IN SITU SOIL PROPERTIES FOR SEISMIC SITE RESPONSE
Takaji KokushoProfessor, Dr. of Eng., Registered Eng.
Chuo University, Tokyo
Major References for presentation(1/2):
Aoyagi, T. (2000): Inversion analysis for soil properties based on vertical array record using the extended Bayesian method, Master’s thesis, Graduate school of Chuo University.
Fukushima, Y. and Midorikawa, S.(1994): Evaluation of site amplification factors based on average characteristics of frequency dependent Q-1 of sedimentary strata, J. Struct. Constr. Eng. Architectural Institute of Japan, No.460, 37-46, 1994.
Iwasaki, and Tatsuoka, F. (1978): Shear moduli of sands under cyclic torsional shear loading, Soils and Foundations, Vol.18, No.1, 39-56.
Kokusho, T. (1980): Cyclic triaxial test of dynamic soil properties for wide strain range. Soils and Foundations: Vol.20, No.2, 45-60.
Kokusho, T. (1982): Dynamic soil properties and nonlinear seismic response of ground, PhD Dissertation presented to Tokyo University (in Japanese)
Kokusho,T., Yoshida,Y. and Esashi,Y. (1982): Dynamic soil properties of soft clay for wide strainrange. Soils and Foundations: Vol.22, No.4, 1-18.
Kokusho,T. and Tanaka,Y. (1994): Dynamic properties of gravel layers investigated by in-situ freezing sampling. Geotechnical Special Publication No44 -Ground Failures under Seismic Conditions, ASCE Convention (Atlanta), 121-140.
Major References for presentation(2/2):
Kokusho, T. Sato, K. and Matsumoto, M. (1996): Nonlinear dynamic soil properties back-calculated from strong motions during Hyogoken-Nambu Earthquake. Proc. 11th World Conference on Earthquake Engineering, Acapulco, CD publication.
Kokusho, T. & Matsumoto, M. (1998): Nonlinearity in site amplification and soil properties during the 1995 Hyogoken-Nambu earthquake. Special Issue on Geotechnical Aspects of the January 17, 1995Hyogoken-Nambu Earthquake, No.2; Soils and Foundations: 1-10.
Kokusho, T. and Aoyagi, T. (2001): Insitu nonlinear soil properties back-calculated from vertical array records of 1995 Kobe earthquake, Proc. International Conference on In Situ Measurement of Soil Properties and Case Histories, Indonesian Geotechnical Society, Bali, 473-480.
Kokusho, T. and Mantani, (2002):. Seismic amplification evaluation in a very deep down hole array site, 12th European Conference on Earthquake Engineering, Paper Reference 797.Nuclear Power Engineering Corporation(2001): Study of Evaluation Methods for Seismic Wave Propagation Characteristics, Report of Siting Reliability Studies relating to seismic design for nuclear power plants (in Japanese).
Seed, H. B., Wong, R. T., Idriss, I. M. and Tokimatsu, K. (1986): Moduli and damping factors for dynamic analyses of cohesionless soils, Journal of Geotechnical Engineering, Vol.112, No.11, 1016-1032.
Suetomi, I. (1997): A Program manual for optimization of soil structure by Bayes Method. Central Research Institute of Sato Kogyo Com. Ltd, (in Japanese).
UNSOLVED PROBLEMS FOR SEISMIC SITE RESPONSE
1. In situ modulus degradation compared with laboratory tests.
2. In situ damping compared with laboratory tests.
3. Combination of strain-dependency effect and pore pressure-buildup effect on degradation for larger strains.
4. Amplification in vertical motions and related soil properties.
5. Amplification in deep ground and related soil properties; Effect of confining stress-dependency and frequency dependency of damping.
Topics of the presentation:
Seismic site amplification during strong earthquakes based on vertical
array recordsBack-calculated in situ soil properties
versus lab dataSoil properties for deep soil response
Locations of 4 vertical array sites around Kobe City
0 500 1000 1500 2000 2500
0
20
40
60
80
100
Vp-initialVs-initialN-value
Vp, Vs by PS-logging (m/s)
0 50 100 150 200 2SPT N-value
Dep
th (m
)
WLGL-0m
GL-16.4m
GL-32.4m
GL-83.4m
PI
GL-4.0S/G
CHS
G/S
G
CH
G
Seismograph
0 1000 2000 3000 4000
0
20
40
60
80
100
Vp-initialVs-initialN-value
Vp, Vs by PS-logging (m/s)
Dep
th (m
)
0 100 200 300 40SPT N-value
C
SF
G
M
SM WL
SF
SF
M/G
M
Rock
KNK
GL-2.0
GL-0m
GL-25m
GL-100m
Seimograph
0 1000 2000 3000
0
20
40
60
80
100
Vp-initialVs-initialN-value
Vp, Vs by PS-logging (m/s)
Dep
th (m
)
0 50 100 150 200 250 300 3SPT N-value
SM
M
G
GS
CH
G
M
TKSWL
GL-2.5G
GL-0m
GL-25m
GL-100m
Seismogragh
0 500 1000 1500 2000 2500 3000 3500
0
20
40
60
80
100
Vp-initialVs-initialN-value
Vp, Vs by PS-logging (m/s)
Dep
th (m
)
SPT N-value
SMMGSC
G
CHSFCH
CH
CH
CH
WL
SGK
GL-2.0
0 50 100 150 200 250 300 35GL-0m
GL-25m
GL-97m
