Geotechnical Earthquake
Engineering
by
Dr. Deepankar Choudhury Humboldt Fellow, JSPS Fellow, BOYSCAST Fellow
Professor
Department of Civil Engineering
IIT Bombay, Powai, Mumbai 400 076, India.
Email: [email protected]
URL: http://www.civil.iitb.ac.in/~dc/
Lecture – 29
IIT Bombay, DC 2
Module – 7
Seismic Hazard Analysis
IIT Bombay, DC 3
Code Implications
UBC 10% probability of exceedance in 50 years.
IIT Bombay, DC 4
Code Implications
AASHTO 2% probability of exceedance in 50 years ??
IIT Bombay, DC 5
Example
Case Study
on Gujarat, India
Ref: Ph.D. Thesis of Jaykumar Shukla (2013),
IIT Bombay, Mumbai, India.
IIT Bombay, DC 6
Location of Gujarat, India
Urban areas selected …
Kachchh
•Anjar
•Bhuj
•Dholavira
•Gandhidham
•Mandavi
Saurashtra
•Amreli
•Bhavnagar
•Dholera
•Dwarka
•Jamnagar
•Junagadh
•Morvi
•Porbundar
•Rajkot
•Surendranagar
•Veraval
Mainland Gujarat
•Ahmedabad
•Baroach
•Gandhinagar
•Mehsana
•Palanpur
•Patan
•Surat
•Vadodara
•Valsad
25 Cities studied representing the all seismic zones in Gujarat. 4 port sites
are also studied for site specific ground motion estimations (Kandla port,
Mundra Port, Hazira Port and Dahej Port ) 7/11/2013
Seismic Zones of Gujarat, IS: 1893-Part I (2002)
Location of urban areas selected…
7/11/2013
Components of hazard study
Seismic Hazard
Estimation for Gujarat region
Earthquake Catalogue
Regional Seismicity
Parameters
Deterministic Seismic Hazard
Analysis
(DSHA) Probabilistic Seismic Hazard
Analysis
(PSHA)
Site specific ground motions
for Ports
Sensitivity Analysis
7/11/2013
Seismicity of Gujarat, India
7/11/2013
Seismic Zones of Gujarat Region as per
IS:1893 – Part I (2002) Epicenters of earthquakes recorded
from 2007 to 2011 ( modified
after ISR report 2010-11)
Seismotectonic setting of the region Gujarat, India
Pakistan
Arabian
Sea
Seismicity across Gujarat Year Region Numbers of Earthquake recorded
Mw≥4 Mw(3 to
3.9)
Mw(2 to
2.9)
Mw<2 Total
2010-
11*
Kachchh 1 51 292 1155 1499
Saurashtra 2 12 101 571 684
Mainland
Gujarat
1 2 11 38 52
2009
Kachchh 4 72 422 1594 2092
Saurashtra --- 6 22 420 448
Mainland
Gujarat
--- --- 36 65 41
2008
Kachchh 5 52 343 493 893
Saurashtra 2 12 221 424 659
Mainland
Gujarat
--- 3 17 26 46
Note: * up to March 2011
7/11/2013
Kachchh Saurashtra Mainland Gujarat
0
10
20
30
40
50
60
70
80
90
100
% o
f to
tal
Eart
hq
uak
e o
ccu
red
in
Gu
jara
t
Gujarat Region
2010-11*
2009
2008
The seismicity within the Gujarat is not same across the Gujarat. The single seismicity
parameter for entire Gujarat may not represent the true seismicity within the Gujarat
Seismicity in Saurashtra
Seismicity migration is observed in the Saurashtra
7/11/2013
Modified after ISR Report 2009
Catalogue completeness
Catalogue
Completeness is
evaluated in the
present study using
CUVI (Tinti and
Mulargia, 1985)
method and Stepp’s
method (Stepp, 1973)
7/11/2013
0
20
40
60
80
100
120
Cu
mu
lati
ve E
art
hq
uak
e o
ccu
ren
ce
1962
1800 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000 2020
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
Eart
hq
uak
e M
om
en
t M
ag
nit
ud
e (
Mw
)
Time (Years)
Shukla and Choudhury (2012) in NHESS, 12, 2019-2037.
