1. The document discusses contributions of field case histories to geotechnical earthquake engineering, focusing on case histories involving liquefaction of cohesionless soils.
2. It outlines the analysis framework used for liquefaction evaluation, which includes functions describing cyclic resistance ratio, cyclic stress ratio, and the effects of initial effective stress, earthquake magnitude, fines content, and depth.
3. Four frequently asked questions about liquefaction evaluation are presented: why published CRR curves differ depending on the model used; whether models can be combined probabilistically; whether site response analyses are needed to obtain cyclic stress ratios; and how to treat liquefaction at depths beyond case histories.
1. The document discusses contributions of field case histories to geotechnical earthquake engineering, focusing on case histories involving liquefaction of cohesionless soils.
2. It outlines the analysis framework used for liquefaction evaluation, which includes functions describing cyclic resistance ratio, cyclic stress ratio, and the effects of initial effective stress, earthquake magnitude, fines content, and depth.
3. Four frequently asked questions about liquefaction evaluation are presented: why published CRR curves differ depending on the model used; whether models can be combined probabilistically; whether site response analyses are needed to obtain cyclic stress ratios; and how to treat liquefaction at depths beyond case histories.
1. The document discusses contributions of field case histories to geotechnical earthquake engineering, focusing on case histories involving liquefaction of cohesionless soils.
2. It outlines the analysis framework used for liquefaction evaluation, which includes functions describing cyclic resistance ratio, cyclic stress ratio, and the effects of initial effective stress, earthquake magnitude, fines content, and depth.
3. Four frequently asked questions about liquefaction evaluation are presented: why published CRR curves differ depending on the model used; whether models can be combined probabilistically; whether site response analyses are needed to obtain cyclic stress ratios; and how to treat liquefaction at depths beyond case histories.
1. The document discusses contributions of field case histories to geotechnical earthquake engineering, focusing on case histories involving liquefaction of cohesionless soils.
2. It outlines the analysis framework used for liquefaction evaluation, which includes functions describing cyclic resistance ratio, cyclic stress ratio, and the effects of initial effective stress, earthquake magnitude, fines content, and depth.
3. Four frequently asked questions about liquefaction evaluation are presented: why published CRR curves differ depending on the model used; whether models can be combined probabilistically; whether site response analyses are needed to obtain cyclic stress ratios; and how to treat liquefaction at depths beyond case histories.
Copyright:
Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online from Scribd
Download as pdf or txt
You are on page 1of 54
1
CONTRI BUTI ONS OF FI ELD CASE HI STORI ES TO
GEOTECHNI CAL EARTHQUAKE ENGI NEERI NG pr esent ed by I . M. I dr i ss, Pr of essor Emer i t us Uni ver si t y of Cal i f or ni a at Davi s e-mai l : i mi dr i ss@aol .c om Pr esent ed at t he di nner meet i ng of t he ASCE SEATTLE SECTI ON -- GEOTECHNI CAL GROUP Seat t l e, Washi ngt on Sept ember 30, 2010 Materials for this talk are based on work by I. M. Idriss and R. W. Boulanger Idriss & Boulanger (2008). "Soil Liquefaction During Earthquakes." Monograph MNO-12, EERI. Idriss & Boulanger (2010). "SPT-based liquefaction triggering procedures. "Report UCD/CGM-10/02, University of California, Davis, CA. The "Peck Lecture", which was presented at ASCE's GeoFlorida Conference on February 21, 2010 by I. M. Idriss. The full text of the "Peck Lecture" (by I. M. Idriss and R. W. Boulanger) is to be published in the Geotechnical Journal of ASCE in 2011 or 2012, depending on the length of the review process. 2 PECK AWARD The Ralph B. Peck Award recognizes an individual's contributions to the geotechnical engineering profession through the publication of a thoughtful, carefully researched case history or histories, or the publication of recommended practices or design methodologies based on the evaluation of case histories. Case Histories have always played a strong role in geotechnical engineering. They have been an essential means for: improving understanding; Calibrating analytical procedures; Designing & interpreting physical model tests; and developing semi-empirical procedures Under static as well as during earthquake and post- earthquake loading conditions. ROLE OF CASE HI STORI ES 3 SI GNI FI CANT EARTHQUAKES SI NCE 1960 1962 Mexico City 1964 ALASKA 1964 NIIGATA 1966 Parkfield 1967 Caracas 1968 Tokachi-Oki 1971 SAN FERNANDO 1975 Oroville 1975 Haicheng 1976 Gazli (USSR) 1976 Tangshan 1978 Miyagiken-Oki 1978 Santa Barbara 1978 Tabas 1979 Coyote Lake 1979 