Seimograph
Soil profiles, Vp, Vs, SPT-N & Seismometer installation levels in 4 vertical array sites
-100
-80
-60
-40
-20
0
0 100 200 300 400 500 600 700 800 900 1000
A cceleration(gal)
EW (PI) N S U D EW (K N K ) N S U D EW (TK S) N S U D EW (SG K ) N S U DD
epth(m
)
Max. Acc. in 4 vertical array sites during Kobe EQ
Vs-ratio (base to surface) versus amplification ratio of maximum acceleration
0 2 4 6 8 10 120
2
4
6
8
10
12
本震直後の余震
)(
Shima(1978)
1.0
0.33
PI MS PI AS(前震) PI AS(余震) KNK MS KNK AS SGK MS SGK AS TKS MS TKS AS
Surface Acc/Base Acc
Vs ratio: Base/Surface
F.SA.S
AS immediatelyafter MS
Max. Base Acc. versus Acc. amplification (Surf./Base)
1 10 100 10000
2
4
6
8
10
本震直後の余震
)(
PI M S PI A S(前震) PI A S(余震) K N K M S K N K A S TK S M S TK S A S SG K M S SG K A S
Surface Acc/Base Acc
B ase A cc (gal)
F.SA.SAS immediately
after MS
0.01 0.1 1 10 1000
2
4
6
8
10
PI M S PI A S(前震) PI A S(余震) K N K M S K N K A S TK S M S TK S A S SG K M S SG K A S 回帰直線
Surface Vel/Base Vel
B ase V el (kine)
Max. Base Vel. versus Max. Vel. Amplification (Surf./Base)
0 1 2 3 4 5 6 7 8 9 100
2
4
6
8
10( After Shock : before Main Shock
PI EW-direction GL-0m/GL-83.4m
Average
94 06281308
94 07281001
94 10241151
94 11092026
94 11100038
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 100
2
4
6
8
10( After Shock : before Main Shock
PI NS-direction GL-0m/GL-83.4m
Average
94 06281308 94 07281001
94 10241151
94 11092026 94 11100038
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 200
2
4
6
8
10
12( After Shock : before Main Shock
PI UD-direction GL-0m/GL-83.4m
Average
94 06281308
94 07281001 94 10241151
94 11092026 94 11100038
Spectrum ratio
Frequency (Hz)
NS EW
UD
Spectrum ratio
(PI: small shocks before main shock)
Comparison of averaged spectrum
ratios of small shocks before & after main
shock (PI)
0 1 2 3 4 5 6 7 8 9 100
2
4
6
8
PI EW-direction GL-0m/GL-83.4m
berore AfterShock First.peak Amp= 4.74 Freq= 0.78(Hz)after AfterShock First.peak Amp= 4.54 Freq= 0.78(Hz)
before AfterShock Average
after AfterShock Average
Spectrum ratio
Frequency (Hz)0 1 2 3 4 5 6 7 8 9 10
0
2
4
6
8
PI NS-direction GL-0m/GL-83.4m
berore AfterShock First.peak Amp= 3.19 Freq= 0.83(Hz)after AfterShock First.peak Amp= 3.57 Freq= 0.88(Hz)
before AfterShock Average
after AfterShock Average
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 200
2
4
6
8
PI UD-direction GL-0m/GL-83.4m
before AfterShock Average
after AfterShock Average
Spectrum ratio
Frequency (Hz)
NS
UD
EWBefore
After
Before
After
Before
After
0 1 2 3 4 5 6 7 8 9 100
2
4
6
8
SGK EW-direction GL-0m/GL-97m
Average
01170738
01170858
01171305
01152316
02182315
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 100
2
4
6
8
SGK NS-direction GL-0m/GL-97m
Average
01170738
01170858
01171305 01152316
02182315
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 200
2
4
6
8
SGK UD-direction GL-0m/GL-97m
Average
01170738
01170858
01171305
01152316
02182315Spectrum ratio
Frequency (Hz)
NS EW
UD
Spectrum ratio
(SGK: aftershocks)
0 1 2 3 4 5 6 7 8 9 100
5
10
15
TKS EW-direction GL-0m/GL-100m
Average
01170628
01170642
01170738
02182137
Spectrum ratio
Frequency (Hz)0 1 2 3 4 5 6 7 8 9 10
0
2
4
6
8
10
TKS NS-direction GL-0m/GL-100m
Average
01170628
01170642
01170738
02182137
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 200
2
4
6
8
10
TKS UD-direction GL-0m/GL-100m
Average
01170628
01170642
01170738
02182137
Spectrum ratio
Frequency (Hz)
NS EW
UD
Spectrum ratio
(TKS: aftershocks)
0 1 2 3 4 5 6 7 8 9 100
10
20
30
KNK EW-direction GL-0m/GL-100m
Average
01170550
01170553
01170629
01170858
01200047
01230243
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 100
10
20
30
40
50
KNK NS-direction GL-0m/GL-100m
Average
01170550
01170553
01170629
01170858
01200047
01230243
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 200
20
40
60
80
KNK UD-direction GL-0m/GL-100m
Average
01170550
01170553
01170629
01170858
01200047
01230243Spectrum ratio
Frequency (Hz)
NS EW
UD
Spectrum ratio
(KNK: aftershocks)
0 1 2 3 4 5 6 7 8 9 100
1
2
3
4
5
6
7
8
9
10(AfterShock : before MainShock)
PI EW-direction GL-0m/GL-83.4m
Main Shock First.peak Amp= 2.91 Freq= 0.83(Hz)After Shock First.peak Amp= 4.74 Freq= 0.78(Hz)
MainShock
AfterShock Average
Average+SD
Average-SD
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 100
1
2
3
4
5
6
7
8(AfterShock : before MainShock)
PI NS-direction GL-0m/GL-83.4m
Main Shock First.peak Amp= 1.79 Freq= 0.68(Hz)After Shock First.peak Amp= 3.19 Freq= 0.83(Hz)
MainShock
AfterShock Average
Average+SD
Average-SD
