, GA 30332-0355. E-mail: paul.mayne@
ce.gatech.edu
4
Principal Geotechnical Engineer, Sinclair Knight Merz 共SKM兲,
P.O. Box 2500, Malvern, Victoria 3144, Australia. E-mail: nramsey@
skm.com.au
Note. Discussion open until April 1, 2009. Separate discussions must
be submitted for individual papers. The manuscript for this paper was
submitted for review and possible publication on March 12, 2007; ap-
proved on March 3, 2008. This paper is part of the Journal of Geotech-
nical and Geoenvironmental Engineering, Vol. 134, No. 11, November
1, 2008. ©ASCE, ISSN 1090-0241/2008/11-1569–1586/$25.00.
curs under conditions of partial drainage, and to what degree pore
pressures are influencing cone resistance. As penetration re-
sistance in normally consolidated soils has been reported to
increase by approximately three 共House et al. 2001兲 to ten times
共McNeilan and Bugno 1985兲 as penetration moves from un-
drained to drained conditions, assessment of drainage conditions
during penetration has significant implications on reliable appli-
cation of design correlations.
Soil behavioral classification charts using normalized piezo-
cone parameters are well known for postprocessing results 共e.g.,
Robertson 1990, 1991兲. Nevertheless, additional complexities and
nuances in soil behavior can elude a proper soil classification
using these simplified charts. For instance, piezocone results in
clay from the North Sea 共Ramsey 2002兲 are presented in Fig. 1
and show discrepancies in that existing chart. The North Sea clay
data plot in five distinct zones of the Q – Bq 共1991兲 soil classifi-
cation chart 共excepting Zones 1 and 2兲, covering sand, silt, clay,
and soil mixtures. This paper presents theoretical considerations
with supporting data that illustrate factors influencing the location
and “direction of movement” of data within tip resistance pore-
soil classification by piezocone using normalized cone tip resis-
tance and normalized excess pore pressures is discussed with em-
phasis on whether penetration is drained, undrained, or partially
drained. Trends in normalized piezocone response observed in
this study may influence design decisions, particularly in nontext-
book geomaterials, such as silts, heavily overconsolidated clays,
loams, sensitive clays, and mixed soil types. The value of 共cor-
rected兲 CPT sleeve friction measurements 共f t兲 and the correspond-
ing normalized parameter 共friction ratio, F兲 in the evaluation of
soil type is recognized, but beyond the scope of detailed analysis
for this paper.

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008 / 1569


Fig. 1. Location of North Sea clay data 共Ramsey 2002兲 on Robertson
共1991兲 normalized classification charts
Framework for Normalized Soil Classification by
Piezocone
Normalized Parameters
As the effective overburden stress increases with depth, cone pen-
etration tip resistance also tends to increase. This increase in read-
ings with depth may cause errors in the interpreted soil
classification using raw piezocone readings 共Robertson 1990兲,
particularly in deep soundings such as those from offshore inves-
tigations. Therefore, normalization of measured parameters is re-
quired for rational evaluation of soil behavior 共Wroth 1984,
1988兲. During undrained penetration, the cone tip resistance is
primarily controlled by the undrained shear strength and as such
can be expressed simply as
qt − ␴v0 qcnet
Nkt = = 共1兲
su su
where qt = total cone tip resistance corrected for area ratio 共Lunne
et al. 1997兲; ␴v0 = total vertical stress; and qcnet = net cone tip re-
sistance. Although it is emphasized that the measured cone tip
resistance 共qc兲 must be corrected for area ratio to qt, the term qcnet
is used for net cone tip resistance to prevent confusion with T-bar
penetration resistance 共e.g., Randolph 2004兲. The cone factor Nkt
can be considered a “normalized” cone tip resistance. However,
as profiles of undrained shear strength are generally unknown at a
given site, this format for normalization is not useful for assess-
ment of normalized resistance in practice. Additionally, the pa-
rameter Nkt will be influenced by a number of different soil and
1570 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008
penetrometer properties including: 共1兲 soil stiffness or rigidity
index; 共2兲 cone and sleeve roughness; 共3兲 ratio of horizontal to
vertical stress; 共4兲 strength anisotropy; 共5兲 soil sensitivity; 共6兲
degree of pore pressure dissipation during penetration; and 共7兲
viscous rate effects, which complicate the analysis of cone tip
resistance or correlations between su and qcnet 共e.g., Teh and
Houlsby 1991; Mayne 1992; Randolph 2004兲. Therefore, normal-
ization of cone tip resistance is typically based on vertical effec-
tive stress, ␴⬘v0, as 共Wroth 1988兲
q t − ␴ v0 qcnet
Q= = 共2兲
␴⬘v0 ␴⬘v0
Although this normalization is simple to apply and accounts for
the influence of stress level on soil strength and stiffness, it is not
a unique parameter for assessing clay state during undrained pen-
etration. Definition of the cone factor 共Nkt兲 in relation to the nor-
malized cone tip resistance 共Q兲, results in the following
expression:
qcnet Q
Nkt = = 共3兲
su su/␴⬘v0
The undrained strength ratio 共su / ␴⬘v0兲 of soils has been shown to
be strongly affected by the vertical yield stress ratio 共YSR兲, or
overconsolidation ratio 共OCR兲, where the YSR= ␴⬘vy / ␴⬘v0 and the
OCR= p⬘c / ␴⬘v0. Although the vertical yield stress, ␴⬘vy, is explicitly
allowed to exceed the vertical preconsolidation pressure, p⬘c , due
to additional effects such as aging, OCR is used in this paper for
consistency with typical nomenclature. The relationship between
OCR and su / ␴⬘v0 is often expressed as
冉 冊
su su
= OCR⌳ 共4兲
␴⬘v0 ␴⬘v0 NC
where ⌳ = plastic volumetric strain ratio and 共su / ␴⬘v0兲NC
= normally consolidated undrained shear strength ratio. The
plastic volumetric strain ratio is expressed as ⌳ = 1 − ␬ / ␭, where
␭ ⬇ Cc / 2.3 or the isotropic compression index, and ␬ ⬇ Cs / 2.3 or
the isotropic swelling index. The influence of vertical yield stress
on qcnet, or OCR on Q, has been explored by a number of re-
searchers, resulting in the following relationships:
␴⬘vy = ␬ pqcnet 共5a兲
␴⬘vy
OCR = = ␬ pQ 共5b兲
␴⬘v0
where ␬ p = empirical coefficient which typically varies from 0.1 to
0.5 with an average of 0.3 共e.g., Mayne et al. 1998兲. The empiri-
cal relationship shows that Q increases with OCR, as suggested
by Eqs. 共3兲 and 共4兲. The influence of soil friction angle 共␾⬘兲 on Q
during undrained penetration is often quantified through the nor-
mally consolidated undrained strength ratio, i.e., 共su / ␴⬘v0兲NC
which tends to decrease with ␾⬘ 共Wroth and Houlsby 1985兲. As
with undrained strength, profiles of OCR 共or ␴⬘vy兲 and ␾⬘ are
typically not known at a given site, and normalization of cone tip
resistance based on Eq. 共2兲 is still the most practical option.
Normalization of penetration pore pressures first requires sepa-
ration of pore pressures that are a function of soil response and
those existing in the ground prior to penetration. Measured pen-
etration pore pressure 共um兲 can be expressed simply as the sum of
the in situ pore pressure 共u0兲 and the excess pore pressure 共⌬um兲
 um = u0 + ⌬um 共6兲
During cone penetration testing, pore pressures are typically mea-
sured behind the cone tip at the shoulder, or u2 location. The
excess pore pressure component 共⌬um兲 can be further separated
into the pore pressure resulting from increase in total stress due to
cone insertion 共octahedral, ⌬uoct兲 and the component induced by
shearing of the soil 共⌬ushear兲 共Randolph et al. 1979; Wroth 1984;
Baligh 1986; Mayne and Bachus 1988兲
um = u0 + ⌬ushear + ⌬uoct 共7兲
It is not possible to separate the measurements of the shear and
octahedral components of pore pressure at the cone shaft, and
these distinctions must be estimated theoretically.
Tip resistance and pore pressure-based soil classification by
piezocone typically uses the pore pressure parameter, Bq
共Senneset and Janbu 1985; Robertson et al. 1986; Robertson
1990; Ramsey 2002兲, defined as
u2 − u0 ⌬u2
Bq = = 共8兲
qt − ␴v0 qcnet
Alternatively, use of the excess pore pressure ⌬u2 directly for soil
classification has been discussed 共Jones and Rust 1982兲, and this
paper will make use of the normalized excess pore pressure,
⌬u2 / ␴⬘v0. Normalized soil classification charts based on Q and Bq
or Q and ⌬u2 / ␴⬘v0 are analogous as
⌬u2 1
Bq = 共9兲
␴⬘v0 Q
This paper presents soil classification charts in both formats
共Q – Bq space or Q – ⌬u2 / ␴⬘v0 space兲 and discusses relative advan-
tages of each.
For assessing partial consolidation effects, a normalized pen-
etration velocity 共V兲 has been proposed as a function of cone
velocity 共v兲, penetrometer diameter 共d兲, and coefficient of con-
solidation 共cv兲 given by 共Finnie and Randolph 1994; Randolph
2004兲
vd
V= 共10兲
cv
Influence of viscous rate effects on penetration resistance evalu-
ated using normalized velocity is discussed by Chung et al.
共2006兲 and Lehane et al. 共2008兲, among others.
For pore pressure dissipation at the shoulder location during a
pause in penetration, the operational coefficient of consolidation
tends to be intermediate between the normally consolidated and
swelling coefficient of consolidation 共Leroueil et al. 1995; Fahey
and Lee Goh 1995兲. During penetration, the appropriate “opera-
tional” coefficient of consolidation 共cv or ch兲 for use in Eq. 共10兲 is
still being researched. As most initial tests used in the develop-
ment of “backbone curves” related to the normalized velocity 共V兲
were performed on isotropic normally consolidated centrifuge
specimens 共e.g., Finnie 1993; Finnie and Randolph 1994; Watson
1999; House et al. 2001; Randolph and Hope 2004; Randolph
2004; Chung et al. 2006兲, the normally consolidated vertical co-
efficient of consolidation based on laboratory Rowe cell tests was
initially used in Eq. 共10兲.
It has been suggested that when using the normally consoli-
dated coefficient of consolidation, penetration occurs undrained
for normalized velocity greater than approximately 30–100
共Finnie and Randolph 1994; Randolph 2004兲. Fully drained be-
havior was originally considered at normalized velocities below
JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008 / 1571
0.01 共Finnie and Randolph 1994兲. A value of V less than 0.01 was
also calculated by Baligh 共as presented in Hight and Leroueil
2003兲 using the strain path method to be the normalized velocity
where excess u1 pore pressures 共measured on the cone face兲 were
zero. Randolph and Hope 共2004兲 found that u2 penetration pore
pressures were essentially hydrostatic for cone penetration in nor-
mally consolidated kaolin at V of approximately 0.1–0.3.
Schneider et al. 共2007兲 also observed that Bq and ⌬u2 / ␴⬘v0 were
approximately zero at V of 0.1–0.3, but that Q increased by ap-
proximately an additional 15% as V decreased from 0.3 to 0.03 in
a normally consolidated kaolin clay. This is a relatively small
change in Q compared to the three times increase in Q as V
decreases from 30 to 0.1 in normally consolidated kaolin.
As it is recognized that the coefficient of consolidation is in-
fluenced by a number of different factors, including the effects of
the penetration process, stress level, void ratio, stress history, and
anisotropy, a more appropriate form of the 共horizontal兲 coefficient
of consolidation 共ch兲 may be expressed as
kh共1 + e兲␴⬘h0
ch = 共11兲
␥w冑␭␬
where the hydraulic conductivity 共k兲 = function of void ratio 共e兲,
stress level, and predominant direction of flow; ␥w = unit weight of
water; and ␭ and ␬ have been previously defined in association
with Eq. 共4兲. Lehane et al. 共2008兲 have used Eqs. 共10兲 and 共11兲 to
achieve consistent backbone curves for velocity-dependent be-
havior in both normally consolidated and overconsolidated ka-
olin. Taking that the ratio of ␭ to ␬ is approximately 5 共e.g.,
Kulhawy and Mayne 1990兲, the ratio of the normally consolidated
to operational coefficient of consolidation using Eq. 共11兲 is ap-
proximately equal to 冑5, and thus the boundaries for V using ch
expressed by Eq. 共11兲 tend to range from 0.05 to 15 for drained
and undrained penetration, respectively. This range is in agree-
ment with experimental data of Lehane et al. 共2008兲 for kaolin
with OCR between 1 and 5. For the constant 20 mm/ s rate of
penetration in a standard 10 cm2 CPTU 共ASTM 2000; ISSMFE
1997兲, penetration will generally occur under conditions of partial
consolidation if cv is between approximately 350 m2 / year and
1 ⫻ 105 m2 / year using the definition of ch in Eq. 共11兲.
Trends in Normalized Parameters
The North Sea clay data in Fig. 1 tend to indicate that as Q 共or
OCR兲 increases, Bq tends to decrease. The trend of Bq decreasing
with OCR has been observed for individual sites 共e.g., Lacasse
and Lunne 1982; Wroth 1984兲, but the correlation between OCR
and Bq is not unique 共e.g., Jamiolkowski et al. 1985兲. Hight et al.
共1994兲, among others, also discuss how Q increases and Bq de-
creases due to the effects of partial consolidation. It is desirable to
be able to separate the effects of OCR and partial consolidation
when developing charts for classification of soils. Fig. 2共a兲 plots
normalized piezocone data 共Q and Bq兲 on the Roberston 共1991兲
soil classification charts for North Sea clays as well as six rela-
tively uniform sites. The data are from two predominantly silt
sites 共Halsen and Vägverket兲, three lightly overconsolidated clay
sites 共Bothkennar, Onsøy, and Troll兲, and one dense fine sand site
共Bugg 40兲. Additional information about these sites and other
sites discussed in to this paper is presented in Table 1.
In Fig. 2共a兲, the Bugg 40 sand site has essentially hydrostatic
pore pressure 共⌬u2 ⬇ 0⬖Bq ⬇ 0兲, with Q generally decreasing with
depth due to decreases in sand state 共e.g., Konrad 1998; Klotz and
Coop 2001兲. Both the silt and the clay sites show a general in-