Earthquake Moment Magnitude (Mw)
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
Log
N
0.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00
2.25
Entire Gujarat
Saurashtra
Mainland Gujarat
Kachchh
Regional Seismicity parameters
7/11/2013
Gutenberg – Richter recurrence
relations are derived using Least
Square Fit method using prepared
earthquake catalogue for Mw ≥ 4.
Region Past seismicity
used (Year) a-value b-value G-R relation R2
Saurashtra 135 4.03 0.64 Log N=4.03 – 0.64 Mw 0.9783
Mainland 175 4.02 0.62 Log N=4.02 – 0.62 Mw 0.9370
Kachchh 189 3.41 0.41 Log N=3.41 – 0.41 Mw 0.9821
Gujarat 189 4.13 0.51 Log N=4.13 – 0.51 Mw 0.9899
Choudhury and Shukla (2011) in Disaster Advances, 4(2), 47-59.
Rastogi et al. (2013)
recommended b value
= 0.67 for Saurashtra
region.
b-value using ML method
• Another popular method for estimation of b-
value is by using Maximum Likelihood (ML)
method (Aki, 1965; Utsu, 1965)
7/11/2013
min
1
ln(10)( )b
u m
u is the sampling average of the magnitudes
Region b-value using ML estimate
Kachchh 0.526
Saurashtra 0.572
Mainland Gujarat 0.642
Probability models for earthquake recurrence
• In time predictable methodology many researchers has applied
various probability models to predict the next earthquake within the
some specified time. In another words fitting the recurrence time of
earthquake using various probability distributions.
• Key researcher are Utsu (1984); Nishenko and Buland (1987);
Sykes and Nishenko (1984); Rikitake (1991); Shimazaki (2002);
Kagan and Knopoff (1987); Papazachos (1989); Ferreas (2003,
2005); Yilmaz et al. (2004); Shankar and Papadimitriou (2004) and
many others.
• Specifically for Indian Peninsula, Pervez and Ram (1997, 1999);
Tripathi (2006); Jaiswal (2006) and Yadav et al. (2008)
7/11/2013
Various Probability Distributions No Year Month Date
Date
(Years)
Latitude
(0N)
Longitude
(0E) Mw
Recurrenc
e Time
(Years
Location
1 1819 6 16 1819.5 24 69 7.8 Kachchh
2 1845 4 19 1845.333 23.8 68.9 6.3 25.833 Lakhpat
3 1848 4 26 1848.333 24.4 72.7 6 3 Mount Abu
4 1856 12 25 1857 20 73 5.7 8.667 Surat
5 1864 4 29 1864.333 22.3 72.8 5.7 7.333 Ahmedabad
6 1871 1 31 1871.083 21.2 72.9 5 6.75 Surat
7 1872 4 14 1872.372 21.75 72.15 5 1.289 Bhavnagar
8 1882 6 10 1882.5 23.2 71.38 5 10.128 Bhachau
9 1903 1 14 1903.083 24 70 5.6 20.583 Kachchh
10 1919 4 21 1919.391 22 72 5.7 16.308 Bhavnagar
11 1921 10 26 1921.833 25 68 5.5 2.442 Kachchh
12 1935 7 20 1935.583 21 72.4 5.7 13.75 Surat
13 1938 3 14 1938.25 21.6 75 6 2.667 Satpura
14 1950 6 14 1950.5 24 71.2 5.3 12.25 Kachchh
15 1956 7 21 1956.583 23.3 70 6 6.083 Kachchh
16 1963 7 13 1963.583 24.9 70.3 5.3 7 Pakistan
17 1965 3 26 1965.25 24.4 70 5.1 1.667 Kachchh
18 1966 5 27 1966.417 24.46 68.69 5 1.167 Pakistan
19 1970 2 13 1970.167 24.6 68.61 5.2 3.75 Kachchh
20 1976 6 4 1976.5 24.51 68.45 5.1 6.333 Allah Band
21 1985 4 7 1985.333 24.36 69.74 5 8.833 Kachchh
22 1993 8 24 1993.732 20.6 71.4 5 8.399 Rajula
23 2001 1 26 2001.083 23.44 70.31 7.7 7.351 Kachchh
24 2006 3 7 2006.25 23.79 70.73 5.7 5.167 Gedi,
25 2007 11 6 2007.933 21.16 70.54 5 1.683 Junagadh
7/11/2013
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Cu
mu
lati
ve P
rob
ab
ilit
yEarthquake (M
w >= 5) in Gujarat region
Pareto Distribution
Rayleigh Distribution
Weibull Distribution
Exponential Distribution
Earthquakes
Choudhury and Shukla (2011) in Disaster Advances, 4(2), 47-59.