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 200
2
4
6
8
10
12(AfterShock : before MainShock)
PI UD-direction GL-0m/GL-83.4m
MainShock AfterShock Average
Average+SD
Average-SD
Spectrum ratio
Frequency (Hz)
NS EW
UDComparison of spectrum ratio of small shocks and
main shock (PI)
0 1 2 3 4 5 6 7 8 9 100
2
4
6
8
10
SGK EW-direction GL-0m/GL-97m
Main Shock First.peak Amp= 8.90 Freq= 0.68(Hz)After Shock First.peak Amp= 3.72 Freq= 0.78(Hz)
MainShock
AfterShock Average
Average+SD
Average-SD
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 100
2
4
6
8
SGK NS-direction GL-0m/GL-97m
Main Shock First.peak Amp= 3.72 Freq= 0.68(Hz)After Shock First.peak Amp= 4.93 Freq= 0.78(Hz)
MainShock
AfterShock Average
Average+SD
Average-SD
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 100
2
4
6
8
SGK UD-direction GL-0m/GL-97m
MainShock
AfterShock Average
Average+SD
Average-SD
Spectrum ratio
Frequency (Hz)
NS EW
UDComparison of averaged spectrum ratio of
aftershocks and main shock (SGK)
0 1 2 3 4 5 6 7 8 9 100
5
10
15
TKS EW-direction GL-0m/GL-100m
Main Shock First.peak Amp=10.96 Freq= 1.12(Hz)After Shock First.peak Amp= 5.83 Freq= 1.22(Hz)
MainShock
AfterShock Average Average+SD
Average-SD
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 100
2
4
6
8
10
TKS NS-direction GL-0m/GL-100m
Main Shock First.peak Amp= 9.46 Freq= 1.12(Hz)After Shock First.peak Amp= 5.49 Freq= 1.42(Hz)
MainShock AfterShock Average
Average+SD
Average-SD
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 200
2
4
6
8
TKS UD-direction GL-0m/GL-100m
MainShock
AfterShock Average
Average+SD
Average-SD
Spectrum ratio
Frequency (Hz)
NS EW
UDComparison of averaged spectrum ratio of
aftershocks and main shock (TKS)
0 1 2 3 4 5 6 7 8 9 100
10
20
30
KNK EW-direction GL-0m/GL-100m
Main Shock First.peak Amp=18.81 Freq= 1.12(Hz)After Shock First.peak Amp=16.34 Freq= 1.12(Hz)
MainShock
AfterShock Average
Average+SD Average-SD
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 100
10
20
30
KNK NS-direction GL-0m/GL-100m
Main Shock First.peak Amp=18.52 Freq= 1.03(Hz)After Shock First.peak Amp=15.65 Freq= 1.07(Hz)
MainShock
AfterShock Average
Average+SD
Average-SD
Spectrum ratio
Frequency (Hz)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 200
10
20
30
40
50
KNK UD-direction GL-0m/GL-100m
MainShock
AfterShock Average
Average+SD
Average-SD
Spectrum ratio
Frequency (Hz)
NS EW
UDComparison of averaged spectrum ratio of
aftershocks and main shock (KNK)
0 5 10 15 20-200
-100
0
100
200
U D
Acceleration(gal)
Tim e(sec)
-800
-400
0
400
800
analysis dom ain
a part of P-waveskip dataAcceleration(gal)
EW
Analysis of vertical motion for initial P-wave
0 2 4 6 8 10 12 14 16 18 200
5
10
15
20
25PI UD-direction -0m /-83.4m
Spectrum ratio
Frequency(Hz)
M ainShock94 6/28 13:0894 7/28 10:0194 10/24 11:5194 11/09 20:2694 11/10 00:38
0 2 4 6 8 10 12 14 16 18 200
5
10
15SG K UD-direction -0m /-97m
M ain Shock 95 1/17 07:38 95 1/17 08:58 95 1/17 13:05 95 1/25 23:16 95 2/18 23:15
Spe
ctrum ratio
Frequency(Hz)
Comparison of spectrum ratios of vertical motion
for initial P-wave between main shock and
aftershocks
0 2 4 6 8 10 12 14 16 18 200
5
10
15
TKS UD -direction 0/100
M ain Shock 1170628 1170642 1170738 2182137
Spectrum ratio
Frequency(H z)
0 2 4 6 8 10 12 14 16 18 200
5
10
15
20
25
30
KNK UD-direction 0/100
M ain Shock 1170550 1170553 1170629 1170858
Spectrum ratio
Frequency(Hz)
Comparison of spectrum ratios of vertical motion
for initial P-wave between main shock and
aftershocks
1 10 100 10000
10
20
30
40
PI Main PI After SGK Main SGK After TKS Main TKS After KNK Main KNK After
Maximum Spectrum ratio
Acceleration at Base Layer (gal)
1 10 100 10000
4
8
12
PI Main PI After SGK Main SGK After TKS Main TKS After KNK Main KNK After
Average Spectrum ratio (0.5Hz-2.5Hz)
Acceleration at Base Layer (gal)
Average values (0.5-2.5Hz)
Peak values
Base Acc. versus Spectrum Ratios of horizontal motions
(Peak & Average values)
1 10 100 10000
10
20
50
60
70
80
PI Main PI After SGK Main SGK After TKS Main TKS After KNK Main KNK After
Maximum Spectrum ratio
Acceleration at Base Layer (gal)
1 10 100 10000
1
2
3
4
5
6
7
8
PI Main PI After SGK MainSGK After
TKS MainTKS After
KNK MainKNK After
Average Spectrum ratio (2.0Hz-8.0Hz)
Acceleration at Base Layer (gal)
Base Acc. versus Spectrum Ratios of vertical motions (Peak & Average values)
Average values (0.5-2.5Hz)
Peak values
SUMMARY ON SITE AMPLIFICATION BASED ON VERTICAL ARRAY RECORDS
•Clear Acc. amplification reduction for increasing base Acc. The same trend for Vel. though milder.