Fig. 2. Differing trends of increasing su / ␴⬘v0 共OCR兲 and coefficient
of consolidation 共partial consolidation兲 for relatively uniform sites in
共a兲 Q – Bq; 共b兲 Q – ⌬u2 / ␴⬘v0 space
crease in Q with a corresponding decrease in Bq. Below 15 m at
the Vägverket silt site, Bq approaches zero and Q approaches 50.
This corresponds to a transition from a silt/sandy silt to a silty
sand/sand with cv above 6000 m2 / year in the lower layer 共Lars-
son 1997兲. Similar, but offset to the right, trends of increasing Q
and decreasing Bq are observed for increases in OCR at the clay
1572 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008
sites. The influence of OCR is therefore mischaracterized as
changes in silt content 共or coefficient of consolidation兲 due to
distortion of the classification chart by incorporating qcnet in the
normalized parameters on both axes. A heavily overconsolidated
clay will not necessarily behave in the same way as a normally
consolidated silty soil, and thus should not have the same “soil
behavior type” based on piezocone response. When classifying
soils based on piezocone data it is desirable to be able to separate
the influence of OCR and coefficient of consolidation. Plotting
CPTU data in Q – ⌬u2 / ␴⬘v0 space, as shown in Fig. 2共b兲, tends to
result in opposite trends for influence of OCR and coefficient of
consolidation, as the octahedral component of penetration pore
pressures often dominates u2 response during undrained penetra-
tion 共Wroth 1984; Baligh 1986兲.
Delineating Soil Types
Piezocone Testing of Clays
Prior to plotting field data on soil classification charts, the influ-
ence of OCR on undrained piezocone response in Q – ⌬u2 / ␴⬘v0
space is explored parametrically using solutions relating net cone
tip resistance, qcnet, and excess penetration pore pressure, ⌬u2, in
clay soils. The evaluation of qcnet using large strain finite element
analyses discussed by Lu et al. 共2004兲 is combined with the hy-
brid critical state-cavity expansion model developed by Mayne
共1992, 2001兲 for this parametric study. Although the cylindrical
cavity expansion analogy for penetration resistance is only par-
tially supported by numerical studies 共Lu et al. 2004兲, these sim-
plistic models are useful for evaluating trends in behavior on
Q – ⌬u2 / ␴⬘v0 共or Bq兲 classification charts due to the influence of
OCR.
As discussed previously, the relationship between undrained
shear strength and undrained cone tip resistance is influenced
by a number of factors. Large displacement numerical studies
共Lu et al. 2004兲 have quantified the influence of rigidity index
共Ir = G / su兲, roughness of the cone face 共␣c兲, and influence of hori-
zontal stress 关⌬ = 共␴v0 − ␴h0兲 / 共2su兲 = ␴⬘v0共1 − K0兲 / 共2su兲兴, on cone
factor, Nkt, as
Nkt ⬇ 3.4 + 1.6 ln共Ir兲 − 1.9⌬ + 1.3␣c 共12a兲
The roughness of the cone face varies between 0 and 1 for per-
fectly smooth and rough interfaces, respectively, and is taken
as 0.2 in these parametric studies. The influence of horizontal
stress will be a function of OCR, and is estimated based on the
correlation among friction angle 共␾⬘兲, OCR, and K0 共Mayne and
Kulhawy 1982兲
K0 = 共1 − sin ␾⬘兲OCRsin ␾ 共12b兲
A friction angle of 30° was used to estimate K0 in these paramet-
ric studies. The normalized cone tip resistance 共Q兲 is estimated by
varying su / ␴⬘v0 with OCR as expressed in Eq. 共4兲 at a constant Ir.
Although rigidity index tends to decrease with increasing OCR
共Keaveny and Mitchell 1986; Mayne 2001兲, evaluation of a range
of Ir values within the parametric study captures this effect. The
cone factor, Nkt, is calculated to vary with Ir per Eq. 共12a兲, and Q
is then calculated from su / ␴⬘v0 and Nkt using Eq. 共3兲.
Corresponding penetration pore pressures are evaluated using
the hybrid critical state-cavity expansion model described by
Mayne 共1992, 2001兲. Based on spherical cavity expansion theory,
Table 1. Data Sets Used in Development of New Piezocone Soil Classification Charts
Depth
Data set 共m兲 OCR Description References
North Sea soil database 70% of data ⬍3 m 1–1,200 CLAY, SAND, SILT, and Ramsey 2002
SAND mixtures
Bothkennar, United Kingdom 1.5–19.8 1–3 Soft CLAY Powell et al. 1988
Onsøy, Norway 1.7–19.4 1–4 Soft CLAY Lacasse and Lunne 1982
Troll East, North Sea 0.8–44 1–2 Soft CLAY Sandven 1990; Amundsen et al. 1985
Vägverket, Sweden 3–21 1–2 SILT to sandy silt with Larsson 1997
clayey silt
Halsen, Norway 4–20 1–2 SILT to clayey silt with sand Sandven 2002
Bugg 40, Arkansas, 4–34 NA Clean SAND to silty SAND Schneider et al. 2001
United States
Global clay database 1–51 1–80 Soft to hard CLAY Mayne et al. 1990; Chen and Mayne 1994;
Chen and Mayne 1996; Mayne et al. 1998;
Gloucester, Canada 3–16.5 1–2 Soft sensitive 共St ⬇ 100兲 CLAY Konrad and Law 1987
Tiller, Norway 6–20 1–2 Soft sensitive 共St ⬇ 250+ 兲 Sandven 1990
CLAY
Lierstranda, Norway 5–24 1–3 Sensitive 共St ⬇ 5 – 20兲 low Lunne and Lacasse 1999
plasticity CLAY
Burswood, Australia 5–8 1–2 Soft CLAY Randolph 2004; Chung et al. 2006
NC and OC kaolin, 3.5–15 1–6.5 Soft to stiff CLAY Schneider et al. 2007
UWA beam centrifuge
Global liquefaction database 1–10 NA Clean, silty, and clayey Moss 2003; Moss et al. 2006a
SAND and sandy SILT
Euripides, The Netherlands 23–45 NA Dense to very dense SAND Kolk et al. 2005
Drammen, Norway 3.5–27 NA Loose SAND Tveldt and Fredriksen 2003
Jamuna Bridge, Bangladesh 6–60 NA Micaceous SAND Tomlinson 2001; Fugro 1995
Overland Corner, South Australia 6.5–49 NA Fine to coarse calcareous Randolph et al. 1996
SAND and calcarenite
Fittja Straits, Sweden 2–27 NA Silty SAND Axelsson 2000
Opelika, United States 4–31 NA Residual sandy SILT to Mayne et al. 2000; Finke et al. 2001
silty SAND
SFOBB, United States 12–16 NA Aged silty SAND Olson and Shantz 2004
Brent Cross, United Kingdom 12 31 Hard CLAY Lunne et al. 1986