IMPERIAL VALLEY 1980 Livermore 1992 Petrolia 1992 Landers 1992 Big Bear 1994 NORTHRIDGE 1995 KOBE 1999 KOCAELI 1999 CHI-CHI 1999 Duzce 2001 Bhuj 2001 Nisqually 2004 Niigata 2010 Chile 1980 Mammoth Lake 1982 Miramichi 1983 Coalinga 1985 Chile 1985 MEXICO CITY 1985 Nahani 1986 NORTH PALM SPRINGS 1987 WHITTIER-NARROWS 1988 Armenia 1988 Saguenay 1989 LOMA PRIETA 1990 Manjil 1990 Philippine 1991 Costa Rica 1991 Sierra Madre 1992 Turkey 1992 Joshua tree OUTLI NE OF THI S TALK Case Histories of large deformations involving soft cohesive soils: Case Histories involving liquefaction of cohesionless soils: 4 LI QUEFACTI ON OF COHESI ONLESS SOI LS Examples of Surface Evidence of Liquefaction 1978 Mi yagi ken-Ok i ear t hquake LI QUEFACTI ON OF COHESI ONLESS SOI LS 5 1964 Ni i gat a ear t hquake (phot o: NI SEE) LI QUEFACTI ON OF COHESI ONLESS SOI LS 1971 San Fer nando ear t hquake (phot o: Cal i f or ni a DWR) LI QUEFACTI ON OF COHESI ONLESS SOI LS 6 LI QUEFACTI ON OF COHESI ONLESS SOI LS 1999 CHI -CHI ear t hquake LI QUEFACTI ON OF COHESI ONLESS SOI LS Information needed for each case history 1. Site information: i. Location, adjacent topography; ii. Adjacent physical features; iii. Surface [Evidence/No Evidence] of liquefaction. 2. Subsurface information: i. Borings, samples methods used; ii. Water table measurements; iii. Standard penetration tests details used; iv. Cone penetration resistance data; v. Shear wave measurements method(s) used. 3. Earthquake & earthquake ground motions information i. M w , distance, nearby recordings, site "classification". 7 LI QUEFACTI ON OF COHESI ONLESS SOI LS Use of liquefaction case histories started in 1968. At that time, there were only 23 cases with observed surface evidence of liquefaction and 12 cases with no observed evidence of liquefaction. These case histories were used in the development of the Seed-Idriss simplified liquefaction procedure, which was published in the Journal of ASCE's SM&FE Division in 1971. LI QUEFACTI ON OF COHESI ONLESS SOI LS Since then, the number of cases has dramatically increased. While in 1968 correlation was made to relative density and SPT blow count only, correlations are now made with: SPT blow count; CPT tip resistance, and V s , shear wave velocity. More recently, correlations with dilatometer measurements have been proposed. 8 ( ) 1 60 N E R B S m N C C C C C N = ( ) . , v M 7 5 1 1 60cs CRR f N o ' = =
=
Anal ysi s f r amewor k CYCLIC RESISTANCE RATIO (CRR) [Framework is similar for SPT, CPT, or V s correlations] ( ) ( ) ( ) 1 1 1 60cs 60 60 N N N = + A ( ) . , , vc M 7 5 1 1 60 CRR f N FC o ' = =
=
( ) 1 60 N E R B S m N C C C C C N = ( ) . , v M 7 5 1 1 60cs CRR f N o ' = =
=
Anal ysi s f r amewor k Cyclic resistance ratio (CRR) [Framework is similar for SPT, CPT, or V s correlations] ( ) ( ) ( ) 1 1 1 60cs 60 60 N N N = + A ( ) . , , vc M 7 5 1 1 60 CRR f N FC o ' = =
=
C N = f(o' v ; D R ; FC) C R = f(depth; rod stick-up length) 9 v v d M v a r CSR 0 65 o o o ' = ' max , . Anal ysi s f r amewor k Earthquake-induced CYCLIC STRESS RATIO (CSR) based on using the Seed-Idriss (1971) Simplified Procedure r d = f(depth; ground motion characteristics; dynamic soil properties) A c c e l e r a t i o n Time A c c e l e r a t i o n Time A c c e l e r a t i o n Time Ef f ec t s of dur at i on M = 5.1 M = 6.5 M = 7.3 ( ) v M, ' vo max d M 7.5 vo CSR a r 1 CSR 0.65 MSF ' MSF o o o = | | = = | \ . 10 Cyclic triaxial test results for clean Fraser Delta sand showing cyclic stress and CRR to cause 3% shear strain in 10 uniform cycles at D R of 31-72% and effective consolidation stresses of 50-400 kPa (data from Vaid & Sivathayalan 1996). EFFECTS OF I NI TI AL EFFECTI VE VERTI CAL STRESS, o' v Cyclic stress to cause 3% strain in 10 uniform cycles versus effective consolidation stress in ICU cyclic triaxial tests on Fraser Delta sand (data from Vaid & Sivathayalan 1996) EFFECTS OF I NI TI AL EFFECTI VE VERTI CAL STRESS, o' v 11 Ef f ec t s of o' v ( ) ( ) ( ) v v M. ' vo max d M 7.5, ' 1 atm vo CSR a r 1 1 CSR 0.65 MSF K ' MSF K o o o o o o = = | | = = | \ . Framework includes 5 functions that describe fundamental aspects of dynamic site response, penetration testing, and soil behavior: r d = f(depth; ground motion characteristics; dynamic soil properties) C N = f(o' v ; D R ; FC) C R = f(depth; rod stick-up length) K o = f(o' v ; D R ; FC) MSF = f(ground motion characteristics; D R ; FC) These functions should be based on a synthesis of experimental and theoretical methods, as they guide the application to conditions outside those that are represented in the case history database. Anal ysi s f r amewor k 12 Many questions have been raised over the years regarding evaluation of liquefaction potential during earthquakes. I will attempt to address in this presentation 4 of the most recurring questions. LI QUEFACTI ON OF COHESI ONLESS SOI LS LI QUEFACTI ON OF COHESI ONLESS SOI LS Corrected standard penetration, (N 1 ) 60 0 10 20 30 40 C y c l i c
s t r e s s