•Steady spectral response for small shocks.
•Reduction in peak frequencies and magnitudes of spectrum ratios between main shocks and small shocks particularly in liquefaction sites.
•Increase in amplification with increasing Vs ratio between base and surface.
•In vertical motions, no difference in peak frequenciesbetween main shocks and small shocks
Topics of the presentation:
Seismic site amplification during strong earthquakes based on vertical
array recordsBack-calculated in situ soil properties
versus lab dataSoil properties for deep soil response
How to back-calculate in situ properties from vertical array records
Optimize Vs (S-wave velocity) and D (damping ratio) based on 1D Multiple Reflection Theory of SH-wave.
Minimize residuals by EBM between observed and computed spectrum ratios.
Main shock properties compared with small shock linear properties for in situ degradation curves.
Identify in situ soil-specific properties.
0 1 2 3 4 5 6 7 8 9 100
5
10
15
PI EW -direction G L-0m /G L-83.4m (A fter Shock) (inversion result)
O BSERV ED A V ERA G E SPEC TRU M RA TIO A V ERA G E+SD
A V ERA G E-SD
O PTIM IZED
Amplification Factor
Frequency(H z)
0 1 2 3 4 5 6 7 8 9 100
5
10
15
PI N S-direction G L-0m /G L-83.4m (A fter Shock) (inversion result)
O BSERV ED A V ERA G E SPEC TRU M RA TIO
A V ERA G E+SD
A V ERA G E-SD
O PTIM IZED
Amplification Factor
Frequency(H z)
Comparison of inversion and observation
(Small shocks before main shock: PI)
NS
EW
0 1 2 3 4 5 6 7 8 9 100
5
10
SG K EW -direction G L-0m /G L-97m (A fter Shock) (inversion result)
O BSERV ED A V ERA G E SPEC TRU M RA TIO
A V ERA G E+SD
A V ERA G E-SD
O PTIM IZED
Amplification Factor
Frequency(H z)
0 1 2 3 4 5 6 7 8 9 100
5
10
SG K N S-direction G L-0m /G L-97m (A fter Shock) (inversion result)
O BSERV ED A V ERA G E SPEC TRU M RA TIO
A V ERA G E+SD
A V ERA G E-SD
O PTIM IZED
Amplification Factor
Frequency(H z)
NS
EW
Comparison of inversion and observation
(Aftershocks:SGK)
0 1 2 3 4 5 6 7 8 9 100
5
10
15
20
TK S EW -direction G L-0m /G L-100m (A fter Shock) (inversion result)
O BSERV ED A V ERA G E SPEC TRU M RA TIO
A V ERA G E+SD
A V ERA G E-SD
O PTIM IZED
Amplification Factor
Frequency(H z)
0 1 2 3 4 5 6 7 8 9 100
5
10
15
20
TK S N S-direction G L-0m /G L-100m (A fter Shock) (inversion result)
O BSERV ED A V ERA G E SPEC TRU M RA TIO
A V ERA G E+SD
A V ERA G E-SD
O PTIM IZED
Amplification Factor
Frequency(H z)
NS
EW
Comparison of inversion and observation
(Aftershocks:TKS)
0 1 2 3 4 5 6 7 8 9 100
10
20
30
40
KNK NS-direction GL-0m/GL-100m (After Shock) (inversion result)
OBSERVED AVERAGE SPECTRUM RATIO
AVERAGE+SD
AVERAGE-SD
OPTIMIZED
Amplification Factor
Frequency(Hz)
0 1 2 3 4 5 6 7 8 9 100
10
20
30
KNK EW-direction GL-0m/GL-100m (After Shock) (inversion result)
OBSERVED AVERAGE SPECTRUM RATIO
AVERAGE+SD
AVERAGE-SD
OPTIMIZED
Amplification Factor
Frequency(Hz)
NS
EW
Comparison of inversion and observation
(Aftershocks:KNK)
0 1 2 3 4 5 6 7 8 9 100
1
2
3
4
P I N S-direction G L -0m /G L -83.4m (M ain Shock) (inversion result)
O B SE R V E D SP E C T R U M R A T IO O P T IM IZ E D
電研
Amplification Factor
F requency(H z)
0 1 2 3 4 5 6 7 8 9 100
2
4
6
8
SG K N S-direction G L -0m /G L -97m (M ain Shock) (inversion result)
O B SE R V E D SP E C T R U M R A T IO
O P T IM IZ E D
電研
Amplification Factor
F requency(H z)
PI NS
SGKNS
Comparison of inversion and observation (Main
shock: PI & SGK) Horizontal
0 1 2 3 4 5 6 7 8 9 100
4
8
12
T K S N S-direction G L -0m /G L -100m (M ain Shock) (inversion result)
O B SE R V E D SP E C T R U M R A T IO O P T IM IZ E D
電研
Amplification Factor
F requency(H z)
0 1 2 3 4 5 6 7 8 9 100
5
10
15
20
K N K N S-direction G L -0m /G L -100m (M ain Shock) (inversion result)
O B SE R V E D SP E C T R U M R A T IO
O P T IM IZ E D
電研Amplification Factor
F requency(H z)
TKSNS
KNKNS
Comparison of inversion and observation (Main shock: TKS & KNK)
-80
-60
-40
-20
0
0 100 200 300 400 500
-80
-60
-40