the octahedral component of penetration pore pressure can be Fig. 3 plots parametric studies of the relationship between Q and
evaluated as a function of the cone factor 共Nkt兲 and rigidity index ⌬u2 / ␴⬘v0 for 共su / ␴⬘v0兲NC ratios of 0.2 and 0.3. For typical rigidity
共Ir兲 from index values between 150 and 500, OCR and 共su / ␴⬘v0兲NC tend to
4
dominate the relationship, although for a given OCR, Q values
⌬uoct = 3 su ln共Ir兲 共13a兲 reduce with decreasing rigidity index due to its influence on Nkt.
As the rigidity index drops from 150 to 10, local shear induced
⌬u2,oct 4 1 pore pressures have a greater relative effect on the measured ⌬u2,
Bq,oct = = ln共Ir兲 共13b兲 leading to a steepening of the curves.
qcnet 3 Nkt
Theoretical parametric studies are compared to two databases
of piezocone measurements in clay soils: 共1兲 a global database of
⌬u2,oct clays with corresponding laboratory measurements of ␴⬘vy 关update
= QBq,oct 共13c兲
␴⬘v0 of database discussed by Mayne et al. 共1990兲 and Chen and
The shear component of penetration pore pressures is evalu- Mayne 共1996兲兴, and 共2兲 the clay subset of the database of North
ated as 共Mayne 2001兲 Sea soils discussed by Ramsey 共2002兲. Table 2 summarizes typi-
cal characteristics 共␴⬘v0, plasticity index 共PI兲, OCR兲 of the clay
⌬u2,shear databases, and Fig. 4 compares field data to the previously dis-
⬇ 1 − 共OCR/2兲⌳ 共14兲 cussed parametric studies. The data are plotted in log–log
␴⬘v0
Q – ⌬u2 / ␴⬘v0 space to achieve additional detail at low ⌬u2 / ␴⬘v0.
The normalized excess penetration pore pressure is the sum of the Three lines intended to bound the analytical studies are included,
octahedral and shear components: along with empirical fits of those curves. Although it is acknowl-
edged that the soil models used in the previously mentioned para-
⌬u2 ⌬u2,oct ⌬u2,shear
= + 共15兲 metric studies are a simplification of actual behavior, the
␴⬘v0 ␴⬘v0 ␴⬘v0 relationship between field data and analytical trends are useful

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008 / 1573

 Fig. 4. Delineation of clay soils in Q – ⌬u2 / ␴⬘v0 space

Additionally, the curve bounding observed North Sea clay

data presented by Ramsey 共2002兲 are included in Fig. 4, and is
calculated as

Q ⬍ max 29冋冉 冊 册⌬u2

␴⬘v0
0.9
,10 共17兲

Eq. 共16a兲 matches well with Eq. 共17兲 for ⌬u2 / ␴⬘v0 greater than 2
to 3, but differences at low excess penetration pore pressures have
implications for delineating silts from clays.
Assuming each curve presented in Eq. 共16兲 might be consid-
ered as a boundary for particular soil types, their potential accu-
racy for correctly classifying a given soil as clay is presented in
Fig. 3. Influence of rigidity index 共Ir = G / su兲 on trends in relationship
Table 3. Of the 476 data points in Fig. 4 only 14 data points do
between qcnet / ␴⬘v0 and ⌬u2 / ␴⬘v0 for 共a兲 共su / ␴⬘v0兲NC of 0.2; 共b兲
not plot between Eq. 共16a兲 and a Bq value of unity, with an even
共su / ␴⬘v0兲NC of 0.3
split of high and low data points. Still, 96% of the Mayne data-
base lies between Eq. 共16b兲 and Bq of unity, whereas only 50% of
the Ramsey 共2002兲 database lies in this same region. It is there-
when assessing implications from modeling piezocone response. fore inferred that the shallow heavily overconsolidated clays of
The soil conditions and empirical curves for each case are pre- the North Sea within the Ramsey database have lower apparent
sented in the following: rigidity index values than typical. The apparently low rigidity
index values may result from a high degree of unloading 共reduc-
⌬u2 Q1.25 tion in total radial stress and pore pressure兲 behind the cone tip,
= + 0.99 共su/␴⬘v0兲NC = 0.2, Ir = 10, Bq 艋 1
␴⬘v0 100 and thus low octahedral pore pressures as compared to cone tip
resistance, or from more negative shear-induced pore pressures
共16a兲 than predicted from Eq. 共14兲. Measurements of the reduction in
total radial stress and pore pressures behind a penetrometer tip
⌬u2 Q0.95 have been discussed by Lehane and Jardine 共1994兲. Eq. 共16c兲
= + 1.05 共su/␴⬘v0兲NC = 0.2, Ir = 30, Bq 艋 1 tends to be a best fit of the Mayne intact clay database rather than
␴⬘v0 5
a “boundary,” with 50% of the data lying on either side of the
共16b兲 equation. Soils plotting with higher normalized penetration pore
pressures for a given tip resistance may result from other facets of
⌬u2 Q0.95 soil behavior, such as sensitivity. Data from CPTUs in highly
= + 1.05 共su/␴⬘v0兲NC = 0.3, Ir = 500, Bq 艋 1 sensitive clays are explored to investigate a possible delineation
␴⬘v0 2 between “typical” and “highly sensitive” clays, such as Zone 1 of
共16c兲 Fig. 1.

1574 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008

Table 2. Characteristics of Clay Databases Used in Chart Development
Mayne Mayne Ramsey 共2002兲 clay Ramsey 共2002兲 clay
Database Database intact clays fissured clays ⌬u2 / ␴⬘v0 ⬎ 0 ⌬u2 / ␴⬘v0 ⬍ 0
subsets subsets n = 313 n = 48 n = 163 n = 35
␴⬘v0 共kPa兲 Minimum 6 23 1 2
Maximum 400 647 1,035 310
␮ ga 79 142 33 28
␴lna 0.83 0.79 1.7 1.3
PI Minimum 8 11 3 5
Maximum 152 55 95 76
␮ ga 38 25 23 18
␴lna 0.53 0.53 0.62 0.56
OCRb Minimum 1 1.6 1 4
Maximum 6.4 122 450 1,200
␮ ga 1.4 6.7 8.5 33
␴lna 0.41 1.2 1.2 1.2
a
Statistical parameters linked to the lognormal distribution; ␮g = geometric mean; ␴ln = standard deviation of natural log of the variable; median values were
typically within 10% of ␮g.
b
OCR for Ramsey 共2002兲 North Sea clays estimated from measured su / ␴⬘v0 and PI.

Piezocone Testing in Sensitive Clays have high values of Q due to overconsolidation ratio and friction
angle, among other soil properties, but tend to have higher Bq 共or
The cone penetration test results in very high strain levels sur-
⌬u2 / ␴⬘v0兲 values than similar clays with lower sensitivity. It is
rounding the probe. In strain softening soils, it is likely that the
judged that for highly sensitive clays:
peak strength value is exceeded. As soils strain soften, shear-
induced pore pressures are likely to be higher than those predicted ⌬u2 Q0.91
using the simple critical state model discussed by Mayne 共1992, ⬎ + 1.1 共18兲
␴⬘v0 1.5
2001兲, which does not include the influence of sensitivity in
its formulation. Fig. 5 plots profiles of “sensitive” clays in Eq. 共18兲 was developed to delineate the quick clays at Tiller and
Q – ⌬u2 / ␴⬘v0 space. Additional information about these sites is Gloucester from the Mayne intact clay database. Approximately
presented in Table 1.
The high sensitivity clays tend to plot near the line represented
by Eq. 共16c兲. The quick clays at Gloucester and Tiller tend to
have higher ⌬u2 / ␴⬘v0 at an equivalent Q than the clay of lower
sensitivity at Lierstranda, although there do not seem to be any
unique trends in location within Fig. 5 based solely on sensitivity.
Fig. 6 plots the same data for sensitive clays in Q – Bq space.
Trends in behavior for the clay soils are visually more apparent in
this representation, although the locations of sensitive clays are
significantly different from the previous recommendations of
Robertson 共1991兲 shown in Fig. 1 共Zone 1兲. Sensitive clays may

