Manual PDF

SECRETARY
DR. SHAILESH NAYAK GOVERNMENT OF INDIA

MINISTRY OF EARTH SCIENCES
MAHASAGAR BHAVAN, BLOCK-12, C.G.O. COMPLEX,
LODHI ROAD, NEW DELHI-110 003
FOREWORD
The seismic vulnerability in urban areas is rapidly increasing due to uncontrolled

growth, destabilization of environment and social framework. According to a conservative
estimate, more than 15 million lives have been lost and damage worth hundred billion
dollars has been inflicted in the recorded history due to earthquakes.
Urban Safety has gained importance in recent years with rapid increase in
construction activities and growth in urban population all over the world. It poses a
challenge to planners, administrators, engineers and architects alike due to its multi-
dimensional nature, which cover areas from structural design, maintenance and
rehabilitation to disaster mitigation.
Twenty seven cities in India have a population of one million or more. Urban population
in India is growing very fast. Most urban centers fall under Seismic Zones V, IV and III. In
this context, development of seismic microzonation maps of major urban centers has,
therefore, been recognized as a priority area of seismic mitigation programme in India. To
this effect, Ministry of Earth Sciences (MoES), Govt. of India constituted a National Steering
Committee in March 2008 to provide overall guidance to undertake microzonation studies
for the identified cities on priority, to be decided upon their locations and population
density. Though a limited work has already been done for a few selected cities in India, the
quantum of work is huge. Also, at present there are no detailed guidelines available for
adopting appropriate investigations. I hope that the present Seismic Microzonation
Manual, an MoES initiative that encapsulates bench-marked principles and protocol of
accomplishing seismic microzonation studies, will prove useful to the geoscientific and
earthquake engineering community of the country.
Shailesh Nayak
Tel. : 00-91-11-24360874, 24362548 E-mail : secretary@moes.gov.in Fax No. : 00-91-11-24362644/24360336
iii
PREAMBLE
The seismic microzonation has emerged as an important issue in the high risk
urban centers across the globe. A microzonation project extends from elementary
to exhaustive data analyses involving innumerous technical aspects underlying
the knowledge base with methodological diversity, but culminating ultimately into
recommendations defining constraints on the national/global regulations with local
considerations.
Even though Seismic Microzonation is a burning issue in Earthquake Disaster

Mitigation and Management, there are no detailed guidelines laid out for adopting
appropriate testing and investigation programs leading ultimately to erroneous
seismic hazard mapping providing inaccurate deliverables thereof. A typical seismic
microzonation endeavor follows the protocol: Source Characterization
Characterization of Wave Path Model Site Characterization. Through the
initiative of the Geoscience Division, Ministry of Earth Sciences, Government of
India, a Holistic Seismic Microzonation Framework is developed alongwith a
comprehensive work sheet as encapsulated in this concise Seismic
Microzonation Manual for the successful implementation of a typical Seismic
Microzonation project in any earthquake vulnerable province of the country on
uniform guidelines and protocol. However the detailed explanation of all the
Geological, Geophysical, Seismological and Geotechnical methodologies available
for systematic seismic microzonation studies is given in the Seismic
Microzonation Handbook for in-depth understanding and exhaustive analyses
to be performed in the research mode.
Prof. Sankar Kumar Nath

Department of Geology & Geophysics
Indian Institute of Technology
Kharagpur 721302
v
CONTENTS
Foreword Secretary, Ministry of Earth Sciences, Government of India
Preamble Prof. S. K. Nath, IIT Kharagpur
1. Introduction 1
2. Seismic Microzonation : Principles and Protocol 2
3. Global Trends 3
4. Microzonation Works carried out in India 4
5. Definition of Microzonation Problem 6
6. Objectives and Deliverables of a Typical Seismic

Microzonation Project 9
7. Project Implementation 11
8. Microzonation Framework 14
9. Microzonation Levels with Scale 18
10. Regional Assessments 18
11. Seismicity of the Study Area and Estimation of its

Ground Motion Parameters 20
12. Local Specific Assessments 24
13. In-Situ Field Testing for Site Characterization 26
14. Routine Geotechnical Laboratory Tests 39
15. Eurocode-8 and NEHRP 41
16. Assessment of Dynamic Properties based on

other Soil Properties 45
17. Spectral Shapes 45
vii
18. Site Response 47
19. Methods of Estimating Site Effects 48
20. Site Effects and Seismic Codes 57
21. 1D Ground Response Analysis using Equivalent Linear Approach 58
22. Predominant Frequency 62
23. Liquefaction Hazard Assessment 63
24. Liquefaction Susceptibility 64
25. Earthquake Induced Landslides 82
26. Simulation of Strong Ground Motion 86
27. Ground Motion Prediction Equations 97
28. Seismic Hazard Analysis 115
29. Composite Hazard Evaluation 119
30. Earthquake Hazard Parameters 120
31. GIS Integration Logic 132
32. Seismic Vulnerability and Risk Assessment 137
33. Concluding Remarks 139
Work Sheet for Seismic Microzonation 141
Appendix I : A Unified Earthquake Catalogue for South Asia

covering the period 1900-2008 145
Appendix II : A Seismogenic Source Framework for the Indian

Subcontinent 155
Appendix III : Probabilistic Seismic Hazard Assessment of India 177
References 195
viii
1. INTRODUCTION
Decisions to mitigate seismic risk require a logical and consistent approach towards evaluating
the effects of future earthquakes on people and structures. The knowledge that enables such
a decision making process comprises four steps as depicted in Figure 1: (i) the PSHA, which
provides a probabilistic description of earthquake characteristics such as ground motion
amplitudes and fault displacement, (ii) the estimation of earthquake damage to artificial and
perhaps natural structures, (iii) the translation of seismic hazards into seismic risks by using
the selected damage or loss functions, and (iv) the formal or informal analysis of earthquake
mitigation, wherein the options, uncertainties, costs, decision criteria, and risk aversion of the
policy makers are incorporated into the decision logic area. The ultimate goal of both the
seismic hazard and seismic risk analyses are to develop the elements that can be used to
make rational decisions on seismic safety. Such a decision process should incorporate
uncertainties in the earthquake occurrence and ground-motion characteristics, uncertainties
in the effects of earthquakes on people and structures, costs of seismic safety and potential
losses, and aversion to risk. Thus, study of seismic hazard and preparation of microzonation
maps will provide an effective solution for urban planning. Seismic hazard and microzonation of
the cities enable to characterize potential seismic vulnerability/risk that needs to be taken into
account while designing new structures or retrofitting the existing ones.
Figure 1: Steps in the mitigation of earthquake risk (after McGuire, 2004).

2 Ministry of Earth Sciences, Government of India
The seismic zonation map of India presents a large scale view and hence inhibits
interpretation of the site behavior during an earthquake due to local variations in soil type and
geology. Earthquake disasters are inevitable but it is possible to minimize the aftermath of
an earthquake if the zones that are more susceptible to undergo maximum ground motion
are identified. Seismic microzonation seems to be an answer to the need for mitigation
against the seismic hazards as it gives a realistic answer in terms of ground motion at a
higher resolution.
Rapid urbanization is a factor that calls for construction of mega-structures. The main
reason for human loss and property damage is when due importance is not given for adequate
preparation for possible seismic hazard. Seismic microzonation is needed in urban areas or
the upcoming urban areas falling under the high hazard zone. Some of the seismic
microzonation works carried out in important Indian cities are Delhi (Parvez et al., 2002;
Mohanty et al., 2007), Sikkim (Nath, 2005; Pal et al., 2008), and Jabalpur (Mishra, 2004),
while the seismic ground motion in large urban areas in other parts of the world are carried in
Bursa (Topal et al., 2003), Bucharest (Moldoveanu et al., 2004), Algiers (Harbi et al., 2004),
Alexandria (El-Sayed et al., 2004), Beijing (Ding et al., 2004), Napoli (Nunziata, 2004), Santiago
de Cuba (Alvarez et al., 2004), Sofia (Slavov et al., 2004), and Zagreb (Herak et al., 2004).
The microzonation map can serve many purposes for the Urban Development Authorities.
It can offer valuable information to the engineers for the seismic design of buildings and
structures, assessment of seismic risk to the existing structures and constructions,
management of landuse and also for the future construction of defense installations, heavy
industry, and important structures like dams, nuclear power stations and other public utility
services.
2. SEISMIC MICROZONATION : PRINCIPLES AND PROTOCOL
The well defined objectives involving mitigation and management of seismic hazards include
(i) evaluation of earthquake and related hazards, (ii) standardization of a global implementation
scheme to facilitate uniform action plans towards adapting urbanization regulations and
codes for design and construction practices, and (iii) seismic vulnerability assessment and
risk prognosis to enable preventive measures against the hazard. The seismic hazard defines
the potentially damaging ground shaking in terms of Peak Ground Acceleration (PGA), Peak
Ground Velocity (PGV), and/or Peak Ground Displacement (PGD) evaluated by both
Seismic Microzonation Manual 3
deterministic and probabilistic approaches. While the former delivers absolute values, the
latter estimates the same in terms of probability of non-exceedance corresponding to a
certain determined level at the site of interest. A quasi-deterministic or quasi-probabilistic
approach employs a hybrid seismological, geological, geomorphological, and geo-technically
guided framework wherein all the potential hazard attributing features are considered with
relative rankings in a logic tree, fuzzy set, or a hierarchical concept (Nath, 2004 and 2005).
Regional hazard zonations do not incorporate local and secondary effects induced by
the earthquakes, leading to its infeasibility in landuse development and planning, hazard
mitigation and management, and structural engineering applications which are site-specific.
It is necessary to overcome these limitations, especially in the highly populated urban centers
with unplanned urbanization practices in vogue. Seismic microzonation is, therefore, envisaged
to subdivide a region into subregions in which different safeguards must be applied to reduce,
and/or prevent damages, loss of life and societal disruptions; in the event of a large devastating
earthquake striking the region.
Typically, a seismic microzonation evaluation consists of two distinct parts:
A technical report, which can and must be flexible according to the risk;
A part concerning regulations, which makes the technical report applicable to
third parties.
In order to succeed, both the parts need a framework and specific criteria and tools.
Professionals are responsible for the first part and public authorities for the second one.
3. GLOBAL TRENDS
In the recent times, several seismic microzonation projects have been reported across the
globe. These mainly include assessment of predominant frequency, effective shear wave
velocity (Vs30) and deterministic seismic scenarios while site specific probabilistic hazard analysis
is slowly emerging for engineering decisions. These aspects can summarily be reviewed as,
1. Single attribute characterization based on ambient noise/microtremor derived

predominant frequency.
2. Geotechnical modeling of response spectra and site response based on borehole
data.
3. Shear wave velocity measurement to achieve Vs30 zonation.
4. Estimation of site response and predominant frequency from earthquake recordings.

5. Strong motion simulations for seismic scenario and deterministic hazard assessments.
6. Probabilistic hazard analysis based on local specific relationships.
Fundamentally, an approach depends on the data availability, the study region/ terrain
under consideration, and the choice of techniques: a few global scenario approaches are
enlisted in Table 1.
Table 1: Global Scenario Approaches
Approach Example/s
Estimation of predominant frequency from Bam City (Motamed et al., 2007),

ambient noise survey Napoli (Nunziata, 2004),
Greater Bangkok (Tuladhar et al., 2004)
Site response estimation from earthquake Alaska (Nath et al., 2002a)

data
1-D geotechnical modeling of predominant Bucharest City (Ehret et al., 2004),

frequencies, and peak amplifications Yenisehir-Bursa (Topal et al., 2003)
Probabilistic seismic microzonation maps Tashkent City (Sokolov and Chernov, 2001)
for various return periods
Predominant frequency and site City of Thessaloniki (Lachet et al., 1996)

amplifications from earthquake records &
ambient noise survey
Site response & response spectra from Santiago de Cuba (Alvarez et al., 2004),
synthetic seismograms Beijing (Ding et al., 2004),
Tehran (Hamzehloo et al., 2007)
Vs30 based site classifications by means of Las Vegas Valley (Scott et al., 2006),
shear wave velocity measurements Linares (Montalvo-Arrieta et al., 2005)
4. MICROZONATION WORKS CARRIED OUT IN INDIA

Seismic microzonation for a few important Indian cities vulnerable to earthquake hazards
has been carried out by different workers. These cities are Delhi (Rao and Mohanty, 2001;
Parvez et al., 2002; Singh et al., 2002; Mukhopadhyay and Bormann, 2004; Mohanty et al.,
2007), Sikkim (Nath, 2005; Pal et al., 2008), Jabalpur (Mishra, 2004), Guwahati (Nath et al.,
2008a), Bangalore (Sitharam et al., 2006; Sitharam and Anbazhagan, 2007; Anbazhagan et
al., 2010), Dehradun (Mahajan et al., 2007), and Mumbai (Raghukanth and Iyengar, 2006).
A summary of the scenarios are enlisted in Table 2.
Table 2 : Indian Scenario
Study Region Approaches
Guwahati Estimation of site response and predominant frequency from strong

motion data and geotechnical borehole modeling (Nath et al.,
2008a).
Seismic scenario for deterministic hazard (Nath et al., 2008b).
Landslide hazard and soil liquefaction zonation, site classification,

and thematic integration on GIS (Nath et al., 2008c).
Sikkim Estimation of site response and predominant frequency from strong

motion array (Nath et al., 2005).
Deterministic seismic scenario and thematic integration based on

analytical hierarchical process on GIS (Nath, 2004 and 2005; Pal
et al., 2008).
Delhi Geotechnical analysis, stochastic simulation of strong ground

motion and shear wave velocity measurements with MASW (Rao
and Satyam, 2005).
Synthetic seismogram (Parvez et al., 2004).
Predominant frequency estimation from ambient noise survey

(Mukhopadhyay et al., 2002).
Site response estimation from earthquake recordings (Nath et al.,

2003).
Probabilistic and deterministic assessment from geotechnical

borehole analysis (Iyengar and Ghosh, 2004).
Thematic integration on GIS (Mohanty et al., 2007).

Study Region Approaches
Bangalore Geotechnical data analysis & deterministic seismic scenario

(Sitharam and Anbazhagan, 2008).
Soil liquefaction potential mapping (Anbazhagan and Sitharam,

2006).
Probabilistic assessment (Anbazhagan et al., 2009).
GIS integration based on AHP (Anbazhagan et al., 2010).
Dehradun Seismic response from geotechnical modeling on the shear wave

velocity derived from MASW and geotechnical data (Mahajan et
al., 2007).
Chennai Site response estimation from geotechnical modeling of the MASW

derived shear wave velocity data and borehole information
(Boominathan et al., 2008).
Chandigarh Vs30 mapping and soil liquefaction assessment (Kandpal et al.,

2009).
Jabalpur Deterministic seismic hazard analysis based on scenario

earthquake; site response studies and evaluation of liquefaction
potential of the area.
Liquefaction susceptibility and frequency dependent amplification

of the strata, site classification based on Vs30 data (Mishra, 2004;
Menon et al., 2008).
5. DEFINITION OF MICROZONATION PROBLEM
Seismic microzonation can be defined as the subdivision of a region that has relatively
similar exposure to various earthquake related activities or the identification of individual
areas having different potential for earthquake effects. The important places of concern for
which seismic microzonation needs to be carried out is the urban or upcoming urban area
that falls under the high seismic hazard zone. The damage pattern due to an earthquake
depends largely on the local site condition and the social infrastructures of the region with
the most important condition being the intensity of ground shaking at the time of the earthquakes.
Contrasting seismic response is observed even within a short distance over small changes in
the geology of the site. Moreover, designing and constructing all structures everywhere to
withstand conceivable future earthquake is economically not viable.
Regional hazard zonations do not incorporate local and secondary effects induced by
the earthquakes leading to its infeasibility in landuse development and planning, hazard
mitigation and management, and structural engineering applications at site-specific terms. It
is necessary to overcome these limitations, especially in the highly populated urban centers
with unplanned urbanization practices in vogue. Seismic microzonation is, therefore, envisaged
to subdivide a region into subregions in which different safeguards must be applied to reduce,
and/or prevent damages, loss of life and societal disruptions; in case a large devastating
earthquake strikes the region.
Development of seismic microzonation of major urban centers has been recognized as

a priority area of seismic hazard mitigation programme in India. Twenty seven cities in India
have a population of one million or more. These 27 cities contain 25.6% of the total urban
population of the country. This is compounded by the fact that while geographically, 57.1% of
the country area is under seismic zones III, IV and V of BIS (2002), 66% of the population
and 63% of the housing are located within these zones. The three Indian metro cities Mumbai,
Kolkata, and New Delhi are listed amongst the 20 largest urban agglomerations of the world.
Their existence in Seismic Zones III and IV places them in moderate damage risk (MSK VII)
in Zone III to high damage risk (MSK VIII) in Zone IV.
Cities in India, which fall in seismic zones III, IV and V and having a population exceeding
half a million, are recommended to have seismic microzonation maps in 1:10,000 to 1:5,000
scales. The list of these cities is given in Table 3 below :
Table 3
Sl. State Name of the District Seismic Zone

No. City Zone Factor
1 Uttaranchal Dehradun Dehradun IV 0.24
2 Delhi Delhi New Delhi IV 0.24


3 Gujarat Jamnagar Jamnagar IV 0.24
4 Gujarat Rajkot Rajkot III 0.16
5 Gujarat Bhavanagar Bhavnagar III 0.16
6 Gujarat Surat Surat III 0.16
7 Maharashtra Greater Mumbai Mumbai III 0.16
8 Maharashtra Bhiwandi Thane III 0.16
9 Maharashtra Nashik Nashik III 0.16
10 Maharashtra Pune Pune III 0.16
11 Orissa Bhubaneshwar Khurda III 0.16
12 Orissa Cuttack Cuttack III 0.16
13 Tamil Nadu Chennai Chennai III 0.16
14 Bihar Patna Patna IV 0.24
15 West Bengal Asansol Bardhaman III 0.16
16 Assam Guwahati Kamrup V 0.36
17 Gujarat Vadodara Vadodara III 0.16
18 Gujarat Ahmedabad Ahmedabad III 0.16
19 Tamil Nadu Coimbatore Coimbatore III 0.16
20 Uttar Pradesh Agra Agra III 0.16
21 Uttar Pradesh Varanasi Varanasi III 0.16
22 Uttar Pradesh Bareilly Bareilly III 0.16
23 Uttar Pradesh Meerut Meerut IV 0.24
24 Uttar Pradesh Lucknow Lucknow III 0.16
25 Uttar Pradesh Kanpur Kanpur Nagar III 0.16


26 West Bengal Kolkata Kolkata III 0.16
27 Jammu & Kashmir Srinagar Srinagar V 0.36
28 Jammu & Kashmir Jammu Jammu IV 0.24
29 Madhya Pradesh Indore Indore III 0.16
30 Madhya Pradesh Jabalpur Jabalpur III 0.16
31 Punjab Amritsar Amritsar IV 0.24
32 Punjab Jalandhar Jalandhar IV 0.24
33 Andhra Pradesh Vijayawada Krishna III 0.16
34 Jharkhand Dhanbad Dhanbad III 0.16
35 Karnataka Mangalore South Canara III 0.16
36 Kerala Kochi Ernakulam III 0.16
37 Kerala Kozhikode Kozhikode III 0.16
38 Kerala Trivandrum Trivandrum III 0.16
6. OBJECTIVES AND DELIVERABLES OF A TYPICAL SEISMIC MICROZONATION

PROJECT
A typical seismic microzonation research project aims at achieving and assimilating the
following aspects in the evaluation of seismic hazard and building up levels of seismic
microzonation of an urban center.
i) Establish the geological and geomorphologic units of the region and its surrounding
areas, including the lithological characteristics and Seismic & Engineering Bedrock
configurations at different locations.
ii) Identify and characterize major/minor faults, lineaments, and seismotectonic units
that are seismically active.
iii) Characterize seismic activities based on historical seismicity and recorded ground
motion data. The attributes pertinent include location of potential sources, magnitude,
intensity, focal mechanism and epicentral distances, etc. Selection of suitable

attenuation laws.
iv) Evaluate spatial variation of shear wave velocity (i.e., average VS30 profiles) through
geophysical surveys, geotechnical borehole logging, and HVSR modeling to develop a
database of shallow subsurface stratigraphy information.
v) Evaluate the spatial distribution of predominant frequency of the soil through Nakamuras
technique.
vi) Establish seismic (ground) response through theoretical and numerical modeling of
wave propagation to identify amplification effects associated with near-surface ground
motions from alluvial deposits during an earthquake, during which critical facilities and
infrastructure are expected to remain operational.
vii) Determine the ground motion parameters such as peak ground acceleration/velocity
distribution at different locations. Determine the ground response spectrum, duration
and the time history of earthquake inputs.
viii) Assess seismic stability and estimation of permanent ground deformation within typical
structures.
ix) Site classification on the basis of the shear wave velocity model, and the geotechnical
assessments.
x) Assess the liquefaction potential from the detailed borehole geotechnical data and
shear wave velocity profiles using suitable model studies. Identification of threshold
value, maximum epicentral distance and depth based on field and laboratory studies.
xi) Local specific deterministic and probabilistic seismic hazard analyses.
xii) Generate thematic hazard maps in terms of site classification, site response,
predominant frequency, spectral accelerations/ response spectra, and peak ground
accelerations on GIS platform.
xiii) Prepare inventory for the demographic, building typology, and landuse patterns.
xiv) Identify building hazards from resonance phenomenon on the basis of predominant
frequency of the basin to demarcate zones of varying risk.
xv) Risk assessment of the urban center through scenario generations to identify damage
levels and achieve a probabilistic account of the same.
The deliverables envisaged from a seismic microzonation project are:
Regional Seismotectonism
Geological & Geomorphological mapping
Engineering Bedrock configuration
Predominant Frequency distribution mapping
Site Classification mapping
Site Amplifications mapping
Site-specific Ground Motion Prediction Equations
Deterministic and Probabilistic Hazard maps
Socio-economic impact evaluations: vulnerability and risk assessment
Provisions and regulations for landuse planning, structural developments, and
recommendations for hazard mitigation actions.
The output maps of microzonation will include:
Site characterization/ classification maps

Spatial variation of Peak Horizontal Acceleration (PHA) and Spectral Acceleration
(SA) values at bed rock level
Spatial variation of PGA and SA values at surface level
Liquefaction Hazard maps
Landslide maps
Comprehensive seismic hazard map prepared by combining all the above given maps
(after giving proper weightage and ranks for each factors and features in each theme).
7. PROJECT IMPLEMENTATION
A microzonation project extends from elementary to exhaustive data analyses involving

numerous technical aspects underlying the knowledge base with methodological diversity,
but culminating ultimately into recommendations defining constraints on the national/global
regulations with local ones. A framework that addresses the pertinent technical issues
highlighted through the state-of-the-art practices and methodologies are :
7.1 Source Characterization
Seismic sources are generally characterized by well defined physical parameters like corner
frequency fc, seismic moment M0, and stress drop derived directly from waveform data
(Nath et al., 2008b). Fault rupture attributes also include rupture dimensions, slip distributions,
maximum slip, strike and dip. The source accounts for the shape of the amplitude spectra
as well as the duration and directivity of ground motions.
The regional level study would involve compilation of fault based seismic sources within
the purview of the city (both near- and far-sources) with associated geometry and, if possible,
slip rates. Areal seismic sources can also be defined in case of associated unknown
complexity or distributed fault system. Development of homogenous moment magnitude
based earthquake catalogues spanning several hundred years (historical and instrumental
periods) is an essential work component to quantify seismicity distribution, the magnitude
frequency relation and the maximum magnitude.
7.2 Characterization of Wave Path Model
The attenuation of seismic waves along the propagation path connecting the earthquake
source and the recording site (observatory) is attributed to the degradation in the elastic
properties such as shear and compressional moduli, and the scattering of seismic waves
caused by heterogeneities in the earths interior. The strong motion attenuation feature as
defined by Wu and Aki (1985) is an exponential function of the type Qs (= Qo f n). The shear
wave Quality factor Qs is related to frequency dependent attenuation of spectral amplitudes,
and scattering within the crustal structure. The geometric spreading function accounts for
variations in the seismic radiations, which evidently is inversely related to a factor of the
distance from the source in the direction of the wave-front propagation. The evaluation of
path attributes is envisaged from available waveform data and crustal models.
7.3 Site Characterization
Site response estimation from geotechnical investigations involve combination of wave

propagation theory with the material properties and the expected ground motion computed at
the site of interest. Several algorithms are available for seismic response analysis for horizontally
layered soil deposits in which recurrent and circular soil behavior can be simulated using linear
equivalent model of a nonlinear phenomenon (Kramer, 1996). Techniques used widely to quantify
site response from waveform data include the standard spectral ratio, generalized and horizontal-
to-vertical spectral ratio (Field and Jacob, 1993).
The ground characterization of a terrain will include the following:
1. Geo-hazard attributes
o Geomorphological (including relief) features
o Landuse patterns
o Near-surface lithological structure
o Tectonic and structural organization.
2. Geotechnical parameters
o Geo-hydrological properties
o Standard Penetration Test (SPT) N values
o Factor of Safety against soil liquefaction
o Density
o Shear Wave Velocity.
Vs10, Vs30, Vs50 and VsAvg distributions corresponding to average shear wave
velocity from the ground surface to depths of 10 m, 30 m, 50 m and the basement.
3. Site Response
o Response Curves or transfer functions
o Response spectra
o Interpreted thematic distribution of:
- Predominant Frequency (F0), and
- Peak Amplification (A0) at the predominant frequency.
At present there are some fundamental disagreements amongst researchers on the

details of the procedures for assessing earthquake site response of several thousand meter
deep alluvial deposits (Pyke, 1979).
7.4 Ground Motion Simulation and Hazard Prediction
The quantitative assessment of seismic hazard necessitates measurement of a peak ground

motion parameter, such as the peak ground acceleration from earthquake records. Paucity
of strong ground motion data records under conditions similar to design earthquakes in
terms of tectonic regime, earthquake size, local geology, and near fault conditions
necessitates analytical or numerical approaches for a realistic prognosis of the possible
seismic effects. The strong ground motion modeling must accommodate: (i) the seismic
wave radiation from a fault rupture, (ii) propagation through the crust, and (iii) modifications
by site conditions. Several techniques are available that differ in their theoretical
considerations, data and computational requirements. Realistic results vis--vis complexity
and computational as well as data requirements are the deciding factors for the applicability
of a particular technique. Some of the existing techniques include: (i) stochastic approach
(Boore and Atkinson, 1987; Beresnev and Atkinson, 1997; Motazedian and Atkinson, 2005),
(ii) Greens Function method (Bouchon and Aki, 1977), (iii) Empirical Greens Function method
(Hartzell, 1978; Irikura, 1983), (iv) Finite Difference method (Panza, 1985; Oprsal and
Zahradnik, 2002), (v) Finite Element method (Frankel, 1989), and (vi) Spectral Finite Element
method (Komatitsch and Tromp, 1999). The strong ground motion study would involve
development of a database comprising of realistic synthetic strong ground motion data, site
specific ground motion prediction equations, and the interpreted thematic distributions of:
- Peak Ground Acceleration (PGA) and

- Period-specific Spectral Acceleration (SA) according to targeted building typology.
8. MICROZONATION FRAMEWORK
Seismic microzonation process is initiated with rudimentary assessments based on existing
regional level hazard estimation, seismotectonic and macro-seismic studies. Several local
specific hazard factors are, thereafter, evaluated and mapped on a Geographical Information
System (GIS) platform with a uniform and consistent georeferencing scheme. From the
structural viewpoint, GIS is very similar to conventional Data Base Management System
(DBMS), except for the fact that the database of GIS is more sophisticated and has the
capability to associate and manipulate enormous volume of spatially referenced interrelated
data (Star et al., 1997; Foresman, 1997; Longley and Batty, 1997; Hanna and Culpepper,
1998). The components of a GIS are pictorially represented in Figure 2.
Figure 2 : Basic components of a Geographical Information System (GIS).

GIS stores spatial and aspatial data into two different databases. The geocoded spatial
data defines an object that has an orientation and relationship with other objects in two (2D)
or three-dimensional (3D) space. It is also known as topological data and is stored in a
topological database. The data that describe the objects are known as attribute data stored
in a relational database. GIS links the two databases by maintaining one-to-one relationship
between records of object location in the topological database and records of the object
attribute in the relational database by using end-user defined common identification index or
code (Marble and Pequet, 1983; Korte, 1997; Hohl, 1998).
GIS uses three types of data to represent a map or any geo-referenced data, namely,
point type, line type, and area or polygon type. It can work with both the vector and the raster
geographic models. The vector model is generally used for describing the discrete features,
while the raster model does it for the continuous features (Burrough, 1986; Davis, 1996;
Burrough and McDonnell, 1998).
A GIS approach comprises three distinct phases (Sander, 1998): (1) data acquisition,
(2) data processing, and (3) data analysis. There are several ways of digitizing the map data
for its incorporation in a GIS. The data can be directly digitized from the map using a digitizing
table or the outline of the required classes may be traced on a transparent overlay by image
processing software.
The GIS can be used for hazard management at different levels of development planning.
At the national level, it can provide a general familiarity with the study area, giving planners a
reference to the overall hazard situation. At the regional level, it can be used in hazard
assessments for resource analysis and project identification. At the local level, it can be
used to formulate investment projects and specific mitigation strategies. It is hard to conceive
a micro-seismic programme without its very intimate coupling with GIS. It is the most efficient
way to integrate a set of factors or state of nature maps into hazard maps, and thereafter,
marry the hazard maps with the respective infrastructure map, to deliver risk assessment. It
is proposed to be developed as a powerful decision support system for the planning of human
settlement in earthquake prone areas, as well as for earthquake disaster management.
A holistic Seismic Microzonation Framework is depicted in Figure 3 along with the

seismological and geological attributions depicted in Figures 4 and 5 respectively. The scheme
outlines compilation of information related to seismicity, identification of potential seismic source
zones, development of seismicity models, and maximum earthquake prognosis at the regional
level supported by earthquake catalogues and other relevant data such as fault database. The
local level assessments involve mapping of surficial geological and geomorphological features
supported by 2D/3D subsurface models, development of geotechnical database, and the
evaluation of different surficial soil attributes (e.g., density, rigidity, compressibility, damping,
water content, etc.), and the basement topography. The prevalent seismic characteristics, in
terms of predominant frequency, site response, path and source attributes, are generally
established through an analytical and numerical treatment of the waveform, microtremor and
geotechnical data, and thereupon, a deterministic assessment is performed by means of
strong ground motion simulations. Additional evaluations include those related to relevant
earthquake induced effects such as soil liquefaction and landslides. Eventually, a composite
assessment is taken up of the geological, geotechnical, and seismological attributes to deliver
the seismic microzonation map in terms of a hazard index map.
Figure 3: Overall perspective of a seismic microzonation project: (a) data considerations and
flow, and (b) framework outlining regional to local hazard assessment.
Figure 4 : Seismological aspects in the seismic microzonation model.
Figure 5 : Geological aspects in the seismic microzonation model.

9. MICROZONATION LEVELS WITH SCALE

A microzonation study is designed to map the different components of local seismic hazard
at the scale of a study area, generally between 1:5,000 and 1:25,000 (Bard et al., 1995).
These include :
active tectonic structures

modification of the seismic signal due to local geomorphological conditions
induced phenomena such as liquefaction, settlements, landslides etc.
The methods used to achieve these vary in complexity according to the degree of precision
sought. Throughout the document, three levels of precision are adopted. These are A, B & C,
which signify a range from the most rudimentary to the most refined level of precision.
Level A
The most rudimentary level of study is generally based on a compilation and interpretation of
available data. It is the least expensive study and the scale of microzonation ranges from
1:100,000 to 1:50,000.
Level B
This intermediate level of study provides much more reliable results than those for Level A.
Specific surveys are generally carried out during this study level, including drilling, trenching,
geological sampling etc. The cost of this study remains reasonable, and the scale of
microzonation generally ranges from 1:25,000 to 1:10,000.
Level C
This study is only carried out in areas where a very detailed level of mapping is required.
Specific surveys and detailed calculations are involved. The cost of this study is high, but
may be necessary in areas of high earthquake hazard risk. This scale of microzonation
ranges from 1:10,000 to 1:5,000.
10. REGIONAL ASSESSMENTS
The regional level analysis encompasses the seismicity, seismic sources, and earthquake
potential based on available historical and instrumental data covering hundreds of years,
micro- and macro-seismicity, regional tectonics and neo-tectonics (faults/lineaments),

seismotectonics, geology, geo-hydrology, crustal structure, landslides incidents, observed
soil liquefactions, etc. (Nath et al., 2008b). Long-term earthquake catalogues are associated
with two important issues namely, data completeness and magnitude scale inhomogeneity,
the former necessitating a temporal segregation of the data according to its completeness
(Kijko, 2004) while in the latter, empirical relations connecting the different magnitude scales
are used to homogenize the magnitude scale to moment magnitude, MW, owing to its
applicability to all magnitude ranges, faulting types, and hypocentral depths of the earthquakes.
A large scale seismicity analysis is envisaged to examine spatial patterns (represented by
b-value and fractal correlation dimension of the epicenters or hypocenters), which along with
the tectonic background delivers a broad seismic source zonation (e.g., Thingbaijam et al.,
2008). Another alternative is seismicity smoothening of Woo (1996) that caters to spatial
distribution of event activity rates.
The maximum, characteristics, or maximum credible earthquakes in a region can be

estimated from the maximum fault-rupture projected on faults/lineaments known or otherwise
established from paleoseismic investigations (Wells and Coppersmith, 1994; Hanks and
Bakun, 2002; Rajendran et al., 2004). Likewise, the estimation may also be derived from the
slip deficiency based on the historical events and geodetic studies (Anderson et al., 1996).
However, association of unknown fault complexities presents limitations in such deterministic
assessments. A general technique employs homogenous earthquake catalogue to derive
appropriate seismicity models (Kijko, 2004) to establish the annual recurrences. The faults
likely to generate major earthquakes can be inferred from observed geological deformation
episodes. Large scale micro-seismicity recordings can enable detecting active faults on the
geological and geomorphologic signatures.
Glossary of Regional Studies:
Faults/lineaments database (historical and instrumental records)

Geological features and crustal structure
Earthquake Cataloging and seismicity analysis (e.g., Thingbaijam et al., 2008;
Thingbaijam and Nath, 2008; Thingbaijam et al., 2009; Thingbaijam and Nath, 2011)
Data completeness
Scale homogeneity
Development of up-to-date seismotectonic map
Seismic source zonation
Based on the tectonic background and distribution of seismicity parameters (e.g.,
Thingbaijam et al., 2008)
Seismicity smoothening (e.g., Woo, 1996; Thingbaijam and Nath, 2011)
The faults likely to generate major earthquakes can be inferred from observed
geological deformation episodes. Large scale micro-seismicity recordings
can enable detecting active faults on the geological and geomorphologic
signatures
Earthquake potential: Assessment of maximum earthquakes
Maximum fault-rupture projected on fault/lineament known or otherwise established
from paleoseismic investigations (Wells and Coppersmith, 1994; Hanks and Bakun,
2002; Rajendran et al., 2004)
Slip deficiency based on the historical events and geodetic studies (Anderson et
al., 1996)
Appropriate seismicity models to establish the annual recurrences, and establish
maximum earthquakes (Kijko, 2004; Thingbaijam and Nath, 2011)
Earthquake recurrence study and prediction of seismicity parameters viz. a-value, b-
value (Thingbaijam and Nath, 2011).
11. SEISMICITY OF THE STUDY AREA AND ESTIMATION OF ITS GROUND

MOTION PARAMETERS
The first step in seismic microzonation is to assess the characteristics of ground motion at
the bedrock level considering the seismic sources and the path effects. Generally, the seismic
study area extending up to 300 km from the boundary of the study area should be identified.
However, if there are any seismic sources which can create a very large earthquake (mega
earthquake), this distance should be increased. The details of the seismic events and the
seismic sources need to be identified from the region of interest. The initial estimates of
ground motion parameters (PGA, PGV or PGD) and/or bed rock motions should be obtained
at the bed rock level using deterministic and probabilistic methods.
11.1 Seismicity Analysis
Typical seismicity analyses (Thingbaijam and Nath, 2011) comprise the following steps:
Development of a homogenous and consistent earthquake catalogue (Thingbaijam

and Nath, 2011)
Identification of spatial seismicity patterns for geodynamic characterization
Maximum Earthquake Prognosis (Thingbaijam and Nath, 2011)
Temporal Seismicity Pattern Analysis for Earthquake Precursors.
The existing and accessible earthquake catalogues generally suffer from temporal
data completeness, and different magnitude types used. The latter is solved by homogenization
to moment magnitude (MW ) scale using conversion equations that connect different magnitude
types. The analyses are carried out using different sub-catalogues for each period of data
completeness. The seismicity patterns allow identification of stress regimes, and earthquake
precursors in the geodynamic framework.
A quantitative approach to seismicity analysis can be carried out with the assessment
of seismicity parameters : a- and b-value, and correlation fractal dimension, DC. The first
two parameters are obtained from the frequency magnitude distribution (FMD) given by the
Gutenberg-Richter (GR) relation (Gutenberg and Richter, 1944)
Log10 N = a bm (1)
Where N is the cumulative frequency of occurrence of magnitude, m in a given earthquake

database. The intercept and slope, a- and b-value respectively signify the background
seismicity level, and the magnitude size distribution.
The maximum earthquake, mmax, is defined as the one that is assessed as physically
capable of occurring within a defined seismic regime in an underlying tectonic setup (Thenhaus
and Campbell, 2003). The implementation of seismic hazard assessment towards hazard
mitigation and disaster prevention, in general, involves expensive infrastructure creation with
planned growth. Therefore, a realistic and pragmatic approach is crucial. The estimation of
mmax can be either probabilistic or deterministic. The deterministic approach involves employing
empirical relationships between mmax and various tectonic and fault parameters or between
mmax and the strain rate or alternatively the rate of seismic-moment release (Wells and
Coppersmith, 1994; Anderson et al., 1996; Anderson and Luco, 1983; Papastamatiou, 1980).
However, in most cases the deterministic approach associates large uncertainties with the
estimated mmax. In the probabilistic approach, mmax is estimated from the seismic event
catalogues and appropriate statistical estimation procedures. Some probabilistic approaches
are linear extrapolation of the GR relation (Gutenberg and Richter, 1944), application of
extreme value theory (Nath et al., 2005), application of gamma distribution for the FMD
(Main, 1996) and seismic moment distribution (Kagan, 2002). The extreme value approach
has strong drawbacks as most of the historical data are not used and large uncertainties are
involved in the estimation (Knopoff and Kagan, 1977). Furthermore, incorporation of seismogenic
perspectives of the earthquakes as criticality phenomena has been a recent development
(Main, 1995; Thingbaijam and Nath, 2011).
11.2 The b-value and Correlation Fractal Dimension Dc
In several studies, the b-value has been found to vary both spatially and temporally, and has
often been employed as one parameter approach for seismicity analysis. A low b-value
implies that majority of earthquakes are of higher magnitude, and a high b-value implies that
the majority of earthquakes are of lower magnitude. The variation of b-value has been seen
to be inversely related to stress distribution (Mogi, 1962; Wesnousky et al., 1983; Schorlemmer
et al., 2005). Furthermore, large material heterogeneities have been reported with higher b-
values (Scholz, 1968). Aftershocks have been reported to have high b-values while foreshocks
have been associated with low b-values (Suyehiro et al., 1964; Nuannin et al., 2005). In
spatial analysis, low b-value can be inferred as growing stress regime unleashing larger
magnitude earthquakes while high b-value indicates low stress buildup with continued stress
release through numerous smaller magnitude earthquakes.
The estimation of b-value is generally performed by the maximum-likelihood method

(Aki, 1965; Bender, 1983; Utsu, 1999) given as,
(2)
Where mmean is the average magnitude and m is the magnitude bin size. Schorlemmer et al.
(2003) suggests a bootstrap method (Chernick, 1999) to estimate the associated standard
deviation, b, of b-value. The approach involves computation of b-value repeatedly for a number
of times, each time employing different replacement events drawn from the associated
catalogue wherein any event can be selected more than once. The error is, then, estimated
as the standard deviation associated with the computed values.
Another commonly used seismicity parameter is the fractal correlation dimension of

earthquakes, DC. The parameter is implied as a power law exponent relating distance and
the number of pair of points of either epicenters or focal depths within the distance (Kagan,
2007). The temporal variations in the fractal dimension can enable assessment of seismic
processes (Kagan and Knopoff, 1980). A lower DC value indicates higher clustering while a
higher Dc value implies uniform or random spatial distribution. DC has been implicated as a
quantifier of crustal deformation in time and space (Turcotte, 1986). However, the variation of
fractal dimension in different seismic zones has also been attributed to the Geological diversity
(Aviles et al., 1987). An estimation of DC can be achieved through correlation integral method
(Grassberger and Procaccia, 1983). The fractal correlation dimension is given as,
(3)
where
(4)
C(N, r) is referred to as the correlation integral. H(x) is the Heaviside step function, whose
value is 0 if x < 0, otherwise 1. N is the total number of points in the query. The coordinates of
the location of points are given by yi and yj. Essentially, C(N, r) accounts for the number of
points that are at a distance r with respect to the total number of pairs of all the points.
A plot of log10(C(N, r)) against log10(r) is used to estimate DC as the slope of the curve in
its linear bound for a specific range of r. In case of an infinite two-dimensional distribution of
epicenters, the plot is a straight line. But, practically for larger values of r, a state of saturation
is attained thereby decreasing the gradient. For smaller values of r, an increase in the gradient
is induced indicating a state of depopulation (Narenberg and Essex, 1990). The specific
range of r is, therefore, projected from the bounds of saturation and depopulation limits.
12. LOCAL SPECIFIC ASSESSMENTS
12.1 Geology and Geomorphology
The geology and geomorphology serves as a significant attribute towards seismic

ground motion depiction at a site of interest (Aki, 1988; Panizza, 1991; Hartzell,
1992; Nath et al., 2002a; Nath et al., 2008b). In the geological and geomorphological
studies, the near-surface signatures pertaining to the recent sedimentary deposits -
alluvium, flood plains, cliffs, slope aspects, etc. can be complemented by borehole
litholog, exploratory drill holes, surface elevation model, land-cover, and basement
topography derived from vertical electrical resistivity soundings and other geophysical
investigations.
The geology and geomorphology serves as a significant attribute towards seismic

ground motion depiction at a site of interest (Aki, 1988; Panizza, 1991; Hartzell,
1992; Nath et al., 2002a; Nath et al., 2008b).
The assessments include:
The near-surface signatures pertaining to the recent sedimentary deposits
- alluvium, flood plains, cliffs, slope aspects, etc.
Complementary studies using borehole litholog, exploratory drill holes,
surface elevation model, and land-cover data
Development of basement topography from vertical electrical resistivity
soundings and other geophysical investigations
Higher resolution mapping from the analysis of satellite imageries.
12.2 Shear Wave Velocity
The shear wave velocity profile of a soil column is used for site response modeling as
well as site classification adhering to National Earthquake Hazard Reduction Program
(NEHRP, Building Seismic Safety Council 2001) and Uniform Building Code (UBC)
(ICBO, 1994) terminology. Several techniques are available to obtain subsurface shear
wave velocity profiles that include (i) using empirical equations between SPT-N values
obtained from geotechnical borelogs, and the average shear wave velocity (e.g., Fumal
and Tinsley, 1985; Imai and Tonouchi, 1982), (ii) Multichannel Analysis of Surface
Waves (MASW) (Park et al., 1999), (iii) Spectral Analysis of Surface Waves (SASW)
(Stokoe et al., 1994), and (iv) Cone Penetration Test (CPT). These techniques are
often employed in combination to authenticate and maintain consistency within
specified uncertainty in the interpretations of different observations. The spatial
mapping is generally done either with the average values for the sediment depth or 30
m from the surface of the soil column. The latter is widely used for site classification.
12.3 Site Characterization
Site class specifications are employed to characterize generic subsurface conditions towards
seismic response of the soil. A site characterization framework is shown in Figure 6. The
average shear wave velocity of the upper soil column is widely used for the purpose; the
ranges of values for each site class correspond to a specific class of soil. NEHRP (Building
Seismic Safety Council, 2001) specifies site class specifications (A-F) on the shear wave
velocity averaged for the upper 30 metres of the soil column VS30 with the exceptions of site
class E and F. The former is identified with the velocity less than 180 m/s or with more than
3 m of soft clay and Plastic index greater than 20, water content more than 40%, and
corresponding average undrained shear strength less than 25 kPa. Site F requires site-
specific evaluations to identify any of the four categories: (1) soils vulnerable to potential
failure or collapse under seismic loading such as liquefiable soils, quick and highly sensitive
clays, and collapsible weakly cemented soils, (2) peats and/or highly organic clays (soil
thickness greater than 3 m) of peat and/or highly organic clay, (3) very high plasticity clays
(soil thickness greater than 8 m with plasticity index greater than 75), and (4) very thick soft/
medium stiff clays (soil thickness greater than 36 m). Universal Building Code (UBC) provisions
also employ average shear wave velocities to describe the soil coefficients (ICBO, 1994).
Geological attributes are often connected to shear wave velocity in view of limited number of
observation sites (e.g., Wills and Silva, 1998; Wills and Clahan, 2006). On the basis of the
overlapping ranges of Vs30 accorded to different geological units, Wills et al. (2000) introduced
intermediate classes namely BC, CD, and DE corresponding to an average Vs30 of 760 m/s,
360 m/s, and 180 m/s, respectively.
Figure 6 : Site Characterization Framework.
13. IN-SITU FIELD TESTING FOR SITE CHARACTERIZATION
Various in-situ field tests available for site characterization are discussed in the subsequent
sub-sections. The most common field tests conducted for site characterization are Standard
Penetration Test (SPT), Cone Penetration Test (CPT), Spectral Analysis of Surface Wave
(SASW), and Multichannel Analysis of Surface Wave (MASW).
13.1 Standard Penetration Test (SPT)
The standard penetration test is done using a split-spoon sampler in a borehole/auger hole.
This sampler consists of a driving shoe, a split-barrel of circular cross-section (longitudinally
split into two parts) and a coupling. The standard procedure for carrying out the standard
penetration test is given below (as given by BIS: 2131, 1981).
A borehole is made to the required depth and the bottom of the hole is cleaned.
The split-spoon sampler, attached to the drill-rods of required length is lowered
into the borehole and is relaxed at the bottom.
The sampler is then driven to a distance of 450 mm in three intervals of 150 mm
each. This is done by dropping a hammer of 63.5 kg from a height of 762 mm (BIS:
2131, 1981). The number of blows required to penetrate the soil is noted down for
the last 300 mm, and this is recorded as the N value. The number of blows required
to penetrate the sampler through the first 150 mm is called the seating drive and is
disregarded. This is because the soil for the first 150 mm is disturbed and is ineffective
for the SPT-N value.
The sampler is then pulled out and is detached from the drill rods. The soil sample,
within the split barrel, is collected taking all precautions so as not to disturb the
moisture content and is then transported to the laboratory, for tests. Sometimes,
a thin liner is placed inside the split barrel. This makes it feasible for collecting the
soil sample, within the liner, by sealing off both the ends of the liner with molten
wax and then taking it away for laboratory test of the contained soil.
The standard penetration test is performed at every 0.75 m intervals in a borehole. If the
depth of the borehole is large the interval can be made 1.50 m. In case, the soil under
consideration consists of rocks or boulders, the SPT-N value can be recorded for the first 300
mm. The test is stopped if:
50 blows are required for any 150 mm penetration