-20
00 100 200 300 400 500
INITIAL OPTIMIZED AS EW OPTIMIZED AS NS OPTIMIZED MS EW OPTIMIZED MS NS Water table GL-4.0m
Shear wave velocity : Vs (m/s)
Depth from G
L (m)
-100
-80
-60
-40
-20
0
0 100 200 300 400 500 600-100
-80
-60
-40
-20
00 100 200 300 400 500 600
Shear wave velocity : V s (m /s)
Depth from G
L (m)
IN ITIA L O PTIM IZED A S EW O PTIM IZED A S N S
O PTIM IZED M S EW O PTIM IZED M S N S W ater table G L-2m
PI
SGK
Optimized Vs distribution along depth for MS & AS
compared with wave-logging (PI & SGK)
-100
-80
-60
-40
-20
0
0 100 200 300 400 500 600 700-100
-80
-60
-40
-20
00 100 200 300 400 500 600 700
Shear wave velocity : Vs (m /s)
Depth from G
L (m)
INITIAL
O PTIM IZED AS EW O PTIM IZED AS NS
O PTIM IZED M S EW O PTIM IZED M S NS
W ater table G L-2m
-100
-80
-60
-40
-20
0
0 200 400 600 8001000120014001600-100
-80
-60
-40
-20
00 200 400 600 8001000120014001600
Shear wave velocity : Vs (m/s)
Depth from G
L (m)
INITIAL OPTIMIZED AS EW OPTIMIZED AS NS OPTIMIZED MS EW OPTIMIZED MS NS Water table GL-2.2m
TKS
KNK
Optimized Vs distribution along depth for MS & AS
compared with wave-logging (TKS & KNK)
0 500 1000 1500 2000 2500
0
20
40
60
80
100
PS検層
本震同定
P波速度 Vp (m/s)
深度 (m)
地下水位埋立土
軟弱粘土
砂
GL 0
GL-16.4
GL-32.4
GL-83.4m
洪積礫
(t=0~5.12sのみを解析)
Comparison of inversion and observation (Main shock: PI ) Vertical
Wave logging
Optimized for Main shockt=0~5.1 s for intial P-wave
P-wave velocity: Vp (m/s)
Dep
th (
m)
Water tableFill
Soft clay
Sand
Gravel
0 2 4 6 8 100
4
8
12
(b) G L.0m /G L-97m
Spectrum ratio
Frequency(Hz)
0
4
8
12
16SG K Aftershock-A (NS)
(a) G L.0m /G L-24.9m
Spectrum ratio
0 2 4 6 8 100
4
8(b) G L.0m /-97m
Spe
ctrum ratio
Frequency (Hz)
M easured Initial guess Back-calculated L.M .S.M .
0
4
8SG K M ain shock (NS)
(a) G L.0m /-29.4m
Spe
ctrum ratio
0 2 4 6 8 100
4
8
12
(b) G L.0m /G L-97m
Spe
ctrum ratio
Frequency(Hz)
0
4
8
12
16 SG K Aftershock-C (NS)
(a) G L.0m /G L-24.9m
Spectrum ratio
M easured Initial guess Back-calculated
0 2 4 6 8 100
4
8
12
(b) G L.0m /G L-97m
Spectrum ratio
Frequency (Hz)
0 2 4 6 8 100
4
8
12
16 SG K Aftershock-B (NS)
(a) G L.0m /G L-24.9m
Spe
ctrum ratio
Measured & Back-calculated spectrum ratios compared in SGK for Aftershocks C, B, A and Main Shock
AS-C AS-B
AS-A MS
-100
-80
-60
-40
-20
00 4 8 12 16 20
(b) Dam ping ratio (%)
M S(EW ) M S(NS) AS-A(EW ) AS-A(NS) AS-B(EW ) AS-B(NS) AS-C(EW ) AS-C(NS)
-100
-80
-60
-40
-20
00 200 400 600
S-wave logging
(a) S-wave velocity Vs (m /s)
Depth (m)
Back-calculated S-wave velocity and Damping in SGK for Aftershocks and Main Shock
( )20 MS ASG G Vs Vs=
5 6 7 8 9 1 0 1 1- 4 0
- 2 0
0
2 0
4 0
(a ) S G K A S - B N S - d ire c tio n
T im e (s )
In p u t W a v e
- 4 0
- 2 0
0
2 0
4 0
GL-24.9m
GL-97m
Acceleration(cm/s
2 ) - 4 0- 2 0
0
2 0
4 0
GL-0m
O b s e rv e d C a lc u la te d
1 0 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 8 1 9 2 0- 5 0 0
- 2 5 0
0
2 5 0
5 0 0
(b ) S G K M S N S - d ire c tio n
G L - 9 7 m
Acceleration(cm/s2)
In p u t W a v e
T im e (s )
- 5 0 0
- 2 5 0
0
2 5 0
5 0 0
G L - 2 4 .9 m
- 5 0 0
- 2 5 0
0
2 5 0
5 0 0
G L - 0 m
O b s e rv e d C a lc u la te d
Measured & Back-calculated Accelerogramscompared in SGK for Aftershocks and Main Shock
-100
-80
-60
-40
-20
0
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20
T K S M ain Shock (01170546)
EW N S
M ax.Shear Strain(%)
Depth from G
L(m
)
-100
-80
-60
-40
-20
0
0.000.010.020.030.040.050.060.070.080.090.10
KNK Main Shock (01170546)
EW NS
Max.Shear Strain (%)
Depth from G
L (m)
-80
-60
-40
-20
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
PI M ain Shock (01170546)
EW N S
M ax.Shear Strain (%)
Depth from G
L (m)
-100
-80
-60
-40
-20
0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
SG K M ain Shock (01170546)
EW N S
M ax.Shear Strain (%)
Depth from G
L (m)
Computed Max. shear strain versus Depth in 4 vertical array sites
max0.65
effγ γ= ⋅
PI SGK
TKS KNK
1E-6 1E-5 1E-4 1E-3 0.01
0.0
0.2
0.4
0.6
0.8
1.0
0