r a t i o 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Curves derived by FCs 5% Seed & Idriss (1982) Seed et al (1984) & NCEER/NSF Workshops (1997) Idriss & Boulanger (2004) Seed (1979) Cetin et al (2004) 1 2 3 5 3 2 1 5 4 4 13 QUESTI ONS RAI SED Q-2. Can we treat these differences as "epistemic" uncertainty and hence can use all models with "assigned weights"? Q-3. Can we use site response analyses to obtain CSR or do we have to always use the simplified stress ratio equation? Q-4. How should we treat liquefaction at depths exceeding those included in the liquefaction case histories? Q-1. Why are the published curves of CRR versus (N 1 ) 60 or versus (N 1 ) 60cs different, depending on whose model is implemented? QUESTI ON No. 1 Q-1. Why are the published curves of CRR versus (N 1 ) 60 or versus (N 1 ) 60cs different, depending on whose model is implemented? In particular, why is the Cetin et al correlation so much lower than the other correlations? Equivalent clean sand corrected standard penetration, (N 1 ) 60cs 0 10 20 30 40 C R R 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Curves derived by 3 5 4 Seed et al (1984) & NCEER/NSF Workshops (1997) Idriss & Boulanger (2004) Cetin et al (2004) 3 4 5 14 QUESTI ON No. 1 The best way to address this question is to examine each model in terms of how the interpretations were made for those case histories that control the position of the correlation. Specifically, it is essential that the derived liquefaction triggering correlation for M = 7.5 and o' v = 1 atm be supported by the case histories with o' v close to 1 atm. Differences in the treatment of key case histories near o' v = 1 atm (where differences in C N and K o are smallest) were found to be the primary cause of differences in the correlations. ( ) 1 60 N E R B S m N C C C C C N = ( ) . , v M 7 5 1 1 60cs CRR f N o ' = =
=
Cycl i c Resi st anc e Rat i o Cyclic resistance ratio (CRR) [Framework is similar for SPT, CPT, or V s correlations] ( ) ( ) ( ) 1 1 1 60cs 60 60 N N N = + A ( ) . , , vc M 7 5 1 1 60 CRR f N FC o ' = =
=
C N = f(o' v ; D R ; FC) C R = f(depth; rod stick-up length) 15 Wher e t he f unc t i ons ar e ( ) ( ) ( ) v v M. ' vo max d M 7.5, ' 1 atm vo CSR a r 1 1 CSR 0.65 MSF K ' MSF K o o o o o o = = | | = = | \ . Shear st r ess i nduc ed by t he ear t hquake gr ound mot i ons Sensitivity of case history interpretation to MSF 16 0 10 20 30 40 50 (N 1 ) 60 16 12 8 4 0 D e p t h
b e l o w
g r o u n d
s u r f a c e
( m ) Liquefaction Marginal No liquefaction 0 0.2 0.4 0.6 0.8 CSR M=7.5,o=1 16 12 8 4 0 5 6 7 8 9 M 16 12 8 4 0 D e p t h
b e l o w
g r o u n d
s u r f a c e
( m ) 0 20 40 60 80 100 FC (%) 16 12 8 4 0 Ef f ec t s of dur at i on Earthquake moment magnitude, M 5 6 7 8 M a g n i t u d e
s c a l i n g
f a c t o r ,
M S F 0.5 1.0 1.5 2.0 2.5 Cetin et al (2004) Idriss & Boulanger (2004) Seed et al (1984) 17 Ef f ec t s of o' v ( ) ( ) ( ) v v M. ' vo max d M 7.5, ' 1 atm vo CSR a r 1 1 CSR 0.65 MSF K ' MSF K o o o o o o = = | | = = | \ . Sensitivity of case history interpretation to K o EFFECTS OF I NI TI AL EFFECTI VE VERTI CAL STRESS, o' v K o relations recommended by Youd et al (2001) for a relative density of 40, 60 and 80% (solid lines) and relation used by Cetin et al (2004) Vertical effective stress, o' v (atm) 0 1 2 3 K o 0.0 0.5 1.0 1.5 Cetin et al (2004) Youd et al (2001); D R = 40, 60 & 80% 18 LI QUEFACTI ON OF COHESI ONLESS SOI LS Vertical effective stress, o' v (atm) 0 1 2 3 K o 0.0 0.5 1.0 1.5 Boulanger & Idriss (2004); D R = 40, 60 & 80% K o relations recommended by Boulanger and Idriss (2004) for a relative density of 40, 60 and 80% LI QUEFACTI ON OF COHESI ONLESS SOI LS Vertical effective stress, o' v (atm) 0.0 0.5 1.0 1.5 2.0 K o 0.0 0.5 1.0 1.5 Boulanger & Idriss (2004); D R = 60 Youd et al (2001); D R = 60 Cetin et al (2004) 19 LI QUEFACTI ON OF COHESI ONLESS SOI LS Vertical effective stress, o' v (atm) 0.6 0.8 1.0 1.2 1.4 1.6 K o 0.8 0.9 1.0 1.1 1.2 Boulanger & Idriss (2004); D R = 60 Youd et al (2001); D R = 60 Cetin et al (2004) Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) Effective vertical stress, o' v (psf) 0 500 1000 1500 2000 2500 3000 C u m .
D i s t r i b u t i o n
( % ) 0 20 40 60 80 100 0.8 atm 1.2 atm 1 atm Values of o' v as listed in Cetin et al (2004) for the "liquefaction" & "marginal" case histories 0.65 atm 20 Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) (N 1 ) 60cs 0 10 20 30 40 C S R
( a d j u s t e d
t o
M
=
7 . 5
&
o ' v
=
1
