1) Microzonation main Volume - Ministry Of Earth Sciences

Front cover page figure description:Map-Deterministic Hazard Index Seismic Microzonation Map (1:20,000) of Bangalore RegionBack cover page figure description:Vidhana SoudhaPrinted in Bangalore, India byM/s AQUILA Consultancy services, Srinagar, Bangalore, India.vinay2014@gmail.com; Mob: 9886311830.

REPORTONSEISMIC MICROZONATION OFBANGALORE URBAN CENTRESeismology Division,Ministry of Earth SciencesGovernment of IndiaNew Delhi -110 0032009MAIN VOLUMEI

III

PREPARED BY WORKING GROUPPARTICIPATING ORGANIZATIONSIndian Institute of Science(IISc), BangaloreNational Geophysical Research Institute (NGRI)-HyderabadIndian Institute of Technology, Kharagpur (IITKGP)India Meteorological Department (IMD), New DelhiGeological Survey of India, KolkataDisaster Mitigation and Management Group, KarnatakaCentre for Mathematical Modelling and ComputerSimulation (CMMACS), BangaloreSurvey of India, DeharadunCompiled and Edited by:Prof. T.G. Sitharam and Dr. P. AnbazhaganIISc, BangaloreV

EXECUTIVE SUMMARYSeismic hazard and microzonation of cities enable us to characterize the potential seismic areas that need tobe taken into account when designing new structures or retrofitting the existing one. Study of seismichazard and preparation of geotechnical microzonation maps will provide an effective solution for cityplanning and input to earthquake resistant design of structures in an area. In the present study an attempt ismade to characterize the site and to study the seismic hazard analysis considering the local site effects andto develop microzonation maps for Bangalore. Seismic hazard analysis with local site and microzonation ofBangalore is addressed in three parts: The first part provides estimation of seismic hazard usingseismotectonic and geological information. From hazard studies it found that expected peak groundacceleration (PGA) for Bangalore at rock level is about 0.15g using deterministic approach (0.136g usingsynthetic ground motion model). Seismic hazard parameter ‘b’ value is estimated as 0.87, which is slightlyhigher than the published values, indicating high activity of the region. The uncertainty involved in thehazard analysis is quantified using probabilistic approach which gives similar peak ground acceleration of0.121g. Second part deals with site characterization using geotechnical and shallow geophysical techniqueof multichannel analysis of surface wave (MASW). Based on soil average shear wave velocity and 30maverage shear wave velocity, as per National Earthquake Hazards Reduction Program (NEHRP) andInternational Building Code (IBC), Bangalore can be classified as “Site class D”. Correlation betweencorrected standard penetration test (SPT) ‘N’ values and measured shear wave velocity has been developed.There were over 150 lakes in Bangalore Mahanagar Palike, though most of them are dried up due toerosion and encroachments leaving only 64 at present in an area of 220 km 2 . The classification as site classD and filled up soils in encroachments emphasize the need to study site effects. In the last part, local siteeffects are assessed by carrying one-dimensional (1-D) ground response analysis (using the programSHAKE 2000) using both borehole SPT data and shear wave velocity survey data within an area of 220km 2 . 1-D site response study shows that the amplification factor is in the range of 1 to 4.7 and predominantfrequency varies from 2 Hz to 12Hz. Ground response parameters evaluated using MASW data showslightly lower values when compared to the parameters obtained using SPT data. Correlation betweencorrected SPT ‘N’ values and low strain shear modulus has been developed. Further, field experimentsusing microtremor studies have also been carried out (jointly with NGRI) for evaluation of predominantfrequency of the soil columns. Predominant frequency obtained from all the three methods matches verywell. Further, Seed and Idriss simplified approach has been adopted to evaluate the liquefactionsusceptibility and liquefaction resistance assessment. Liquefaction study shows that Bangalore is safeagainst liquefaction except at few locations where the overburden is sandy silt with presence of shallowwater table. From this study about 31 Microzonation maps have been prepared for Bangalore city on a scaleof 1:20000 and correlation between SPT corrected N values to measured shear wave velocity and low strainshear modulus have also been generated. The rocklevel PGA map for 2 ½ and 10 % probability ofexcedence in 50 years corresponding to the return period of 2475 and 475 years has been presented. Thestudy brings out that both the probabilistic and deterministic approaches lead to similar answer andprovides an insight to the seismic hazard assessment.VI

CONTENTSTITLE PAGEIFOREWORDIIPREAMBLEIIIPARTICIPATING ORGANIZATIONSVEXECUTIVE SUMMARYVICONTENTSVIIChapter 1Introduction and Background Information1.1 General Background 11.21.3Need for the StudyObjectives and scope of the Study121.4 Seismic Hazard and Microzonation Terminologies 21.4.1 Macrozonation and Microzonation 3Chapter 2Study Area and Methodology2.1 Introduction 42.2 Study Area 42.2.1 Location of study area 42.2.2 Background of Bangalore 42.2.3 Geology of the Seismic Study Area 52.2.4 Geomorphology of the Seismic Study Area 82.3 Methodology 102.3.1 Seismic Hazard and Microzonation 102.3.2 Flow Chart for Seismic Hazard Analysis and Microzonation 112.3.3 Methodology adopted for Deterministic Seismic Hazard Analysis 112.3.4 Methodology adopted for Probabilistic Seismic Hazard Analysis 112.3.5 Methodology adopted for Site Characterization 132.3.6 Methodology adopted for Local Site Effects and Site ResponseStudy2.3.7 Methodology adopted for Liquefaction Analysis1416Chapter 3Seismic Hazard Analysis3.1 Introduction 193.2 Seismic Study Area 203.3 Seismic Sources 203.4 Seismicity of the Study Area 213.5 Preparation of Seismotectonic Map of Bangalore 273.6 Geotechnical Data for Rock Depth 273.7 Deterministic Seismic Hazard Analysis 293.7.1 Rock PGA Using Past Earthquake 303.7.2 Rock PGA using Subsurface Fault Rupture Relationship 313.8 Synthetic Earthquake Model 353.9 Development of Rock Level PGA Map 383.10 Deterministic Seismic Hazard Analysis 383.11 Data Completeness Analysis and Seismic Hazard Parameter 403.11.1 Analysis of Completeness of Seismic Data 403.11.2 Seismic Hazard Parameters 443.12 Comparative Analysis 463.13 Seismogenic Sources 473.14 Probabilistic Seismic Hazard Analysis 483.14.1 Regional Recurrence Model 493.14.2 Deaggregation 503.14.3 Uncertainty in the Hypocentral Distance 513.14.4 Probability of Ground Motion52VII

CONTENTS3.14.5 Hazard Curves 533.15 Summary from Probabilistic Approach 57Chapter 4Site Characterization Using Geotechnical and Geophysical Technique4.1 Site Characterization using Geotechnical Data 584.2 Preparation of Base Map 584.3 Geotechnical Data and Distribution 604.4 Subsurface 3-D model 614.5 Artificial Neural Network (ANN) for rock depth 654.6 Corrections Applied for SPT “N” Values 654.6.1 Correction for Overburden Pressure 654.6.2 Correction for hammer energy ratio 654.6.3 Other correction factors 664.7 Summary from Geotechnical Technique 684.8 Site Characterization using Multichannel Analysis of Surface Wave (MASW) Survey 684.9 Testing Programme 684.10 MASW Testing 694.10.1 Dispersion Curves 724.10.2 One-Dimensional Shear Wave Velocity Profiles 724.11 Mapping of Subsurface Layers using Two-Dimensional “Vs” Profiling 744.12 Average Shear Wave Velocity 764.13 Shear Wave Velocity Distribution in Bangalore 774.14 Calculation of Dynamic Properties 784.15 Correlation between (N 1 ) 60cs and V s 784.16 Summary From MASW Technique 88Chapter 5Local Site Effects and Site Response Analysis5.1 Introduction 895.2 Need for the Study 895.3 Synthetic Ground Motions and Peak Acceleration Map 905.4 1-D Ground Response Analysis Using Equivalent Linear Approach 935.4.1 Geotechnical Data and Rock Depth 945.5 Ground Response Analysis based on SPT data 955.5.1 Spatial Variation of Site Effects in Bangalore 985.5.2 Peak Ground Acceleration 985.5.3 Amplification Factor 985.5.4 Amplification versus Overburden Thickness 1025.5.5 Period of Soil Columns 1025.5.6 Peak Spectral Parameters 1025.6 Ground Response using Shear Wave Velocity 1125.6.1 Site Response Results based on MASW Data 1135.6.2 Comparison of Predominant frequency obtained using SPT and 1165.7MASW DataCorrelation between (N 1 ) 60cs and G max 1165.8 Site response Using Micro Tremor Studies 1185.8.1 Instrument and Methodology 1185.8.2 Testing Locations 1205.8.3 Microtremor Survey Results 1205.8.4 Comparison of Predominant Frequency from Site Response Study 123using Vs and Microtremor5.9 Natural Frequency and Period of Typical Structure 1245.10 Probabilistic Seismic Hazard Analysis with Local Site Effects1255.11 Conclusions127Chapter 6Liquefaction Hazard Assessment6.1 Introduction6.2 Liquefaction Susceptibility Map130130VIII

CONTENTSCONTENTSCONTENTS6.3 Factor of Safety against Liquefaction Assessment 1306.3.1 Peak Ground Acceleration 1316.3.2 Cyclic Stress Ratio(CSR) 1326.3.3 Cyclic Resistance Ratio(CRR) 1326.3.4 Magnitude Scaling Factor(MSF) 1326.3.5 Factor of Safety Calculation 1336.4 Liquefaction Hazard Map 1346.5 Results and Discussion 1346.5.1 Cyclic Triaxial Experiments on Undisturbed Soil Samples 1406.6 Concluding Remarks 142Chapter 7Integration of hazard maps on GIS Platform7.1 Introduction 1437.2 Geographic Information System (GIS) 1437.2.1 GIS Integration Logic 1447.3 Earthquake hazard parameters 1457.3.1 Geomorphological Attributes 1457.3.2 Seismological Attributes 1507.4 Integration of different layers (Themes) 1537.5 Summary 160Chapter 8Summary and Conclusions8.1 Summary 1618.2 Deterministic Seismic Hazard Analysis 1618.3 Probabilistic Seismic Hazard Analysis 1618.4 Site characterization using Geotechnical Borehole Data 1628.5 Site Characterization using Multichannel Analysis of 162Surface Wave (MASW) Survey8.6 Local Site Effects and Site Response 1638.7 Liquefaction Hazard Assessment 1638.8 Integration of hazard maps on GIS Platform8.9 Conclusions8.10 Recommendations for Further StudyREFERENCES164164165166IX

CHAPTER 1INTRODUCTION AND BACKGROUND INFORMATION1.1 GENERAL BACKGROUNDEarthquakes are one of the most devastating among the various natural hazards. The hazards associated withearthquakes are referred to as seismic hazards. The practice of earthquake engineering involves theidentification and mitigation of seismic hazards. The basis of earthquake geotechnical engineering andmicrozonation is to model the rupture mechanism at the source of an earthquake, evaluate the propagation ofwaves through the earth to the top of bed rock, determine the effect of local soil profile and thus develop ahazard map indicating the vulnerability of the area to potential seismic hazard. In short, the geotechnicalengineer is responsible for providing the structural engineer with appropriate site-specific design groundmotions for earthquake resistant design of structures. Many earthquakes in past have left many lessons to belearned which are very essential to plan infrastructure and even to mitigate such calamities in future. India hasbeen facing threat from earthquakes since ancient times. The planning of cities becomes an important issue ascities are growing and expanding due to increasing migration of people to cities. The role of geological andgeotechnical data is becoming very important in the planning of city urban infrastructure, which can recognize,control and prevent geological hazards (Bell et al., 1987; Legget, 1987; Hake, 1987; Rau, 1994; Dai et al., 1994,2001; Van Rooy and Stiff, 2001). Study of seismic hazard and preparation of geotechnical microzonation mapswill provide an effective solution for city planning. Seismic hazard and microzonation of cities enable tocharacterize potential seismic vulnerability/risk that need to be taken into account when designing newstructures or retrofitting existing ones. Even this will help in designing buried lifelines such as tunnels, waterand sewage lines, gas and oil lines, and power and communication lines.Seismicity of an area is the basic issue to be examined in seismic hazard analysis for evaluating seismic risk forthe purpose of microzonation planning of urban centres. Detailed knowledge of active faults and lineaments andassociated seismicity is required to quantify seismic hazard and risk. Indian peninsular shield, which was onceconsidered to be seismically stable, has shown that it is becoming active. In the last three decades largeearthquakes have caused massive loss of lives and extensive physical destruction throughout the world(Armenia, 1988; Iran, 1990; US, 1994; Japan, 1995; Turkey, 1999; Taiwan, 1999, India 2001, Sumatra 2004,Pakistan, 2005). In India, the recent destructive earthquakes are Killari (1993), Jabalpur (1997), Bhuj (2001),Sumatra (2004) and Indo-Pakistan (2005). Seismic activity of India is clearly evident from these recentearthquakes within the intra plate and also along the boundaries of Indo-Australian Plate and Eurasian Plate.Many researchers address the intra plate earthquakes and seismicity of South India (Purnachandra Rao, 1999;Ramalingeswara Rao, 2000; Iyengar and Raghukanth, 2004).Very preliminary process of reducing the effects of earthquake is by assessing the hazard itself. The seismichazard analysis is concerned with estimate of strong ground motion parameters at a site for the purpose ofearthquake resistant design. The deterministic and the probabilistic seismic hazard analysis are the two basicmethodologies used for this purpose. The complete seismic hazard analysis and microzonation of Bangalore hasbeen carried out using extensive experimental and theoretical work. As part of the national level microzonationprogramme, Department of Science and Technology, Govt. of India has initiated microzonation of Bangaloreregion.1.2 NEED FOR THE STUDYPeninsular India once believed to be a stable continent, has experienced many earthquakes, in particular, Laturearthquake on 30 th September 1993 (M 6.3), Jabalpur earthquake 22 nd May 1997(M 6.0) and Bhuj earthquake26th January 2001(M 7.9). These earthquakes have influenced the need for study of earthquakes and theireffects on Indian cities. As part of the national level microzonation programme, Department of Science andTechnology, Govt. of India has initiated microzonation of Bangalore region. The seismic hazard analysis andmicrozonation of Bangalore region is carried out as part of this project. Apart from this, the specific need tostudy the seismic hazard and microzonation of Bangalore is as follows.Bangalore is densely populated, economically and industrially important city in India. It is one of thefastest growing cities in Asia.The population of Bangalore city is over 6 million and Bangalore city is the fifth biggest city in India.Bangalore is the political capital of the state of Karnataka. Besides political activities, Bangalorepossesses many national laboratories, defense establishments, small and large-scale industries and1

Introduction and Background InformationInformation Technology Companies. Bangalore is also called as Silicon Valley of India/Science cityof India. These establishments have made Bangalore a very important and strategic city.1.3 OBJECTIVES AND SCOPE OF THE STUDYThe objective of this research is to study the seismic hazard and microzonation of Bangalore and develop theseismic hazard and microzonation procedures in Indian contest with the following topic of interests.Estimation of Seismic hazard in terms of deterministic and probabilistic approaches for Bangalore anddevelop the rock level peak ground acceleration map and Hazard curves.Development of synthetic ground models for Bangalore region.3-D subsurface modeling of the geotechnical data using standard penetration tests along with boreloginformation.Measurement of shear wave velocities and evaluation of dynamic properties of soil in Bangalore usingmultichannel analysis of surface wave (MASW) method.Generate correlation between corrected standard penetration test “N” values and measured shear wavevelocity.Study of local site effects and site response parameters of soil using the synthetic ground motion,standard penetration test borelogs and shear wave velocity from MASW.Study of site response of soil using micro tremor experiments.Evaluation of liquefaction potential of Bangalore soil using standard penetration test borelogs andmapping of liquefaction hazard.Preparation of microzonation map using Geographical Information System (GIS)1.4 SEISMIC HAZARD AND MICROZONATION TERMINOLOGIES Earthquake is a vibration of the earth's surface usually triggered by the release of stored strain energyin the earth crust along fault lines. This release causes movement in masses of rock and resulting shockwaves. Seismic hazard is the study of expected earthquake ground motions at any point on the earth. Faults are localized areas of weakness in the surface of the earth, sometimes the plate boundary itself.A fracture (crack) in the earth, where the two sides move past each other and the relative motion isparallel to the fracture. Lineament is a mappable simple or composite linear feature of a surface whose parts align in a straightor slightly curving relationship and that differs distinctly from the patterns of adjacent features. Active Lineament is lineament associated with earthquakes, it may be termed as fault. Seismotectonic sources are earthquake sources activated by the tectonic forces. Seismogenic sources are the faults and area sources have to be delineated describing the geometricdistribution of earthquake occurrence in the investigated area. Non-seismotectonic sources are earthquake sources related to the human activities, such as the fillingof large reservoirs. Focus is the point of origin of an earthquake. Epicenter is the vertical projection of focus in the ground surface. Epicenter Distance is the distance between the epicenter to the site or instrument. Focal distance is the distance between the focus to the site or instrument. Intraplate earthquake is an earthquake that occurs within a plate, as opposed to those occurring at aplate boundary. Earthquake Intensity is the qualitative assessment of earthquake, based on how strong earthquakefeels to the observer, qualitative assessment of the kinds of damage done by an earthquake, depends ondistance to earthquake & strength of earthquake. It is usually determined from the intensity of shakingand damage from the earthquake. Earthquake Magnitude is the quantitative estimate of earthquakes related to energy release.Quantitative measurement of the amount of energy released by an earthquake. Depends on the size ofthe fault that breaks and determined from Seismic Records. Microzonation is the process of sub division of region in to number of zones based on the earthquakeeffects in the local scale.2

Introduction and Background Information Microzonation mapping of seismic hazards can be expressed in relative or absolute terms, on an urbanblock-by-block scale, based on local conditions (such as soil types) that affect ground shaking levels orvulnerability to soil liquefaction. Intensity (I O ) is represented in the Modified Mercalli Intensity (MMI) scale which is commonly used inthe world by seismologists. Intensity ratings are expressed as Roman numerals between I at the low endand XII at the high end. Magnitude concept was introduced by Richter (1935) to provide an objective instrumental measure ofthe size of earthquakes. Magnitudes are evaluated from measured amplitude and period of seismicsignals recorded at a seismic station. For a given earthquake, the amplitude decreases with increasingdistance (due to attenuation of the signals) and a distance dependent correction is applied. Themagnitudes are classified based on different calculation procedures and the definition for differentmagnitudes scales followed in the report is given below and their relation with moment magnitude isalso given. Local magnitude (M L ) scale is referred as Richter magnitude scale. Charles Richter used Wood-Anderson Seismometer to define a magnitude scale for shallow and local (epicenteral distance less thanabout 600km) earthquakes in southern California (Richter 1935). This magnitude scale is the bestknown and most commonly used magnitude scale (Day, 2002). Surface wave magnitude (M S ) is calculated based on the amplitude of surface (Rayleigh) waves havinga period of about 20s (Gutenberg and Richter, 1956). The surface wave magnitude uses the maximumground displacement, which can be obtained from any type of seismograph. But in local magnitude themaximum trace amplitude of Wood-Anderson seismograph is used. This magnitude scale can be usedfor moderate to large earthquakes having a shallow focal depth but the seismograph should be 1000kmaway from the epicenter (Day, 2002). Body wave magnitude (m b ) is used to measure the deep-focus earthquakes, which is based onamplitude of the first few cycles of P-waves (Gutenberg, 1945). Intraplate earthquake can berepresented by Body wave magnitude (m bLg ), which is estimated from the amplitude of one-second –period, higher –mode of Rayleigh waves (Kramer, 1996). Coda magnitude (M C ) is obtained from characteristics of backscattered waves (Aki, 1969) that followpassage of primary (unreflected) body and surface waves. Moment magnitude (Mw) is world wide more commonly used scale to determine the magnitude oflarge earthquakes. It represents the entire size of the earthquake and is calculated from seismic moment(Kanamori, 1977) and is not based on ground shaking levels.1.4.1 Macrozonation and MicrozonationSeismic zonation is usually carried out in two parts; one is macro level and another one micro level. For a largerarea like zonation of country or continent macro level is adopted. Macrozonation are carried out considering theseismicity, geology in lager scales without considering geotechnical aspects. But microzonation is carried out insmaller scale by considering regional seismicity, geology and local site conditions. This microzonation isleveled or graded based on scale of the study and details of geology and geotechnical inputs.3