5
10
15
20
25
G /G0
SG K PI KNK
C lay: Ip=40-83by Kokusho et al.(1982)
(a) C lay
Shear m
odulus ratio G/G
0
Effective shear strain γeff
D SG K PI KNK
Dam
ping ratio D (%)
1E-6 1E-5 1E-4 1E-3 0.01
0.0
0.2
0.4
0.6
0.8
1.0
0
5
10
15
20
25
Shear m
odulus ratio G/G
0
Effective shear strain γeff
G /G0
SG K PI TKS KNK
Liquefied
(c) Sand
Dam
ping ratio D (%)
D SG K PI TKS KNK
1E-6 1E-5 1E-4 1E-3 0.01 0.1
0.0
0.2
0.4
0.6
0.8
1.0
0
5
10
15
20
25
C lay: Ip=40-83by Kokusho et al.(1982)
Sand: σc'=98 kPa
by Kokusho (1980)
Shear m
odulus ratio G/G
0
Effective shear strain γeff
Dam
ping ratio D (%)
(b) Silt
1E-6 1E-5 1E-4 1E-3 0.01
0.0
0.2
0.4
0.6
0.8
1.0
0
5
10
15
20
25
Shear m
odulus ratio G/G
0
Effective shear strain γeff
(d) G ravel
Dam
ping ratio D (%)
1E-6 1E-5 1E-4 1E-3 0.01
0.0
0.2
0.4
0.6
0.8
1.0
0
5
10
15
20
25
Shear m
odulus ratio G/G
0
N orm alized effective shear strain γeff/(σ'
c/p
0)0.5
G ravel: σc'=98 kPa
by Kokusho and Tanaka (1994)
(f) G ravel (Norm alized by confining stress)
Dam
ping ratio D (%)
1E-6 1E-5 1E-4 1E-3 0.01
0.0
0.2
0.4
0.6
0.8
1.0
0
5
10
15
20
25
SandIw asaki et al.(1978)
Shear m
odulus ratio G/G
0
N orm alized effective shear strain γeff/(σ'
c/p
0)0.5
Sandby Seed-Idriss(1970)
Sand: σc'=98 kPa
by Kokusho(1980)
(e) Sand (Norm alized by confining stress)
Dam
ping ratio D (%) G/G0~γeff
relationships for different soils
In Japan, How to evaluate strain-dependent shear modulus and damping ratio in laboratory.
1) Use triaxial device with inner load cell.
2) Employ high-sensitivity gap sensors or LDT for vertical disp. measurement.
3) Undrained cyclic loading test for f=1~0.1 Hz shear modulusand damping for wide strain range of 10-6-10-2.
4) For small strain modulus, pulse tests or bender element test are also employed. Resonant column test is not so popular.
LDT
Measured by gap sensor kokusho (1980)
Measured by conventional sensor
Shear modulus versus strain of sand for different confining stress
0
0.2
0.4
0.6
0.8
1
1.E-06 1.E-05 1.E-04 1.E-03 1.E-02
SHEAR STRAIN; γ
SHEA
R MO
DULU
S RA
TIO; G
/G0
σc'=300 KN /m 2
σc'=200 KN /m 2
σc'=100 KN /m 2
σc'=50 KN /m 2
σc'=20 KN /m 2
0
50
100
150
200
250
1.E-06 1.E-05 1.E-04 1.E-03 1.E-02
SHEAR STRAIN; γ
SHEA
R MODU
LUS ; G
(x 100
0 kP
a)
N o.58(e=0.645, σc'=300 KN /m 2) N o.48(e=0.651, σc'=200 KN /m 2)
N o.49(e=0.643, σc'=200 KN /m 2) N o.53(e=0.641, σc'=200 KN /m 2)
N o.47(e=0.649, σc'=100 KN /m 2) N o.50(e=0.644, σc'=100 KN /m 2)
N o.57(e=0.640, σc'=100 KN /m 2) N o.51(e=0.644, σc'=50 KN /m 2)
N o.52(e=0.641, σc'=50 KN /m 2) N o.55(e=0.644, σc'=50 KN /m 2)
N o.54(e=0.646, σc'=20 KN /m 2) N o.56(e=0.646, σc'=20 KN /m 2)
N o.59(e=0.636, σc'=20 KN /m 2)
Toyoura sand
Hysteretic damping ratio of sand for different confining stress
0
0.05
0.1
0.15
0.2
0.25
1.E-06 1.E-05 1.E-04 1.E-03 1.E-02
SHEAR STRAIN; γ
HYSTERETIC DAMPING RAITO; D
N o.58(σc'=300 KN/m 2)
No.48,49,53(σc'=200 KN/m 2)
No.47,50,57(σc'=100 KN/m 2)
No.51,52,55(σc'=50 KN/m 2)
No.54,56,59(σc'=20 KN/m 2)
Toyoura sand
Shear modulus ratio versus strain of clay for different Ip
0
0.2
0.4
0.6
0.8
1
1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01
SHEAR STRAIN; γ
SHEAR M
ODULUS RATIO; G/G
0
S -1-1 (IP=90,σc'=16 KN/m 2) S-1-3 (IP=38,σc'=16 KN/m 2) S-2-1 (IP=96,σc'=16 KN/m 2)
S-2-3 (IP=85,σc'=16 KN/m 2) S-4-1 (IP=52,σc'=31 KN/m 2) S-4-3 (IP=54,σc'=29 KN/m 2)S-5-1 (IP=57,σc'=29 KN/m 2) S-5-3 (IP=52,σc'=36 KN/m 2) S-6-1 (IP=46,σc'=46 KN/m 2)
S-6-2 (IP=56,σc'=45 KN/m 2) S-6-1 (IP=49,σc'=46 KN/m 2) S-8-1 (IP=41,σc'=62 KN/m 2)S-9-1 (IP=NP,σc'=69 KN/m 2) S-9-2 (IP=14,σc'=68 KN/m 2) Toyoura sand σc'=100 KN/m 2
Ip > 80
Ip = 50-60
Ip = 40-50
Ip = 30-40
Ip = 10-20
N PToyoura sand
Intact Teganuma clay
Effect of confining stress on modulus degradation
0
0.2
0.4
0.6
0.8
1
1.E-05 1.E-04 1.E-03 1.E-02 1.E-01
SHEAR STRAIN; γ
SHEAR M
ODULUS RATIO; G
/G
0
S -6-2 (IP=56,σc'=45 KN/m 2)
S-6-4 (IP=44,σc'=100 KN/m 2)
S-6-5 (IP=38,σc'=300 KN/m 2)
S-6-3 (IP=54,σc'=500 KN/m 2)
Intact Teganuma clay