a t m ) 0.0 0.1 0.2 0.3 0.4 Cetin et al (2004) M = 7.5; o' v = 1 atm 1 2 3 4 5 6 7 8 9 10 11 Triangles: 1984 cases; Circles: 2000 cases; Squares: Kobe proprietary cases. Filled-in symbols: liquefaction; Open symbols: no liquefaction; Cyan symbol: marginal. Cases for o' v = 0.65 to 1.5 atm Data and parameters from Cetin et al (2004); Points 1 -- 11 identified for further examination as described in text. Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) Point identified in next figure Site name Earthquake 1 Miller Farm CMF-10 1989 Loma Prieta earthquake; M = 6.9 2 Malden Street, Unit D 1994 Northridge earthquake; M = 6.7 3 Kobe #6 1995 Hyogoken-Nambu (Kobe) earthquake; M = 6.9 4 Kobe #7 1995 Hyogoken-Nambu (Kobe) earthquake; M = 6.9 5 Miller Farm CMF-5 1989 Loma Prieta earthquake; M = 6.9 6 Rail Road #2 1964 Niigata earthquake; M = 7.6 7 Port of Oakland POO7-2 1989 Loma Prieta earthquake; M = 6.9 8 Port of Oakland POO7-3 1989 Loma Prieta earthquake; M = 6.9 9 Panjin Chemical Fertilizer Plant 1975 Haicheng earthquake; M = 7.0 10 Shuang Tai Zi River 1975 Haicheng earthquake; M = 7.0 11 San Juan B-3 1974 Argentina earthquake; M = 7.4 Sites identified for further examination because they dictate the location of the liquefaction triggering curve for M = 7.5 & o' v = 1 atm 21 Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) Point identified in next figure Site name Earthquake 1 Miller Farm CMF-10 1989 Loma Prieta earthquake; M = 6.9 2 Malden Street, Unit D 1994 Northridge earthquake; M = 6.7 3 Kobe #6 1995 Hyogoken-Nambu (Kobe) earthquake; M = 6.9 4 Kobe #7 1995 Hyogoken-Nambu (Kobe) earthquake; M = 6.9 5 Miller Farm CMF-5 1989 Loma Prieta earthquake; M = 6.9 6 Rail Road #2 1964 Niigata earthquake; M = 7.6 7 Port of Oakland POO7-2 1989 Loma Prieta earthquake; M = 6.9 8 Port of Oakland POO7-3 1989 Loma Prieta earthquake; M = 6.9 9 Panjin Chemical Fertilizer Plant 1975 Haicheng earthquake; M = 7.0 10 Shuang Tai Zi River 1975 Haicheng earthquake; M = 7.0 11 San Juan B-3 1974 Argentina earthquake; M = 7.4 Sites identified for further examination because they dictate the location of the liquefaction triggering curve for M = 7.5 & o' v = 1 atm misclassified cases Poi nt 1 Mi l l er Far m CMF-10 Profile across the failure zone at the Miller (south side of Pajaro River) during the 1989 Loma Prieta Earthquake (Holzer et al. 1994) 22 Cet i n et al (2004) From Cetin et al (2000) Geotechnical Engineering Research Report No. UCB/GT-2000/09 Poi nt 2 Mal den St . Uni t D Poi nt 2: Mal den St r eet , Uni t D Profile across the failure zone at the Malden Street site during the 1994 Northridge Earthquake (Holzer et al. 1998) 23 Expanded profile across the failure zone (Holzer et al. 1998) [additional details in Bennett et al. 1998] Poi nt 2: Mal den St r eet , Uni t D Poi nt 3 Kobe pr opr i et ar y si t e 6 Original table from Tokimatsu (2010) From Cetin et al (2000) Geotechnical Engineering Research Report No. UCB/GT-2000/09 24 Point 10 Shuang Tai Zi River Poi nt 10 Shuang Tai Zi Ri ver From original source: Shengcong & Tatsuoka (1984) 25 Point 10 Shuang Tai Zi River From Seed et al (1984) Points 1, 2, 3 & 10 were designated as "No Liquefaction" by the original investigators of these sites; Cetin et al (2004) listed these as "Liquefaction" sites. Point 1 Miller Farm CMF 10 o' v ~ 0.70 atm Point 2 Malden Street o' v ~ 1.2 atm Point 3 Kobe No. 6 o' v ~ 0.68 atm Point 10 Shuang Tai Zi R. o' v ~ 0.69 atm Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) 26 Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) (N 1 ) 60cs 0 10 20 30 40 C S R
( a d j u s t e d
t o
M
=
7 . 5
&
o ' v
=
1
a t m ) 0.0 0.1 0.2 0.3 0.4 Cetin et al (2004) M = 7.5; o' v = 1 atm 1 2 3 4 5 6 7 8 9 10 11 Triangles: 1984 cases; Circles: 2000 cases; Squares: Kobe proprietary cases. Filled-in symbols: liquefaction; Open symbols: no liquefaction; Cyan symbol: marginal. Cases for o' v = 0.65 to 1.5 atm Data and parameters from Cetin et al (2004); Points 1, 2, 3 & 10 were designated as "No Liquefaction" by the original investigators of these sites; Cetin et al (2004) listed these as "Liquefaction" sites . Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 r d values from summary tables in Cetin et al (2004) 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 r d
v a l u e s
c o m p u t e d
u s i n g
C e t i n
e t
a l ' s
e q u a t i o n ( o n l y
f o r
c a s e s
w i t h o u t
s i t e
r e s p o n s e
c a l c u l a t i o n s s ) Issue: The r d values computed using the Cetin et al (2004) equation do not agree with the r d values they used in processing the case histories. Discrepancy between r d values used in the Cetin et al (2004) database and the r d values computed using their referenced r d equation 27 Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) Point identified in next figure Site name Earthquake 1 Miller Farm CMF-10 1989 Loma Prieta earthquake; M = 6.9 2 Malden Street, Unit D 1994 Northridge earthquake; M = 6.7 3 Kobe #6 1995 Hyogoken-Nambu (Kobe) earthquake; M = 6.9 4 Kobe #7 1995 Hyogoken-Nambu (Kobe) earthquake; M = 6.9 5 Miller Farm CMF-5 1989 Loma Prieta earthquake; M = 6.9 6 Rail Road #2 1964 Niigata earthquake; M = 7.6 7 Port of Oakland POO7-2 1989 Loma Prieta earthquake; M = 6.9 8 Port of Oakland POO7-3 1989 Loma Prieta earthquake; M = 6.9 9 Panjin Chemical Fertilizer Plant 1975 Haicheng earthquake; M = 7.0 10 Shuang Tai Zi River 1975 Haicheng earthquake; M = 7.0 11 San Juan B-3 1974 Argentina earthquake; M = 7.4 Sites identified for further examination because they dictate the location of the liquefaction triggering curve for M = 7.5 & o' v = 1 atm r d Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) (N 1 ) 60cs 0 10 20 30 40 C S R