Table 3. Performance of Equations Used to Classify Intact Clay Soils

Based on Piezocone Parameters
Fraction correctly classified
Ramsey
Mayne intact 共2002兲 clay Both
Equation clays ⌬u2 / ␴⬘v0 ⬎ 0 databases
number Equation n = 313 n = 163 n = 476
共16a兲 ⌬u2 Q1.25 1.00 0.96 0.99
⬎ + 0.99
␴⬘v0 100

共16b兲 ⌬u2 Q0.95 0.96 0.50 0.80

⬎ + 1.05
␴⬘v0 5

共16c兲 ⌬u2 Q0.95 0.50 0.87 0.63

⬍ + 1.05
␴⬘v0 2

共18兲 ⌬u2 Q0.91 0.91 0.98 0.94

⬍ + 1.1
␴⬘v0 1.5 Fig. 5. Location of “sensitive clays” in Q – ⌬u2 / ␴⬘v0 space

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008 / 1575

 Fig. 6. Location of “sensitive clays” in Q – Bq space

90% of the data points in the Mayne intact clay database

have ⌬u2 / ␴⬘v0 values lower than that expressed using Eq. 共18兲
共Table 3兲, whereas 90% of the data from Tiller and Gloucester
have ⌬u2 / ␴⬘v0 values higher than Eq. 共18兲. It is noted that clays of
high sensitivity, such as Lierstranda in Figs. 5 and 6, may plot
with ⌬u2 / ␴⬘v0 values less than Eq. 共18兲, and the trend is not in-
tended to exclude clays as sensitive, but rather highlight soils
which may have a highly collapsible structure. It is also noted that
soils with high sensitivity may have Bq values in excess of one.
The line representing Bq of one in Fig. 5, as well as in other
figures in this paper, is not intended to represent a boundary, but
is included as a link between Q – ⌬u2 / ␴⬘v0 and Q – Bq charts.

Partial Consolidation during Penetration Testing

in Clays
As the degree of pore pressure dissipation during penetration be-
comes greater, increases in cone tip resistance are generally re-
flecting by lower excess penetration pore pressures 共e.g.,
Campanella et al. 1983; Randolph 2004兲. These trends in behav-
ior differ from those resulting from increasing OCR in clays,
which typically results in an increase in both Q and ⌬u2 / ␴⬘v0.
Using the concept of a normalized velocity as expressed in
Eq. 共10兲, changes in cone velocity can conceptually be related to
changes in coefficient of consolidation, or soil type 共e.g.,Hight
and Leroueil 2003; Randolph 2004兲. A series of variable rate pen-
etration tests, or “twitch” tests 共House et al. 2001兲, were per-
formed at the Burswood soft clay site in Perth, Western Australia,
to explore the influence of normalized cone velocity on piezocone
Fig. 7. Influence of cone velocity of piezocone parameters at
parameters.
Burswood, Western Australia
Two piezocone soundings with twitch tests were performed at

1576 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008


Fig. 8. Influence of partial consolidation from variable rate penetra-
tion tests on evaluation of soil behavior type in Q – ⌬u2 / ␴⬘v0 space
the Burswood site, with data summarized in Fig. 7. During pen-
etration between depths of 6.1 and 7.1 m, the cone velocity
was successively halved 共i.e., 20 to 10 to 5 mm/ s, etc.兲 for each
of 10 to 11 steps, before reverting to the standard 20 mm/ s cone
velocity. The cone was advanced by at least two diameters at each
step to ensure a steady cone tip resistance was reached. Cone
velocities as low as 0.01 mm/ s were achieved, or a normalized
velocity as low as four to six based on the laboratory coefficient
of consolidation of 4 – 5 m2 / year and a ratio of ␭ to ␬ of ap-
proximately five for this depth range. Additional details on the
Burswood site and penetration twitch testing are discussed by
Chung et al. 共2006兲.
The twitch tests help to identify two facets of clay behavior
during penetration: 共1兲 viscous rate effects; and 共2兲 partial con-
solidation. This results in a “saddle” behavior around a minimum
Q value, with higher cone tip resistance at higher velocity due to
viscous rate effects and higher cone tip resistance at lower veloci-
ties due to partial consolidation. Viscous rate effects also tend to
increase ⌬u2 / ␴⬘v0 proportionally to increases in Q in this essen-
tially normally consolidated clay, whereas partial consolidation
results in a reduction in penetration pore pressures at low veloci-
ties. This results in relatively constant Bq during undrained pen-
etration at this site, with Bq dropping more rapidly than ⌬u2 / ␴⬘v0
during partial consolidation due to the compounding effects of
⌬u2 decreasing and qcnet increasing.
Fig. 8 presents the influence of partial consolidation during
penetration on soil classification using piezocone data. Variable
rate penetration test data from the Burswood field site are com-
bined with centrifuge data in normally and overconsolidated ka-
JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008 / 1577
Fig. 9. Location of silts at Värgverket, Sweden on soil classification
charts
olin discussed by Schneider et al. 共2007兲. The “direction” of
trends relating to the influence of OCR and partial consolidation
are significantly different in Q – ⌬u2 / ␴⬘v0 space, but are similar for
both the centrifuge and field data. Even when penetration occurs
under conditions of partial consolidation, data may lie to the right
of Eq. 共16a兲, which could result in CPTU classification as a
“clay” soil. Fig. 8 highlights that not only the location on the
Q – ⌬u2 / ␴⬘v0 chart is important, but also the “direction of move-
ment” within the chart. As the OCR of clays tends to reduce with
depth, most overconsolidated clay soils will also show Q and
⌬u2 / ␴⬘v0 decreasing with depth. Soils in which penetration is par-
tially drained often tend to vary in material type and coefficient of
consolidation, which tends to result in increases in Q and de-
creases in ⌬u2 / ␴⬘v0 共or Bq兲 with increases in coefficient of con-
solidation 共e.g., Hight et al. 1994; Watson and Humpheson 2005兲.
Piezocone Testing in Silts
Delineation of silty soils by piezocone as well as engineering
analyses in these materials is complicated by effects of partial
consolidation, with drainage conditions during penetration testing
often differing from those of the design application. Consider-
ations for analyses of silts and soils that are partially consolidated
during in situ testing include the following:
• If essentially all penetration pore pressures are dissipated
within 5 min of stopping penetration, then loading rates for
normal onshore construction practices will occur under free
draining conditions 共Larsson 1997兲. Conversely, if significant
excess pore pressures are generated during penetration, qcnet
does not reflect the fully drained soil response and direct cor-
relations between penetration testing and engineering behavior
may be less reliable.