100 blows are required for any 300 mm penetration
10 consecutive blows produce no advance.
However, it should be noted that the SPT-N value obtained from the above set of
procedures has to be corrected before it can be used for any of the empirical relations.
Necessary Corrections to be applied for the SPT Values
The SPT-N value collected from the field is without applying any corrections. The N values
that are obtained in the field are corrected for various errors, such as: overburden pressure,
hammer energy, borehole diameter, and presence of line, rod length and fines content.
The SPT-N value is corrected as follows :
(N1)60 = (N).CN.CR.CS.CB.CE (5)
Where, CR - correction for rod length; CS - correction for sampler configuration; CB - correction
for borehole diameter; CE - correction for hammer energy efficiency; It should also be noted
that in the current studies, the corrected SPT-N value (N1,60) is further corrected for its fine
contents as follows,
(N1)60,CS = (N1)60 x CFINES (6)
The fines correction is 1 for fine contents of FC < 5% and reaches a maximum value of
FC > 35% (where FC is the percentage fine content). The regressed relation for CFINES is
however given by the regression relationship as follows :
(7)
Where, FC, the percentage fines content is expressed as an integer (e.g., 10% fines is
expressed as 10); N1,60 is in units of blows/ft. The correction factors to be used are given in
Tables 4 7.
Table 4: Hammer Correction factor (CE)
Factor Correction
Donut Hammer 0.5-1.0
Safety Hammer 0.7-1.2
Automatic-trip Donut Hammers 0.8-1.3
Table 5: Correction for B.H. Diameter (CB)

Factor Size Correction
Borehole Diameter 65-115 mm 1.00
150 mm 1.05
200 mm 1.15
Table 6: Correction for Rod Length (CR)

<3 m 0.75
Rod Length 3-4 m 0.80
4-6 m 0.85
6-10 m 0.95
10-30 m 1.00
Table 7: Correction for Sampler based on method (CS)

Standard samplers 1.00
Sampling method
Sampler without liners 1.1-1.3
The formula used to find the correction for energy ratio is :
(8)
Where ER efficiency ratio is the fraction or percentage of the theoretical SPT impact energy
that is actually transferred to the sampler.
13.2 Cone Penetration Test (CPT)

Cone Penetration Test (CPT) is an in-situ test done to determine the soil properties and to
get the soil stratigraphy. This test was initially developed by the Dutch Laboratory for Soil
Mechanics in 1955 and hence it is sometimes known as the Dutch cone test. On a broad
scale the CPT test can be divided into two Static Cone Penetration Test (BIS: 4968, Part -
3, 1976) and Dynamic Cone Penetration Test.
(i) Static Cone Penetration Test
The cone with an apex angle of 60 and an end area of 10 cm2 will be pushed through the
ground at a controlled rate (2 cm/sec). In the static test the cone is pushed into the ground
and not driven. During the penetration of cone penetrometer through the ground surface, the
forces on the cone tip (qc) and sleeve friction (fS) are measured. The measurements are
carried out using electronic transfer and data logging, with a measurement frequency that
can secure the detailed data about soil contents and its characteristics. The Friction Ratio
(FR = fs/qc), varies with the soil types and is an important parameter.
(ii) Dynamic Cone Penetration Test
Dynamic test is conducted by driving the cone by hammer blows. The dynamic cone resistance
is estimated by measuring the number of blows required for driving the cone through a
specified distance. Usually this test is performed with a 50 mm cone without bentonite slurry
or using a 65 mm cone with bentonite slurry. The hammer weighs 65 kg and the height of fall
is 75 cm. The test is done in a cased borehole to eliminate the skin friction.
There are many correlations available to evaluate soil properties based on the CPT
value (either static or dynamic).
13.3 Seismic Cone Penetration Test (SCPT)
The seismic cone penetration test uses a standard cone penetrometer with two geophones.
One set of geophones is located behind the friction jacket and the other set is located one
meter above the first set.
The test method consists of measuring the travel time of seismic waves propagating
between a wave source and ground surface. These waves will comprise of shear waves and
compressional or primary waves. The velocity of seismic waves in the ground will give shear
modulus and poissons ratio and the soil profile.
13.4 Field Vane Shear Test
Vane shear test is done on fully saturated clays for the evaluation of undrained shear strength.
This is suitable for soft clays whose shear strength (less than 100 kPa) is changed
considerably by sampling. The equipment consists of stainless steel vane (with four blades)
connected to the end of a high tensile rod. The rod is enclosed by a sleeve packed by
grease. The typical dimension of the vanes is usually 50 mm by 100 mm or 75 mm by 150

mm and the diameter of the rod should not exceed 12.5 mm (BIS : 4434, 1978). After pushing
the vane and the rod into the clay, torque is gradually applied to the top end of the rod till the
clay fails in shear. The shear strength can be calculated based on the torque applied and the
test is repeated at the desired depths.
13.5 Dilatometer Test
Dilatometer consists of a stainless steel blade with a thin flat circular expandable steel
membrane on one side. Dilatometer is advanced into a borehole from the ground surface and
tests are conducted at an interval of 10 to 20 cm. At each interval, the dilatometer is stopped
and the membrane is inflated under gas pressure. The readings of inflation of the membrane
and the corresponding gas pressure are recorded. There are correlations available to relate
the test results with low-strain soil stiffness and liquefaction resistance of the soil.
13.6 Pressuremeter Test
Pressuremeter test is conducted using a pressuremeter. It is a cylindrical device that

uses a flexible membrane for the application of uniform pressure to the walls of a borehole.
It measures the stress-deformation behavior of the soil by measuring volume of fluid
injected into the flexible membrane and the corresponding pressure applied. This is the
only in-situ test that can measure stress-strain as well as strength behavior of in-situ
soil.
13.7 Suspension Logging Test
This test is most commonly used in petroleum explorations. It principally consists of a

probe of length 5 m to 6 m long. It is lowered into an uncased borehole filled with water and
drilling fluid. A reversible-polarity solenoid located at the end of the probe produces an
impulsive pressure wave in the drilling fluid, which travels to the bottom of the borehole and
produces P and S waves in the surrounding soil. The P- and S-waves thus produced transmit
through the soil and reach the geophones attached to the probe 1 m away from the solenoid.
This technique allows measurements of P- and S-waves in a single, uncased borehole, but
at very high frequencies (1000 to 3000 Hz for P-waves and 500 to 2000 Hz for S-waves),
which are much higher than the frequency range of interest of geotechnical earthquake
engineering.
13.8 Spectral Analysis of Surface Wave (SASW)
This method is used to measure the shear wave profiles of soil. This method depends on the
dispersive characteristics of Rayleigh waves traveling through a layered soil medium. A dynamic
source is used to create surface waves of different frequencies and these are monitored by
two receivers at known distances. Using the SASW methods, a large area can be covered
and the soil profiles can be obtained. However, a more advanced MASW method is preferred
to SASW as it is very tedious to conduct SASW tests which depend only on two geophones
for exploration.
13.9 Multichannel Analysis of Surface Wave (MASW)
Shear wave velocity (Vs) is an essential parameter for evaluating the dynamic properties of
soil in the shallow subsurface. A number of geophysical methods have been proposed for
near-surface characterization and the measurement of shear wave velocity by using a great
variety of testing configurations, processing techniques, and inversion algorithms. The most
widely-used technique is the Multichannel Analysis of Surface Waves (MASW), which was
first introduced in Geophysics by Park et al. (1999).
An MASW system consists of a number of geophones (usually more than twelve)

arranged at equal intervals. The seismic waves are created by an impulsive source (sledge
hammer) and captured by the geophones/receivers. The captured Rayleigh wave is further
analyzed using suitable software to generate Vs data. This is being done in three steps: i)
preparation of a Multichannel record (some times called a shot gather), ii) dispersion curve
analysis, and iii) inversion. The term Multichannel record indicates a seismic data set
acquired by using a recording instrument with more than one channel. MASW has been
effectively used with higher signal-to-noise ratio (S/N) of surface waves.
The generation of a dispersion curve is a critical step in the MASW method. A dispersion
curve (Figure 7) is generally displayed as a function of phase velocity versus frequency.
Phase velocity can be calculated from the linear slope of each component on the swept-
frequency record. The lowest analyzable frequency in this dispersion curve is around 4 Hz
and the highest frequency of 75 Hz is considered.
Empirical relations have been proposed to correlate the penetration test results between
CPT and SPT (Robertson et al., 1983) as well as with the shear-wave velocities (Ohta and
Goto, 1978; Mayne and Rix, 1995; yisan, 1996). A list of some of the relationships proposed
to calculate shear wave velocity in terms of SPT-N value is given in Table 8.
Figure 7: Dispersion AnalysisMultichannel Approach (2D Wave-field Transformation).
Table 8: Proposed Relationships to Estimate Shear Wave Velocity from SPT Values
Soil type VS (m/sec) Author Region
Clay Vs=5.3 N+134 Fumal and Tinsley (1985) USA
Vs=114.4 N0.31 Lee (1990) USA
Vs=165.7 N0.19 Pitilakis et al. (1992) Greece
Vs=105.7 N0.33 Raptakis et al. (1995) Greece
Vs=27.0 N0.73 Jafari et al. (2002) South of Tehran
Vs=80.2 N0.292 Imai (1977) Japan
Vs=97.9 N0.269 Hasancebi and Ulusay (2007) Turkey

Clay Vs=132(N60)0.178 Pitilakis et al. (1999) Greece
Vs=80(N)0.33 Anbazhagan and Sitharam Bangalore, India

(2010)
Vs=89.31 N0.358 Uma Maheswari et al. (2010) Chennai, India
Vs=44 N0.48 Dikmen (2009) Western Taiwan
Vs=100.0 N0.33 Jafari et al. (2002) Japan
Sand Vs=57.4 N0.49 Lee (1990) USA
Vs=56.4 N0.50 Seed et al. (1983) USA
Vs=162.0 N0.17 Pitilakis et al. (1992) Greece
Vs=80.6 N0.331 Imai (1977) Japan
Vs=90.8 N0.319 Hasancebi and Ulusay(2007) Turkey
Vs=22.0 N0.76 Chein et al. (2000) Western Taiwan
Vs=79.0 N0.434 Hanumantharao and Ramana Delhi

(2008)
Vs=32 N0.5 Shibata (1970) Japan
Vs=87 N0.36 Ohta et al. (1972) Japan
Vs=145(N60)0.178 Pitilakis et al. (1999) Greece
Vs=57((N1)60cs)0.44 Anbazhagan and Sitharam Bangalore, India

(2010)
Vs=100.53 N0.265 Uma Maheswari et al. (2010) Chennai, India
Vs=5.1 N+152 Fumal and Tinsley (1985) USA
Vs=88.0 N0.34 Ohta and Goto (1978) Japan
Vs=73 N0.33 Dikmen (2009) Western Taiwan

Sand Vs=125 N0.3 Okamota et al. (1989)

(P.sand)
Vs = 80.0 N0.33 Jafari et al. (2002) Japan
All Vs=91.0 N0.34 Imai (1977) Japan
Vs=85.35 N0.348 Ohta and Goto (1978) Japan
Vs=121.0 N0.27 Jafari et al. (2002) Japan
Vs=61.0 N0.50 Seed and Idriss (1981) USA
Vs=107.6 N0.36 Athanasopoulos (1995) Greece
Vs=22.0 N0.85 Jafari et al. (1997) Iran
Vs=116.1 Jinan (1987)

(N + 0.3185)0.202
Vs=51.5 N0.516 Iyisan (1996) Turkey
Vs=97.0 N0.314 Imai and Tonouchi (1982) Japan
Vs=82.0 N0.39 Ohsaki and Iwasaki (1973) Japan
Vs=92.1 N0.337 Fujiwara (1972)
Vs=90.0 N0.309 Hasancebi and Ulusay (2007) Turkey
Vs = 82.6 N0.43 Hanumantharao and Ramana Delhi

(2008)
Vs = 76 N0.33 Imai and Yoahimura (1970) Japan
Vs = 19 N0.6 Kanai et al. (1966)
Vs = 92 N0.329 Imai and Yoahimura (1975) Japan
Vs = 90 N0.341 Imai et al. (1975) Japan
Vs = 243.8 e0.4 LIQUFAC
Vs = 152.4 e0.3 LIQUFAC
Vs = 95.64 N0.301 Uma Maheswari et al. (2010) Chennai, India

Vs = 32.8 N0.51 Sisman (1995) Turkey

Vs = 68.3 N0.292 Kiku et al. (2001) Turkey
Vs = 58 N 0.39
Dikmen (2009) Western Taiwan
Vs = 76.2 N0.24 Kalteziotis et al. (1992) Greece
Silt loam/ Vs = 4.3 N+218 Fumal and Tinsley (1985) USA
sandy clay
Silty Sand/ Vs = 86.0 N0.42 Hanumantharao and Ramana Delhi
Sandy slit (2008)
Silts Vs = 106 N0.32 Lee (1990) USA
Silts Vs = 22.0 N 0.77
Jafari et al. (2002) South of Tehran
Alluvial Soil Vs = 84 N0.31 Ohba and Toriumi (1970) Japan
Fine grained Vs = 19.0 N0.85 Jafari et al. (2002) South of Tehran
Soil
Cohesionless Vs = 59 N0.47 Ohsaki and Iwasaki (1973) Japan
Soils
Granular Vs=100.5 N0.29 Sykora and Stokoe (1983) USA
Soils
Sands and Vs=152+5.1 N0.27 Fumal and Tinsley (1985) USA
gravelly
sand soils
Granular Vs=175+3.75 N Kayabali (1996) Turkey
Soils
Gravel Vs = 94.0 N0.34 Ohta and Goto (1978) Japan
Gravel Vs=192.4 N0.13 Raptakis et al. (1995) Greece
Holocene Vs=63.0 N0.43 Sykora and Koester (1988) USA
gravels
Pleistocene Vs=132 N0.32 Sykora and Koester (1988) USA
gravels
Holocene Vs=63.0 N600.43 Rollons et al. (1998) USA
gravels
Pleistocene Vs=132.0 N600.32 Rollons et al. (1998) USA

gravels
Cohesive Vs=76.6 N0.45 Kalteziotis et al. (1992) Greek
Soils
Cohesionless Vs=49.1 N0.50 Kalteziotis et al. (1992) Greek
Soils
Where, N - SPT value; e - effective vertical stress of the soils; Vs - shear wave velocity; P- Pleistocene.
13.10 Other Low Strain Field Tests
The strain levels in these types of tests will be around 0.0001%, some of the important low
strain tests are discussed below.
13.10.1 Seismic Reflection Test
This test is used to evaluate the wave propagation velocity and the thickness of soil layers.
The test setup consists of a source producing a seismic impulse and a receiver to identify
the arrival of seismic waves and the travel time from source to receiver is measured. Based
on these measurements, the thickness of soil layer can be evaluated.
13.10.2 Seismic Refraction Test
This test uses the arrival time of the first seismic wave at the receiver. Using the results
obtained from this test the delineation of major stratigraphic units is possible.
13.10.3 Steady State Vibration Test
In this test, the wave propagation velocities are measured from steady state vibration
characteristics. However, these tests can be useful for determining the near-surface shear
wave velocity and they fail to provide the details of highly variable soil profiles.
13.10.4 Seismic Cross-hole Test
In a seismic cross-hole test, the wave velocities are measured using more than one borehole.
In the simplest case, two boreholes are used one with an impulse source and the other
with a receiver and both are kept at the same depth. The test is repeated at various depths
to get the soil profile.
13.10.5 Seismic Down-hole (up-hole) Test
This test is used to measure the travel time of seismic waves from source to receiver. It is
performed using a single borehole. In a seismic down-hole test, the receiver is kept at the
ground surface and the impulse source is kept at different depths. The up-hole test is done
with the receiver at the ground surface and the impulse source in the borehole. This test is
not effective for depths greater than 30 m to 60 m.
13.10.6 Resonant Column Test
This is a most common low strain test which is being used to evaluate the dynamic properties
of soil. In this test the soil samples in the form of either solid or hollow cylindrical samples
are subjected to torsion or axial loading using an electro magnetic loading system. The
fundamental frequency of the sample can be evaluated and this, in turn, will give the value of
the shear modulus of the soil specimen. Even though the resonant column test is a very
good alternative in evaluating the damping and strain dependent properties of soils, the
response will depend on the response of the apparatus also.
13.10.7 Bender Element Test
In this test, the shear wave velocity of a laboratory specimen can be measured using a
piezoelectric bender element. A transmitter and the receiver elements (piezoelectric
transducers) are placed at each end of the sample. There will be change in the dimension
of these piezoelectric elements when subjected to change in voltage. An electric pulse
applied to the transmitter causes it to deform rapidly and produces a stress wave that will
travel through the specimen towards the receiver. When the stress wave reaches the receiver,
it generates a voltage pulse which is measured. The wave speed is calculated from the
arrival time and the known distance between the transmitter and the receiver.
13.10.8 Cyclic Triaxial Tests
The test device consists of the standard triaxial testing equipment with a cyclic axial loading
unit. In some cases, the cell pressure is also applied cyclically and it is possible to simulate
isotropic or anisotropic initial stress conditions. The values of the shear modulus and damping
ratio can be obtained from the stress strain response of the samples. The average slope of the
hysteresis loop yields shear modulus and the area under the loop provides the damping ratio.
Typically, five to ten specimens are tested under different levels of cyclic shear strain amplitudes
chosen in the range of 10-4 to 10-1 for the determination of dynamic properties. Dynamic
deformation characteristics are influenced by the effective confining pressure during the test.
When an undisturbed sample of normally consolidated soil is obtained, the effective vertical
pressure at the depth of sampling is isotropically applied by cell pressure to avoid the influence
of over consolidation. In order to obtain in-situ shear modulus of the soil from the laboratory test
results, correction of these results is necessary so that a shear modulus corresponding to the
average effective principal stress at the sample depth is obtained.
13.10.9 Cyclic Direct Simple Shear Test
The cyclic direct simple shear test can simulate the earthquake loading more precisely than the
cyclic triaxial test and hence this is one of the tests which is commonly used for liquefaction
testing. When a cyclic shear stress is applied to the top or bottom of the specimen, the deformation
is similar to that of a soil element in which there is a vertical propagation of S-wave.
13.11 Model Tests
13.11.1 Centrifuge Modeling
In usual lab tests, it is extremely difficult to simulate very high body forces. This deficiency
can be overcome with the use of centrifuge techniques, in which models are subjected to
predetermined, high acceleration levels to simulate the field conditions satisfactorily. Some
of the important problems in which the centrifuge modeling can be applied are earth dams,
tunnels, offshore foundations, geo-environmental problems, problems of nuclear waste
disposal, seismic studies of earth structures and foundations etc.
13.11.2 Shake Table
Characteristics of earthquake ground motions and the behavior of structures during earthquakes
can be simulated more precisely by constructing scaled models on a shake table and subjecting
them to real time ground motions. The results obtained can be used to improve seismic
design of various structures. Shake table models are usually built in laminar boxes, to reduce
frictional effects and to facilitate smooth relative displacement between the adjacent layers.
Single to six degrees of freedom shake tables in various sizes are available.
14. ROUTINE GEOTECHNICAL LABORATORY TESTS

The routine geotechnical laboratory tests for soils and rock samples are as follows:
14.1 Index Properties of Soil and Rock Samples
For Soil samples: Grain Size Analysis of the representative samples can be obtained from
Sieve and Hydrometer analysis, (BIS: 2720, Part 4-1985) or by deploying a laser analyzer
(BIS: 2720, Part 4-1985). This is to evaluate the soil particle sizes and gradation. Coarser
particles are separated in the sieve analysis portion, and the finer particles are analyzed
with a hydrometer (75m size is chosen to make a distinction between the coarse and fine
particles). The sieve analysis is done using an automatic sieve shaker wherein the sample
passes progressively through to smaller mesh sizes to assess its gradation. The hydrometer
analysis uses the rate of sedimentation to determine particle gradation.
The Atterberg limits are a basic measure of the nature of a fine-grained soil. Depending
on the water content of the soil, it may appear in any of the four states: solid, semi-solid,
plastic and liquid. In each state, the consistency and behavior of a soil is different and so
are its engineering properties. Thus, the boundary between each state can be defined based
on a change in the soils behavior and they are represented by Atterbergs Limits (Liquid
Limit, LL; Plastic Limit, PL; Shrinkage Limit, SL). These Atterberg Limits can be determined
in the laboratory following BIS: 2720 (Part 5)-1985. The difference between the liquid Limit
and plastic limit is called the plasticity index (IP). The Shrinkage Limit (SL) is the water
content where further loss of moisture will not result in any more volume reduction.
Natural water content (w%) can be calculated as per BIS: 2720 (Part 2)-1973. Specific
Gravity, In-situ Density and Moisture Content can be obtained as per BIS: 2720 (Part 3)-
1980. Relative Density of cohesionless soils can be evaluated as described in BIS: 2720
(Part 14)-1983. Free swell index of soil as per BIS: 2720 (Part XL)1977 also termed as free
swell or differential free swell is the increase in the volume of soil without any external
constraint when subjected to submergence in water. Bulk density ( ) is defined as the
mass of soil particles of the material divided by the total volume they occupy.
Permeability characteristics of the soils can be determined using falling head or fixed
head permeameter as per BIS: 2720 (Part 17)-1986. Compressibility characteristics can be
obtained from oedometer tests as per Bureau of Indian Standards (BIS: 2720 (Part 15) -1986).
Strength characteristics can also be obtained using triaxial, direct shear and vane shear tests.
For Rock Samples: Following tests alongwith BIS codes are used for rock samples:
Unconfined Compressive Strength of rock samples [BIS: 9143-1979]

Dynamic Modulus of rock core specimen [BIS: 10782-1983]

Modulus of Elasticity, Poissons Ratio in uniaxial compression [BIS: 9221-1979]
Point Load Strength Index [BIS: 8764-1998].
14.2 Tests for Shear Strength Parameters and Consolidation Characteristics
Tests for shear and consolidation are preferably performed on undisturbed samples and in
some cases on remolded samples. The direct shear test (Direct shear Test: BIS: 2720
(Part 13)-1986) determines the consolidated drained strength properties of a sample. The
test is performed with different normal loads to evaluate the shear strength parameters (c
and ). The methods of testing for soils for the determination of Shear Strength parameters
of soil from consolidated undrained triaxial compression test with or without pore water
measurement are provided in BIS: 2720 (Part XII)1981. Triaxial Shear tests comprise UU,
CU (Consolidated Undrained test with and without Pore Water Pressure Measurement) or
CD (Consolidated Drained) tests.
15. EUROCODE-8 AND NEHRP
A site classification scheme based on Vs30 values was proposed by Borcherdt (1994) and a
similar scheme was adopted by the National Earthquake Hazard Reduction Program
(NEHRP) also. The NEHRP (BSSC, 2003) site classification scheme is given in Table 9.
Eurocode-8 (2003) has also classified the site based on Vs30 Standard Penetration Test
(SPT) and Cone Penetration Test (CPT) values. The classification given by Eurocode-8 is
given in Table 10. Eventhough both the schemes use similar methods to identify site classes,
the range of Vs30 values specified for each site class is different in both the methods.
Table 9: Site classification as per NEHRP scheme (BSSC, 2003)
NEHRP Site Description Vs30

Class
A Hard rock >1500 m/s

B Firm and hard rock 760 1500 m/s
C Dense soil, soft rock 360 760 m/s
D Stiff soil 180 360 m/s
NEHRP Site Description Vs30

Class
E Soft clays <180 m/s
F Special sandy soils, e.g., Liquefiable soils,
sensitive clays, organic soils, soft clays
> 36 m thick
Table 10: Site classification adopted by Eurocode 8 (2003)
Ground Description of Stratigraphic Profile Parameters

Type (m/s) SPT Cu (KPa)
A Rock or other rock-like geological >800
formation, including utmost 5m of weaker
material at the surface.
B Deposits of very dense sand, gravel, or 360 800 >50 >250

very stiff clay, at least several tens of
meters in thickness, characterized by a
gradual increase of mechanical
properties with depth.
C Deep deposits of dense or medium 180 360 15 50 70 250

dense sand, gravel or stiff clay with
thickness from several tens to many
hundreds of meters.
D Deposits of loose-to-medium <180 <15 <70

cohesionless soil (with or without some
soft cohesive layers), or of predominantly
soft-to-firm cohesive soil.
E A soil profile consisting of a surface

alluvium layer with Vs30 values of type C
or D and thickness varying between about
5 m and 20 m, underlain by stiffer material
with Vs30 >800 m/s.
Ground Description of Stratigraphic Profile Parameters

Type Vs30 (m/s) SPT Cu (KPa)
S1 Deposits consisting, or containing a <100 10 - 20

layer at least 10 m thick of soft clays/ (indicative)
silts with a high plasticity index (PI >
40) and high water content
S2 Deposits of liquefiable soils, of sensi-

tive clays, or any other soil profile not
included in types A E or S1
In many locations, the rock depth is shallow (less than 30 m) and hence the evaluation
of V value is not possible. In those cases, extrapolation of the available Vs values has to be
s
30
done to evaluate the Vs30 values. The method proposed by Boore (2004) can be used for this
purpose. He has suggested different models to extrapolate the shear wave velocities for
depths less than 30 m to get the Vs30 value. The first method is extrapolation based on
constant velocity. In this model it is assumed that the shear wave velocity remains constant
from the deepest velocity measurement to the 30 m level as
(9)
Where tt(d) is the travel time to depth d and Veff = Vs(d), Vs(d) is the timed average velocity
to a depth of d.
Even though this method is simple, it is found to underestimate the Vs30 values, since
in most of the soils, the shear wave velocity is found to increase with depth. Another relation
proposed by Boore (2004) was based on a power law relation as:
(10)
Where is the velocity at a depth of d m (10 < d < 30). The values of the regression
coefficients a and b can be obtained from Boore (2004). The extrapolation of Vs values can
also be done based on the velocity statistics (Boore, 2004),
(11)
Where is the probability of exceedance of (see Boore 2004 for

further details of this analysis).
A modified site classification system based on geotechnical data was proposed by

Rodriguez-Marek et al. (2001). In this system the stiffness of soil was also taken into account
for site classification. This system is presented in Table 11. The main advantage of this system
is that it correlates the Vs30 values with the geotechnical and surface geological features.
Table 11: Classification based on geotechnical features (after Rodriguez-Marek et al.,

2001)
Site Description Comments
A Hard rock Crystalline bedrock; Vs30 1500 m/s
B Competent bed rock Vs30 >600 m/s or < 6 m of soil. Most unweathered
California rock cases
C1 Weathered rock V s 30 ~300 m/s increasing to > 600 m/s,

weathering zone > 6 m and < 30 m
C2 Shallow stiff soil Soil depth > 6 m and < 30 m
C3 Intermediate depth stiff soil Soil depth > 30 m and < 60 m
D1 Deep stiff Holocene soil Soil depth > 60 m and < 200 m
D2 Deep stiff Pleistocene soil Soil depth > 60 m and < 200 m
D3 Very deep stiff soil Soil depth > 200 m
E1 Medium thickness soft clay Thickness of soft clay layer 3 12 m
E2 Deep soft clay Thickness of soft clay layer > 12 m
F Potentially liquefiable sand Holocene loose sand with high water table, Zw
6 m
16. ASSESSMENT OF DYNAMIC PROPERTIES BASED ON OTHER SOIL

PROPERTIES
There may be some cases when the evaluation of the shear modulus (Gmax) is not possible
based on any of the above mentioned methods. In those cases the evaluation can be done
based on other soil properties which are available. One of the initial relations to find the
value of Gmax based on void ratio was proposed by Richart et al. (1970).
(12)
Where e is the void ratio and is the effective mean principle stress.
The shear modulus varies with effective stress alone. The relation showing the variation
of Gmax was proposed by Chung et al. (1984). The effect of soil types on damping ratio was
studied by Kokusho (1987).
17. SPECTRAL SHAPES
Within the scope of Eurocode 8, the earthquake motions are represented by an elastic
ground acceleration response spectrum, dependent on sub-soil class defined and on the
magnitude value.
For application purposes only, two different response spectra, Type 1 and Type 2,
have been introduced as shown in Figures 8 and 9; and are to be adopted respectively in
high (Mw>5.5) and low seismicity regions (Mw>5.5). The epicentral distance is not considered
because it controls the amplitude of the spectra but not their shape, and, therefore, has no
effect on normalized spectra. For other applications and tectonic environments, the adoptation
of more spectral shapes may be recommended, especially for the very large magnitude
values (Mw>7.5), for which Tc and Td may be higher.
Figure 8: Type 1 elastic response spectra for the 5 subsoil classes and corresponding soil
parameter (S) and control periods (Tb, Tc, Td).
Figure 9: Type 2 elastic response spectra for the 5 subsoil classes and corresponding soil
parameter (S) and control periods (Tb, Tc, Td).
In situations where the 30 m depth is not enough to reach a bed-rock type soil formation,
as in deep sedimentary basins, it might be necessary to introduce more classes, which
are essentially the same as above, but with specific deep ground geological constitution.
In conclusion, one can say that there are several possibilities to deal with the soil
problem in impact studies, depending on the working scale, the knowledge of the
geotechnical situation and the software availability. But it also depends on the details of
knowledge of the other components of the entire process for estimating the seismic impact.
18. SITE RESPONSE
The site amplification of ground motion is primarily attributed to either the geomorphological
features that produce scattering, focusing, or defocusing of incident energy (topographic
effect) or thick alluvium-filled terrain that causes reverberations due to trapped energy (basin
Figure 10: Site Response Study Framework.

Figure 11: Site amplification factor versus frequency evaluated through HVSR method from
strong motion data (in bold curve) calibrated against those estimated through 1D
geotechnical analysis (in lighter shade curves) at borehole sites located closely to
the strong ground motion station (after Nath et al., 2008a).
effect). Techniques used widely to quantify site response in terms of site amplification
factors include (i) horizontal-to-vertical spectral ratio, (ii) generalized inversion approach,
(iii) standard spectral ratio method with reference to a rock site, (iv) coda wave technique,
and (v) geotechnical evaluation of the soil transfer function using available softwares like
SHAKE, SHAKE 2000, WESHAKE, ShakeEdit, Deepsoil etc. (Nath et al., 2002b; Kramer
1996; Kato et al., 1995; Lermo and Chavez-Garcia, 1993; Hartzell, 1992). The site
amplification factor estimated using the flow chart shown in Figure 10 from geotechnical
data analysis can complement the ones assessed from strong ground motion waveform. An
example of calibration of HVSR site response with those estimated through geotechnical
analysis is depicted in Figure 11 application of multiple techniques allows resolving ambiguity
associated with the estimation. Validations can be achieved through macro-seismic intensity
distribution for previous/historical earthquakes (e.g., Hough and Bilham, 2008).
19. METHODS OF ESTIMATING SITE EFFECTS
There are various methods that may be used for site effect evaluation. The choice of the
method is usually related to the significance of the engineering project for which it is applied.
Generally, the methods are classified into five main categories.
Experimental empirical techniques that utilize recordings of ground motion or

ambient noise to estimate the basic characteristics of the expected ground motion
usually in the frequency domain.
Empirical methods that evaluate parameters of earthquake motions such as

acceleration, velocity and response spectra based on site classification, average
S-wave velocity, topography, earthquake magnitude and existing amplification
relationships; usually these methods are incorporated in seismic code provisions.
Semi-empirical methods that compute time histories of earthquake motion by

combining recorded earthquake motion of smaller earthquakes as element motion
(i. e., Greens Functions); these methods may account for the detailed fault
rupture process and the effects of asperities.
Theoretical methods where site effects are computed through an analytical and
more often numerical 1D, 2D or 3D wave propagation model; different wave types
with different incident angles may be used; the main advantage of these methods
is the possibility to use complex constitutive relationships for describing soil
behavior under dynamic loading conditions and the ability to model accurately
site stratigraphy inclusive of basin topography.
Hybrid methods that compute time histories of earthquake motions by coupling a

longer period component determined by a theoretical seismic fault model with a
computational seismic wave propagation model having a shorter period component
is determined by a semi-empirical method.
The use of each method depends on many parameters and, in any case, requires an
increased level of expertise.
19.1 Theoretical Considerations and Site Response Estimation
The seismograms of the selected events are first corrected for the system response. Next
the S-wave packets recorded by the seismographs are windowed with a window width
containing the maximum amplitude. The window length is selected following the results of
Seekins et al. (1996). A Hanning taper is applied to the time windowed data and then
butterworth bandpass filtered before the amplitude spectra are computed.
Let the S-wave spectral amplitude and that of the background noise be O(rij, fk) and
B(rij, fk) respectively at the hypocentral distance rij. Then the signal amplitude spectrum at
the frequency fk can be expressed as,
(13)
The corrected spectra are smoothed in order to reduce the data variance using a five-
point smoothing window and a spline interpolator at 0.1 Hz interval.
Suppose a network has recorded I events at J stations. The amplitude spectrum of the
ith event recorded at the jth station for the kth frequency, A(rij, fk) can be written in the frequency
domain as a product of a source term SOi(fk), a propagation path term P(rij, fk), and a site
effect term SIj(fk), (Lermo and Chvez-Garca, 1993; Nath et al., 2002a and 2002b):
(14)
19.2 Standard Spectral Ratio (SSR) Technique
The most popular and widely used technique to characterize site amplification has been the
Standard Spectral Ratio (SSR) (Borcherdt, 1970), which is defined as the ratio of the Fourier
amplitude spectra of a soil-site record to that of a nearby rock-site record from the same
earthquake and component of motion (Figure 12). Suppose, there is a reference site (j=R)
that is assumed to have a negligible site response (lnSR=0) and if the interstation spacing
is too small compared to the epicentral distances, so that , then the site
response at each seismological station can be estimated from,
(15)
Equation (15) constitutes the geometric average spectral ratio. If the reference site
has a non-negligible site response, then the spectral ratios become relative site-response
estimates or site amplification factors. In the classical spectral ratio method, the source
and path effects are considered identical for every site and the reference one only when the
epicentral distance is appreciably larger compared to the interstation spacing. As a result,
the normalized SSR method is expected to perform better for the regional events.
Figure 12: General description of the Standard Spectral Ratio (SSR) Technique.
A usual option for the selection of the reference station is a site of outcropping rock,
while less frequently, a bedrock site having a down-hole accelerometer installed in a borehole
is used for this purpose. The basic conditions for the application of this particular technique
are: (a) the existence of simultaneous recordings at a soil site and at the reference site, (b)
the reference site has to be free of any kind of site effects (sediments and topography), and
(c) the distance between the soil site and the reference one ought to be small (i. e., smaller
than the epicentral distance), in order to consider that the effect of the propagating path of
the seismic energy is the same for the two sites.
However, the condition that an outcrop rock reference site should be free of any kind of
site effects often is not valid. For this reason, a careful examination of the reference site is
obligatory in order to correctly estimate amplification in sedimentary sites (Stiedl et al.,
1996).
19.3 Generalized Inversion Scheme (GIS) Technique
Andrews (1986) having considering the fact that that the Standard Spectra Ratio technique
is reliably applicable only to data from dense, local arrays, proposed a generalized technique
to look for all source, path and site effects in large data sets recorded in local or regional
networks, by applying the solution of a large inverse problem.
In Equation (14) the propagation path term can be expressed as,
(16)
Where G(rij) accounts for geometrical spreading, QS(fk) and are S-wave frequency-depen-
dent quality factor and velocity, respectively. Following Ordaz and Singh (1992), Castro et
al. (1996) and others,
for r <100 km (17)
Or,
for r >100 km (18)
to take into account possible arrival of surface waves in the windowed data.
Equation (14) can further be written as,
(19)
Where is the spreading and attenuation corrected amplitude spectrum. Taking

the natural logarithm on both the sides of Equation (19) and multiplying by (Hartzell,
1992),
(20)
Or, (21)
Where for the kth frequency fk, Si = ijk lnSOi(fk), Rj = ijk lnSIj(fk) and Dij = ijk ln A(rij,fk). is a
weighted factor and an estimate of the standard deviation of the data given by the ratio of
the signal spectrum to a noise sample spectrum, Nij(fk). The algorithm for estimating
tried in this study is,
(22)
Where is normalized and limited to the range of 1.0 to 0.2. This imposes a balance in the
quality of the data with the need to include as many observations as possible to average out
radiation-pattern and source-directivity effects. Nij(fk) is obtained from a noise sample im-
mediately preceding the shear-wave arrival. Equation (21) can be written in the matrix form
following the notation of Menke (1989) and solved by generalized inversion scheme using
singular value decomposition (Nath et al., 2002b). The method requires a reference site with
known response value in the frequency range of interest to minimize the trade-off between
the source and site parameter values as observed by Andrews (1986).
(23)
Where G, the data kernel matrix, is a sparse matrix, having two nonzero elements per row
or column for each frequency. Equation (23) is solved using singular value decomposition
technique and the source and site terms are computed.
(24)
The main advantage of this technique is the reliable estimation of the source and site
effect terms from the whole data set especially in cases where there are no simultaneous
records of all earthquakes at all sites of the network (Field and Jacob, 1995).
19.4. Horizontal to Vertical Spectral Ratio (HVSR) Technique
All techniques mentioned above are using a reference site but in practice, appropriate
reference sites are not always available. For this reason different methods that do not
depend on reference sites have been developed. One of them is the Horizontal-to-Vertical
Spectral Ratio. This extremely simple technique consists of using the spectral ratio of the
horizontal to the vertical component of ground motion and estimates the Fourier amplitudes
in different frequencies accordingly. The receiver function HVSRij(fk) can be computed at
each j site for the ith event at the central frequency fk from the root mean square average of
the amplitude spectra as,
(25a)
Where,
Hij(fk)|NS - Fourier spectra of the NS component,
Hij(fk)|EW - Fourier spectra of the EW component and
Vij(fk) - Fourier spectra of the vertical component.
Finally the event average receiver function (Field and Jacob, 1995) is
computed at each j site for the k frequency to consider the contribution of all the seismic
th
events recorded at that station. The basic assumption of the method is that the vertical
component of the ground motion in case where the soil stratigraphy is flat and horizontal is
supposed to be free of any kind of influence related to the soil conditions at the recording
site. Figure 13 shows the general layout of the method which was first applied to the S-wave
portion of the earthquake recordings obtained at three sites in Mexico City by Lermo and
Chavez-Garcia (1993). Generally, the Fourier spectral ratio exhibit similarities between SSR
and HVSR technique, with a better fit in frequencies rather than amplitudes of the resonant
peaks.
Detailed comparisons between SSR technique and other reference station techniques
(Field et al., 1992; Stiedl, 1993; Field and Jacob, 1995) led to few basic qualitative conclusions
such as; (a) the estimation of site effects, which with the use of SSR technique is relatively
stable even if records are quite noisy, (b) the process should be based on a significant
number of earthquake recordings (the use of a limited number of records should be avoided),
and (c) the amplification level determined with SSR technique is quite similar with that
determined from other techniques.
Figure 13: Description of the Horizontal to Vertical Spectral Ratio (HVSR) Technique .
Other comparison between results of SSR and HVSR techniques led to controversial
conclusions. As stated above, Lermo and Chavez-Garcia (1993) applied the HVSR to the S-
wave portion of the earthquake recordings and found similarities between standard spectral
ratio and HVSR with a good fit in both the frequencies and amplitude of the resonant peaks.
Some other researchers used the HVSR technique on data sets from weak and strong
motion records and concluded that the shape of the spectral ratio presents a very good
statistical stability with minor dependency on source and path effects and that it is quite
well correlated with surface geology, while their amplification level seems to depend on the
type of incident wave, a fact that does not affect the fundamental resonant frequency. Field
and Jacob (1995) after systematic comparisons with other techniques concluded that the
shape of the transfer function is satisfactorily reproduced by the HVSR technique, although
there is an underestimation of the amplification factor compared to SSR. On the same
issue Raptakis et al. (1998 and 2000), using a large and high quality data set from
EUROSEISTEST experimental site, proved that the significant differences between SSR
and HVSR amplitudes at the fundamental frequency are attributed to the considerable
amplification of the vertical component due to diffracted Rayleigh waves at the lateral
discontinuities of the basin.
In conclusion, both SSR and HVSR techniques are reliable in estimating the
fundamental frequency of the soil profile. However the amplification amplitude is comparable
only when the soil layering is horizontal and there are no lateral geometrical variations. In
those cases, due to the presence of in-ward propagating surface waves, it is expected that
part of them will affect the vertical component and hence the amplitude of the HVSR method.
19.5 Horizontal to Vertical Response Spectral Ratio (HVRSR) Technique
The Horizontal to Vertical Response Spectral Ratio (HVRSR) is obtained using geometric
mean of the response spectra of the horizontal components and the response spectra of
the vertical component as follows,
HVRSRij(fk) = (Aij(fk)|T Aij(fk)|R )0.5 / Aij(fk)|V (25b)
Where A stands for spectral acceleration at each j site for the ith event at the central frequency
fk , T, R and V representing the transverse, radial and vertical components respectively.
19.6 Empirical Methods
Empirical methods are practically used either for preliminary analyses or in the frame of
seismic code prescriptions with well specified amplification factors defined according to the
soil classification and the earthquake intensity. Simple relationships giving the amplification
factor for the peak acceleration or/and velocity with the average shear wave velocity of the
soil profile are proposed in the literature (Joyner and Fumal, 1984; Midorikawa, 1987; Borcherdt
et al., 1991). All these relationships should be used only for preliminary studies and with
extreme caution.
19.7 Semi-Empirical Methods
The semi-empirical methods compute time histories of earthquake motions caused by a

large scenario earthquake by combining recorded earthquake motions due to smaller seismic
events. The Greens Function technique is based on the idea that the total motion at a
particular site is equal to the sum of the motions produced by a series of independent
ruptures of many small parts on a causative fault. The method requires an approximate
definition or estimation of certain parameters such as the geometry of the source, the slip
functions describing the slip displacement vector with time for each elementary source, the
velocity structure of the crustal materials between the source and the site and the Greens
Functions that describe the motion at the site due to an instantaneous unit slip at each
elementary source. Normally the Greens Function at each site, account implicitly for the
particular site specific ground behavior in the linear elastic range.
The Empirical Greens Function (EGF) technique (Hartzell, 1978) bypasses these
complicated computations by using the weak motions of small earthquakes as empirical
Greens Functions to simulate strong motion. The method is essentially a deterministic one
as it computes time histories for a defined earthquake scenario and other parameters. However
it is possible to use statistical Greens Functions which are computed as the statistical
average of the recorded earthquake motions for different small seismic events.
19.8 Theoretical (Numerical and Analytical Methods)
When the geological structure of an area and the geotechnical characteristics of the site are
known, site effects can be estimated through theoretical analysis. The prerequisite of sufficient
geotechnical knowledge of the soil structure including surface and deep topography is,
therefore, obvious. Such an approach requires an in-depth understanding of the constitutive
models describing the soil behavior under dynamic solicitations and the methods used to
solve the wave propagation problem in 1, 2 or 3 dimensions.
20. SITE EFFECTS AND SEISMIC CODES
Seismic ground response characteristics, defined generally as site effects, are inevitably
reflected in seismic code provisions. The selection of appropriate elastic response spectra
according to soil categories and seismic intensity is the simplest way to account for site
effects both for engineering projects and for a general purpose microzonation study.
Modern seismic codes (IBC2000, UBC97, NEHRP, EC8) all introduced in the last few
years, after the recent strong earthquakes in America, Europe, Japan and worldwide that
produced numerous valuable data, have incorporated the most important experimental and
theoretical results with the adjustments and simplifications for purely practical reasons.
The main improvement is that amplification factors of spectral values are varying with
the seismic intensity; lower shaking intensity earthquakes introduce higher amplification
factors due to the more linear elastic soil behaviors, contrary to higher intensities where soils
exhibit a non-linear behavior resulting in a decrease of peak spectral values. Additionally, a
more accurate soil categorization is introduced based on a better description of soil profiles
using standard geotechnical parameters (i.e., plasticity index PI, undrained shear strength
Su) and average Vs values. In IBC2000 and other codes of the same family, a special
attention is given to near-field conditions introducing higher amplification factors for the same
earthquake magnitude. Also, for soil layers of small thickness presenting high impedance
contrast, the new version of codes attribute higher amplification factors which is compatible
with observations and theory.
In general, the parameters describing site effects in seismic codes are expressed
through (a) soil categorization and (b) spectral amplification factors and shapes. 1D site
effect computations using the equivalent linear model is the main and almost universal tool
for all improvement and modifications introduced so far (i.e., Dickenson and Seed, 1996;
Pitilakis, 2003). The increasing number of records during the last two decades gave the
minimum necessary validation background to these theoretical efforts and simulations. Full
non-liner and elasto-plastic models are not really used mainly because of the difficulty in
defining soil parameters for all soil categories, while it is known that probably for strong
earthquakes, the use of elasto-plastic models shall lead to an even more important decrease
of ground amplification, especially in low resistance soil layers (Pitilakis et al., 1999).
Modern codes are certainly a serious step forward for a better evaluation of design input
motions at least for ordinary buildings. Future improvements should be directed towards
addressing the following issues: Azimuthal effects (i.e., different spectral values for the two
horizontal components), basin edge and deep basin effects, evaluation of ground motion for
large and very large shear strains, vertical component, topographic effects, velocity and
displacement spectra, spatial variability of ground motion etc.
21. 1D GROUND RESPONSE ANALYSIS USING EQUIVALENT LINEAR APPROACH
The Geotechnical site response is computed by considering (a) a geological unit which is
defined by shear wave velocity, damping, total unit weight, and thickness and (b) an input
motion at the rock level which has to be computed synthetically or using the observed
earthquake. Kramer (1996) has given the general theory for the computation of geotechnical
site response.
Any input signal can be treated as the sum of sine waves of different amplitude, frequency
and phase. A simpler solution of the site response at each frequency is used to compute
the response of the soil at each of the input sine waves.
The site amplification of multilayer can be computed using the formulations as that the
transfer function for this type of soil deposit must account for the transmission and reflection
of waves at boundaries between adjacent layers. Due to constant change in shear modulus
of the soil; the nonlinear behavior cant be ignored in the analysis. The effect of nonlinear
behavior in seismically induced dynamic loading has been a debatable issue amongst
seismologists. The prime assumption in the geotechnical analysis is that site response is
dependent on strain amplitude (Finn, 1991). However seismologists have avoided the
nonlinearity due to lack of direct evidence from strong motion data (Aki and Richards,
1980).
As proposed by Kramer, an equivalent linear model of non-linear behavior of soil can

be taken into consideration. It performs the iterative adjustment on the soil properties which
is to be consistent with effective level of shear strain in the soil. The relationship between
shear modulus and shear strain can be characterized by a modulus reduction curve. The
first iteration in the equivalent linear analysis is performed using shear modulus and damping
ratio. The effective shear strain can be defined as,
(26)
Where R is a stress reduction factor often taken as,
R = , M is the magnitude of earthquake (27)
The process is repeated until there is not much variation from one iteration to the next
and this is called convergence of the solution. It must be noted that although it is a linear
equivalent of nonlinear phenomenon, it is still a linear analysis. However, this approach has
been shown to provide a reasonable estimate of soil response. Site response using the
above algorithm has been coded in many types of software (SHAKE 2000, SHAKE91,
PROSHAKE, DEEPSOIL etc.).
In an equivalent linear approach as defined in Figure 14, the non-linearity of the shear
modulus and damping is accounted for the use of equivalent linear soil properties using an
iterative procedure to obtain values for modulus and damping compatible with the effective
strains in each layer. In this approach, first, a known time history of bedrock motion is
represented as a Fourier series, usually using the Fast Fourier Transform (FFT). Second,
the Transfer Functions for the different layers are determined using the current properties of
the soil profile. The transfer functions give the amplification factor in terms of frequency for
a given profile. In the third step, the Fourier spectrum is multiplied by the soil profile transfer
function to obtain an amplification spectrum transferred to the specified layer. Then, the
acceleration time history is determined for that layer by the Inverse Fourier Transformation
in step four. With the peak acceleration from the acceleration time history obtained and with
the properties of the soil layer, the shear stress and strain time histories are determined in
step five. In step six, new values of soil damping and shear modulus are obtained from the
damping ratio and shear modulus degradation curves corresponding to the effective strain
from the strain time history. With these new soil properties, new transfer functions are
obtained and the process is repeated until the difference between the old and new properties
fit in a specified range.
Figure 14: Equivalent Linear Approach.