2.1 INTRODUCTIONCHAPTER 2STUDY AREA AND METHODOLOGYRapidly growing cities with increasing population are most vulnerable to natural hazards due to agglomerationof the population at one place. Preparation of the geotechnical microzonation maps provides an effectivesolution to overcome the hazards to some extent. Seismic microzonation has been carried out to understand theeffects of earthquake generated ground motions due to local soil or/and man-made structures. The mainobjective of a microzonation study is to use the obtained variation of the selected parameters for land use andcity planning. Therefore it is very important that the selected microzonation parameters should be meaningfulfor city planners as well as for public officials. Ansal (2004) recommends that the national seismic zoning mapsare mostly at small scale level (1:1,000,000 or less) and are mostly based on seismic source zones defined atsimilar scales. The seismic microzonation for a town requires 1:5,000 or even 1:1,000 scale studies and needs tobe based on seismic hazard studies at similar scales. The general trend in conventional microzonation studies inIndia was to simplify the applied methodology by adopting the macrozonation seismic hazard maps as theprimary source to estimate the earthquake hazard. In addition, due to lack of sufficient geological andgeotechnical data, a second simplification is to define the site conditions with respect to local geological units.Even though these two simplifications may appear logical, to many engineers and scientists they are the mainsource of incorrectness in any microzonation study. One possible reason for this weakness and multiplicity inthe seismic microzonation studies is that in most cases, seismic zonation studies, whether they are at macro ormicro scale are generally conducted by earth scientists (Seismologists or Geologist). But seismic microzonationrequires an essential input from civil engineering, especially in the field of geotechnical engineering (DRM,2003). Keeping these things in mind, in this study an attempt has been made to develop microzonationmethodology for Indian context with an example case study of Bangalore city.2.2 STUDY AREAThe section of this chapter gives a brief review of the development of study area (Bangalore city) from the timeit was founded. The background, geology and Geomorphology of the study area are discussed in the subsequentsections.2.2.1 Location of Study AreaThe Bangalore city covers an area of approximately 696.17 Sq. km (Greater Bangalore). The area of study islimited to Bangalore Metropolis area (Bangalore Mahanagar Palike) of about 220 sq.km. Bangalore is situatedon a latitude of 12 o 58' North and longitude of 77 o 36' East and is at an average altitude of around 910 m abovemean sea level (MSL). It is the principal administrative, industrial, commercial, educational and cultural capitalof Karnataka state and lies in the South- Western part of India (see Figure 2.1). Bangalore city is the fastestgrown city and fifth biggest city in India. Besides political activities, Bangalore possesses many nationallaboratories, defence establishments, small and large-scale industries and Information Technology Companies.These establishments have made Bangalore a very important and strategic city. It experiences temperate andsalubrious climate and an annual rainfall of around 940 mm. There were over 150 lakes, though most of themare dried up due to erosion and encroachments leaving only 64 at present in an area of 220 sq km. These tankswere once distributed throughout the city for better water supply but presently in a dried up condition and theresidual silt and silty sand form thick deposits over which buildings/structures have been built, which aresusceptible for site amplification. Study area covers about 220 square kilometers in Bangalore city.2.2.2 Background of BangaloreThis section provides a brief history about the development of Bangalore to a cosmopolitan city till today.Bangalore in its present context was founded when a mud fort was built by Kempe Gowda I in 1537 AD andmade it his capital. There has been a consistent development of the city since then. Bangalore has been thecapital of4

Study Area and MethodologyBangalore MunicipalCorporationIndiaSeismicStudy AreaFigure 2.1: Study area with India mapKarnataka state from 1947 and several public sector industries were set up from 1940-1970 transforming it intoa science and technology centre. By 1961, Bangalore had become the 6 th largest city in India with a populationof 1,207,000. Between 1971 and 1981, Bangalore’s growth rate was 76%, the fastest in Asia. In 1889, openspace in Bangalore was four times the built up area and by 1980 the built up area was four times the open spacearea. This indicates the rapid growth of infrastructure in the city. By 1988 the electronic city had been developedand Bangalore emerged as India’s software capital. Consequently, there was a huge construction boom in the1990’s. Blessed with a strong educational and technological base and agreeable climate, Bangalore is stillwitnessing a tremendous growth in industry, trade and commerce leading to a rapid growth of the city and largescale urbanization. The population of Bangalore region is over 6 million. Because of density of population,mushrooming of buildings of all kinds from mud buildings to RCC framed structures and steel construction and,improper and low quality construction practice, Bangalore is vulnerable even against average earthquakes(Sitharam et al, 2006). As per BIS 1893 (2002) Bangalore has been upgraded to Zone II from Zone I in theseismic zonation map. Recent studies by Ganesha Raj (2001), Sitharam and Anbazhagan (2007) and Sitharam etal. (2006) suggested that Bangalore need to be upgraded from seismic zone II to zone III based on the regionalseismotectonic details and hazard analysis.2.2.3 Geology of the Seismic Study AreaFor the purpose of seismic hazard analysis, circular area with a radius of about 350km around Bangalore hasbeen selected for the seismicity study as per Regulatory Guide 1.165(1997). Regional geological andseismological details for the Bangalore city were collected from literature review, study of maps and remotesensing data. The study area marked in the India map is shown in Figure 2.1. The geological study area has the5Scale 1:20000

Study Area and Methodologycenter point as Bangalore city with a circular area of 350km radius (which covers the latitude 9.8 o N to 16.2 o Nand longitude of 74.5 o E to 80.7 o E). Geology of the study area presented in the Seismotectonic Atlas of India(SEISAT, 2000) published by Geological Survey of India, is used here. Geological formation of the study areais similar to the Indian Peninsula, which is geologically considered as one of the oldest land masses of theearth’s crust. Tectonic/Geological map of the study area is shown in Figure 2.2. Most of the study area isclassified as Gneissic complex/Gneissic granulite with major inoculation of greenstone and allied supracrustalbelt, which are believed to have occurred between 3400 to 3000 million years ago giving rise to an extensivegroup of grey gneisses designated as the “older gneiss complex”. These gneisses act as the basement for awidespread belt of schist’s. The younger group of gneissic rocks mostly of granodiomitic and granitiacomposition is found in the eastern part of the state, representing remobilized parts of an older crust withabundant additions of newer granite material, for which the name “younger gneiss complex” was given(Radhakrishnan and Vaidyanadhan, 1997). The rocks in this group range in age from 2700 to 2500 millionyears. The oldest rocks of Karnataka are the Sargur Group of rocks, which is followed by Peninsular GneissicComplex, Dharwar Super Group, Closepet Granite, Kaladgi, Bhima’s, and Deccan Traps; these are furtherfollowed by laterite and alluvium. The Peninsular Gneissic Complex is the dominant unit and covers about twothirdsof the area, which includes granites, gneisses and magmatites. Sargur group comprises ultramorphicrocks, amphibolites, Quartzite banded magnetites- quartzite occurring as small bands and lenses within themagmatites and gneisses. The bed rocks essentially consist of granites and gneisses intruded by number of basicdykes. The geology deposits close to the eastern and western side of the geological study area has coastline withalluvial fill in pericratonic rift. The soils of Bangalore district consist of red laterite and red fine loamy to clayeysoils with a wide variation of overburden thickness.The Peninsular Gneissic Complex occupies major part of the study area. Further, findings from geologists haveshown that in the southern Karnataka the faults/ lineaments are reactivated. Valdiya (1998) highlighted that theseismic activity is generally confined to linear belts related to transcurrent and terrain-bounding faults and shearzones, implying that the Precambrian faults are being reactivated in the present time based on purely geologicalstudies. The morphology of Karnataka shows that there are a series of water falls, cascades and rabid along theCauvery river, particularly between Sivasamudram in Karnataka and Mettur in Tamil Nadu. This is attributed toreactivation of Precambrian faults across part of the old course here and lateral displacement of the upliftedblocks, giving rise to change in the course of the river as shown in Figure 2.3 (Valdiya, 1998). Figure 2.3 showsthe active faults speculated at present by Valdiya (1998) in south of Bangalore on either side within 100 kms.Similarly, in the north, the Arkavathi River that follows a remarkably straight fault valley in the Manchenabele-Aganahalli-Ramagiri tract is shown in Figure 2.4. Valdiya (1998) highlighted that the recent uplift is in the orderof 7 to 10 m on the eastern side formed gully erosion on the Manchenabele reservoir area corroborating to therecent movement of the faults. Figure 2.4 shows faults and lineaments identified by Valdiya (1998) close toBangalore at a distance of about 20 kms to 50kms; having a length varying from about 35 kms to 90 kms. Largenumber of earthquakes with different magnitudes has occurred very often in this region (Ramalingeswara Raoand Sitapathi Rao, 1984; Bansal and Gupta, 1998). In recent years much of the seismic activity in the state ofKarnataka has been in the south, in the Mysore-Bangalore region (Ganesha Raj and Nijagunappa, 2004).Seismotectonic map from Project Vasundhara (1994) also shows that there are active faults that triggeredearthquake magnitude of 2 to 4 close to Bangalore. Ganesha Raj and Nijagunappa (2004) have also highlightedthe need to upgrade the seismic zonation of Karnataka; particularly the areas surrounding Bangalore, Mandyaand Kolar to zone III rather than the current zone II as these areas are quite active, based on the analysis carriedout using remote sensing data and neotectonic activity in the area.6

Study Area and MethodologyBangalore12 o 58’N; 77 o 36’EScale 1:5,00,000Greenstone and allied supracrustal beltGneissic Complex / Gnesis GranuliteTerrestrial facies cover in linear grabenAlluvial fill in pericratonic riftComposite Batholithic ComplexAlkaline Plutonic ComplexShelf facies cover intracratonic sagFigure 2.2: Tectonic/Geologic map around Bangalore/study area.Epiric sea / marginal overlap cover7

Study Area and MethodologyFigure 2.3: Active faults present close to Bangalore (after Valdiya, 1998)Bangalore city lies over a hard and moderately dense Gneissic basement dated back to the Archean era (2500-3500 mya). A large granitic intrusion in the south central part of the city extends from the Golf Course in thenorth central to Vasantpur (VV Nagar) in the south of the city (almost 13 km in length) and on an average 4 kmfrom east to west along the way. A migmatite intrusion formed within the granitic one extends forapproximately 7.3 kms running parallel with Krishna Rajendra Road/ Kanakpura Road from Puttanna ChettyRoad in Chamrajpet till Bikaspura Road in the south. A 2.25km Quatrzite formation is found in Jahahalli East.Dike swarms are seen around the western outskirts of the city (west of the Outer Ring Road), most of themstriking approximately N15 o E. However random east west trending ones are also seen. They appear to strikeparallel to the strike of the vertical foliation of the country rock at that area. These basic intrusive which markthe close of the Archean era (Lower Proterozoic; 1600-2500 mya) mainly constitute of hard massive rocks suchas Gabbro, Dolerite, Norite and Pyroxenite.2.2.4 Geomorphology of the Seismic Study AreaBangalore city lies within the South Pennar Basin. There are over 150 freshwater lakes within the city which arelinked to one another dendritically. The Vrishabhavathi, a minor tributary of the Arkavati enters the city almostdiagonally from the Southwest (Kengeri). On entering, it branches off giving rise to the Nagarbhavi Thorai atthe intersection of the Mysore Road and the Bangalore University Road. The river Arkavati branches north fromCauvery, most probably due to some structural control imposed by presently active faults and other North southtrending lineaments in its course. The river Ponnaiyar enters the city from the south east and branches into two,one stream terminates in the Bellandur Lake, where as the other continues northwestwards.Bangalore city is subject to a moderate annual soil erosion rate of 10 Mg/ha. The basic geomorphology of thecity comprises of a central Denudational Plateau and Pediment (towards the west) with flat valleys that areformed by the present drainage patterns (See Figure 2.5). The central Denudational Plateau is almost void of anytopology and the erosion and transportation of sediments carried out by the drainage network gives rise to thelateritic clayey alluvium seen throughout the central area of the city. The Pediment/Pediplain is a low relief areathat abruptly joins the plateau. The area might have uplifted along active lineaments and been eroded by theriver Arkavati and its sub tributaries. The resulting alluvial fan deposits have been deposited or transportedalong the waterways.8

Study Area and Methodology17° 30'Mindupur13°30'DevarayadurgaTumkurMadhugiri1158Bondarahalli125510391123Gouribidanur290Makalidurga 144511951023DodBallapur1211105913461009BagepalliGudibanda9301466 ChikballapurNandidurgaPapaghni R.AmbadurgaChintamaniBamangarhSrinivaspur13°30'1386ShivagangaPalacolakeSolurSavandurgaKumudvati1226Mahadevapura953921Devanhalli1017935HoskoteBANGALORE9961127AntaragangaKolarKendattibettaPalar R.900 mMulbagal13°MaddurSivasamudramKaveriKollegalPalacolakeFalls1033ReservoirKanakpurArkavathi979ManchanbetePalacolakeHosadurgaSangam1049Hosdurga113513631303 KaveriPalacolake10091040 1436Raudalli11191064Anekal151410261013Devarahalli994Pinakini R.HosurTottikereDenknkota13951290Chinari R.HogenakalFalls 40mTectonic hill> > 450 mup to 1750 mPalacolakeFaultLineamentAreaillustratedelsewhere0 20 km13°Figure 2.4: Active faults present in southeast of Karnataka (after Valdiya, 1998).9

Study Area and Methodology2.3 METHODOLOGYFigure 2.5: Geomorphology of Bangalore areaThe earthquake damage basically depends on three groups of factors: earthquake source and path characteristics,local geological and geotechnical site conditions, structural design and construction features. Seismicmicrozonation should address the assessment of the first two groups of factors. In general terms, seismicmicrozonation is the process of estimating the response of soil layers for earthquake excitations and thus thevariation of earthquake characteristics is represented on the ground surface. Seismic microzonation is the initialphase of earthquake risk mitigation and requires multidisciplinary approach with major contributions fromgeology, seismology, geotechnical and structural engineering (Ansal, 2004).2.3.1 Seismic Hazard and MicrozonationSeismic Microzonation falls into the category of “applied research”. That is why it needs to be upgraded andrevised based on the latest information. “Microzonation” is the subdivision of a region into zones that haverelatively similar exposure to various earthquake related effects. This exercise is similar to the macro levelhazard evaluation but requires more rigorous input about the site specific geological conditions, groundresponses to earthquake motions and their effects on the safety of the constructions taking into consideration thedesign aspects of the buildings, ground conditions which would enhance the earthquake effects like theliquefaction of soil, the ground water conditions and the static and dynamic characteristics of foundations or ofstability of slopes in the hilly terrain” –DST Expert Group on Microzonation of Delhi Chaired by Dr. A.S. Aryain 1998 and the definition was endorsed by the DST subcommittee on Microzonation, chaired by Mr. P.K.Narula in 2001. For the present investigation the seismic microzonation has been subdivided into three majoritems:1) Evaluation of the expected input motion2) Local Site effects and ground Response analysis3) Preparation of microzonation maps.The microzonation is graded based on the scale of the investigation and details of the study carried out. Thetechnical committee on earthquake geotechnical engineering (TC4) of the International society of soilmechanics and foundation engineering (TC4-ISSMGE 1993) states that the first grade (Level I) map can beprepared with scale of 1:1,000,000 – 1:50,000 and the ground motion was assessed based on the historicalearthquakes and existing information of geological and geomorphological maps. If the scale of the mapping is10

Study Area and Methodology1:100,000-1:10,000 and ground motion is assessed based on the microtremor and simplified geotechnical studiesthen it is called second grade (Level II) map. In the third grade (Level III) map ground motion has been assessedbased on the complete geotechnical investigations and ground response analysis with a scale of 1:25,000-1:5,000.Expert groups chaired by Dr. N.C. Nigam and later on by Dr. A.S. Arya recommended seismic microzonationmapping on a scale of 1:25000 for India. The present investigation was carried out with a scale of 1:20,000 andground motion are arrived based on the detailed geotechnical/geophysical investigations and ground responseanalysis. The methodology followed for the seismic hazard and microzonation in this research will bediscussed in the next section.2.3.2 Flow Chart for Seismic Hazard Analysis and MicrozonationEven though the step involved seismic hazard analysis and microzonation grouped in to three major groups asdiscussed in previous section, it need to adopt step by step procedure to arrive at the final map of microzonation.The steps followed in seismic hazard and microzonation of Bangalore in the present investigation is illustratedin the form of a flow chart in Figure 2.6. The first step illustrates the assessment of the expected ground motionusing the deterministic and probabilistic seismic hazard analysis which is discussed in the section 2.3.3 and2.3.4. The site characterized for the study area at local scale of 1:20,000 using geotechnical and shallowsubsurface geophysical data are discussed in the step 2.3.5. Third step is the study of local site effects using firstand second part output data and producing the ground level hazard parameters; the methodology is discussed inthe section 2.3.6. Fourth step is the assessment of liquefaction potential in terms of factor of safety againstliquefaction. Steps involved in this step discussed in section 2.3.7. As the ground is fairly flat in most of the areaexcept in north and northwestern of Bangalore, landslide possibility is considered as remote. Finallymicrozonation maps are prepared in terms of ground motion parameters and factor safety against liquefaction.2.3.3 Methodology adopted for Deterministic Seismic Hazard AnalysisThe methodology followed for DSHA described as four steps and illustrated in Figure 2.7:1. Source characterization, which includes identification and characterization of all earthquake sourceswhich may cause significant ground motion in the study area.2. Selection of the shortest distance between the source and the site/area of interest.3. Selection of controlling earthquake i.e. the earthquake that is expected to produce the strongest level ofshaking.4. Defining the hazard at the site formally in terms of the ground motions produced at the site bycontrolling earthquake.2.3.4 Methodology adopted for Probabilistic Seismic Hazard AnalysisThe PSHA procedure can also be described in four steps as illustrated in Figure 2.81. Identification of earthquake sources such as active faults, which may affect the study area. Characterizethe probability distribution of potential rupture locations within the source.2. Characterization of the seismicity of each source zone using a recurrence relationship, which specifiesthe average rate at which an earthquake of some size will be exceeded (recurrence relationship).3. Estimation of the ground motion produced at the site by earthquakes of any possible size occurring atany possible point in each source zone using predictive relationships.4. Obtaining the probability that the ground motion parameter will be exceeded during a particular period.11

Study Area and MethodologyInputo Geology datao Seismology datao Seismotectonic datao Deep Geophysical datao Remote sensing datao Regional Attenuation lawo Geotechnical datao Shallow Geophysicaldatao Soil Mappingo Rock motion datao Soil Datao Dynamic Propertieso Experimental Study-Microtremoro Ground PGAo Magnitude of EQo Soil properties withcorrected “N” valueo Experimental studieso Geology andSeismologyo Rock deptho Soil characterizationo Response resultso Liquefaction resultsSeismic Hazard AnalysisDeterministicSite CharacterizationTheoreticalSite ResponseProbabilisticExperimentalLiquefaction AssessmentIntegration of HazardsOutput Maximum Credible Earthquake Vulnerable Sources Synthetic Ground Motions Hazard parameters Rock level Peak GroundAcceleration maps Hazard curves Rock depth Mapping Subsurface Models 3-D Borehole models SPT ‘N’ Corrections Vs Mapping Vs 30 Mapping (N 1 ) 60 versus Vs Relations Amplification Maps Ground Peak Acceleration map Period of soil column map Spectral acceleration fordifferent frequency Response spectrum Comparative study (N 1 ) 60 versus G max Relations Liquefaction susceptibilitymap Factor of safety Table Factor of safety map Liquefaction mapping Microzonation maps Hazard Map Data for Vulnerability Study Data for Risk analysisFigure 2.6: Flow chart for seismic hazard and microzonation12

Study Area and MethodologySource 1Source 3SiteR 3M 1M 3R 1R 2M 2Source 2STEP 1 STEP 2Identification of seismic sourcesShortest distance from the sourceGround motionparameter, YM 3M 1M 2.. .ControllingearthquakeY={}Y 1Y 2..Y NR 3R 2R 1DistanceSTEP 3PGA from Attenuation relation for each sourceSTEP 4PGA for controlling EQFigure 2.7: Different steps for deterministic seismic hazard analysis (after Kramer, 1996).2.3.5 Methodology adopted for Site CharacterizationA complete site characterization is essential for the seismic site classification and site response studies, both ofthem can be used together for seismic microzonation. Site Characterization should include an evaluation ofsubsurface features, sub surface material types, subsurface material properties and buried/hollow structures todetermine whether the site is safe against earthquake effects. Site characterization should provide data on thefollowing:‣ Site description and location‣ Geotechnical data‣ Soil conditions‣ Geological data‣ Hydrogeology/ ground water data‣ Aquifer or permeable characteristics‣ Presence and distribution of contaminants (if any)‣ Climatic conditions (if needed)Assessment of available data should include an analysis of the sufficiency and validity of the data in relation tothe proposed application/ study.As part of the site characterization, experimental data should be collected, interpolated and represented in theform of maps. The representation maps can be further used for the site classification and seismic studies. Herean attempt has been made to characterize the Bangalore site using Geotechnical and Geophysical experimental13

Study Area and Methodologydata. About 850 collected geotechnical borehole information with standard penetration “N” values data and 58geophysical data of multichannel analysis of surface wave field test results are used for Bangalore sitecharacterization. Figure 2.9 shows the steps followed in site characterization of Bangalore. The detaileddiscussion about the SPT data, interpretations, mappings and corrections applied for SPT ‘N’ are presented inchapter 5. About geophysical method of MASW, field testing, results, evaluation of dynamic properties of soil,mapping of shear wave velocity, site classification and correlation between SPT corrected “N” values and shearwave velocity is discussed in chapter 6.Source 1 Source 3RSiteR12Source 2Rlog (# earthquake > m)3Magnitude, xSTEP 1Identification & characterizationof EQ sourcesSTEP 2Recurrence relationshipsGround motion parameter, YDistance, RUncertainty in the predictive relationships(attenuation relation.)STEP 3P[ Y > y*]Parameter value, y*Mean annual rate of Exceedance plotSTEP 4Figure 2.8: Different steps for probabilistic seismic hazard analysis (after Kramer, 1996).2.3.6 Methodology adopted for Local Site Effects and Site Response StudyThe site response study of Bangalore has been carried using one-dimensional ground response analysis andexperimental method of microtremor. Site response analysis aims at determining the response of a soil deposit tothe motion of the bedrock immediately beneath it. The overburden plays a very important role in determiningthe characteristics of the ground surface motion thus emphasizing the need for ground response analysis. Anumber of techniques have been developed for ground response analysis. These techniques can be grouped asone-, two-, and three-dimensional analyses according to thedimensionality of the problems they can address. A one- dimensional method can be used if the soil structure isessentially horizontal and is widely used in earthquake geotechnical engineering:Site response studies mainly deal with the determination of peak frequency of soft soil, amplification and thenature of response curve defining the transfer function at the site which forms an important input for evaluatingand characterizing the ground motion for seismic hazard quantification. Present study has been carried out byadopting the Nakamura method for obtaining the transfer function at various sites in Bangalore usingMicrotremor studies (details are given in 7). The surface sources for the ambient noise generate Rayleigh waveswhich affect the vertical and horizontal motion equally in the surface layer. The spectral ratio of the horizontalcomponent by the vertical component of the time series provides the transfer function at a given site. Thedominant peak is well correlated with the fundamental resonant frequency. The complete steps followed in siteresponse study are shown in the form of a flow chart in Figure 2.10.Assumptions in one-dimensional ground response analysis are:14