Fig. 10. Location of silts at Halsen, Norway on soil classification
chart
• For offshore foundation solutions, such as spudcans and shal-
low foundations, normalized rates of loading will push silty
soils into the undrained region, even though cone tests may be
drained or partially drained. The resulting differences between
penetrometer resistance and foundation performance need
careful consideration 共Erbrich 2005兲.
• Evaluation of the coefficient of consolidation from piezocone
dissipation tests in soils that exhibit partial consolidation dur-
ing installation may underestimate ch. Theoretical studies
共Silva et al. 2006兲 and experimental evidence 共Schneider et al.
2007兲 have shown that t50 increases compared to the undrained
case due to partial consolidation during penetration.
Figs. 9 and 10 plot in log–log Q – ⌬u2 / ␴⬘v0 space the location
of layers from the Vägverket and Halsen silt sites, respectively.
The soils generally plot between the line that represents Eq. 共16b兲
and ⌬u2 / ␴⬘v0 near zero. All dissipation tests at Vägverket essen-
tially reached 90% dissipation within 5 min, except for the test at
4.5 m depth. Longer consolidation times for this layer are re-
flected by a position to the right of Eq. 共16b兲. Dissipation tests
in the Halsen silt typically took longer than 5 min to reach 90%
pore pressure dissipation, and laboratory consolidation tests in-
dicated cv values between 300 and 3500 m2 / year 共V of 75 to 7兲.
As CPTU penetration occurs under conditions of partial consoli-
dation for cv values between approximately 350 m2 / year and
1 ⫻ 105 m2 / year, these two sites are good examples of transitional
soils.
The difficulties of separating heavily overconsolidated clays
and silty soils based on Q and ⌬u2 / ␴⬘v0 can be seen as both silts
in Figs. 9 and 10 plot between Eqs. 共16a兲 and 共16b兲 for portions
of the profiles. This is due to the potential for low octahedral
pore pressures in both material types. In the overconsolidated
clays this tends to result from reduction in radial total stress 共and
1578 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008
pore pressure兲 as the soil flows around the cone tip as well as
negative shear-induced pore pressures, whereas in silts ⌬uoct
tends to decrease more from the effects of partial consolidation.
This influence of partial consolidation on the location of data
within soil classification charts, as shown in Figs. 8–10, appears
consistent for velocity effects in clays and coefficient of con-
solidation effects in silts. The direction of movement within the
charts can help to separate silty soils from heavily overcon-
solidated clays. Fig. 8 illustrated that in Q – ⌬u2 / ␴⬘v0 space, CPTU
data tend to “move” up and to the right as OCR increases during
undrained penetration, whereas data tend to move up and to
the left as penetration approaches drained conditions. If heavily
overconsolidated clays are expected, separation of silts and
clays using Eq. 共16a兲 may fit the data better, or if the region
contains more silty soils, Eq. 共16b兲 may be more appropriate.
As with most site characterization activities, knowledge of
the local geology, supplemental boreholes, dissipation testing, as
well as variable rate penetration tests will aid in the engineering
process.
Piezocone Testing in Sands
The observation of essentially hydrostatic pore pressure is a good
indication of drained penetration. Drained penetration will typi-
cally result in higher Q values than undrained or partially drained
penetration for soils at an equivalent state, but in looser sands
there may be an overlap of soil types. Despite cone tip resistance
measured in calibration chamber testing of young unaged sands
normalizing approximately to 冑␴h0 ⬘ 共e.g., Parkin 1988兲, or 冑␴⬘v0
in normally consolidated sand, normalization to ␴⬘v0 is used for
all materials within these classification charts for a number of
reasons:
1. A uniform approach is presented where no iteration is re-
quired to decide on appropriate normalization in different
soil types.
2. Even if an iterative solution is used to assess appropriate
normalization based on soil type, that normalization is still a
simplification of actual behavior and prone to error.
3. Potential errors in classification would only occur in loose to
very loose normally consolidated sands at great depths.
Loose to very loose sands arguably have a normalization
exponent 关n, as in 共␴⬘v0/Pref兲n兴 between 0.8 and 1 共e.g., Olsen
and Mitchell 1995; Moss et al. 2006b兲, rather than the com-
monly assumed value of 0.5 for sands. This trend in behavior
minimizes the potential for normalization errors when using
Q 共i.e., n = 1兲 for classification of sandy soils.
4. Increasing K0 will also increase Q in sands at equivalent
relative densities, which will not lead to errors in soil classi-
fication.
5. Conversion between Q – Bq space and Q – ⌬u2 / ␴⬘v0 space will
not be stress dependent.
It is noted that while the above-mentioned reasons can be ac-
counted for in analyses for Q – Bq or Q – ⌬u2 / ␴⬘v0 classification
charts, linear stress normalization may not be valid for all soil
types / states when using friction ratio 共Q – F兲 based soil classifi-
cation charts.
As ⌬u2 / ␴⬘v0 and Bq are approximately zero in drained sands,
development of the delineation of sands from silts and clays is
strongly based on the minimum value of Q that is expected.
A minimum value of Q = 20 is selected based on a review of
field data. Theoretical trends are compared to two example cases
in a very loose sand 共Drammen兲 and a dense to very dense sand
共Euripides兲 in Fig. 11. The program CONPOINT 共Salgado 1993;
Fig. 11. Profiles of Q with depth in siliceous sands and silty sands
Salgado et al. 1997兲, with properties for the Ticino reference sand,
was used to calculate profiles of Q with depth for comparison
to field data. The reduction in Q with depth 共i.e., stress exponent
n ⬍ 1兲 is more significant for the very dense sand as compared to
the loose sand, and trends based on the theoretical model matches
trends in field data well. Additionally, sands with a fines content
less than 50% 共which may or may not have liquefied兲 from the
Moss et al. 共2006a兲 liquefaction database are included in Fig. 11
for additional examples of CPT qcnet behavior at shallow depths.
The liquefaction database is used as it is expected to contain some
of the loosest sands, and thus lowest expected Q values. Although
shallow, none of these data have Q values less than 20, and tend
Fig. 12. Profiles of Q with depth in compressible micaceous and
calcareous sands
JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008 / 1579
Fig. 13. Location of sandy soils on soil classification charts
to fall between the Dr limits for Ticino sand. Still, 10 共out of the
200兲 data points within the full database 共not limited to fines
content less than 50%兲 of Moss et al. 共2006a兲 have Q values
below 20, but all data with Q less than 20 that have soils descrip-
tions are classified as low plasticity silts 共ML兲 or silty sands to
sandy silts 共SM/ML兲 with fines content greater than 50%. It is
likely that penetration pore pressures are influencing cone tip re-
sistance in these silty soils 共e.g., D. Elsworth and M. Fitzgerald
personal communication, 2007兲, although no u2 data are avail-
able. For clean siliceous sands, field data and theoretical studies
support the selected minimum Q value of 20.
Normalized cone tip resistance is known to be lower in more
compressible sands at the same relative density and stress level
共e.g., Robertson and Campanella 1983兲. Fig. 12 illustrates trends
of Q with depth for a compressible micaceous sand 共Jamuna
Bridge兲 and a compressible calcareous sand 共Overland Corner兲.
The bias of Q decreasing with depth 共which would occur if the
stress exponent n should be less than 1兲 decreases with depth and
appears to be minimal below 40 m. Both the micaceous and cal-
careous sands have Q values in excess of the proposed minimum
value of 20. Minimum Q values of approximately ten have been
observed for calcareous silty sand and sandy silt at Mossel Bay,
South Africa 共Ebelhar et al. 1988兲 as well as for some calcareous
silty sand sites offshore Western Australia. Most observed occur-
rences of Q values below 20 in sands occur for CPTs performed
in highly compressible or silty deposits where pore pressures may
influence tip resistance, or at depths greater than 100 m.
The delineation of essentially drained sands from clays and
transitional soil is based on the previously discussed logic and
uses the following equation for Q greater than the selected mini-
mum value of 20:
Table 4. Performance of Equations Used to Classify Sandy Soils Based
on Piezocone Parameters
Does not
Depth Classifies classify Fraction
Site 共m兲 as sand as sand sand
Euripides 23–45 1,053 55 0.95
Drammen 3.5–27 1,085 100 0.92
Jamuna Bridge 6–60 2,540 165 0.94
Fittja Straits 2–27 820 422 0.66
All four sites 2–60 5,498 692 0.89

冋
max − 0.5 ln 冉 冊 册
Q
20
,− 1 艋
⌬u2
␴⬘v0
冋 冉 冊册
艋 min 0.5 ln
Q
20
,1

共19兲
In sharp contrast with the maximum ⌬u2 / ␴⬘v0 value of one used in
Eq. 共19兲 ⌬u2 / ␴⬘v0 values on the order of 25 may be observed for
soils plotting in Zones 6 and 7, sands and gravels, based on Fig. 1.
This does not imply that sandy soils may have very high normal-
ized excess pore pressure values, but is a function of chart distor-
tion resulting from qcnet being incorporated in both the axes of the
Bq charts.
CPTU data from the Drammen, Euripides, and Jamuna Bridge
siliceous and micaceous sands are compared to data for a silty
sand at Fittja Straits and Eq. 共19兲 in Fig. 13. For sands, the lower
Q value of 20 tends to capture loose siliceous sands and mica-
ceous sands, as well as silty sands at Fittja Straits. The transition Fig. 15. Location of select transitional soils in Q – Bq space
from ⌬u2 / ␴⬘v0 of 0 to 1 roughly parallels Eq. 共16a兲, and tends to
adequately separate the nearly drained silts from the predomi-
nantly sandy soils. Although it appears that a number of data points. Of the data points which do not classify within the bounds
points in Fig. 13 plot outside of the “essentially drained sand” of Eq. 共19兲, over half are from the silty sand at Fittja Straits.
zone, Table 4 indicates that this occurs for only 11% of the 5500 Penetration pore pressures for points that fall outside the essen-
tially drained sand zone indicate partial consolidation during pen-
etration, and correlations between qcnet and drained engineering
behavior are considered less reliable. Additional variable rate pen-
etration tests in silty and clayey sands are needed to provide in-
formation on the transition from drained to partially consolidated
penetration in these material types.

Negative u2 Penetration Pore Pressures

Fig. 14. Location of select transitional soils in Q – ⌬u2 / ␴⬘v0 space
Significant negative penetration pore pressures have been re-
corded at the shoulder 共u2兲 location during piezocone penetration
in overconsolidated clays, stiff glacial clays, alluvial silty sands,
residual silty sands and sandy silts, among other soil types 共e.g.,
Finke and Mayne 2002兲. While these are actual measurements of
penetration behavior, the measurements are often limited by the
performance of the piezocone pore pressure measurement system
due to cavitation 共Jefferies and Davies 1991; Robertson 1991兲.
Fig. 14 plots data from a silty marine sand 共SFOBB兲, a residual
soil 共Opelika兲, and a database of fissured clays updated from that
discussed by Chen and Mayne 共1996兲. These terrestrial deposits,
although having significantly different engineering behavior, tend
to plot in very similar areas of the classification charts. Additional
data from North Sea clays 共Ramsey 2002兲 are added to the data
sets evaluated for Fig. 15. The offshore soils tend to have lower
共larger negative兲 Bq values due to the influence of the larger u0
values which occurs in offshore testing 共and thus greater −⌬u2
values at cavitation兲. The relative magnitude of these negative
values is primarily a function of instrument behavior rather than
soil response, but the fact that negative ⌬u2 values are recorded is
quite significant when classifying soil by piezocone.

1580 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008

Fig. 16. Differences in dissipation behavior for soils which plot in
similar locations on the soil behavior type chart

When observing negative ⌬u2 penetration pore pressures in

unfamiliar situations, it is highly recommended to perform dissi-
pation tests. If the soil is a silty sand or sandy silt, full dissipation
to u0 should occur within 5 min. As shown in Fig. 16, pore pres-
sures during dissipation tests in heavily overconsolidated clays
may start as negative, increase past hydrostatic, and then start to
decay monotonically. This behavior results from very large nega-
tive shear-induced penetration pore pressures adjacent to the cone
shaft, which dissipate more rapidly than the positive octahedral Fig. 17. Location of “transitional” North Sea soils in Q – ⌬u2 / ␴⬘v0
pore pressures located away from the cone shaft 共e.g., Bond and space
Jardine 1991; Burns and Mayne 1998兲.