Modulus & Damping ratio versus strain of gravel for different confining stress
0
0.2
0.4
0.6
0.8
1
1.2
1.E-06 1.E-05 1.E-04 1.E-03 1.E-02
SHEAR STRAIN; γ
SHEA
R MODULU
S RATIO; G
/G0
σc'=75 KN/m 2, G 0=103 M Pa
σc'=100 KN/m 2, G 0=100 M Pa
σc'=200 KN/m 2, G 0=206 M Pa
σc'=400 KN/m 2, G 0=288 M Pa
0
0.05
0.1
0.15
0.2
1.E-06 1.E-05 1.E-04 1.E-03 1.E-02
SHEAR STRAIN; γ
HYST
ERET
IC DAM
PING
RAITO
; h
Intact Pleistocene
gravel
1E-6 1E-5 1E-4 1E-3 0.01 0.1
0.0
0.2
0.4
0.6
0.8
1.0
Lab test results
G ravel(98kPa)
Sand(98kPa)
C lay(Ip=40-83)
(a) Shear m odulus ratio
Liquefied
C lay Silt Sand G ravel
Shear m
odulus ratio G/G
0
Effective shear strain γeff or γ
eff/(σ'
c/p
0)0.5
Back-calculated strain-dependent modulus degradations of 4 types of soils compared with previous lab tests
1E-6 1E-5 1E-4 1E-3 0.01 0.1
0
5
10
15
20
25
G ravel(98kPa)
C lay(Ip=40-83)
Sand (98kPa)
C lay Silt Sand G ravel
(b) D am ping ratio
Dam
ping ratio D (%)
Effective shear strain γeff or γ
eff/(σ'
c/p
0)0.5
Back-calculated damping ratios versus strain of 4 types of soils compared with previous lab tests
1E-6 1E-5 1E-4 1E-3 0.01
0.0
0.2
0.4
0.6
0.8
1.0
0
5
10
15
20
25
G /G0
SG K PI KNK
C lay: Ip=40-83by Kokusho et al.(1982)
(a) C lay
Shear m
odulus ratio G/G
0
Effective shear strain γeff
D SG K PI KNK
Dam
ping ratio D (%)
Back-calculated modulus degradation and damping compared with lab test (Clay)
1E-6 1E-5 1E-4 1E-3 0.01 0.1
0.0
0.2
0.4
0.6
0.8
1.0
0
5
10
15
20
25
C lay: Ip=40-83by Kokusho et al.(1982)
Sand: σc'=98 kPa
by Kokusho (1980)
Shear m
odulus ratio G/G
0
Effective shear strain γeff
Dam
ping ratio D (%)
(b) Silt
Back-calculated modulus degradation and damping compared with lab test (Silt)
1E-6 1E-5 1E-4 1E-3 0.01
0.0
0.2
0.4
0.6
0.8
1.0
0
5
10
15
20
25
SandIw asaki et al.(1978)
Shear m
odulus ratio G/G
0
N orm alized effective shear strain γeff/(σ'
c/p
0)0.5
Sandby Seed-Idriss(1970)
Sand: σc'=98 kPa
by Kokusho(1980)
(e) Sand (Norm alized by confining stress)
Dam
ping ratio D (%)
Back-calculated modulus degradation and damping compared with lab test (Sand)
1E-6 1E-5 1E-4 1E-3 0.01
0.0
0.2
0.4
0.6
0.8
1.0
0
5
10
15
20
25
Shear m
odulus ratio G/G
0
N orm alized effective shear strain γeff/(σ'
c/p
0)0.5
G ravel: σc'=98 kPa
by Kokusho and Tanaka (1994)
(f) G ravel (Norm alized by confining stress)
Dam
ping ratio D (%)
Back-calculated modulus degradation and damping compared with lab test (Gravel)
SUMMARY ON BACK-CALCULATED PROPERTIES (I)In spectrum ratios, back-calculation overestimates lower frequency peaks and under-estimates higher frequency peaks compared to observation presumably due to strain-dependency and frequency-dependency of properties.
Clear differences in S-wave velocities are recognized not only between the main shock and aftershocks but also among aftershocks of different intensities.
Damping ratios for the main shock are evidently larger than aftershocks.
Clear modulus degradations can be identified from the back-calculated S-wave velocities.
Degradations are almost consistent at 4 sites, from which soil-specific curves can be differentiated for clay, silt, sand and gravel.
SUMMARY ON BACK-CALCULATED PROPERTIES (II)Confining stresses have significant effect on the degradations of non-cohesive soils in situ, too.
Back-calculated damping ratios increase with increasing effective strains for 10-4 or larger. Damping ratio in liquefied sand layers may be back-calculated as very large values (46-52%).
The majority of back-calculated damping ratios in small strain ranges are a few percent higher than laboratory test results despite apparent splits in the back-calculated damping values (1-6%).
A fair agreement in modulus degradation can be recognized between back-calculation and lab test results for clays and sands except for large strain level.
For gravels, back-calculated degradations are milder than laboratory tests presumably reflecting appreciable fine content in in situ soils.