Study Area and MethodologyArea selection for sitecharacterizationNoExperimental StudiesYesSPT Data CollectionMASW ExperimentsData AnalysisField ExperimentBase Map Preparation withlocation of data PointsData InterpretationAnalysis and ResultsData InterpolationsMapping andData RepresentationsComparisons and correlationbetween both dataSite classificationFigure 2.9: Steps involved for site characterizationThe soil layer boundaries are horizontal and extend infinitely in the horizontal direction.Response of the soil deposit is predominantly caused by SH-waves propagating vertically from theunderlying bedrock.A number of different techniques are available for one-dimensional ground response analysis. These methodscan be broadly grouped into the following three categories:15

Study Area and Methodology‣ Linear analysis‣ Equivalent linear analysis‣ Nonlinear analysisOf the above, the most popular method used in professional practice is the “equivalent linear” approach which isincorporated in the computer program SHAKE. The equivalent linear analysis is discussed in detail in chapter 7.One-dimensional ground response analysis software SHAKE2000 was used in the present study. It requiresthree input parameters such as bedrock motion, dynamic material properties and site specific soil properties. Thepeak surface acceleration, ground response spectrum and period of soil column are obtained as output from thisanalysis. These values are used to get map indicating zones of amplification potential, spectral accelerationmaps at various frequencies and period of soil column map for Bangalore city.2.3.7 Methodology adopted for Liquefaction AnalysisThe first step in calculation of liquefaction potential is to determine if the soil has the potential to liquefy duringthe earthquake. Liquefaction potential analysis usually carried out by using simplified empirical procedure,originally developed by Seed and Idriss (1971). This simplified procedure represents the standard of practice inNorth America and in many other countries across the globe. Simplified procedures are reviewed periodicallyby groups of experts who make recommendations and changes according to data collected from newearthquakes and new developments in liquefaction hazard assessment. The potential for liquefaction is assessedwith the aid of liquefaction charts or semi empirical equations, which are based on observations of whetherliquefaction did or did not occur at specific sites during numerous past earthquakes (Chen and Scawthorn,2003). In this study simplified approach has been used with SPT bore log information collected. A simplespread sheet is developed using Excel macros to calculate the factor of safety against liquefaction for 720locations. The earthquake loading is evaluated in terms of cyclic stress ratio using Seed and Idriss (1971)simplified approach. Cyclic resistance ratio (CRR) is arrived based on corrected “N” value using plots of CRRversus corrected “N” value from a large amount of laboratory and field data of earthquake using equationproposed by Idriss and Boulanger (2005).A detailed procedures and calculation with tables are discussed inchapter 8.16

Study Area and MethodologySite Response StudyTheoretical ModelExperimentalInput shear modulusand dampingSite SelectionInput Ground Motionfrom DSHASite Response AnalysisInput Soil properties(SPT or MASW data)Processing DataField ExperimentResponse ParametersPredominant FrequencyComparison &Map preparationFigure 2.10: Flow chart for site response study17

Study Area and MethodologyDataGround Level PGASPT Corrected “N” ValueIf LL>32YESNOCalculation of Cyclic StressFor NextLocatiionCalculation of CyclicResistance RatioMagnitude Scaling FactorYESDetailed studyCalculation of Factor of SafetyEnd for one LocationLeast Factor of safety forMappingFigure 2.11: Flow chart for liquefaction assessment18

CHAPTER 3SEISMIC HAZARD ANALYSIS3.1 INTRODUCTIONIn this chapter, the deterministic hazard analysis for Bangalore is presented. Seismic hazard analyses involve thequantitative estimation of ground shaking hazards at a particular area. Seismic hazards can be analyzeddeterministically as and when a particular earthquake scenario is assumed, or probabilistically, in whichuncertainties in earthquake size, location, and time of occurrence are explicitly considered (Kramer, 1996).Probabilistic seismic hazard analysis provides not one, two, or three choices, but infinite choices for the user anddecision-makers (Wang, 2005). Krinitzsky (2005) comments on the problems in the application of probabilisticmethods and gives an account on a deterministic alternative which highlights that “A Deterministic SeismicHazard Analysis (DSHA) uses geology and seismic history to identify earthquake sources and to interpret thestrongest earthquake each source is capable of producing regardless of time, because that earthquake mighthappen tomorrow. Those are the Maximum Credible Earthquakes (MCEs), the largest earthquakes that canreasonably be expected. As we cannot safely predict when an earthquake will happen, the MCEs are what acritical structure should be designed for if the structure is to avoid surprises”.A critical part of seismic hazard analysis is the determination of Peak Ground Acceleration (PGA) and responseacceleration (spectral acceleration) for an area/site. Spectral acceleration (Sa) is preferred for the design of civilengineering structures. It is an accepted trend in engineering practice to develop the design response spectrumfor the different types of foundation materials such as rock, hard soil and weak soils. Seismic hazard analysisand determination of PGA is crucial and very important for any earthquake resistant design and Microzonation.To evaluate seismic hazards for a particular site or region, all possible sources of seismic activity must beidentified and their potential for generating future strong ground motion should be evaluated. Analysis oflineaments and faults helps in understanding the regional seismotectonic activity of the area. Lineaments arelinear features seen on the surface of earth which represents faults, features, shear zones, joints, litho contacts,dykes, etc; and are of great relevance to geoscientists. Scientists believe that a lineament is a deep crustal,ancient, episodically reactivated linear feature that exerts control on the make up of the crust and associateddistribution of ore and hydrocarbons (O’ leary et al, 1976, Ganesha Raj and Nijagunappa, 2004). This chapterhighlights the identification of seismic source by using remote sensing data in conjunction with collateral datalike seismotectonic atlas of India, drainage patterns using TOPO sheets of the area and published literature.From these identified sources, Maximum Credible Earthquake has been evaluated by three methods. Furthersynthetic ground motion is generated for MCE using regional seismotectonic parameters to fulfill therequirements of ground motion data for future study.To evaluate seismic hazards for a particular site or region, all possible sources of seismic activity must beidentified and their potential for generating future strong ground motion should be evaluated. Analysis oflineaments and faults helps in understanding the regional seismotectonic activity of the area. The study has beencarried out for an area of about 350 km radius around Bangalore as per the guideline available in RegulatoryGuide of U.S. Nuclear Regulatory Commission (1997). The recent seismic activity of Bangalore has beenstudied based on the seismic sources and earthquake events in the area. A new seismotectonic map has beenprepared by considering all the earthquake sources such as faults, lineaments, and shear zones. The pastearthquake events (of more than 350 moderate events from 3.5 to 6.2 in moment magnitude and 1150 smalltremors < 3.5 moment magnitude) are superimposed on this map with available latitude and longitude details.The peak horizontal accelerations at rock level has been calculated by considering the regional attenuationrelation from all the possible sources, through which seismogenic sources have been identified. Maximumcredible earthquake for the city has also been determined using Wells and Coppersmith (1994) relation. For thepurpose of future site response studies the synthetic ground motion has been generated considering regionalseismotectonic parameters using the synthetic ground motion model. Seismic hazard analysis details andsynthetic ground motion generated for Bangalore is presented in this chapter. Further the complete probabilistichazard analysis has been carried out. In this seismic hazard parameter ‘b’ has been evaluated considering theavailable earthquake data using (1) Gutenberg–Richter (G-R) relationship and (2) Kijko and Sellevoll (1989,1992) method utilizing extreme and complete catalogs (mixed data). A mixed data file contains information ofcomplete as well as incomplete earthquake events. Seismic hazard parameters such as “b” of the magnitudefrequency relationship R the mean return period, M max maximum regional magnitude were also evaluated. formost seismogenic sources found in DSHA. Hazard curves of peak ground acceleration versus mean annual rateof exceedance and spectral acceleration versus mean annual rate of exceedance at rock level have beengenerated. The present study results are compared with the earlier studies using PSHA for Mumbai (in south19

Deterministic Seismic Hazard AnalysisIndia). Uniform hazard response spectra for Bangalore city at rock level with different natural periods for 10%probability of exceedance in 50 years of exposure period have been generated.3.2 SEISMIC STUDY AREASeismic study area having a circular area of radius 350km has been selected for the seismicity study as perRegulatory Guide 1.165(1997). Regional, geological and seismological details for the Bangalore city has beencollected by using available literature, study of maps, remote sensing data. The study area marked in the Indiamap is shown in Figure 3.1. The study area having the center point as Bangalore city (with latitude of 12 o 58’’ Nand longitude of 77 o 36’’ E) has a radius of 350km (which covers the latitude 9.8 o N to 16.2 o N and longitude of74.5 o E to 80.7 o E). Study area covers major part of the Karnataka state, northern part of Tamil Nadu state,portion of Kerala and Andhra Pradesh states.MumbaiBangalore cityLatitude of 12 o 58’’NLongitude of 77 o 36’’EStudy Area350km around Bangalore city3.3 SEISMIC SOURCESFigure 3.1: Seismic study area in India map.In this research, as per Regulatory Guide 1.168(1997), regional geological and seismological investigations forthe Bangalore city has been carried out considering a circular area with a radius of 350km around the point ofinterest to identify seismic sources by using literature review, study of maps, remote sensing data and groundreconnaissance study. Study area lies between latitudes 9° 50”north to17°12” north and longitudes 74° 24”eastto 81°42” east. Seismotectonic details includes geology, rock type, fault orientation with length, lineaments withlengths, shear zones with length and seismic earthquake events. The well defined and documented seismicsources are present in the Seismotectonic Atlas-2000 published by Geological Survey of India. Geologicalsurvey of India has compiled all the available geological, geophysical and seismological data for entire India20

Deterministic Seismic Hazard Analysisand has published a seismotectonic map in 2000. Seismotectonic atlas (SEISAT, 2000) contains 43 maps in 42sheets of 3 o x4 o sizes with scale of 1:1 million, which also describes the tectonic frame work and seismicity.SEISAT is prepared with the intention that it can be used for the seismic hazard analysis of Indian cities. In thisanalysis about 6 SEISAT maps are merged, seismic sources with 350km in a circular area having radius aroundBangalore are used. Typical seismotectonic map is shown in Figure 3.2. Seismicity and activity of the regionwill always change based on neotectonic activity of the region. Thus it is necessary that any seismic hazardshould include recent seismicity. An extensive literature has been carried out to collect seismic sources fromrecent publications. Ganesha Raj and Nijagunappa (2004) have mapped major lineaments of Karnataka Statewith length more than 100 km using satellite remote sensing data and correlated with the earthquakeoccurrences. They have highlighted that there are 43 major lineaments and 33 earthquake occurrences withmagnitude above 3 (since 1828) in the State. About 23 of these earthquakes were associated with 8 majorlineaments, which they have named as active lineaments. The Mandya-Channapatna-Bangalore lineament,Lakshman Thirtha-KRS-Bangalore lineament, and Chelur-Kolar-Battipalle lineament are some of theseismically active lineaments identified by the authors. They have also stated that earthquakes are confined tothe southern part of the state indicating that south Karnataka is seismically more active. The authors have alsorecommended the need to upgrade the seismic zonation map of Karnataka especially for areas surroundingMandya, Bangalore, and Kolar. Karnataka lineaments published by Ganesha Raj and Nijagunappa (2004) usingremote sensing data are also considered in the present study which is shown in Figure 3.3. These sources arecompiled and seismic source map for Bangalore has been prepared which is shown in Figure 3.4. In seismichazard analysis other than the earthquake data and sources, ground motion attenuation relation is also importantwhich is discussed in the next section.3.4 SEISMICITY OF THE STUDY AREASeismicity of an area is the basic issue to be examined in seismic hazard analysis for evaluating seismic risk forthe purpose of microzonation planning of urban centres. Detailed knowledge of active faults and lineaments andassociated seismicity is required to quantify seismic hazard and risk. Indian peninsular shield, which was onceconsidered to be seismically stable, has shown that it is quite active. Seismic activity of the south India isstudied by Srinivasan and Sreenivas (1977), Valdiya (1998), Purnachandra Rao (1999), Ramalingeswara Rao(2000), Subrahmanya (2002 and 1996), Ganesha Raj (2001), Sridevi Jade (2004), Ganesha Raj and Nijagunappa(2004), Sitharam et al (2006) and Sitharam and Anbazhagan (2007). Srinivasan and Sreenivas (1977) have usedfield studies of bore well yield data and discussed the reactivation of dormant or inactive lineaments inducingseismicity in relative stable terrains of the continents. Purnachandra Rao (1999) highlighted the occurrence ofearthquakes in last few decades due to enhanced seismic activity in the interior of the Indian Plate which resultsfrom pre-existing faults under the influence of the ambient stress field due to the India-Eurasia plate collisionforces, oriented NS to NNE. Ramalingeswara Rao (2000) carried out strain rate and heat flow study in thesouthern India and characterized as medium to low seismicity region. Subrahmanya (2002 and 1996) highlightsthat the entire study area becoming seismically active due to the up warp of Mulki-Pulicat Lake (MPL) axiswhich connects 13 o N in west to 13.4 o N in east. He concludes that there is lot of seismic activity around thisMulki-Pulicat Lake axis and in particular he highlights that micro to meso-seimicity to the south and megaseismicity to the north of the MLP axis. In recent years much of the seismic activity in the state of Karnatakahas been in the south, in the Mysore-Bangalore region (Ganesha Raj and Nijagunappa, 2004). Recently, SrideviJade (2004) has estimated the plate velocity and crustal deformation in the Indian subcontinent using GPSmeasurements. The author concludes that southern peninsular India consists of large zones of complex folding,major and minor faults and granulite exposures, and this region cannot be classified as an area of low seismicactivity. All these authors highlight that the seismic activity of the south India has shown an increasing trend.Seismic data collected from various agencies [United State Geological Survey (USGS), Indian MetrologicalDepartment (IMD), NewDelhi; Geological Survey of India (GSI) and Amateur Seismic Centre (ASC), NationalGeophysical Research Institute (NGRI),Hyderabad; Centre for Earth Science Studies (CESS), Akkulam, Keralaand Gauribindanur (GB) Seismic station] contain information about the earthquake size in different scales suchas intensity, local magnitude or Richter magnitude and body wave magnitudes. These magnitudes are convertedto moment magnitudes (M w ) to achieve the status of the uniform magnitude by using magnitude relations givenby Heaton et al (1986). Some of the earthquake data is given in Table 3.1 and remaining is listed in Annexure A.The earthquake events collated are about 1421 with minimum moment21

Deterministic Seismic Hazard AnalysisFigure 3.2: Typical seismotectonic map used (after SEISAT, 2000)22

Deterministic Seismic Hazard AnalysisFigure 3.3: Lineaments map of Karnataka (Ganesha Raj and Nijagunappa 2004)23

Deterministic Seismic Hazard AnalysisTable 3.1: List of selected major earthquake data used in this studyDate Latitude Longitude Io Mb Ms Depth Data Mw( o N) ( o E)(Km) Source23.05.1974 12.8 78.3 - - 3.9 - INR 4.625.02.1975 15.3 79.6 3.9 INR 4.627.07.1959 11.5 75.25 4 IMD 4.731.07.1974 12.8 78.3 - - 4.1 - INR 4.713.03.1982 13.06 78..22 4 GBA 4.727.11.1984 12.53 78.69 - - 4.1 - GBA 4.717.01.1971 12.4 77 4.2 UMC 4.806.03.1971 12.4 77 4.2 UMC 4.807.06.1988 9.8 77.2 4.2 5 NGRI 4.817.12.1959 11.7 78.1 - - 4.3 - GUB 4.90.10.1964 11.3 75.8 4.3 GUB 4.927.03.1971 12.4 77 4.3 UMC 4.920.03.1984 12.55 77.77 4.4 GBA 4.903.12.1984 12.57 78.73 - - 4.3 - GBA 4.916.9.1816 13.1 80.3 V 4.4 OLD 520.06.1819 12 79.6 V - 4.4 - OLD 531.12.1820 14.5 80 V 4.4 OLD 522.08.1828 13 75 V - 4.4 - OLD 513.03.1829 13 77.6 V - 4.4 - OLD 513.8.1858 11.4 76 V 4.4 OLD 502.08.1865 12.7 78.7 V - 4.4 - OLD 501.09.1869 14.5 80 V 4.4 OLD 517.05.1972 12.4 77 4.5 UMC 516.05.1972 12.4 77 4.6 UMC 5.126.11.1972 12.8 78.3 - - 4.7 - INR 5.210.12.1807 13.1 80.3 VI 5 OLD 5.329.01.1822 12.5 79.7 VI - 5 - OLD 5.302.03.1823 13 80 VI 5 OLD 5.307.01.1916 13 77.3 V - 5 - IMD 5.329.07.1972 11 77 5 IMD 5.425.09.2001 11.79 80.31 5.6 23 IMD 5.501.04.1843 15.2 76.9 VII - 5.6 - OLD 5.703.07.1867 12 79.6 VII - 5.6 - MIL 5.712.02.1866 14.48 75.43 IV - 3.8 - OLD 628.2.1882 11.46 76.7 VIII 6.2 MIL 6.208.02.1900 10.8 76.8 VIII 6.2 UC 6.2magnitude of 1.0 and a maximum of 6.2 and earthquake magnitudes and are shown as various symbols withdifferent colors in Figure 3.5. About 1421 earthquakes have been collated and their magnitudes were convertedto moment magnitude scale. The data set contains 394 events which are less than moment magnitude 3, 790events from 3 to 3.9, 212 events from 4 to 4.9, 22 events from 5 to 5.9 and 3 events which are more thanmoment magnitude 6. Maximum earthquake magnitude out of about 1421 events reported in the study area is6.2. The earthquake events collated with latitudes and longitudes are used to prepare the seismicity map ofBangalore region which is shown in Figure 3.5. Out of 1421 seismic data about 1340 data collected from therecord of Gauribidanur seismic array (GBA), which is in operation for long time, having geographic coordinatesof the array center point, 13 o 36’15’’N, 77 o 26’10”E. GBA seismic station is about 85km away from the center ofthe study area. The GBA has an L-shaped configuration with dimensions of about 22 x 22 km 2 and a tightstation interval of about 2.5 km (see Figure 3.6). Data set is unique; it is the only array data available in the25

Deterministic Seismic Hazard Analysis10▲▲▲▲▲ 05▲▲▲▲ 01▲Gauribindanur Array ▲ 11▲(GBA)▲▲ ▲ 15▲▲22.4km▲▲20 NFigure 3.6: Pattern of the location of receivers in Gauribindanur Array.3.5 PREPARATION OF SEISMOTECTONIC MAP OF BANGALORESeismotectonic map showing the geology, geomorphology, water features, faults, lineaments, shear zone andpast earthquake events has been prepared for Bangalore which is as shown in Figure 3.7. Seismotectonic detailsof study area have been collected in a circular area having about 350 km radius around Bangalore. The sourcesidentified from SEISAT (2000) and remote sensing studies are compiled and a map has been prepared usingAdobe Illustrator version 9.0. The seismotectonic map contains 65 numbers of faults with length varying from9.73 km to 323.5km, 34 lineaments and 14 shear zones. The earthquake events collated and converted has beensuper imposed on base map with available latitudes and longitudes. The earthquake events collated are about1421 with minimum moment magnitude of 1.0 and a maximum of 6.2 and earthquake magnitudes are shown assymbols with different shape and colours. Maximum source magnitude is assigned based on the maximum sizeof the earthquake close to the each of the sources.3.6 GEOTECHNICAL DATA FOR ROCK DEPTHGeotechnical data was basically collated from geotechnical investigations carried out for several major projectsin Bangalore. The GIS model developed currently consists of about 850 borehole locations marked on thedigitized Bangalore map of 1:20000 scale which is shown in Figure 3.8. The data consists of visual soilclassification, standard penetration test results, ground water level, time during which test has been carried out,other physical and engineering properties of soil. Most investigations for residential and commercial complexeswere below 15m and wherever bedrock has been encountered investigation has been terminated at that depth forthese projects. This rock depth information has been used to prepare rock depth map for Bangalore, details arepresented in chapter 6. The rock depth information with latitude and longitude are obtained from the 653boreholes out of 850, which is used for rock level PGA Mapping. Further reduced level (RL) of each bore logrock depth is calculated from contour map (see the Figure 3.9.) developed in GIS model using TOPO sheets.Figure 3.9 shows the location of boreholes with the elevation contours at 10m intervals. The ground RL variesfrom 845 m to 910(mostly in the north and north western part of Bangalore, which gives information on slopingterrains or valleys and could be used to locate the RL of ground as well as rock surface.27