silty fine sands and slightly silty fine to medium sands to classify
Transitional Soils of the North Sea largely as transitional soils or clay deposits is unexpected. The
The above presented discussion has been based primarily on rela- data in Fig. 17 are from different sites and are summarized as
tively uniform graded soils, i.e., sands, silts, and clays. Mixed soil single points. As previously mentioned, and, for example, illus-
types may frequently be encountered during site investigations, trated by Hight et al. 共1994兲 for a clayey sand from the North Sea
and it is of interest to explore if these materials tend to lie in 共using Q – Bq space兲, trends in direction of movement in
typical locations in Q – ⌬u2 / ␴⬘v0 space. The database of North Sea Q – ⌬u2 / ␴⬘v0 soil classification charts for profiles of CPTU param-
soils discussed by Ramsey 共2002兲 is used to explore general pi- eters with depth may be of more use in site characterization than
ezocone behavior of clayey sands, silty sands, silty sands with evaluation of individual points. Additional research in sand mix-
clay, clayey sands with silt, slightly silty fine sands, and slightly tures, such as variable rate penetration testing, is warranted to
silty fine to medium sands, as shown in Fig. 17. One would not assess if these measurements of penetration pore pressures indi-
expect a large difference between slightly silty fine sands and
slightly silty fine to medium sands, but these two soil types are
separated due to differences in observed behavior in this database. Table 5. Classification of North Sea “Transitional” Soils Used on
It is noted that these shallow sand mixtures have the potential for Piezocone Parameters
a very high overconsolidation ratio 共and thus K0兲, with ␴⬘vy on the Fraction as Fraction as
order of 1 – 2 MPa due to glaciation. Specific trends in location essentially Fraction as SANDS or
within Q – ⌬u2 / ␴⬘v0 space are not apparent, with most material Data drained transitional transitional
types plotting in both essentially drained sands and transitional Category points SANDS soils soils
soils. Locations of each soil type are explored further in Table 5. Clayey SAND 5 0.0 1.0 1.0
Most soil types are split equally between essentially drained Silty SAND 20 0.65 0.30 0.95
sands and transitional soils. Clayey sands and silty sands with
Silty SAND w/clay 11 0.0 0.73 0.73
clay are plotted as predominantly transitional soils, whereas
Clayey SAND w/silt 7 0.57 0.43 1.0
clayey sands with silt tended to plot predominantly as essentially
Slightly silty fn SAND 31 0.35 0.55 0.90
drained sands. These differences are likely due to variation in
coefficient of consolidation and yield stress ratio of the individual Slightly silty fn to 9 0.33 0.11 0.44
med SAND
sites, as u2 pore pressures are very sensitive to these effects in
Total 83 0.37 0.48 0.86
silty materials 共e.g., Schneider et al. 2007兲. The trend of slightly

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008 / 1581


Fig. 18. Newly proposed piezocone soil classification chart in
various plotting formats
1582 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008
Fig. 19. Comparison of newly proposed chart 共dark black lines兲 with
Robertson 共1991兲 chart 共light gray lines兲
cate a significant degree of partial consolidation during penetra-
tion, which may influence cone resistance and thus reliability of
design correlations.
General Tip Resistance—Pore Pressure-Based Soil
Classification Charts
General soil classification charts were developed using parametric
studies of analytical solutions, field data, and judgment based
on the previous discussions. Fig. 18 shows the recommended
soil classification charts based on normalized piezocone param-
eters which have evolved from this study. It is noted that the
“zones” in these three charts are exactly the same, but the plots
are shown in different formats: 共1兲 log–log Q – ⌬u2 / ␴⬘v0 space; 共2兲
semilog Q – ⌬u2 / ␴⬘v0 space; and 共3兲 semilog Q – Bq space. These
formats are best used in the cases of: 共1兲 clays, clayey silts, silts,
sandy silts, and sands with no negative penetration pore pres-
sures; 共2兲 sands and transitional soils with small negative excess
penetration pore pressures; and 共3兲 clay soils with large negative
excess penetration pore pressures, respectively. The newly recom-
mended classification charts are overlain on the recommendations
of Robertson 共1991兲 in Q – Bq space in Fig. 19. These proposed
charts and data at heavily overconsolidated North Sea clay sites
共Fig. 1兲 are significantly different from the recommendations of
Robertson 共1991兲, highlighting the need for design charts based
on a combination of analytical studies and field data.
Table 6 summarizes equations used for this assessment of soil
behavior type using piezocone data. A simple algorithm, which
checks if the soil is essentially drained or essentially undrained
Table 6. Description of Equations for Trends in New Soil Classification Charta
Zone Equation Q range Description/notes
1a ⌬u2 Q 1.25
Q艌1 Below this line and above Eq. 共16b兲 data tend to indicate
⬎ + 0.99
␴⬘v0 100 penetration in apparently “low Ir” CLAYS and SILTS.
Penetration may occur under conditions of partial
consolidation or in soils with high unloading stiffness/large
negative local shear induced pore pressures.

1b ⌬u2 Q0.95 Q艌1 Below this line and above Eq. 共16c兲 data tend to represent
⬎ + 1.05
␴⬘v0 5 undrained penetration in CLAYS.

1c ⌬u2 Q0.91 Q艌1 Piezocone data tend to indicate CLAYS with a higher than
⬎ + 1.1
␴⬘v0 1.5 average sensitivity. This zone is not intended to exclude soils
as possibly being sensitive. Figs. 5 and 6 illustrate clays with
high sensitivity also plotting in Zone 1b.

2
关
max −0.5 ln 共 兲 兴
Q
20
, −1 艋
⌬u2
␴⬘v0
and
⌬u2
␴ v0
⬘
关 共 兲兴
艋 min 0.5 ln
Q
20
,1
Q ⬎ 20 Between these trend lines data tend to indicate essentially
drained penetration in SANDS, GRAVELS, and silty SANDS.
The existence of u2 pore pressures during drained penetration
is not well understood but could result from local soil
behavior adjacent to the filter element as well as filter
compression induced by high horizontal stresses in these
materials. Penetration in very loose or compressible sands at
great depths may result in Q below 20 during drained
penetration, and misclassification as a “Transitional Soil.”

3 Not as previously described All “Transitional” soils. A wide variety of soils fall within this
range, and soil classification should be augmented using CPT
friction ratio 共F兲, dissipation tests, variable rate penetration
testing, and/or sampling. It is important to recognize that
typical correlations based on drained or undrained penetration
likely have lower levels of reliability if applied to soils in
Zone 3.
a
Conversion between Bq and ⌬u2 / ␴⬘v0 by means of Eq. 共9兲.

during penetration will highlight transitional soils, which likely
require dissipation or variable rate penetration testing to supple-
ment standard piezocone parameters. Soils that plot outside Zones
1b or 2 will likely have lower levels of reliability if applying
standard design correlations intended for application to drained or
undrained CPTU parameters. When analyzing soil behavior for
transitional soils it is necessary to assess the influence of normal-
ized velocity on Q, whether through dissipation tests or variable
rate penetration testing.
Tip Resistance—Friction Ratio-Based Soil
Classification Charts
The soil classification charts based on Q and ⌬u2 / ␴⬘v0 developed
in this paper were primarily derived to aid in separating whether
penetration is drained, undrained, or partially drained. Provided
that a soil layer is below the water table and that the piezocone
filter element is properly saturated and does not lose saturation
during penetration, Q – ⌬u2 / ␴⬘v0 charts can be relatively reliable in
this effect when positive pore pressures are measured. In onshore
geotechnical practice, soil classification using the friction ratio
关F共%兲 = f t / qcnet ⫻ 100兴 often receives more use than pore
pressure-based classification charts. One reason for this trend
results from lack of pore pressure readings above the water
table, and a second is due to difficulties in maintaining element
saturation in stiff, dilatant deposits or when passing through
partially saturated material above the water table. Still, the ten-
dency for both OCR 共state兲 and partial consolidation to increase
the normalized penetration resistance, Q, leads to the potential
for overlap of soil types in Q – F classification charts and some
uncertainty in assessment of whether penetration is drained,
undrained, or partially drained. As compared to the undrained
case, decreases in friction ratio will occur due to reductions in
radial total stress behind the cone tip. This reduction is generally
more significant during drained, or partially drained, penetration
due to volume change 共i.e., cavity contraction兲, but is also influ-
enced by cementation and soil sensitivity during undrained pen-
etration 共e.g., Lehane 1992; Karlsrud et al. 1993兲. During
undrained penetration, the combination of high OCR and sensi-
tivity, or the presence of cementation, tends to lead to high Q
values and low F values, which may lead to difficulties in the
separation of OCR and partial consolidation using Q – F charts.
Variable rate penetration tests, which include pore pressure and
sleeve friction measurements, are needed to aid in analysis of the
separation of partial consolidation and OCR when using CPT
friction ratio.
Conclusions
Soil behavior is generally controlled by degree of pore pressure
dissipation during loading, stress level at failure, and initial yield
stress ratio. These features of behavior will lead to differing soil
response during piezocone penetration, particularly the relative
values of cone tip resistance and penetration pore pressures.
When using piezocone data to classify soil type, it is concluded
that:
• Depth bias, which is not a function of soil behavior, will occur
if qcnet and ⌬u2 are not normalized. Normalization to initial