The basic approach of one dimensional site response study is the vertical propagation
of shear waves through soil layers lying on an elastic layer of the rock which extends to
infinite depth.
Soil behavior under irregular cyclic loading is modeled using modulus reduction
(G/G max) and damping ratio () vs. strain curves. The non-linearity of the shear modulus and
damping is accounted for by the use of equivalent linear soil properties using an iterative
procedure to obtain values for modulus and damping compatible with the effective strains in
each layer as discussed above. The degradation curves for sand and rock spectral response
analyses used for those proposed by Seed and Idriss (1970) and Schnabel (1973) respectively
are shown in Figures 15 and 16 respectively. These are included in the SHAKE database
and can be selected as input using option command.
Figure 15: Shear modulus reduction and damping ratio curves considered for sand (Seed
and Idriss, 1970).
Figure 16: Shear modulus reduction and damping ratio curves considered for rock (Schnabel,
1973).
22. PREDOMINANT FREQUENCY

The predominant frequency corresponds to the maximum amplitudes of the ground motion
in frequency domain. The proximity of predominant frequency of the soil layers and natural
frequency of the buildings indicate higher vulnerability of the built-environment owing to
resonance effects (Navarro and Oliveiram, 2006).
Figure 17: Local specific relations between averages shear wave velocity, and predominant
frequency with the basement depth.
The assessment of predominant frequency distribution in a particular terrain can be

performed through waveform data analysis that may be strong ground motion (accelerations),
broadband (velocity) or ambient noise (microtremors). The Horizontal-to-Vertical Spectral
Ratio (HVSR) analysis on ambient noise measurements purposed by Nakamura (1989)
offers advantage of easy data acquisition, besides being inexpensive and reliable so far as
the basin response effect is considered. Local specific relations between average shear
wave velocity and predominant frequency with the basement depth can be established to
enable durability in the spatial extrapolation/ interpolation of these parameters (e.g., Ibs-von
Seht and Wohlenberg, 1999; Parolai et al., 2002). Figure 17 depicts such relations observed
in the Guwahati City.
Liquefaction, landslides and subsidence are other topics connected to soil performance
with great importance essentially for stabilization of foundations and performance of lifelines.
The level of water content in the soil stratum is of major importance and, consequently, the
epoch of the year when the event takes place does have a great influence on the potential
for liquefaction, landslide etc.
23. LIQUEFACTION HAZARD ASSESSMENT

Liquefaction is one of the most important, interesting, complex, and controversial topic in
geotechnical earthquake engineering. Its devastating effects sprang to the attention of
geotechnical engineers in a three-month period in 1964 when the Good Friday earthquake
(Mw = 9.2) in Alaska was followed by the Niigata earthquake (Ms = 7.5) in Japan. Both
earthquakes produced spectacular examples of liquefaction-induced damage, including slope
failures, bridge and building foundation failures, and flotation of buried structures. In the 30
years since these earthquakes, liquefaction has been studied extensively by hundreds of
researchers around the world. Different terminologies, procedures, and methods of analysis
have been proposed, and a prevailing approach has been slow to emerge.
Evaluation of Liquefaction Hazards
Both flow liquefaction and cyclic mobility can produce damage at a particular site, and a
complete evaluation of liquefaction hazards requires that the potential for each be addressed.
When faced with such a problem, the geotechnical earthquake engineer can systematically
evaluate potential liquefaction hazards by addressing the following questions:
1. Is the soil susceptible to liquefaction?

2. If the soil is susceptible, will liquefaction be triggered?
3. If liquefaction is triggered, will damage occur?
If the answer to the first question is no, the liquefaction hazard evaluation can be
terminated with the conclusion that liquefaction hazards do not exist. If the answer is yes,
the next question must be addressed. In some cases it may be more efficient to reverse
the order of the second and third questions, particularly when damage appears unlikely. If
the answers to all three are yes, a problem exists; if the anticipated level of damage is
unacceptable, the site must be abandoned or improved or on-site structures strengthened.
These questions pertain to the three most critical aspects of liquefaction hazard evaluation:
susceptibility, initiation, and effects. All the three must be considered in a comprehensive
evaluation of liquefaction hazards.
24. LIQUEFACTION SUSCEPTIBILITY

Not all soils are susceptible to liquefaction; consequently, the first step in a liquefaction
hazard evaluation is usually the evaluation of liquefaction susceptibility. If the soil at a
particular site is not susceptible, liquefaction hazards do not exist and the liquefaction
hazard evaluation can be ended. If the soil is susceptible, however, the matters of liquefaction
initiation and effects must be addressed. There are several criteria by which liquefaction
susceptibility can be judged, and some are different for flow liquefaction and cyclic mobility.
These include historical, geologic, compositional, and state criteria.
24.1 Geological Criteria
Soil deposits that are susceptible to liquefaction are formed within a relatively narrow range
of geological environments (Youd, 1991). The depositional environment, hydrological
environment, and age of a soil deposit all contribute to its liquefaction susceptibility (Youd
and Hoose, 1977).
Geologic processes that sort soils into uniform grain size distributions and deposit
them in loose states produce soil deposits with high liquefaction susceptibility. Consequently,
fluvial deposits, and colluvial and aeolian deposits when saturated, are likely to be susceptible
to liquefaction. Liquefaction has also been observed in alluvial-fan, alluvial-plain, beach,
terrace, playa, and estuarine deposits. The susceptibility of older soil deposits to liquefaction
is generally lower than that of newer deposits. Soils of Holocene age are more susceptible
than soils of Pleistocene age, although susceptibility decreases with age within the Holocene.
Liquefaction of pre-Pleistocene deposits is rare.
Liquefaction occurs only in saturated soils, so the depth to groundwater (either free or
perched) influences liquefaction susceptibility. Liquefaction susceptibility decreases with
increasing groundwater depth; the effects of liquefaction are most commonly observed at
sites where groundwater is within a few meters of the ground surface. At sites where
groundwater levels fluctuate significantly, liquefaction hazards may also fluctuate.
24.2 Compositional Criteria
Since liquefaction requires the development of excess pore pressure, liquefaction susceptibility
is influenced by the compositional characteristics that influence volume change behavior.
Compositional characteristics associated with high volume change potential are associated
with high liquefaction susceptibility. These characteristics include particle size, shape, and
gradation.
For many years, liquefaction-related phenomena were thought to be limited to sands.

Finer-grained soils were considered incapable of generating high pore pressures commonly
associated with liquefaction, and coarser-grained soils were considered too permeable to
sustain any generated pore pressure long enough for liquefaction to develop. More recently,
the bounds on gradation criteria for liquefaction susceptibility have broadened.
Liquefaction of nonplastic silts has been observed (Ishihara, 1984 and 1985) in the
laboratory and the field, indicating that plasticity characteristics rather than grain size alone
influence the liquefaction susceptibility of fine-grained soils. Coarse silts with bulky particle
shape, which is nonplastic and cohesionless, are fully susceptible to liquefaction (Ishihara,
1993); finer silts with flaky or platelike particles generally exhibit sufficient cohesion to
inhibit liquefaction. Clays remain nonsusceptible to liquefaction, although sensitive clays
can exhibit strain-softening behavior similar to that of liquefied soil. Fine-grained soils that
satisfy each of the following four Chinese criteria (Wang, 1979) may be considered susceptible
to significant strength loss:
Fraction finer than 0.005 mm 15%

Liquid limit, LL 35%
Natural water content 0.9 LL

Liquidity index 0.75.
To account for the differences in Chinese and U.S. practice, the U.S. Army Crops of
Engineers modified the measured index properties (by decreasing the fines content by 5%,
increasing the liquid limit by 1%, and increasing the natural water content by 2%) before
applying the Chinese criteria to a clayey silt in the foundation of Sardis Dam (Finn et al.,
1994).
On the other end of the grain size spectrum, liquefaction of gravels has been observed in
the field (Coulter and Migliaccio, 1966; Wang, 1984; Youd et al., 1985; Yegian et al., 1994)
and in the laboratory (Wong et al., 1974; Evans and Seed, 1987). The effects of membrane
penetration are now thought to be responsible for the high liquefaction resistance observed
in early laboratory investigations of gravelly soils.
Liquefaction susceptibility is influenced by gradation. Well-graded soils are generally

less susceptible to liquefaction than poorly graded soils; the filling of voids between larger
particles by smaller particles in a well-graded soil results in the lower volume change potential
under drained conditions, and consequently lower excess pore pressures under undrained
conditions.
Particle shape can also influence liquefaction susceptibility. Soils with rounded particle
shapes are known to densify more easily than soils with angular grains. Consequently, they
are usually more susceptible to liquefaction than angular-grained soils. Particle founding
frequently occurs in the fluvial and alluvial environments where loosely deposited saturated
soils are frequently found, and liquefaction susceptibility is often high in those areas.
24.3 State Criteria

Even if a soil meets all of the preceding criteria for liquefaction susceptibility, it still may or
may not be susceptible to liquefaction. Liquefaction susceptibility also depends on the
initial state of the soil (i.e., its stress and density characteristics at the time of the earthquake).
Since the tendency to generate excess pore pressure of a particular soil is strongly influenced
by both density and initial stress conditions, liquefaction susceptibility depends strongly on
the initial state of the soil. These liquefaction susceptibility criteria are different for flow
liquefaction and cyclic mobility.
24.4 Identification of Liquefiable Soil and Liquefaction Studies
The following soils must be considered as potential subjects of liquefaction:
a) Muddy sand and silt with the following characteristics

degree of water saturation around 100%
Diameter at 50%, D50, between 0.05 mm and 1.5 mm.
b) Clayey soil with the following characteristics
diameter at 15%, D15, greater than 0.005 mm;
liquid limit, WL, less than 35%;
water content greater than 0.9 WL
In contrast, the following soil types may be considered as exempt from risk:
a) Soil with a grain-size diameter at 10% D10, greater than 2 mm

b) Soils which simultaneously have:

(Plasticity index)
Where the indications above reveal a possibility of liquefaction, there are good grounds for
quantifying the liquefaction risk by the methods described below.
24.4.1 Level A Studies
An initial qualitative study of the sensitivity to liquefaction is carried out using geological,
hydrogeological and geomorphological data. The zonation at Level A will result in a map of
liquefaction susceptibility. It will comprise of three categories:
Zones of very low or non-existent susceptibility

Zones of low to moderate susceptibility
Zones of high to very high susceptibility.
24.4.2 Level B Studies
Level B studies, and, therefore, those at Level C, use the mechanical characteristics of the
soil to create a liquefaction hazard map. In order to accomplish this, a simplified method
is used, originally developed based on the use of Standard Penetration Tests (SPT). A
certain number of boreholes, and SPT and/or Cone Penetration Tests (CPT) are required.
The cyclic shear stress e generated by an earthquake at a given depth where the
total vertical stress is v is
(28)
Where amax is the maximum acceleration at the surface, derived from the seismic-hazard
study, possibly modified to take into consideration the site effects, and rd is a coefficient of
stress attenuation as a function of soil rigidity. It can be assumed that:
(29)
The soil cyclic shear strength is calculated from its resistance to standard penetra-
tion. If N is the number of SPT blow, it can be defined as:
(30)
Where v, the effective vertical stress is expressed in kN/m2. Cyclic shear strength is given
by the equation:
(31)
Where the coefficient A is dependent on the fines content of the soil, and on the zone of
seismicity. The following values adapted from Ambraseys (1988) are appropriate:
Fines content of soil 5% > 10%
Seismicity zone I II III I II III
A 1.7x10-2 1x10-2 5x10-3 2.5x10-2 1.2x10-2 5x10-3
The values in zone III for material with more than 10% fines are slightly conservative due to
the lack of experimental data.
The safety coefficient (FS) is given by:

(32)
The results of the analyses are generally presented in the map format, with four zones
distinguished according to the value of the safety coefficient:
no liquefaction: non-liquefiable zones or FS 2.0

liquefaction unlikely: 1.5 FS 2.0
liquefaction probable: 1.0 FS < 1.5
liquefaction almost certain: FS < 1.0
The maximum investigation depth should be 25 m.
The geotechnical investigations required for the evaluation of the liquefaction hazard
are not generally sufficiently numerous to allow the tracing of precise zone boundaries,
particularly where the study area is very large. Geological criteria are, therefore, used to
make sure that quantitative mechanical information according to the type of sedimentary
deposit is included.
24.4.3 Level C Studies
This level of study is distinguished from Level B by the volume of investigations involved. It
will commonly be desirable to carry out complementary drilling with SPT and CPT. Undisturbed
samples should be taken in suspect formations.
The cyclic shear stress produced by an earthquake will be calculated on some

characteristic stratigraphic profiles by dynamic computation of wave propagation. These
calculations will enable further refinement of the variation of the coefficient (FS) rd, particularly
where profiles are very different.
Undrained cyclic shear strength will be further controlled through cyclic triaxial laboratory
tests, for areas where Level B studies revealed a safety coefficient FS between 1 and 1.5
(liquefaction probable).
For shallow foundations of small buildings, the impact of the depth and thickness of
the liquefiable layer will be evaluated. The coefficient of safety, FS, being calculated
as indicated in the following integral can be evaluated:
(33)
Where,
FL = 1 FS When FS 1
FL = 0 When FS 1
z is the depth in meters.
The value IL varies between 0 for a non-liquefiable zone and 1.00 for an area where
liquefaction is highly likely. The result of the analysis is presented in a map format, with four
zones distinguished according to their IL value:
No liquefaction : non-liquefiable zone or IL = 0

Liquefaction unlikely : 0 < IL 5
Liquefaction probable : 5 < IL 15
Liquefaction almost certain : IL > 15.
24.5 Liquefaction Susceptibility Map

Liquefaction susceptibility is a measure of a soils inherent resistance to liquefaction, and
can range from not susceptible, regardless of seismic loading, to highly susceptible, which
means that very little seismic energy is required to induce liquefaction. Susceptibility has
been evolved by comparing the properties of top soil deposits of the region under study to
the other soil deposits where liquefaction has been observed in the past (based on Seed et
al., 1985). Liquefaction susceptibility is evaluated based on the primary relevant soil properties
such as grain size, fine content, and density, degree of saturation, SPT-N values and age of
the soil deposit in each of the borelogs. These susceptible areas are identified by considering
the approach of Pearce and Baldwin (2005). Soil is susceptible for liquefaction if (1) there
are presence of sand layers at depths less than 20 m, (2) it encounters water table depth
less than 10 m, and (3) SPT value N blow counts is less than 20. By interpolation, a
susceptibility map can be prepared.
24.6 Factor of Safety against Liquefaction Assessment

Factor of Safety against liquefaction of soil layer are evaluated based on the simplified
procedure (Seed and Idriss, 1971) and subsequent revisions of the simplified procedures
(Seed et al., 1983 and 1985; Youd et al., 2001; Cetin et al., 2004). In this study, the earthquake
induced loading is expressed in terms of cyclic shear stress and this is compared with the
liquefaction resistance of the soil. Liquefaction estimation requires two variables for the
evaluation of liquefaction resistance of soils. Two variables are defined based on cyclic
stress approaches which are as follows.
1. The seismic demand of a soil layer is represented by a Cyclic Stress Ratio (CSR).
2. The capacity of soil to resist liquefaction is represented by Cyclic Resistance Ratio (CRR).
Here liquefaction resistance is estimated using an in-situ test based on corrected

SPT-N values. Steps involved in the calculation of liquefaction hazard are shown in Figure
18 as a flow chart.
Figure 18: Flow chart for liquefaction hazard assessment.

24.6.1 Factor of Safety against Liquefaction
If the cyclic stress ratio caused by the earthquake is greater than the cyclic resistance
ratio of in-situ soil, then liquefaction could occur during an earthquake. The factor of safety
against liquefaction is defined as follows:
(34)
Here subscript 7.5 for CRR denotes that CRR values calculated for the earthquake
moment magnitude of Mw 7.5. MSF is the magnitude scaling factor. The higher factor of
safety means that the soil has more resistance to liquefaction.
24.6.2 Peak Ground Acceleration
Estimation of factor of safety against liquefaction of soil layer requires the ground level peak
acceleration due to an earthquake. Ground level peak horizontal accelerations have to be
estimated by the site response studies using equivalent linear response analysis software
(using SHAKE2000 program).This peak ground acceleration is further used to estimate the
Cyclic Stress Ratio (CSR).
24.6.3 Cyclic Stress Ratio (CSR)
The excess pore pressure generation to initiate liquefaction depends on the amplitude and
the duration of the earthquake induced cyclic loading. In the cyclic stress approach, the pore
pressure generation is related to the cyclic shear stresses; hence the earthquake loading is
represented in terms of cyclic shear stresses. The earthquake loading can be evaluated by
using the simplified approach of Seed and Idriss (1971). The earthquake loading is evaluated
in terms of uniform cyclic shear stress amplitude and it is as given below:
Cyclic Stress Ratio (CSR) = 0.65 (35)
In this equation 0.65 represents 65% of the peak cyclic shear stress, amax is the
peak ground surface acceleration, g is the acceleration due gravity, and are the
total and effective vertical stresses and rd is the stress reduction coefficient. For the calcu-
lation of stress reduction coefficient many correlations are available which are discussed in
detail in a 1996 NCEER workshop report (Youd et al., 2001). Youd et al. (2001) recom-
mends that for routine practice and non-critical projects, the equations given by Liao and
Whitman (1986) may be used to estimate average values of rd as given below:
rd=1.00.00765z for z 9.15 m (36)
rd=1.1740.0267z for 9.15m < z 23 m (37)
24.6.4 Cyclic Resistance Ratio (CRR)
Liquefaction resistance of soil depends on how close the initial state of soil is to the state
corresponding to failure.The liquefaction resistance can be calculated based on laboratory
tests and in-situ tests. Here, liquefaction resistance using in-situ test based on SPT-N
values is usually attempted. Cyclic Resistance Ratio (CRR) is arrived at, based on the
corrected N value in accordance with Seed et al. (1985), Youd et al. (2001) and Cetin et al.
(2004). Seed et al. (1985) present a plot of CRR versus corrected N value from a large
volume of laboratory and field data. However, the corrected N values are used to calculate
CRR for magnitude Mw 7.5 earthquakes using the equation proposed by Idriss and Boulanger
(2005) as given below:
(38)
24.6.5 Magnitude Scaling Factor (MSF)
The CRR curves are generated by either using the SPT-N values or CPT qc values or shear
wave velocity Vs corresponding to an earthquake of magnitude Mw 7.5. Seed and Idriss
(1982) suggested the use of magnitude scaling factors (MSF) for earthquakes of magnitude
other than Mw 7.5. The available MSF are Seed and Idriss (1982) scaling factors, Revised
Idriss (1985) scaling factors, Ambraseys (1988) scaling factors, Arango (1996) scaling factors,
Andrus and Stokoe (1997) scaling factors and Youd and Noble (1997) scaling factors. Detailed
discussion and comparison of these scaling factors are available in Youd et al. (2001) and
Bhandari et al. (2003). An NCEER - 1996 and 1998 NCEER/NSF workshop (Youd et al.
2001) recommends the revised Idriss (1985) scaling factors and it was used by Yilmaz and
Bagci (2006) for soil liquefaction susceptibility and hazard mapping in the residential area of
Kutahya (Turkey). The magnitude-scaling factor used is the revised Idriss scaling factor for
the magnitude less than Mw 7.5 and it is given as,
MSF = (39)
24.6.6 Factor of Safety Calculation
Using the available geotechnical borelog data base, after applying necessary corrections to
SPT-N values, the corrected N [(N1)60cs] values are obtained. A simple excel spread sheet
is developed to automate these calculations for all the borelogs with depth. The factor of
safety for each layer of soil is arrived at considering the corresponding (N1)60cs values. A
typical liquefaction analysis is shown in (Table 12). It is to be noted here that, apart from the
recommendation of Seed et al. (1983), the fines content in the soil is considered using
representative parameters such as liquid limit (LL) and/ Plasticity Index (PI). The soil having
the liquid limit of more than 32 is recommended (Boulanger and Idriss, 2004) for the detailed
study (DS), which can account for the strength loss in plastic silts or clays during cyclic or
seismic loading.
Table 12: Typical liquefaction analysis for a borehole
Magnitude, Mw = 5.1 Peak Acceleration = 0.35g

Depth Corrected vo vo rd CSR FC Liquid CRR MSF FS
N value Limit
(m) (N1)60cs kN/m2 kN/m2 %
1.50 4 30.00 30.00 0.99 0.22 46.2 0 0.08 2.68 0.94

3.20 3 64.00 47.32 0.98 0.30 40.9 0 0.08 2.68 0.69
4.20 21 84.00 74.19 0.97 0.25 53.3 26 0.22 2.68 2.36
5.20 20 104.00 94.19 0.96 0.24 53.1 31 0.21 2.68 2.35
7.00 44 140.00 122.34 0.95 0.25 57.1 25 19.54 2.68 NL
8.50 102 170.00 155.29 0.93 0.23 59.2 27 NL 2.68 NL
24.6.7 Liquefaction Hazard Map
Liquefaction hazard mapping has been done by many researchers using SPT data (Palmer
et al., 2003; Brankman et al., 2004; Pearce and Baldwin, 2005; Yilmaz and Bagci, 2006).
The liquefaction hazard map is prepared for the desired maximum moment magnitude. The
minimum factor of safety from each borelog is considered to represent the factor of safety
against liquefaction at that location, which is used for the mapping. These factors of safety
against liquefaction are grouped into 4 classes as shown in Table 13 below.
Table 13: Factor of safety and severity index of liquefaction
Group Factor of safety range Severity Index

1 <1 Very Critical
2 1 to 2 Critical
3 2 to 3 Low critical
4 >3 Non liquefiable
24.7 Sample Computations of Liquefaction Hazard in Terms of Factor of Safety
The simplified procedures for liquefaction analyses of Seed and Idriss (1971 and 1981) are
widely used. A soil column is considered to be a rigid body, which gets excited at the base
with the propagation of shear wave to the ground surface. The shear stress is generated in
the soil column and can be calculated through the following expression,
(40)
Where ( max)r denotes maximum shear stress for rigid body, 0 is the total overburden pressure,
amax is the peak horizontal acceleration on the ground surface, and g represents acceleration
due to gravity.
In reality, the soil behaves as a deformable body instead of as a rigid body. Hence,
the rigid body shear stress should be reduced with a correction factor to obtain the deformable
body maximum shear stress ( max)d . This correction factor is called the stress reduction
coefficient (rd) and can be computed as follows,
(41)
The value of the stress reduction coefficient decreases with depth to a value of unity on the
ground surface as shown in Figure 19.
Figure 19: The variation of stress reduction factor (rd ) with depth (after Cetin et al., 2004).
The following equations have been derived for calculating the average value of rd (Cetin
et al., 2004),
, d < 20 m (42)
, d 20 m (43)
The above relations do not incorporate shear wave velocity. Different relations are
available in case shear wave velocity is to be employed (Cetin et al., 2004). On substitut-
ing, Equation (40) in Equation (41), the maximum shear stress for the deformable body
( max)d can be calculated as,
(44)
Seed and Idriss (1971) suggested that a value of 65% of the maximum shear stress
( max) is reasonably accurate. They based their prediction by appropriate weighting of labo-
ratory test data. Therefore, Equation (44) can be written in terms of equivalent average of
shear stress ( ave) as,
(45)
If the equivalent average of shear stress ( ave) is normalized with the initial effective
overburden pressure 0 , the term is called the seismic demand of a soil layer or CSR
(Cyclic Stress Ratio).
(46)
Another parameter that is employed is the Cyclic Resistance Ratio (CRR). It represents
the capacity of the soil to resist liquefaction. The SPT-N value data have been computed at
a depth spacing of 1.5 m at each of the boreholes in the study region. The effective overburden
pressure is given as below,
(47)
Where dj is the depth interval of each N-value, di is the depth at which the soil lithology
changes and i is the average unit weight of the material.
According to the classifications given by Youd et al. (2001), three relations can be
formulated as shown in Figure 20 for the increasing level of silt content or fine content (FC)
in the soil. These are given as follows,
y = 0.00000031*x5 0.000021*x4 + 0.00053*x3 0.0054*x2 + 0.03*x + 0.011, FC = 5% (48)

y = 0.0000012*x5 0.000087*x4 + 0.0024*x3 0.03*x2 + 0.18*x 0.32, FC = 15% (49)
y = 0.00000027*x5 0.00001*x4 + 0.00018*x3 0.0016*x2 + 0.02*x + 0.04, FC = 35% (50)
Where y=CRR and X= (N1)60.
The correlation between CRR and Corrected blow count is given in Figure 20. Thereafter,
the Factor of Safety (FOS) against the liquefaction hazard is computed as,
FOS = CRR / CSR (51)
The factor of safety represents a criteria which is given as
(i) FOS < 1 for liquefiable soil i.e., failed state.

(ii) FOS = 1 at the point of failure indicating vulnerability, and
(iii) FOS > 1 for non liquefiable soil i.e., stable state.
Figure 20: Correlations between CRR and Corrected blow count (N1)60 (after Youd et at., 2001).
24.8 Liquefaction Susceptibility of Soils with Plastic Fines
Since liquefaction requires development of excess pore pressure, liquefaction susceptibility

is influenced by the compositional characteristics that influence volume change behavior
and composition characteristics are associated with high volume change potential tending
to be associated with high susceptibility. These characteristics include particle size, shape
and gradation.
According to the observations after earthquakes in China, Wang (1979) established

that any clayey soil containing less than 1520% particles by weight, smaller than 0.005mm
and having water content (wc) to liquid limit (LL) ratio greater than 0.9 is susceptible to
liquefaction. Based on these data, Seed and Idriss (1982) stated that clayey soils could be
susceptible to liquefaction only if all three of the following conditions are met : (1) percent of
particles less than 0.005 mm is 15%, (2) LL<35, and (3) wc/LL>0.9. Due to its origin, this
standard is referred to as Chinese Criterion. It has been modified by different workers since
then. Based on the results of the cyclic testing performed in the study (Bray and Sancio,
2006), a soil may be susceptible to liquefaction if the ratio of the water content to liquid limit
Figure 21: Plasticity Index versus wc/LL exhibiting boreholes associated with different
liquefaction potentials non-susceptible, moderate susceptible and susceptible.
is greater than 0.85 (wc/LL>0.85) and the soil plasticity index is less than12 (12<PI). Soils
that do not meet these conditions but have plasticity index less than 18 (PI<18) and water
content to liquid limit ratio greater than 0.8 (wc/LL>0.8) may be moderately susceptible to
liquefaction. These soils, especially those satisfying the first set of requirements, should be
tested in the laboratory to assess their liquefaction susceptibility and strain potential under
the loading conditions existing in the field. Soils with PI>18 do not liquefy at low effective
stresses. However, structures founded on these soils, and for that matter, any soil, may
undergo significant deformations if the cyclic loads approach or exceed the dynamic strength
of the soil. The plasticity index versus wc/LL plot depicting clustering of points in different
liquefaction potential zones are shown in Figure 21 and the chart to determine volume strain
as a function of safety is given in Figure 22.
Figure 22: Chart to determine volumetric strain as a function of safety.
Finally, all the results from the application of simplified procedure as well as the fine
sand criteria are merged to prepare a FOS zonation map. An example for Guwahati Region,
India is depicted in Figure 23 where the Geographical Information System (GIS) is used to
create a buffer area of 500 m around the observation site and thereafter modified according
to surface geology of the site.
Figure 23: A Factor of Safety (FOS) zonation map in the Guwahati City against soil liquefaction
hazard assessed for a deterministic seismic hazard scenario of MW 8.7 nucleating
from Shillong plateau at the hypocenter of the 1897 Shillong Earthquake.
24.9 Factor of Safety before Earthquake
Factor of safety has been calculated using the volumetric strain which is calculated using
the following approach of Ishihara and Yoshimine (1992):
Specific density (52)
e0 = ef + e
where
and Dr (%) Before Earthquake = (1 e0)2

So from Dr (%) before Earthquake, one can calculate the (N1)60 before earthquake.
Where vol is the volumetric strain.

After estimating the factor of safety before and after an earthquake, one can see the
effect of the earthquake and interpret its impact on the settlement in terms of liquefaction
potential.
25. EARTHQUAKE INDUCED LANDSLIDES

Landslides occur on a regular basis throughout the world as part of the ongoing evolution of
landscapes. Many landslides occur in natural slopes, but slides also occur in man-made
slopes from time to time. Hence, at any given point of time, slopes exist in states ranging
from very stable to marginally stable. When an earthquake occurs, the effects of earthquake-
induced ground shaking is often sufficient to cause failure of slopes that were marginally to
moderately stable before the earthquake.
Earthquake induced landslides, which have been documented from as early as 1789
B.C. have caused tremendous amounts of damage throughout the history. In many
earthquakes, landslides have been responsible for as much or more damage than all other
seismic hazards combined. In the 1964 Alaska earthquake, for example, an estimated 56%
of the total cost of damage was caused by earthquake-induced landslides (Youd and Perkins,
1978; Wilson and Keefer, 1985). Kobayashi (1981) found that more than half of all deaths in
large (M>6.9) earthquakes in Japan between 1964 and 1980 were caused by landslides.
The 1920 Haiyuan earthquake (M=8.5) in the Ningxia Province of China produced hundreds
of large landslides that caused more than 100,000 deaths.
25.1 Types of Earthquake Induced Landslides

Many factors, including geologic and hydrologic conditions, topography, climate, weathering,
and landuse influence the stability of slopes and the characteristics of landslides. A number
of procedures for the classification of landslides have been proposed; that of Varnes (1978)
being perhaps the most widely used one in the United States. Similar principles and
terminology can be used to classify earthquake-induced landslides on the basis of material
type (soil or rock), character of movement (disrupted or coherent), and other attributes,
such as velocity, depth and water content. Earthquake induced landslides can be divided
into three main categories: disrupted slides and falls, coherent slides, and lateral spreads
and flows.
Disrupted slides and falls include rock falls, rock slides, rock avalanches, soil falls,
disrupted soil slides, and soil avalanches. The earth materials involved in such failures are
sheared, broken, and disturbed into a nearly random order. These types of failures, usually
found in steep terrain, can produce extremely rapid movements and devastating damages;
rock avalanches and rock falls have historically been among the leading causes of death
from earthquake induced landslides.
Coherent slides, such as rock and soil slumps, rock and soil black slides, and slow
earth flows, generally consist of a few coherent blocks that translate or rotate on somewhat
deeper failure surfaces in moderate to steeply sloping terrains. Most coherent slides occur
at lower velocities than the disrupted slides and falls. Lateral spreads and flows generally
involve liquefiable soils, although sensitive clays can produce landslides with very similar
characteristics. Due to the low residual strength of these materials, sliding can occur on
remarkably flat slopes and produce very high velocities.
The different types of earthquake induced landslides occur with different frequencies.
Rock falls, disrupted soil slides, and rock slides appear to be the most common types of
landslides observed in the historical earthquakes. Subaqueous landslides, slow earth flows,
rock block slides, and rock avalanches are least common, although the difficulty of observing
subaqueous slides may contribute to their apparent rarity.
For preliminary stability evaluations, knowledge of the conditions under which earthquake
induced landslides have occurred due to past earthquakes is useful. It is logical to expect
that the extent of earthquake-induced landslide activity should increase with increasing
earthquake magnitude and that there could be a minimum magnitude below which earthquake-
induced landsliding would rarely occur. It is equally logical to expect that the extent of
earthquake-induced landslide activity should decrease with increasing source to site distance
and that there could be a distance beyond which landslides would not be expected in
earthquakes of a given size.
A study of 300 U.S. earthquakes between 1958 and 1977 showed that the smallest
earthquakes noted to have produced landslides had magnitudes of about 4.0 (Keefer, 1984).
Where magnitudes were not available, minimum Modified Mercalli Intensity (MMI) values of
IV and V have been observed for disrupted slides or falls and other types of slides, respectively.
Although these empirically based limits are useful, their approximate nature must be
recognized; failure of slopes that are near the brink of failure under static conditions could
be produced by quite weak earthquake shaking.
The maximum sources to site distances at which landslides have been produced in
historical earthquakes are different for different types of landslides. Disrupted slides or falls,
for example, have rarely been found beyond epicentral distances of about 15 km for M=5
events but have been observed as far as about 200 km in M=7 earthquakes. Note that the
curve for lateral spreads and flows correlates reasonably well with the magnitude distance
curve for liquefaction. Regional differences in attenuation behavior have little apparent influence
on the area of earthquake induced landsliding.
25.2 Landslide Hazard and Risk Analysis in India at a Regional Scale

Several geological hazards (active deformation and faulting, landslide, rock falling,
earthquakes) affect at least 15% of land area of India. The southern, central and western
mountainous regions namely, the Western Ghats, Satpura and Vindhyan Range and Aravalis
are geologically very old and stable formations as compared to the Himalayas and Shiwalik
Ranges in the North. These recent formations which lie in the active seismic zones are
geologically unstable, and are still in the upheaval stage (Valdiya, 1975). Recently, there
has been a very rapid increase in the developmental activities such as large scale construction
of mountain roads, mining, overgrazing, urbanization, deforestation and opening up of steep
land for agriculture in the Himalaya and the peninsular India. Mass movement and rock
falling are two major types of landslides that occur in these regions. Landslides coupled
with earthquakes in this high-risk areas act as double disasters for the villages, farmlands
and roads as well as the exacerbation of erosion of the land surface. Detecting landslide
and monitoring their activities using Remote Sensing and GIS tools provide a wealth of
information for assessing and mapping landslide hazards. The major advantages of using
GIS in Landslide Hazard Zonation is the utilization of multiple datasets as vector layers
deciphering their importance for preparing hazard zonation and subsequent risk analysis.
Many methodologies have been proposed for the spatial landslide hazard zonation. Two
approaches are the most promising ones, namely the methods based on the statistical
analysis of geo-environmental factors related to the occurrence of landslides (Information
Value Based Method), and the deterministic modeling based on simple mechanical laws
that control slope instability (Guzzetti et al., 1999). The Information Value Based Method
(IVBM) (Yin and Yan, 1988; Wu et al., 2000) is the one used for the recent investigations
(Dasadhikari et al., 2011). This statistical approach is based on the observed relationship
between each factor and the past and present landslide distributions (Carrara et al., 1991).
The input thematic datasets are converted into a grid format (i.e., rastarization) and various
factor layers that control the landsliding are prepared in ArcGIS platform. The Landslide
Hazard Zonation and Risk Assessment have been accomplished in six stages as shown in
a Logic Tree Framework in Figure 24 (Dasadhikari et al., 2011).
Figure 24: Methodology for landslide hazard zonation and risk analysis.1,25,6 are repre-
sent six stages respectively; 1. Thematic Layer, 2. Data Quantification, 3. Calcula-
tion of Score Factor, 4. Integration, 5. Model Testing Zonation & Classification, and
6. Risk analysis (after Dasadhikari et al., 2011).
25.2.1 Data Quantification

By using remote sensing and GIS, all thematic maps are quantified and rasterised to a
specific pixel size (1 km x 1 km). Entire qualitative thematic information (vector base) have
been converted to quantitative (raster base) data sets. A binary method is used for cross
matching of each parameter with respect to the landslide occurrence map.
25.3 Landslide Risk Analysis

In a vulnerable area, landslides mainly affect the population, buildings, transportation
networks, and agricultural activities. Landslide risk can be estimated based on the following
equation:
Risk = hazard * exposure (53)
The exposure factor is estimated by the combination of population density, landuse /

landcover, and road network data layers. Generally, larger values are assigned to population
areas, built-up areas and road networks to reflect their higher significance in determining
the impact of landslides.
26. SIMULATION OF STRONG GROUND MOTION

Recordings of high-frequency ground motion provide the information about the seismic
sources. At the same time geotechnical engineers use the strong motion data for structural
design, urban planning and management. The paucity of strong ground motion recordings
at all magnitude ranges necessitates the generation of synthetic data. The shape, scaling
of source spectra and seismic moment form the basis of simulating ground motions. Several
techniques are available that differ in the theoretical considerations, data and computational
requirements. Realistic results vis--vis complexity and computational as well as data
requirements are the deciding factors for the applicability of a particular technique. Some of
the techniques widely used include: (i) stochastic approach (Boore and Atkinson, 1987;
Beresnev and Atkinson, 1997; Motazedian and Atkinson, 2005), (ii) Green functions method
(Bouchon and Aki, 1977), (iii) Empirical Green Functions (EGF) method (Hartzell, 1978;
Irikura, 1983), (iv) Finite Difference method (Panza, 1985; Oprsal and Zahradnik, 2002), (v)
Finite Element method (Frankel, 1989), and (vi) Spectral Finite Element method (Komatitsch
and Tromp, 1999). The Frequency-Wave number (F-K) integration method has proved to be
very useful in understanding wave progression in layered media as well as in the complex
geological domains. Most often, the F-K integration approach accurately reflects the wave
propagation phenomena and is useful for site-specific simulations. In the Empirical Greens
Function method, a moderate event is simulated taking into account some a-priori information
about source and path using small recorded events.
The quantitative assessment of seismic hazard necessitates measurement of peak

ground motion parameter (e.g., PGA/PGV/PGD) from earthquake records. Paucity of strong
ground motion data records under conditions similar to design earthquakes in terms of
tectonic regime, earthquake size, local geology, and near fault conditions necessitates
analytical or numerical approaches for a realistic prognosis of the possible seismic effects.
The strong ground motion modeling must accommodate: (i) the seismic wave radiation from
a fault rupture, (ii) propagation through the crust, and (iii) modifications by the site conditions.
Two types of modeling exist in seismology: forward modeling and inverse modeling.
Forward modeling deals with the estimation of ground motion at the surface by modeling
the earthquake faulting process, the medium of propagation between the earthquake source
and the station, and local site effects near the station, such as modeling of topography,
basin structure, and soft soil conditions. Inverse modeling, on the other hand, uses recorded
ground motion data and tries to model the earthquake source processes. For engineering
purposes, estimation of ground motion at a location for a future large earthquake is important.
Therefore, forward modeling is used on many occasions for strong ground motion estimation,
using the results of inverse modeling as input, if available. In the following paragraphs
forward modeling will be dealt with.
26.1 Stochastic Simulation
The high frequency accelerations are highly incoherent, and are the result of irregularities of
the rupture front, heterogeneity of co-seismic slip distribution, fault geometry irregularity,
propagation transfer, rupture bifurcation, and sudden changes in rupture velocity and slip
amplitude. These details are unknown and unpredictable and probably impossible to model
in a deterministic manner. Therefore, in strong ground motion simulations the incoherence
of ground accelerations is accounted for using stochastic methods. Stochastic approaches
to the simulation of strong ground motion filter and window the white-noise time series
according to seismologically determined average spectra and duration (Boore, 1983). This
is a widely used and easily applicable method for the simulation of higher-frequency motions;
however, it lacks low frequency information. In a recent study by Tumarkin and Archuleta
(1997), there have been efforts towards incorporating the rupture directivity effect into the
stochastic method by adjusting the corner frequency, and towards increased low frequency
content by the use of a new filter.
The stochastic approach is a powerful simulator of high frequency ground motion

(f>0.1 Hz). It has been widely used to predict the ground motion around the globe where
earthquake recordings are meager. This method has been used to match the observed waveform
caused even by very high magnitude or seismic moment earthquakes in different tectonic
environments. One of the essential characteristics of the method is that it uses the simple
functional forms of various factors affecting the ground motions (source, path, and site).
It uses the convolution model to simulate spectral acceleration. Time domain simulation
is done as in the following steps:
White noise (Gaussian or uniform) is generated for a given duration of the strong motion
It is windowed to give a shape like an accelerograms using some shaping window.
A Boxcar function or window given by Saragoni and Hart (1974) can be used for
the purpose
The windowed noise is then transformed into frequency domain and it is normalized
The convolution model is then multiplied with this normalized spectrum, and
Inverse Fourier transformation is performed to simulate the signal in time domain.
In the entire process, Safak and Boore (1988) showed that the order of windowing and
filtering is important; if the white noise is first filtered and then windowed the long-period level
of the motion is distorted.
26.1.1 Stochastic Simulations for Point Source Earthquakes

This is essentially an engineering approach to the simulation of strong ground motion.
Ground acceleration is modeled as a filtered Gaussian white noise modulated by a
deterministic envelope function (Safak, 1988). The filter parameters are determined by
either matching the empirical properties of the spectrum of the strong ground motion (e.g.,
Trifunac et al., 1988), theoretical spectral shapes (i.e., the Kanai-Tajimi spectrum; Housner
and Jennings, 1964), or are determined on the basis of reliable physical characteristics of
the earthquake source and propagation media (e.g., Hanks and McGuire, 1981; Boore,
1983). The model used in the latter stochastic simulations is essentially an S-wave ground
motion spectrum based on the far-field model of Brune (1970). It has been found to satisfy
the main parameters of high frequency ground motion for earthquakes within a wide magnitude
range (McGuire and Hanks, 1980). The theoretical considerations are given below.
The amplitude spectrum A() can be written, in the frequency domain, as the product
of source function SO(, c), a propagation path term P(), and a site function SI() (Nath
et al., 2005; Lermo and Chavez-Garcia, 1993; Boore, 1983) as
A() = SO(, c) SI().P() (54)
Acceleration spectra often shows a sharp decrease with increasing frequency so a

high cut filter F(, m) is incorporated in Equation (54) such that,
A() = SO(, c) SI().P().F(, m) (55)
Papageorgiou and Aki (1983) related m to source processes while Hanks (1982) as-
sociated m to attenuation near the recording site. However, the high-cut filter F(, m) has
been taken to be
F(, m)=[1+( / m)2s]1/2 (56)
Where s controls the decay rate at higher frequencies. After comparing with several
observed spectra, it has been assigned a value of 4 (Boore, 1983). Sometimes the high-
cut filter F() given by Anderson and Hough (1984) has also been used
(57)
Where k is a spectral decay parameter and controls the decay rate at higher frequencies.
Brunes 2 circular crack source model (Hwang and Huo, 1997) defines earthquake
source to be a circular fault for which the ground acceleration spectrum decays with -2 for
frequencies higher than source corner frequencies. The point-source spectrum of this model
is given as,
(58)
Where R (=0.63) is the radiation pattern averaged over an appropriate range of azimuths
and take-off angles, FS (=2.0) accounts for the amplification of the seismic wave at the free
surface, is the crustal density of the continental crust at the focal depth and is the
shear-wave velocity at the source region. M0 and c are the scalar moment and corner
frequency respectively. The corner frequency (fc=c/2) given in terms of moment is,
fc = 4.9 106
* * * (59)
where is the stress drop in bars, fc is in Hertz, in kilometers/sec and M0 in dyne-cm