Topics of the presentation:
Seismic site amplification during strong earthquakes based on vertical
array recordsBack-calculated in situ soil properties
versus lab dataSoil properties for deep soil response
Higashinadadeep hole site
VsVp
Higashinada deep hole site (GL-1670m), seismometer installation levels and Vs, Vp distribution along depth
Seismometer
Acceleration amplification in deep soil ground for 3 larger earthquakes
(M>6.0, Max.Surf. Acc~10 cm/s2)
EQ1 EQ2 EQ3
Spectrum ratio (3 larger earthquakes M>6.0)
Spec
trum
ratio
Spec
trum
ratio
Spec
trum
ratio
Spec
trum
ratio
GL-0m/GL-20m
GL-0m/GL-50m GL-0m/GL-1670m
GL-0m/GL-750m
Variation of D0 and m along depth for frequency-dependent damping ratio
0.01 0.1 1
0
500
1000
1500
2000
Damping at f=1.0 Hz (D0) or m
Depth(m)
0mD D f −=
mD0
Dam
ping
ratio
x 2
In situ frequency-dependent damping ratio based on previous papers (Fukushima and Midorikawa, 1994)
-1600
-1400
-1200
-1000
-800
-600
-400
-200
0
0.0 0.5 1.0 1.5 2.0 2.5
Density
Density(t/m3)
Depth(m
)
0 1000 2000 3000
-1600
-1400
-1200
-1000
-800
-600
-400
-200
0 S-wave velocity
Vs(m/s)
Soil model, Vs and density distribution along depth
0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 0
- 6 0 0
- 4 0 0
- 2 0 0
0
2 0 0
4 0 0
6 0 0 (a ) m a x = 6 0 9 .4 (c m / s 2 ):tim e = 5 .3 2 (s)
Acc
eleration (cm
/s2)
T im e (s)
0.1 1 100
50
100
150
200
250
300 (b)
A nalyzed frequency range(U p to 20 H z)
Fourier spec. (cm/s2*s)
F re quency (H z)
Accelerogram and Fourier spectrum of input motion (PI)
1
10
100
1000
0.1 1 10 100
Dep
th fr
om G
L (m
)
Damping ratio D0 (%)
Kokusho & Matsumoto 1998
Kokusho et al.1992Yamamizu et al.1983
Shear strain:
γ≤10 −4
γ>10 −310−3≥ γ≥10 −4
(1995 Kobe EQ)
Damping ratio in deep soil ground assuming 0mD D f −= ⋅
-1400
-1200
-1000
-800
-600
-400
-200
0
0.0 0.2 0.4 0.6 0.8 1.0
Case-A
m =0.5 m =1.0
(G0-G)/G
0
Depth(m
)
( )0.5r vγ σ ′∝For cohesive soils,too.-1400
-1200
-1000
-800
-600
-400
-200
0
0.0 0.2 0.4 0.6 0.8 1.0
Case-B
m=0 m=0.5 m=1.0
(G0-G)/G
0
Depth(m
)Degree of
nonlinearity
versus depth
( )0 0G G G−
.r co n stγ = for cohesive soils.
( )0.5r vγ σ ′∝ for cohesive soils,too.
-1600
-1400
-1200
-1000
-800
-600
-400
-200
0
100 1000 10000
Linear
Equivalent linear
Case-A
m=0 m=0.5 m=1.0
M axim um acceleration (gal)
Depth(m
)
-1600
-1400
-1200
-1000
-800
-600
-400
-200
0
100 1000 10000
Linear
Equivalent linear
Case-B
m=0.0 m=0.5 m=1.0
M aximum acceleration (gal)
Depth (m)
Effect of m on Max. Acc. versus Depth relationship
.r co n stγ = for cohesive soils.
( )0.5r vγ σ ′∝ for cohesive soils,too.
-1600
-1400
-1200
-1000
-800
-600
-400
-200
0
0 1000 2000 3000 4000 5000 6000
Linear
Equivalent linear
M axim um Acceleration (gal)
Depth(m
) Basic case * 1/2 * 1/4
Effect of reducing damping ratio on
Max.Acc. distribution
10 100 10000.1
0.2
0.4
0.6
0.81
2
Dam
ping ratio D
0 (%)
Basic value (from Fig.4) x1/2 x1/4
D epth(m )
0.0 0.2 0.4 0.6 0.8 1.00
500
1000
1500
2000
2500
3000
3500(cm /s2)
Maximum
base acceleration
C onstant m
γ'=const.: for cohesive soils. : for cohesive soils, too.( )0.5r vγ σ ′∝
0mD D f −= ⋅
Effect of m on Max. Base Acc. (GL-1670 m)
SUMMARY ON RESPONSE IN DEEP SOIL GROUND
Even during the destructive Kobe earthquakes, calculated modulus degradation is milder than 20% deeper than about 300m.
Acceleration tends to increase with increasing depth in analyses based on frequency-independent damping, indicating that the damping should inevitably be frequency-dependent in analyzing deep ground.
Assumptions on soil properties such as frequency-dependency in damping or reference strain have a great effect in analytical seismic response in deep soil ground.
SUMMARY & FURTHER RESEARCH (I)
In Situ PropertiesLaboratory modulus degradation for individual soil types are applicable to the field at least up to medium strain range. Field damping is only qualitatively consistent with lab test damping.
More quantitative research for in situ properties; -Modulus degradation for large strain range considering
strain-dependency and PP-buildup. --Damping ratio for small strain and its mechanism -Large strain damping considering cyclic mobility and
PP-buildup.
SUMMARY & FURTHER RESEARCH (II)In vertical motions, little nonlinearity of amplification and back-calculated Vp.
Deep Soil PropertiesStrain-dependent soil nonlinearity may be ignorable in depth of a few hundred meters of soil ground even during destructive earthquakes.
Based on deep-hole data analyses, damping is no doubt frequency-dependent.
Effect of high overburden stress on modulus and damping in deep soil should be investigated further.