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008 / 1583

 vertical effective stress is the most practical option, resulting
in trends as a function of overconsolidation ratio. These trends
are a function of soil behavior and can be assessed analytically
for development of generalized soil classification charts.
• Soil classification in Q – Bq space and Q – ⌬u2 / ␴⬘v0 space are
analogous, although different aspects of behavior are high-
lighted depending on the presentation. Q – B2 space tends to
provide additional detail for separating clay behavior, whereas
Q – ⌬u2 / ␴⬘v0 space is more useful for identifying sands, silts
and clays, or soil type.
• Newly developed soil classification charts in this paper based
on Q and Bq, or the analogous Q and ⌬u2 / ␴⬘v0 versions of the
charts, are significantly different than corresponding charts
typically used in practice. These differences are not solely due
to differences in databases used for calibration, but theoretical
considerations that affect cone tip resistance and penetration
pore pressures. Databases agree well with the analytical solu-
tions discussed, and provide a new framework for evaluation
of soil type by piezocone.
• Octahedral pore pressures tend to dominate the ⌬u2 response
except at low “apparent” values of rigidity index 共Ir兲. Low
apparent Ir may result from large reductions in radial total
stress behind the cone tip, large negative shear induced pore
pressures adjacent to the cone shaft, or partial consolidation
during penetration, resulting in an overlap of regions within
the classification charts.
• As large reductions in radial total stress behind the cone tip
tend to occur in highly preconsolidated clays, and OCR tends
to reduce with depth in these materials due to a relatively
constant ␴⬘vy, the direction of movement within classification
charts due to changes in depth is useful for separating over-
consolidated clays from silty soils.
For potential areas of overlap in classification charts, informa-
tion in addition to qcnet and ⌬u2 are necessary for assessment of
soil type from piezocone tests. This information may come from
knowledge of local geology, drilling and sampling, CPT friction
ratio, piezocone dissipation tests, or variable rate penetration
tests. Variable rate penetration tests provide additional levels of
information, which cannot be achieved through dissipation test-
ing, and additional experience with this testing procedure in a
variety of soil types can help to improve on the analyses dis-
cussed in this paper.
Acknowledgments
The first writer would like to thank Rolf Sandven of NTH for data
and discussions of the Halsen silt site, as well as Rolf Larsson
of SGI for discussion on piezocone tests at silt sites in Sweden.
Professor Rodrigo Salgado of Purdue Univ. is thanked for use
of the program CONPOINT. Support for the first writer from
the Australian Research Council 共ARC兲, an International Post-
graduate Research Scholarship, and a University Postgraduate
Awards from the University of Western Australia is gratefully
acknowledged.
References
Amundsen, T., Lunne, T., and Christophersen, H. P. 共1985兲. “Advanced
deep-water soil investigation at the Troll East field.” Advances in
underwater technology, offshore site investigation, Vol. 3, Grahan &
Trotman, London, 165–186.
1584 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008
ASTM. 共2000兲. “Standard test method for performing electronic friction
cone and piezocone penetration testing of soils.” D5778-95, ASTM
book of standards, Vol. 04–09, Philadelphia.
Axelsson, G. 共2000兲. “Long term setup of driven piles in sand.” Ph.D.
thesis, Royal Institute of Technology, Stockholm, Sweden.
Baligh, M. M. 共1986兲. “Undrained deep penetration. II: Pore pressures.”
Geotechnique, 36共4兲, 487–501.
Bond, A. J., and Jardine, R. J. 共1991兲. “Effects of installing displacement
piles in a high OCR clay.” Geotechnique, 41共3兲, 341–363.
Burns, S. E., and Mayne, P. W. 共1998兲. “Monotonic and dilatory pore-
water pressures during piezocone dissipation tests in clay.” Can. Geo-
tech. J., 35共6兲, 1063–1073.
Campanella, R. G., Robertson, P. K., and Gillespie, D. 共1983兲. “Cone
penetration testing in deltaic soils.” Can. Geotech. J., 20共1兲, 23–35.
Chen, B. S. Y., and Mayne, P. W. 共1994兲. “Profiling the overconsolidation
ratio of clays by piezocone tests.” Rep. No. GIT-CEEGEO-94-1,
School of Civil and Environmental Engineering, Georgia Institute of
Technology, Atlanta.
Chen, B. S. Y., and Mayne, P. W. 共1996兲. “Statistical relationships be-
tween piezocone measurements and stress history of clays.” Can.
Geotech. J., 33共3兲, 488–498.
Chung, S. F., Randolph, M. F., and Schneider, J. A. 共2006兲. “Effect of
penetration rate on penetrometer resistance.” J. Geotech. Geoenviron.
Eng., 132共9兲, 1188–1196.
Ebelhar, R. J., Young, A. G., and Stieben, G. P. 共1988兲. “Cone penetrom-
eter and conductor pullout tests in carbonate soils offshore Africa.”
Proc., Int. Conf. on Calcareous Sediments, Balkema, Rotterdam, The
Netherlands, 155–163.
Erbrich, C. T. 共2005兲. “Australian frontiers—Spudcans on the edge.”
Proc., Int. Symp. on Frontiers Offshore Geomechanics, Taylor &
Francis, London, 49–74.
Fahey, M., and Lee Goh, A. 共1995兲. “A comparison of pressuremeter and
piezocone methods of determining the coefficient of consolidation.”
Proc., 4th Int. Symp. on the Pressuremeter and Its New Avenues,
Telford, London, 153–160.
Finke, K. A., and Mayne, P. W. 共2002兲. “Closure to piezocone penetration
testing in Atlantic Piedmont residuum.” J. Geotech. Geoenviron. Eng.,
128共5兲, 443–446.
Finke, K. A., Mayne, P. W., and Klopp, R. A. 共2001兲. “Piezocone testing
in Atlantic Piedmont residuum.” J. Geotech. Geoenviron. Eng.,
127共1兲, 48–54.
Finnie, I. M. S. 共1993兲. “Performance of shallow foundations in calcare-
ous soil.” Ph.D. thesis, Dept. of Civil Engineering, Univ. of Western
Australia, Crawley, Western Australia.
Finnie, I. M. S., and Randolph, M. F. 共1994兲. “Punch-through and lique-
faction induced failure of shallow foundations on calcareous sedi-
ments.” Proc., Int. Conf. on Behavior of Offshore Structures, BOSS
’94, Elsevier Science Ltd., Tarrytown, N.Y., 217–230.
Fugro Engineers, B. V. 共Fugro兲. 共1995兲. “Factual report on reduced scale
pile load tests, Jamuna Bridge, Bangladesh.” Rep. No. K2380-206,
Leidschendam, The Netherlands.
Hight, D. W., Georgiannou, V. N., and Ford, C. J. 共1994兲. “Characteriza-
tion of clayey sands.” Proc., Int. Conf. on Behavior of Offshore Struc-
tures, BOSS ’94, Elsevier Science Ltd., Tarrytown, N.Y., 321–340.
Hight, D. W., and Leroueil, S. 共2003兲. “Characterisation of soils for en-
gineering purposes.” Characterisation and engineering properties of
natural soils, Swets & Zeitlinger, Lisse, The Netherlands, 255–360.
House, A. R., Oliveira, J. R. M. S., and Randolph, M. F. 共2001兲. “Evalu-
ating the coefficient of consolidation using penetration tests.” Int. J.
Phys. Modell. Geotech., 1共3兲, 17–25.
International Society of Soil Mechanics and Foundation Engineering
共ISSMFE兲. 共1997兲. International reference test procedure for cone
penetration test (CPT), Linköping, Sweden.
Jamiolkowski, M., Ladd, C. C., Germaine, J., and Lancellotta, R. 共1985兲.
“New developments in field and lab testing of soils.” Proc., 11th Int.
Conf. on Soil Mechanics and Foundation Engineering, Vol. 1,
Balkema, Rotterdam, The Netherlands, 57–154.
Jefferies, M. G., and Davies, M. P. 共1991兲. “Soil classification by the cone
penetration test: Discussion.” Can. Geotech. J., 28共1兲, 173–176.
Jones, G., and Rust, E. 共1982兲. “Piezometer penetration testing.” Proc.,
2nd European Symp. on Penetration Testing, Vol. 2, Balkema, Rotter-
dam, The Netherlands, 607–613.
Karlsrud, K., Nowacki, F., and Kalsnes, B. 共1993兲. “Response in soft clay
and silt deposits to static and cyclic loading based on recent instru-
mented pile load tests.” Offshore site investigation and foundation
behavior, Society for Underwater Technology, London, 549–584.
Keaveny, J. M., and Mitchell, J. K. 共1986兲. “Strength of fine-grained
soils using the piezocone.” In-Situ ’86, GSP 6, ASCE, Reston, Va.,
668–685.
Klotz, E. U., and Coop, M. R. 共2001兲. “An investigation of the effect of
soil state on the capacity of driven piles in sands.” Geotechnique,
51共9兲, 733–751.
Kolk, H. J., Baaijens, A. E., and Vergobi, P. 共2005兲. “Results of axial
load tests on pipe piles in very dense sands: The EURIPIDES JIP.”
Proc., Int. Symp. Frontiers Offshore Geomechanics ISFOG, Taylor &
Francis, London, 661–667.
Konrad, J. M. 共1998兲. “Sand state from cone penetrometer tests: A
framework considering grain crushing stress.” Geotechnique, 48共2兲,
201–215.
Konrad, J.-M., and Law, K. T. 共1987兲. “Preconsolidation pressure from
piezocone tests in marine clays.” Geotechnique, 37共2兲, 177–190.
Kulhawy, F. H., and Mayne, P. W. 共1990兲. “Manual on estimating soil
properties for foundation design.” Rep. No. EL-6800, Electric Power
Research Institute, Palo Alto, Calif.
Lacasse, S., and Lunne, T. 共1982兲. “Penetration testing in two Norwegian