Selection of the best distribution
7/11/2013
Exponential
Distribution
Rayleigh
Distribution
Pareto
Distribution
Weibull
Distribution
Parameters μ= 7.851 δ=11.436 α=0.573 θ=0.118
Maximum Log
Likelihood
-76.016 -74.255 -79.189 -71.868
A-D value 0.760 2.643 0.841 0.273
Modified K-S test
value
0.164 0.277 0.170 0.105
P-value 0.554 0.051 0.502 0.955
( )
T
μef T
μ2 2
( )2
T Tf T Exp
δ δ
( 1)
0 0( , , ) α αf T α x αx T
1( ) ( )ηη θTf T θη T e
Weibull Model Exponential Model
Pareto Model
Rayleigh Model
Recurrence Estimation
7/11/2013
Probability
Distribution
Model
Recurrence
interval
(years)
Predicted
Last Event
occurred on
Next
Earthquake
Expected on
Study Date
Considered
(Nov 10th
2009)
Year Left
from Present
Date
Next
earthquake
expected
before
Exponential 7.853 2007.933 2015.786 2009.85 5.936 Oct 2015*
Rayleigh 16.173 2007.933 2024.106 2009.85 14.256 Feb 2024
Pareto 3.135 2007.933 2011.068 2009.85 1.218 Jan 2011
Weibull 7.011 2007.933 2014.944 2009.85 5.094 Dec 2014*
*Note: This research output published in Journal Disaster Advances in Aug.
2011 was validated by actual occurrence of earthquake of September 2011.
Choudhury and Shukla (2011) in Disaster Advances, 4(2), 47-59.
Comparison of Earthquake return periods
7/11/2013 Choudhury and Shukla (2011) in Disaster Advances, 4(2), 47-59.
b-value proposed & those by other researchers
Study
Number
Application area b- value Reference Periods for
study taken
1 Kachchh 0.417 Based on least square fit, Present Study (1820-2008)
2 Saurashtra 0.64 (1872-2008)
3 Mainland 0.62 (1872-2008)
4 Entire 0.51 (1820-2008)
5 Kachchh 0.526 Based on ML estimate, Present Study (1820-2009)
6 Saurashtra 0.572 (1872-2009)
7 Mainland 0.642 (1872-2009)
8 Saurashtra 0.67 Rastogi et al. (2013) (1970-2010)
9 Gujarat
0.87
( 0.06)
WCE NDMA (2010) (*1800-2009)
10 Gujarat 0.72 Tripathi et al., (2005) -
11 Kachchh 0.43 Ashara et al., (2006) -
12 Kachchh 0.71
0.03
Jaiswal (2006) (1842-2002)
13 Gujarat
0.7 to 0.9
0.07
Raghukanth (2010) (1250-2008)
14 Gujarat
0.4 to 0.6 Kolathayar et al. (2011)
for Clustered catalogue
(250 B.C. -2010)
15 Gujarat
0.4 to 0.8 Kolathayar et al. (2011)
for declustered catalogue
(250 B.C. -2010)
16 Peninsular 0.92 Jaiswal and Sinha (2007) (1842-2002)
17 Gujarat region 0.55 Bhatia et al. (1999) -
18 Gujarat 0.89 Thaker et al. (2012) 1818-2008
Shukla and Choudhury (2012) in NHESS, 12, 2019-2037.