(Brune, 1970). The propagation path term can be expressed as,
(60)
Where G(R) accounts for geometrical spreading, and Q() is a shear wave frequency-
dependent quality factor of the medium. Following Ordaz and Singh (1992) and Castro et al.
(1996) the following has been considered to take into account the possible arrival of surface
waves in the windowed data,
G(R) = 1/R for R<100 km

= (R*100)-0.5 for R>100 km. (61)
26.1.2 Random Vibration Theory
Peak ground accelerations can be predicted using random vibration theory (Lai, 1982) without
calculating the time series. Although the time series has been generated but this theory is
preferred, since we can avoid reliance on the additional comparison that may be required for
calculating the time series. Here E (arms) is determined by arms using the following relations
(Boore, 1983). The kth moments mk of energy density spectrum of acceleration are defined
as,
mk= (62)
The rms acceleration is given by,
(63)
Where T is the duration of an accelerogram. For accelerograms in which the number of

extremes N is less than 2, N is arbitrarily set equal to 2. If the value of N exceeds 20, an
asymptotic approximation given below should be used for E(amax) (Cartwright and Longuet-
Higgins, 1956).
(64)
Where N is determined by (65)

and f is predominant frequency and is calculated by
(66)
The accelerograms are computed by stochastic simulation both in time as well as

frequency domains taking into consideration the source, site and path effect to verify the
derived parameters from strong motion data.
26.1.3 Stochastic Simulations for Finite Fault Rupture Earthquakes
Beresnev and Atkinson (1997) have developed a procedure (FINSIM) for the stochastic simulation
of strong ground motion from finite fault ruptures. The fault rupture plane is modeled with an
array of subfaults. The radiation from each subfault is modeled as a point source with a 2
spectrum, similar to Boore (1983). The fault rupture initiates at the hypocenter and spreads
uniformly along the fault plane with a constant rupture velocity triggering radiation from subfaults
in succession. Simulations from each subfault, properly lagged and summed at the observation
point, provide the simulation of ground motion from the modeled finite fault rupture. The size
of the subfaults controls the overall spectral shape at medium frequencies. Optimal subfault
size ( ) is given as (Beresnev and Atkinson, 1999):
log ( ) = 2.0 + 0.4 Mw (67)
The total number of subfaults is controlled by the constraint that the total seismic
moment of the subfaults must be equal to the target seismic moment. The quantitative
assessment of seismic hazard necessitates measurement of peak ground motion parameter
(e.g., PGA) from earthquake records. Paucity of strong ground motion data records under
conditions similar to design earthquakes in terms of tectonic regime, earthquake size, local
geology, and near fault conditions necessitates analytical or numerical approaches for a
realistic prognosis of the possible seismic effects.
The current trend in ground-motion prediction is shifting towards an extended, or finite-

fault, source representation. The point source approximation is clearly unable to characterize
key features of ground motions from large earthquakes, such as their long duration and the
dependence of amplitudes and duration on the azimuth of the observation point (source
directivity). Finite-fault effects contribute not only to the duration and directivity of ground
motions; they also affect the shape of the spectra of seismic waves. In this model, the
finite-fault plane is subdivided into elements (subfaults), and the radiation from a large
earthquake is obtained as the sum of contributions from all the elements, each of which
acts as an independent (sub) source. In the typical implementation, the rupture starts at a
hypocentral point on the fault and propagates radially from it, triggering the subfaults as it
passes them. The fields from all subevents are geometrically delayed and added together
at the observation point.
The stochastic method for strong ground motion simulation of Beresnev and Atkinson
(1997) employs a finite fault. The dimension of the sub-fault can be calculated using a
relation given by Beresnev and Atkinson (1997). However, a large uncertainty is associated
with this equation due to paucity of large earthquake magnitude data. To overcome this
limitation to a large extent, the algorithm is further modified by Motazedian and Atkinson
(2005) by introducing dynamic corner frequencies. The enhanced approach conserves the
radiated energy at high frequencies at any sampling of sub-fault size and controls the
relative amplitude of higher versus lower frequencies.
Accordingly the Fourier amplitude spectrum due to the nth sub-fault is given as,
(68)
Where C is a scaling factor, Mn is the nth sub-fault moment, Hn is the spreading factor
responsible for conserving the energy at high frequency spectral level of the sub-fault, and
fon(t) is the dynamic corner frequency. The scaling factor is given as,
C= (69)
The moment Mn of the nth sub-fault is calculated using the slip distribution as follows,
Mn = (70)
Where D n is the corresponding average slip. The dynamic corner frequency is expressed
as follows,
fon(t) = 4.9 x 106 (71)
Where N R (t) is the number of rupture sub-faults at a time, t. The spreading factor in Equa-
tion (68) is computed as,
Hn =
(72)
Where f 0 is the corner frequency at the end of the rupture.
The strong ground motion modeling must accommodate:
Seismic wave radiation from a fault rupture,

Propagation through the crust, and
Modifications by the site conditions.
26.2 Dynamic Corner Frequency Concept
In finite-fault modeling, ruptured area is time dependent. It is initially zero and is finally
considered equal to the entire fault area. If the rupture stops at the end of the first subfault,
the corner frequency is inversely proportional to the area of the first subfault. Eventually,
when the rupture stops at the end of the Nth subfault, the corner frequency is inversely
proportional to the entire ruptured area. Thus, if corner frequency is inversely proportional to
the ruptured area, it follows that the corner frequency in finite-fault modeling can be considered
as a function of time. Similarly, it also follows that corner frequency should be decreasing
as the signal duration builds. Thus, at each moment of time the corner frequency depends
on the cumulative ruptured area. The rupture begins with higher corner frequencies and
progresses to lower corner frequencies.
26.3 Pulsating Area
In an actual earthquake rupture, the slip may only be occurring on a part of the fault at any
one time. The Self-healing model proposed by Heaton (1990) considers that duration of
slip at any location on the fault is short. In the stochastic modeling, this is possible only
when some part of the fault is actively pulsatinging at any time. In such a model, the areas
that are actively pulsating contribute to the ground-motion radiation, but the other areas on
the fault are passive. The passive cells will have no effect on the dynamic corner frequency.
The active area will move along the fault as the rupture progresses. By decreasing the
pulsating area, the amplitudes at low frequencies decrease; thus, a narrow pulsating area
on the fault results in lower amplitudes at longer periods and lower energy radiation. Variation
of this parameter can be used to adjust relative amplitudes of low-frequency motion in finite-
fault modeling. Note that the pulsating length cannot be less than the length of one subfault
(Motazedian and Atkinson, 2005).
26.4 Waveform Synthesis using F-K Integration
For the computation of synthetic accelerograms, an impulsive source has been used as a
first approximation for the near-field effect. The Wavenumber integration method of Herrmann
and Mandal (1986) is then followed.
The generation of synthetic seismograms for point sources in simply layered structures
has made rapid advances in the past decade. Two approaches involving Laplace Transform
and Fourier Transform are actively being pursued. The Laplace Transform or Cagniard-de
Hoop technique, usually referred to as the generalized ray method (Helmberger, 1968),
constructs the solution by tracking the individual seismic arrivals ray by ray from the source
to receivers. This method is valid at higher frequencies and works well at predicting particular
phases, but is poorly suited to models with many layers and larger distances when a
complete seismogram is desired. The other approach involves expressing the solutions in
terms of a double integral transformation over wavenumber and frequency (Hudson, 1969).
The complete solution rather than individual rays is considered in such a full wave theory
approach. This method can handle a larger number of plane layers, but requires considerable
computational efforts especially at higher frequencies.
Suppose that an earthquake can be represented by a double-couple without moment

source model with the symbols n and f representing the vector normal to the fault and the
direction of force respectively (as used by Haskell, 1964). The Fourier transform of vertical,
radial and tangential components of the displacement can be given as,
(73)
(74)
(75)
where ZDD, ZDS, ZSS, RSS, RDS, RDD, TSS and TDS are referred to as Greens Functions.
RSS and RDS in Equation (74) also include near-field terms. These terms decrease faster
than the others, and therefore, are important only at short distances.
The inverse Fourier transforms (73), (74) and (75) on multiplication of 2, needs a
convolution of the source spectra for the generation of acceleration time history of the
vertical, radial and tangential components of ground motion as given below,
(76)
Where S() is the Fourier spectra of the impulse source function as described by Herrmann
(1979).
26.5 Factors Affecting Earthquake Strong Ground Motions
Findings (e.g., Somerville, 2000) indicate that while the average ground motions from one
large earthquake are similar to those of another, there are conditions that cause the ground
motions to vary significantly from one location to another at the same distance from a given
event. This variability is related to earthquake source process, propagation and site response.
It has been well recognized that earthquake ground motion is affected by earthquake
source conditions, source-to-site transmission path properties, and site conditions. The
source conditions include the stress drop, source depth, size of the rupture area, slip
distribution, rise time, type of faulting, and rupture directivity. The transmission path properties
include the crustal structure and the shear-wave velocity and damping characteristics of
the crustal rock. The site conditions include the rock properties beneath the site to depths
of up to about few kilometers, the local soil conditions, and the topography of the site.
26.5.1 Effects of the Earthquake Source
Seismic Moment, Stress Drop, Effective Stress and the Corner Frequency are the main
parameters of the earthquake source that influences the strong ground motion characteristics.
A short review of these parameters is provided below.
Seismic Moment, M0, is the most recognized measure of the earthquake size given
by the multiplication of the shear modulus (Lames constant) of the medium, the average
total dislocation (i.e., mean fault offset or slip) and the area of the dislocation surface (i.e.,
fault rupture surface). The seismic moment, M0, is generally regarded as the best available
single number to describe the size of an earthquake and can be estimated from the low
frequency asymptote of the Fourier transform of the displacement seismogram. The moment
magnitude (Mw) is derived from the seismic moment on the basis of the following equation
(Kanamori, 1977)
Mw = (2/3) log Mo 10.73 (77)
Effective Stress is the difference between the initial static stress and frictional stress
in existence during the rupture process.
Corner Frequency is the frequency where the high and low frequency trends of the
Fourier Amplitude Spectrum intersect. It is related to the inverse of the rise time (rate of
growth in dislocation or roughly the time duration of rupture). By measuring the corner
frequency from the Fourier Amplitude Spectrum the apparent duration of faulting at the
source and hence the fault dimension can be estimated. A large magnitude earthquake will
generally have a large fault dimension and hence a small corner frequency, implying a
longer rupture and also strong motion duration.
26.5.2 Subduction Zone and Shallow Crustal Earthquakes
The collision of tectonic plates in subduction zones causes large and deep earthquakes.
Ground motion data from subduction zone earthquakes are associated with slower rate of
attenuation compared to those from shallow crustal earthquakes. Analysis of ground motion
data also indicates that the response spectral shapes obtained from subduction zone
earthquakes have smaller amplitudes in the log-period range than the response spectral
shapes from shallow crustal earthquakes.
26.5.3 Effects of Distance
Attenuation is strongly influenced by distance for both the geometric spreading and the
material damping. Excluding material damping and considering only geometric attenuation
it can be observed that the cylindrical body waves are attenuated with inverse of distance
and spherical body waves attenuate with the inverse of the distance squared. The material
attenuation is generally given by the following expression:
(78)
27. GROUND MOTION PREDICTION EQUATIONS
An evaluation of seismic hazard, whether deterministic (scenario based) or probabilistic,

requires an estimate of the expected ground motion at the site of interest. The most common
means of estimating this ground motion in engineering practices, including probabilistic
seismic hazard analysis, is the use of attenuation relationships. An attenuation relation is
a mathematical expression that relates a specific strong motion parameter of ground shaking
to one or more seismological parameters of an earthquake. These seismological parameters
quantitatively characterize the earthquake source, the wave propagation path between the
source and the site, and the soil and geological profiles beneath the site. Attenuation
relationships have been established for many ground motion parameters including
Peak Ground Acceleration (PGA)

Peak Ground Velocity (PGV)
Peak Ground Displacement (PGD)
Spectral Quantities.
Attenuation relationships are developed, by statistical evaluation of a large set of ground

motion data, for different regions and fault types (e.g., interplate versus intraplate, strike
slip versus subduction). Various investigators have proposed attenuation relationships for
PGA. All the commonly used attenuation relationships (Figure 25) assume that the ground
motion values are lognormally distributed (i.e., the logarithm of the value is normally
distributed).
Figure 25: Example of ground motion prediction equations.
27.1 Development of Ground Motion Prediction Equations
The rapid estimations of the ground motion parameters at a site of interest are often achieved
by using a ground motion prediction relationship that relates a specific strong ground motion
parameter of ground shaking to one or more attributes of an earthquake (e.g., Sadigh et al.,
1997; Abrahamson and Silva, 1997; Campbell and Bozorgnia, 2003; Atkinson and Boore,
2003; Nath et al., 2005; Nath et al., 2009). Generally employed ground motion parameters
include PGA, PGV, pseudo-spectral acceleration or velocity PSA or PSV, and intensity.
These parameters are found to increase with magnitude while decreasing with the epicenter
distance and are also controlled by the fault-rupture directivity, and site conditions. Some
typical global attenuation relationships widely used are given in Table 14A while those used
for seismic hazard estimation in Indian conditions developed for various tectonic provinces
are given in Table 14B.
Table 14A: Some Typical Global Attenuation Relationships
Sl. Data Source Relationship Reference

No.
1. San Fernando log PGA = 190 / R1.83 Donovan

earthquake Where R is the distance to the source in (1973)
(February 9, 1971) km.
2. California PGA = yo/(1+ (R / h)2) Blume

earthquake Where log yo = (b+3) + 0.81M 0.027M2 (1965)
and b is a site factor,
M = Earthquake Magnitude,
R = Reserve energy reduction coefficient,
h = Focal Depth (miles).
(0.61M plog R + 0.167 1.83 / R)

3. California and 10 Kanai
Japanese Where M = Earthquake Magnitude, (1966)

earthquakes R = Epicentral Distance,
TG = Time period.
4. Cloud (1963) PGA = 0.0069e1.64M / (1.1e1.1M + R2) Milne and

Where M = Earthquake magnitude, Davenport
R is the distance to the source in km. (1969)
5. Cloud (1963) PGA = 1.254e0.8M / (R + 25)2 Esteva

Where M = Earthquake magnitude, (1970)
R is the distance to the source in km.

No.
6. U.S.C. and G.S. log PGA = (6.5 2 log(R +80)) / 981 Cloud and
Where R = Epicentral Distance in km. Perez(1971)
7. 303 Instrumental PGA = 1.325e0.67M / (R + 25)1.6 Donovan

Values Where M = Earthquake magnitude, (1973)
8. Western U.S. PGA = 0.0193e0.8M / (R2 + 400) Donovan

records Where M = Earthquake magnitude, (1973)
9. U.S., Japan PGA = 1.35e0.58M / (R + 25)1.52 Donovan

Where M = Earthquake magnitude, R is the (1973)
distance to the source in km.
10. Western U.S. In PGA = 3.99 + 1.28M 1.75 ln [R + Campbell

records, USSR 0.147e0.732M] (1981)
and Iran Where M is the surface wave magnitude for M
greater than or equal to 6, or it is the local
magnitude for M less than 6, R is fault distance.
11. Western U.S. log PGA = 1.02 + 0.249M Joyner and

records and log 0.00255 Boore
worldwide Where PGA is peak horizontal acceleration in g, (1981)
M is moment magnitude,
R is the closest distance to the surface
projection of the fault rupture in km.
12. Western U.S. log PGA = 0.49 + 0.23(M 6) Joyner and

records and log 0.0027 Boore
worldwide Where, M is moment magnitude, (1982)
projection of the fault rupture in km.

No.
13. Western U.S. In PGA = ln (M) (M) In(R+20) Idriss

records Where M is the surface wave magnitude for M (1985)
greater than or equal to 6, or it is the local
magnitude for smaller M,
R is the closest distance to source for M
greater than 6 and hypocentral distance for M
smaller than 6,
(M) and (M) are magnitude-dependent
coefficients.
14. Italian records In PGA = 1.562 + 0.306M log Sabetta and

+ 0.169S Pugliese
Where M is Earthquake Magnitude, (1987)
projection of the fault rupture in kilometers,
S is 1.0 for soft sites or 0.0 for rock.
15. Western U. S. and For M less than 6.5, Sadigh et al.

worldwide (soil In PGA = 2.611 + 1.1M 1.75 In[R + (1986)
sites) 0.822e0.418M]
For M greater than or equal to 6.5,
In PGA = 2.611 + 1.1M 1.75 In[R +
0.316e0.629M]
Where M is Earthquake Magnitude,
projection of the fault rupture in kilometers.
16. Western U. S. and For M less than 6.5, Sadigh

worldwide (rock In PGA = 1.406 + 1.1M 2.05 In[R + et al.
sites) 1.353e0.406M] (1986)
For M greater than or equal to 6.5,
In PGA = 1.406 + 1.1M 2.05 In[R +
0.579e0.537M]

No.
Where, M is Earthquake Magnitude,

projection of the fault rupture in kilometers.
17. Worldwide In PGA = 3.512 + 0.904M 1.328 Campbell

earthquakes In + [1.125 0.112 and
In R 0.0957M] F + [0.440 0.171 Bozorgnia
In R]Ssr + [0.405 0.222 In R]Shr (1994)
Where F = 0 for strike-slip and normal fault

earthquakes and 1 for reverse, reverse-oblique,
and thrust fault earthquakes,
Ssr = 1 for soft rock and 0 for hard rock and
alluvium,
Shr = 1 for hard rock and 0 for soft rock and
alluvium,
R = fault distance,
M = Earthquake Magnitude.
18. Western North In PGA = b + 0.527(M 6.0) 0.778 Boore et al.

American
In (1997)
earthquakes
Where b = 0.313 for strike-slip earthquakes,
0.117 for reverse-slip earthquakes, and
0.242 if mechanism is not specified,
Vs is the average shear wave velocity of the
soil in (m/sec) over the upper 30 meters,
The equation can be used for magnitudes (M)
of 5.5 to 7.5 and for distances not greater than
80 km,
R is fault distance.
Table 14B: Typical Ground Motion Prediction Equations used for Seismic Hazard
Estimation in Indian conditions developed for various tectonic provinces
Sl. Region Relationship Reference

No.
1. Sikkim Himalaya, First order attenuation relation: Nath et al.

Shallow active ln(Y) = 3.6 + 0.72M 1.08 ln r + 0.007r (2005)
tectonic crust Where Y is the strong motion parameter (PGA),
M is earthquake magnitude,
r is a measure of source to site distance.
Second order attenuation relation:
ln(PGA) = ln(SA) a1 (a2 a3M) + a4h a5 ln(SR)
Where h is the site elevation,
SR the site response,
SA spectral acceleration at respective
frequencies.
2. Peninsular India, ln(ybr) = C1 + C2(M 6) + C3(M 6)2 ln(r) C4r + ln( br) Raghukanth
SCR Where C1, C2,,C4 are regression coefficients, and Iyengar
ybr = (Sa/g) stands for the ratio of spectral (2007)
acceleration at bedrock level to acceleration
due to gravity,
M and r refers to moment magnitude and
hypocentral distance respectively,
ln( br) is the error term.
3. Himalayas and log A = b1+b2Mw b3 log +b5S + b6H Sharma

Zagros region, Where b1, b2, .., b6 are regression coefficients, et al. (2009)
Shallow active A is the spectral acceleration in terms of
tectonic crust m/sec2,
S is 1 for rock site and 0 otherwise,
H is 1 for strike slip mechanism and 0 for
reverse mechanism,
Mw is the magnitude and Rjb represents the
distance to the surface projection of the rupture.

No.
4. Gujarat, SCR ln(Y)=7.9527+1.4043Mwln(r2jb+19.822)1/20.0682S Mandal

For 3.1 < Mw 7.7 et al. (2009)
std.dev. (): 0.8243
Where Y is peak horizontal acceleration in g,
Mw is moment magnitude,
rjb is the closest distance to the surface
projection of the fault rupture in kilometers,
S is a variable taking the values of 0 and 1
according to the local site geology, S is 0 for
a rock site, and, S is 1 for a soil site.
5. Northeast India, Spectral attenuation relation: Nath et al.

Active intraplate ln(PGA) = C1 + C2M + C3(10 M)3 (2009)
deformation + C4 ln [rrup + C5 exp (C6M)]
+ C7Sv + C8 ln(SR) + C9 ln(SA)
Where PGA is in g,
M is the earthquake moment magnitude,
rrup is the rupture distance (km),
Sv represents the effective shear wave velocity
averaged over the top 30 meters overburden,
SR is the site response, and
SA is the spectral acceleration.
6. Northeast India, log[PSV(T)] = C1(T) + C2(T)M + C3(T)h Das et al.

Shallow crust & + C4(T)log + C5(T)v (2006)
Subduction zone Where M is the earthquake magnitude,
R is the epicentral distance,
h is the focal depth,
v is an index variable taken as zero for
horizontal motion and 1 for vertical motion.
The spectral amplitude PSV(T) at natural
period T, of a single-degree-of freedom

No.
oscillator is defined in the units of m/s. Further,

c1(T), c2(T), c3(T), c4(T), and c5(T) are the
coefficients to be evaluated by the regression
analysis on the available PSV data.
7. Shillong region, Equation for peak ground acceleration aPG Baruah et

Active Intraplate estimated for Shillong plateau is al. (2009)
deformation lgaPG = 0.086Ms 0.547lgR + 0.185 0.18
Where aPG indicates peak ground acceleration,
Ms earthquake magnitude,
R epicentral distance. Predominant period for
the Shillong plateau which is expressed by :
lgt = 0.04Ms + 0.01lg R 1.06 0.09
Where t indicates predominant period,
Ms earthquake magnitude,
R epicentral distance.
8. Indo-Myanmar Arc, log Y = C1 + C2M + C3h + C4R g log R Gupta

Subduction zone + C5slSC + C6slSD + C7slSE (2010)
Where Y represents the random horizontal
component of the peak ground acceleration or
5% damped pseudo-acceleration in cm/sec2,
M is the moment magnitude (limited to 8.5 for
interface and to 8.0 for in-slab events with
larger magnitude),
h the focal depth in km (limited to100 km for
deeper events),
R a distance metric with near source
saturation effects taken into account,
g the geometric attenuation factor,
The last three terms in equation accounts for
the effect of local soil condition defined by the
NEHRP site classes based on the average

No.
shear wave - velocity, VS30, in the top 30 m of

ground. All these terms are taken as zero for
the generic rock type of site (NEHRP class B
defined by VS30 >760 m/s), where as only
one of SC, SD, and SE is taken equal to 1 for
soil classes C, D, and E, respectively.
Further, the scaling factor sl in these terms
accounts for the nonlinear soil behavior, which
is defined as a function of peak ground
acceleration.
9. Entire India log A = -1.02 + 0.249M - log r - 0.00255r + 0.26P Bhatia et al.
r = (d2 + 7.32)1/2 5.0 M 7.7 (1999)
logV = 0.67 + 0.489M logr 0.00256r +
0.17S + 0.22P
r = (d2 + 4.02)1/2 5.3 M 7.4
(by Joyner and Boore, 1981)
Where A is peak horizontal acceleration in
g, V is peak horizontal velocity in cm/sec,
d is the closest distance to the surface
projection of the fault rupture in km,
S takes on the value of zero at rock sites and
one at soil sites, and P is zero for 50 percentile
values and one for 84 percentile values.
10. Gujarat ln Y = C1 + C2 (M 6) + C3 (M 6)2 Petersen et
C4 ln RM (C5 C4) max C6RM al. (2004)
+ e + a
(After Toro et al. 1997)
RM =

No.
In the above equation, M is moment magnitude,

R is horizontal distance, e is epistemic
uncertainty, and a is aleatory uncertainty.
11. Northeast India ln(A) = C2M b ln(X + exp (C3M)) + C4SSR Sharma and
(After Sharma and Bungum, 2006) Malik (2006)
Where M is the magnitude, X is the distance
parameter; b is fixed to be 1.21 and SSR is the
site condition.
ln(PGA)ij=C1*+C2Mi+C3* ln
+ C5Zss + C8Zt + C9Hi + i + ij;

C1* = C1 + C3C4 C3* C4*
C3* = C3 + C6Zs
C4* = C4 + C7Zs
(After Youngs et al., 1997)
Where i is the earthquake index,
J is the recording station index for the ith event,
PGA (in units of g) is the geometrical mean of
the two horizontal components of peak ground
acceleration,
M is the moment magnitude,
rup is source to site distance (in km),
H is the focal depth (in km). The terms H
and Zss are addition to the Youngs et al. (1988)
model.
i Representing earthquake to earthquake
variability of ground motions, and an intra event
component ij representing within earthquake
variability of ground motions.

No.
12. Peninsular India ln(PGA / g) = C1 + C2 (M 6) Jaiswal and

+ C3 (M 6) ln (R) C4R + ln
2
Sinha (2007)
(After Iyengar and Raghukanth, 2004)
Where y, M, R and refer to PGA(g),
moment magnitude, and hypocentral distance
and error associated with the regression
respectively.
ln(PSA) = C1 + C2 (M 6) + C3 (M 6)2
ln (R) C4R
(After Atkinson and Boore, 1995)
ln Y = C1 + C2 (M 6) + C3 (M 6)2
C4 lnRM (C5 C4) max C6RM
+ e + a
(After Toro et al.,1997)
RM =
In the above equation, M is moment magnitude,
R is horizontal distance, e is epistemic
uncertainty, and a is aleatory uncertainty.
13. Northwest log(y) = C1 + C2M + C3r + C4 log(r) + P MonaLisa

Himalays, (After Ambraseys et al.,1996) et al. (2007)
Pakistan Where y is the parameter being predicted, in
this case peak horizontal ground acceleration
(PGA) in g, Ms is the surface wave magnitude,
and
r=
Where d is the shortest distance from the
station to the surface projection of the fault
rupture, in km, and ho is a constant to be
determined with C1, C2, C3 and C4. The standard

No.
deviation of log(y) is 0, and the constant P

takes a value of 0 for mean values and 1 for
84-percentile values of log(y).
14. Bangalore ln(y) = C1 + C2 (M 6) + C3 (M 6)2 Anbazhagan

ln (R) C4R + ln et al. (2009a)
(After Raghukanth, 2005)
Where y, M, R and refer to PGA/spectral
acceleration (g), moment magnitude,
hypocentral distance, and error associated
with the regression, respectively.
15. Eastern coast of log10 (Y) = a + bM + c log10 (r2 + h2)1/2 Kanagarath-

India + e1S1 + e2S2 , inam et al.
(After Sabetta and Pugliese, 1996) (2008)
Where M is magnitude,
R is distance (fault or epicentral) in kilometers,
a is the standard deviation of the logarithm
of Y,
The dummy variables S1 and S2 refer to the
site classification and take the value of 1 for
shallow and deep alluvium sites, respectively,
and zero otherwise,
The parameter h is a fictitious depth determined
by the regression and incorporates all of the
factors that tend to limit the motion near the
source, a property normally referred to as
saturation with distance.
log Y = b1 + b2(M 6) + b3 (M 6)2 + b4r
+ b5 logr + b6GB + b7GC + r + e,
Where r = (d2 + h2)1/2
(After Boore et al., 1993)

No.
In this equation Y is the ground motion

parameter; the predictor variable are magnitude
(M), distance (d, in km), and site classification
(GB = 1 for class B and zero otherwise; GC = 1
for class C and zero otherwise); r is an
independent random variable that takes a
specific value for each record; e is an
independent variable that takes on a specific
value for each earthquake,
b1, b2 b7 and h is the coefficients to be
determined.
log (A) = 1.072 + 0.3903M
1.21 log (X + e0.5873M)
(After Sharma, 1998)
Where A is the peak ground acceleration (g),
M is the magnitude, and X is the hypocentral
distance from the source.
ln y = C1 + C2 (M 6) + C3 (M 6)2
ln (R) C4R + ln
(After Iyengar and Raghukanth, 2004)
Where y, M, R and refer to PGA/spectral
acceleration (g), moment magnitude,
hypocentral distance, and error associated
with the regression, respectively.
16. Afghanistan log(y) = C1 + C2M + C3r + C4 log(r) + P Boyd et al.

(After Ambraseys et al.,1996) (2007)
Where y is the parameter being predicted, in
this case peak horizontal ground acceleration
in g, Ms is the surface wave magnitude, and
r=

No.
Where d is the shortest distance from the

station to the surface projection of the fault
rupture, in km, and ho is a constant to be
determined with C1, C2, C3 and C4.
The standard deviation of log(y) is 0, and the
constant P takes a value of 0 for mean values
and 1 for 84-percentile values of log(y).
ln Sa(g) = f1(M, rrup) + Ff3(M) + HWf4 (M, rrup)
+ Sf5 (pgarock)
(After Abrahamson and Silva 1997)
Where Sa(g) is the spectral acceleration in
g, M is moment magnitude, rrup is the
closest distance to the rupture plane in km, F
is the fault type, HW is the dummy variable for
hanging wall sites and S is the dummy variable
for site class.
log Y = C1 + f1(Mw) + C4 ln +
f3(F) + f4(S) + f 5(HW, F, Mw, rseis) +
(After Cambell and Bozorgnia, 2003)
Where Y is either the vertical component,YV,
or the average horizontal component, YH, of
PGA or 5% damped PSA in g (g : 981 cm/sec2),
MW is moment magnitude,
rseis is the closest distance to seismogenic
rupture in kilometers,
rjb is the closest distance to the surface
projection of fault rupture in kilometers,
Is a random error term with zero mean and
standard deviation equal to ln Y .
log Y = fn(M) + c3h + c4R g log R
+ c5slSC + c6slSD + c7slSE

No.

Where Y = peak ground acceleration or 5%
damped pseudoacceleration (PSA) in cm/sec
random horizontal component,
M = moment magnitude use M 8.5 for interface
events of M > 8.5, M 8.0 for in-slab events of
M 8); fn(M) =c1 + c2 M,
h = focal depth in kilometers.
R= With Dfault being closest
distance to fault surface, in kilometers and a
near source saturation term,
SC = 1(NEHRP C),
SD = 1 (NEHRP D),
SE = 1(NEHRP E).
ln(PGA)ij=C1*+C2Mi+C3* ln
+ C5Zss + C8Zt + C9Hi + i + ij;

C1* = C1 + C3C4 C3* C4*
C3* = C3 + C6Zs
C4* = C4 + C7Zs
(After Youngs et al.,1997)
Where i is the earthquake index,
j is the recording station index for the ith event,
PGA (in units of g) is the geometrical mean of
the two horizontal components of peak ground
acceleration,
M is the moment magnitude,
rrup is source to site distance (in km),
H is the focal depth (in km). The terms H and

No.
Zss are addition to the Youngs et al. (1988)

model,
i representing earthquake to earthquake
variability of ground motions, and an intra event
component ij representing within earthquake
variability of ground motions.
17. Northwest log10 av(g) = 1.15 + 0.245M 1.096 log10 Mahajan

0.256M
Himalaya, (r + e ) + 0.096F 0.0011Er et al. (2010)
India (After Abrahamson and Litechiser, 1989)
Where M is magnitude,
r is the distance in kilometers to the closest
approach of the zone of energy release,
F is a dummy variable that is 1 for reverse or
reverse oblique events and 0 otherwise, and
E is a dummy variable that is 1 for interpolate
events and 0 for intraplate events. The standard
error of log10 av is 0.296.
18. Tamil Nadu ln Sa(g) = f1(M, rrup) + Ff3(M) + HWf4(M, rrup) Menon et al.
+ Sf5 (pgarock) (2010)
(After Abrahamson and Silva, 1997)
Where Sa(g) is the spectral acceleration in g,
rrup is the closest distance to the rupture
plane in km,
F is the fault type,
HW is the dummy variable for hanging wall
sites and S is the dummy variable for site
class.
ln(ybr)=c1 + c2(M 6) + c3(m 6)2 ln(r) c4r + ln( br)
(After Raghukanth and Iyengar, 2007)

No.
Where ybr = (Sa/g) stands for the ratio of

spectral acceleration at bedrock level to
acceleration due to gravity,
M and r refers to moment magnitude and
hypocentral distance respectively,
ln( br) is the error term.
19. West Bengal log Y = C1 + f1(Mw) + C4 ln Nath et al.

+ f3(F) + f4(S) (2011d)
(Modified Cambell and Bozorgnia, 2003)
Where Y is either the vertical component,
YV, or the average horizontal component,
YH, of PGA or 5% damped PSA in g
(g 981 cm/sec2),
MW is moment magnitude,
rseis is the closest distance to seismogenic
rupture in kilometers.
log PSA = C1 + C2M + C2M2 + (C4 + C5M) f1 +
(C6 + C7M) f2 + (C8 + C9M)f0 + C10Rcd
Where
f0 = max(log(R0 / Rcd), 0);
f1 = min(log Rcd, R1);
f2 = max(log(Rcd /R2),0); R0 = 10, R1 = 70, R2 = 140.
M is the Earthquake magnitude,
Rcd is the closed distance to the faults.
27.2 Next Generation Attenuation Models

The Next Generation Attenuation (NGA) equations were developed for shallow crustal
earthquakes in active tectonic region of Western United States. However, these new equations
have been purported to be valid in other regions of similar tectonics. The models aimed at
the prediction of different ground motion parameters namely, Peak Ground Acceleration (PGA),
Peak Ground Velocity (PGV), Peak Ground Displacement (PGD), and 5% damped elastic
Pseudo-absolute response Spectral Acceleration (PSA) for 0-10 sec. Abrahamson et al.
(2008) compared the five NGA equations developed by Abrahamson and Silva (2008), Boore
and Atkinson (2008), Campbell and Bozorgnia (2008), Chiou and Youngs (2008), and Idriss
(2008) on the datasets used, adopted model parameters and constraints, and the predictions
of the ground motions. The difference was pointed out in use/discard of aftershocks data,
incorporation of rupture characteristics (i.e., buried or not), modeling of soil effects, treatment
of the magnitude dependence and nonlinear site response. Owing to the data employed,
NGA models have been normally considered valid for shallow continental earthquakes in
shallow active tectonic regimes with magnitude MW > 4.0, MW 8.5 for strike-slip faulting, MW
8.0 for reverse faulting, and MW 7.5-8.0 for normal faulting. Applicability of NGA has been
tested and found reasonably suitable in several regions across the globe, e.g. ShojaTaheri
et al. (2010) in the Iranian Plateau, Stafford et al. (2008) in the Euro-Mediterranean Region,
and Campbell and Bozorgnia (2006) in Europe. Scasserra et al. (2008) reported the necessity
of minor adjustments of the models for Italy. Akkar and Bommer (2007) also observed that
the systematic differences between the NGA models and those from Europe to be insignificant.
Conformity of the ground motion characteristics across tectonically active regions of shallow
crustal seismicity has been suggested.
28. SEISMIC HAZARD ANALYSIS

The seismic hazard analysis is concerned with getting an estimate of the strong-motion
parameters at a site for the purpose of earthquake resistant design or seismic safety
assessment. For generalized applications, seismic hazard analyses can also be used to
prepare macro or micro-zoning maps of an area by estimating the strong-motion parameters
for a closely spaced grid of sites. Two basic methodologies used for the purpose are the
deterministic and the probabilistic seismic hazard analysis approaches. In the
Deterministic Seismic Hazard Analysis (DSHA), the strong-motion parameters are estimated
for the maximum potential earthquake, assumed to occur at the closest possible distance
from the site of interest, without considering the likelihood of its occurrence during a specified
exposure period. On the other hand, the Probabilistic Seismic Hazard Analysis (PSHA)
integrates the effects of all the earthquakes expected to occur at different locations during a
specified life period, with the associated uncertainties and randomness taken into account.
28.1 Conventional Deterministic Seismic Hazard Analysis

In the early years of geotechnical earthquake engineering, the use of Deterministic Seismic
Hazard Analysis (DSHA) was prevalent. A DSHA involves the development of a particular
seismic scenario upon which a ground motion hazard evaluation is based. The scenario consists
of the postulated occurrence of an earthquake of a specified size occurring at a specified
location. A typical DSHA can be described as a four-step process consisting of:
1. Identification and characterization of all earthquake courses capable of producing significant

ground motion at the site. Source characterization includes definition of earth sources
geometry (the source zone) and earthquake potential.
2. Selection of a source-to-site distance parameter for each source zone. In most DSHAs,
the shortest distance between the source zone and the site of interest is selected. The
distance may be expressed as an epicentral distance or hypocentral distance, depending
on the measure of distance of the predictive relationship(s) used in the following step.
3. Selection of the controlling earthquake (i.e., the earthquake that is expected to produce
the strongest level of shaking), generally expressed in terms of some ground motion
parameter, at the site. The selection is made by comparing the levels of shaking produced
by earthquakes (identified in step 1) assumed to occur at the distances identified in step
2. The controlling earthquake is described in terms of its size (usually expressed as
magnitude) and distance from the site.
4. The hazard at the site is formally defined, usually in terms of the ground motions produced
at the site by the controlling earthquake. Its characteristics are usually described by
one or more ground motion parameters obtained from predictive relationships of the
types presented in the previous sections. Peak acceleration, peak velocity, and response
spectrum ordinates are commonly used to characterize the seismic hazard.
The DSHA procedure is shown schematically in Figure 26. Expressed in these four
compact steps, DSHA appears to be a very simple procedure, and in many respects it is.
When applied to structures for which failure could have catastrophic consequences, such as
nuclear power plants and large dams, DSHA provides a straight forward framework for the
evaluation of worst-case ground motions. However, it provides no information on the likelihood
of occurrence of the controlling earthquake, the likelihood of it occurring where it is assumed
to occur, the level of shaking that might be expected during a finite period of time (such as
the useful lifetime of a particular structure or facility), or the effects of uncertainties in the
various steps required to compute the resulting ground motion characteristics.
Figure 26: Four steps of a deterministic seismic hazard analysis.
28.2 Probabilistic Seismic Hazard Analysis (PSHA)

PSHA rectifies a number of the problems inherent in DSHA by quantifying uncertainty and
the probability of earthquake occurrence. As noted by Kramer (1996), PSHA follows similar
steps to DSHA but uncertainty is quantified by a probability distribution at every step in the
process. Probability distributions are determined for the magnitude of each earthquake on
each source, the location of the earthquake in or along each source, and the prediction of
the response parameter of interest. Kramer (1996) describes PSHA as a four-step process
as enumerated and depicted in Figure 27.
Identify and characterize (geometry and potential) all earthquake sources capable of
generating significant shaking at the site. Three sources surrounding the site are shown
in Figure 27. For each source, develop the probability distribution of rupture locations
within the source. (A uniform probability distribution is generally chosen, which means
that earthquakes are equally likely of occurring at any point along or in the source).
Combine these distributions with the source geometry to obtain the probability distribution
of source-to-site distance. (Contrast this with DSHA that assumes that the probability
of occurrence is 1 at the points in each source zone closest to the site and 0 elsewhere).
Develop a seismicity or temporal distribution of earthquake occurrence. A recurrence
relationship, which specifies the average rate at which an earthquake of some size will
be exceeded, is used to characterize the seismicity of each source zone. (The recurrence
relationship may accommodate the maximum earthquake but is not limited to that
earthquake, as DSHA often does).
The ground motion produced at the site by earthquakes of any possible size (magnitude)
occurring at any possible point in each source zone must be determined with the use of
predictive (attenuation) relationships. (The uncertainty inherent in the attenuation
relationship is also considered explicitly in PSHA unlike DSHA).
The uncertainties in earthquake location, size, and ground motion prediction are combined
to obtain the probability that the ground motion parameter (e.g., PHA, spectral
acceleration) will be exceeded in a particular time period (say 10% in 50 years).
Figure 27: Four steps of a Probabilistic Seismic Hazard Analysis.

28.3 Establishment of Seismic Occurrence Scenarios

The establishment of feasible scenarios require the study of past seismic occurrences, the
definition of different seismogenic zones affecting the area under study, and the
characterization of their most important parameters (maximum expected magnitude and
frequency or probability). It may also involve the knowledge of attenuation functions, if
scenarios have to do anything with ground motion parameters, or the minimization of any
other objective function, as total losses inflicted within a given time period.
In many applications not a single criterion prevails. Hypotheses to be analysed are :

the consideration of a largest historical event, as a measure of an extreme type event; the
50, 100 or 1000 year mean return period event, requiring an hazard and deaggregation
analysis; the most probable measure of impact over the entire stock in the study area; or
the eventual rupture of a possible fault structure.
All the seismological and geological themes have to be generated using the Geographical
Information System (GIS), a tool for computer based data and storage and manipulation
that can link geological, seismological and geotechnical data with information on urban
development for studies of the impact of hypothetical/historical/maximum credible/scenario
earthquakes on human activities. The GIS facilitates numerous repetitive calculations required
to produce the individual seismic hazard map that the overall relative hazard map is based
on, as well as linking the seismic hazard data with information on buildings and infrastructure
of the damage and loss assessment.
A typical Probabilistic Seismic Hazard Assessment of India (Nath and Thingbaijam,

2011b) is given in AppendixIII.
29. COMPOSITE HAZARD EVALUATION
The evaluation of damages of the selected elements in risk, for each scenario, is obtained
using a simulation model that integrates all the above mentioned aspects, by summing up
all possible contributions. The damages of each element in risk are classified in several
different limit states, usually defined at five levels: no damage, slight damage, moderate
damage, heavy damage, and collapse. Depending on the element under study, this damage
classification may be lumped into coarser categories or adapted to operational/ non-operational
terms, as in the case of sections of lifelines etc.
Scenario studies are viewed in different perspectives according to the objectives to be

targeted. These define the scale of intervention which is very critical to obtain the elements
necessary for the analysis.
A holistic multiple hazard considerations to deliver a decision support tool for landuse
planning and developmental projects are :
The composite hazard assessment incorporates multiple attributes through multi-criteria

evaluation technique for the spatial delineation.
Fuzzy sets enabled scheme for the representation and manipulation of uncertainty
related to the classification of individual locations according to their attribute values can
be aided by the Analytic Hierarchy Process (AHP) - a mathematical method introduced
by Saaty (1980) to determine priority of criteria in the decision making process (Nath,
2005).
AHP uses hierarchical structures to represent a problem and then develop priorities for
the alternatives based on the judgment of the experts. Pair-wise comparisons are
employed to form judgments between two particular elements rather than attempting to
prioritize an entire list of elements.
The process of allocating weights is a subjective one and can be done in the participatory
mode in which a group of decision makers may be encouraged to reach a consensus of
opinions about the relative importance of factors.
The values within each thematic map/layer varying significantly are classified into various
ranges or types, which are referred to as the features of a layer. These features are then
assigned ranks or scores within each layer, normalized to ensure that no layer exerts
an influence beyond its determined weight.
30. EARTHQUAKE HAZARD PARAMETERS

Seismic microzonation is subdividing a region into smaller areas having a different potential
for hazardous earthquake effects. The earthquake effects depend on ground geomorphological
attributes consisting of geological, geomorphological and geotechnical information. The
parameters of geology and geomorphology, soil coverage/thickness, and rock outcrop/depth
are some of the important geomorphological attributes. Other attributes are the earthquake
parameters, which are estimated by hazard analysis and effects of local soil contributing to
hazard (local site response for an earthquake). The Peak Ground Acceleration (PGA) [from
deterministic or probabilistic approach], amplification/ site response, predominant frequency,
liquefaction and landslide due to earthquakes are some of the important seismological
attributes. Weight of the attributes depends on the region and decision maker, for example
flat terrain has weight of 0 value for landslide and deep soil terrain has highest weight for
site response or liquefaction.
In the following sections, the Geomorphological, Geotechnical and Seismological

themes (as listed from i to ix below) considered for the microzonation of the Greater Guwahati
City are outlined, for a better understanding (Nath, 2007; Nath et al., 2008b; Nath and
Thingbaijam, 2009):
i) Geology and Geomorphology (Base map)

ii) Effective Shear-wave Velocity (calculated from Lithologs/ SPT blow counts/ MASW/
SASW data)
iii) Liquefaction Potential (Liquefaction Susceptibilty / Factor of Safety)
iv) Landuse Map
v) Basement configuration and Thickness of Valley Fill
vi) Landslide Hazard Zonation

vii) Site Response (considered at Predominant Frequency)
viii) Predominant Frequency
ix) Peak Ground Acceleration (PGA calculated using F-K integration).
30.1 Landuse
For the preparation of Landuse map of Guwahati City, the 1990 landuse map prepared by the
Assam State Remote Sensing Application Center, Guwahati was used as primary guide to
carry out the baseline classification and ground checking to arrive at an appropriate landuse.
Satellite images were used for finer classification of different landuse classes. A detailed
road to road GPS based survey was conducted to arrive at cultural parameters such as
residential, commercial and institutional areas in order to further sub-classify the broad
habitation areas shown in greenish hue on the image. Ground checking was also conducted
to collect data on other landuse classes such as water bodies, swamps, and agricultural
field. A survey of the industrial areas was also conducted and shown in Figure 28.
The landuse classes that have been derived are Residential areas, Commercial areas,
Industrial areas, Public Utility areas, Educational areas, Residential areas located in a hill,
Army/Police reserve, Airport, Field/Open areas, Agricultural land, Hills with Dense Forest,
Hills with Light Forest, River, Sandbars, River Island, Waterbody/Bell and Swampy areas.
Figure 28: Landuse map of Guwahati Region.
30.2 Geology and Geomorphology (BASE MAP)

In order to facilitate compilation and collation of data and maps from varied sources on a
single platform, the first task was to bring out a Base map of the area surrounding Guwahati
where microzonation was contemplated. For this purpose, Department of Science &
Technology, Govt. of India took up the matter with the Survey of India, which provided the
1:25,000 scale topographic sheets. The required information, coverage wise, such as roads,
contour, railways, streams and certain administrative boundaries were traced on a
transparency and digitized using onscreen digitization techniques. Further, IRS PAN &
LISS III satellite images were acquired in digital format for the Greater Guwahati Region.
The images were corrected and georeferenced with respect to the topographic map points
and GPS readings taken of the ground control points. Thereafter, the LISS and PAN scenes
were merged to produce a sharpened high resolution FCC image having a ground resolution
of 5.8 m. The stream network, road network, river and water bodies were, thereafter updated
using the satellite image. Further, an extensive GPS based point survey was done in the
greater Guwahati to capture all possible road networks, important places, and other requisite
features on the ground. Based on the above exercise, a comprehensive map of Guwahati
was prepared. This map has been used all throughout the microzonation exercise as a
landmark study. All other maps, data and GPS point surveys were collated and corrected
with respect to this base map, which made it possible to weed out errors in various data
sets for Guwahatmi Seismic Microzonation.
Broadly the Guwahati Region consists of two main geological formations, viz.
a) Precambrian granitic rocks forming the hill tracts and isolated hillocks, and
b) Quaternary alluvium occupying the valleys, deposited over the uneven eroded and faulted
basement of granitic rocks.
Figure 29: Geology and Geomorphology map of Guwahati Region (Base Map).
The granite and granite gneissic rocks are well foliated and jointed, allowing deep
weathering along the joint and fault planes, and covered in most places by 1 to 3 m thick
ferruginous soil capping.
The Quaternary alluvium that probably forms the flood plain deposit of the Brahmaputra
River could be classified into 5 aggradational units based on lithological characters, state of
weathering, order of superposition and unconformity between them. As shown in Figure 29 they
are, in order of increasing antiquity.
1. Active flood plain and levee deposit (AFP)

2. Digaru Surface (T1)
3. Bordang Surface (T2)
4. Sonapur Surface (T3) and
5. Pediment Sruface (PD).
30.3 Geomorphology / Terrain Configuration
The overall topography of the area is rugged with high relief due to the presence of steep
sided hillocks carved out of the Meghalaya Plateau that occupies the southern and the
eastern fringe of the Guwahati area. The denuded and continuous hill tracts from Rani to
Khanapara RF in the south and the Amchang Hills RF in the east rise in altitude from 200 m
to 400 m above MSL. Isolated hillocks within the valley occasionally rise up to 300 m above
MSL. The highest hillock is located in the north-western part of the area (Silapahar - 381 m)
and the lowest elevated hillock, Odalbakra 145 m, lies in the center of the area. The
relative relief is high varying between 80 m and 300 m. The general elevation of the valley
area varies from 25 m to 50 m above MSL.
The major landform units can be mapped as
a) Denuded hills
b) Valley filled alluvium with almost flat surface, and
c) Swampy landmass/bels/water bodies.
Figure 30: Basement zonation map of Guwahati Region.
30.4 Basement Configuration and Thickness of Valley Fill

The basement zonation map shown in Figure 30 forms N10oW-S10oE trending steep V
shaped valley with maximum depth of 150 m in between the Fatasil and Kalapahar hillocks
that is traversed by a fault. Another NE-SW trending valley occurs in between the engineering
College hill and the Fatasil hill that extend up to the junction of Neelachal hill and the fatasil
hill with a maximum depth of 250 m. This valley is also underlain by a fault. A NE-SW
trending valley passing along the Tepar beel by the western side of the Japorigong hills
reaches the depth of 100 m and it is relatively wide. An E-W deep valley passes from
Panjabri to Kahabari across the Deepar bell. Toward east the basin has highs and lows but
to the west the basin gradually gets deeper and attains a depth of 300 m. A fault also runs
along this valley. Thus it is seen the valley area has a variable depth to basement, having a
steep gradient along some zones like the western, eastern and the southern margins of the
Fatasil hill; western margin of the Kalapahar hill and the northern periphery of the Rani-
Khanapara hill tract. These zones will have pronounced basin edge effect during earthquake
shaking.
30.5 Seismotectonics
The city of Guwahati is located in an area surrounded on all sides by highly active tectonic
blocks. Generally speaking the area is buttressed in between the Himalayan collision zone
to the north and the northeast, Indo-Myanmar subduction interface of Indian plate to the
east and the Meghalaya Plateau Mikir hills tectonic block to its south. Strictly speaking
the Guwahati area falls in the domain of Meghalaya Plateau and Mikir hill block. Juxtaposition
of ongoing collision-subduction tectonic processes has made the area one of the most
intense seismic zones of the world.
Analysis of contemporary tectonics in the region reveals that active Himalayan frontal
thrusts and cross faults cutting across these thrusts have been generating many relatively
shallow, small and moderate earthquakes. Strike-slip movements along a NW-SE trending
fault (Po Chu) in the Mishmi block had produced the 1950 Great Assam earthquake (Mw 8.6)
inflicting catastrophic damage in the Upper Assam area. All the other thrusts/faults in this
block viz. Mishmi thrust, Lohit thrust and Tidding suture may be classified as capable
faults, as shown in Figure 31.
The active subduction process along the Indo-Myanmar mobile belt and the conjugate
faults lying across this belt have been producing many large and major earthquakes that
shake the Guwahati area.
It may be mentioned here that recent release of stress along a 1200 km long subduction
interface in the southern part of the Indo-Maynmar-Andaman-Sunda subduction zone by
the 26th December, 2004, 9.1 magnitude earthquake has made it highly probable that next
rupture may take place along the northern sector of the subduction zone in Indo-Myanmar
Region.
30.6 Landslide Hazard Zonation

Detailed work by Keefer (1984) on 40 historical earthquakes and numerous landslides induced
by them revealed that most of the slides get triggered at intensity VI and above (MM Scale)
with very few slides occurring at lower intensity zones. The thickness of weathered zone/
soil is fairly high in the granite hillocks in and around the Guwahati city. Many of the
landslide incidences (baring a few in the Kalapahar area) are due to anthropogenic activities.
A total of 6 landslides of different categories have been recorded. Most of them are slump
type followed by debris slide and rock fall.
Figure 31: Seismotectonic map of the Guwahati Region.