Deterministic Seismic Hazard AnalysisBore holeLocationScale1:20000Figure 3.8: Borehole locations in Bangalore3.7 DETERMINISTIC SEISMIC HAZARD ANALYSISSeismic hazard analysis has been carried out using deterministic approach. Deterministic seismic hazardassessment is carried out to identify the Maximum Credible Earthquake (MCE) that will affect a site. The MCEis the largest earthquake that appears possible along a recognized fault under the presently known or presumedtectonic activityBore holeLocationScale1:20000Figure 3.9: Contour map of Bangalore29

Deterministic Seismic Hazard Analysis(USCOLD, 1995), which will cause the most severe consequences to the site. MCE assessment gives littleconsideration to the probability of future fault movements. For the seismogenic earthquake source identificationthe minimum moment magnitude considered was 3.5 and above. In the present study, 52 faults and lineamentsand out of 1421 earthquake events, 470 events that are more than 3.5 moment magnitude have been considered.Shortest distances from source to Bangalore city centre have been measured from the seismotectonic map shownin Figure 3.7 and they are also listed in Table 3.2. With these distance and moment magnitude, Peak GroundAcceleration (PGA) is calculated at bedrock level by considering a focal depth of the earthquake of about 15 kmfrom the surface, which is arrived at based on the past earthquake data.3.7.1 Rock PGA Using Past EarthquakeIn this method the PGA for Bangalore has been calculated using the attenuation relation developed for southIndia by Iyengar and Raghukanth (2004) and the past earthquakes. By considering shortest hypocentral distancefor each source and the largest past earthquake close to the source the PGA is calculated. The attenuationrelation used to calculate PGA is given below (Iyengar and Raghukanth, 2004):2ln y = c1 + c2( M − 6) + c3( M − 6) − ln R − c4R+ ln ∈(3.1)Where y, M and R refer to PGA (g), moment magnitude and hypocentral distance respectively. Since PGA isknown to be distributed nearly as a lognormal random variable ln y would be normally distributed with theaverage of (ln ε) being almost zero. Hence with ε = 1, coefficients for the southern region are: (Iyengar andRaghukanth, 2004):c 1 = 1.7816; c 2 =0.9205; c 3 =-0.0673; c 4 =0.0035; σ(ln ε)=0.3136 (taken as zero) (3.2)The calculation shows that the minimum PGA value is 0.001g and maximum PGA value is 0.146g (caused fromMandya-Channapatna-Bangalore lineament, see L15 in Figure 3.10). Totally 10 sources have generated thehigher PGA values close to Bangalore. Among the 10 sources, the active lineament of Mandya-Channapatna-Bangalore lineament having a length of about 105km (which is 5.2km away from the Bangalore) causing aPGA value of 0.146g due to an earthquake event (M w of 5.1 occurred on 16th May 1972; corresponds to alatitude of 12.4 o N and longitude of 77.0 o E). This is a measured earthquake event having a surface wavemagnitude (M S ) of 4.6.Number and Name of SourceTable 3.2: Rock level PGA obtained using past earthquake eventsDistance(km)HypocentralDistance (km)OccurredEarthquake(Mw)F1 Periyar Fault 336 337 4.8 0.002F2 Vaigai River - Fault 326 326 4.6 0.001F3 Ottipalam - Kuttampuzah Fault 282 283 4.2 0.001F6 Valparai-Anaimudi Fault 290 290 4.5 0.002F9 Pattikkad - kollengol Fault 281 281 6.2 0.009F10 Cauveri Fault 224 225 5.4 0.007F13 Crystalline-Sedimentary Contact Fault 243 244 5.3 0.005F14 Attur Fault 198 199 4.5 0.003F16 Amirdi Fault 172 172 4.6 0.005F17 Main Fault 137 138 4.9 0.009F19 Mettur East Fault 97 98 4.6 0.010F20 Tirukkavilur Pondicherry Fault 219 220 5.7 0.009F21 Javadi Hills Fault 162 163 5 0.008F22 Pambar River Fault 124 125 4.6 0.007F23 Main Fault 143 144 4.9 0.008F24 264 264 4.2 0.001F25 Palar River Fault 175 176 5 0.007F30 Karkambadi -Swarnamukhi Fault 211 211 5 0.005F31 Tirumala Fault 216 216 5 0.005F32 Gulcheru Fault 181 182 4.4 0.003F35 Papaghani Fault 204 205 4 0.002PGA(g)30

Deterministic Seismic Hazard AnalysisF36 Badvel Fault 276 276 4.1 0.001F41 Wajrakarur Fault 246 247 5.7 0.008F43 Gani - Kalva Fault 284 284 4.4 0.001F45 Kumadavati - Narihalli Fault 271 271 6 0.008F47 Arkavati Fault 51 53 4.7 0.025F48 Chitradurga Fault 182 183 4.6 0.004F50 Sakleshpur - Bettadpur Fault 181 182 4 0.002F52 Bhavani Fault 217 217 6.2 0.015F65 Cudapah Eastern Magin Shear 269 269 4 0.001L2 Kabini Lineament 100 101 4.6 0.010L6 Netravathi Hemavathy Lineament 145 146 4.6 0.006L9 Yagachi Lineament 173 173 4.6 0.005L10 Mangalore-Shimoga-Tunga Lineament 251 251 5 0.004L11 Subramanya- Byadagi Gadag Lineament 235 235 6 0.011L14 Kunigal- Arkavathi Lineament 44 46 4.1 0.015L15 Mandya-Channapatna- Bangalore Lineament 5 16 5.1 0.146L16 Arakavathi- Doddaballapur Lineament 18 24 4.7 0.063L17 Arkavathi - Madhugiri Lineament 30 33 4.2 0.024L18 Doddabelvangala- Pavagada Lineament 24 28 4.1 0.026L20 Chelur-Kolar-Battipalle Lineament 58 60 5.2 0.037L22 Nelamangala- Shravanabelagula Lineament 26 30 5.3 0.089L23 Shimoga Lineament 265 265 4.5 0.002L24 Sorab-Narihalla Lineament 265 266 6 0.009L25 Vedavathi-Vanivilas Sagar Lineament 158 159 4.6 0.005L26 Holalkere- Herur Lineament 158 159 6 0.021L31Molakalmur-Hospet-Kushtagi- KrishnaLineament59 61 4 0.010L34 Sindhnur- Krishna Lineament 55 57 4.2 0.0133.7.2 Rock PGA using Subsurface Fault Rupture RelationshipWells and Coppersmith (1994) estimated subsurface rupture length using the length of the best-definedaftershock zone. They highlighted that accuracy of the size of the aftershock zone depends on the accuracy ofthe locations of individual aftershocks, which depends on the azimuths and proximity of the recording stationsand the accuracy of the subsurface structure velocity model. Wells and Coppersmith (1994) developedrelationships for magnitude versus subsurface rupture length, magnitude versus rupture width, and magnitudeversus rupture area. The developed regression relationships for subsurface rupture length and rupture area alsoprovide a basis for estimating the magnitudes of earthquakes that may occur on subsurface seismic sources suchas blind faults, which cannot be evaluated from surface observations. These relations on subsurface parametersinclude data for moderate-magnitude earthquakes (in the range of magnitude 5 to 6), allowing thecharacterization of relatively small seismic sources that may not rupture the surface. They believed thatsubsurface rupture length relations are appropriate for estimating magnitudes for expected ruptures along singleor multiple fault segments. These relations are determined from shallow-focus (crustal) continental interplate orintraplate earthquakes (stable and non stable continental) on the basis of a rather comprehensive data base ofhistorical events. Different correlation coefficients for these relations are given for strike-slip, reverse, normalfaulting and also the average relation for all slip types are developed to be appropriate for most applications.Best established are the relationships between moment magnitude Mw and subsurface rupture length (RLD) isvalid for the magnitude range of 4.8 to 8.1 and length/width range of 1.1 to 350 km.DSHA has carried out by considering subsurface fault rupture length for 52 seismic sources. Mark (1977)recommends that the surface rupture length may be assumed as 1/3 to 1/2 of the total fault length (TFL) basedon the worldwide data. However, assuming such large subsurface rupture length yields very large momentmagnitude and also it does not match with the past earthquake data. Wells and Coppersmith (1994) developedempirical relation between moment magnitude and subsurface fault length using past world wide earthquakes,which is as follows:31

Deterministic Seismic Hazard Analysislog( RLD ) 0.57M− 2.33(3.3)=WThe relation between moment magnitude and subsurface rupture length (RLD) was developed using reliablesource parameters and this is applicable for all types of faults, shallow earthquakes and interplate or intraplateearthquakes (Wells and Coppersmith, 1994). Using Wells and Coppersmith (1994) Equation 3.3 along with aparametric study, it is found that subsurface fault rupture length of about 3.8% of total fault length givesmoment magnitudes closely matching with the past earthquakes. Table 3.3 shows the RLD calculations,expected magnitude, and corresponding PGA from all the sources. The revised PGA lies in between minimumvalue of 0.001g and maximum value of 0.146g. In total, 9 sources have generated PGA value of more then0.045g at rock level close to Bangalore region. Among the nine sources, 3 sources have generated the higherPGA values close to Bangalore city. (i) the Arkavati fault (F47 in Figure 3.10) which is 51.24km away fromBangalore and having a length of about 125km.with a PGA of 0.025g (0.047g from RLD approach) due to anearthquake having a moment magnitude MW of 4.7. (ii) Chelur-Kolar-Battipalle Lineament (L20 in Figure3.10) having a length of about 111 km, which is 57.6km away from Bangalore causing a PGA value of0.037g(0.038g from RLD approach) due to an earthquake with M W of 5.2. (iii) Mandya-Channapatna-Bangalorelineament (L15 in Figure 3.10) having a length of about 105km which is 5.2km away from Bangalore causing aPGA value of 0.146g due to an earthquake with M W of 5.1 occurred on 16 th May 1972 (corresponds to a latitudeof 12.4 o N and longitude of 77.0 o E), RLD works out to be 3.8% of the total fault length corresponding to a PGAof 0.146g.Figure 3.10: Three seismogenic sources close to Bangalore.32

Deterministic Seismic Hazard AnalysisNumber and Name of SourceTable 3.3: PGA obtained from RLD approachLength(km)RLD(km)3.8(%)TFLExpected (3.8%)Magnitude (Mw)OccurredMagnitude (Mw)Distance(km)HypocentralDistance (km)F1 Periyar Fault 69 3 3 4.8 4.8 336 337 0.002F2 Vaigai River - Fault 32 2 1 4.2 4.6 326 326 0.001F3 Ottipalam - Kuttampuzah Fault 103 1 4 5.1 4.2 282 283 0.003F6 Valparai-Anaimudi Fault 46 2 2 4.5 4.5 290 290 0.002F9 Pattikkad - kollengol Fault 42 16 2 4.4 6.2 281 281 0.002F10 Cauveri Fault 323 6 12 6.0 5.4 224 225 0.012F13 Crystalline-Sedimentary Contact Fault 222 5 8 5.7 5.3 243 244 0.008F14 Attur Fault 167 2 6 5.5 4.5 198 199 0.009F16 Amirdi Fault 100 2 4 5.1 4.6 172 172 0.008F17 Main Fault 129 3 5 5.3 4.9 137 138 0.013F19 Mettur East Fault 38 2 1 4.4 4.6 97 98 0.008F20 Tirukkavilur Pondicherry Fault 67 8 3 4.8 5.7 219 220 0.004F21 Javadi Hills Fault 90 3 3 5.0 5 162 163 0.008F22 Pambar River Fault 99 2 4 5.1 4.6 124 125 0.013F23 Main Fault 82 3 3 5.0 4.9 143 144 0.009F24 52 1 2 4.6 4.2 264 264 0.002F25 Palar River Fault 136 3 5 5.3 5 175 176 0.009F30 Karkambadi -Swarnamukhi Fault 106 3 4 5.1 5 211 211 0.006F31 Tirumala Fault 48 3 2 4.6 5 216 216 0.003F32 Gulcheru Fault 22 2 1 4.0 4.4 181 182 0.002F35 Papaghani Fault 55 1 2 4.6 4 204 205 0.003F36 Badvel Fault 55 1 2 4.6 4.1 276 276 0.002F41 Wajrakarur Fault 39 8 1 4.4 5.7 246 247 0.002F43 Gani - Kalva Fault 144 2 5 5.4 4.4 284 284 0.004F45 Kumadavati - Narihalli Fault 148 12 6 5.4 6 271 271 0.005F47 Arkavati Fault 125 2 5 5.3 4.7 51 53 0.047F48 Chitradurga Fault 79 2 3 4.9 4.6 182 183 0.006F50 Sakleshpur - Bettadpur Fault 86 1 3 5.0 4 181 182 0.006F52 Bhavani Fault 90 16 3 5.0 6.2 217 217 0.005F65 Cudapah Eastern Magin Shear 94 1 4 5.1 4 269 269 0.004PGA(g)33

Deterministic Seismic Hazard AnalysisL2 Kabini 130 2 5 5.3 4.6 100 101 0.021L6 Netravathi Hemavathy 169 2 6 5.5 4.6 145 146 0.015L9 Yagachi 102 2 4 5.1 4.6 173 173 0.008L10 Mangalore-Shimoga-Tunga 134 3 5 5.3 5 251 251 0.005L11 Subramanya- Byadagi Gadag 318 12 12 6.0 6 235 235 0.011L14 Kunigal- Arkavathi 101 1 4 5.1 4.1 44 46 0.045L15 Mandya-Channapatna- Bangalore 105 4 4 5.1 5.1 5 16 0.146L16 Arakavathi- Doddaballapur 109 2 4 5.2 4.7 18 24 0.107L17 Arkavathi - Madhugiri 156 1 6 5.4 4.2 30 33 0.089L18 Doddabelvangala- Pavagada 125 1 5 5.3 4.1 24 28 0.096L20 Chelur-Kolar-Battipalle 111 4 4 5.2 5.2 58 60 0.037L22 Nelamangala- Shravanabelagula 130 5 5 5.3 5.3 26 30 0.089L23 Shimoga 130 2 5 5.3 4.5 265 265 0.004L24 Sorab-Narihalla 249 12 9 5.8 6 265 266 0.007L25 Vedavathi-Vanivilas Sagar 163 2 6 5.5 4.6 158 159 0.013L26 Holalkere- Herur 172 12 7 5.5 6 158 159 0.013L31 Molakalmur-Hospet-Kushtagi- Krishna 190 1 7 5.6 4 59 61 0.054L34 Sindhnur- Krishna 223 1 8 5.7 4.2 55 57 0.06434

Deterministic Seismic Hazard Analysis3.8 SYNTHETIC EARTHQUAKE MODELFor microzonation, the study of local sites effects for a particular earthquake found in the seismic hazardanalysis need to be carried out. To study the effects of earthquake in the local scale level, the particularearthquake record/ground motion in the form of time series is required. For the area having poor seismic record,synthetic ground motion models is the alternative. The study area lacks ground motion records. Iyengar andRaghukanth (2004) have developed ground motion attenuation relation based on the statistically simulatedseismological model. Modeling of strong motion helps to estimate future hazard of the region and study thelocal effects in local scale. Seismological model by Boore (1983) is used for generation of syntheticacceleration-time response (Atkinson and Boore 1995, Hwang and Huo 1997). Iyengar and Raghukanth (2004)have been used this seismological model and developed attenuation equation for Peninsular India and suggestedcoefficients for southern India. Boore (1983, 2003) gives the details of estimating ground motion based on theFourier amplitude spectrum of acceleration at bedrock and this is expressed as:A ( f ) = C[ S( f )] D( f ) P( f )(3.4)where, S ( f ) is the source spectral function, D( f ) is the diminution function characterizing the attenuation,and P ( f ) is a filter to shape acceleration amplitudes beyond a high cut-off frequency f m,and C is a scalingfactor. In the present study, the single corner frequency model has been used (Brune, 1970) which is given as;2 M0S( f ) = ( 2Πf)(3.5)[ 1+( f / f c)]Where, f c is the corner frequency, M 0is the seismic moment, and, ∆σ is the stress drop and they are relatedthrough:136⎛ ∆σ⎞fc = 4.9Χ10V⎜⎟S(3.6)⎝ M0 ⎠Here, the shear wave velocity (V S ) for the source region is taken as 3.65 km/sec and seismic rock density istaken as 2.7g/cc (Parvez et.al, 2003). The diminution function D(f) is defined as:⎡ − ΠfR⎤( f ) G expD = ⎢( ) ⎥ (3.7)⎣VSQf ⎦In which, G refers to the geometric attenuation and the other term to an elastic attenuation. In this equation, Q(f)is the quality factor of the region.The high-cut filter in the seismological model is:1−28⎡ ⎛ f ⎞ ⎤P ( f , f ) ⎢1⎜⎟m= + ⎥(3.8)⎢⎣⎝ fm ⎠ ⎥⎦Where f mcontrols the high frequency fall of the spectrum.The scaling factor C is given by:< R > 2C =θΦ4∏ρVs3Where R > is the radiation coefficient averaged over appropriate range of azimuths and take-off angles. 100km. For Southern Indian region, Rao et al (1998) used strong motion records of small magnitude earthquakes0.83and estimated Q value to be 460 f (after Iyengar and Raghukanth, 2004). However, for Bangalore region,Tripathi and Ugalde (2004) developed Q factor by using seismic array from the Gauribidanur seismic recordingstation that is about 85 km from Bangalore, They estimated and reported for the different frequency range of1Hz to 10 Hz. For Bangalore, the natural frequency is in the range of 3 to 6 Hz and the corresponding Q value is0.88488 f (Tripathi and Ugalde, 2004). (3.10)The sources (15 numbers) causing the PGA value of 0.01g and more, from the method I, have been used forgenerating the synthetic ground motion. From the above sources, using method II, 8 sources have been(3.9)35

Deterministic Seismic Hazard Analysisidentified for generating the synthetic ground motion. These 8 sources considered for synthetic ground motiongeneration along with past earthquake events are shown in Table 3.4. The PGA values obtained form syntheticground motion model using regional seismotectonic parameters varies from 0.005g to 0.136g. These values foreach source are shown in Table 3.5. The lineament L15 gives the highest PGA value of 0.136g by takinghypocentral distance of 15.88km. The response spectrums for the simulated ground motions are plotted; it showsthat the predominant period of synthetic ground motion is 0.06 seconds irrespective of the magnitude andsources. Further, PGA obtained from the model for the L15 matches well with the PGA values from both theabove approaches. The synthetic ground motion for Mandya-Channapatna-Bangalore lineament (L15) is shownin Figure 3.11. Rock level spectral acceleration for Mandya-Channapatna-Bangalore lineament (L15) is shownin Figure 3.12. The generated the shape spectral acceleration in study matches with the shape of uniform hazardspectrum.0.200.15Acceleration (g)0.100.050.0029 30 31 32 33 34 35-0.05-0.10-0.15Time ( sec)Figure 3.11: Synthetic ground motion generated from Mandya-Channapatna-Bangalore lineament (L15)0.35Spectral Acceleration (g)0.300.250.200.150.100.050.000 0.25 0.5 0.75 1 1.25 1.5 1.75 2Period (sec)Figure 3.12: Rock level spectral acceleration from synthetic ground motion36

Deterministic Seismic Hazard AnalysisNumber and Name of SourceTable 3.4: Sources used for synthetic ground motion generationLength(km)RLD(km)3.8(%)TFLDistance(km)HypocentralDistance (km)Expected(3.8%)Magnitude(Mw)PGA(g)OccurredMagnitude(Mw)F19 Mettur East Fault 38 2 1 97 98 4.4 0.008 4.6 0.010F47 Arkavati Fault 125 2 5 51 53 5.3 0.047 4.7 0.025L11 Subramanya- Byadagi Gadag 318 12 12 235 235 6.0 0.011 6 0.011L15 Mandya-Channapatna- Bangalore 105 4 4 5 16 5.1 0.146 5.1 0.146L16 Arakavathi- Doddaballapur 109 2 4 18 24 5.2 0.107 4.7 0.063L20 Chelur-Kolar-Battipalle 111 4 4 58 60 5.2 0.037 5.2 0.037L22 Nelamangala- Shravanabelagula 130 5 5 26 30 5.3 0.089 5.3 0.089L26 Holalkere- Herur 172 12 7 158 159 5.5 0.013 6 0.021PGA(g)Number and Name of SourceTable 3.5: PGA obtained from synthetic ground motion modelLength(km)Distance(km)Maximum(expected)Magnitude(Mw)HypocentralDistance (km)F19 Mettur East Fault 38 97 4.6 98 0.005F47 Arkavati Fault 125 51 4.7 53 0.014L11 Subramanya- Byadagi Gadag 318 235 6 235 0.012L15 Mandya-Channapatna- Bangalore 105 5 5.1 16 0.136L16 Arakavathi- Doddaballapur 109 18 5.2 24 0.085L20 Chelur-Kolar-Battipalle 111 58 5.2 60 0.022L22 Nelamangala- Shravanabelagula 130 26 5.3 30 0.064L26 Holalkere- Herur 172 158 6 159 0.020PGA(g)37