( a d j u s t e d
t o
M
=
7 . 5
&
o ' v
=
1
a t m ) 0.0 0.1 0.2 0.3 0.4 1 2 3 4 5 6 7 8 9 10 11 Cetin et al (2004) M = 7.5; o' v = 1 atm Triangles: 1984 cases; Circles: 2000 cases; Squares: Kobe proprietary cases. Filled-in symbols: liquefaction; Open symbols: no liquefaction; Cyan symbol: marginal. Cases for o' v = 0.65 to 1.5 atm Data and parameters from Cetin et al (2004); CSR for Points 3, 4, 6, 9. 10 & 11 recalculated using equation for r d in Cetin et al (2004) in lieu of their listed values. 28 Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) Point identified in next figure Site name Earthquake 1 Miller Farm CMF-10 1989 Loma Prieta earthquake; M = 6.9 2 Malden Street, Unit D 1994 Northridge earthquake; M = 6.7 3 Kobe #6 1995 Hyogoken-Nambu (Kobe) earthquake; M = 6.9 4 Kobe #7 1995 Hyogoken-Nambu (Kobe) earthquake; M = 6.9 5 Miller Farm CMF-5 1989 Loma Prieta earthquake; M = 6.9 6 Rail Road #2 1964 Niigata earthquake; M = 7.6 7 Port of Oakland POO7-2 1989 Loma Prieta earthquake; M = 6.9 8 Port of Oakland POO7-3 1989 Loma Prieta earthquake; M = 6.9 9 Panjin Chemical Fertilizer Plant 1975 Haicheng earthquake; M = 7.0 10 Shuang Tai Zi River 1975 Haicheng earthquake; M = 7.0 11 San Juan B-3 1974 Argentina earthquake; M = 7.4 Sites identified for further examination because they dictate the location of the liquefaction triggering curve for M = 7.5 & o' v = 1 atm SPT data not included Poi nt 4 Kobe Pr opr i et ar y Si t e No. 7 (f r om Cet i n et al (2000) Point 4 (Kobe No. 7 site) 29 Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) Point 4 (Kobe No. 7 site) o' v ~ 0.8 atm Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) Selection of a representative (N 1 ) 60 for Point 4 (Kobe No. 7 site) Average o' v ~ 0.86 atm Avg depth (m) Depthto GWT(m) o vc(kPa) o'vc(kPa) (Nm) (N1)60 CB CE CN CR CS FC(%) (N1)60,cs 3.3 3.2 62 60 8 10.4 1 1.22 1.26 0.85 1 0 10.4 4.3 3.2 82 71 21 28.2 1 1.22 1.16 0.95 1 0 28.2 6.3 3.2 124 93 32 37.7 1 1.22 1.02 0.95 1 12 39.8 7.3 3.2 144 104 23 25.6 1 1.22 0.96 0.95 1 0 25.6 8.3 3.2 165 114 21 23.4 1 1.22 0.92 1 1 0 23.4 Averages: 5.8 113 87 18.3 21.9 Average= 0 21.9 30 Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) (N 1 ) 60cs 0 10 20 30 40 C S R
( a d j u s t e d
t o
M
=
7 . 5
&
o ' v
=
1
a t m ) 0.0 0.1 0.2 0.3 0.4 1 2 3 4 5 6 7 8 9 10 11 Cetin et al (2004) M = 7.5; o' v = 1 atm Triangles: 1984 cases; Circles: 2000 cases; Squares: Kobe proprietary cases. Filled-in symbols: liquefaction; Open symbols: no liquefaction; Cyan symbol: marginal. Cases for o' v = 0.65 to 1.5 atm Data and parameters from Cetin et al (2004); CSR & (N 1 ) 60 for Point 4 recalculated to include a sublayer below the water table with N = 8, which had not been used by Cetin et al (2004). Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) (N 1 ) 60cs 0 10 20 30 40 C S R
( a d j u s t e d
t o
M
=
7 . 5
&
o ' v
=
1
a t m ) 0.0 0.1 0.2 0.3 0.4 1 2 3 4 5 6 7 8 9 10 11 Cetin et al (2004) M = 7.5; o' v = 1 atm Triangles: 1984 cases; Circles: 2000 cases; Squares: Kobe proprietary cases. Filled-in symbols: liquefaction; Open symbols: no liquefaction; Cyan symbol: marginal. Cases for o' v = 0.65 to 1.5 atm Data and parameters from Cetin et al (2004); CSR & (N 1 ) 60 for Point 4 recalculated to include a sublayer below the water table with N = 8, which had not been used by Cetin et al (2004). 31 Poi nt 6 Rai l Road-2 (f r om Cet i n et al (2000) Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) 10 12 14 16 18 20 22 Average total unit weight (kN/m 3 ) 12 10 8 6 4 2 0 D e p t h
b e l o w
g r o u n d
s u r f a c e
( m ) Idriss & Boulanger (this study) Cetin et al (2004) Seed et al. (1984), plus Averages of all values Kobe proprietary (Tokimatsu) 32 Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) (N 1 ) 60cs 0 10 20 30 40 C S R
( a d j u s t e d
t o
M
=
7 . 5
&
o ' v
=
1
a t m ) 0.0 0.1 0.2 0.3 0.4 1 2 3 4 5 6 7 8 9 10 11 Cetin et al (2004) M = 7.5; o' v = 1 atm Triangles: 1984 cases; Circles: 2000 cases; Squares: Kobe proprietary cases. Filled-in symbols: liquefaction; Open symbols: no liquefaction; Cyan symbol: marginal. Cases for o' v = 0.65 to 1.5 atm Data and parameters from Cetin et al (2004); CSR & (N 1 ) 60 for Points 1 -- 11 recalculated using unit weights described in text. (N 1 ) 60cs 0 10 20 30 40 C S R
( a d j u s t e d
t o
M
=
7 . 5
&
o ' v
=
1
a t m ) 0.0 0.1 0.2 0.3 0.4 1 2 3 4 5 6 7 8 9 10 11 Triangles: 1984 cases; Circles: 2000 cases; Squares: Kobe proprietary cases. Filled-in symbols: liquefaction; Open symbols: no liquefaction; Cyan symbol: marginal. Cases for o' v = 0.65 to 1.5 atm Idriss & Boulanger (2004) M = 7.5; o' v = 1 atm NCEER/Youd (2001) M = 7.5; o' v = 1 atm Data and parameters from Cetin et al (2004); Changes to Points 1 -- 11 described in text. 33 Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) 10 12 14 16 18 20 22 Average total unit weight (kN/m 3 ) 12 10 8 6 4 2 0 D e p t h
b e l o w
g r o u n d
s u r f a c e