Twenty two landslides are concentrated in and around the Guwahati city, especially in
its central part. Kalapahar area experienced 10 slides followed by Dhirenpara having 4
slides. The area along the G-S road has the record of 9 landslides. Based on geo-
environmental parameters like slope angle, lithology, structure, relative relief, landuse,
landcover, hydrological correlation, seismicity, rainfall and landslide incidences were
considered for preparation of landslide hazard zonation map. Eight thematic maps were
first prepared viz., facet map, slope morphometry map, relative relief map, lithological map,
structural map, landuse, landcover map, drainage map, landslide incidence map, on 1:50,000
scales and then enlarged to 1:25,000 scales. The landslide hazard zonation map has been
prepared according to total estimated hazard (TEH) of each face by superimposing the
slope facet map successively one by one over all the thematic maps. The TEH of the facet
is calculated after adding the values of landslide hazard evaluation factor (LHEF) of all 9
geo-environmental parameters encompassing the particular facet. The derived landslide
hazard zonation map thus prepared showed only three categories of hazard zone such as
Low Category of Hazard Zone (LHZ), Moderate Hazard Zone (MHZ) and High Hazard Zone
(HHZ) as shown in Figure 32.
Figure 32: Landslide Hazard Zonation map of Guwahati Region.

30.7 Shear Wave Velocity (Vs30) Map

Shear Wave Velocity Distribution Map of Guwahati Region is shown in Figure 33.
Figure 33: Shear Wave Velocity Distribution Map of Guwahati Region.
30.8 Predominant Frequency Map

Predominant Frequency Distribution Map of Guwahati Region is given in Figure 34.
30.9 Site Response Map

The Site Response is a seismological theme. Site Response Distribution Map is presented
in Figure 35.
30.10 Factor of Safety Map

For simplicity, a component subjected to a solitary load will be considered in the first
instance. This load is interpreted in the context of the components nature and duty - thus
load usually implies a transverse force in the case of a beam component, or a longitudinal
compressive force in a column, or a torque in the case of a shaft, or a pressure in a fluid
containment vessel, and so on.
Figure 34: Predominant Frequency Distribution Map of Guwahati Region.
Figure 35: Site Response Distribution Map of Guwahati Region.

There are two completely different manifestations of the load, which have important
consequences for the component:
The extrinsic actual load is the load exerted on the component by its surrounds
The intrinsic maximum load is the largest load that the component can withstand without
failure; the maximum load is a property of the component, a function of its dimensions
and material.
Clearly a component is safe only if the actual load applied to the component does not
exceed the components inherent maximum sustainable load. The degree of safety is usually
expressed by the safety factor FS
FS = maximum load / actual load = Fmax / F and it follows that :

if FS = 1 then the component is on the point of failure
if FS < 1 then the component is in a failed state
if FS > 1 then the component is safe.
Factor of Safety distribution map of Guwahati Region is shown in Figure 36.
Figure 36: Factor of Safety Zonation Map of Guwahati Region. Accordingly Guwahati is clas-
sified in two zones namely safe and unsafe.
30.11 Site Classification Map

Detailed Site Classification of Guwahati City is shown in Figure 37.
Figure 37: Site Classification Map of Guwahati Region.
30.12 Peak Ground Acceleration (PGA) Map
The PGA at various locales in Guwahati Region is estimated using two methods, but for
microzonation purposes, the PGA estimated by F-K integration is used and shown in Figure 38.
31. GIS INTEGRATION LOGIC

The representation and interpretation of uncertainty related to the classification of individual
locations is provided by the fuzzy logic based on location attribute values. Fuzzy logic
implements classes or groupings of data with boundaries. The central idea of fuzzy sets is
aided by the Analytic Hierarchy Process (AHP). AHP is a multi-criteria decision method that
uses hierarchical structures to represent a problem and then develop priorities for the alternatives
based on the judgment of the user (Saaty, 1980). The idea of multi-criteria decision-making
was based on the concept of McHarg (1968). McHarg (1968) introduced a systematic landuse
planning by using the concept of compatibility of multiple landuses. He mentioned that the
factors affecting land and its relative values are different and, therefore, it is difficult to think of
optimizing them for a single use. It can be optimized for multiple compatible uses. He introduced
a simple matrix system for determining the degree of compatibility.
Figure 38: Peak Ground Acceleration (compute by F-K integration) map of Guwahati Region.
Saaty (1968) has shown that weighting activities in multi-criteria decision-making can be
effectively dealt with hierarchical structuring and pair-wise comparisons. Pair-wise comparisons
are based on forming judgments between two particular elements rather than attempting to
prioritize an entire list of elements (Saaty, 1980). For multi-criteria evaluation, Saatys Analytical
Hierarchy Process (AHP) is used to determine the weights of each individual criterion (Saaty,
1990). AHP is a mathematical method to determine priority of criteria in the decision making
process. It is a popular tool used by decision makers in the multi-attribute decisions.
Saatys Analytical Hierarchy process constructs a matrix of pair-wise comparisons

(ratios) between the factors of Earthquake Hazard Parameters (EHP). The constructed matrix
shows the relative importance of the EHP based on their weights. If 9 earthquake hazard
parameters are scaled from 1 to 9, 1 meaning that the two factors are equally important, and
9 indicating that one factor is more important than the other, the reciprocals of 1 to 9 (i.e., 1/
1 to 1/9) show that one is less important than the others. The allocation of weights for the
identical EHP depends on the relative importance of factors and participatory group of decision
makers. Then the individual normalized weights of each EHP are derived from the matrix
developed by pair-wise comparisons between the factors of EHP. This operation is performed
by calculating the principal Eigen vector of the matrix. The results are in the range of 0 to 1
and their sum adds up to 1 in each column. The weights for each attribute can be calculated
by averaging the values in each row of the matrix. These weights will also sum to 1 and can
be used in deriving the weighted sums of rating or scores for each region of cells or polygon
of the mapped layers (Jones, 1997).
Since EHPs vary significantly and depend on several factors, they need to be classified into
various ranges or types, which are known as the features of a layer. Hence each EHP features
are rated or scored within EHP and then this rate is normalized to ensure that no layer exerts an
influence beyond its determined weight. Therefore, a raw rating for each feature of EHP is allocated
initially on a standard scale such as 1 to 10 and then normalized using the relation,
(79)
Where, Ri is the rating assigned for features with single EHP, Rmin and Rmax are the minimum and
maximum rates of a particular EHP.
For seismic microzonation and hazard delineation of the above themes of Greater Guwahati
Region, both the geomophological and seismological themes are reclassified into a 1st phase:
geohazard map and 2nd phase: seismic microzonation map with PGA distribution for a Scenario
Earthquake Magnitude (SEM) of Mw 8.7. A typical two phase microzonation procedure follows
hazard zonation to regional hazard zonation, mapping on a GIS platform and finally to seismic
microzonation, as shown in the detailed roadmap given in Figures 3, 4 and 5.
The geological vector layers that have been used for microzonation are Geology and
Geomorphology (GG), Basement (BS), Landslide hazard zones (LS) and Landuse (LU), where as
the seismological themes are Shear wave velocity (Vs30), Site Response (SR), Peak Ground
Acceleration (PGA), Predominant Frequency (PF) and Factor of Safety (FS).
The geological and seismological themes are in weight scale of 9:1 depending on their
contribution to seismic hazard, the highest being attached to shear Geological &
Geomorphological layer with a normalized weight of 0.2000. Basement has got the next
weightage with a normalized value of 0.1778; Landslide hazard and Landuse have got the
next weightage with normalized value of 0.1556 and 0.1333 where as the shear wave velocity
(Vs30), Peak Ground Acceleration (PGA), site response, predominant frequency and factor
of safety are assigned the values 0.1111, 0.0889, 0.0667, 0.0444 & 0.0222 respectively.
In this method, a matrix of pair-wise comparisons (ratio) between the factors is built, which
is used to derive the individual normalized weights of each factor. The pair-wise comparison is
performed by calculating the principal eigen vector of the matrix and the elements of the matrix are
in the range of 0 to 1 summing to 1 in each column. The weights for each theme can be calculated
by averaging the values in each row of the matrix. These weights will also sum to 1 and can be
used in deriving the weighted sum of rating or scores of each region of cells or polygon of the
mapped layers. Since the values within each thematic map/layer vary significantly, those are
classified into various ranges or types known as the features of a layer. These features are then
assigned ratings (r) or scores within each layer, normalized to 0-1 as shown in Table 15 below.
Table 15: Normalized feature ratings of the thematic maps
Theme Weight Feature Rating Normalized

Rating
Geology 0.2000 River, Water Bodies & Swampy area 8 1.0000
(GG) Active Flood Plain 7 0.8571
Natural Levee 6 0.7143
Pediment 5 0.5714
Sonapur Surface 4 0.4286
Digaru Surface 3 0.2857
Bordang Surface 2 0.1429
Denuded Hills 1 0.0000
Basement 0.1778 >600 7 1.0000
(BS) 500 - 600 6 0.8333
400 - 500 5 0.6667
300 - 400 4 0.5000
200 - 300 3 0.3333

Rating
100 - 200 2 0.1667
<50 -100 1 0.0000
Landslide 0.1556 High Hazard Zone 4 1.0000
Hazard Medium Hazard Zone 3 0.6667
(LS) Low Hazard Zone 2 0.3333
River, Sand Bar & Hazard 1 0.0000
Free Zone
Landuse 0.1333 Residential area 7 1.0000
(LU) Educational, Army/Police Reserve, 6 0.8333
Commercial area
Sandbars, River Island & 5 0.6667
Swampy area
Field/open space & 4 0.5000
Agricultural Area
Residential Areas in Hill 3 0.3333
Hill with dense & light forest 2 0.1667
River, water bodies/Beel 1 0.0000
Shear 0.1111 200 - 240 4 1.0000
Wave 240 - 280 3 0.6667
Velocity 280 - 320 2 0.3333
(Vs30 ) 320 - 360 1 0.0000
Peak 0.0889 >= 0.75 6 1.0000
Ground 0.60 - 0.75 5 0.8000
Acceleration 0.45 - 0.60 4 0.6000
(PGA) 0.30 - 0.45 3 0.4000
0.15 - 0.30 2 0.2000
< 0.15 1 0.0000
Site 0.0667 >= 5.5 5 1.0000
Response 4.5 - 5.5 4 0.7500
(SR) 3.0 - 4.5 3 0.5000

Rating
1.5 - 3.0 2 0.2500
<1.5 1 0.0000
Predominant 0.0444 <0.5 8 1.0000
Frequency 0.5 - 1.0 7 0.8571
(PF) 1.0 - 2.0 6 0.7143
2.0 - 3.0 5 0.5714
3.0 - 4.0 4 0.4286
4.0 - 5.0 3 0.2857
5.0 - 7.0 2 0.1429
> 7.0 1 0.0000
Factor of 0.0222 1 1 1
Safety (FS) >1 0 0
The probabilistic seismic hazard zonation map is obtained through the integration of all
the above themes using the following relation.
PSHI= GGwGGr + BwBr + LSwLSr + LUwLUr + Vs30w Vs30r + PGAwPGAr + SRwSRr + PFwPFr + FSwFSr/w
(80)
The notations have their usual meanings. The integration scheme is shown in Figure
39. Four zones are mapped as shown in Figure 40 where the average PSHI indices are
>0.55, 0.44, 0.35, 0.26 and <0.16. These zones are termed as very high, high, moderate,
low and very low hazard regions. Figure 40 shows the microzonation map of Guwahati Region
overlain by PGA computed by Greens Function approximation.
32. SEISMIC VULNERABILITY AND RISK ASSESSMENT

The exposures of the vulnerability components such as human population, buildings, etc.
to the seismic hazard characterize seismic risk of a region. The seismic hazard is generally
assumed to be stable over a long geological time while the typical vulnerability (and therefore,
the risk) to the hazard changes (McGuire, 2004). In a simple formulation, the risk is assessed
as a convolution function of the hazard and the vulnerability:
Risk = Hazard * Vulnerability (81)
The risk appraisals, aimed at promoting reasonable hazard mitigation regulations, are
generally based on vulnerability aspects such as landuse, demographic distributions, building
typology, etc. The computation of risk is fundamentally influenced by that of the hazard.
Likewise, seismic risk assessment could be deterministic or probabilistic.
Figure 39 : The seismic microzonation scheme for Guwahati city with the weights assigned
to each theme labeled accordingly leading to the final hazard map. A representative
accelerogram simulated at a borehole location for the projected maximum
earthquake of Mw 8.3 nucleating from the Shillong seismic zone is also depicted
(after Nath and Thingbaijam, 2009).
The former involves direct assessment of possible losses based on the results of
deterministic hazard analysis with no involvement of reference time period but yielding to
the current status. The assessment could, otherwise, follow either mean values or take into
account the uncertainties related to frequency of event occurrences (hazard) and damage
levels (vulnerability) yielding to a probabilistic account of the expected losses (Giacomo et
al., 2005). These approaches allow estimation of risk on a reference period of time. Another
approach is to generate the probable damage scenario by random simulations based on
post earthquake damage studies (Barbat et al., 1996).
FIGURE 40 : Seismic Microzonation Map of Guwahati Regions (using PGA computed by F-K
integration).
33. CONCLUDING REMARKS

The seismic microzonation has emerged as an important issue in high risk urban centers
across the globe. The compilation of data pertaining to geological, geophysical, geotechnical
and seismological aspects comprises a major part of the venture, which necessitates a
consortium of several public and private organizations engaged in diversified but related
domains. The effort towards enhancing our understanding of the seismic hazard and related
effects is an on-going process, and therefore, the framework and tools for the seismic
microzonation studies presented here needs to be continuously updated in the light of
ongoing advancements as well as experiences gained during earthquakes. It is expected
that seismic microzonation will enable updating building codes as well as formulate actions
for hazard mitigation at subregional and local levels. Active programs related to infrastructural
improvements and response planning can lead to reduction of seismic risk.
Structural mitigation measures are the key to making a significant impact towards
earthquake safety in our country. For successful earthquake mitigation, it has to be ensured
that all new constructions in the seismic zones are complaint with the BIS Codes and for
this purpose a techno legal regime has to be put in place. Though Bureau of Indian Standards
(BIS) has laid down the national standards for construction in seismically vulnerable areas,
these are not mandatory in nature. In many States, building bylaws are non-existent, and
even in states where there are by-laws, which have considerations for seismic safety, the
enforcement mechanisms leave a lot to be desired.
WORK SHEET FOR SEISMIC MICROZONATION
1. Geological & Geomorphological Characterization:

(i) Compile a Geological map that primarily indicates the local geological and
geomorphological units including the lithological characteristics and bedrock
configuration of the study region and its surrounding areas based on the maps prepared
by the Geological Survey of India (GSI). These maps are available on 1:2000,000 scale
on GSI portal (www.portal.gsi.gov.in). Also GSI has completed geological mapping at
1:50,000 scale and these maps will be made available on GSI portal shortly.
(ii) Identify and characterize major/minor faults, lineaments, and seismotectonic units that are
seismically active in the region using the Seismotectonic Atlas of GSI (Dasgupta et al.,
2000), digital version being available at http://www.portal.gsi.gov.in/, or from published
literatures.
(iii) The crustal structure for a given region to be used for ground motion simulation at a
later stage (exclusively for deterministic seismic hazard assessment) of the
microzonation protocol may be evaluated based on the active/passive seismological
results from published literature. If no such information for the study region is available,
the same may be adopted from a similar tectonic province in India.
2. Seismic Source Characterization:

Characterize the seismic source zones in the vicinity of the study region (300 km radius for
stable continental regions and 1000 km radius in seismically active regions) through:
(i) Preparation and Homogenization of the Earthquake Catalog. The catalog of the
earthquakes compiled by Nath et al. (2010) as highlighted in APPENDIXI may be
used. The Homogenous Earthquake Catalogue of South Asia in Mw scale can be
downloaded from http://earthqhaz.net/sacat/ as: Early instrumental and instrumental
period (1900-2008) [.txt format, 3.6 MB], Pre-instrumental period (Prior to 1900) [.txt
format, 18.0 KB] & readme file [.pdf format, 20.0 KB]. The catalog has to be updated
by including earthquakes that occurred subsequently. Based on the latitude and
longitude coverage of the study region, the requisite portion of the catalog may be
extracted for further use.
(ii) Delineation of seismic source zones by superimposing the seismicity on the faults/
lineaments obtained at Step No. 1(ii). If no obvious linear source zones can be identified,
areal clustering of seismicity may be considered as a source region. The seismic
source zones identified by Thingbaijam and Nath (2011) and given in APPENDIXII may
be used as ready reference.
(iii) In seismically passive regions with known faults and lineaments, the empirical relations
given by Wells and Coppersmith (1994) are to be used for the estimation of Maximum
Credible Earthquake (MCE) for each tectonic fabric for the assessment of Deterministic
Seismic Hazard.
(iv) In regions devoid of seismicity and known tectonic features, the Peak Ground
Acceleration (PGA) values at engineering bed rock (NEHRP Site Class-B) applicable
for the study region should be adopted from the Probabilistic Seismic Hazard Assessment
of India given by Nath and Thingbaijam (2011b) in APPENDIX III.
(v) Estimation of a and b values for each of the source zones using the maximum likelihood
approach. These values for 172 source zones in four hypocentral depth ranges: upper
crust (0-25 km), lower-crust (25-70 km), lower crust (70-180 km), and deep-seated (beyond
180 km) for India and adjoining regions are provided in Figure II.6 in APPENDIX II.
3. Site Characterization:
The shear wave velocity profile of a soil column is to be used for site classification adhering
to the National Earthquake Hazard Reduction Program (NEHRP, Building Seismic Safety
Council 2001, [Table 9]) and Uniform Building Code (UBC, Table 10) terminology. The spatial
variation of shear wave velocity (average Vs30 profiles) has to be evaluated using empirical
equations between SPT-N values obtained from geotechnical borehole litholog, and the average
shear wave velocity as given in Table 8, Multichannel Analysis of Surface Waves (MASW),
Spectral Analysis of Surface Waves (SASW) and Cone Penetration Test (CPT) as discussed
in Section 12. Follow the Site Characterization Framework of Figure 6.
4. Site Effects :
(i) Estimate Site Effects following the Framework given in Figure 10, using seismological
data from 3 component seismic stations (preferably accelerations) uniformly sampling
all the representative geological and geomorphological units of the study region. At
each of the sites, the amplification and predominant frequency values are to be
determined from the waveforms of local and regional earthquakes using methods like
Standard Spectral Ratio (Figure 12), Horizontal-to-Vertical Spectral Ratio/ Receiver
Function (Figure 13), Horizontal-to-Vertical Response Spectral Ratio and Generalized
Inversion techniques.
(ii) In regions devoid of earthquakes, one has to estimate the predominant frequencies by
employing the ambient noise (microtremor) measurements and applying Nakamura
technique.
(iii) Estimate site amplifications using geotechnical studies by applying an equivalent linear
model of the soil column and ground motion at the bedrock following the protocol given
in Figure 14, using standard codes like SHAKE 2000 / DEEPSOIL. Standard shear
modulus reduction and damping ratio curves to be used in this analysis are given in
Figures 15 and 16 for sand and rock respectively.
5. Ground Motion Prediction:

Select a suitable ground motion prediction (attenuation) equation valid for the study region
from the list provided in Table 14B. In the event of availability of more than one relation, all the
relations should be considered based on the logic tree framework indicated by Nath and
Thingbaijam (2011b) in Figure III.2 of APPENDIX III.
6. Seismic Hazard Assessment:

Assess the seismic hazard at the bedrock level by considering the inputs assembled from
the previous steps and prepare a map depicting the PGA values using either the deterministic
or probabilistic approaches with10% probability of exceedence in 50 years (475 years return
period). Follow the protocol given in Figures 26 and 27 respectively. While the former
overestimates the hazard, the latter yields definitive estimates. Since most engineers do not
rely on the PGA values only, the Pseudo Spectral Acceleration/Response Spectra (PSA) at
short (0.2 sec) and long (1 sec) periods are also to be computed.
An updated Probabilistic Hazard Model of India comprising of spatial distribution of

PGA and PSA at 0.2 sec and 1 sec for 10% and 2% probability of exceedence at 0.2o grid
spacing (Figure III.6), the hazard truncation curves (Figure III.8) and Design Spectra (Figure
III.7) at major cities in India published by Nath and Thingbaijam (2011b) have been provided in
APPENDIXIII for immediate use.
7. Liquefaction Hazard Assessment:

Evaluate Factor of Safety (FS) against liquefaction of soil layer based on the simplified
procedure of Seed and Idriss (1971) and subsequent revisions of the simplified procedures of
Seed et al. (1983 and 1985), Youd et al. (2001), and Cetin et al. (2004). Follow the framework
for liquefaction hazard assessment given in Figure 18; calculate Cyclic Stress Ration (CSR)
[Equation (35)] and Cyclic Resistance Ration (CRR) [Equation (38)] for the earthquake moment
magnitude Mw 7.5, the Magnitude Scaling Factor (MSF) [Equation (39)] and finally Equation
(34) for computing FS. Follow the sample computation procedure for Liquefaction Hazard in
terms of FS as given in Section 24.7. Ultimately, a factor of safety zonation map is to be
prepared on GIS platform
8. Landslide Hazard Assessment:

Wherever applicable, prepare a landslide hazard zonation map on a GIS platform, in accordance
with the guidelines specified by the Bureau of Indian standards and the framework given in
Figure 24.
9. Thematic Mapping:
Generate thematic maps on a GIS platform for:
(i) Geological/Geomorphological attributes, that serves as the Base Map of the study
region in the scale of the desired level of Microzonation,
(ii) Site Classification based on NEHRP Nomenclature,
(iii) Site Amplification,
(iv) Predominant Frequency,
(v) Pseudo Spectral Acceleration at both short and long periods, and
(vi) Peak Ground Acceleration.
10. GIS Integration for Composite Hazard Microzonation:

Prepare a comprehensive site specific Hazard Microzonation map by integrating the maps
generated above. For this purpose, appropriate weights are to be assigned to each theme in
order of their contribution to region specific hazard and ranks to all features in each theme
following the GIS Integration Logic given in Section 31, preferably using an Analytical Hierarchy
Process as described in Nath (2005 and 2007) and Nath et al. (2008b). Follow the normalized
feature rating scheme adopted for Guwahati Seismic Microzonation (Table 15).
APPENDIX - I
A Unified Earthquake Catalogue for South Asia covering the period 1900-2008
(Nath et al., 2011c)
Nath et al. (2011c) compiled an earthquake catalog of South Asia for 1900-2008 period in a
homogenous Mw magnitude framework. The considered data sources include International
Seismological Center, Global Centroid Moment Tensor database, and several other
publications. The uniform magnitude scaling in MW is achieved through connecting relationships
between the different magnitude types. These relationships are derived by regression analysis
on the available data-pairs. It is seen that on average, mb is lower than MW by 0.16 units while
ML is lower than MW by 0.6-0.8 units depending on the magnitude range. The compiled
instrumental catalogue is declustered to remove foreshocks and aftershocks to derive a
main-shock catalogue. A windowing algorithm is employed for the lower magnitude range
(MW<6.5). For higher magnitudes, declustering is done on the zone of the aftermath seismicity
and subsequent assessment of the temporal coverage of events occurring within the zone
based on the Omoris decay rate law. Aftershock sequences of major earthquakes prior to
2008 are observed to have terminated, except those for 2004 Sumatra MW 9.1 and 2005
Kashmir MW 7.6 earthquakes. The temporal variations of the data completeness for the time-
period 1900-2008 evaluated on the basis of stable number of events for the magnitude range
indicate three verges: first one spanning the entire period, the second one from 1964 onwards
and the third one with considerable improvement in the reportings during the last two decades.
Regional Context and Data Sources
The entire study region comprises of different tectonics: inter-plate collision, subduction
zones, plate divergence, and intra-plate settings. The region is, therefore, broadly classified
into eight sub-regions; namely East Iran (EI), Northwest India Eurasia Convergence (NWIEC),
Central Himalaya (CH), Tibetan Plateau (TP), Northeast India Eurasia Convergence (NEIEC),
Andaman-Nicobar (AN), Mid-Plate (MP), and Northwest Carlsberg Ridge (NCR) as depicted
in Figure I.1. These delineations are anticipated to allow investigations at sub-regional levels.
Figure I.1: The sub-regions of South Asia are depicted along with the seismicity for a period
1976-2008 derived from GCMT database. The beach balls pertain to largest event
within a sliding spatial window of 2ox 2o. The topographic information is adapted
from global digital elevation model: GTOPO30 of US Geological Survey (after Nath
et al., 2011c).
Three major data sources, namely, International Seismological Centre (ISC, http://
www.isc.ac.uk, last accessed August, 2009), US Geological Survey/National Earthquake
Information Center (USGS/NEIC, http://neic.usgs.gov.us, last accessed April, 2009), and
Global Centroid Moment Tensor (GCMT, http://www.globalcmt.org, last accessed April, 2009)
are considered. Predominantly employed magnitude types for earthquake reporting are listed
in Table I.1. As depicted in Figure I.2 (a), the catalog from ISC is seen having higher data
volume with longer temporal coverage corroborating the similar observation by Willemann
and Storchak (2001). The total number of events in the region reported by ISC is ~7.4% of the
total global events (Figure I.2b). The catalog from ISC is employed as the primary data
source owing to its higher data volume and other advantages such as magnitude error provided
for the number of entries, multiple entries in different magnitude types, and multiple estimates
for the same magnitude type. India Meteorological Department (IMD) reports regional and
local earthquakes at the website: http://www.imd.ernet.in, since 2006. Jaiswal and Sinha
(2004) prepared an earthquake catalog for south India based on Rao and Rao (1984), Seeber
et al. (1999), and USGS/NEIC (Jaiswal and Sinha, 2007).
Table I.1: Different magnitude types that are predominantly employed for earthquake
reportings (after McGuire, 2004). Approximate magnitude at which the
magnitude type saturates is also listed
Magnitude type Saturation Reference/s

Local/Richter, ML 6.8 Richter (1935)
Short period body-wave, mb 7.0 Kanamori (1983)
Surface-wave, MS 8.0 Gutenberg (1945)
Duration, MD #
- Real and Teng (1973)
Vertical p-wave, mpv 7.0$ Hori (1969), Bune et al. (1973)
Moment, MW None Hanks and Kanamori (1979)
Long-period body-wave, mB 8.0 Gutenberg and Richter (1956)
mB based moment, MW(mB) 8.3 Bormann and Saul (2008)
Lg wave, Mn or MbLg 7.0 Nuttli (1973)
Vertical surface-wave, MLV 8.0 Hori (1969)
#
Duration magnitudes are used for small earthquakes; $
Based on characteristics of short

period P-waves; comparable to ML; vis--vis characteristics of surface waves.
Figure I.2: (a) Annual reportings of earthquakes with magnitude (all types) 3.0 in South Asia
from three major global agencies: ISC, USGS/NEIC, and GCMT database. The
inset depicts the temporal coverages. (b) The reported events by ISC is ~7.4 % of
the reported global events for magnitudes (all types) 3.0 during 1999-2008 (after
Nath et al., 2011c).
Figure I.3: Spatial coverage of the derived datasets for the magnitude type indicated on each
map (after Nath et al., 2011c).
Figure 1.3 depicts spatial coverage of datasets for different magnitude types. The
dataset for Mw is derived from the GCMT database, and those for other magnitude types,
except ML,IMD and MW,JS, are derived from the ISC catalog. MW, mb, MS, and ML are widespread,
although MW and MS are scanty in the mid-plate regions and ML is negligible in the Northwest
Carlsberg Ridge province. The mpv specified events are confined to northern parts of the
study region in the HindukushPamir province. Events reported in MD are seen across the
northwest and central Himalayas, AndamanNicobar, and peninsular India. It is noted that
the ML magnitude type in ISC are standard (i.e., scaled to Richters definition for California)
within the organizational framework. The data from IMD are also considered with the magnitude
type designated as ML,IMD to differentiate it from the local magnitudes given by ISC. The
records from Jaiswal and Sinha (2004) with the magnitude type designated as MW,JS are
seen clustered within South India.
Regression Analyses
Appropriate correlations between different magnitude types through regression analysis are
envisaged to enable magnitude type conversions (e.g., Castellaro and Bormann, 2007;
Bormann et al., 2007; Thingbaijam et al., 2008, 2009; Yadav et al., 2009). Castellaro et al.
(2006) demonstrated that Generalized Orthogonal Regression (GOR) technique is more
appropriate for the earthquake magnitudes compared to the standard linear least square
technique. However, in situations necessitating treatment of difference in magnitude errors,
more complicated approach such as chi-square regression of Stromeyer et al. (2004) may
be necessary. Nonetheless, GOR is employed in the present analysis owing to the use of
instrumental dataset and lack of comprehensive error estimates. Moreover, the technique
has been successfully employed in the earlier studies. The regression is based on error
variance ratio ( = y2/x2) between the error variances of the variables on vertical and horizontal
axis, respectively. The main objective is the functional interchangeability of the variables
i.e., MY= MX+ and MX= MY. The analyses make use of the average errors computed from
sample error distributions available on the applicable data subset or the estimated average
value in case of non-availability of associated errors. Linear models of the type: MY= MX+
are adopted. In case of the expected linear compatibility between the connecting magnitudes,
linear fits with slope=1 are also examined. The following correlations have been worked out
between different magnitude scales.
Correlation between M S and M W

M W = 0.700(0.018) MS +1.865(0.090), n = 1824, = 0.226 (I.1)
Correlation between m b and M W

M W = 1.000(0.049) mb +0.159(0.249), mb < 6.3, n = 1002 (I.2)
Correlation between M L and M W
The compatibility of the two magnitude types is observed by Cassidy et al. (2005) for the
continental regions. Heaton et al. (1986) also suggested ML and MW to be roughly equivalent.
Correlation between m b and M L
mb =1.00(0.129) ML + 0.007(0.446), n = 3130. (I.3)
m b =1.069(0.034) ML 0.250(0.127), n = 3130, = 0.27. (I.4)

Also see Table I.2.
Table I.2 : The correlation of mb and ML such that mb = ML+a, where n is the number of
data-pairs used for the regression
Sub-region n
EI 0.0200.411 20
NWIEC 0.2230.431 404
CH 0.1390.375 184
TP 0.1440.399 479
NEIEC 0.0360.430 266
AN 0.1060.426 1756
MP 0.6330.896 18
NCR 3
Combined 0.0070.446 3130
Correlation between ML and MD

ML= 1.000(0.140) MD 0.017(0.346), n = 741 (I.5)
ML= 0.991 (0.011) MD 0.009(0.031), n = 741, = 0.62 (I.6)
Correlation between mb and MD
mb =1.00(0.148) MD+ 0.189(0.498), n = 46. (I.7)
mb = 1.031(0.070) MD+ 0.063(0.294), n = 46, = 0.62. (I.8)
Correlation between mb and mpv
(1.9)
Other magnitude types

mb = 1.00(0.031) MW(mB) 0.022(0.192), mb6.2, n = 9 (I.10)
mb = 0.964(0.154) MW(mB)+ 0.21(0.797), = 0.623, n = 9 (I.11)
MW(mB) = 1.113(0.045) mB 0.898(0.267), = 0.902, n = 26 (I.12)
The data-pairs associated with MLV are seen to be highly scattered, and hence, are not
feasible for any correlation. Hori (1969) related MLV and mpv as follows,
MLV = 1.13(0.13) mpv- 0.18(0.06) (I.13)
In case of Mn, no correlating pairs are found with the available 106 entries. Patton
(2001) suggested that Mn and ML have one-to-one correspondences based on a dataset
prepared from events associated with diverse sources. Mn and ML are, therefore, considered
to be practically equivalent. This also agrees with the Mn MW relation (within the uncertainty
bounds) given by Atkinson and Sonley (2005) for southeast Canada. The uniform magnitude
scaling in MW is achieved through connecting relationships between the different magnitude
types. These relationships are derived by regression analysis on the available data-pairs. It
is seen that on average, mb is lower than MW by 0.16 units while ML is lower than MW by 0.6
0.8 units depending on the magnitude range.
Data compilation
The following criteria are implemented for selecting the entries into the compilation:
1. The records derived using the dataset from ISC are employed as primary data.
2. Any entry found to match in the dataset from GCMT, the entry is replaced with the one
obtained from it.
3. Entries in the dataset from GCMT not found in the compilation (after step 2) are directly
adopted in the compilation.
4. If an entry in the catalog given by Jaiswal and Sinha (2004) does not have a match in
the compilation (after step 3), it is accepted in the compilation.
5. The events reported by IMD without any clear match with entries of the compilation
(after step 4) are entered in the compilation.
6. Entries in the present compilation are updated with respect to the published reporting
of magnitude in MW, if available and are found having one-to-one correspondence event
wise.
7. In case the reported event is not found in the compilation (after step 6), it is inserted into
the compilation with appropriate magnitude scaling.
The present compilation is, thus, achieved with higher data volume compared to the
original sources. Figure I.4 depicts the spatial distributions of the events.
Figure I.4: A seismicity map prepared using the compiled catalog (after Nath et al., 2011c).
Figure I.5: A seismicity map of India and the adjoining regions with the man-shock events
with magnitude MW4.0 during the period 19002008 categorized according to
the hypocentral depths (after Nath et al., 2011c).
The spatial distribution of the earthquakes at different hypocentral depth ranges is

depicted in Figure I.5. Figure I.6(a) depicts the distributions of the hypocentral depth uncertainty
( depth) for main-shock events with MW 4.0. Lowered average standard deviation suggests
significant improvement. A histogram of hypocentral depth for the events is depicted in Figure
I.6(b). Data are observed to be clustered at four different depth-ranges (in km): 025, 2570,
70180 and 180300. The data confidence levels are evaluated as percentage of data within
the depth-range as given by the ratio between the total entries found to be within the specified
range considering the uncertainty and that without considering the uncertainty. Mean absolute
deviation for Gaussian distribution is considered, which is equal to about 0.8 times the
standard deviation. The inset in Figure I.6(b) depicts the histogram for the depth ranges. The
confidence levels are found to be 87, 85, 86 and 89 percent for the depth bins, respectively.
These define the limitations of the data employed in the present analysis, and are significantly
close to the acceptable 90% confidence bounds; as such the degree of uncertainty associated
with the earthquake data is generally high (e.g., Wiemer and Wyss, 2000; Thingbaijam et
al., 2008).
Figure I.6: (a) Distribution of hypocentral depth uncertainty ( depth) for the main-shock events
with MW 4.0, (b) Hypocentral depth distribution for the main-shock events binned
at 5 km spacing. The inset depicts histograms for the different depth ranges.
Inscribed are % entries found within the range taking into account the associated
data uncertainty (after Nath et al., 2011c).
The compiled instrumental catalogue is declustered to remove foreshocks and aftershocks

to derive a main-shock catalogue. A windowing algorithm is employed for the lower magnitude
range (MW<6.5). For higher magnitudes, declustering is done on the zone of the aftermath
seismicity and subsequent assessment of the temporal coverage of events occurring within the
zone based on the Omoris decay rate law. Aftershock sequences of major earthquakes prior to
2008 are observed to have terminated, except those for 2004 Sumatra MW 9.1 and 2005 Kashmir
MW 7.6 earthquakes. The temporal variations of the data completeness for the time-period 1900
2008 evaluated on the basis of stable number of events for the magnitude range indicate three
verges: first one spanning the entire period, the second one from 1964 onwards and the third one
with considerable improvement in the reportings during the last two decades.
A seismicity map prepared using the derived main-shock catalog is depicted in Figure I.7.
Figure I.7: Declustered seismicity of South Asia covering a period: 1900-2008 comprising of
30364 events (after Nath et al., 2011c).
The Homogenous Earthquake Catalogue of South Asia in Mw scale can be accessed

at http://earthqhaz.net/sacat/ for the downloadable files for the catalogue as: Early
instrumental and instrumental period (1900-2008) [.txt format, 3.6 MB], Pre-instrumental
period (Prior to 1900) [.txt format, 18.0 KB] & readme file [.pdf format, 20.0 KB].
APPENDIX - II
A Seismogenic Source Framework for the Indian Subcontinent
(Thingbaijam and Nath, 2011)
Fault database, focal mechanisms, and slip rates

The fault database is compiled on Geographical Information System (ArcGIS 9.1) platform.
The sources include seismotectonic map of India published by Geological Survey of India
Figure II.1: A seismotectonic map of India and the adjoining regions prepared with fault patterns
adapted from Wellman (1966), Curray (1991), Dasgupta et al. (2000), and He and
Tsukuda (2003). The depicted earthquake rupture areas were adapted from Bendick
et al. (2007), Ortiz and Bilham (2003), Byrne et al. (1992), and Cummins (2004)
except those enclosed with dashed lines, which are conjectured. Those in lighter
shades have higher uncertainty. Epicenters of events occurring post 1900 are
depicted with stars. The beach balls are adapted from different reporting (after
Thingbaijam and Nath, 2011).
(Dasgupta et al., 2000), seismotectonic map of Afghanistan and adjacent areas published
by US Geological Survey (Wheeler and Rukstales, 2007) and that given by Wellman (1966),
the tectonic framework of Andaman-Sumatra belt given by Curray (1991), and the fault map
of Tibet compiled by He and Tsukuda (2003).
The focal mechanism data employed in the present study are primarily derived from
GCMT Global Centroid Moment Tensor (GCMT, www.globalcmt.org, last access April, 2009),
which covers a period from 1976 to 2008. Those published by several researchers are also
consulted, which include Ben-Menahem et al. (1974), Singh et al. (1975), Banghar (1976),
Chen et al. (1976), Chandra (1977), Singh and Gupta (1980), Byrne et al. (1992), Rastogi
(1992), Chung (1993), Chung and Gao (1995), Chen and Molnar (1990), Ramesh and Estabrook
(1998), Dasgupta et al. (2000), and Bilham and England (2001). A seismotectonic map of the
study region is depicted in Figure II.1.
General Methodologies
Time independent seismicity model
Earthquake occurrences across the globe are universally accepted to follow Gutenberg and
Richter (GR, 1944) relationship,
log10 (m) = a bm (II.1)
where (m) is the cumulative number of earthquakes with magnitude m. The parameters a-
and b-value relates to the background seismicity level and the magnitude size distribution,
respectively. The b-value is often employed as an indicator of stress disparity (Scholz, 1968;
Schorlemmer et al., 2005) due to subtle tectonic reinforcement (Mogi, 1967), and hence can
be utilized to define seismotectonic fabric of the seismogenic provinces (Thingbaijam et al.,
2008; Gulia and Wiemer, 2010). Also, it has been observed to be relatively stable in comparison
to the a-value (Tinti and Mulargia, 1985; Qin et al., 1999). A maximum likelihood method for
estimating b-value given by Aki (1965) and later modified by Bender (1983) is as follows,
(II.2)
where mmean is the average magnitude, mt is the minimum magnitude of completeness,

and m is the magnitude b in size (= 0.1 in the present study). The standard deviation
of b-value b can be computed by means of a bootstrapping approach suggested by

Schorlemmer et al. (2003) which involves the repeated computation each time employing
different replacement events drawn from the catalog. A minimum magnitude constraint
is generally applied on the GR relation given by Equation (II.1) on the basis of the
magnitude of completeness entailed by the linearity of the GR relation on the lower magnitude
range. Apparently, data incompleteness is attributed to instrument insensitivity to lower
magnitudes. At the same time, an upper magnitude has been also suggested in accordance
with physical dissipation of energy and constraints due to the tectonic framework (e.g.,
Kagan, 2002; Kijko, 2004; Thingbaijam and Nath, 2008). This is achieved by establishing the
maximum earthquake mmax; physically capable of occurring within a defined seismic regime
in an underlying tectonic setup. The magnitude distribution is, therefore, truncated at mmax
such that mmax>>mmin. A modified version of Equation (II.1) is a truncated exponential distribution
incorporating an upper bound as follows,
(II.3)
This equation was formulated by Cornell and Vanmarke (1969) and can be found in
Berril and Davis (1980), and Reiter (1990). Berril and Davis (1980) observed that the expression
satisfies the maximum entropy considerations imposed by the mmax constraint. Different
truncation forms have also been proposed by several researchers (e.g., Main, 1996; Utsu,
1999; Kagan, 2002). Fault geodetic measurement often supports characteristic earthquakes
near the maximum earthquake to have lower recurrence periods than the one predicted by
GR relation. In such cases, a delta function or a Gaussian spike in the region of the mmax is
used to formulate characteristic recurrence models (e.g., Schwartz and Coppersmith, 1984;
Youngs and Coppersmith, 1985).
Commonly employed approaches for the estimation of mmax are either based on
seismicity data, fault dimension, geodetic measures or a hybrid one. An earthquake catalog
based approaches include the maximum likelihood method based on seismicity models,
extreme value distribution, and linear extrapolation on the GR trend (e.g., Kijko 2004; Nath et
al., 2005). Probabilistic extrapolation with the truncation of the frequency-magnitude curve at
specific value of annual probability of exceedance is also used (Nuttli, 1981). Alternatively, a
fault based approach involves estimation of mmax based on maximum fault-rupture dimensions
using the relations between magnitude and fault rupture dimensions (e.g., Wells and
Coppersmith, 1994; Hanks and Bakun, 2002). The procedure purposed by Ward (1997)
considers the fault-rupture constrained by the strength and configuration of the faults. Fault
slip rates are often used to assess mmax as well as constrain the recurrence periods (e.g.,
Youngs and Coppersmith, 1985; McGuire, 2004; Pace et al., 2006).
Seismogenic source zonation