Deterministic Seismic Hazard Analysis3.9 DEVELOPMENT OF ROCK LEVEL PGA MAPFrom the above three approaches the highest PGA value for the Bangalore is obtained from the Mandya-Channapatna-Bangalore lineament (L15) for the past earthquake of 5.1 in moment magnitude. This clearlyindicates that the maximum credible earthquake for the region is 5.1 in moment magnitude. Seismogenic sourceis the lineament of Mandya-Channapatna-Bangalore lineament (L15). Further an attempt has been to map therock level PGA considering the Mandya-Channapatna-Bangalore lineament (L15) as the source. The rock levelPGA has been calculated using rock depth information from geotechnical data (using 653 boreholes drilled forgeotechnical investigations). The lineament L15 location is superimposed with borehole locations. The shortestdistance between the each borehole to the lineament has been measured. Using the depth of 15km and shortestdistance, the hypocenter distance of each bore log is evaluated. The peak ground acceleration for each boreholelocation considering MCE of 5.1 for the seismogenic source of L15 has been evaluated. Using these PGA valuesat rock level PGA map has been prepared, which is as shown in Figure 3.13. The rock level PGA valuesdetermined from model matches well with the regional attenuation relation.3.10 DETERMINISTIC SEISMIC HAZARD ANALYSIS.Deterministic seismic hazard analysis has been carried out considering all the sources within a circular areahaving a radius about 350 Km around Bangalore and the past earthquake events. The PGA at base rock level hasbeen calculated by using an appropriate attenuation relation by assigning the maximum reported earthquake tothe source and also by assuming the subsurface rupture length (RLD corresponds to 3.8% of total fault length).The highest PGA value (0.146g) is attributed to the source Mandya-Channapatna-Bangalore lineament (L15)and it is considered to be the seismogenic source for Bangalore. The earthquake event of Mw of 5.1 isconsidered as MCE for the Bangalore (which occurred on 16th May 1972; corresponds to latitude of 12.4 o N andlongitude of 77.0 o E). This event was a measured/recorded earthquake event with surface wave magnitude (Ms)of 4.6. The synthetic ground motion has been generated using regional seismotectonic parameters for thedifferent identified sources. The source Mandya-Channapatna-Bangalore lineament gives the highest PGA valueof 0.136g for MCE of 5.1. The following conclusions are drawn.From detailed deterministic seismic hazard analysis, the seismogenic source for Bangalore is Mandya-Channapatna-Bangalore lineament (L15) and maximum credible earthquake is 5.1 inmoment magnitude.The synthetic ground motion and spectral acceleration has been generated and the shape of the spectralacceleration matches with the shape of uniform hazard spectrum.The rock level PGA map has been prepared and such maps will be of use for the purpose of seismicmicrozonation, ground response analysis and design of important structures.From this study, it is very clear that Bangalore area can be described as seismically moderately activeregion. It is also recommended that southern part of Karnataka in particular Bangalore, Mandya andKolar, need to be upgraded from current Indian Seismic Zone II to Seismic Zone III.38

Deterministic Seismic Hazard AnalysisFigure 3.13: Rock level peak ground acceleration map39

Probabilistic Seismic Hazard Analysis3.11 DATA COMPLETENESS ANALYSIS AND SEISMIC HAZARDPARAMETERUnlike deterministic seismic hazard analysis probabilistic analysis allows the uncertainties in the size, location,rate of recurrence and effect of earthquakes to be explicitly considered in the evaluation of seismic hazard. Thissection deals with the completeness analysis of seismic data and estimation of seismic hazard parameters for theseismic study area.3.11.1 Analysis of Completeness of Seismic DataImportant step in any earthquake data analysis is to investigate the available data set to assess its nature anddegree of completeness. Incompleteness of available earthquake data make it difficult to obtain fits ofGutenberg-Richter recurrence law (equation 4.1) that is thought to represent true long term recurrence rate.Almost all earthquake catalogs are biased against small shocks, because of seismographs station density or inthe early records population density (Stepp, 1972). Uncertainty in size of earthquakes produced by each sourcezone can be described by various recurrence laws. The Gutenberg-Richter recurrence law that assumes anexponential distribution of magnitude is commonly used with modification to account for minimum andmaximum magnitudes and is given by:LogN = a − bM(3.11)For a certain range and time interval, equation 4.1 will provide the number of earthquakes, (N) with magnitude,(M) where ‘a’ and ‘b’ are positive, real constants. ‘a’ describes the seismic activity (log number of events withM=0) and ‘b’ which is typically close to 1 is a tectonics parameter describing the relative abundance of large tosmaller shocks.The number of earthquakes per decade was divided in to five magnitude range such as 2 < M < 2.9; 3 < M < 3.9;4 < M < 4.9; M > 5. Table 3.6 describes the number of earthquakes reported in each decade (Stepp, 1972;Shankar and Sharma, 1997) since the beginning of the available historical record. Figure 3.14 shows thehistogram representing the data listed in Table 3.6 for the whole catalogue from 1807 through 2006. The wholecatalog shows that the period 1807 to 1976 the data is poorly reported which may be due to the lack ofobservations. However from 1976 – 1996 better recording of the data can be observed. From 1997 to 2006 lessdata this is attributed to non availability of earthquake data from GBA (as the organization does not release thelast 10 year data for the public consumption).Data Completeness AnalysisThe analysis in previous section highlight that the data is severely incomplete. This is to determine the mean rateof occurrence, λ = N/year, from the complete 200 year sample leads to underestimation of λ for the middle andlow intensity levels. To overcome this problem the method proposed by Stepp (1972) was adopted for thepresent investigation in order to check the completeness of the earthquake data. Analysis has been carried out bygrouping the earthquake data into several magnitude classes and each magnitude class is modeled as a pointprocess in time. By taking the advantage of the property of statistical estimation that variance of the estimate ofa sample mean is inversely proportional to the number of observations in the sample (Stepp, 1972). Thus thevariance can be made as small as desired by making the number of observation in theTable 3.6: Number of earthquakes reported in each decade since the beginning of the available historicalrecords for Bangalore regionTime inNumber of earthquakesyears 1

Probabilistic Seismic Hazard Analysis1887-1896 4 41897-1906 1 11907-1916 1 11917-1926 01927-1936 01937-1946 01947-1956 01957-1966 4 41967-1976 1 17 4 221977-1986 443 104 5471987-1996 13 380 341 64 7981997-2006 4 4 2 1410sample large enough, provided that reporting is complete in time and the process is stationary i.e. the meanvariance and other moments of each observations remains the same. In order to obtain an efficient estimate ofthe variance of the sample mean, it is assumed that the earthquake sequence can be modeled by the Poissondistribution. If k 1 , k 2 , k 3 …..k n are the number of earthquakes per unit time interval, then an unbiased estimate ofthe mean rate per unit time interval of this sample is:=1nn∑ k ii=1λ (3.12)and its variance is:σ 2λnλ= (3.13)Where n is the number of unit time intervals. Taking the unit time interval to be one year gives a standarddeviation of:λσ λ= (3.14)Tas the stationary deviation of the estimate of the mean, where T is the sample length. Hence by assuming1stationary process, one can expect that σλbehaves as in the subintervals, in which the mean rate ofoccurrence in a magnitude class is constant. In other words when λ is constant, and then the standard deviation1σλvaries as where T is the time interval of the sample. If the mean rate of occurrence is constant weTexpect stability to occur only in the subinterval that is long enough to give a good estimate of the mean but shortenough that it does not include intervals in which reports are complete (Stepp, 1972).The rate of earthquake occurrence as a function of time interval is listed in Table 3.7 for the range ofmagnitudes. The rate is given as N/T where N is the cumulative number of earthquakes in the time interval T,for subintervals of the 200-year sample shown in first column. These data are used to determine the standarddeviation of the estimate of the mean through equation (3.14). The results are shown in Figure 3.15.Table 3.7 and Figure 3.15 reveals several features significant to statistical treatment of earthquake dataregardless of whether one uses empirical relationship log N = a – b M with the extreme value distribution orother statistical approaches. For each magnitude interval in Figure 3.15 the plotted points are supposed to definea straight line relation as long as the data set for the magnitude interval is complete. For a given seismic regionthe slope of the lines for all magnitude intervals should be same. It can be observed from Figure 3.15 that dataset for all magnitude intervals seems to be complete for last 40 years (1967-2006).T41

Probabilistic Seismic Hazard AnalysisNumber of earthquakes1000100101

Probabilistic Seismic Hazard AnalysisTable 3.7: Earthquake distribution in time and magnitude intervals for Bangalore regionTime periodTimeInterval1

Probabilistic Seismic Hazard Analysis1011 < M < 1.92 < M < 2.93 < M < 3.94 < M < 4.9M > 51 / sqrt (T)σ λ0.10.0110 100 1000Time (yrs)Figure 3.15: Variance of seismicity rate for different magnitude intervals and different lengths of movingtime windows3.11.2 Seismic Hazard ParametersIt has been observed from the earlier analysis that the data set is not complete for the interval 1807 through1967. Generally “b” value is computed from the analysis of whole set of data without testing the completenessof the data which gives error in the estimation of “b” value. Following the method proposed by Stepp (1972), itwas found that data set is complete for the last 40 years. Hence, computation of “b” value has been carried outusing the data set from 1967 through 2006. Figure 3.16 presents the logarithm of the cumulative earthquake peryear for M, where M is the magnitude in particular interval. An interval of 0.5 is taken for grouping the datawhile computing the ‘b’ value. A straight line fit in least square sense for the complete set of each magnituderange which is as follows:log( N)= 3.6 − 0. 89M(3.15)From the above equation seismic hazard parameter “a” is 3.6 and “b” is 0.89 with a correlation coefficient of0.96. Recurrence relation arrived for the region does not include major earthquake in the historic times and alsoit includes micro seismic data less than Mw of 3.5. Hence G-R relation also developed by considering all thedata as two groups, one is magnitude range of 3.5 to 6.2 and another is 4 to 6.2. Figure 3.17 presents thelogarithm of the cumulative earthquake per year for M versus magnitude of Mw of 3.5 to 6.2, a straight line fitin least square sense for the complete set of each magnitude range which is as follows:log( N)= 3.56 − 0. 87M(3.16)Figure 3.18 presents the logarithm of the cumulative earthquake per year for M versus magnitude for Mw of 4 to6.2, a straight line fit in least square sense for the complete set of each magnitude range which is as follows:log( N)= 3.82 − 0. 92M(3.17)From the above three equations seismic parameter ‘b’ value of the region varies from 0.87 to 0.92. Furtherseismic hazard parameters were also evaluated using all the earthquake data set, which is also termed as mixeddata set. Kijko and Sellevoll (1989, 1992) have presented a versatile statistical method based on the maximumlikelihood estimation of earthquake hazard parameters for the mixed data set. These values are reevaluated andconfirmed by Indian Institute of Technology Kharagpur and National Geophysical Research Institute byadopting their methodology.44

Probabilistic Seismic Hazard Analysis1Log (Cumulative. no of events/yrear)0.80.60.40.20-0.2-0.4-0.6log(N) = -0.89M + 3.56R 2 = 0.96-0.82.5 3 3.5 4 4.5 5Magnitude (Mw)Figure 3.16: Frequency magnitude relationship for study area for complete data1Log (Cumulative. no of events/yrear)0.50-0.5-1-1.5log(N) = -0.87M + 3.56R 2 = 0.98-23 3.5 4 4.5 5 5.5 6 6.5Magnitude (Mw)Figure 3.17: Frequency magnitude relationship for study area using Mw of 3.5 to 6.2Analysis Based on Kijko and Sellevoll MethodSeismic hazard parameters were also evaluated based on the complete as well as incomplete data which is alsotermed as mixed data. Kijko and Sellevoll (1989, 1992) have presented a versatile statistical method foranalyzing such mixed data set. Kijko and Sellevoll have also released a computer program (HN2, Release 2.10,2005) which is used for these calculations. The program was developed by assuming the earthquake occurrenceas Poisson’s model and the doubly truncated Gutenberg-Richter magnitude distribution for maximum likelihoodestimation of the slope of the recurrence relationship. In the present investigation, a threshold magnitude valueof 3.00 and standard deviation value of 0.2 is used. From the analysis it was observed that the beta (β) value is2.00 ± 0.07, Lambda ( λ ) value is 0.386 (for Mw of 5.0), seismic parameter of ‘b’ value is 0.87 ±0.03 andmaximum magnitude is (M max ) 6.0± 0.54. Figure 3.19 shows the variation of return period with magnitude asobtained by mixed data analysis. Figure 3.20 shows that lower the magnitude shorter the return period. Withincreasing magnitude, the return period becomes longer.45

Probabilistic Seismic Hazard Analysis0.5Log (Cumulative. no of events/yrear)0-0.5-1-1.5log(N) = -0.92M + 3.82R 2 = 0.98-23 3.5 4 4.5 5 5.5 6 6.5Magnitude (Mw)Figure 3.18: Frequency magnitude relationship for Study area using Mw of 4 to 6.2100Return Period (Years)1010.13 3.5 4 4.5 5 5.5 6Magnitude (M w )Figure 3.19: Return periods estimated using Kijko and Sellevoll (1992) methodProbability of the magnitudes for the time period (exposure time) of 50, 100 and 1000 years is shown in Figure4.8. Probability of occurring of the lower magnitude is 100% in the specified return period. Further for highermagnitudes, probability of occurrence gets decreasing. These probability values are only based on magnitudeand thus it is necessary to revise the same by considering other uncertainties involved inseismic hazard analysis.3.12 COMPARATIVE ANALYSISTable 3.8 compares the values of ‘b’ obtained from the two methods presented in this chapter. Also values of“b” reported by other researchers for south India with considering earthquake data stretched over differentperiods. It can be observed from the Table 4.3, that “b” values obtained from both the analysis (G-R relation and46

Probabilistic Seismic Hazard AnalysisKijko and Sellevoll, 1989, 1992) are quite comparable. The ‘b’ value obtained in this study matches well withthe previous studies of Ram and Rathor (1970), Kaila et al (1972), Ramalingeswara Rao and Sitapathi Rao(1984) and Jaiswal and Sinha, (2006) for southern India. Further it can be observed that “b value” reported inthis paper is higher when compared to the earlier investigators. This higher “b” values attributed to thegeological material heterogeneity and increased seismicity data in the study area. This clearly highlight thatseismic activity of region is showing increasing trend when compared to the past.1Probability0.10.010.001T=50 YearT=100 yearT=1000 Year0.00010.000012 2.5 3 3.5 4 4.5 5 5.5 6 6.5Magnitude (M w )Sl noFigure 3.20: Probability- magnitude diagrams using Kijko and Sellevoll (1992) methodTable 3.8: Values of ‘b’ compared with published literatureAuthorsYear ofpublicationValue of ‘b’Data analyzedfor a period(years)1 Avadh Ram and Rathor 1970 0.81 702 Kaila et al 1972 0.7 143 Ramalingeswara Rao and Sitpathi Rao 1984 0.85 1704 Jaiswal and Sinha 2006 0.84 to1.0 1605Present work (Bangalore region)G-R relation -completed data- Mw of 3.5 to 6.2- Mw of 4.0 to 6.2Kijko and Sellvoll method----0.890.870.920.87±0.03402002002003.13 SEISMOGENIC SOURCESIn the deterministic seismic hazard analysis of Bangalore, 8 seismogenic seismic sources were identified withina circular area having a radius of 350km. Among the 8 sources, Subramanya - Byadagi Gadag lineament andHolalkere- Herur lineament are far away from Bangalore (more than 150 km and also moment magnitude ofmore than 4 are very few on these two lineaments), hence the remaining 6 sources are considered for theprobabilistic seismic hazard analysis. The details of 6 sources with reported number of earthquake data close to47

Probabilistic Seismic Hazard Analysiseach source are shown in Table 3.9. In Table 3.9, the shortest and longest distance are presented, which is usedto calculate the hypocentral distances by assuming a focal depth of 15km for all the sources as listed in Table3.10. The sources considered for the PSHA with respect to Bangalore is shown in Figure 3.21.Figure 3.21: Seismogenic sources considered for the PSHA.Table 3.9: Seismogenic sources considered for the PSHA.HypocentralDistance(km)No EQclosetosource3.14 PROBABILISTIC SEISMIC HAZARD ANALYSISWeighting factorLengthNumber and Name of Source(km)Min Maxα s β s AverageF19 Mettur East Fault 98 117 38 15 0.061 0.011 0.036F47 Arkavati Fault 53 89 125 20 0.202 0.015 0.108Mandya-Channapatna-105L15 Bangalore 16105 25 0.170 0.018 0.094L16 Arakavathi- Doddaballapur 24 78 109 12 0.176 0.009 0.093L20 Chelur-Kolar-Battipalle 60 105 111 50 0.180 0.037 0.108Nelamangala-151L22 Shravanabelagula 30130 14 0.210 0.010 0.110PSHA is the most common approach to evaluate the seismic design load for the important engineering projects.PSHA method was initially developed by Cornell (1968) and its computer form was developed by McGuire(1976 and 1978) and Algermissen and Perkins (1976). McGuire developed EqRisk in the year 1976 and FRISKin the year 1978. Algermissen and Perkins (1976) developed RISK4a, presently called as SeisRisk III. Siteground motions are estimated for selected values of the probability of ground motion exceedance in a designTable 3.10: Source recurrence relation weighting factors48

Probabilistic Seismic Hazard AnalysisNumber and Name of SourceF19 Mettur East Fault 4.6 38 97 116 15F47 Arkavati Fault 4.7 125 51 88 20L15 Mandya-Channapatna- Bangalore 5.1 105 5.2 104 25L16 Arakavathi- Doddaballapur 5.2 109 18 77 12L20 Chelur-Kolar-Battipalle 5.2 111 58 104 50L22 Nelamangala- Shravanabelagula 5.3 130 26 150 14period of the structures or for selected values of annual frequency or return period for ground motionexceedance. The principal output from a PSHA is a hazard curve showing the variation of a selected groundmotionparameter, such as peak ground acceleration (PGA) or spectral acceleration (SA), against the annualfrequency of exceedance (or its reciprocal, return period). The occurrence of earthquakes in a seismic source isassumed as the Poisson distribution. The probability distribution is defined in terms of the annual rate ofexceedance of the ground motion level z at the site under consideration ( ν (z)), due to all possible pairs (M, R)of the magnitude and epicentral distance of the earthquake event expected around the site, considering itsrandom nature. The probability of ground motion parameter at a given site, Z, will exceed a specified level, z,during a specified time, T and it is represented by the expression:−ν( z)TP(Z > z)= 1−e ≤ν( z)T(3.18)Where ν (z)is (mean annual rate of exceedance) the average frequency during time period T at which the levelof ground motion parameters, Z, exceed level z at a given site. The function ν (z)incorporates the uncertaintyin time, size and location of future earthquakes and uncertainty in the level of ground motion they produce at thesite. It is given by:uNm∞⎡⎤ν ( z)= ∑ Nn( m0 ) ∫ fn( m)⎢ ∫ fn( r | m)P(Z > z | m,r)dr⎥dm(3.19)n= 10m= m ⎣r= 0⎦Where N n (m 0 ) is the frequency of earthquakes on seismic source n above a minimum magnitude m 0 that is takenas 4.0 in this work (magnitude less than 4 is considered to be insignificant in the engineering construction). f n (m)is the probability density function for minimum magnitude of m 0 and maximum magnitude of m u ; f n (r|m) is theconditional probability density function for distance to earthquake rupture; P(Z>z|m, r) is the probability thatgiven a magnitude ‘m’ earthquake at a distance ‘r’ from the site, the ground motion exceeds level z. Theintegral in equation (4.9) is replaced by summation and the density function f n (m) and f n (r|m) are replaced bydiscrete mass functions. The resulting expression for ν (z)is given by:uN m =r = ri m ⎡ j max⎤ν ( z)= ∑ ∑λn( mi) ⎢ ∑ Pn( R = rj| mi) P(Z > z | mi, rj⎥(3.20)0n=1 mi= m ⎢⎣rj= rmin⎥⎦Where m )λ is the frequency of events of magnitude m i occurring on source n obtained by discretizing then(iearthquake recurrence relationship for source n. The estimation of uncertainty involved in magnitude, distanceand peak ground acceleration are discussed below.3.14.1 Regional Recurrence ModelMaximumMagnitude(Mw)Length(km)ShorterDistance(km)LongerDistance(km)No of EQclose tosourceEach seismic source has a maximum earthquake that can not exceed. In PSHA, the lower magnitude can betaken from 4.0 to 5.0 magnitudes, since smaller than this will not cause significant damage to the engineeringstructures and larger magnitude can be evaluated by considering the seismotectonic of the region and historicearthquake data.The magnitude recurrence model for a seismic source specifies the frequency of seismic eventsof various sizes per year. For Bangalore region the seismic parameters are determined using Gutenberg-Richter(G-R) magnitude-frequency relationship which is given in Equation 3.16. The recurrence relation of each faultcapable of producing earthquake magnitude in the range m 0 to m u is calculated using the truncated exponentialrecurrence model developed by Cornel and Van Mark (1969), and it is given by the following expression:49