( m ) Idriss & Boulanger (this study) Cetin et al (2004) Seed et al. (1984), plus Averages of all values Kobe proprietary (Tokimatsu) Case histories of Liquefaction/ No Liquefaction published by Cetin et al (2004) Point identified in next figure Site name Earthquake 1 Miller Farm CMF-10 1989 Loma Prieta earthquake; M = 6.9 2 Malden Street, Unit D 1994 Northridge earthquake; M = 6.7 3 Kobe #6 1995 Hyogoken-Nambu (Kobe) earthquake; M = 6.9 4 Kobe #7 1995 Hyogoken-Nambu (Kobe) earthquake; M = 6.9 5 Miller Farm CMF-5 1989 Loma Prieta earthquake; M = 6.9 6 Rail Road #2 1964 Niigata earthquake; M = 7.6 7 Port of Oakland POO7-2 1989 Loma Prieta earthquake; M = 6.9 8 Port of Oakland POO7-3 1989 Loma Prieta earthquake; M = 6.9 9 Panjin Chemical Fertilizer Plant 1975 Haicheng earthquake; M = 7.0 10 Shuang Tai Zi River 1975 Haicheng earthquake; M = 7.0 11 San Juan B-3 1974 Argentina earthquake; M = 7.4 Sites identified for further examination because they dictate the location of the liquefaction triggering curve for M = 7.5 & o' v = 1 atm Low total unit weights 34 (N 1 ) 60cs 0 10 20 30 40 C S R
( a d j u s t e d
t o
M
=
7 . 5
&
o ' v
=
1
a t m ) 0.0 0.1 0.2 0.3 0.4 1 2 3 4 5 6 7 8 9 10 11 Triangles: 1984 cases; Circles: 2000 cases; Squares: Kobe proprietary cases. Filled-in symbols: liquefaction; Open symbols: no liquefaction; Cyan symbol: marginal. Cases for o' v = 0.65 to 1.5 atm Cetin et al (2004) M = 7.5; o' v = 1 atm Idriss & Boulanger (2004) M = 7.5; o' v = 1 atm Data and parameters from Cetin et al (2004); Changes to Points 1 -- 11 described in text. Concl usi ons r e: Quest i on No. 1 The Cetin et al. triggering correlation, if it were updated after correcting the above problems, would thus be expected to move close to the Idriss-Boulanger correlation at overburden stresses of 0.65-1.5 atm. This would also cause the Cetin et al. K o relationship to become flatter because it is regressed as part of their analyses and higher CRR values at higher confining stresses would dictate a flatter K s relationship. Q-1. Why are the published curves of CRR versus (N 1 ) 60 or versus (N 1 ) 60cs different, depending on whose model is implemented? In particular, why is the Cetin et al correlation so much lower than the other correlations? 35 The combination of these changes would be expected to reduce the degree to which the Cetin et al. procedure predicts significantly smaller CRR values than the other liquefaction triggering correlations as depth increases. Until these issues are addressed, however, the Cetin et al. procedure shoul d not be used. Concl usi ons r e: Quest i on No. 1 Quest i on No. 2 Q-2. Can we treat these differences as "epistemic" uncertainty and hence can use all models with "assigned weights"? No, we should not treat these differences as "epistemic" uncertainty and hence can use all models with "assigned weights". The examination I just summarized emphasizes the need to fully examine any model before it is adopted for use. 36 Quest i on No. 3 Q-3. Can we use site response analyses to obtain CSR or do we have to always use the simplified stress ratio equation? The answer is it depends. 37 Shear wave velocity (m/sec) 0 200 400 600 800 D e p t h
( m ) 0 20 40 60 80 100 120 140 160 V s profile used in 1990 V s profile used in 1993 and 1996 1996 1993 Period (sec) 0.01 0.1 1 10 S p e c t r a l
a c c e l e r a t i o n
( g )
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 damping = 5 percent Calculated Motion using 1990 V s profile Recorded Motion at Treasure Island Rock Outcrop (Yerba Buena Island) Calculated Motion using 1993 V s profile Calculated Motion using 1996 V s profile 38 Period (sec) 0.01 0.1 1 10 S p e c t r a l
a c c e l e r a t i o n
( g )
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 damping = 5 percent Spectral values for motion recorded at Treasure Island Spectral values calculated using recording at Yerba Buena as input motion 39 Period (sec) 0.01 0.1 1 10 S p e c t r a l
a c c e l e r a t i o n
( g )
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 damping = 5 percent Spectral values for motion recorded at Treasure Island Spectral values calculated using recordings at other rock sites in the Bay Area as input motions recording at Yerba Buena as input motion Maximum shear stress (kPa) 0 10 20 30 40 D e p t h
( m ) 0 5 10 15 20 Maximum shear stresses calculated using: recordings at other rock sites in the Bay Area as input motions recording at Yerba Buena as input motion average shear stresses for all cases 40 Period (sec) 0.01 0.1 1 10 S p e c t r a l
a c c e l e r a t i o n
/
P G A 0.0 0.5 1.0 1.5 2.0 2.5 3.0 damping = 5 percent Target spectrum -- M = 6.9 at 80 km Spectrum compatible motion -- SYN1 Spectrum compatible motion -- SYN2 Maximum shear stress (kPa) 0 10 20 30 40 D e p t h
( m ) 0 5 10 15 20 Maximum shear stresses calculated using: recordings at other rock sites in the Bay Area as input motions recording at Yerba Buena as input motion average shear stresses for all cases Input: SYN1 41 Maximum shear stress (kPa) 0 10 20 30 40 D e p t h