A popular approach in the seismogenic localization process is areal source zonation,
wherein the objective is to capture uniform seismicity. However, given the temporal span
of seismicity records and varying completeness thereof, the uniformity of the long-term
seismicity within a classified zone cannot be fully confirmed or dismissed. A viewpoint
given by Woo (1996) is that the earthquake magnitudes have spatial organization in
accordance with the fractal law, which the magnitude-independent areal zonation fails to
address. He explains that b-value for the demarcated zone should be same as those of
sub-zones existing within the zone, and the source geometry should pertain to magnitude
distribution. This does not necessarily conflict with the regional-tectonic specific b-
value, which has been often employed in seismic hazard analysis, e.g., Frankel et al.
(2002), Jaiswal and Sinha (2007), Petersen et al. (2008) among others. An implication
is drawn from the fact that the b-value conforms to the prevalent fault kinematics within
the demarcated zone and vice versa. Incidentally, the maximum earthquake m max has
also been entailed as region specific constrained by the tectonic framework.
In the real situations, it is usually difficult to establish definite tectonic classification for a
given seismogenic zone. In most of the cases, observed faulting mechanisms are non-uniform
necessitating provisions to account for the epistemic (or aleatory?) uncertainties. While seismic
source zonation becomes a case of tectonic dismantling, the standard zoning procedure is
prone to possibly biased hazard estimates (Mucciarelli et al., 2008). The segregation of uniform
tectonic regimes leading to reduced seismogenic zone dimensions with sparse earthquake
occurrence would obscure the seismicity parameterization. In that respect, seismicity
smoothening or zone-free approach has been viewed as a pragmatic one, which accounts for
tectonic uncertainties while adhering to the observed spatial distribution of the earthquake
occurrences. The approach also complies with the fact that the locations of future large
earthquakes tend to follow the spatial distribution of the past seismicity (e.g., Kafka, 2007;
Parsons, 2008). The zone-free methodology has been in vogue since the works of Vere-Jones
(1992), Kagan and Jackson (1994), and Frankel (1995). Several researchers have come up
with different procedural treatments (e.g., Woo, 1996; Cao et al., 1996; Jackson and Kagan,
1999; Frankel et al., 2002; Lapajne et al., 2003). Mucciarelli et al. (2008) observed that the
standard areal zonation approach provides hazard estimates relatively lower than zone-free
approach. On the other hand, Beauval et al. (2006) asserts that in high seismic activity regions,
the smoothing approach yields systematically lower estimates than the zoning method. In
that respect, recent most studies have combined the techniques say seismicity smoothing
for small-moderate earthquakes and fault specific zonation for larger earthquakes (e.g., Petersen
et al., 2008; Kalkan et al., 2009). Alternately, unified approach can be formulated that accounts
for regional seismicity models based on area zonation associated with larger earthquakes for
assessment of b-value and mmax while seismicity smoothing is used to establish activity rate in
order to accommodate absence of fault associability. This delineates the grid cells according
to regions of homogenous seismotectonic characteristics. Eventually, the methodology adopted
in the present study can be outlined into three aspects: (1) delineation of areal source zones
on the basis of seismicity distribution and fault patterns complemented by available focal
mechanism data, (2) formulation of seismicity model and associated uncertainty values for
each source zone, and (3) application of seismicity smoothening algorithm to obtain activity
rates for specific threshold magnitude/s.
Seismogenic source models

Layered seismogenic zones
The seismicity patterns and seismic source dynamics has been observed to have significant
variations with depth (e.g., Prozorov and Dziewonski, 1982; Christova, 1992; Tsapanos, 2000;
Allen et al., 2004) that assuming a single set of seismicity parameters over the entire depth
range can produce incorrect hazard estimations. Accordingly, four hypocentral depth ranges
are considered: upper crust (025 km), lower-crust (2570 km), lower crust (70180 km), and
deep-seated (beyond 180 km). The source zonation at each layer is carried out by considering
the seismicity patterns, fault networks and similarity in the style of focal mechanisms (e.g.,
Cceres et al., 2005). The assessment yields a total of 172 seismic source zones. The
demarcated seismogenic sources zones for the four different seismogenic layers in India
and adjoining regions, as depicted in Figure II.2, encompass areas with implicit fault multiplicity
wherein the frequency magnitude distribution can be expected to be uniform. The layered
seismogenic source model in the present study is expected to facilitate resolving the source
characteristics more precisely than single layer schemes that has been considered hitherto
in the study region.
In order to establish the seismotectonic description at each layer, representative

focal mechanism tensor (i.e., F) is constructed by calculating the weighted average of
the known moment tensors as follows,
(II.4)
n
where N is the total number of the focal mechanisms; M 0 is the scalar moment of the nth
n
focal mechanism, and Fij is a function of the strike, dip, and rake of this focal mechanism
(Aki and Richards, 1980). The average focal mechanism thus obtained for each zone is
depicted in Figure II.2. These enabled characterizing each zone (wherever data is available)
in terms of focal mechanism and their most likely rupture mechanism of the future
earthquakes, especially large magnitude ones (e.g. Meletti et al., 2008).
Seismicity parameters
Two aspects of the seismicity parameters are considered: correlation between the parameters,
and tectonic affinity of the b-value distribution. The b-value is estimated by means of Equation
(II.2). The a-value and consequently the annual activity rate are based on Equations (II.1) and
(II.4). In several cases, zones with similar tectonics are merged to overcome lack of sufficient
number of events i.e., 50 while achieving acceptable uncertainty with the estimated b-
values eventually producing 103 zones out of the 172 zones. The assessment of mmax across
the zones and the results for the zone-wise analyses are summarized in Figure II.6. Figure
II.3 depicts a summary of the seismicity parameters by means of cross plots. It is seen that
a-value and b-value have strong positive correlation while b-value have weak negative correlation
with the maximum earthquake mmax and the observed maximum earthquake mmax,obs. However,
it has been noted that these observations do not imply casual relationships. On the other
hand, the correlation between a-value and mmax or mmax,obs is insignificant. The annual rate
activity for mean magnitude, considered MW 5.0 for the ranges of the magnitudes, exhibits no
correlation with the b-value. Figure II.4 (ae) depicts the mean b-value calculated from the
estimated values for each zone grouped according to the underlying predominant fault type.
The classification of fault type has been done according to the rake, as depicted in Figure II.4
(f). In the zones with predominantly reverse faulting type, b-value ranges between 0.720.96
with lowest mean value of 0.82(0.07) while mean b-value of 1.05(0.11) is seen with the
strike-slip regimes, where it ranges from 0.90 to 1.28. The regimes with predominantly normal
faulting have b-values in the range 1.281.57 and associate the highest mean b-value of
1.37(0.10). The zones with assorted reverse and strike-slip faulting and those with mixed
normal with strike-slip faulting types, respectively have fuzzy b-value ranges exhibiting affinity
towards that of strike-slip faulting type. These observations suggest entailment of the b-value
with the underlying tectonics while existence of assorted faulting types in a zone can generate
fuzzy association between b-value vis--vis the seismotectonic classes.
Figure II.2: A layered seismogenic source model for India and adjoining regions considering
hypocentral depth ranges: (a) 0-25 km, (b) 25-70 km, (c) 70-180 km, and (d) 180-
300 km (after Thingbaijam and Nath, 2011).
Figure II.3: The correlations between the zone-specific seismicity parameters indicated by
Pearsons linear correlation coefficient r, such that r -1 (perfect negative correlation)
to r +1 (perfect positive correlation) and r 0 for no relationship (after Thingbaijam
and Nath, 2011).
Figure II.4: e) The zone specific b-value distribution according to the underlying predominant
(a
focal mechanism types, viz. RR: reverse, SS: strike slip, NN: normal, RS: reverse
with strike slip component, and NS: normal with strike slip component, (f) The
classification of faulting type according the rake, (g) Mean b-value, and (h) distribution
of b corresponding to the different fault kinematics (after Thingbaijam and Nath,
2011).
Construction of seismicity grids
The seismicity smoothening allows modeling of the discrete earthquake distributions into
spatially continuous probability distributions. The fundamental gridded seismicity smoothening
technique of Frankel (1995) is considered. The approach has been previously employed by
several workers across the globe; to name a few, Stirling et al. (2002) in New Zealand,
Frankel et al. (2002) for nation-wide seismic hazard mapping in United States, Akinci et al.
(2004) to generate hazard maps for Northern and Central Italy, Pelaez Montilla et al. (2003) in
Northern Algeria, Lapajne et al. (2003) in Slovenia, and Jaiswal and Sinha (2007) in Peninsular
India.
In the present analysis, the study region is gridded at a regular interval of 0.1; each grid
point encompassing a cell of 0.1o x 0.1o. The seismic activity rate is computed as follows,
(II.5)
where nj(mr) is the number of events with magnitude mr, ij is the distance between ith and
jth cells, and c denotes the correlation distance. The annual activity rate mr is computed
each time as N(mr)/T where T is the (sub) catalog period. Different models are generally
envisaged according to sub-catalogs pertaining to different data completeness to
accommodate the pertinent model uncertainties. Threshold magnitudes of MW 4.0, MW 4.5
and MW 5.5 are employed. Description of the sub-catalogs pertaining to each seismogenic
depth sections alongwith the time-frames for the threshold magnitudes is given in Table II.1.
The temporal coverages conform to the earlier assessments of Nath et al. (2011c), Thingbaijam
et al. (2008 and 2009), and Thingbaijam and Nath (2008). The correlation distances of 55 km,
65 km, and 85 km are decided for the respective cases after calibrating output of several runs
of the smoothing algorithm with the observed seismicity. The corrections for the magnitude
uncertainty are applied as well. Figure II.5 depicts the models obtained at each seismogenic
layer. A significant disparity of seismic activity is observed across the study region suggesting
spatial clustering to be evident with the seismicity. At same time, remarkable conformity to
the underlying tectonics is observed. It is observed that the activity rates are higher in the
plate boundary regions of Hindukush-Pamir, Himalayas, Indo-Myanmar arc, and Andaman-
Sumatra zone. In the intra-plate terrains, Kutch province and Koyna-Warna zone exhibits
higher activity rates. While the activity rates at the intermediate hypocentral depth range are
high along the Indo-Myanmar arc and Andaman-Sumatra tracts, the Hindukush-Pamir tracts
shows exceptionally high activity rates. The seismicity at hypocentral depths>180 km are
mostly confined along these three seismogenic provinces.
Table II.1: The sub-catalogs for the three different threshold magnitudes considered for
the construction of seismicity grids
Sub-catalog
Depth-range (km)
MW 4.0 MW 4.5 MW 5.5
0-25 1994-2008 1964-2008 1903-2008
25-70 1990-2008 1964-2008 1902-2008
70-180 1996-2008 1964-2008 1914-2008
180-300 1970-2008 1984-2008 1912-2008
Seismicity Models for the Source Zones

Time independent recurrence model
In the Global Seismic Hazard Assessment Programme exercise for the India and adjoining
regions, Bhatia et al. (1999) employed conjectural mmax based on the past seismicity but did
not elaborate on the applied estimation technique. More recently, there have been several
investigations on the distribution of mmax in the study region. Thingbaijam and Nath (2008)
carried out assessment for broad seismogenic zones in the northeast India and observed
that seismicity models conform to mmax of MW 8.2-8.7 in the region. Jaiswal and Sinha (2008)
defined the distribution of mmax estimated from seismicity data according to the geologic
zones of craton and rift in the peninsular India with the mean values ranging from MW 5.6 to
MW 8.3. Bilham et al. (2001) observed that the slip potential across the Himalayan tracts
could yield to several great earthquakes of MW ~8.6. In the central Himalayan terrains, Feldl
and Bilham (2006) observed that recurrence of the historical 1505 Nepal Earthquake would
cause an earthquake of magnitude in order of MW ~8.6. Thingbaijam et al. (2009) observed
that the broad source zones in the northwest frontier province of the India-Eurasia plate
convergence zone covering the fault tracts of Sulaiman-Kirthar and Hindukush-Pamir have
mmax estimates in the range of MW 8.05-8.31. These regional assessments have been taken
into account in the present study. Apart from this, the catalog based maximum likelihood
method (Kijko, 2004; Thingbaijam and Nath, 2008) and the probabilistic extrapolation for
annual probability of exceedance of 0.001 is employed in according to the data feasibility. In
case of the zone associated with the 2004 Sumatra earthquake of MW 9.2(0.1), the catalog
based maximum likelihood method yields a maximum earthquake of MW 9.4(0.2).
Figure II.5: Smoothened seismicity models computed at the different hypocentral depth ranges
in terms of activity rate for 100 years considering different sub-catalogs with thresh-
old magnitudes of MW 4.0, MW 4.5 and MW 5.5, respectively (after Thingbaijam and
Nath, 2011).
Table II.2: A comparison of the return period for the maximum earthquake, mmax, in selected
zones estimated from fault slip rate Sr through slip ratio technique, and model
fit of the linear GR and the tapering GR relations. These are denoted as TSR, TGR
and TTGR, respectively. The value of Sr for each zone is decided from the estimated
values as given in the cited reference/s.
Zone mmax Sr TSR TGR TTGR Reference#
(MW) (mm/yr)
1 8.0 16.0 4.97e02 1.56e03 1.56e04 1
2 8.0 16.0 4.97e02 6.28e02 6.27e03 1
3 8.0 16.0 4.97e02 2.72e02 2.72e03 1
4 7.4 12.0 2.19e02 7.63e02 7.63e03 2
5 7.8 7.0 7.85e02 7.16e02 7.15e03 3
6 8.2 10.0 1.17e03 2.61e04 2.67e03 3, 4
7 7.6 6.0 6.3e02 2.0e03 2.00e04 5
8 8.3 8.0 1.73e03 1.37e03 1.37e04 4*
9 8.0 14.0 6.12e02 9.80e02 9.00e03 4
10+34 7.1 8.0 6.12e02 9.80e02 9.05e03 6
11+33 7.8 6.0 6.87e02 3.57e02 3.56e03 7
12 8.6 14.0 1.60e03 5.34e02 4.87e03 8
13 8.6 6.5 3.43e03 1.61e03 1.45e04 9
15 8.3 6.0 2.30e03 1.96e03 1.96e04 10
16 8.3 6.0 2.39e03 9.35e02 8.98e03 10
17 7.0 6.0 2.51e02 3.1e02 3.005e03 11
21 8.6 6.3 3.81e03 2.87e03 2.87e04 9
22 8.8 12 2.89e03 1.17e03 1.17e03 12
23+27+29 7.7 3.0 1.44e03 7.10e02 6.0e03 13
24+25+26 8.5 10 2.0e03 3.67e03 3.67e04 12, 14
28+30+31 7.7 6.5 6.65e02 2.31e02 1.99e03 14
Zone mmax Sr TSR TGR TTGR Reference#

(MW) (mm/yr)
41+42 7.60 1.0 3.80e03 7.73e03 7.73e04 15
47+48+53+54+55 8.3 14.0 9.89e02 1.40e03 1.40e04 16
63 8.2 6.0 1.15e03 7.65e02 7.65e03 12, 13
119 8.3 0.7 1.97e04 7.20e03 7.20e04 15
# 1Burtman and Molnar (1993), 2Van Der Woerd et al. (1998), 3Cowgill (2007), 4Mohadjer et al. (2010),
5
Bendick et al. (2000), 6Valli et al. (2008), 7Murphy et al. (2000), 8Burgess et al. (2009), 9Malik and
Nakata (2003), 10Bernard et al. (2000), 11Pennock et al. (1989), 12Jade (2004), 13Paul et al. (2001),
14
Chen et al. (2004), 15Clark and Bilham (2008), 16Wang et al. (2008).
* Average of Chaman-Gardiz-Konar system, Darvaz-Karakul fault zone, and Herat and Talas-
Ferghana faults.
Figure II.6 depicts the frequency magnitude distribution plots for main-shock events in
each of the seismogenic source zones. The b-value and a-value are estimated by applying
the maximum likelihood method (Aki, 1965; Bender, 1983) on the instrumental catalog. The
incomplete data (including the historical data) are rendered return periods according to the
models, namely, Gutenberg-Richter (GR) and Truncated Gutenberg-Richter (TGR) models.
The return periods for the earthquakes (near mmax) have also been evaluated using average
lower estimate of the fault slip rates in the zone. While the linear GR relation can statistically
accommodate large events if seismic source zone is of appropriate size and the temporal
coverage of the catalog is also long enough (e.g., Molchan et al., 1997), TGR model is
reckoned to be more appropriate considering the energy dissipations at the larger magnitudes.
Possible cases for characteristic models are also noted. However, the overall average slip
rates in the source zones could be still lesser than those employed here owing to the
consideration of the large spatial bounds as well as likely presence of creeping blocks. As
such Characteristic recurrence models entail fault-specific analyses, e.g., Petersen et al.
(2004) in the Gujarat seismic province.
Figure II.6: Frequency magnitude distribution plots for the seismogenic source zones (after
Thingbaijam and Nath, 2011).
APPENDIX - III
Probabilistic Seismic Hazard Assessment of India

(Nath and Thingbaijam, 2011b)
Seismic hazard analysis studies are facilitated by appraisal of underlying seismotectonic

regimes. The usually employed data comprises of earthquake catalog (instrumental and
historical), fault map, focal mechanism data, palaeoseismic information, and fault-slip rates.
Seismogenic Sources and Seismicity Models
The Indian subcontinent encompasses different seismotectonic regimes. Figure III.1 depicts
a seismicity map of the region. The tectonically active interplate regions include the
Himalayas, southern Tibetan Plateau, northwest frontier province, Indo-Myanmar arc, and
Andaman-Sumatra region. Distinct from the tectonic plate boundaries is the intraplate margin
of northeast India that encompasses the Shillong plateau. On the other hand, Peninsular
India is delineated as Stable Continental Region (SCR). The subduction zones include that
of Hindukush-Pamir in the northwest frontier province, Indo-Myanmar arc, and Andaman-
Sumatra seismic belt. Subduction interface earthquakes are also observed across the
Himalayas and the northwestern flanks of the Indian plate boundary.
Thingbaijam and Nath (2011) carried out an extensive study to demarcate and
parameterize the underlying seismogenic source zones in the Indian subcontinent. The
authors employed the earthquake catalog (Nath et al., 2011c), supplemented by records of
historical earthquakes (with reported significant events occurring as late as 0819 AD), focal
mechanism data, fault-slip rates and palaeoseismicity findings. They formulated a layered
seismogenic source zonation corresponding to four hypocentral depth ranges (in km): 0-25,
25-70, 70-180 and 180-300. In total, 172 areal source zones were delineated on the basis of
seismicity, fault patterns, and similarity in fault plane solutions. Owing to the data inadequacy,
several zones were merged to facilitate computation of the seismicity parameters (Figure
II.2). The Gutenberg and Richter (GR) parameters, i.e. a-value and b-value, for the zones
were estimated by means of maximum likelihood method while the maximum earthquakes
mmax decided according to results of earlier investigations or the seismicity models. The
earthquake occurrences are considered to be random. Smoothened seismicity models
represented by activity rates at regular interval of 0.2 were also constructed for threshold
magnitudes of MW 4.0, MW 4.5 and MW 5.5 respectively.
Figure III.1: A seismicity map of India and adjoining regions depicting epicentral locations of the
main-shock events covering the period 0819-2008; broadly classified tectonic prov-
inces excluding active shallow crustal and interface regions are delineated with
different shades (after Nath and Thingbaijam, 2011b).
Ground Motion Prediction Equations

16 Ground Motion Prediction Equations (GMPEs) as listed in Table III.1 are adopted in the
present analysis for hazard computations. The GMPEs are selected according to the
assessment carried out by Nath and Thingbaijam (2011a). Number of candidate GMPEs
were selected for the different seismotectonic provinces across India based on the criteria
given by Bommer et al. (2010), and then subjected to suitability test proposed by Scherbaum
et al. (2009). Nath and Thingbaijam (2011a) reported that there is general conformity amongst
the GMPEs developed for tectonically active shallow crust while those developed for intraplate
regions catering to higher ground motions have lower ranks (or suitability), and the subduction
zones have significant regional implications. In the present selection, included lower-ranked
equations are included such that they are either equally matched or outnumbered by those
of higher ranks. Figure III.2 depicts the logic tree constructed for the GMPEs in the present
analysis. The ranking analyses were carried out using macroseismic intensity data due to
which the ranking parameter (i.e., log-likelihood) does not have large variations in several
cases (cf., Delavaud et al. 2009). Nevertheless, the ranking analysis indicated important
considerations to be taken while adopting the relevant GMPEs such as regional corrections
suggested by the developers, and elimination due to rather poor conformity. The decision to
assign equal weights is taken in order to avoid clear-cut preference. Higher influence by the
GMPEs with higher rank collectively, nonetheless, is achieved owing to the selection.
The equations selected for tectonically active shallow crust regions include that of
Akkar and Bommer (2010), Boore and Atkinson (2008), Campbell and Boore (2008), and
Sharma et al. (2009). The equations developed by Kanno et al. (2006) and Zhao et al. (2006)
addresses tectonically active shallow crust and subduction zones. For the SCR, the adopted
equations include that of Atkinson and Boore (2006), Toro (2002), Raghukanth and Iyengar
(2007), and Campbell (2003). In case of intraplate margin region of northeast India, the
equation developed by Nath et al. (2011a) for the region is included. In order to appropriately
associate active deformations, the equation given by Sharma et al. (2009) is also included.
For the subduction zones, several considerations are imposed according to the
observations of Nath and Thingbaijam (2011a). The equation developed by Atkinson and
Boore (2003) is incorporated with correction for Japan in case of the Himalayas and northwest
India-Eurasia convergence, and with correction for Cascadia in case of Indo-Myanmar and
Andaman-Sumatra subduction zones respectively. The equation developed for Atkinson and
Macias (2009) is restricted to interface earthquakes with magnitude MW 7.5. Likewise, the
equations developed by Gupta (2010), and Lin and Lee (2008) are restricted to intraslab
regions of Indo-Myanmar, and Himalayas/Hindukush-Pamir respectively. The equation
developed by Youngs et al. (1997) based on the world-wide data is also limited to the
intraslab-subduction zones.
The adjustments for compatibility between the GMPEs are applied in the same manner
as carried out by Nath and Thingbaijam (2011a). The mean peak horizontal component of
the ground motion is homogenized in terms of new geometric mean definition, namely
GMRotI50, as given by Boore et al. (2006). The conversion factors given by Beyer and
Bommer (2006), and Campbell and Bozorgnia (2008) are used accordingly. The different
source-to-site distance measures namely, RJB (Joyner-Boore distance), REPIC (epicentral
distance), RRUP (rupture distance), RHYPO (hypocentral distance), and RCF (distance to the
site from center of the fault rupture) for larger earthquakes (MW>6.4) are calculated by
constructing finite-fault models. The predominant focal mechanisms as depicted in Figure
II.2 are used for the purpose. This approach has been adopted instead of using the relations
developed by Scherbaum et al. (2004) for tectonically active shallow crustal regions owing
to different seismotectonic regimes in the present case. In case of smaller magnitude
earthquakes (MW<6.5), coincidence is assumed between RJB and REPIC, RRUP and RHYPO, and
RCF and RHYPO respectively. To estimate the rupture dimensions (i.e., length and width), we
use the relations given by Wells and Coppersmith (1994) for crustal events and those given
by Strasser et al. (2010) for the subduction earthquakes. For the large intraplate earthquakes
with reverse faulting, the fault-rupture area estimated from the magnitude is constrained by
a factor of 2 (Nath and Thingbaijam, 2011a). Following the observation of Mai et al. (2005)
that the hypocenters in strike-slip and crustal dip-slip events mostly occur in deeper sections
of the fault plane, the location of hypocenter is placed on the plane decided by 0.5 and 0.8
(reverse faulting), 0.5 and 0.4 (strike-slip faulting), and 0.5 and 0.2 (normal faulting) of the
rupture length and width respectively from the fault location. The fault location is the top
corner of the fault plane such that the dip is on the right-hand side. Depth to shear-wave
velocity VS=1.0 km/s (Z1.0) is estimated using the relation between Z1.0 and VS30 given by
Chiou and Youngs (2008) while depth to VS=2.5 km (Z2.5) is assigned 2 km following Boore
and Atkinson (2008). The computed scenarios consider varying hypocentral depths with
homogenous distribution of source-to-site distance not less than 15 km for the shallow
crustal zones. This is to avoid estimating ground motions at very near-source locations.
Resolving the uncertainty associated with near-source ground motions is currently a topic
of active research (Mai, 2009). In case of the lower crust and subduction zones, the minimum
hypocentral depths are assigned for the source locations.
Figure III.2: The logic tree framework for the ground motion prediction equations; the assigned
weights are given inside the square brackets and the references for the codes
used for the equations are listed in Table III.1 (after Nath and Thingbaijam 2011b).
Table III.1: Selected Ground Motion Prediction Equations
Tectonic province Reference and code in brackets
Tectonically active shallow crust Akkar and Bommer, 2010 (AKBO10);

Boore and Atkinson, 2008 (BOAT08);
Campbell and Bozorgnia, 2008 (CABO08);
Sharma et al., 2009 (SHAR09)
Active shallow crust/ Subduction Kanno et al., 2006 (KAN06);

Zhao et al., 2006 (ZHAO06)
Tectonic province Reference and code in brackets
Subduction Atkinson and Boore, 2003 (ATBO03);

Atkinson and Macias, 2009 (ATMA09);
Gupta, 2010 (GUPT10);
Lin and Lee, 2008 (LILE08);
Youngs et al., 1997 (YOU97)
Stable continental region Toro, 2002 (TOR02);

Campbell, 2003 (CAM03);
Atkinson and Boore, 2006 (ATBO06);
Raghukanth and Iyengar, 2007 (RAIY07);
Intraplate margin Toro, 2002 (TOR02);

Atkinson and Boore, 2006 (ATBO06);
Sharma et al., 2009 (SHAR09);
Nath et al., 2011a (NATH11)
Seismic Site Conditions

Site conditions considered in previous PSHA studies in India discussed earlier include hard-
rock, rock, and stiff soil conditions. Nath and Thingbaijam (2010) observed that the engineering
bedrock in Guwahati city, northeast India conforms to average shear-wave velocity for upper
30 m soil-column VS30 values ranging from 760-1500 m/s. In Garhwal Himalaya, Mahajan and
Rai (2011) observed that sandstone and conglomerate bedrock have VS in the range of 750-
800 m/s and 950-1000 m/s. In Bangalore, south India, Anbazhagan and Sitharam (2009)
observed that the bedrock profile for weathered (soft) rock with VS~33030 m/s extends from
1 m to ~21 m while that of engineering bedrock with VS~76060 m/s extends from 1 m to
50 m. Maheswari et al. (2010) reported site class D, C and B in Chennai, south India;
shallow bedrock areas conforming to site class B.
Nath et al. (2011b) attempted a first-order nation-wide assessment of site conditions in

India. They observed that ~70% of the total landmass comes under site classes D and C. In
view of the site characterization studies across the country, firm rock site condition (standard
engineering bedrock) is considered to be more realistic for the regional hazard computations.
The standard engineering bedrock conforms to VS30 ~760 m/s (defined as boundary site-
class BC). The GMPEs employed in the present analysis are accordingly adopted for the
respective site condition (cf. Nath and Thingbaijam, 2011b). At the same time, a correction
factor of 1 with negligible uncertainty is considered between site-class BC and site-class B
(e.g., Boore and Atkinson, 2008).
General Methodology
In the probabilistic seismic hazard analysis, annual rate of ground motion exceeding a specific
value is computed to account for different return periods of the hazard. Contributions from all
the relevant sources and possible events are considered. The computational formulation as
developed by Cornell (1968), Esteva (1970), and McGuire (1976) is given as follows,
(a > A) = i P(a > A | m, r , ) f (m) f (r) f (s) dm dr ds (III.1)

m r m r s
i
where (a>A) is the annual frequency of exceedance of ground motion amplitude A, vi, is
the annual activity rate for ith seismogenic source for a threshold magnitude, function P
yields probability of the ground motion parameter a exceeding A given magnitude m at
source-to-site distance r. The standard deviation of the residuals (in log-normal distribution)
associated with GMPE, denoted by is also considered. The corresponding probability
density functions are represented by fm(m), fr(r), and f(). The probability density function
for the magnitudes is generally derived from the GR relation (Gutenberg and Richter, 1944).
The present implementation makes use of the truncated exponential density function given
by Cornell and Vanmarcke (1969),
(III.2)
where = b ln(10), b refers to the b-value of GR relation. The distribution is bounded within
minimum magnitude mmin and maximum magnitude mmax. Instead of considering probability
function for the source-to-distance measure explicitly, point source locations are adopted,
wherein finite fault-ruptures are constructed based on the magnitude and underlying tectonic
regime as discussed earlier. Two schemes adopted are smoothened gridded seismicity
and uniform-seismicity areal zones (or uniformly smoothened) respectively. The smoothened
seismicity is used to predict the activity rates representing spatial feature of earthquake
occurrences. Each grid-point is associated with finite-fault rupturing formulated for the
magnitude and the underlying focal mechanism. In this implementation, the annual activity
rates vary spatially within the source zone but b-value and mmax remain fixed. This assumes
b-value and activity rate to be uncorrelated and a non-uniform distribution of earthquake

probability within a zone. On the other hand, the uniform-seismicity consideration postulates
each point within the areal zone to have equal probability for earthquake occurrences.
Ground motion variability constitute aleatory uncertainty intrinsic to the definition of

GMPEs and consequently to that of PHSA. Computations based only on the median ground
motions ignoring the associated variability are known to underestimate the hazards, especially
at low annual frequencies of exceedance (Bender, 1984; Bommer and Abrahamson, 2006).
Explicit treatment of uncertainty is generally achieved by integrating a number of times the
standard deviation; the number of standard deviations is denoted by and its maximum
value by max. The ground motions become unrealistically high with increasing max necessitating
truncation at a specific value. Truncation at max <3 has been found to be inappropriate (e.g.,
Bommer and Abrahamson, 2006; Strasser et al., 2008). The upper bound of the ground
motions (or constraining the physical limits on the ground motion values) is a topic of
ongoing research (Bommer et al., 2004; Strasser and Bommer, 2009). The values of max
ranging from 2 to 4 are usually employed (e.g., Bernreuter et al. 1989; Romeo and Prestininzi,
2000; Marin et al., 2004). These aspects allow considering max= 3 to be pragmatic, and is
accordingly adopted uniformly for all the GMPEs in the present study.
The hazard computation is performed on grid-points covering the entire study region at
a spacing of 0.2. Logic tree framework is employed in computation at each site to incorporate
multiple models in source considerations, GMPEs and seismicity parameters. Figure III.3
depicts a logic tree formulation at a site. In the present study, the seismogenic source
framework represented by smoothed-gridded seismicity is collectively assigned weight equal
to 0.6. The adopted two models corresponding to the threshold magnitude of MW 4.5 and MW
5.5 are further assigned weights equal to 0.45 and 0.55 respectively. The latter was derived
using earthquake catalog having a longer period compared to the former, and therefore,
entail higher weight. The seismicity model parameters are assigned weights of 0.36 while
the respective 1 standard deviation gets weight equal to 0.32. Similar weight allotment is
assigned to mmax.
The computations are performed with the minimum magnitude equal to MW 4.5. This
consideration is corroborated by seismic intensity attenuation models provided by Szeliga
et al. (2010). The hazard distributions are computed for the source zones at each depth-
section separately, and thereafter, integrated.
Figure III.3: A logic tree formulation at a site for the source specified as that of tectonically active
shallow crust region with predominant strike-slip faulting; the GPME framework for
different tectonic regions is given in Figure III.2 (after Nath and Thingbaijam 2011b).
Deliverables
Figure III.4 depicts spatial distributions of PGA at 10% probability of exceedance in 50
years estimated for each hypocentral depth-section across active Himalayan tracts and
northeast India. The smoothed-gridded seismicity source zonations have been exclusively
considered in this case. The hazard contributions from the upper crust zones (0-25 km
hypocentral depth range) cover the entire region. In case of lower crust zone, i.e. 25-70 km
hypocentral depth range, higher hazards are concentrated in two regions namely west-
central Himalayas, and northeast India. The intraslab earthquakes in Indo-Myanmar arc
occurring in the depth range of 70-180 km constitute a major hazard contributor in northeast
India. On the other hand, western parts of Kashmir are exposed to shallow as well as
subduction earthquakes. Although not depicted in the figure, hazards from shallow as well
as deep-seated earthquakes can also be noted for Andaman-Nicobar Islands. The results
obtained for each depth range are integrated to establish the overall hazard distribution in
the country. Figure III.5 depicts hazard curves obtained at major cities in India.
The seismic hazard maps are presented in Figure III.6. These correspond to spatial
distribution of PGA, PSA at 0.2 sec and 1 sec computed for 10% and 2% probability of
exceedance in 50 years, which correspond to return periods of 475 years and 2475 years
respectively. In the tectonically active region, higher hazard areas include extent of the
Garhwal Himalayas, parts of western Kashmir, and northeast India. Western Gujarat and
Koyna-Warna regions in the stable continental region exhibit higher hazard. Furthermore,
regions in and around Delhi, Jabalpur, Satpura, Latur, Bhadrachalam, Ongole, Bangalore,
Chennai, Coimbatore, and Bengal basin, respectively have relatively higher hazards.
Figure III.4: The spatial distribution of peak ground acceleration estimated for each hypocentral
depth ranges. The computation corresponds to a return period of 475 years and is
exclusively based on smoothed-gridded seismicity source zonations (after Nath
and Thingbaijam 2011b).
Figure III.5: The seismic hazard curves for selected cities (as indicated on each plot) computed
for PGA, PSA at 0.2 sec and 1 sec, respectively for uniform firm rock site (after Nath
and Thingbaijam 2011b).
Seismic Hazard Perspectives

Table III.2 compares the computed PGA with those indicated by BIS (2002), GSHAP and
earlier studies carried out at selected major cities in India. The computation carried out for
10% probability of exceedance in 50 years is considered for the purpose of comparisons.
The present study yields comparatively higher hazard, which is pronounced in the high
seismogenic regions. Jaiswal and Sinha (2007) estimated lower hazard in the highly
seismogenic zones of Kutch, Gujarat, and Koyna-Warna regions. Same is the case with
Menon et al. (2010) in Tamil Nadu. The present results are, however, similar to those of
Anbazhagan et al. (2009) in Bangalore. In the northwestern Himalayas, Mahajan et al.
(2010) estimated maximum PGA as high as 0.75 g. The present analysis associates the
(a) (b)
(c) (d)
(e) (f)
Figure III.6: Seismic hazard distribution in India in terms of PGA, PSA at 0.2 sec. and 0.1 sec. for
firm rock site conditions. Also included in the maps are the data for Nepal, Bhutan,
Bangladesh and Srilanka (after Nath and Thingbaijam 2011b).
region with a maximum of about 0.60 g. In northeast India, the differences between results of
Sharma and Malik (2006) and the present study are observed to be about 1.3-2.0 times less.
In order to evaluate the updated seismic hazard analysis vis--vis the current provisions,
eight cities are selected; two located in the specific seismic zone classified by BIS (2002).
Figure III.7 depicts plots of design response-spectra at 5% damping for firm rock site conditions
at each city. The PSA at 0.2 sec and 1 sec respectively for 2475 years return period are
employed following the scheme outlined by IBC (2006 and 2009). It is observed that the provision
given by BIS (2002) greatly underestimate the hazard distribution. The differences in the estimated
hazard distribution compared to previous studies can be attributed to several factors:
(1) In the present study, the GMPEs have been used as appropriate for different
seismotectonic regimes. This aspect has been overlooked in most of the earlier
studies; for instance Bhatia et al. (1999) employed a single equation for the
entire country disregarding the different seismotectonic provinces, and Menon et
al. (2010) inappropriately employed equations developed for tectonically active
regions although their study region comes under stable continental region. More
details have been given by Nath and Thingbaijam (2011a).
(2) The layered seismogenic source framework based on hypocentral-depth

distribution for the areal zonation, and smoothed-gridded seismicity models
employed in the present study conforms to the variation of seismotectonic attributes
with hypocentral depth, especially in subduction zones. This is a significant
improvement over the previous studies where seismotectonic attribution has been
oversimplified.
(3) Multiple models for the seismogenic source, and seismicity parameters were not
considered in most of the previous studies, except for Jaiswal and Sinha (2007)
and Menon et al. (2010).
(4) The hazard computation in the present study incorporates the ground-motion
variability, which incidentally could be a major reason for the estimation of
comparatively higher hazard. Previous studies delivered the hazard estimates in
terms of median (or mean) ground-motion values. As depicted in Figure III.8, the
observations at different cities indicate that the median ground-motion values are
significantly lower, especially at lower annual exceedance rates.
Table III.2: Estimated peak ground accelerations with 10% probability of exceedance in
50 years at selected major cities across India by Bureau of Indian Standards
(BIS, 2002), Global Seismic Hazard Assessment Program (GSHAP, Bhatia et
al., 1999), present study, and other independent studies are listed. Except for
BIS and the present study, the PGA estimate is the largest value obtained
from the published map contours. The estimate is given for rock site condition
or is otherwise indicated.
City Peak Ground Acceleration (g)
(Latitude, Longitude) BIS (zone)1 GSHAP Present Additional

study notes
Ahmedabad 0.08 (III) 0.05 0.11 0.10 (Jaiswal and Sinha,

(23.03N, 72.61E) 2007)2
Bangalore 0.05 (II) 0.05 0.11 0.10 (Jaiswal and Sinha

(12.98N, 77.58E) (2007)2,
0.15 (Anbazhagan et al.,
2009)
Bhuj 0.18 (V) 0.20 0.42 0.25 (Jaiswal and Sinha,

(23.25N, 69.66E) 2007)2,
0.200.70 (Petersen et al.
2004)3
Chennai 0.08 (III) 0.05 0.12 0.089 (Menon et al.,

(13.00N, 80.18E) 2010),
0.10 (Jaiswal and Sinha,
2007)2
Dehradun 0.12 (IV) 0.30 0.47 0.45 (Mahajan et al.,

(30.33N, 78.04E) 2010)
Guwahati 0.18 (V) 0.30 0.66 0.50 (Sharma and Malik

(26.18N, 91.73E) 2006)
City Peak Ground Acceleration (g)
(Latitude, Longitude) BIS (zone) 1 GSHAP Present Additional

study notes
Hyderabad 0.05 (II) 0.05 0.09 0.08 (Jaiswal and Sinha

(17.45N, 78.46E) 2007)2
Imphal 0.18 (V) 0.45 0.68 0.50 (Sharma and Malik,

(24.78N, 93.94E) 2006)
Jabalpur 0.08 (III) 0.10 0.19 0.15 (Jaiswal and Sinha,

(23.20N, 79.95E) 2007) 2
Kolkata 0.08 (III) 0.10 0.15 0.10 (Jaiswal and Sinha,

(22.65N, 88.45E) 2007)2
Koyna 0.12 (IV) 0.25 0.47 0.25 (Jaiswal and Sinha,

(17.40N, 73.75E) 2007)2
Mumbai 0.08 (III) 0.10 0.16 0.15 (Jaiswal and Sinha,

(19.11N, 72.85E) 2007)2
New Delhi 0.12 (IV) 0.15 0.24 -

(28.56N, 77.11E)
Patna 0.12 (IV) 0.05 0.13 -

(25.60N, 85.10E)
Port Blair 0.18 (V) 0.25 0.71 -

(11.61N, 92.72E)
Shillong 0.18 (V) 0.30 0.72 0.45 (Sharma and Malik,

(25.48N, 92.11E) 2006)
Srinagar 0.18 (V) 0.25 0.33 -

(34.08N, 74.80E)
Thiruvananthapuram 0.08 (III) 0.05 0.07 -

(8.50N, 76.95E)
1
After Menon et al. (2010); 2 Hard rock site; 3 The authors considered only three fault sources.
Figure III.7: Design response spectra (5% damped) for selected cities (after Nath and
Thingbaijam 2011b).
Figure III.8: Seismic hazard curves for PGA derived using present logic tree formulation with the
ground motion truncated at different levels of standard deviation at major cities in
India (after Nath and Thingbaijam 2011b).
References
Abrahamson, N. A., and J. J. Litcheister (1989). Attenuation of vertical peak acceleration, Bull.
Seismol. Soc. Am. 79, 549-580.
Abrahamson, N. A., and K. M. Shedlock (1997). Overview, Seismol. Res. Lett. 68, 9-23.
Abrahamson, N. A., and W. J. Silva (1997). Empirical response spectral attenuation relations for
shallow crustal earthquakes, Seismol. Res. Lett. 68, 94-127.
Abrahamson, N. A., and W. J. Silva (2008). Summary of the Abrahamson and Silva NGA ground
motion relations, Earthq. Spectra 24, 67-97.
Abrahamson, N. A., G. Atkinson, D. Boore, Y. Bozorgnia, K. Campbell, B. Chiou, I. M. Idriss, W. Silva,

and R. Youngs (2008). Comparisons of the NGA Ground-Motion Relations, Earthq. Spectra
24, 45-66.
Aki, K. (1965). Maximum likelihood estimate of b in the formula log N = a - bM and its confidence
limits, Bull. Earthq. Res. Inst. Univ. Tokyo 43, 237-239.
Aki, K., and P. Richards (1980). Quantitative Seismology: Theory and Methods, I, II W. H. Freeman
and Company, San Francisco, 948.
Aki, K. (1988). Local site effects on strong ground motion, Earthq. Eng. Soil. Dynam. II Recent
Advances in Ground Motion Evaluation. June 27-30, Park City, Utah.
Akinci, A., C. Mueller, L. Malagnini, and A. M. Lombardi (2004). A probabilistic seismic hazard
assessment for the Alps and Apennines (Italy) using historical seismicity and new predictive
ground-motion relationships, Boll. Geofis. Teor. Appl. 45, 285-304.
Akkar, S., and J. J. Bommer (2007). Prediction of elastic displacement response spectra in Europe
and the Middle East, Earthq. Eng. Struct. Dynam. 36, 1275-1301.
Akkar, S., and J. J. Bommer (2010). Empirical equations for the prediction of PGA, PGV, and
spectral accelerations in Europe, the Mediterranean region, and the Middle East, Seismol.
Res. Lett. 81, 195-206.
Allen, T. I., G. Gibson, A. Brown, and J. P. Cull (2004). Depth variation of seismic source scaling
relations: Implications for earthquake hazard in southeastern Australia,
Tectonophysics 390, 5-24.
Alvarez, L., J. Garca, F. Vaccari, G. F. Panza, B. Gonzlez, C. Reyes, B. Fernndez, R. Pico, J. A.