Probabilistic Seismic Hazard Analysis−β( m−m0 )βe0uN(m)= Ni( m )for m ≤ m < m(3.21)0u 0−β( m −m)1−eWhere β =b ln (10) and N i (m 0 ) proposal weightage factor for particular source based on the deaggregation.3.14.2 DeaggregationThe recurrence relation (Equation 3.16) developed for the study area represents the entire region and it is notfor specific source. Each source recurrence is necessary to discriminate nearby sources from far-off sources andto differentiate activity rate among sources. Such seismic source recurrence relation is rarely known due topaucity of large-scale data accruing in historical times. An alternative is to empirically calculate the “b” valuefrom known measured slip rate of each seismic source. For the sources under consideration, no such slip ratemeasurements are reported. Moreover, PI earthquakes are associated with poor surface expressions of faults andhence reliable estimation of slip rates has not been yet possible (Rajendran and Rajendran, 1999; RaghuKanthand Iyengar, 2006). Hence, it is necessary to proceed on a heuristic basis invoking the principle of conservationof seismic activity. According to this, the regional seismicity measured in terms of the number of earthquakes0per year with m ≥ m , should be equal to the sum of such events occurring on individual source. Deaggregationprocedure followed by Iyengar and Ghosh (2004); RaghuKanth and Iyengar (2006) for PSHA of Delhi andMumbai (in south India) have been used here to find the weightage factor for each source based on the length(α) and number of earthquakes (χ) for the corresponding source. The length weighting factor for the source iLhave been arrived from α =iiand earthquake event weighting factor (χ∑ Li ) has been taken as the ratio ofithe past events associated with source i to the total number of events in the region as given below:Number of earthquakeclosetothe sourceχi=(3.22)Total number of earthquake inthe regionThe recurrence relation of source i have been arrived by averaging both weighting and multiplying the regionalrecurrence relation as given below:Ni( m0 ) = 0.5( αi+ χ ) N(m0)(3.23)iThe weighting factors calculated for each source have considered the source length and number of events,which are shown in Table 3.9. Finally, the probability density function (PDF) for each source has beenevaluated. Typical plot of PDF versus magnitude is shown in Figure 3.22.0.60Probability density fuction0.500.400.300.200.100.004.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50Magnitude. MFigure3.22: Typical probability density function of magnitude50

Probabilistic Seismic Hazard Analysis3.14.3 Uncertainty in the Hypocentral DistanceIn the PSHA, the other uncertainty involved is the distance of each source to the site. In a seismogenic source,each point/segment of the source can rupture and generate an earthquake. The geometries of seismic sourcesdepend on the tectonic processes involved in their formation. The seismic sources like faults and lineaments,earthquake can occur at any point along their length which can be referred as area source or linear sources.Earthquakes are usually assumed to be uniformly distributed with in a particular fault or lineaments. A uniformdistribution of source to site distance is expressed in ground motion parameter in terms of some measure ofsource to site distance; the uncertainty must be described with respect to the appropriate distance parameter.The uncertainty involved in the source to site distance is described by a probability density function. Thus therelative orientation of each source with respect to the site becomes important. The shortest and longest distancefrom each source to city center has been evaluated from Bangalore seismotectonic map published by Sitharamand Anbazhagan (2007). The hypocentral distance has been evaluated by considering focal depth of 15km fromthe ground level similar to the one used for DSHA in Sitharam et al (2006) and Sitharam and Anbazhagan(2007). A probable source zone depth of 10km has been considered by Bhatia et al (1997) in an exercise todevelop seismic hazard map of the shield region of India in GSHAP. However, in this study a source zone depthof 15km has been considered. Shortest and longest hypocentral distances are presented in Table 3.9. Theprobability distribution for the hypocenter distance, from any site to earthquake rupture on the source, iscomputed conditionally for the earthquake magnitude. Generally, the rupture length is a function of magnitude.The conditional probability distribution function of the hypocentral distance R for on earthquake magnitudeM=m for a ruptured segment, is assumed to be uniformly distributed along a fault and is given by Kiureghianand Ang (1977),which is as follows:2 2 1/ 2P ( R < r | M = m)= 0 for R < ( D + L0)(3.24)2 2 1/ 2( r − d ) − L0P(R < r | M = m)=forL − X ( m)2 222L { [ ] } 1/ 20) ≤ R < D + L + L0− X ( )22P ( R < r | M = m)= 1 for R { D + [ L + L − X (m)]} 1/ 2( D + m(3.25)>0(3.26)Where X(m) is the rupture length in kilometers for the event of magnitude m is estimated using the Wells andCoppersmith (1994) equation, which is as given below:iX ( m)min[ 10fault length]( −2.44+ 0.59( m ))= ,(3.27)The notations used in the equations 3.24, 3.25 and 3.26 are explained in Figure 3.23. Typical probability densityfunction of the hypocentral distance for Mandya-Channapatna- Bangalore lineament (L15) is shown in Figure3.24.SiteSourcerRuptureXhDr oL 0LdFigure 3.23: Schematic representation of fault rupture model51

Probabilistic Seismic Hazard Analysis0.070.06Probability Density Function0.050.040.030.020.01040 45 50 55 60 65 70 75 80 85 903.14.4 Probability of Ground MotionDistance (km)Figure 3.24: Typical PDF for the source L15Among the critical elements required in seismic hazard analysis, the attenuation relationship of peak horizontalacceleration (PGA) is very important. Ground motion attenuation relationship express the variation of peakground acceleration at specific structural periods of vibration and damping ratios with earthquake magnitudeand source to site distance. Strong ground motions depends on the characteristics of the earthquake source, thecrustal wave propagation path, and the local site geology. Accordingly, attenuation relations typically aredeveloped for specific tectonic environments. As the study area is located in peninsular India, the attenuationrelation (for peak ground acceleration and spectral acceleration) for the rock site in Peninsular India developedby RaghuKanth and Iyengar (2004) and RaghuKanth (2005) has been used here:2ln y = c1 + c2( M − 6) + c3( M − 6) − ln R − c4R+ ln( ∈)(3.28)Where y, M, R and ∈ refer to PGA/spectral acceleration (g), moment magnitude, hypocentral distance and errorassociated with the regression respectively. The coefficients in equation (3.28), c 1 , c 2, c 3, and c 4 are obtainedfrom RaghuKanth and Iyengar (2004) and RaghuKanth (2005). The coefficients corresponding to PGA is givenin equation 3.2 and for spectral acceleration is given in Table 3.11.Table 3.11: Coefficients of spectral acceleration equation for rock for southern IndiaPeriod(s) c 1 c 2 c 3 c 4 σ (ln ε br )0.0000 1.7816 0.9205 −0.0673 0.0035 0.31360.0100 1.8375 0.9196 −0.0666 0.0035 0.31720.0150 1.9657 0.9136 −0.0643 0.0036 0.33830.0200 2.2153 0.9054 −0.0607 0.0037 0.39200.0300 2.7418 0.8988 −0.0570 0.0037 0.31710.0400 2.9025 0.9034 −0.0578 0.0036 0.33440.0500 2.8652 0.9113 −0.0604 0.0035 0.30000.0600 2.7795 0.9202 −0.0637 0.0034 0.29170.0750 2.6483 0.9343 −0.0693 0.0032 0.286552

Probabilistic Seismic Hazard Analysis0.0900 2.5333 0.9492 −0.0757 0.0031 0.28250.1000 2.4651 0.9595 −0.0803 0.0030 0.28010.1500 2.1941 1.0139 −0.1058 0.0027 0.27030.2000 1.9917 1.0708 −0.1331 0.0025 0.26370.3000 1.6832 1.1830 −0.1846 0.0021 0.25630.4000 1.4379 1.2859 −0.2269 0.0019 0.25100.5000 1.2262 1.3770 −0.2592 0.0017 0.24500.6000 1.0361 1.4571 −0.2830 0.0015 0.23860.7000 0.8621 1.5276 −0.3001 0.0014 0.23230.7500 0.7800 1.5598 −0.3067 0.0013 0.22900.8000 0.7008 1.5900 −0.3121 0.0013 0.22680.9000 0.5501 1.6456 −0.3203 0.0012 0.22251.0000 0.4087 1.6955 −0.3255 0.0012 0.21941.2000 0.1489 1.7814 −0.3298 0.0011 0.21631.5000 −0.1943 1.8847 −0.3268 0.0010 0.21752.0000 −0.6755 2.0119 −0.3105 0.0001 0.22652.5000 −1.0762 2.1041 −0.2895 0.0010 0.23653.0000 −1.4191 2.1741 −0.2680 0.0010 0.24474.0000 −1.9847 2.2730 −0.2287 0.0011 0.2544In hazard analysis the ground motion parameters are estimated from the predictive ground motion relation(equation 3.28) in terms of PGA and spectral acceleration. These attenuation relations are obtained fromregression, which is associated with the randomness of predictive equations. Uncertainty involved in theseequations can be accounted by calculating the probability of exceedance of a particular value by the attenuationequation. The normal cumulative distribution function has a value which is most efficiently expressed in termsof the standard normal variables (z) which can be computed for any random variables using transformation asgiven below (Kramer, 1996):ln PGA − ln PGAz = (3.29)σln PGAWhere ln PGA is the various targeted peak accelerations level will be exceeded, ln PGA is the accelerationcalculated using attenuation relationship andσ ln PGAis the uncertainty in the attenuation relation expressed bythe standard deviation.3.14.5 Hazard CurvesThe summation of all the probabilities is termed as hazard curve, which is plotted as the mean annual rate ofexceedance (and its reciprocal is defined return period) versus the corresponding ground motion. The meanannual rate of exceedance has been calculated for six sources separately and summation of these representingthe cumulative hazard curves. To carryout the repetitive calculation involved in the PSHA, a program has beendeveloped. The mean annual rate of exceedance versus peak horizontal acceleration for all the sources at rocklevel is shown in Figure 3.25. This clearly highlights that the sources close to Bangalore produce more hazardwhen compared to the source far away from Bangalore. The return periods corresponding to PGA at rock levelfor Mumbai and Bangalore are presented in the Table3.12. The return period for Bangalore is slightly morewhen compare to Mumbai. Further to define the seismic hazard at rock level for the study area, PGA at eachgrind point has been estimated. These values are used to prepare PGA distribution maps for 2% and 10%probability exceedance in 50 years, which corresponds to return periods of 2475 and 475 years. Rock levelPGA distribution map for 2% and 10% probability for Bangalore is shown in Figures 3.26 and 3.27, PGAvalues varies from 0.17g to 0.25g. These values are comparable with the PGA map at rock level published bySitharam et.al., (2006) using deterministic approach. Even though PGA is used to characterize the groundmotion, the spectral acceleration is generally used for design of engineering structures. Similar to mean annualrate of exceedance calculation for PGA, the spectral acceleration at period of 1 second and 5% damping are53

Probabilistic Seismic Hazard AnalysisTable 3.12: Return periods for Mumbai and BangalorePGA (g)Return period (years)Mumbai* Bangalore0.1 606 6660.2 3225 46720.3 11337 166660.4 30959 44444Mean Annual rate of Exceedance1.0E+001.0E-011.0E-021.0E-031.0E-041.0E-051.0E-06L15F47F19L16L20L22Cumulative0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6Peak Ground Acceleration (g)Figure 3.25: Hazard curve in terms of PGA at rock level for Bangaloreevaluated for all the sources. Cumulative mean annual rate of exceedance versus spectral acceleration for aperiod of 1 sec and 5% damping (represented as hazard curve) is shown in Figure 3.28. The return periods forMumbai and Bangalore corresponding to spectral acceleration at rock level for a period of 1 sec and 5%damping are presented in Table 3.13. The return periods for Bangalore are slightly large when compared toMumbai. For the design of structures, a uniform hazard response spectrum (UHRS) / equivalent hazardspectrum is used. UHRS is developed from a probabilistic ground motion analysis that has an equal probabilityof being exceeded at each period of vibration. For finding the UHRS, seismic hazard curves of spectralacceleration (S a ) are computed for a range of the frequency values. From theses hazard curves, response spectrafor a specified probability of exceedance over the entire frequency range of interest are evaluated. The responsespectrum of Bangalore at rock level for 5% damping has been evaluated for 10% probability of exceedance in50 years. Figure 3.29 shows the UHRS (with damping of 5%) at rock level for 10% probability of exceedance in50 years. The shape of the spectrum compared is similar to the Mumbai city UHRS developed by RaghuKanthand Iyengar (2006). The zero-period spectral acceleration (ZPA = PGA) evaluated in this study (0.121g) iscomparable with the value obtained in deterministic approach (0.129g).54

Probabilistic Seismic Hazard Analysis0.205Scale1:20,0000.2100.2150.2200.2000.2300.2250.2350.2400.1950.2450.1900.1850.1800.1750.1700.2500.255Figure 3.26: Peak ground acceleration contours at rock level with 10% probability of exceedance in 50years.Acceleration in gFigure 3.27: Peak ground acceleration contours at rock level with 10% probability of exceedance in 50years.55

Probabilistic Seismic Hazard AnalysisTable 3.13: Return periods for Mumbai and BangaloreSpectralacceleration (g)Mumbai*Return period (years)Bangalore0.1 550 10000.2 3030 840340.3 14104 465116*Data from Raghukanth and Iyengar (2006)Mean Anuual rate of Exceedance1.E+001.E-011.E-021.E-031.E-041.E-051.E-061.E-071.E-08F47L15L16F19L20L22Cumulative0 0.1 0.2 0.3 0.4 0.5 0.6Spectral Aceeleration (g)Figure 3.28: Spectral acceleration at rock level corresponding to period of 1s and 5% damping forBangalore.Spectral acceleration (g)0.450.40.350.30.250.20.150.10.0500 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2period (sec)Figure 3.29: Uniform Hazard Response Spectrum at rock level with 10% probability of exceedance in 50years (5% damping) for Bangalore.56

Probabilistic Seismic Hazard AnalysisThis clearly brings out that the probabilistic method is inclusive of all the deterministic events with finiteprobability of occurrence and both PSHA and DSHA methods complement each other to provide additionalinsights to the seismic hazard problem. Also the reported PGA values are quite higher than the GSHAP study ofIndia (which considers the focal depth of 10km). From this study, PGA at rock level for the focal depth of 15kmis much higher than PGA obtained in GSHAP for this area. If the focal depth of 10km is considered, hazardvalues will be much higher for this area. Based on this study, it is recommended that study area aroundBangalore need to be upgraded from zone II to zone III in Indian seismic code BIS1893-2002 as highlightedearlier by Ganesha Raj and Nijagunappa (2004) and Sitharam et al (2006).3.15 SUMMARY FROM PROBABILISTIC APPROACHThe analysis of historical earthquake data as performed in this paper, lends support to the fact that Bangaloreregion has been in an area of increased earthquake activity especially during the last few decades. This chapterpresents the probabilistic seismic hazard analysis for Bangalore along with completeness analysis of seismicdata and estimation of seismic hazard parameter. The following conclusions are arrived from this study.From the analysis, it was found that the seismic data is homogenous for the last four decadesirrespective of the magnitude.The analysis of historical data of earthquakes around Bangalore has shown increased seismic activityduring the last few decades.The frequency magnitude relationship has been established for the study area after carrying out thecompleteness analysis as per Stepp (1972).Completeness of the data has been observed for the last 4 decades for Bangalore region and a seismichazard parameter of “b value” estimated using G-R relation varies from 0.87 to 0.92 based on themagnitudes.Whereas, a “b value” of 0.87 ±0.03 was obtained based on the analysis of Kijko and Sellevoll, (1989,1992) using mixed data set.The probabilistic seismic hazard analysis for Bangalore city has been carried out using the regionalrecurrence relation with appropriate deaggregation weighting factors for the seismogenic sources inthe study area.The focal depth of 15km is taken as probable source zone depth in the study area. If a focal depth of10km is used (as considered in GSHAP study for the shield region of south India by Bhatia et. al.,1997) the PGA and spectral acceleration values would have been increased considerably.The curves of mean annual rate of exceedance for peak horizontal acceleration and spectralacceleration have also been generated at rock level.The rock level PGA map for 2% and 10% probability of exceedance in 50 years corresponding to thereturn period of 2475 and 475years has been presented..The study brings out that both the probabilistic and deterministic approaches will lead to similaranswers complementing each other and provides additional insight to the seismic hazard assessment.Further, the uniform hazard response spectrum at rock level with 5% damping for Bangalore has beengenerated for 10% probability of exceedance in 50 years. The shape of response spectrum developedin this study is similar to the one presented by other researchers for sites in southern India.The hazard values presented here are much higher than what is reported by GSHAP study of Bhatia etal (1997) for the study area.57

CHAPTER 4SITE CHARACTERIZATION USING GEOTECHNICAL ANDGEOPHYSICAL TECHNIQUE4.1 SITE CHARACTERIZATION USING GEOTECHNICAL DATAFor the complete microzonation and site response study the subsurface material has to be characterizedconsidering the local sites. This chapter presents the subsurface soil properties using geotechnical borehole,geotechnical properties and SPT ‘N’ values with depth for site characterization. This chapter describes the borelogs of geotechnical data by developing a geographical information system (GIS) based subsurface model. Thegeotechnical data was collected from Archives of Torsteel Research Foundation, Bangalore and Indian Instituteof Science, Bangalore. GIS based model helps in data management, to develop geostatistical functions, 3-dimentional (3-D) visualization of subsurface with geo-processing capability and future scope for web basedsubsurface mapping tool. This is envisaged not only in economizing geotechnical investigations and to aid indetailed site investigations for major projects. The three major tasks carried out are as presented below:(1) Development of digitized map of Bangalore city with several layers of information. (2) Development of GISdatabase for collating and synthesizing geotechnical data available with different sources.(3) Development of 3-dimensional view of subsoil stratums presenting various geotechnical properties such aslocation details, physical properties, grain size distribution, Atterberg limits, SPT ‘N’ values and strengthproperties for soil and rock along with depth in appropriate format. GIS has been used in geosciences for datacollection, organization and distribution of geological data and maps. Visualization is another GIS activity,which is very useful for the purpose of microzonation. ArcGIS 8.1 has been used here with 3-D analystextension.The rock level spatial variability of the bed/hard rock with reference to ground surface is vital for manyapplications in geosciences. Rock depth is a very useful parameter to the geotechnical earthquake engineers tofind their basic requirement of hard stratum and engineering rock depth for site response studies. In most of thegeotechnical investigations, knowledge of the hard strata or rock is essential to decide the type of foundationsand design a suitable foundation for a structure. In the ground response analysis, Peak Ground Acceleration(PGA) and response spectrum for the particular site is evaluated at the rock depth levels and further on at theground level considering local site effects. In ground response analysis, the response of the soil deposit isdetermined from the motion at the bed rock level. In all these problems, it is essential to evaluate the depth ofthe hard rock from the ground level. With an objective of predicting the spatial variability of the reduced levelof the bed/hard rock in Bangalore, an attempt has been made to develop model using Artificial Neural Network(ANN). It is also aimed at comparing the performance of the developed model based on ANN for the availabledata in Bangalore.4.2 PREPARATION OF BASE MAPThe Bangalore map forms the base layer for the development of GIS model. The map entities were developed inview of two aspects, first, for locating the bore logs to the utmost accuracy on a scale of 1:20000 and secondlyfor identification of bore logs by the end user. With this scope of work, the map was developed with severallayers of information. Some of the important layers considered are the boundaries (Outer and Administrative),Contours, Highways, Major roads, Minor roads, Streets, Rail roads, Water bodies, Drains, Landmarks and Borelocations. Digitized map was developed mainly using hard copy of Bangalore guide map, published by Surveyof India in 1983 and several other maps from standard publishers were used for reference. Digitization of themap layers was done in AutoCAD and then imported to Arc view GIS 8.1. Few combinations of layers that canbe viewed in the map for various information are shown in Figures 4.1 and 4.2. Figure 4.1 depicts the locationof boreholes with respect to water features like tanks, lakes and drains within the corporate boundary ofBangalore along with outer boundary circumscribing the ring road. It gives a clear view of the spatialdistribution of boreholes in Bangalore region. Figure 4.2 shows the location of boreholes58

Site Characterization using Geotechnical Borehole DataDrainsNWNEVidhana SoudhaLat-Long:(77 o35.46’; 12° 58.67')CorporateBoundaryOuterBoundarySWWaterBodiesFigure 4.1: GIS model of borehole locations along with water body featuresSEBoreholeLocationContours withinterval of 10mBoreholelocationsFigure 4.2: GIS model of borehole locations along with contourswith the ground elevation contours at 10m interval. This would give information on availability of reliablegeotechnical data on sloping terrains or valleys and could be used for land use planning or other planningpurposes. Several other layers of information are also available on the map like road (Highways, streets etc) andrail network, bridges, culverts, Landmarks and monumental structures etc, which would cater to the interest of59