( m ) 0 5 10 15 20 Maximum shear stresses calculated using: recordings at other rock sites in the Bay Area as input motions recording at Yerba Buena as input motion average shear stresses for all cases Input: SYN2 Period (sec) 0.01 0.1 1 10 S p e c t r a l
a c c e l e r a t i o n
/
P G A 0.0 0.5 1.0 1.5 2.0 2.5 3.0 damping = 5 percent Target spectrum -- pre-NGA Target spectrum -- NGA 42 Period (sec) 0.01 0.1 1 10 S p e c t r a l
a c c e l e r a t i o n
/
P G A 0.0 0.5 1.0 1.5 2.0 2.5 3.0 damping = 5 percent Target spectrum -- NGA Spectra -- synthetic time series Maximum shear stress (kPa) 0 10 20 30 40 D e p t h
( m ) 0 5 10 15 20 Maximum shear stresses calculated using: NGA-compatible time series as input motions recording at Yerba Buena as input motion 43 Maximum shear stress (kPa) 0 10 20 30 40 D e p t h
( m ) 0 5 10 15 20 Maximum shear stresses calculated using: NGA-compatible time series as input motions recording at Yerba Buena as input motion average shear stresses synth time series Using simplified equation (F = ma): ( ) ( ) surf max v d surf a r g a 0.16g t o | | = | \ . = 44 Stress reduction coefficient, r d 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 D e p t h
b e l o w
g r o u n d
s u r f a c e
( m ) 0 4 8 12 16 20 24 28 M = 7 M = 8 Magnitude: M = 5 M = 6 Average of Range Published by Seed & Idriss (1971) M = 6.9 Maximum shear stress (kPa) 0 10 20 30 40 D e p t h
( m ) 0 5 10 15 20 Maximum shear stresses calculated using: recordings at other rock sites in the Bay Area as input motions recording at Yerba Buena as input motion average shear stresses for all cases Calculated using r d for M = 6.9 45 Concl usi ons r e: Quest i on No. 3 Use of an appropriate r d is adequate for most cases. For site response studies, you need to use at least 7 different rock outcrop motions. Q-3. Can we use site response analyses to obtain CSR or do we have to always use the simplified stress ratio equation? Q-4: How should we evaluate liquefaction at depths that exceed those represented in liquefaction case histories? Two critical parameters affecting liquefaction potential with depth are C N and K o . Studies at Perris Dam provide valuable data on C N at large depths Studies at Duncan Dam provide a valuable check on the complete liquefaction analysis procedure for large depths. Quest i on No. 4 46 A critical parameter affecting liquefaction potential with depth is the value of C N . Boulanger and Idriss (2004) recommended: ( ) m a N vo 1 60 P C 1.7 m 0.784 0.0768 N o | | = s | ' \ . = Note that m = , originally derived by Liao & Whitman has been extensively used, but it can produce unreasonably low C N values as the depth increases. The investigations carried out at Perris Dam (CDWR 2005, Wehling and Rennie 2008) are very helpful is assessing the value of the exponent m as a function of denseness. Per r i s Dam and C N Aerial photo and boring locations at Perris Dam (Wehling & Rennie 2008) 47 SPT data by location and percentile groupings (Wehling & Rennie 2008) 0 20 40 60 80 SPT (N 1 ) 60 0 0.2 0.4 0.6 0.8 E x p o n e n t
m 0 20 40 60 80 SPT (N 1 ) 60,CS 0 0.2 0.4 0.6 0.8 E x p o n e n t
m C N = (P a /o' v ) m Idriss & Boulanger (2008) [using (N 1 ) 60 as input] Idriss & Boulanger (2008) Perris dam foundation (Wehling & Rennie 2008) Perris dam foundation (Wehling & Rennie 2008) C N = (P a /o' v ) m 48 Overburden normalization factor C N : (a) dependence on denseness, and (b) simpler approximations often used at shallower depths. 0 0.5 1 1.5 C N 10 8 6 4 2 0 V e r t i c a l e f f e c t i v e s t r e s s , o ' v / P a (N 1 ) 60cs =40 (N 1 ) 60cs =30 (N 1 ) 60cs =20 (N 1 ) 60cs =10 (N 1 ) 60cs =4 0 0.5 1 1.5 C N 2 1.5 1 0.5 0 V e r t i c a l e f f e c t i v e s t r e s s , o ' v / P a (a) (b) (N 1 ) 60cs =4 (N 1 ) 60cs =30 Liao & Whitman (1986) C N = (P a /o' v ) 0.5 The investigations carried out at Duncan Dam (Special collection of papers in the Canadian Geotechnical Journal, 1994) are helpful in assessing the application of liquefaction triggering procedures to large depths. Dunc an Dam 49 Frozen sand samples obtained from Unit 3c at the toe, and tested at confining stresses of 2 to 12 atm. Dunc an Dam Table 5.2. Summary of SPT and laboratory test data for Duncan Dam
SPT data DSS tests Triaxial tests Conversion to o' v = 1 atm o' v
Notes: (1) Original data from Pillai and Byrne (1994). (2) Average ratio of CRR DSS /CRR TX = 0.85 is used to convert triaxial test results to field simple shear conditions. (3) Cyclic strengths multiplied by 0.937 to convert from 10 to 15 equivalent uniform cycles (based on slope of CRR versus number of uniform cycles curves). (4) Cyclic strengths multiplied by 0.90 to convert from 1D to 2D cyclic loading conditions. (5) Final value for field CRR M=7.5 taken as average of strengths from DSS and Triaxial tests. SPT-based prediction of CRR M=7.5 versus depth (confining stress) depends on combination of triggering curve, C N , and K o . Dunc an Dam Corrected standard penetration, (N 1 ) 60 0 10 20 30 40 C y c l i c
s t r e s s