Zapata, and E. Arango (2004). Ground motion zoning of Santiago de Cuba: an Approach by
SH waves modelling, Pure. Appl. Geophys. 161, 1041-1059.
Ambraseys, N. N. (1988). Engineering seismology, Earthq. Engng. Struct. Dyn. 17, 1-105.
Ambraseys, N. N., K. A. Simpson, and J. J. Bommer (1996). The Prediction of Horizontal Response
Spectra in Europe, Earthq. Engng. Struct. Dyn. 25, 371-400.
Anbazhagan, P., and T. G. Sitharam (2006). Evaluations of Dynamic Properties and Ground Profiling
using MASW: Correlation between Vs and N60, 13th Symposium on Earthquake Engineering,
Roorkee, India, 390-400 pp.
Anbazhagan, P., and T. G. Sitharam (2009). Mapping of average shear wave velocity for Bangalore
region: A case study, Jour. Environ. Eng. Geophys. 13, 69-84.
Anbazhagan, P., J. S. Vinod, and T. G. Sitharam (2009). Probabilistic seismic hazard analysis for
Bangalore, Nat. Hazards 48,145-166.
Anbazhagan, P., and T. G. Sitharam (2010). Correlation between Low Strain Shear Modulus and
Standard Penetration Test N Values, Proceedings of Fifth International Conference on Recent
Advances in Geotechnical Earthquake Engineering and Soil Dynamics. San Diego, California
1.13b, 10 pp.
Anbazhagan, P., K. K. S. Thingbaijam, S. K. Nath, J. N. Narendara Kumar, and T. G. Sitharam

(2010). Multi-Criteria Seismic Hazard Evaluation for Bangalore City, India, J. Asian Earth Sci.
38, 186-198.
Anderson, J. G., and J. E. Luco (1983). Parametric study of near-field ground motions for oblique-
slip and dip-slip dislocation models, Bull. Seismol. Soc. Am. 73, 45-57.
Anderson, J. G., and S. E. Hough (1984). A model for the shape of the Fourier amplitude spectrum
of acceleration at high frequencies, Bull. Seismol. Soc. Am. 74, 1969-1993.
Anderson, J. G., S. G. Wesnousky, and M. W. Stirling (1996). Earthquake size as a function of fault
slip rate, Bull. Seismol. Soc. Am. 86, 683-690.
Andrews, D. J. (1986). Objective determination of source parameters and similarity of earthquakes

of different size, in Earthquake Source Mechanics, S. Das, J. Boatwright, and C. H. Scholz,
Eds., American Geophysical Union, Washington D.C., 259-268.
Andrus, R. D., and K. H. Stokoe (1997). Liquefaction resistance based on shear wave velocity.
Proc., NCEER Workshop on Evaluation of Liquefaction Resistance of Soils, Nat. Ctr. for
Earthquake Engrg. Res., State Univ. of New York at Buffalo, 89-128.
Arango, I. (1996). Magnitude scaling factors for soil liquefaction evaluations, J. Geotech. Eng.
ASCE 122, 929-936.
Athanasopoulos, G. A. (1995). Empirical correlations Vso-NSPT for soils of Greece: A comparative

study of reliability, Proc. 7th Int. Conf. Soil Dyn. Earthq. Engg., Computation Mechanics
Publications, Southampton, Boston, 19-25.
Atkinson, G. M., and D. M. Boore (1995). Ground Motion Relations for Eastern North America, Bull.
Atkinson, G. M., and D. M. Boore (2003). Empirical ground-motion relations for subduction-zone
earthquakes and their applications to Cascadia and other regions, Bull. Seismol. Soc. Am.
93, 1703-1717.
Atkinson, G. M., and E. Sonley (2005). Empirical relationship between moment magnitude and
Nuttli magnitude for small-magnitude earthquakes in southeastern Canada, Seism. Res.
Lett. 76, 752-755.
Atkinson, G. M., and D. M. Boore (2006). Earthquake ground-motion predictions for eastern North
America, Bull. Seismol. Soc. Am. 96, 2181-2205.
Atkinson, G. M., and M. Macias (2009). Predicted ground motions for Great Interface Earthquakes
in the Cascadia Subduction Zone, Bull. Seismol. Soc. Am. 99, 1552-1578.
Aviles, C. A., C. H. Scholz, and J. Boatwright (1987). Fractal analysis applied to characteristic
segments of the San Andreas fault. J. Geophys. Res. 92, 331-344.
Banghar, A. R. (1976). Mechanism solution of Kinnaur (Himachal Pradesh, India) earthquake of

January 19, 1975, Tectonophysics 31, T5-T11.
Barbat, A., F. Yepez, and J. A. Canas (1996). Damage Scenario Simulation for Risk Assessment in
Urban Zones, Oakland, California. Earthq. Spectra 12, 371-394.
Bard, P. Y., C. Czitrom, J. L. Durville, P. Godefroy, J. P. Meneroud, P. Mouroux, and A. Pecker (1995).
Guidelines for Seismic Microzonation Studies, Delegation of MajorRisks of the French Ministry
of the Environment-Direction for prevention, Pollution and Risks, 50 pp.
Baruah S., S. Baruah, N. K. Gogoi, O. Erteleva., F. Aptikaev, and J. R. Kayal (2009). Ground motion
parameters of Shillong plateau: One of the most seismically active zones of northeastern
India, Earthquake Scienc. 22, 283-291.
Beauval, C., S. Oona, and B. Fabian, (2006). The role of seismicity models in probabilistic seismic
hazard estimation: comparison of a zoning and a smoothing approach, Geophys. J. Int.
165, 584-595.
Bender, B. (1983). Maximum likelihood estimation of b-values for magnitude grouped data, Bull.
Bender, B. (1984). Incorporating acceleration variability into seismic hazard analysis. Bull. Seismol.
Soc. Am. 74, 1451-1462.
Bendick, R., R. Bilham, J. Freymueller, K. Larson, and G. Yin, (2000). Geodetic evidence for a low
slip rate in the Altyn Tagh fault system, Nature 404, 69-72.
Bendick, R., R. Bilham, M. A. Khan, and S. F. Khan (2007). Slip on an active wedge thrust from
geodetic observations of the 8 October 2005 Kashmir earthquake, Geology 35, 267-270.
Ben-Menahem, A., E. Aboodi, and R. Schild (1974). The source of the great Assam earthquakean
interplate wedge motion, Phys. Earth. and Planet. Int. 9, 265-289.
Beresnev, I. A., and G. M. Atkinson (1997). Modelling finite fault radiation from the n spectrum, Bull.
Beresnev, I. A., and G. Atkinson (1999). Generic Finite-Fault Model for Ground Motion Prediction in
Eastern North America, Bull. Seismol. Soc. Am. 89, 608-625.
Bernard, M., B. Shen-Tu, W. E. Holt, and D. M. Davis (2000). Kinematics of active deformation in the
Sulaiman lobe and range, Pakistan, J. Geophys. Res. 105, 13253-13279, doi: 10.1029/
1999JB900405.
Bernreuter, D. L., J. B. Savy, R. W. Mensing, and J. C. Chen (1989). Seismic hazard characterization
of 69 nuclear power plant sites east of the Rocky Mountains. US Nuclear Regulatory
Commission Report NUREG/CR-5250.
Berril, J. B., and R. O. Davis (1980). Maximum entropy and the magnitude distribution, Bull. Seismol.
Soc. Am. 70, 1823-1831.
Beyer, K., and J. J. Bommer (2006). Relationships between median values and between aleatory
variability for different definitions of the horizontal component of motion, Bull. Seismol. Soc.
Am. 96(4), 1512-1522.
Bhandari, R. K., S. Rajarathnam, G. Janardhanan, and G. P. Ganapathy (2003). Seismic

Microzonation-A Training progarmme guide, published by Centre for Disaster Mitigation and
Management, Anna University, Chennai, India, 140 pp.
Bhatia, S. C., M. Ravi Kumar, and H. K. Gupta (1999). A probabilistic seismic hazard map of India
and adjoining regions, Ann. Geofis. 42, 1153-1164.
Bilham, R., and P. England (2001). Plateau pop-up in the 1897 Assam earthquake, Nature 410,
806-809.
Bilham, R., V. K. Gaur, and P. Molnar (2001). Himalayan Seismic Hazard, Science 293, 1442-1444.
BIS (2002). IS 18932002 (Part 1): Indian Standard Criteria for Earthquake Resistant Design of
Structures, Part 1General Provisions and Buildings, Bureau of Indian Standards, New
Delhi.
Blume, J. A. (1965). Earthquake Ground Motion and Engineering Procedures for Important
Installations near Active Faults, Proc. 3rd World Conf. Earthq. Eng., New Zealand 4, 53-67.
Bommer, J. J., N. A. Abrahamson, F. O. Strasser, A. Pecker, P.-Y. Bard, H. Bungum, F. Cotton, D. Fh,
F. Sabetta, F. Scherbaum, and J. Studer (2004). The challenge of defining upper bounds on
earthquake ground motions, Seismol. Res. Lett. 75, 82-95.
Bommer, J. J., and N. A. Abrahamson (2006). Why Do Modern Probabilistic Seismic-Hazard Analysis
Often Lead to Increase Hazard Estimates? Bull. Seismol. Soc. Am. 96, 1967-1977.
Bommer, J. J., J. Douglas, F. Scherbaum, F. Cotton, H. Bungum, and D. Fh (2010). On the

selection of ground-motion prediction equations for seismic hazard analysis, Seismol.Res.
Lett. 81, 794-801.
Boominathan, A., G. R. Dodagoudar, A. Suganthi, and R. Uma Maheswari (2008). Seismic hazard
assessment of Chennai city considering local site effects, J. Earth Sys. Sci. 117, 853-863.
Boore, D. M. (1983). Stochastic Simulation of High-Frequency Ground Motions Based on

Seismological Models of the Radiated Spectra, Bull. Seismol. Soc. Am. 73, 1865-1894.
Boore, D. M., and G. M. Atkinson (1987). Stochastic prediction of ground motion and spectral
response parameters at hard-rock sites in eastern North America, Bull. Seismol. Soc. Am.
77, 440-467.
Boore, D. M., W. B. Joyner, and T. E. Fumal (1993). Estimation of Response spectra and Peak
Accelerations from Western North America Earthquakes. An Interim Report, Open-File-Report
93-509, U. S. Geological Survey, Reston, Virginia.
Boore, D. M., B. William Joyner, and E. Thomas Fumal (1997). Equations for Estimating Horizontal
Response Spectra and Peak Acceleration from Western North American Earethquakes: A
Summary of Recent Work, Seismol. Res. Lett. 68, 128-153.
Boore, D. M. (2004). Estimating Vs30 (or NEHRP Site Classes) from Shallow Velocity Models
(Depths < 30 m), Bull. Seismol. Soc. Am. 94(2), 591-597.
Boore, D. M., J. Watson-Lamprey, and N. A. Abrahamson (2006) Orientation-Independent Measures

of Ground Motion. Bull. Seismol. Soc. Am. 96, 1502-1511.
Boore, D. M., and G. M. Atkinson (2008). Ground-motion prediction equations for the average
horizontal component of PGA, PGV, and 5%-damped PSA at spectral periods between 0.01s
and 10.0s, Earthq. Spectra 24, 99-138.
Borcherdt, R. D. (1970). Effects of local geology on ground motion near San Francisco Bay, Bull.
Borcherdt, R. D., C. M. Wentworth, A. Jannsen, T. Fumal, and J. Gibbs (1991). Methodology for
Predictive GIS Mapping of Special Study Zones for Strong Ground Shaking in the San Francisco
Bay Region, California, Proceedings of the Fourth International Conference on Seismic
Zonation, Stanford, Calfornia 3, 545-552.
Borcherdt, R. D. (1994). Estimates of site-dependent response spectra for design (methodology

and justification), Earthq. Spectra 10, 617-653.
Bormann, P., R. Liu, X. Ren, R. Gutdeutsch, D. Kaiser, and S. Castellaro (2007). Chinese national
network magnitudes, their relation to NEIC magnitudes, and recommendations for new
IASPEI magnitude standards, Bull. Seismol. Soc. Am. 97, 114-127.
Bormann, P., and J. Saul (2008). The new IASPEI standard broadband magnitude mB, Seism.
Res. Lett., 79, 698-706.
Bouchon, M., and K. Aki (1977). Discrete wave-number representation of seismic-source wave-
fields, Bull. Seismol. Soc. Am. 67, 259-277.
Boulanger, R. W., and I. M. Idriss (2004). State normalization of penetration resistance and the
effect of overburden stress on liquefaction resistance. Proceedings 11th Int. Conf. on Soil
Dynamics and Earthquake Engrg. and 33d Int. Conf. on Earthquake Geotech. Engrg. Berkeley,
484-491.
Boyd, O. S., C. S. Mueller, and K. S. Rukstales (2007). Preliminary probabilistic seismic hazard
map for Afghanistan, U. S. Geological Survey Open-File Report, 113.
Brankman, C. M., L. G. Baise, R. B. Higgins, and K. M. Dawson (2004). Liquefaction hazard mapping
in Boston, Massachusetts: Collaborative Research with William Lettis and Associates, Inc.,
and Tudts University, Published in web page, http://earthquake.usgs.gov/research/external/
reports/02hqgr0040.pdf.
Bray, J. D., and R. B. Sancio (2006). Assessment of the Liquefaction Susceptibility of Fine-Grained
Soils, Journal of Geotechnical and Geoenvironmental Engg. ASCE 132, 1165-1177, doi:
10.1061/(ASCE)1090-0241(2006)132:9(1165).
Brune, J. N. (1970). Tectonic Stress and the Spectra of Seismic Shear Waves from Earthquakes, J.
Geophys. Res. 75, 4997-5009.
BSSC (2003). NEHRP recommended provisions for seismic regulations for new buildings and
other structures (FEMA 450), Part 1: Provisions, Building Seismic Safety Council for the
Federal Emergency Management Agency, Washington, DC, USA.
Bune, V. I., N. A. Vvedenskaya, I. V. Gorbunova, K. K. Zapolsky, N. V. Kondorskaya, and I. V. Fedorova

(1973). To the problem of earthquake magnitudes determination, Pageoph 103, 350-361.
Burgess, W., A. Yin, and C. S. Dubey (2009). Investigation of the geometry and kinematics of the
Main Frontal Thrust Zone of the eastern Himalaya reveals eastward increasing Quaternary
Himalayan convergence rate, American Geophysical Union, Fall Meeting, Abstract #T43C-
2131.
Burrough, P. A. (1986). Principles of Geographical Information Systems for Land Resources

Assessment, Glarendon Press, Oxford, UK, 194 pp.
Burrough, P. A., and R. A. McDonnell (1998). Principles of geographic information system, Oxford
University Press, New York, USA, 333 pp.
Burtman, V. S., and P. Molnar (1993). Geological and geophysical evidence for deep subduction of
continental crust beneath the Pamir, Tech. Report Spec. Paper 281, Geological Society of
America.
Byrne, D. E., L. R. Sykes, and D. M. Davis (1992). Great thrust earthquakes and aseismic slip along
the plate boundary of Makran subduction zone, J. Geophys. Res. 97, 449-478.
Cceres, D., D. Monterroso, and B. Tavakoli (2005). Crustal deformation in northern Central
America, Tectonophysics 404, 119-131.
Campbell, K. W. (1981). Near-source attenuation of peak horizontal accelerations, Bull. Seismol.

Soc. Am. 71, 2039-2070.
Campbell, K. W., and Y. Bozorgnia (1994). Near-Source Attenuation of Peak Horizontal Acceleration
from Worldwide Accelerograms Recorded from 1957 to1993, Proceedings 5th U. S. National
Conf. Earthquake Eng. 3, 283-292, Chicago, Illinois.
Campbell, K. W. (2003). Prediction of strong ground motion using the hybrid empirical method and
its use in the development of ground motion (attenuation) relations in eastern North America,
Bull. Seismol. Soc. Am. 93, 1012-1033.
Campbell, K. W., and Y. Bozorgnia (2003). Updated near-source ground-motion (attenuation)

relations for the horizontal and vertical components of peak ground acceleration and
acceleration response spectra, Bull. Seismol. Soc. Am. 93, 314-331.
Campbell, K. W., and Y. Bozorgnia (2006). Next Generation Attenuation (NGA) empirical ground
motion models: Can they be used in Europe? First European Conference on Earthquake
Engineering and Seismology, A joint event of the 13th ECEE & 30th General Assembly of the
ESC., Geneva, Switzerland, 3-8 September, 458 pp.
Campbell, K. W., and Y. Bozorgnia (2008). NGA ground motion model for the geometric mean
horizontal component of PGA, PGV, PGD and 5% damped linear elastic response spectra for
periods ranging from 0.01 to 10s, Earthq. Spectra 24, 139-171.
Cao, T., M. D. Petersen, and M. S. Reichle (1996). Seismic hazard estimate from background
seismicity in southern California, Bull. Seismol. Soc. Am. 86(5), 1372-1381.
Carrara, A., M. Cardinali, F. Detti, F. Guzzetti, V. Pasqui, and P. Reichenbach (1991). GIS techniques
and statistical models in evaluating landslide hazard, Earth Surface Processes and
Landforms 16(5), 427-445.
Cartwright, D. E., and M. S. Longuet-Higgins (1956). The statistical distribution of the maxima of a
random function, Proceedings Roy. Soc. London, Ser. A237, 212-223.
Cassidy, J. F., G. C. Rogers, and J. Ristau (2005). Seismicity in the vicinity of the SNORCLE
corridors of the northern Canadian Cordillera, Can. J. Earth Sci. 42, 1137-1148.
Castellaro, S., F. Mulargia, and Y. Y. Kagan (2006). Regression problems for magnitudes, Geophys.
J. Int. 165, 913-930.
Castellaro, S., and P. Bormann (2007). Performance of different regression procedures on the
magnitude conversion problem, Bull. Seismol. Soc. Am. 97, 1167-1175.
Castro, R. R., F. Pacor, A. Sala, and C. Petrungaro (1996). S-wave attenuation and site effects in the
region of Friuli, Italy, J. Geophys. Res. 101, 22355-22369.
Cetin, K. O., R. B. Seed, A. D. Kiureghian, K. Tokimatsu, Jr. L. F. Harder, R. E. Kayen, and R. E. S.

Moss (2004). Standard penetration test-based probabilistic and deterministic assessment
of seismic soil liquefaction potential, J. Geotech. Geoenviron. Engg. 12, 1314-1340.
Chandra, U. (1977). Earthquakes of peninsular India A seismotectonic study, Bull. Seismol. Soc.
Am. 67, 1387-1413.
Chein, L. K., M. C. Lin, and Y. N. Oh (2000). Shear wave velocity and SPT-N values of in situ
reclaimed soil in west Taiwan, J. Geotech. Eng. Southeast Asian Geotechnical Engineering
Society 31, 63-77.
Chen, Q., J. T. Freymueller, Z. Yang, C. Xu, W. Jiang, Q. Wang, and J. Liu (2004). Spatially variable
extension in southern Tibet based on GPS measurements, J. Geophys. Res. 109, B09401,
doi: 10.1029/2002JB002350.
Chen, W. P., J. L. Nblek, T. J. Fitch, and P. Molnar (1976). An intermediate depth earthquake
beneath Tibet: Source characteristics of the event of September 14, 1976, J. Geophys. Res.
86, 2863-2876.
Chen, W. P., and P. Molnar (1990). Source Parameters of Earthquakes and Intraplate deformation
beneath the Shillong Plateau and the Northern Indoburman Ranges, J. Geophys. Res. 95,
12527-12552.
Chernick, M. R. (1999). Bootstrap Methods: A Practitioners Guide, Wiley Series in Probability and
Statistics, Wiley, New York.
Chiou, B., and R. R. Youngs (2008). An NGA model for the average horizontal component of peak
ground motion and response spectra, Earthq. Spectra 24, 173-215.
Christova, C. (1992). Seismicity depth pattern, seismic energy and b value depth variation in the
Hellenic Wadati-Benioff zone, Phys. Earth Planet. Int. 72, 38-48.
Chung, R. M., F. Y. Yo Kel, and V. P. Drnevich (1984). Evaluation of dynamic properties of sands by
resonant column testing, Geotechnical Testing Journal 7(2), 60-69.
Chung, W. Y. (1993). Source parameters of two rift-associated intraplate earthquakes in peninsular

India: the Bhadrachalam earthquake of April 13, 1969 and the Broach earthquake of March
23, 1970, Tectonophysics 225, 219-230.
Chung, W. Y., and H. Gao (1995). Source parameters of the Anjar earthquake of July 21, 1956, India
and its seismotectonic implications for the Kutch rift basin, Tectonophysics 242, 281-292.
Clark, M., and R. Bilham (2008). Miocene rise of the Shillong Plateau and the Beginning of the end
for the Eastern Himalaya, Earth Planet. Sci. Lett. 269(3), 337-351.
Cloud, W. K., and V. Perez (1971). Unusual Accelerograms Recorded at Lima, Peru, Bull. Seismol.
Soc. Am. 61, 633-640.
Cornell, C. A. (1968). Engineering seismic risk analysis, Bull. Seismol. Soc. Am. 58, 1583-1606.
Cornell, C. A., and E. H. Vanmarke (1969). The Major Influences on Seismic Risk, Proceedings of
the Third World Conference on Earthquake Engineering, Santiago, Chile 1, 69-93.
Coulter, M., and L. Migliaccio (1966). Effects of the earthquake of March 27, 1964 at Valdez, Alaska,
Professional Paper 542-C, U. S. Geological Survey, U.S. Department of the Interior,
Washington, D.C.
Cowgill, E. (2007). Cenozoic right-slip faulting along the east flank of the Pamirs, NW China:
Implications for the kinematic evolution of a major salient in the northern margin of the
Himalayan-Tibetan orogen, American Geophysical Union, Fall Meeting 2007, Abstract #T23D-
1642.
Cummins, P. R. (2004). The potential for giant tsunamigenic earthquakes in the northern Bay of
Bengal, Nature 449, 75-78.
Curray, J. R. (1991). Possible greenschist metamorphism at the base of a 22-km sedimentary

section, Bay of Bengal, Geology 19, 1097-1100.
Das, S., I. D. Gupta, and V. K. Gupta (2006). A Probabilistic Seismic Hazard Analysis of Northeast
India, Earthq. Spectra 22(1), 1-27.
Dasadhikari, M., S. K. Nath, and S. K. Maiti (2011). Landslide Hazard and Risk Analysis in India at
a Regional Scale, Disaster Advances 4(2), 26-39.
Dasgupta, S., P. Pande, D. Ganguly, Z. Iqbal, K. Sanyal, N. V. Venaktraman, S. Dasgupta, B. Sural,

L. Harendranath, K. Mazumdar, S. Sanyal, A. Roy, L. K. Das, P. S. Misra, and H. Gupta (2000).
Seismotectonic Atlas of India and its Environs, Geological Survey of India. Calcutta, India,
Spl. Publicaiton 59, 87 pp.
Davis, B. E. (1996). Geographic Information Systems: A Visual Approach, OnWord Press, Albany,
New York, 400.
Dickenson, S. E., and R. B. Seed (1996). Nonlinear dynamic response of soft and deep cohesive
soil deposits, Proceedings of Int. Workshop on Site Response, Yokosuka, Japan 2, 67-81.
Dikmen, U. (2009). Statistical correlations of shear wave velocity and penetration resistance for
soils, Journal of Geophysics and Engineering 6, 61-72.
Ding, Z, Y. T. Chen, and G. F. Panza (2004). Estimation of site effects in Beijing City, Pure. Appl.
Geophys. 161, 1107-1123.
Donovan, N. C. (1973). Earthquake Hazards for Buildings, Building Practices for Disaster Mitigation,
National Bureau of Standards, US, Department of Commerce, Building Research Services
46, 82-111.
Ehret, D., A. Kienzle, D. Hannich, and W. Wirth (2004). Seismic Microzonation based on geotechnical
Parameters - Estimation of Site Effects in Bucharest (Romania), Geophys. Res. Abstr. 6,
EGU04-A-03708.
El-Sayed, A., I. Korrat, and H. M. Hussein (2004). Seismicity and Seismic Hazard in Alexandria
(Egypt) and its Surroundings, Pure. Appl. Geophys. 161, 1003-1019.
Esteva, L. (1970). Seismic Risk and Seismic Design Decisions, Seismic Design for Nuclear
Power Plants, R. J. Hansen, Editor, MIT Press.
Evans, M. D., and H. B. Seed (1987). Undrained cyclic triaxial testing of gravels: the effect of
membrane compliance, Report UCB/EERC-87/08, Earthquake Engineering Research
Center, University of California, Berkeley, California.
Feldl, N., and R. Bilham (2006). Great Himalayan earthquakes and the Tibetan plateau, Nature 444,
165-170.
Field, E. H., K. H. Jacob, and S. E. Hough (1992). Earthquake Site Response Estimation: A weak
motion case study, Bull. Seismol. Soc. Am. 82, 2283-2307.
Field, E. H., and K. H. Jacob (1993). The theoretical response of sedimentary layers to ambient
seismic noise, Geophys. Res. Lett. 20, 2925-2928.
Field, E. H., and K. H. Jacob (1995). A comparison and test of various site response estimation
techniques, including three that are not reference site dependent, Bull. Seismol. Soc. Am.
85, 1127-1143.
Finn, W. D. L. (1991). Geotechnical engineering aspects of Microzonation, Proceedings of the

Fourth International Conference on Seismic Zonation, Stanford, California 1, 199-259.
Finn, W. D. L., R. H. Ledbetter, and G. Wu (1994). Liquefaction in silty soils: Design and analysis,
Ground Failures under Seismic Conditions, Geotechnical Special Publication, ASCE, New
York 44, 51-76.
Foresman, T. W. (1997). The History of Geographic Information Systems: Perspectives from the
Pioneers. Prentice Hall, Englewood Cliffs, New Jersey, USA, 320 pp.
Frankel, A. (1989). A review of numerical experiments on seismic wave scattering, Pure. Appl.
Geophys. 131, 639-685.
Frankel, A. (1995). Mapping Seismic Hazard in the central and eastern United States, Bull. Seismol.
Soc. Am. 66, 8-21.
Frankel, A. D., M. D. Petersen, C. S. Mueller, K. M. Haller, R. L. Wheeler, E. V. Leyendecker, R. L.

Wesson, S. C. Harmsen, C. H. Cramer, D. M. Perkins, and K. S. Rukstales (2002).
Documentation for the 2002 update of national seismic hazards maps, U.S. Geological
Survey Open-File Report, 02-420.
Fujiwara, T. (1972). Estimation of ground movements in actual destructive earthquakes, Proc, 4th
European Symp. Earthq. Eng. London, 125-132.
Fumal, T. E., and J. C. Tinsley (1985). Mapping shear-wave velocities of near-surface geologic
materials; Evaluating Earthquake Hazards in the Los Angeles Region - An Earth-Science
Perspective J. I. Ziony (Ed.), U. S. Geological Survey Prof. Paper 1360, 127-149.
Giacomo, D., M. R. Gallipoli, M. Mucciarelli, S. Parolai, and S. M. Richwalski (2005). Analysis and
modeling of HVSR in the presence of a velocity inversion: the case of Venosa, Italy, Bull.
Seismol. Soc. Am. 95, 2364-2372.
Grassberger, P., and I. Procaccia (1983). Measuring the strangeness of strange attractors, Physica
9, 189-208.
Gulia, L., and S. Wiemer (2010). The influence of tectonic regimes on the earthquake size
distribution: A case study for Italy, EGU General Assembly 2010, Geophys. Res. Abstr. 12,
EGU2010-2331.
Gupta, I. D. (2010). Response spectral attenuation relations for in-slab earthquakes in Indo-
Burmese subduction zone, Soil Dyn. Earthq. Eng. 30(5), 368-377.
Gutenberg, B., and C. F. Richter (1944). Frequency of earthquakes in California, Bull. Seismol.
Soc. Am. 34, 185-188.
Gutenberg, B. (1945). Magnitude determination for deep-focus earthquakes, Bull. Seismol. Soc.
Am. 35, 117-130.
Gutenberg, B., and C. F. Richter (1956). Magnitude and energy of earthquakes, Ann. Geofis. 9, 1-
15.
Guzzetti, F., A. Carrarra, M. Cardinali, and P. Reichenbach (1999). Landslide hazard evaluation: a
review of current techniques and their application in a multi-scale study, Central Italy,
Geomorphology 31, 181-216.
Hamzehloo, H., F. Vaccari, and G. F. Panza (2007). Towards a reliable seismic microzonation in
Tehran, Iran, Eng. Geo. 93, 1-16.
Hanks, T. C., and H. Kanamori (1979). A moment magnitude scale, J. Geophys. Res. 84, 2348-
2350.
Hanks, T. C., and R. K. McGuire (1981). The Character of High-Frequency Strong Ground Motion,
Hanks, T. C. (1982). fmax, Bull. Seismol. Soc. Am. 72, 1867-1880.
Hanks, T. C., and W. H. Bakun (2002). A bilinear source-scaling model for M-log A observations of
continental earthquakes, Bull. Seismol. Soc. Am. 92, 1841-1846.
Hanna, K. C., and R. B. Culpepper (1998). GIS and Site Design: New Tools for Design
Professionals, John Wiley & Sons Ltd., New York, USA, 240 pp.
Hanumantharao. C., and G. V. Ramana (2008). Dynamic soil properties for microzonation of Delhi,
India, J. Earth Syst. Sci. 117, 719-730.
Harbi, A., S. Maouche, A. Ayadi, D. Benouar, G. F. Panza, and H. Benhallou (2004). Seismicity and
Tectonic Structures in the Site of Algiers and its Surroundings: A Step towards Microzonation,
Pure. Appl. Geophys. 161(5-6), 949-967.
Hartzell, S. H. (1978). Earthquake aftershocks as Greens functions, Geophys. Res. Lett. 5, 1-4.
Hartzell, S. H. (1992). Site response estimation from earthquake data, Bull. Seismol. Soc. Am. 82,
2308-2327.
Hasancebi, N., and R. Ulusay (2007). Empirical correlations between shear wave velocity and
penetration resistance for ground shaking assessments, Bull. Engg. Geo. Environ. 66, 203-
213.
Haskell, N. A. (1964). Total energy and energy spectra density of elastic waves from propagating
faults, Bull. Seismol. Soc. Am. 54, 1811-1841.
He, H., and E. Tsukuda (2003). Recent progresses of active fault research in China, J. Geo. 112(4),
489-520.
Heaton, T. H., F. Tajima, and A. W. Mori (1986). Estimating ground motion using recorded
accelerograms, Surv. Geophys. 8, 25-83.
Heaton, T. H. (1990). Evidence for and implications of self-healing pulses of slip in earthquake
rupture, Phys. Earth Planet. Int. 64, 1-20.
Helmberger, D. V. (1968). The crust-mantle transition in the Bering Sea, Bull. Seismol. Soc. Am.
58, 179-214.
Herak, M., I. Lokmer, F. Vaccari, and G. F. Panza (2004). Linear Amplification of Horizontal Strong
Ground Motion in Zagreb (Croatia) for a Realistic Range of Scaled Point Sources, Pure. Appl.
Geophys. 161(56), 1021-1040.
Herrmann, R. B. (1979). SH wave generation by Dislocation sources-a numerical study, Bull.

Herrmann, R. B., and B. Mandal (1986). A study of wavenumber integration techniques, Earthquake
Notes 57, 33-40.
Hohl, P. (1998). GIS Data Conversion: Strategies, Techniques and Management. OnWord Press,
Albany, New York, 423 pp.
Hori, M. (1969). CGS magnitudes for Japanese earthquakes, Bull. Earth. Res. Inst. 47, 1003-1014.
Hough, S. E., and R. Bilham (2008). Site response of the Ganges basin inferred from re-evaluated
macro-seismic observations from the 1897 Shillong, 1905 Kangra, and 1934 Nepal
earthquakes, J. Earth Sys. Sci. 117, 773-782.
Housner, G. W., and P. C. Jennings (1964). Generation of Artificial Earthquakes, ASCE, J. Eng.
Mech. Div. 90, 113-150.
Hudson, J. A. (1969). A quantitative evaluation of seismic signals at teleseismic distances II.

Body waves and surface waves from and extended source, J. Geophys. 18, 353-370.
Hwang, H., and J. R. Huo (1997). Attenuation relations of ground motion for rock and soil sites in
eastern United States, Soil Dyn. Earthq. Eng. 16, 363-372.
IBC (2000). International Building Code, International Code Council: Inc. 5th Edition, Falls
Church, VA.
IBC (2006). International Building Code, International Code Council.

IBC (2009). International Building Code, International Code Council, Inc., Country Club Hills,
Illinois.
IbsVon, M. Seht., and J. Wohlenberg (1999). Microtremor measurements used to map thickness
of soft sediments, Bull. Seismol. Soc. Am. 89, 250-259.
ICBO (1994). Uniform Building Code, International Conference of Building Officials, ICBO,
Whittier, CA.
Idriss, I. M. (1985). Evaluating Seismic Risk in Engineering Practice, Chapter 6, Proceedings 11th
International Conf. Soil Mechanics and Foundation Eng., San Francisco, 255-320.
Idriss, I. M., and R. W. Boulanger (2005). Evaluation of Liquefaction Potential, Consequences and
Mitigation, Invited Expert Lectures, Proc. Indian Geotechnical Conference-2005, 17-19
December 2005, Ahmedabad, India, 3-25.
Idriss, I. M. (2008). An NGA empirical model for estimating the horizontal spectral values generated
by shallow crustal earthquakes, Earthq. Spectra 24(S1), 217-242.
Imai, T., and Y. Yoshimura (1970). Elastic wave velocity and soil properties in soft soil (in Japanese),
Tsuchito-Kiso. 18(1), 17-22.
Imai, T., and Y. Yoshimura (1975). The relation of mechanical properties of soils to P and S-wave
velocities for ground in Japan, Technical Note, OYO Corporation.
Imai, T., H. Fumoto, and K. Yokota (1975). The relation of mechanical properties of soil to P- and S-
wave velocities in Japan (in Japanese), Proc. 4th Japan Earthquake Engineering Symp., 89-
96.
Imai, T. (1977). P- and S-wave velocities of the ground in Japan, Proc. 9th Int. Conf. on Soil
Mechanics and Foundation Eng. 2, 127-132.
Imai, T., and K. Tonouchi (1982). Correlations of N value with S-wave velocity and shear modulus.
Proceedings of the 2nd European Symposium on Penetration Testing, 24-27.
Irikura, K. (1983). Semi-empirical estimation of strong ground motions during large earthquakes,
Bull. Dis. Prev. Res. Inst., Kyoto University 33, 63-104.
Ishihara, K. (1984). Post-earthquake failure of a tailings dam due to liquefaction of the pond
deposit, Proceedings, International Conference on Case Histories in Geotechnical
Engineering, University of Missouri, St. Louis 3, 1129-1143.
Ishihara, K. (1985). Stability of natural deposits during earthquakes, Proceedings, 11th International
conf. on Soil Mechanics and Foundation Eng. 1, 321-376.
Ishihara, K., and M. Yoshimine (1992). Evaluation of settlements in sand deposits following
liquefaction during earthquakes, Soils and Foundations 32, 173-188.
Ishihara, K. (1993). Liquefaction and flow failure during earthquakes, Geotechnique 43, 351-415.
Iyengar, R. N., and S. Ghosh (2004). Microzonation of earthquake hazard in Greater Delhi area,
Curr. Sci. 87, 1193-1202.
Iyengar, R. N., and S. T. G. Raghukanth (2004). Attenuation of Strong Ground Motion in Peninsular
India, Seismol. Res. Lett. 75, 530-540.
Iyisan, R. (1996). Correlations between shear wave velocity and in-situ penetration test results (in
Turkish), Chamber of Civil Engineers of Turkey. Teknik Dergi 7(2),1187-1199.
Jackson, D. D., and Y. Y. Kagan (1999). Testable earthquake forecasts for 1999, Seismol. Res.
Lett. 70, 393-403.
Jade, S. (2004). Estimates of plate velocity and crustal deformation in the Indian subcontinent
using GPS geodesy, Curr. Sci. 86, 1443-1448.
Jafari, M. K., A. Asghari, and I. Rahmani (1997). Empirical correlation between shear wave velocity
(Vs) and SPT-N value for south of Tehran soils, In: Proceedings of the 4th international
conference on civil engineering, Tehran, Iran (in Persian).
Jafari, M. K., A. Shafiee., and A. Razmkhah (2002). Dynamic properties of fine grained soils in south
of Tehran, Soil Dyn. Earthq. Eng. 4, 25-35.
Jaiswal, K., and R. Sinha (2004). Web portal on earthquake disaster awareness in India,
www.earthquakeinfo.org.
Jaiswal, K., and R. Sinha (2007). Probabilistic seismic-hazard estimation for Peninsular India,
Jaiswal, K., and R. Sinha (2008). Spatial-temporal variability of seismic hazard in peninsular India,
J. Earth Syst. Sci. 117(S2), 707-718.
Jinan, Z. (1987). Correlation between seismic wave velocity and the number of blow of SPT and
depth, Selected papers from the Chinese Journal of Geotechnical Engineering, American
Society of Civil Engineers, 92-100.
Jones, D. T. (1997). Successful ab initio prediction of the tertiary structure of NK-lysin using multiple
sequences and recognized supersecondary structural motifs, Proteins: Struct. Funct. Genet.
S1, 185-191.
Joyner, W. B., and D. M. Boore (1981). Peak Horizontal Acceleration and Velocity from Strong Motion
Records Including Records from the 1979 Imperial Valley, California, Earthquake, Bull.
Seismol. Soc. Am. 71, 2011-2038.
Joyner, W. B., and D. M. Boore (1982). Prediction of Earthquake Response Spectra, U. S. Geol.
Surv., Open File Report, 82-977.
Joyner, W., and T. E. Fumal (1984). Use of measured S-wave velocity for predicting geologic site
effects on strong ground motion, Proceedings of the Eight World Congerence on Earthquake
Engineering, San Francisco, Calif. 2, 777-783.
Kafka, A. L. (2007). Does seismicity delineate zones where future large earthquakes are likely to
occur in intraplate environments? Geol. Soc. Am. Sp. Papers 425, 35-48.
Kagan, Y. Y., and L. Knopoff (1980). Spatial distribution of earthquakes: the two point correlation
function, Geophys. J. R. Astr. Soc. 62, 303-320.
Kagan, Y. Y., and D. D. Jackson (1994). Long-term probabilistic forecasting of earthquakes, J.

Geophys. Res. 99(B7), 13685-13700.
Kagan, Y. Y. (2002). Seismic moment distribution revisited: I. Statistical results, Geophys. J.

Int. 148(3), 520-541.
Kagan, Y. Y. (2007). Earthquake spatial distribution: the correlation dimension, Geophys. J. Int.
168, 1175-1194.
Kalkan, E., P. Gulkan, N. Yilmaz, and M. Celebi (2009). Reassessment of probabilistic seismic
hazard in the Marmara region, Bull. Seismol. Soc. Am. 99, 2127-2146.
Kalteziotis, N., N. Sabatakakis, and J. Vassiliou (1992). Evaluation of dynamic characteristics of

Greek soil formations, In: Second Hellenic Conference on Geotechnical Engineering 2, 239-
246 (in Greek).
Kanagarathinam, L., G. R. Dodagoudar, and A. Boominathan (2008). Probabilistic seismic hazard

studies of east coast region of India, The 14th World Conference on Earthquake Engineering,
October 12-17, Beijing, China.
Kanai, K. (1966). Improved Empirical Formula for Characteristics of Stray Earthquake Motions,
Proc. Japan Earthquake Symposium, 1-4.
Kanai, K., T. Tanaka, T. Morishita and K. Osada (1966). Observation of Microtremors XI, Bull.
Earthq. Res. Ins. University of Tokyo, 44, 1297-1333.
Kanamori, H. (1977). The energy release in great earthquakes, J. Geophys. Res. 82, 2981-2987.
Kanamori, H. (1983). Magnitude scale and quantification of earthquakes, Tectonophys 93, 185-
199.
Kandpal, G. C., B. John, and K. C. Joshi (2009). Geotechnical Studies in Relation to Seismic
Microzonation of Union Territory of Chandigarh, J. Ind. Geophys. 13, 75-83.
Kanno, T., A. Narita, N. Morikawa, H. Fujiwara, and Y. Fukushima (2006). A new attenuation relation
for strong ground motion in Japan based on recorded data, Bull. Seismal. Soc. Am. 96, 879-
897.
Kato, K., K. Aki, and M. Takemura (1995). Site amplification from coda waves: Validation and
application to S-wave site response, Bull. Seismol. Soc. Am. 85, 467-477.
Kayabali, K. (1996). Soil liquefaction evaluation using shear wave velocity, Engg. Geol. 44(1), 121-
127.
Keefer, D. K. (1984). Landslides caused by earthquakes, Geol. Soc. Am. Bull. 95, 406-421.
Kijko, A. (2004). Estimation of the Maximum Earthquake Magnitude, Pure. Appl. Geophys. 161,
1655-1681.
Kiku, H., N. Yoshida, S. Yasuda, T. Irisawa, H. Nakazawa, Y. Shimizu, A. Ansal, and A. Erkan (2001).
In-situ penetration tests and soil profiling in Adapazari, Turkey, In: Proceedings of the ICSMGE/
TC4 satellite conference on lessons learned from recent strong earthquakes, 259-265.
Knopoff, L., and Y. Kagan (1977). Analysis of theory of external values as Applied to Earthquake
Problems, J. Geophys. Res. 82, 5647-5657.
Kobayashi, K. D. (1981). Modeling of dormancy and cold hardiness in red-osier dogwood (Cornus
serica L.), Ph.D. Thesis, Oregon State University, Corvallis, 175 pp.
Kokusho, T. (1987). In-situ Dynamic Soil Properties and their Evaluations, Proc. 8th Asian Regional
Conference on Soil Mechanics and Foundation Engineering. Kyoto, Japan 2, 215-240.
Komatitsch, D., and J. Tromp (1999). Introduction to the spectral-element method for 3-D seismic
wave propagation, Geophys. J. Int. 139, 806-822.
Korte, G. B. (1997). The GIS Book: Understanding the Value and Implementation of Geographic
Information Systems, OnWord Press, Albany, New York, 440 pp.
Kramer, S. L. (1996). Geotechnical Earthquake Engineering, Prentice Hall, Upper Saddle River,
NJ, 653 pp.
Lachet, C., D. Hatzfeld, P. Bard, N. Theodulidis, C. Papaioannou, and A. Savvaidis (1996). Site
Effect and Microzonation in the City of Thessaloniky (Greece). Comparison of Different
Approaches, Bull. Seismol. Soc. Am. 86, 1692-1703.
Lai, S. P. (1982). Statistical characterization of strong ground motions using Power Spectral Density
Function, Bull. Seismol. Soc. Am. 72, 259-274.
Lapajne, J., B. S. Motnikar, and P. Zupancic (2003). Probabilistic seismic hazard assessment
methodology for distributed seismicity, Bull. Seismol. Soc. Am. 93, 2502-2515.
Lee, S. H. H. (1990). Regression models of shear wave velocities, Journal of the Chinese Institute
of Engineers 13, 519-532.
Lermo, J., and F. J. Chavez-Garcia (1993). Site effect evaluation using spectral ratios with only one
station, Bull. Seismol. Soc. Am. 83, 1574-1594.
Liao, S. S. C., and R. V. Whitman (1986). Catalogue of liquefaction and non-liquefaction occurrences
during earthquakes, Res. Rep., Dept. of Civ. Engrg., Massachusetts Institute of Technology,
Cambridge, Mass.
Lin, P. S., and C. T. Lee (2008). Ground-motion attenuation relationships for subduction-zone
earthquakes in northeastern Taiwan, Bull. Seismol. Soc. Am. 98, 220-240.
Longley, P., and M. Batty (1997). Spatial Analysis: Modelling in a GIS Environment, John Wiley &
Sons Ltd., New York, USA, 392 pp.
Mahajan, A. K., S. Slob, R. Ranjan, R. Sporry, P. K. C. Ray, and C. J. Van Westen (2007). Seismic
microzonation of Dehradun City using geophysical and geotechnical characteristics in the
upper 30m of soil column, J. Seismol. 11, 355-370.
Mahajan, A. K., V. C. Thakur., M. L. Sharma., and M. Chauhan (2010). Probabilistic seismic hazard
map of NW Himalaya and its adjoining area, India, Nat. Hazards 53, 443-457.
Mahajan, A. K., and N. Rai (2011). Using MASW to map depth to bedrock underneath Dehradun fan
deposits in NW Himalaya, Curr. Sci. 100, 233-238.
Maheswari, R. U., A. Boominathan, and G. R. Dodagoudar (2010) Seismic site classification and
site period mapping of Chennai City using geophysical and geotechnical data, J. Appl.
Geophys. 72, 152-168.
Mai, P. M. (2009). Ground motion: Complexity and scaling in the near field of earthquake ruptures.
In Encyclopedia of Complexity and Systems Science, R. Meyers (Editor), Springer, New
York 4, 4435-4474.
Main, I. G. (1995). Earthquakes as critical phenomena (1995), Implications for probabilistic Seismic
Hazard Analysis, Bull. Seismol. Soc. Am. 85, 1299-1308.
Main, I. G. (1996). Statistical physics, seismogeneis and seismic hazard, Reviews of Geophysics
34, 433-462.
Malik, J. N., and T. Nakata (2003). Active faults and related late Quaternary deformation along the
northwestern Himalayan frontal zone, India, Ann. Geophys. 46, 917-936.
Mandal, P., N. Kumar, C. Satyamurthy, and I. P. Raju (2009). Ground-motion Attenuation Relation
from Strong-motion Records of the 2001 Mw 7.7 Bhuj Earthquake Sequence (20012006),
Gujarat, India, Pure. App. Geophys. 166, 451-469.
Marble, D. F., and D. J. Pequet (1983). Geographic Information System and Remote Sensing, In
Manual of Remote Sensing, American Society of Photogrammetry, Falls Church, VA.
Marin, S., J. P. Avouac, M. Nicolas, and A. Schlupp (2004). A probabilistic approach to seismic
hazard in metropolitan France, Bull. Seismol. Soc. Am. 94, 2137-2163.
Mayne, P. W., and G. J. Rix (1995). Correlations between shear wave velocity and cone tip resistance
in natural clays, Soils and Foundations 35(2), 193-194.
McGuire, R. K. (1976). FORTRAN Computer Program for Seismic Risk Analysis, U.S. Geological
Survey, Open file report 76-67.
McGuire, R. K., and T. C. Hanks (1980). RMS Acceleration and Spectral Amplitudes of Strong
Ground Motion during the San Fernando, California, Earthquake, Bull. Seismol. Soc. Am. 70,
1907-1919.
McGuire, R. K. (2004). Seismic hazard and risk analysis, EERI Publication, No. MN0-10, ISBN 0-
943198-01-1.
McHarg, I. (1968). Design with Nature, National History Press, New York.
Meletti, C., F. Galadini, G. Valensise, M. Stucchi, R. Basili, S. Barba, G. Vannucci, and E. Boschi
(2008). A seismic source zone model for the seismic hazard assessment of the Italian
territory, Tectonophysics 420, 85-108.
Menke, W. (1989). Geophysical Data Analysis: Discrete Inverse Theory, Academic Press, New
York, 260 pp.
Menon, C. D., S. Pimprikarsad, and P. Mishra (2008). Seismic hazard and risk microzonation of
Jabalpur urban area - A case history, GHZ-04 Earthquake hazard assessment and
geotechnics, International Geological Congress OSLO.
Menon, A., T. Ornthammarath., M. Corigliano, and C. G. Lai (2010). Probalistic Seismic Hazard
Microzonation of Tamil nadu in Southern India, Bull. Seismol. Soc. Am. 100(3), 1320-1341.
Midorikawa, S. (1987). Prediction of Isoseismal Map in Kanto Plain due to Hypothetical Earthquake,
J. Struc. Dynam. 33, 43-48.
Milne, W. G., and A. G. Davenport (1969). Distribution of Earthquake Risk in Canada, Bull. Seismol.
Soc. Am. 59, 754-779.
Mishra, P. S. (2004). Seismic hazard and risk microzonation of Jabalpur, Workshop on Seismic
Hazard and Risk Microzonation of Jabalpur, National Geophysical Research Institute,
Hyderabad, India.
Mogi, K. (1962). Magnitude-frequency relationship for elastic shocks accompanying fractures of

various materials and some related problems in earthquakes, Bull. Earthq. Res. Inst.
University of Tokyo 40, 831-883.
Mogi, K. (1967). Regional variations in magnitudefrequency relation of earthquakes, Tokyo

University, Bull. Earthq. Res. Inst. 5, 67-86.
Mohadjer, S., R. Bendick, A. Ischuk, S. Kuzikov, A. Kostuk, U. Saydullaev, S. Lodi, D. M. Kakar, A.