Site Characterization using Geotechnical Borehole Datamany other engineering groups. Also the isometric view of all boreholes can be viewed by overlapping of layersto get a 3-D projection. This view gives a visual idea of the depth to which geotechnical data is available in eachborehole.4.3 GEOTECHNICAL DATA AND DISTRIBUTIONGeotechnical borelog data was basically collated from archives of Torsteel Research Foundation in India (TRFI)and Indian Institute of Science (IISc) from geotechnical investigations carried out for several major projects inBangalore. The data collected is carried out by senior geotechnical engineers for important projects in Bangaloreduring the years 1990-2003. So far about 850 borelogs information has been keyed into the database. Most ofthe data so far selected for the database is on an average to a depth of 20m below the ground level with aninterval of 0.5m. The bar chart shown in Figure 4.3 clearly gives the distribution of the boreholes classifiedbased on the depth of borehole below ground level. Majority of the bores with depths greater than 15m werecarried out for several grade separator projects. Most investigations for residential and commercial complexeswere below 15m. But wherever bedrock has been encountered investigation has been terminated at that depth.The properties keyed into the database at a particular depth are location details, physical properties, grain sizedistribution, Atterberg limits, and strength properties for soil and rock. These boreholes spatially cover mostparts of the Bangalore city and more densely in the areas of high land use as shown in Figures 4.1 and 4.2.Distribution of boreholes in four quadrants of Bangalore city considering Vidhana Soudha (shown in Figure 4.1)as the centre point is shown in Figure 4.4. There is greater scope for geostatistical analysis of the collated data tostudy the subsurface features and their variation. The ‘Standard Penetration Test’, commonly known as the‘SPT’, is carried out in a borehole, by driving a standard ‘split spoon’ sampler using repeated blows of a 63.5kghammer falling through 762mm. The hammer is operated at the top of the borehole, and is connected to the splitspoon sampler by rods.Nos of boreholes50045040035030025020015010050018230Bore depth in metersFigure 4.3: Distribution of boreholes based on depthThe split spoon sampler is lowered to the bottom of the hole, and is then driven a distance of 450mm in three150mm intervals, and the blows are counted for each 150mm penetration. The penetration resistance (N) is thenumber of blows required to drive the split spoon for the last 300mm of penetration. The penetration resistanceduring the first 150 mm of penetration is ignored, because the soil is considered to have been disturbed. SPTborelog data includes location of wells, SPT test results, ground water level, grain size distribution, Liquid limit,Plastic limit and strength properties for soil and rock. Figure 4.5 shows the rock level and soil overburdenthickness of Bangalore; this view gives a visual idea of the depth of the soil overburden available in eachborehole locations.60

Site Characterization using Geotechnical Borehole DataNW;15819%NE;18221%SW;26031%SE;25029%Figure 4.4: Distribution of boreholes in different quadrants of Bangalore city4.4 SUBSURFACE 3-D MODELFigure 4.5: Rock level and overburden thicknessThe GIS model developed currently consists of 850 borehole locations marked on the digitized Bangalore mapof 1:20000 scale. The boreholes are represented as 3- dimensional objects projecting below the map layer in0.5m intervals. Also image files of bore logs and properties table has been attached to location in plan. These 3-D boreholes are generated with several layers with a bore location in each layer overlapping one below the otherand each layer representing 0.5 m interval of the subsurface. Figure 4.6 shows a typical borehole viewed inisometric view. It consists of several donut elements in different layers placed coinciding one below the other. Asingle donut as shown on the right in Figure 4.6 represents 0.5 m depth of ground in the model. Topmost donutrepresents the 0.5m depth of surface strata and thereon each donut cumulates to 0.5m below ground level. Eachborehole in this model is attached with geotechnical data along the depth. The data consists of visual soilclassification, standard penetration test results, ground water level, time during which test has been carried out,other physical and engineering properties of soil (Table 4.1). The model provides two options to view the data ateach borehole in order to cater for various groups. In 2-D, by clicking on a borehole will display the standardbore log information as shown in Figure 4.7 and the respective properties table consisting of index properties &shear strength parameters. Apart from this each donut of any borehole is attached with soil/rock properties atthat particular depth. As such when this model is viewed in 3-D, geotechnical information on any borehole atany depth can be obtained by clicking at that level (donut). Figure 4.8 shows a view of some boreholes belowBangalore city map to get a 3-D projection. The data consists of visual soil classification, standard penetrationtest results, ground water level, time during which the test has been carried out, other physical and engineeringproperties of soil. Typical soil profiles for the purpose of general classification of soil layers the Bangalore areais shown in Table 4.2.61

Site Characterization using Geotechnical Borehole Data0.5 mSingle layer of bore representing0.5m depthFigure 4.6: A Typical borehole in 3-DTable 4.1: Bore log details at particular depth1. BH No. 14.Plastic limit2. Location 15.Cohesion3. Ground RL 16.Angle of internal friction4. Depth 17.Soil specific gravity5. Ground water table 18.Compression index6. Duration 19.Pre consolidation pressure7. Visual soil Classification 20.Permeability8. Thickness 21.Safe bearing capacity9. SPT 22.Water absorption10.Bulk density11.Water content12.Grain size distribution (Percentage gravel, sand,sit and clay)13.Liquid limit23.Rock density24.Rock specific gravity25. Rock unconfined compressivestrength26. Rock Porosity62

Site Characterization using Geotechnical Borehole DataTable 4.2: General soil distribution in BangaloreLayerFirst LayerSecondlayerThirdLayerFourthlayerFifth LayerSoil Description with depth and DirectionNorthwest Southwest Northeast SoutheastSilty sandwith clay0-3mMedium todense silty sand3m-6mWeatheredRock6m-17mHard RockBelow the 17mSilty sand withgravel0-1.7mClayey sand1.7m-3.5mWeatheredRock3.5m-8.5mHard RockBelow 8.5mClayey sand0-1.5mClayey sandwith gravel1.5m-4mSilty sand withGravel4m-15.5mWeathered rock15.5m-27.5mHard Rock Hard Rock Hard RockBelow 27.5mFilled up soil0-1.5mSilty clay1.5m-4.5mSandy clay4.5m-17.5mWeatheredRock17.5m-38.5mHard RockBelow 38.5mFigure 4.7: A typical display of bore log sheet63

Site Characterization using Geotechnical Borehole DataFigure 4.8: GIS model of borehole locations in 3-D view64

Site Characterization using Geotechnical Borehole Data4.5 ARTIFICIAL NEURAL NETWORK (ANN) FOR ROCK DEPTHThe Artificial Neural Network (ANN) is used as a tool to predict reduced level of rock or rock depth values ofBangalore considering a large data distributed over 220 sq.km area. For ANN model, the procedures todetermine data division, data normalizing technique, network architecture selection, transfer function and no ofepochs are outlined. In order to compare between the ANN model and actual values five points have beenchosen randomly from known reduced level of rock values of 652 points in the subsurface model of Bangalore.The predicted values of these points are shown in Table 4.3. It can be seen from the table that the ANN modelhas given best prediction.Table 4.3: Comparison between actual and ANN predicted valuesBoreholenoLongitude(degree) ELatitude(degree) NActual reduced levelof rock(m)Predicted reduced level ofrock(m) by ANN71-2 77.5765 12.9448 907.64 908.8553-6 77.6237 12.9447 901 899.74725-39B 77.6641 12.9924 905.86 904.8551-9 77.5874 12.9331 927 926.8087-4 77.5368 13.0293 889 889.134.6 CORRECTIONS APPLIED FOR SPT “N” VALUESThe SPT data collected is field ‘N’ values with out applying any corrections. Usually for engineering use of siteresponse study and liquefaction analysis the SPT “N” values need to be corrected with various corrections and aseismic borelog has to be obtained. The seismic bore log contain information about depth, observed SPT ‘N’values, density of soil, total stress, effective stress, fines content, correction factors for observed “N” values, andcorrected “N” value. The ‘N’ values measured in the field using Standard penetration test procedure have beencorrected for various corrections, such as:(a) Overburden Pressure (C N ), (b) Hammer energy (C E ), (c) Boreholediameter (C B ), (d) presence or absence of liner (C S ), (e) Rod length (C R ) and (f) fines content (C fines ) (Seed et al.,1983, 1985; Youd et al., 2001; Cetin et al., 2004, Skempton; 1986 and Pearce and Baldwin, 2005). Corrected‘N’ value i.e., (N 1 ) 60 is obtained using the following equation:N ) = N × ( C × C × C × C × C )(4.1)(1 60N E B S R4.6.1 Correction for Overburden PressureThe effective use of SPT blow count for seismic study requires that the effects of soil density and effectiveconfining stress on penetration resistance be separated. Consequently, Seed et al (1975) included thenormalization of penetration resistance in sand to an equivalentσ of one atmosphere as part of the semiempirical procedure. SPT N-values recorded in the field increase with increasing effective overburden stress;hence overburden stress correction factor is applied (Seed and Idriss 1982). This factor is commonly calculatedfrom equation developed by Liao and Whitman (1986). However Kayen et al. (1992) suggested the followingequation, which limits the maximum C N value to 1.7, provides a better fit to the original curve specified by Seedand Idriss (1982):'CN= 2.2 /(1.2 + σνo/ Pa)(4.2)'Where, σvo= effective overburden pressure, P a = 100 kPa, and C N should not exceed a value of 1.7. Thisempirical overburden correction factor is also recommended by Youd et al (2001). For high pressures (300kPa),which are generally below the depth for which the simplified procedure has been verified, C N should beestimated by other means (Youd et al, 2001).4.6.2 Correction for hammer energy ratioAnother important factor which affects the SPT –N value is the energy transferred from the falling hammer tothe SPT sampler. The energy ratio (ER) delivered to the sampler depends on the type of hammer, anvil, lifting'vo65

Site Characterization using Geotechnical Borehole Datamechanism, and the method of hammer release. Approximate correction factors to modify the SPT results to a60% energy ratio for various types of hammers and anvils are listed in Table 4.4 (Robertson and Wride 1998).Because of variations in drilling and testing equipment and differences in testing procedures, a rather wide rangein the energy correction factor C ER has been observed as noted in the table. Even when procedures are carefullymonitored to confirm the established standards some variation in C E may occur because of minor variations intesting procedures. Measured energies at a single site indicate that variations in energy ratio between blows orbetween tests in a single borehole typically vary by as much as 10%. The workshop participants of NCEER1996&1998 (Youd et al, 2001) recommend measurement of the hammer energy frequently at each site wherethe SPT is used. Where measurements cannot be made, careful observation and notation of the equipment andprocedures are required to estimate a C E value. Use of good-quality testing equipment and carefully controlledtesting procedures will generally yield more consistent energy ratios. For Liquefaction calculation Yilmaz andBagci (2006) had taken the C E value as 0.7 for SPT hammer energy donut type for soil liquefactionsusceptibility and hazard mapping in Kutahya, Turkey. Similar kind of hammer is used for soil investigations,hence the value of 0.7 is taken for C E.Table 4.4: Hammer correction factors (Robertson and Wride 1998)Type of Hammers Notation Range of correctionDonut Hammer C E 0.5-1.0Safety Hammer C E 0.7-1.2Automatic-trip Donut Hammer C E 0.8-1.34.6.3 Other correction factorsThe other correction factors adopted such as correction for borehole diameter, rod length and sampling methodsmodified from Skempton (1986) and listed by Robertson and Wride (1998) are presented in Table 4.5.Correction for bore hole diameter (C B ) is used as 1.05 for 150 mm borehole diameter, Rod length (C R ) is takenfrom the table based on the rod length, and presence or absence of liner (C S ) is taken as 1.0 for standard sampler.The corrected “N” Value (N 1 ) 60 is further corrected for fines content based on the revised boundary curvesderived by Idriss and Boulanger (2004) for cohesionless soils as described below:(4.3)( Ncs∆ N1)60= ( N1)60+ (1)602⎡ 9.7 ⎛ 15.7 ⎞ ⎤∆(N 1)60= exp⎢1.63+− ⎜ ⎟ ⎥⎢⎣FC + 0.001 ⎝ FC + 0.001⎠(4.4)⎥⎦FC = percent fines content (percent dry weight finer than 0.074mm).A typical corrected bore log generated for a borehole data is shown in Table 4.6; corrected borelog has beengenerated for all the bore holes using Windows macros in excel.Table 4.5: Correction factors for Borehole Diameter (C B ), Rod Length (C R ) andSampling Method (C S )Factor Equipment Variable Notation CorrectionBorehole Dia. 65-115mm C B 1.00Borehole Dia. 150mm C B 1.05Borehole Dia. 200mm C B 1.15Rod Length

Site Characterization using Geotechnical Borehole DataTable 4.6: Typical “N” correction table for borelogBorehole Water Table = 1.4 m/19-11-2005Corrected NDepth Field Density T.S E.S Correction Factors For F.CvalueC N (N 1 ) 60 ∆ (N 1) 60m N Value kN/m 3 kN/m 2 kN/m 2 Hammer Bore hole Rod Sample%(NEffect Dia Length Method1 ) 60cs1.50 19 20.00 30.00 30.00 1.47 0.7 1.05 0.75 1 15.36 48 5.613 213.50 28 20.00 70.00 50.38 1.29 0.7 1.05 0.8 1 21.26 43 5.597 274.50 26 20.00 90.00 60.57 1.22 0.7 1.05 0.85 1 19.79 60 5.602 256.00 41 20.00 120.00 75.86 1.12 0.7 1.05 0.85 1 28.77 48 5.613 347.50 55 20.00 150.00 91.14 1.04 0.7 1.05 0.95 1 40.02 37 5.541 469.00 100 20.00 180.00 106.43 0.97 0.7 1.05 0.95 1 67.84 28 5.270 7310.50 100 20.00 210.00 121.71 0.91 0.7 1.05 1 1 66.90 28 5.270 7212.50 100 20.00 250.00 142.09 0.84 0.7 1.05 1 1 61.70 28 5.270 67T.S - Total StressE.S - Effective StressC N- Correction for overburden correction(N 1 ) 60 - Corrected ‘N’ Value before correction for fines contentF.C - Fines content∆ (N ) - Correction for Fines content1 60(N 1 ) 60cs - Corrected ‘N’ Value67

Site characterization using Multichannel Analysis of Surface Wave (MASW) Survey4.7 SUMMARY FROM GEOTECHNICAL TECHNIQUEDigitized map of Bangalore on a scale of 1:20000 with about 12 layers of information has been prepared andpresented. Geotechnical data for 850 boreholes up to a depth of about 40m has been collated and attached to thelocations in the map using ARC View GIS 8.1 with 3D analyst package. 3-D model has been prepared bysuperposing Bangalore city map with important layers such as buildings, streets, roads, railway network, waterbodies, contours and bore logs with geotechnical test data. The SPT “N” value from 3-D subsurface model wascorrected by applying necessary corrections. These geotechnical data and corrected “N” can be further used togenerate correlation between measured shear wave velocity and corrected (N 1 ) 60cs SPT “N” values and alsoliquefaction hazard mapping. These geotechnical data and corrected “N” can be further used to generatecorrelation between measured shear wave velocity and corrected (N 1 ) 60cs SPT “N” values and also liquefactionhazard mapping. The ANN model is used as tool to predict reduced level of rock or rock depth. The ANN modeldeveloped with the available data for predicting reduced level of rock values in the subsurface of Bangalore,reasonably predicts the bedrock depth.4.8 SITE CHARACTERIZATION USING MULTICHANNEL ANALYSIS OFSURFACE WAVE (MASW) SURVEYIncreasingly being applied to earthquake geotechnical engineering for microzonation and site response studies.In particular, the MASW is used in geotechnical engineering for the To ascertain the manifestation ofearthquake shaking on the ground surface, site characterization by evaluating velocity parameters at a shallowlevel would be of consequence. Shear wave velocity (Vs) is an essential parameter for evaluating the dynamicproperties of soil in the shallow subsurface. A number of geophysical methods have been proposed for nearsurfacecharacterization and measurement of shear wave velocity by using a great variety of testingconfigurations, processing techniques, and inversion algorithms. The most widely-used techniques are SASW(Spectral Analysis of Surface Waves) and MASW (Multichannel Analysis of Surface Waves). The spectralanalysis of surface wave (SASW) method has been used for site investigation for several decades (e.g., Nazarianet al., 1983; Al-Hunaidi, 1992; Stokoe et al., 1994; Tokimatsu, 1995; Ganji et al., 1997). SASW method uses thespectral analysis of a surface wave generated by an impulsive source and recorded by a pair of receivers.Evaluating and distinguishing signal from noise with only a pair of receivers by this method is difficult. Thus toimprove inherent difficulties, a new technique incorporating multichannel analysis of surface waves using activesources, named as MASW, was developed (Park et al., 1999; Xia et al., 1999; Xu et al., 2006). The MASW hasbeen found to be a more efficient method for unraveling the shallow subsurface properties (Park et al., 1999;Xia et al., 1999; Zhang et al., 2004). Multichannel Analysis of Surface wave (MASW) is measurement of shearwave velocity and dynamic properties, identification of subsurface material boundaries and spatial variations ofshear wave velocity. MASW is a non-intrusive and less time consuming geophysical method. It is a seismicmethod that can be used for geotechnical characterization of near surface materials (Park et al, 1999; Xia et al,1999; Miller et al, 1999a; Park et al, 2005a; Kanli et al, 2006).MASW identifies each type of seismic waves on a multichannel record based on the normal pattern recognitiontechnique that has been used in oil exploration for several decades. The identification leads to an optimum fieldconfiguration that assures the highest signal-to-noise ratio (S/N). Effectiveness in signal analysis is then furtherenhanced by in the data processing step (Ivanov et. al, 2005). MASW is also used to generate a 2-D shear wavevelocity profile. In this chapter, the average shear wave velocity for 5m, 10m, 15m, 20m, 25m and 30m (Vs 30 )has been evaluated and mapped for Bangalore covering an area of 220sq.km in Bangalore Municipalcorporation limits. The study has been carried out for assigning soil classification for seismic local site effectevaluation and also to generate the correlation between corrected SPT ‘N’ value [(N 1 ) 60cs ] available in the studyarea and measured shear wave velocity (Vs).4.9 TESTING PROGRAMMEThe MASW test locations are selected based on the space required for the testing and close to importantbuilding such as hospitals, temples, government buildings and schools. About 58 one-dimensional (1-D)MASW surveys and 20 two-dimensional (2-D) MASW surveys have been carried out. The locations of thetesting points in Bangalore are shown in Figure 4.9. The test locations are selected such a way that theserepresent the entire city subsurface information. In total 58 one-dimensional (1-D) surveys and 20 twodimensional(2-D) surveys have been carried out. Most of the survey locations are selected in flat ground andalso in important places like parks, hospitals, schools and temple yards etc. In about 38 locations MASW survey68

Site characterization using Multichannel Analysis of Surface Wave (MASW) Surveypoints are very close to the available SPT borehole locations and these are used to generate correlation betweenshear wave velocity and corrected SPT ‘N’ values .Bangalore Mahanagar Palike1-D MASW2-D MASWMASW TestsScale 1:20,0004.10 MASW TESTINGFigure 4.9: MASW testing locations in Bangalore cityMASW is a geophysical method, which generates a shear-wave velocity (Vs) profile (i.e., Vs versus depth) byanalyzing Raleigh-type surface waves on a multichannel record. MASW system consisting of 24 channelsGeode seismograph with 24 geophones of 4.5 Hz capacity have been used in this investigation. The seismicwaves are created by impulsive source of 10 pound (sledge hammer) with 300mmx300mm size hammer platewith ten shots. These waves are captured by 24 geophones/receivers of 4.5Hz capacity. The complete MASWinstrument with accessories is shown in Figure 4.10a. The capturedRayleigh wave is further analyzed using SurfSeis software. The test locations are selected in such a way thatthese represent the entire city subsurface information (see Figure 6.1). A typical testing arrangement in the fieldis shown in Figure 4.10b. The captured Rayleigh wave is further analyzed using SurfSeis software. SurfSeis isdesigned to generate Vs data (either in 1-D or 2-D format) using a simple three-step procedure: i) preparation ofa Multichannel record (some times called a shot gather or a field file), ii) dispersion-curve analysis, and iii)inversion. The term “Multichannel record” indicates a seismic data set acquired by using a recording instrumentwith more than one channel using geode seismograph. MASW has been effectively used with highest signal-tonoiseratio (S/N) of surface waves. The optimum field parameters such as source to first and last receiver,receiver spacing and spread length of survey lines are selected in such a way that required depth of informationcan be obtained. These are in conformity with the recommendations of Park et al. (2002). All tests have beencarried out with geophone interval of 1m. Source has been kept on both side of the spread and source to the firstand last receiver were also varied from 5m, 10m and 15m to avoid the effects of near- field and far-field. Thesesource distances will help to record good signals in very soft, soft and hard soils. The exploration services69

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFigure 4.10a: MASW instrument with components70

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFigure 4.10b: A typical testing arrangement in the field71

Site characterization using Multichannel Analysis of Surface Wave (MASW) Surveysection at the Kansas Geological Survey (KGS) has suggested offset distance for very soft, soft and hard soil as 1mto 5m, 5m to 10m and 10m to 15m respectively (Xu et al., 2006). Typical recorded surface wave arrivals usingsource to first receiver distance as 5m with recording length of 1000 ms is shown in Figure 4.11.4.10.1 Dispersion CurvesFigure 4.11: Typical seismic waves recorded in geode seismographThe generation of a dispersion curve is a critical step in MASW method. A dispersion curve is generally displayedas a function of phase velocity versus frequency. Phase velocity can be calculated from the linear slope of eachcomponent on the swept-frequency record. The lowest analyzable frequency in this dispersion curve isaround 4 Hz and highest frequency of 75Hz has been considered. Typical dispersion curve is shown in Figure 4.12.Each dispersion curve obtained for corresponding locations has a very high signal to noise ratio of 80 and above.4.10.2 One-Dimensional Shear Wave Velocity ProfilesA shear wave velocity profile has been calculated using an iterative inversion process that requires the dispersioncurve developed earlier as input. A least-squares approach allows automation of the process (Xia et al., 1999) whichis inbuilt in SurfSeis. Shear wave velocity has been updated after completion of each iteration with parameters such72

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFigure 4.12: Typical dispersion curve obtained from MASW73