r a t i o 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Curves derived by FCs 5% Seed & Idriss (1982) Seed et al (1984) & NCEER/NSF Workshops (1997) Idriss & Boulanger (2004) Seed (1979) Cetin et al (2004) 1 2 3 5 3 2 1 5 4 4 50 Pillai & Byrne (1994) used the Seed et al. (1984) triggering curve, in-situ SPT data, and laboratory test data on frozen sand samples to derive site-specific C N and K o relationships. Dunc an Dam 0 0.2 0.4 0.6 0.8 1 1.2 C N 12 10 8 6 4 2 0 V e r t i c a l e f f e c t i v e s t r e s s , o ' v / P a Boulanger & Idriss (2004): (N 1 ) 60 =10 (N 1 ) 60 =20 Liao & Whitman (1986) Pillai & Byrne (1994) 0 0.2 0.4 0.6 0.8 1 1.2 K o 12 10 8 6 4 2 0 V e r t i c a l e f f e c t i v e s t r e s s , o ' v / P a (a) (b) Boulanger & Idriss (2004): (N 1 ) 60 =10 (N 1 ) 60 =20 Hynes & Olsen (1999); f = 0.722 Pillai & Byrne (1994) Kayen et al (1992) CRR M=7.5 predicted using the Pillai & Byrne (1994) site-specific relationships with the Seed et al. (1984) triggering curve. Dunc an Dam 0 10 20 30 40 50 60 SPT N 60 values 12 10 8 6 4 2 0 V e r t i c a l
e f f e c t i v e
s t r e s s
( a t m ) 0 10 20 30 (N 1 ) 60 12 10 8 6 4 2 0 0 0.1 0.2 0.3 CRR M=7.5 12 10 8 6 4 2 0 Computed using relations by Pillai & Byrne (1994) CRR M7.5 from TX & DSS tests on frozen samples (Pillai & Byrne 1994) Duncan Dam - Unit 3c: (Pillai & Stewart 1994) 51 Dunc an Dam 0 10 20 30 40 50 60 SPT N 60 values 12 10 8 6 4 2 0 V e r t i c a l
e f f e c t i v e
s t r e s s
( a t m ) 0 10 20 30 (N 1 ) 60 12 10 8 6 4 2 0 0 0.1 0.2 0.3 CRR M=7.5 12 10 8 6 4 2 0 Computed using relations by Idriss & Boulanger (2008) CRR M7.5 from TX & DSS tests on frozen samples (Pillai & Byrne 1994) Duncan Dam - Unit 3c: (Pillai & Stewart 1994) CRR M=7.5 predicted using the Idriss & Boulanger (2004, 2008) liquefaction triggering procedures. Dunc an Dam 0 10 20 30 40 50 60 SPT N 60 values 12 10 8 6 4 2 0 V e r t i c a l
e f f e c t i v e
s t r e s s
( a t m ) 0 10 20 30 (N 1 ) 60 12 10 8 6 4 2 0 0 0.1 0.2 0.3 CRR M=7.5 12 10 8 6 4 2 0 Computed using relations by NCEER/NSF (Youd et al. 2001) CRR M7.5 from TX & DSS tests on frozen samples (Pillai & Byrne 1994) Duncan Dam - Unit 3c: (Pillai & Stewart 1994) CRR M=7.5 predicted using the NCEER/NSF (Youd et al. 2001) liquefaction triggering procedures. 52 Dunc an Dam 0 10 20 30 40 50 60 SPT N 60 values 12 10 8 6 4 2 0 V e r t i c a l
e f f e c t i v e
s t r e s s
( a t m ) 0 10 20 30 (N 1 ) 60 12 10 8 6 4 2 0 0 0.1 0.2 0.3 CRR M=7.5 12 10 8 6 4 2 0 Computed using relations by Cetin et al. (2004) CRR M7.5 from TX & DSS tests on frozen samples (Pillai & Byrne 1994) Duncan Dam - Unit 3c: (Pillai & Stewart 1994) CRR M=7.5 predicted using the Cetin et al. (2004) liquefaction triggering procedures. Cetin et al. (2004) and Moss et al. (2006) used the same statistical analysis procedures to regress K o from SPT and CPT case histories, respectively. Regr essi ng K o f r om c ase hi st or i es 0 2 4 6 8 10 Effective consolidation stress (atm) 0 0.5 1 1.5 K o Moss et al. (2006): From Bayesian regression of CPT-based liquefaction triggering database 15 (N 1 ) 60 = 5 25 Boulanger & Idriss (2004): From combination of lab- & field-derived CRR-c R correlations Cetin et al. (2004): From Bayesian regression of SPT-based liquefaction triggering database 53 Q-4: How should we evaluate liquefaction at depths that exceed those represented in liquefaction case histories? C N describes how penetration resistance varies with confining stress, and it fundamentally depends not only on o' v but also on soil denseness. For o' v > 2 atm, the Liao-Whitman (1986) or Kayen et al. (1992) relationships for C N , as adopted for the NCEER/NSF (Youd et al. 2001) procedures, can lead to a significant under-estimation of (N 1 ) 60 values for denser soils. For o' v > 2 atm, the Boulanger-Idriss (2004) relationship for C N produces more realistic (N 1 ) 60 values for denser soils, as supported by calibration chamber test data, penetration theory, and field studies. Concl usi ons r e: Quest i on No. 4 K o describes a fundamental soil behavior that also depends on o' v and on soil denseness. The K o relationships regressed from case history data by Cetin et al. (2004) & Moss et al. (2006) are not justifiable and should not be used. The K o relationships by Boulanger & Idriss (2004) or Hynes & Olsen (1998) are reasonable options. The procedures by Idriss & Boulanger were in good agreement with data for Duncan Dam. The NCEER/NSF (Youd et al. 2001) procedures with the Hynes-Olsen K o relationship under-estimated CRR for the larger depths. Concl usi ons r e: Quest i on No. 4 54 Three recurring questions regarding assessment of liquefaction potential were addressed. 1. Why are the published curves of CRR versus (N 1 ) 60 or versus (N 1 ) 60cs different so different if they are based on largely the same case history data? 2. Can we treat these differences as "epistemic" uncertainty and hence can use all models with "assigned weights"? 1. Can we use site response analyses to obtain CSR or do we have to always use the simplified stress ratio equation? 2. How should we treat liquefaction at depth exceeding those included in liquefaction case histories? Summar y