clays.” Penetration testing, Vol. 2, Balkema, Rotterdam, The Nether-
lands, 607–613.
Larsson, R. 共1997兲. “Investigations and load tests in silty soils.” Rep. No.
54, Swedish Geotechnical Institute, Linköping, Sweden.
Lehane, B. M. 共1992兲. “Experimental investigations of pile behavior
using instrumented field piles.” Ph.D. thesis, Imperial College,
London.
Lehane, B. M., and Jardine, R. J. 共1994兲. “Displacement-pile behavior in
a soft marine clay.” Can. Geotech. J., 31共2兲, 181–191.
Lehane, B. M., O’Loughlin, C. D., Gaudin, C., and Randolph, M. F.
共2008兲. “Rate effects on penetrometer resistance in kaolin.” Geotech-
nique, in press.
Leroueil, S., Demers, D., La Rochelle, P., Martel, G., and Virely, D.
共1995兲. “Practical applications of the piezocone in Champlain sea
clays.” Proc., Int. Symp. on Cone Penetration Testing, CPT-95, SGI,
Linköping, Sweden, 515–522.
Lu, Q., Randolph, M. F., Hu, Y., and Bugarski, I. C. 共2004兲. “A nu-
merical study of cone penetration in clay.” Geotechnique, 54共4兲,
257–267.
Lunne, T., Eidsmoen, T. E., Powell, J. J. M., and Quarterman, R. S. T.
共1986兲. “Piezocone testing in overconsolidated clays.” Proc., 39th Ca-
nadian Geotechnical Conf., CGS, Ontario, Canada, 209–218.
Lunne, T., and Lacasse, S. 共1999兲. “Geotechnical characteristics of low
plasticity Drammen clay.” Characterization of soft marine clays,
Balkema, Rotterdam, The Netherlands, 33–56.
Lunne, T., Robertson, P. K., and Powell, J. J. M. 共1997兲. Cone penetration
testing in geotechnical practice, Blakie Academic and Professional,
Melbourne, Australia.
Mayne, P. W. 共1992兲. “In situ determination of clay stress history
by piezocone model.” Predictive soil mechanics, Telford, London,
483–495.
Mayne, P. W. 共2001兲. “Stress-strain-strength-flow parameters from
enhanced in-situ tests.” Proc., Int. Conf. on In-Situ Measurement of
Soil Properties and Case Histories (In-Situ 2001), Parahyangan
Catholic University Publishers, Bandung, Indonesia, 27–47.
Mayne, P. W., and Bachus, R. C. 共1988兲. “Profiling OCR in clays by
piezocone.” Penetration testing 1988, Balkema, Rotterdam, The Neth-
erlands, 857–864.
Mayne, P. W., Brown, D., Vinson, J., Schneider, J. A., and Finke, K. A.
JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008 / 1585
共2000兲. “Site characterization of Piedmont residual soils at the Na-
tional Geotechnical Experimentation Site, Opelika, Alabama.”
National geotechnical experimentation sites, GSP 93, ASCE, Reston,
Va., 160–185.
Mayne, P. W., and Kulhawy, F. H. 共1982兲. “K0-OCR relationships in
soil.” J. Geotech. Engrg. Div., 108共GT6兲, 851–872.
Mayne, P. W., Kulhawy, F. H., and Kay, J. N. 共1990兲. “Observations on
the development of pore-water stresses during piezocone penetration
in clays.” Can. Geotech. J., 27共4兲, 418–428.
Mayne, P. W., Robertson, P. K., and Lunne, T. 共1998兲. “Clay stress history
evaluated from seismic piezocone tests.” Geotechnical site character-
ization, Balkema, Rotterdam, The Netherlands, 1113–1118.
McNeilan, T. W., and Bugno, W. T. 共1985兲. “Cone penetration test results
in offshore California silts.” Strength testing of marine sediments:
Laboratory and in situ test measurements, ASTM STP 833, ASTM,
Philadelphia, 55–71.
Moss, R. E. S. 共2003兲. “CPT-based probabilistic assessment of seismic
soil liquefaction initiation.” Ph.D. thesis, Univ. of California, Berke-
ley, Berkeley, Calif.
Moss, R. E. S., et al. 共2006a兲. “CPT-based probabilistic and deterministic
assessment of in situ seismic soil liquefaction potential.” J. Geotech.
Geoenviron. Eng., 132共8兲, 1032–1051.
Moss, R. E. S., Seed, R. B., and Olsen, R. S. 共2006b兲. “Normalizing the
CPT for overburden stress.” J. Geotech. Geoenviron. Eng., 132共3兲,
378–387.
Olsen, R. S., and Mitchell, J. K. 共1995兲. “CPT stress normalization and
prediction of soil classification.” Proc., Int. Symp. on Cone Penetra-
tion Testing, CPT’95, SGF Rep. No. 3(95), SGI, Vol. 2, Linköping,
Sweden, Geotechnical Society, 257–262.
Olson, R. E., and Shantz, T. J. 共2004兲. “Axial load capacity of piles in
California in cohesionless soils.” Deep Foundations 2002, Geotechni-
cal Special Publication No. 116, ASCE Reston, Va., 1–15.
Parkin, A. K. 共1988兲. “The calibration of cone penetrometers.”
Penetration testing 1988, Balkema, Rotterdam, The Netherlands,
221–243.
Powell, J. J. M., Quarterman, R. S. T., and Lunne, T. 共1988兲. “Interpre-
tation and use of the piezocone test in UK clays.” Penetration testing
in the UK, Telford, London, 151–156.
Ramsey, N. 共2002兲. “A calibrated model for the interpretation of cone
penetration tests 共CPTs兲 in North Sea quaternary soils.” Offshore site
investigation and geotechnics—Diversity and sustainability, Society
for Underwater Technology, London, 341–356.
Randolph, M. F. 共2004兲. “Characterisation of soft sediments for offshore
applications.” Proc., 2nd Int. Conf. on Site Characterization (Proc.
ISC’2, Porto), Vol. 1, Millpress, Rotterdam, The Netherlands,
209–232.
Randolph, M. F., Carter, J. P., and Wroth, C. P. 共1979兲. “Driven piles in
clay—The effects of installation and subsequent consolidation.” Geo-
technique, 29共4兲, 361–393.
Randolph, M. F., and Hope, S. 共2004兲. “Effect of cone velocity on cone
resistance and excess pore pressures.” Proc., Int. Symp. on Engineer-
ing Practice and Performance of Soft Deposits, Yodagawa Kogisha
Co., Ltd., Japan, 147–152.
Randolph, M. F., Joer, H. A., Khorshid, M. S., and Hyden, A. M. 共1996兲.
“Field and laboratory data from pile load tests in calcareous soil.”
OTC 7992, Proc. 28th Offshore Tech. Conf., OTC, Houston, 327–336.
Robertson, P. K. 共1990兲. “Soil classification using the cone penetration
test.” Can. Geotech. J., 27共1兲, 151–158.
Robertson, P. K. 共1991兲. “Soil classification by the cone penetration test:
Reply.” Can. Geotech. J., 28共1兲, 176–178.
Robertson, P. K., and Campanella, R. G. 共1983兲. “Interpretation of cone
penetration tests. I: Sand.” Can. Geotech. J., 20共4兲, 718–733.
Robertson, P. K., Campanella, R. G., Gillespie, D., and Grieg, J. 共1986兲.
“Use of piezometer cone data.” Use of in situ tests in geotechnical
engineering, GSP 6, ASCE, Reston, Va., 1263–1280.
Salgado, R. 共1993兲. “Analysis of penetration resistance in sands.” Ph.D.
thesis, Univ. of California, Berkeley, Berkeley, Calif.
Salgado, R., Mitchell, J. K., and Jamiolkowski, M. 共1997兲. “Cavity ex-
pansion and penetration resistance in sand.” J. Geotech. Geoenviron.
Eng., 123共4兲, 344–354.
Sandven, R. 共1990兲. “Strength and deformation properties of fine grained
soils obtained from piezocone tests.” Ph.D. thesis, Norwegian Insti-
tute of Technology, Trondheim, Norway.
Sandven, R. 共2002兲. “Geotechnical properties of a natural silt deposit
obtained from field and laboratory tests.” Characterisation and engi-
neering properties of natural soils, Swets & Zeitlinger, Lisse, The
Netherlands, 1237–1276.
Schneider, J. A., Lehane, B. M., and Schnaid, F. 共2007兲. “Velocity effects
on piezocone tests in normally and overconsolidated clays.” Int. J.
Phys. Modell. Geotech., 7共2兲, 23–34.
Schneider, J. A., Mayne, P. W., and Rix, G. J. 共2001兲. “Geotechnical site
characterization in the greater Memphis area using cone penetration
tests.” Eng. Geol. (Amsterdam), 62, 169–184.
Senneset, K., and Janbu, N. 共1985兲. “Shear strength parameters obtained
from static cone penetration tests.” Strength testing of marine sedi-
ments: Laboratory and in situ test measurements, ASTM STP 833,
ASTM, Philadelphia, 41–54.
Silva, M. F., White, D. J., and Bolton, M. D. 共2006兲. “An analytical study
of the effect of penetration rate on piezocone tests in clay.” Int. J.
Numer. Analyt. Meth. Geomech., 30共6兲, 501–527.
1586 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / NOVEMBER 2008
View publication stats
Teh, C. I., and Houlsby, G. T. 共1991兲. “An analytical study of the cone
penetration test in clay.” Geotechnique, 41共1兲, 17–34.
Tomlinson, M. J. 共2001兲. Foundation design and construction, 7th Ed.,
Longman Group, London.
Tveldt, G., and Fredriksen, F. 共2003兲. “N18 Ny motorvegbru I Drammen
Prøvebelasting av peler.” Conf. on Rock and Blasting and Geotech-
nics, Oslo, Norway, 37.1–37.32.
Watson, P. G. 共1999兲. “Performance of skirted foundations for offshore
structures.” Ph.D. thesis, Dept. of Civil and Resource Engineering,
Univ. of Western Australia, Crawley, Western Australia, Australia.
Watson, P. G., and Humpheson, C. 共2005兲. “Geotechnical interpretation
for the Yolla A platform.” Proc., Int. Symp. on Frontiers Offshore
Geomechanics (ISFOG), Taylor & Francis, London, 343–349.
Wroth, C. P. 共1984兲. “The interpretation of in situ soil tests.” Geotech-
nique, 34共4兲, 449–489.
Wroth, C. P. 共1988兲. “Penetration testing—A more rigorous approach to
interpretation.” Penetration Testing 1988, (Proc. ISOPT, Orlando),
Vol. 1, Balkema, Rotterdam, The Netherlands, 303–311.
Wroth, C. P., and Houlsby, G. T. 共1985兲. “Soil mechanics: Property char-
acterization and analysis procedures.” Proc., 11th Int. Conf. on Soil
Mechanics and Foundation Engineering, ICSMFE, Vol. 1, Balkema,
Rotterdam, The Netherlands, 1–56.