• Entire Gujarat is divided into three regions
– Kachchh
– Saurashtra
– Mainland Gujarat
• Earthquake catalogue is divided as per these three regions
• Only fault sources are used as seismic sources
• Poisson distribution for earthquake occurrence
• All the faults are Normal faults, depth ranging 10 to 15km from
ground surface.
DSHA - Some starting points
DSHA requirements…
Seismicity model:
• describes geographical distribution of potential active
source zones (seismotectonic sources) and distribution of
magnitudes in each source. (Fault Map and Seismicity
parameters- maximum earthquake magnitude)
Attenuation model:
describes effect of an earthquake originating from a specific
seismotectonic source, at any given site, as function of
magnitude and source-to-site distance (Ground Motion
Prediction Equations –GMPEs)
7/11/2013
1 • Describes the potential for dangerous, earthquake-related natural
phenomena i.e. Maximum Considered Earthquake (MCE)
2
• “The earthquake hazard for the site is a peak ground acceleration of 0.57g resulting from an earthquake of magnitude 5.7 on the Narmada Son Fault at a distance of 11.42 km from the site. ”
3 • Sometimes called Deterministic Scenario in Magnitude,
Distance pair i.e. (5.7, 11.42)
DSHA
Selection of urban areas
7/11/2013
67 68 69 70 71 72 73 74 75
67 68 69 70 71 72 73 74 75
19
20
21
22
23
24
25
26
19
20
21
22
23
24
25
26
F18
F17
F14 F15
F13
F12 F25A F5
F2
F1
F4
F3F6
F7
F8
F10
F9
F33
F35
F34
F37
F38 F42
F41
F43
F45 F46
F31
F32
F24
F49F23
F48
F27
F26
F21
F28
F29
F30
Legend :
n th Fault, Fn
Fault Map of study region
7/11/2013
• Total 40 major faults
are considered.
• Length derived from
referred literature and
maps.
• Maximum earthquake
magnitude calculated
from relationships
recommended by few
researchers
considering one third
length as rupture
surface.
Shukla and Choudhury (2012) in NHESS, 12, 2019-2037.
GMPEs selected
7/11/2013
GMPE Applicability Remark
Abrahamson and
Silva (1997)
Worldwide shallow crustal
earthquake
Boore et al.
(1997)
Shallow crustal earthquake
of Western north America
(Rock site definition is in accordance
with NEHRP seismic code)
Campbell (1997) Worldwide shallow crustal
earthquake
(for Mw > 5 and sites with distance to
seismogenic rupture ≤ 60 km in active
tectonic region)
Sadig (1997) Shallow crustal earthquake
of California
(Moment magnitude Mw = 4 to 8 and
distance up to 100 km).
Toro et al. (1997) Crustal earthquake of
Intraplate region in Eastern
and Central North America
(For spectral period less than 0.2 sec,
values limited to 1.5 g and periods less
than 1 sec are limited to 3 g.)
Frankel et al.
(1996)
Intraplate region of Central
and Eastern North America
Raghukanth and
Iyengar (2007)
Peninsular India (For sites with shear wave velocity Vs
≥ 3.6 km/sec.)
Various GMPEs
7/11/2013
0 20 40 60 80 100 120 140 160 180 200
1E-3
0.01
0.1
Sp
ectr
al
accele
rati
on
s (g
)
Distance from Hypocenter (km)
Abra.-Silva (1997)
Boore-Joyner-Fumal (1997)
Campbell (1997)
Frankel (1996)
Sadigh (1997) Rock
Toro (1997)
Raghu Kanth & Iyengar (2007)