Wasy, M. A. Khan, P. Molnar, R. Bilham, and A. V. Zubovich (2010). Partitioning of India-Eurasia
convergence in the Pamir-Hindu Kush from GPS measurements, Geophys. Res. Lett. 37,
L04305, doi: 10.1029/2009GL041737.
Mohanty, W. K., M. Y. Walling, S. K. Nath, and I. Pal (2007). First Order Seismic Microzonation of
Delhi, India Using Geographic Information System (GIS), Nat. Hazards 40, 245-260.
Molchan, G., T. Kronrod, and G. F. Panza (1997). Multi-scale seismicity model for seismic risk, Bull.
Seismol. Soc. Am. 87(5), 1220-1229.
Moldoveanu, C. L., M. Radulian, G. H. Marmureanu, and G. F. Panza (2004). Microzonation of

Bucharest: State-of-the-Art, Pure. Appl. Geophys. 161(5-6), 1125-1147.
MonaLisa, A. A. Khwaja., and M. Q. Jan (2007). Seismic Hazard Assessment of the NW Himalayan
Fold-and-Thrust Belt, Pakistan, using probabilistic approach, J. Earthq. Eng. 11, 257-301.
Montalvo-Arrieta, J. C., Y. Quintanilla, A. Tamez, M. Meneses, L. Ramos, and D. Masuch (2005).

Microzonation of the Linares, Northeast Mexico area, based on geology and shear-wave
velocity measurements, Geofsica Internacional 44, 331-340.
Motamed, R., A. Ghalandarzadeh, I. Tawhata, and S. H. Tabatabaei (2007). Seismic microzonation

and damage assessment of Bam City, Southeastern Iran, J. Earthq. Eng. 11, 110-132.
Motazedian, D., and G. M. Atkinson (2005). Stochastic finite-fault modeling based on a dynamic
corner frequency, Bull. Seismol. Soc. Am. 95, 995-1010.
Mucciarelli, M., D. Albarello, and V. DAmico (2008). Comparison of probabilistic seismic hazard
estimates in Italy, Bull. Seismol. Soc. Am. 98(6), 2652-2664.
Mukhopadhyay, S., Y. Pandey, R. Dharmaraju, P. K. S. Chauhan, P. Singh, and A. Dev (2002). Seismic
microzonation of Delhi for ground shaking site effect, Curr. Sci. 82, 877-881.
Mukhopadhyay, S., and P. Bormann (2004). Low cost seismic microzonation using microtremor
data: an example from Delhi, India, J. Asian Earth Sci. 24(3), 271-280.
Murphy, M. A., A. Yin, P. Kapp, T. M. Harrison, D. Lin, and G. Jinghui (2000). Southward propagation
of the Karakoram fault system, southwest Tibet: Timing and magnitude of slip, Geology
28(5) 451-454.
Nakamura, Y. (1989). A Method for Dynamic Characteristics Estimations of Subsurface using

Microtremors on the ground Surface, QR RTRI 30, 25-33.
Narenberg, M. A. H., and C. Essex (1990). Correlation Dimension and Systematic Geometric
Effects, Phys. Rev. A. 42, 7065-7074.
Nath, S. K., N. N. Biswas, M. Dravinski, and A. Papageorgiou (2002a). Determination of S-wave site
response in Anchorage, Alaska in the 1-9Hz frequency band, Pure. Appl. Geophys. 159,
2673-2698.
Nath, S. K., P. Sengupta, and J. R. Kayal (2002b). Determination of S-Wave Site Response at
Garhwal Himalaya from the aftershock sequence of 1999 Chamoli Earthquake, Bull. Seismol.
Soc. Am. 92, 1072-1081.
Nath, S. K., P. Sengupta, S. K. Srivastav, S. N. Bhattacharya, R. S. Dattatrayam, R. Prakash, and H. V.

Gupta (2003). Estimation of S-wave site response in and around Delhi region from weak
motion data, Proceedings Indian Acad. Sci. Earth Planet. Sci. 112, 441-462.
Nath, S. K. (2004). Seismic hazard mapping and microzonation in the Sikkim Himalaya through GIS
integration of site effects and strong ground motion attributes, Nat. Hazards 31, 319-342.
Nath, S. K. (2005). An Initial Model of Seismic Microzonation of Sikkim Himalaya through thematic
mapping and GIS Integration of Geological and Strong Motion Features, J. Asian Earth Sci.
25, 329-343.
Nath, S. K., M. Vyas, I. Pal, and P. Sengupta (2005). A Seismic Hazard Scenario in the Sikkim
Himalaya from Seismotectonics, Spectral Amplification, Source Parameterization, and Spectral
Attenuation Laws using Strong Motion Seismometry, J. Geophys. Res. 110:B01301,
doi:10.1029/2004jb003199.
Nath, S. K. (2007). Seismic Microzoantion Atlas of Guwahati Region, Department of Science &
Technology, Government of India, New Delhi.
Nath, S. K., A. Raj, J. Sharma, K. K. S. Thingbaijam, A. Kumar, D. R. Nandy, M. K. Yadav, S.

Dasgupta, K. Majumdar, J. R. Kayal, A. K. Shukla, S. K. Deb, J. Pathak, P. J. Hazarika, D. K.
Paul, and B. K. Bansal (2008a). Site Amplification, Qs, and Source Parameterization in
Guwahati Region from Seismic and Geotechnical Analysis, Seismol. Res. Lett. 79(4), 526-
539.
Nath, S. K., K. K. S. Thingbaijam, and A. Raj (2008b). Earthquake hazard in the northeast India A
seismic microzonation approach with typical case studies from Sikkim Himalaya and
Guwahati city, J. Earth. Sys. Sci. 117, 809-831.
Nath, S. K., K. Shukla, and M. Vyas (2008c). Seismic hazard scenario and attenuation model of the
Garhwal Himalaya using near-field synthesis from weak motion seismometry, J. Earth Syst.
Sci. 117, 649-670.
Nath, S. K., and K. K. S. Thingbaijam (2009). Seismic hazard assessment A holistic microzonation
approach, Nat. Haz. Earth Sys. Sci. 9, 1445-1459.
Nath, S. K., A. Raj, K. K. S. Thingbaijam, and A. Kumar (2009). Ground motion synthesis and
seismic scenario in Guwahati City A stochastic approach, Seismol. Res. Lett. 80(2), 233-
242.
Nath, S. K., and K. K. S. Thingbaijam (2010). Assessment of Seismic Site Conditions : A Case
Study from Guwahati City, Northeast India, Pure Appl. Geophys., doi: 10.1007/s00024-010-
0197-7.
Nath, S. K., and K. K. S. Thingbaijam (2011a). Peak ground motion predictions in India: an
appraisal for rock sites, J. Seismol., doi: 10.1007/s10950-010-9224-5.
Nath, S. K., and K. K. S. Thingbaijam (2011b). Probabilistic Seismic Hazard Assessment of India,
Seismol. Res. Lett. (In press).
Nath, S. K., K. K. S. Thingbaijam, S. K. Maiti, and A. Nayak (2011a). Ground motion prediction
equation in Shillong region, northeast India, J. Earth Syst. Sci. (under review).
Nath, S. K., K. K. S. Thingbaijam, M. Dasadhikari, N. Devaraj, A. Nayak, S. K. Ghosh, and A. K.

Mahajan (2011b). Topographic gradient based site characterization in India complemented
by strong ground motion spectral attributes, Soil Dyn. Earthq. Eng. (revised manuscript
under review).
Nath, S. K., K. K. S. Thingbaijam, and S. K. Ghosh (2011c). A Unified Earthquake catalogue for
South Asia covering the period 1900 - 2008, Acta Geophysica (manuscript under review).
Nath, S. K., S. K. Maiti, M. Dasadhikari, S. Koley, K. K. S. Thingbaijam, and P. Sengupta (2011d).

Probablistic Seismic Hazard Assessment of West Bengal, J. Seismol. (manuscript under
review).
Navarro, N., and C. S. Oliveiram (2006). Experimental techniques for assessment of dynamic
behavior of buildings, Assessing and managing earthquake risk: geo-scientific and
engineering knowledge for earthquake risk mitigation: developments, tools, techniques,
Geotechnical and Earthquake Engineering, Oliveira, C. S., Roca, A., Goula, X. (Eds.), Springer.
Nuannin, P., O. Kulhanek, and L. Persson (2005). Spatial and temporal b value anomalies preceding
the devastating off coast of NW Sumatra earthquake of December 26, 2004, Geophys. Res.
Lett. 32:L11307, doi: 10.1029/2005GL022679.
Nunziata, C. (2004). Seismic ground motion in Napoli for the 1980 Irpinia earthquake, Pure. Appl.
Geophys. 161, 1239-1264.
Nuttli, O. W. (1973). Seismic wave attenuation and magnitude relations for eastern North America,
J. Geophys. Res., 78, 876-885.
Nuttli, O. W. (1981). On the problem of maximum magnitude of earthquakes, USGS Open Report,
13 pp.
Ohba, S., and I. Toriumi (1970). Dynamic response characteristics of Osaka Plain, Proceedings of
the annual meeting AIJ (in Japanese).
Ohsaki, Y., and R. Iwasaki (1973). On dynamic shear moduli and Poissons ratio of soil deposits,
Soil Found. 13(4), 61-73.
Ohta, T., A. Hara, M. Niwa, and T. Sakano (1972). Elastic shear moduli as estimated from N-value,
Proc. 7th Ann. Convention of Japan Society of Soil Mechanics and Foundation Engineering,
265-268.
Ohta, Y., and N. Goto (1978). Empirical Shear Wave Velocity Equations in terms of Characteristics
Soil Indexes, Earthq. Engng. Struct. Dyn. 6, 167-187.
Okamoto, T., T. Kokusho, Y. Yoshida, and K. Kusuonoki (1989). Comparison of surface versus
subsurface wave source for PS logging in sand layer, In: Proc. 44th. Annual Conf. JSCE. 3,
996-997.
Oprsal, I., and J. Zahradnik (2002). Three-dimensional finite difference method and hybrid modeling
of earthquake ground motion, J. Geophys. Res. 107, doi: 10.1029/2000JB000082.
Ordaz, M., and S. K. Singh (1992). Source spectra and spectral attenuation of seismic waves from
Mexican earthquakes, and evidence of amplification in the hill zone of Mexico City, Bull.
Ortiz, M., and R. Bilham (2003). Source area and rupture parameters of the 31 December 1881 MW
= 7.9 Car Nicobar earthquake estimated from tsunamis recorded in the Bay of Bengal, J.
Geophys. Res. 108, 2215, doi: 10.1029/2002JB001941.
Pace, B., L. Peruzza, G. Lavecchia, and P. Boncio (2006). Layered seismogenic source model and
probabilistic seismic-hazard analyses in Central Italy, Bull. Seismol. Soc. Am. 96, 107-132.
Pal, I., S. K. Nath, K. Shukla, D. K. Pal, A. Raj, K. K. S. Thingbaijam, and B. K. Bansal (2008).
Earthquake hazard zonation of Sikkim Himalaya using a GIS platform, Nat. Hazards 45, 333-
377.
Palmer, G. E., A. Cashmore, and J. Sturtevant (2003). Candida albicans VPS11 is required for
vacuole biogenesis and germ tube formation, Eukaryot Cell 2, 411-421.
Panizza, M. (1991). Geomorphology and seismic risk, Earth Sci. Rev. 31, 11-20.
Panza, G. F. (1985). Synthetic seismograms: The Rayleigh waves modal summation, J. Geophys.
58, 125-145.
Papageorgiou, A. S., and K. Aki (1983). A specific barrier for the quantitative description of
inhomogeneous faulting and the prediction of strong motion: II Application of the model, Bull.
Papastamatiou, D. (1980). Incorporation of crustal deformation to seismic hazard analysis, Bull.

Seismol. Soc. Am. 70, 1321-1335.
Park, C. B., R. D. Miller, and J. Xia (1999). Multi-channel analysis of surface waves, Geophysics 64,
800-808.
Parolai, S., P. Bormann, and C. Milkereit (2002). New relationships between Vs, thickness of
sediments and resonance frequency calculated by the H/V ratio of seismic noise for the
Cologne area (Germany), Bull. Seismol. Soc. Am. 92, 2521-2527.
Parsons, T. (2008). Persistent earthquake clusters and gaps from slip on irregular faults, Nature
Geosci. 1, 59-63.
Parvez, I. A., G. F. Panza, A. A. Gusev, and F. Vaccari (2002). Strong-motion amplitudes in Himalayas
and a pilot study for the deterministic first-order microzonation in a part of Delhi city, Curr. Sci.
82, 158-166.
Parvez, I. A., F. Vaccari, and G. F. Panza (2004). Site-specific microzonation study in Delhi metropolitan
city by 2-D modeling of SH and P-SV waves, Pure. Appl. Geophys. 161, 1165-1184.
Patton, H. J. (2001). Regional magnitude scaling, transportability, and Ms:mb discrimination at

small magnitudes, Pageoph 158, 1951-2015.
Paul, J., R. Burgmann, V. K. Gaur, R. Bilham, K. M. Larson, M. B. Ananda, S. Jade, M. Mukal, T. S.
Anupama, G. Satyal, and D. Kumar (2001). The motion and active deformation of India,
Geophys. Res. Lett. 28, 647- 650.
Pearce, J. T., and J. N. Baldwin (2005). Liquefaction Susceptibility Mapping ST. Louis, Missouri
and Illinois - Final Technical Report, published in web.er.usgs.gov/reports/abstract/2003/cu/
03HQGR0029.pdf.
Pelaez Montilla, J. A., M. Hamdache, and C. L. Casado (2003). Seismic hazard in Northern Algeria
using spatially smoothed seismicity. Results for peak ground acceleration,Tectonophysics
372(1-2), 105-119.
Pennock, E. S., R. J. Lillie, A. S. H. Zaman, and M. Yousaf (1989). Structural Interpretation of
Seismic Reflection Data from Eastern Salt Range and Potwar Plateau, Pakistan, AAPG Bull.
73, 841-857.
Petersen, M. D., B. K. Rastogi, E. S. Schweig, S. C. Harmsen, and J. S. Gomberg (2004). Sensitivity
analysis of seismic hazard for the northwestern portion of the state of Gujarat, India,
Tectonophysics 390(14), 105-115.
Petersen, M. D., A. D. Frankel, S. C. Harmsen, C. S. Mueller, K. M. Haller, R. L. Wheeler, R. L.

Wesson, Y. Zeng, O. S. Boyd, D. M. Perkins, N. Luco, E. H. Field, C. J. Wills, and K. S.
Rukstales (2008). Documentation for the 2008 Update of the United States National Seismic
Hazard Maps: U.S. Geological Survey Open-File Report, 2008-1128.
Pitilakis, K. D., A. Anastasiadis, and D. Raptakis (1992). Field and laboratory determination of
dynamic properties of natural soil deposits, Proc. 10th World Conf. Earthq. Engg., Rotterdam,
1275-1280.
Pitilakis, K., D. Raptakis, K. Lontzetidis, Th. Tika-Vassilikou, and D. Jongmans (1999). Geotechnical
and geophysical description of EURO-SEISTEST, using fiels, laboratory tests and moderate
strong motion recordings, J. Earthq. Eng. 3, 381-409.
Pitilakis, K. (2003). Experimental and theoretical analyses of site effects toward the improvement
of soil classification and design spectra in EC8 and Greek Seismic Code, Final Report,
Organization of Seismic Planning and Protection.
Prozorov, A. G., and A. M. Dziewonski (1982). A method of studying variations in the clustering
property of earthquakes: application to the analysis of global seismicity, J. Geophys. Res. 87,
2829-2839.x.
Pyke, R. (1979). Nonlinear soil models for irregular cyclic loadings, ASCE, J. Geotech. Engg. Div.
105 (GT6), 715-726.
Qin, C., E. E. Papadimitriou, B. C. Papazachos, and G. F. Karakaisis (1999). Spatial distribution of

time-independent seismicity in China, Pure. Appl. Geophys. 154, 101-119.
Raghukanth, S. T. G. (2005). Engineering models for earthquake sources, Ph.D. Thesis, Indian
Institute of Science, Bangalore, India
Raghukanth, S. T. G., and R. N. Iyengar (2006). Seismic hazard estimation for Mumbai city, Curr.
Sci. 9, 1486-1494.
Raghukanth, S. T. G., and R. N. Iyengar (2007). Estimation of seismic spectral acceleration in

Peninsular India, J. Earth Syst. Sci. 116, 199-214.
Rajendran, C. P., K. Rajendran, B. P. Duarah, S. Baruah, and A. Earnest (2004). Interpreting the
style of faulting and paleo-seismicity associated with the 1897 Shillong, northeast India,
earthquake: Implications for regional tectonism, Tectonics 23, TC4009.
Ramesh, D. S., and C. H. Estabrook (1998). Rupture histories of two stable continental region
earthquakes of India, J. Earth Sys. Sci. 107, L225-L233.
Rao, B. R., and P. S. Rao (1984). Historical seismicity of Peninsular India, Bull. Seism. Soc. Am.
74, 2519-2533.
Rao, K. S., and W. K. Mohanty (2001). Microzonation of Delhi region: An approach, Journal of Indian
Building Congress 8, 102-114.
Rao, K. S., and D. N. Satyam (2005). Seismic Microzonation Studies for Delhi Region, Symposium
on Seismic Hazard Analysis and Microzonation, Roorkee, 213-234.
Raptakis, D. G., S. A. J. Anastasiadis, K. D. Pitilakis, and K. S. Lontzetidis (1995). Shear wave

velocities and damping of Greek natural soils, Proc. 10th European Conf. Earthq. Engg.,
Vienna, 477-482.
Raptakis, D., N. Theodulidis, and K. Pitilakis (1998). Data analysis of the Euroseistest strong
motion array in Volvi (Greece): standard and horizontal to vertical ratio techniques, Earthq.
Spectra 14, 203-224.
Raptakis, D. G., F. Chavez-Garcia, K. A. Makra, and K. D. Pitilakis (2000). Site Eeffects at Euroseistest-
I. 2D Determination of the Valley Stucture and Confronation of the Observations with 1D
Analysis, Soil Dyn. Earthq. Eng. 19, 1-22.
Rastogi, B. K. (1992). Seismotectonics inferred from earthquakes and earthquake sequences in

India during the 1980s, Curr. Sci. 62, 101-108.
Real, C. R., and T. Teng (1973). Local Richter magnitude and total signal duration in Southern
California, Bull. Seismol. Soc. Am. 63, 1809-1827.
Reiter, L. (1990). Earthquake Hazard Analysis: Issues and Insights, Colombia University Press,
New York, ISBN 0-231-06534-5.
Richart, F. E., R. D. Woods, and J. R. Hall (1970). Vibration of Soils and Foundations, Prentice-Hall,
Englewood Cliffs, USA.
Richter, C. F. (1935). An instrumental earthquake scale, Bull. Seism. Soc. Am. 25, 1-32.
Robertson, R. K., R. G. Campanella, and A. Wightman (1983). SPT-CPT Correlations, Proc. ASCE
109, GT11, 1449-1459.
Rodriguez-Marek, A., J. D. Bray, and N. A. Abrahamson (2001). An empirical geotechnical seismic

site response procedure, Earthq. Spectra 17(1), 65-87.
Rollins, K. M., N. B. Diehl, and T. J. Weaver (1998). Implications of Vs-BPT (N1)60 correlations for
liquefaction assessment in gravels, GSP 75, ASCE 506-517.
Romeo, R, and A. Prestininzi (2000). Probabilistic versus deterministic hazard analysis: an

integrated approach for siting problems, Soil Dynam. Earthq. Engg. 20, 75-84.
Saaty, T. L. (1968). Mathematical models of arms control and disarmament: Application of

mathematical structures in politics, Wiley (New York) 14, ISBN: 0471748102, 190 pp.
Saaty, T. L. (1980). The Analytic Hierarchy Process, McGraw-Hill International, New York, USA.
Saaty, T. L. (1990). Multicriteria Decision Making: The Analytic Hierarchy Process, RWS Publications,
Pittsburgh, PA.
Sabetta, F., and A. Pugliese (1987). Attenuation of Peak Horizontal Acceleration and Velocity from
Italian Strong-Motion Records, Bull. Seismol. Soc. Am. 77, 1491-1513.
Sabetta, F., and A. Pugliese (1996). Estimation of Response Spectra and Simulation of Non-
stationary Earthquake Ground Motions, Bull. Seismol. Soc. Am. 86, 337-352.
Sadigh, K., J. Egan, and R. Youngs (1986). Specification of Ground Motion for Seismic Design of
Long Period Structures, Earthquake Notes 57, 13.
Sadigh, K., C. Y. Chang, J. A. Egan, F. Makdisi, and R. R. Youngs (1997). Attenuation relationships
for shallow crustal earthquakes based on California strong motion data, Seismol. Res. Lett.
68, 180-189.
Safak, E. (1988). Analytical Approach to Calculation of response Spectra from Seismological Models
of Ground Motion, Earthq. Eng. Struct. Dynam. 16, 121-134.
Safak, E., and D. M. Boore (1988). On low-frequency errors of uniformly modulated filtered white-
noise models for ground motions, Earthq. Eng. Struct. Dynam. 16, 381-388.
Sander, P. (1998). Hard rock aquifers in arid and semi arid zones. Water Resources of Hard Rock
Aquifers in Arid and (Semi-Arid) Zones - Studies and Reports in Hydrology, 58 (edited by J. W.
Lloyd), UNESCO Publishing, Paris, France, 224 pp.
Saragoni, G. R., and G. C. Hart (1974). Simulation of Artificial Earthaquakes, Earthq. Eng. Struct.
Dynam. 2, 249-268.
Scasserra, G., J. P. Stewart., P. Bazzurro., G. Lanzo., and F. Mollaioli (2008). A Comparison of NGA
Ground-Motion Prediction Equations to Italian Data, Bull. Seismol. Soc. Am. 99, 2961-2978.
Schnabel, P. B. (1973). Effects of Local Geology and Distance from Source on Earthquake Ground
Motion, Ph.D. Thesis, University of California, Berkeley.
Scholz, C. H. (1968). The frequency-magnitude relation of microfracturing in rock and its relation to
earthquakes, Bull. Seismol. Soc. Am. 58(1), 399-415.
Schorlemmer, D., G. Neri, S. Wiemer, and A. Mostaccio (2003). Stability and significance tests for b-
value anomalies: Example from the Tyrrhenian Sea, Geophys. Res. Lett. 30, 1835 pp, doi:
10.1029/2003GL017335.
Schorlemmer, D., S. Wiemer, M. Wyss, and D. D. Jackson (2004). Earthquake statistics at Parkfield:
2. Probabilistic forecasting and testing, J. Geophys. Res. B12308, doi: 10.1029/
2004JB003235.
Schorlemmer, D., S. Wiemer, and M. Wyss (2005). Variations in earthquake-size distribution across
different stress regimes, Nature 437, 539-542.
Schwartz, D. D., and K. J. Coppersmith (1984). Fault behavior and characteristic earthquakes:
Examples for the Wasatch and San Andreas Fault zones, Bull. J. Geophys. Res. 89B, 5681-
5698.
Scott, J. B., T. Rasmussen, B. Luke, W. J. Taylor, J. L. Wagoner, S. B. Smith, and J. N. Louie (2006).
Shallow shear velocity and seismic microzonation of the urban Las Vegas, Nevada Basin,
Seeber, L., J. G. Armbruster, and K. H. Jacob (1999). Probabilistic assessment of earthquake

hazard for the State of Maharashtra, Government of Maharashtra, Earthquake Rehabilitation
Cell, Mumbai.
Seed, H. B., and I. M. Idriss (1970). Soil Moduli and Damping Factors for Dynamic Response
Analyses, Rep. No. EERC-70/10, Earthquake Engineering Research Center, University of
California, Berkeley, California.
Seed, H. B., and I. M. Idriss (1971). Simplified Procedure for Evaluating Soil Liquefaction Potential,
Journal of the Soil Mechanics and Foundations Division, ASCE 97, 1249-1273.
Seed, H. B., and I. M. Idriss (1981). Evaluation of liquefaction potential sand deposits based on
observation of performance in previous earthquakes, ASCE National Convention (MO), 81-
544.
Seed, H. B., and I. M. Idriss (1982). Ground motions and soil liquefaction during earthquakes,
Earthquake Engineering Research Institute, Oakland, USA.
Seed, H. B., I. M. Idriss, and I. Arango (1983). Evaluation of Liquefaction Potential using Field
Performance Data, J. Geotech. Eng. 109, 458-482.
Seed, H. B., K. Tokimatsu, L. F. Harder, and R. M. Chung (1985). The influence of SPT procedures
in soil liquefaction resistance evaluations, J. Geotech. Eng. ASCE 111, 1425-1445.
Seekins, L. C., L. Wennerberg, L. Margheriti, and H. P. Liu (1996). Site amplification at five locations
in San Francisco, California; a comparison of S-waves, codas, and microtremors, Bull.
Sharma, M. L. (1998). Attenuation relationship for estimation of peak ground horizontal acceleration
using data from strong motion array in India, Bull. Seismol. Sco. Am. 88, 1063-1069.
Sharma, M. L., and S. Malik (2006). Probabilistic Seismic Hazard Analysis and Estimation of
Spectral Strong Ground Motion on Bed Rock in North East India, 4th International Conference
on Earthquake Engineering, Taipei, Taiwan, Paper No. 15.
Sharma, M., and H. Bungum (2006). New strong ground-motion spectral acceleration relations for
the Himalayan region, Proceedings, First European Conference on Earthquake Engineering
and Seismology (ECEES), Paper 1312.
Sharma, M. L., J. Douglas., H. Bungum., and J. Kotadia (2009). Ground-Motion Prediction Equations
Based on Data from the Himalayan and Zagros Regions, J. Earthq. Eng. 13(8), 1191-1210.
Shibata, T. (1970). Analysis of liquefaction of saturated sand during cyclic loading, Disaster
Prevention Res. Inst. Bull. 13, 563-570.
Shoja-Taheri, J., S. Naserieh, and G. Hadi (2010). A Test of the Applicability of NGA Models to the
Strong Ground-Motion Data in the Iranian Plateau, J. Earthq. Eng. 14, 278-292.
Singh, D. D., B. K. Rastogi, and H. K. Gupta (1975). Surface-wave radiation pattern and source
parameters of Koyna earthquake of December 10, 1967, Bull. Seismol. Soc. Am. 65, 711-
731.
Singh, D. D., and H. K. Gupta (1980). Source dynamics of two great earthquakes of the Indian
Subcontinent: the Bihar-Nepal earthquake of January 15, 1934 and the Quetta earthquake of
May 30, 1935, Bull. Seismol. Soc. Am. 70, 757-773.
Singh, R. P., S. Bhoi, and A. K. Sahoo (2002). Changes observed on land and ocean after Gujarat
earthquake 26 January 2001 using IRS data, Int. J. Remot. 23, 3123-3128.
Sisman, H. (1995). An investigation on relationships between shear wave velocity, and SPT and
pressuremeter test results, M.Sc. Thesis, Ankara University, Geophysical Engineering
Department, Ankara (in Turkish, unpublished).
Sitharam, T. G., P. Anbazhagan, and K. Ganesha Raj (2006). Use of remote sensing and
seismotectonic parameters for seismic hazard analysis of Bangalore, Nat. Hazards 6, 927-
939.
Sitharam, T. G., and P. Anbazhagan (2007). Seismic Hazard Analysis for the Bangalore Region,
Nat. Hazards 40, 261-278.
Sitharam, T. G., and P. Anbazhagan (2008). Seismic microzonation: Principles, practices and
experiments, EJGE Special Volume Bouquet 08(61), http://www.ejge.com/Bouquet08/
Sitharam/Sitharam_ppr.pdf.
Slavov, S., I. Paskaleva, M. Kouteva, F. Vaccari, and G. F. Panza (2004). Deterministic Earthquake
Scenarios for the City of Sofia, Pure. Appl. Geophys. 161, 1221-1237.
Sokolov, V. Y., and Y. K. Chernov (2001). Probabilistic microzonation of the urban territories: a case
of Tashkent city, Pure. Appl. Geophys. 158, 2295-2312.
Somerville, P. G. (2000). Magnitude scaling of near fault ground motions, Earthquak Engineering
and Engineering Seismology 2, 15-24.
Stafford, P. J., F. O. Strasser, and J. J. Bommer (2008). An evaluation of the applicability of the NGA
models to ground-motion prediction in the Euro-Mediterranean region, Bull. Earthq. Eng. 6,
149-177.
Star, S., K. C. McGwire, and J. E. Estes (1997). Integration of Geographic information Systems and
Remote Sensing, Cambridge University Press, Cambridge, UK, 248 pp.
Steidl, J. H. (1993). Variation of site response at the UCSB dense array of portable accelerometers,
Earthq. Spectra 9, 289-302.
Stiedl, J. H., A. G. Tumarkin, and R. J. Archuleta (1996). What is a reference site? Bull. Seismol. Soc.
Am. 86, 1733-1748.
Stirling, M., G. H. McVerry, and K. R. Berryman (2002). A new seismic hazard model for New Zealand,
Stokoe, K. H., G. W. Wright, A. B. James, M. R. Jose (1994). Characterization of geotechnical sites

by SASW method, In: Woods, R. D. (Ed.), Geophysical Characterization of Sites: ISSMFE,
Oxford & IBH Publishing Co. Pvt. Ltd., New Delhi.
Strasser, F. O., Bommer, J. J., and Abrahamson N. A. (2008). Truncation of the distribution of
ground-motion residuals. Journal of Seismology 12, 79-105.
Strasser, F. O., and J. J. Bommer (2009). Strong ground motions - have we seen the worst? Bull.
Seismol. Soc. Am. 99, 2613-2637.
Strasser, F. O., M. C. Arango, and J. J. Bommer (2010) Scaling of the source dimensions of
interface and intraslab subduction-zone earthquakes with moment magnitude, Seismol.
Res. Lett. 81, 941-950.
Stromeyer, D., G. Grnthal, and R. Wahlstrm (2004). Chisquare regression for seismic strength
parameter relations, and their uncertainties, with applications to an Mw based earthquake
catalogue for central, northern and northwestern Europe, J. Seismol. 8, 143-153.
Suyehiro, S., T. Asada, and M. Ohtake (1964). Foreshocks and aftershocks accompanying a
perceptible earthquake in central Japan: on the peculiar nature of foreshocks, Meteorology
and Geophysics 19, 427-435.
Sykora, D. W., and K. H. Stokoe (1983). Correlations of In-situ Measurements in Sands of Shear
Wave Velocity, Soil Characteristics, and Site Conditions, Geotechnical Eng. Report. GR83-
33, The University of Texas at Austin.
Sykora, D. W., and P. J Koester (1988). Correlations between dynamic shear resistance and
standard penetration resistance in soils. Earthq. Eng. Soil Dyn. 2, 389-404.
Szeliga, W., S. Hough, S. Martin, and R. Bilham (2010). Intensity, magnitude, location, and attenuation
in India for felt earthquakes since 1762, Bull. Seismol. Soc. Am. 100, 570-584.
Thenhaus, P. C., and W. Campbell (2003). Seismic Hazard Analysis. In W. F. Chen and C. Scawthorn
(eds.), Earthquake Engineering Handbook, CRC Press, Boca Raton, 1-43.
Thingbaijam, K. K. S., and S. K. Nath (2008). Estimation of maximum earthquakes in northeast

India region, Pure. Appl. Geophys. 165, 889-901.
Thingbaijam, K. K. S., S. K. Nath, A. Yadav, A. Raj, Y. M. Walling, and W. K. Mohanty (2008). Recent
seismicity in northeast India and its adjoining region, J. Seismol. 12, 107-123.
Thingbaijam, K. K. S., P. Chingtham, and S. K. Nath (2009). Seismicity in the northwest frontier
province of Indian-Eurasian Plate Convergences, Seismol. Res. Lett. 80(4), 599-608.
Thingbaijam, K. K. S., and S. K. Nath (2011). A Seismogenic Source Framework for the Indian
Subcontinent, Bull. Seismol. Soc. Am. (revised manuscript under review).
Tinti, S., and F. Mulargia (1985). Effects of magnitude uncertainties on estimating the parameters
in the Gutenberg-Richter frequency-magnitude law, Bull. Seismol. Soc. Am. 75, 1681-1697.
Topal, T., V. Doyuran, N. Karahanoglu, V. Toprak, M. L. Suzen, and E. Yesilnacar (2003). Microzonation
for earthquake hazards: Yenisehir settlement, Bursa, Turkey, Engg. Geo. 70, 93-108.
Toro, G. R., N. A. Abrahamson, and J. F. Schneider (1997). Strong ground motion attenuation
relationships for subduction zone earthquakes, Seismol. Res. Lett. 68, 41-57.
Toro, G. R. (2002). Modification of the Toro et al. (1997) Attenuation Equations for Large Magnitudes
and Short Distances, Risk Engineering, Inc.
Trifunac, M. D., V. W. Lee, H. Cao, and M. I. Todorovska (1988). Attenuation of Seismic Intensity in
Balkan Countries, Report CE 88-01, University of Southern California, Los Angeles, USA.
Tsapanos, T. M. (2000). The depth distribution of seismicity parameters estimated for the South
American area, Earth Planet. Sci. Lett. 180, 103-115.
Tuladhar, R., F. Yamazaki, P. Warnitchai, and J. Saita (2004). Seismic microzonation of the Greater
Bangkok area using microtremor observations, Earthq. Eng. Struct. Dynam. 33, 211-225.
Tumarkin, A. G., and R. J. Archuleta (1997). Recent Advances in Prediction and Processing of
Strong Ground Motions, Nat. Hazards 15, 199-215.
Turcotte, D. L. (1986). A fractal Model for Crustal deformation, Tectonophysics 132, 261-269.
Uma Maheswari, R., D. Boominathan, and G. R. Dodagoudar (2010). Use of Surface Waves in
Statistical Correlations of Shear Wave Velocity and Penetration Resistance of Chennai Soils,
Geotech. Geol. Eng. 28, 119-137.
Utsu, T. (1999). Representation and analysis of the earthquake size distribution: a historical
review and some new approaches, Pure. Appl. Geophys. 155, 509-535.
Valdiya, K. S. (1975). Lithology and age of the Tal Formation in Garhwal and implication on
stratigraphic scheme of Krol Belt in Kumaun Himalaya, J. Geol. Soc. India 16(2), 119-134.
Valli, F., P. H. Leloup, J. L. Paquette, N. Arnaud, H. Li, P. Tapponnier, R. Lacassin, S. Guillot, D. Liu,
E. Deloule, Z. Xu, and G. Mahe (2008). New U-Th/Pb constraints on timing of shearing and
long-term slip-rate on the Karakorum fault, Tectonics 27, TC5007, doi: 10.1029/
2007TC002184.
Van Der Woerd, J., F. J. Ryerson, P. Tapponnier, Y. Gaudemer, R. Finkel, A. S. Meriaux, M. Caffee, Z.
Guoguang, and H. Qunlu (1998). Holocene left-slip rate determined by cosmogenic surface
dating on the Xidatan segment of the Kunlun fault (Qinghai, China), Geology 8, 695-698.
Varnes, D. J. (1978). Slope movement types and processes. In: Schuster R. L. & Krizek R. J. Ed.,
Landslides, analysis and control, Transportation Research Board Sp. Rep. No. 176, Nat.
Acad. Sci., 11-33.
Vere-Jones, D. (1992). Statistical methods for the description and display of earthquake
catalogs, Statistics in the Environmental and Earth Sciences, A. T. Walden and P. Guttorp
(Editors), Arnold Publishers, London, 220-246.
Wang, W. (1979). Some findings in soil liquefaction, Water Conservancy and Hydroelectric Power
Scientific research Institute, Beijing, China.
Wang, W. S. (1984). Earthquake damages to earth dams and levees in relation to soil liquefaction
and weakness in soft clays, Proceedings, International Conference on Case Histories in
Geotechnical Engineering 1, 511-521.
Wang, Y., T. Aung, S. Min, K. Lin, S. Tun, K. Sieh, and U. Myint (2008). Offset ancient city wall yields
plausible slip rate for the Sagaing fault, Burma (Myanmar), American Geophysical Union,
Fall Meeting 2008, Abstract #T51C-06.
Ward, S. N. (1997). More on Mmax, Bull. Seismol. Soc. Am. 87, 1199-1208.
Wellman, H. N. (1966). Active wrench faults of Iran, Afghanistan and Pakistan, Geol. Rundsch. 55,
716-735.
Wells, D. L., and K. J. Coppersmith (1994). New empirical relations among magnitude rupture
length, rupture width, rupture area, and surface displacement, Bull. Seismol. Soc. Am. 84,
974-1002.
Wesnousky, S. G., C. H. Scholz, K. Shimazaki, and T. Matsuda (1983). Earthquake frequency

distribution and the mechanics of faulting, J. Geophys. Res. 88, 9331-9340.
Wheeler, R. L., and K. S. Rukstales (2007). Seismotectonic map of Afghanistan and adjacent
areas, Open-File Report 2007, 1104.
Wiemer, S., and M. Wyss (2000). Minimum magnitude of complete reporting in earthquake catalogs:
examples from Alaska, the Western United States and Japan, Bull. Seismol. Soc. Am. 90,
859-869.
Willemann, R. J., and D. A. Storchak (2001). Data collection at the international seismological
centre, Seismol. Res. Lett. 72, 440-453.
Wills, C. J., and W. J. Silva (1998). Shear-wave velocity characteristics of geologic units in California,
Earthq. Spectra 14, 533-556.
Wills, C. J., M. D. Petersen, W. A. Bryant, M. S. Reichle, G. J. Saucedo, S. S. Tan, G. C. Taylor, and J.

A. Treiman (2000). A site-conditions map for California based on geology and shear wave
velocity, Bull. Seismol. Soc. Am. 90, S187-S208.
Wills, C. J., and K. B. Clahan (2006). Developing a map of geologically defined site-condition
categories for California, Bull. Seismol. Soc. Am. 96, 1483-1501.
Wilson, R. C., and D. K. Keefer (1985). Predicting the areal limits of earthquakes induced landslides.
In: Ziony, J. (Ed.), Evaluating Earthquake Hazards in the Los Angeles Metropolitan Area, U. S.
Geological Survey 1360, 317-345.
Wong, R. T., H. B. Seed, and C. K. Chan (1974). Liquefaction of gravelly soil under cyclic loading
conditions, Journal of the Geotechnical Engineering Division, ASCE 101, 571-583.
Woo, G. (1996). Kernel Estimation Methods for Seismic Hazard Area Source Modeling, Bull. Seismol.
Soc. Am. 86, 353-362.
Wu, R. S., and K. Aki, (1985). Elastic wave scattering by a random medium and the small scale
inhomogeneities in the lithosphere, J. Geophys. Res. 90, 10261-10273.
Wu, R. S., S. Jin, and X. B. Xie (2000). Energy Partition and Attenuation of Lg Waves by Numerical
Simulation Using Screen Propagators, Phys. Earth. Planet. Inter. 120, 227-243.
Yadav, R. B. S., P. Bormann, B. K. Rastogi, M. C. Das, and S. Chopra (2009). A homogeneous and
complete earthquake catalog for northeast India and the adjoining region, Seism. Res. Lett.
80, 609-627.
Yegian, M. K., V. G. Gharaman, and R. N. Harutiunyan (1994). Liquefaction and embankment

failure case histories, 1988 Armenai earthquake, J. Geotech. Eng., ASCE 120, 581-596.
Yilmaz, I., and A. Bagci (2006). Soil liquefaction susceptibility and hazard mapping in the residential
area of Kutahya (Turkey), Environ. Geol. 49, 708-719.
Yin, K. L., and T. Z. Yan (1988). Statistical prediction model for slope instability of metamorphosed
rocks, Proceeding of 5th Int. Symp. on landslide, Lausanne 2, 1269-1272.
Youd, T. L., and S. N. Hoose (1977). Liquefaction susceptibility and geologic setting, Proceedings,
6th World Conference on Earthquake Engineering, New Delhi 3, 2189-2194.
Youd, T. L., and D. M. Perkins (1978). Mapping of liquefaction induced ground failure potential,
Journal of the Geotechnical Engineering Division, ASCE 140(GT4), 433-446.
Youd, T. L., E. L. Harp, D. K. Keefer, and R. C. Wilson (1985). The Borah Peak, Idaho earthquake of
October 28, 1983 Liquefaction, Earthq. Spectra 2, 71-89.
Youd, T. L. (1991). Mapping of earthquake-induced liquefaction for seismic zonation, Proceedings,

4 th International Conference on Seismic Zonation, Earthquake Engineering Research
Institute, Stanford University 1, 111-147.
Youd, T. L., and S. K. Noble (1997). Magnitude scaling factors. Proc., NCEER Workshop on
Evaluation of Liquefaction Resistance of Soils, Nat. Ctr. for Earthquake Engrg. Res., State
Univ. of New York at Buffalo, 149-165.
Youd, T. L., I. M. Idriss, R. D. Andrus, I. Arango, G. Castro, J. T. Christian, R. Dobry, W. D. L. Finn, L.

F. Harder Jr., M. E. Hynes, K. Ishihara, J. P. Koester, S. S. C. Liao, W. F. Marcuson-III, G. R.
Martin, J. K. Mitchell, Y. Moriwaki, M. S. Power, P. K. Robertson, R. B. Seed, and K. H. Stokoe-
II (2001). Liquefaction Resistance of Soils: Summary Report from the 1996 NCEER and
1998 NCEER/NSF Workshops on Evaluation of Liquefaction Resistance of Soils, ASCE, J.
Geotech. Geoenviron. Engg. 127, 817-833.
Youngs, R. R., and K. J. Coppersmith (1985). Implications of fault slip rates and earthquake
recurrence models to probabilistic seismic hazard estimates, Bull. Seismol. Soc. Am. 75(4),
939-964.
Youngs, R., S. Day, and J. Stevens (1988). Near field ground motions on rock for large subduction
zone earthquakes, in Earthquake Engineering and Soil Dynamics II, American Society of
Civil Engineers Geotech. Special Publication No. 20, 445-462.
Youngs, R. R., S. J. Chiou, W. J. Silva, and J. R. Humphrey (1997). Strong ground motion relationships
for subduction earthquakes, Seismol. Res. Lett. 68, 58-73.
Zhao, J. X., K. Irikura, and J. Zhang (2006). An empirical site-classification method for strong-
motion stations in Japan using H/V response spectral ratio, Bull. Seismol. Soc. Am. 96, 914-
925.
www.cygnusadvertising.in

Manual PDF

Uploaded by

Copyright:

Available Formats

Manual PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Manual PDF

Uploaded by

Copyright:

Available Formats

SECRETARY

DR. SHAILESH NAYAK GOVERNMENT OF INDIA

The seismic vulnerability in urban areas is rapidly increasing due to uncontrolled

Tel. : 00-91-11-24360874, 24362548 E-mail : secretary@moes.gov.in Fax No. : 00-91-11-24362644/24360336

Even though Seismic Microzonation is a burning issue in Earthquake Disaster

Prof. Sankar Kumar Nath

Preamble Prof. S. K. Nath, IIT Kharagpur

2. Seismic Microzonation : Principles and Protocol 2

4. Microzonation Works carried out in India 4

5. Definition of Microzonation Problem 6

6. Objectives and Deliverables of a Typical Seismic

9. Microzonation Levels with Scale 18

10. Regional Assessments 18

11. Seismicity of the Study Area and Estimation of its

12. Local Specific Assessments 24

13. In-Situ Field Testing for Site Characterization 26

14. Routine Geotechnical Laboratory Tests 39

15. Eurocode-8 and NEHRP 41

16. Assessment of Dynamic Properties based on

17. Spectral Shapes 45

19. Methods of Estimating Site Effects 48

20. Site Effects and Seismic Codes 57

21. 1D Ground Response Analysis using Equivalent Linear Approach 58

22. Predominant Frequency 62

23. Liquefaction Hazard Assessment 63

24. Liquefaction Susceptibility 64

25. Earthquake Induced Landslides 82

26. Simulation of Strong Ground Motion 86

27. Ground Motion Prediction Equations 97

28. Seismic Hazard Analysis 115

29. Composite Hazard Evaluation 119

30. Earthquake Hazard Parameters 120

31. GIS Integration Logic 132

32. Seismic Vulnerability and Risk Assessment 137

33. Concluding Remarks 139

Work Sheet for Seismic Microzonation 141

Appendix I : A Unified Earthquake Catalogue for South Asia

Appendix II : A Seismogenic Source Framework for the Indian

Appendix III : Probabilistic Seismic Hazard Assessment of India 177

Figure 1: Steps in the mitigation of earthquake risk (after McGuire, 2004).

2. SEISMIC MICROZONATION : PRINCIPLES AND PROTOCOL

Typically, a seismic microzonation evaluation consists of two distinct parts:

1. Single attribute characterization based on ambient noise/microtremor derived

4. Estimation of site response and predominant frequency from earthquake recordings.

Table 1: Global Scenario Approaches

Estimation of predominant frequency from Bam City (Motamed et al., 2007),

Site response estimation from earthquake Alaska (Nath et al., 2002a)

1-D geotechnical modeling of predominant Bucharest City (Ehret et al., 2004),

Predominant frequency and site City of Thessaloniki (Lachet et al., 1996)

4. MICROZONATION WORKS CARRIED OUT IN INDIA

Table 2 : Indian Scenario

Study Region Approaches

Guwahati Estimation of site response and predominant frequency from strong

Seismic scenario for deterministic hazard (Nath et al., 2008b).

Landslide hazard and soil liquefaction zonation, site classification,

Sikkim Estimation of site response and predominant frequency from strong

Deterministic seismic scenario and thematic integration based on

Delhi Geotechnical analysis, stochastic simulation of strong ground

Synthetic seismogram (Parvez et al., 2004).