Site characterization using Multichannel Analysis of Surface Wave (MASW) Surveyas Poisson’s ratio, density, and thickness of the model remaining unchanged. An initial earth model is specifiedto begin the iterative inversion process. The earth model consists of velocity (P-wave and S-wave velocity),density, and thickness parameters. Typical 1-D Vs and Vp profile are shown in Figure 4.13. The shear wavevelocity values obtained from each survey line for the different layers falls within the recommendations ofNEHRP “Vs”- soil classification of site categories (Martin, 1994) and IBC code site classification (IBC-2000).VpVsFigure 4.13: Typical 1-D Vs and Vp Plot4.11 MAPPING OF SUBSURFACE LAYERS USING TWO-DIMENSIONAL “VS”PROFILINGTo get the two-dimensional (2-D) shear wave velocity profiles, a multiple number of shot gathers are acquiredin a consecutive manner along the survey line by moving both source and receiver spread simultaneously by afixed amount of distance after each shot. Each shot gather is then analyzed for 1-D Vs profile in a mannerpreviously stated. In this way a multiple number of Vs profiles are generated. The Vs data are assigned into 2-D(x-z) grid, various types of data processing techniques can be applied to get 2-D Vs. A countering, a simpleinterpolation, data smoothing, or combination of these are applied at this stage. When the Vs data are assignedto the grid, there is ambiguity in the horizontal coordinate (x) to be assigned because each Vs profile wasobtained from a shot gather that spanned a distance too large to be considered as a single point. It seemsreasonable that the center of the receiver spread be the most appropriate point because the analyzed Vs profilerepresents an average property within the spread length (Park et. al, 2005a). The 2-D velocity profile has beenused to find the layer thickness, subsurface layering information and rock dipping directions. 2-D velocityprofiling has been carried out to find the spatial variations and subsurface anomaly.2-D MASW test has been carried out at 20 locations with minimum length of 12m. Inbuilt bilinear/krigingoperation has been used to make interpolation of each mid point velocity and generate the 2-D Vs profile for amid point of first spread line to mid point of last spread line. Typical 2-D velocity profile is shown in Figure 6.6.From Figure 4.14, it is clear that shallow depth shear wave velocities are with in the range of 360m/s. Whendepth increases, the shear wave velocities also increase. General observation from the 2-D Vs profiles, materiallayers of velocity 300m/s and above is dipping, falling and tilting, which may be due to the undulation andvariation in original ground elevation. These undulations can be clearly seen in Figure 4.14 after 6m depth,Figure 4.15 after 16m depth. Also there is no considerable ground layering anomaly present in the subsurfaceand few locations where filledup soil are found (earlier tank beds which are encroached for habitation).However these filled up soil are loosely packed soil with bigger stones which results in the lower velocity(

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFigure 4.14: 2-D spatial variation of shear wave velocityFigure 4.15: Ground anomalies due to filled up soilheterogeneity. This may be resulting due to the presence of filled up soil with boulders. From the depth of6m/8m to 15m/18m shear wave velocity varies from 300m/s to 500m/s, which corresponds to soft weatheredrock as identified in borehole. For a depth of 15m/18m to 20m velocity varies from 500m/s to 700m/s and thiscan be classified as hard weathered rock (this matches well with the drilled borehole data). In the borelog from18m depth, hard rock has been identified while drilling. Shear wave velocity of more than 700m/s is also foundin the 2-D velocity profiles at that depth. MASW can be effectively used for mapping the geotechnicalsubsurface layers and their spatial variations.75

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyBORE LOGLocation RVCE Campus Date of commencement 09.05.2006BH No RVCE-1 Date of completion 12.05.2006Ground Water Table15mDepth Thickness Details of Sampling SPTBelow Soil Description of Strata Legend Type Depth N ValueGL(m) (m) (m)0.0Filled Up Soil/ Gravelly Soil2.0 with Boulders4.0 SPT 5.0 10+RFilled Up Soil/ Gravelly Soil5.0 with Boulders6.07.06Weathered soilwith boulders6m to 8m 2SPT 7.5 8+RSPT 9 R12.0 Soft Weathered RockSPT 10.5 R13.08 m to 15 m14.015.0 7SPT 15 R18.020.0Hard Weathered Rock15 m to 18 mHard RockFrom 18 m322.0Bore hole Terminated at 22.0mNoteSPT Standard Penetration Test R ReboundFigure 4.16: Bore log corresponding to Figure 4.1514.12 AVERAGE SHEAR WAVE VELOCITYThe average shear wave velocity for the depth of “d” of soil is referred as V H . The average shear wave velocityup to a depth of H (V H ) is computed in accordance as follows:H∑ di=⎛ d ⎞i∑⎜⎟⎝ vi⎠H ∑ dV (4.5)Where= = cumulative depth in miFor 30m average depth, shear wave velocity is written as;76

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyVs =(4.6)N⎛ di⎞∑⎜⎟i=1⎝ vv⎠where d i and v i denote the thickness (in meters) and shear-wave velocity (at a shear strain level of 10 −5 or less,m/s) of the i th formation or layer respectively, in a total of N layers, existing in the top 30 m. Vs 30 is accepted forsite classification as per NEHRP classifications and UBC (Uniform Building Code in 1997) [Dobry et al. 2000,Kanli et. al, 2006].n order to figure out the average shear wave velocity distribution in Bangalore, the average velocity has beencalculated using the equation (6.1) for each borelog location. A simple spread sheet has been generated to carryout the calculation, as shown in Table 4.7. The Vs average has been calculated for every 5m depth interval up toa depth of 30m and also average Vs for the soil overburden has been calculated based on the boreloginformation. Usually, for amplification and site response study the 30m average Vs is considered. However, ifthe rock is found within a depth of about 30m, near surface shear wave velocity of soil has to be considered.Otherwise, Vs 30 obtained will be higher due to the velocity of the rock mass. Site characterization using SPTdata shows that, the soil overburden thickness in Bangalore varies from 1m to about 40m. Hence, foroverburden soil alone average Vs has also been calculated based on the soil thickness corresponding to thelocation, which is also shown in column 4 of Table 4.7.30 30Table 4.7: Typical average shear wave velocity calculationDepth(m)Vs(m/s)Soilthickness[d i ] (m)AverageVs Soil-7.2mAverageVs-5mAverageVs-10mAverageVs-15mAverageVs-20mAverageVs-25mAverageVs-30m-1.22 316 -1.2 259 265 286 310 338 362 306-2.74 250 -1.5-4.64 255 -1.9-7.02 241 -2.4-10.00 388 -3.0-13.71 355 -3.7-18.36 435 -4.6-24.17 527 -5.8-31.43 424 -7.3-39.29 687 -7.94.13 SHEAR WAVE VELOCITY DISTRIBUTION IN BANGALOREThe rock depth/soil overburden thickness distribution map has been generated based on the bore holes close tothe MASW test locations, which is shown in Figure 4.17. Figure 4.17 show that the north western part has lesseroverburden thickness. However, eastern part and other areas have the overburden thickness of 4m to 26m. Thecalculated average shear wave velocities are grouped according to the NEHRP site classes (see Table 4.8) andmap has been generated. The average shear wave velocity calculated for 5m, 10m, 15m, 20m, 25m and 30mdepth are mapped and shown in Figures 4.18,4.19,4.20,4.21,4.22 and 4.23 respectively. For mapping of shearwave velocities the surfer plotting software has been used with minimum curvature gridding method.Table 4.8: Site classes for average shear wave velocitySite ClassRange of average shearwave velocity (m/s)A - Hard Rock 1500

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFrom Figure 4.18, the average velocity up to a depth of 5m covers most of the study area having velocity rangeof 180m/s to 360m/s. Few locations in south western part and a smaller portion of southeastern part ofBangalore have the velocity less than 180m/s indicating soft soil. The depth also may extend beyond 5m,matching with the rock level map shown in Figure 4.17. The average shear wave velocity for 10m depth variesfrom 180m/s to 360m/s (see Figure 4.19) in the 10m average map, very dense soil/soft rock velocity range of360m/s to 760m/s is found in western part of the study area. In this location, the rock depth is found with in 10mas seen in Figure 4.17. Figure 4.20 show that the area covered has a very dense soil/soft rock, which is increasedfrom Figure 4.19. In this map northwestern part having the average velocity is more than 360m/s, matching withthe rock depth map (see Figure 4.17). When depth of average velocity increases, the rock velocity influences theaverage velocity values, which is seen in Figure 4.21, 4.22 and 4.23. The area covered by the velocity of 360m/sgets reduced with increasing depth. Similar increased area of higher velocity is found in average depth of 20mand 25m shear wave velocity profiles (see Figures 4.21 and 4.22). Figure 4.23 shows the map of average shearwave velocity for a depth of 30m. Even though the average shear wave velocity is calculated for every 5m depthintervals and up to a maximum depth of 30m, these maps does not show the average shear wave velocity of soilbecause of the wide variation in the soil overburden/ rock level. Hence, the average shear wave velocity of soilhas been calculated based on the overburden thickness above obtained from bore holes close to the MASWtesting locations. The average shear wave velocity for soil overburden in the study area is shown in Figure 4.24.Figure shows that whole study area has medium to dense soil with a velocity range of 180m/s to 360m/s fallingin to “site class D” as per Table 4.84.14 CALCULATION OF DYNAMIC PROPERTIESDynamic property i.e, shear modulus values at low strain (Gmax) of soil has been evaluated from the shearwave velocity, which is an important parameter for seismic site characterization. Shear modulus at low strainlevel for site soil layers has been determined using shear wave velocity from MASW and density fromlaboratory test. Poisson’s ratio was estimated by assuming a constant distribution of Poisson’s ratio for a layerby ensuring Vp model matches with the Vs model with depth. From the above value of Poisson’s ratio, Young’smodulus was also calculated. Soil and rock layer shear modulus and elastic Young’s modulus have beenestimated for 58 locations. Typical calculations are given in Table 6.9.Table 4.9 Typical shear and elastic modulus value of soil layersDepth (m)Vs Density Shear ModulusYoung’s Modulus(m/sec) (g/cc) (MN/m 2 Poisson Ratio)(MN/m 2 )0-10.5 220 1.90 92 0.30 23910.5-23 320 2.00 205 0.20 492>23 520 2.20 595 0.20 14284.15 CORRELATION BETWEEN (N 1 ) 60CS AND V SPrediction of ground shaking response at soil sites requires knowledge of shear modulus of the soil, which isdirectly expressed in terms of shear wave velocity. It is preferable to measure Vs directly by using field tests.However, presently it is not feasible to make Vs measurements at all the locations. Hence to make use ofabundant available penetration measurements to obtain Vs values, correlation between Vs and penetrationresistance are being done. Among the 58 MASW testing points, 38 MASW testing points were close to the SPTborehole locations. These 38 MASW locations are shown in Figure 6.25 and these have been used to generatethe correlation between the Vs to the corrected ‘N’ values. Velocities calculated using 1-D MASW whichrepresents Vs at mid point of each survey line, has been used for this purpose. About 162 data pairs of Vs andSPT corrected blow count have been used. The Vs values are selected from the 1-D MASW resultscorresponding to SPT “N” value at different depths.The regression equation developed between Vs and (N 1 ) 60cs is given in equation 4.7 (with regression coefficientof 0.84):Vs = (4.7)0.4078[(N1 )60cs]Where, Vs is the shear wave velocity in m/s and (N 1 ) 60cs is the corrected SPT ‘N’ value.Figure 6.26 shows the actual data along with the fitted equation along with upper bound and lower boundcurves. Regression equation corresponding to upper and lower bound curves are given in equations 4.8 and 4.9respectively.78

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFigure 4.17: Rock depth map of the study area in Bangalore79

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFigure 4.18: Average shear wave velocity for 5m depth with test locations80

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFigure 4.19: Average shear wave velocity for 10m depth81

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFigure 4.20: Average shear wave velocity for 15m depth82

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFigure 4.21: Average shear wave velocity for 20m depth83

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFigure 4.22: Average shear wave velocity for 25m depth84

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFigure 4.23: Average shear wave velocity for 30m depth85

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyFigure 4.24: Average shear wave velocity for soil86

400m 200 0 800 16002 cm to 400m2400 metersSite characterization using Multichannel Analysis of Surface Wave (MASW) SurveyBANGALOREScale 1:200002-D MASWSPT Borelogs2-D MASW1-D MASWSPT1-D MASWFigure 4.25: SPT and MASW locations used for developing relation betweenVs and (N 1 ) 60cs0.40Vs = 103[(N1 )60cs] - Upper bound (4.8)0.40Vs = [( N ) - Lower bound (4.9)531 60cs]600Upper boundMeasured shear wave velocity (m/s)500400300200100Vs =0.4078[(N1 )60cs]Lower bound00 10 20 30 40 50 60 70 80 90SPT Corrected 'N' valueFigure 4.26: Shear wave velocity versus corrected SPT ‘N’ values87

Site characterization using Multichannel Analysis of Surface Wave (MASW) SurveyJapan Road Association (JRA, 1980) equations (see equation 6.6 -for clayey soil and equation 6.7- for sandysoil), relating Vs and N 60 are also given below:1/ 3Vs = 100(N60) (JRA, 1980)- For clayey soil (4.10)1/ 3Vs = 80(N60) (JRA, 1980) - For Sandy soil (4.11)The comparison between JRA equations with newly developed equation (4.7) are given in Figure 4.27. Thecoefficients are close to the value for the sandy soil. From Figure 6.27, it is clear that the fitted equation liesbetween the JRA equations for sandy and clay equations for wide range of “(N 1 ) 60cs ” values, because the soiloverburden in Bangalore has sand and silt with some percentage of clay content.500Measured Shear Wave Velocity (m/sec)450400350300250200150100500Present relation as per equation 7.3Sand as per equation 7.7-JRAClay as per equation 7.6-JRA0 20 40 60 80 100SPT Corrected 'N' Value [(N 1 ) 60cs ]Figure 4.27: Comparison of shear wave velocity versus SPT ‘N’ value relations4.16 SUMMARY FROM MASW TECHNIQUEIn this study MASW one-dimensional survey at 58 locations and two-dimensional surveys at 20 locations havebeen carried out in an area of 220 sq.km in Bangalore city. The shear wave velocity profiles (Vs versus depth);spatial variability of shear wave velocity (Vs versus depth and length) and ground layer anomalies have beenpresented. The average shear wave velocity has been estimated for 5m, 10m, 15m, 20m, 25m and 30m depth inthe study area. Also average shear wave velocity for the soil depth, which is estimated based on overburdenthickness, is also presented. Site soil classification has been carried out by considering the NEHRP and IBCclassification. The estimated Vs 30 for Bangalore soil can be classified as “site class D” as per NEHRP and IBCclassification. Among total 58 locations of MASW survey carried out, 38 locations were very close to the SPTborehole locations and these are used to generate correlation between Vs and corrected ‘N’ values. A power fitregression equation has been developed using 162 pair of Vs and (N 1 ) 60cs with a regression coefficient of 0.84.The upper and lower bound values of shear wave velocities and corresponding regression equations arepresented. The obtained correlation is one of first measured relation for Peninsular India soil. The shear wavevelocity versus corrected ‘N’ value relation can be effectively used to find out the shear modulus for groundshaking response studies.88

CHAPTER 5LOCAL SITE EFFECTS AND SITE RESPONSE ANALYSIS5.1 INTRODUCTIONSite amplification of seismic energy due to soil conditions and damage to built environment was amplydemonstrated by many earthquakes during the last century. The wide spread destruction caused by Guerreroearthquake (1985) in Mexico city, Spitak earthquake (1988) in Leninakan, Loma Prieta earthquake (1989) inSan Francisco Bay area, Kobe earthquake (1995), Kocaeli earthquake (1999) in Adapazari are importantexamples of site specific amplification of ground motion, even at location far away (100-300km) from theepicenter (Ansal, 2004). The recent 2001 Gujarat-Bhuj earthquake in India is another example, with notabledamage at a distance of 250km from the epicenter (Sitharam et. al 2001, and Govinda Raju et. al 2004). Thesefailures resulted from the effect of soil condition on the ground motion that translates to higher amplitude; it alsomodifies the spectral content and duration of ground motion. As seismic waves travel from bedrock to thesurface, the soil deposits that they pass through change certain characteristics of the waves, such as amplitudeand frequency content. This process can transfer large accelerations to structures causing large destruction,particularly when the resulting seismic wave frequency matches with the resonant frequencies of the structures.Site specific ground response analysis aims at determining this effect of local soil conditions on amplification ofseismic waves and hence estimating the ground response spectra for future design purposes. The response of asoil deposit is dependent upon the frequency of the base motion and the geometry and material properties of thesoil layer above the bedrock. Although not known as seismically very active, Bangalore, a fast growing urbancenter, with low to moderate earthquake history (Sitharam and Anbazhagan ,2007) and highly altered soilstructure (due to large reclamation of land) has been the focus of many of our recent studies.A peculiar feature of the study region, falling in zone II in the seismic zoning map of India, is that it hasreclaimed land from silted lakes/tanks leading to significant variations in ground response. In the present study,an attempt has been made to study site response using Geotechnical, geophysical data and experimental studies.The subsurface profiling with about 170 bore logs is selected from the data base in the study area of 220sq.km.About 58 MASW data have been used to study the site response. The synthetic ground motion for each borehole has been generated from the ground motion model for Bangalore as discussed in chapter 4. The soilproperties and synthetic ground motions for each borehole locations are further used to study the local siteeffects using 1-D ground response analysis with program, SHAKE2000. The response and amplificationspectrum have been evaluated for each soil layer at corresponding borehole location. The natural period of thesoil column, peak spectral acceleration and frequency at peak spectral acceleration of each borehole has beenevaluated and presented as maps. The predominant frequency obtained using SPT ‘N’ values and MASW shearwave velocities are compared. The correlation between corrected SPT ‘N’ value [(N 1 ) 60cs ] available in the studyarea and low strain shear modulus from shear wave velocity (Vs) has been generated. The site response studiesis also carried out experimentally based on recording the ambient noise for a selected period of duration. Thenoise was recorded at 64 different locations using L4-3D short period sensors equipped with digital acquisitionsystems. The predominant frequency of soil column of each location is arrived using horizontal to verticalspectral ratios of recorded noise. The predominant frequency obtained by SHAKE2000 and experimental studyare compared. Further, the uniform hazard response spectrum considering PSHA with local site condition hasbeen developed. Spectral acceleration hazard curve and response spectrum at ground surface with 5% dampingfor Bangalore has been generated for 10% probability of exceedance in 50 years for “site class D”.5.2 NEED FOR THE STUDYThe responses of man-made structures during earthquakes are not only related to structural features but they arecontrolled by two main factors: earthquake ground motion and local site conditions. Any seismic microzonationstudy neglecting the probable earthquake ground motion characteristics would be incomplete. Bangalore and itsvicinity have experienced slight to moderate earthquakes during historic to recent times. However, these werenot considered to be of serious consequence as the cities, located in relatively stable region were considered freeof earthquake effects. However, in the past pace of development, the unscientific approach of land use,including the reclamation of land (silted tanks and lakes) has increased the vulnerability of this city, evenagainst a moderate earthquake. The high density of population, mushrooming of buildings of all kinds,improper and low quality construction practices unscientific land use and heavy traffic conditions only add tothe damage scenario. Most of the damage due to the earthquake results from the improper constructions and89

Local site effects and Site Responseagglomerations of populations in the cities. Hence it is necessary to study the behavior of soil for an earthquakeand also to develop design parameters for earthquake resistant and construction in the cities. Here such anattempt has been made for Bangalore city. To cite an example of unscientific land use, this city that once hadmore than 150 lakes has reclaimed much of these silted water bodies for construction activities, leaving only 64,today (see Figure 5.1). Many dry silted tanks of the city are being encroached, for real estate developments. Theexample of few tanks that once existed in the city and now have been encroached due to rapid urbanization islisted in Table 5.1. These tanks were once distributed throughout the city for better water supply facilities, arepresently in a dried up condition with the residual silt and silty sand forming thick deposits over whichbuildings/structures have been built. The amplification is major scenario, which happens in the filled up soilcausing drastic unbalanced forces to structures resting on the filled up soils. Study of amplification is mandatoryfor design of new structures and retrofits the old structures in the area. Bangalore happens to be an importantcity in India, thus it is necessary to study the amplification potential due to the filled up soil in this area.Table 5.1: Details of the lakes that once existed in Bangalore city and have now been encroached due tourbanizationSl NoOld LakesNow encroached to form a Residential/ Industriallayouts1 Vidyaranyapura Lake Vidyaranyapura(Jalahalli East)2 Gokula Tank Mattikere3 Geddalahalli Lake RMV 2nd Stage, 1st Block4 Nagashettihalli Lake RMV 2nd Stage, 2nd Block5 Tumkur Lake Mysore Lamps6 Ramshetty Palya kere Milk Colony (Playground)7 Oddarapalaya Lake Rajajinagar (Industrial Area)8 Ketamaranahalli Lake Rajajinagar (Mahalakshmipuram)9 Kurubarahalli Lake Basaveshwaranagar (Chord Road)10 Agasana Lake Gayathri Devi Park11 Jakkarayana kere Krishna Floor Mills12 Dharmambudhi Lake Kempegowda Bus Terminal13 Vijayanagar Chord Rd Lake Vijayanagar14 Marenahallli Lake Marenahalli15 Sampangi Lake Kanteerva Stadium16 Kalasipalya Lake Kalasipalya17 Siddapura Lake Siddapura/Jayanagar 1 stBlock18 Tyagarajanagar Lake Tyagarajanagar19 Kadirenahalli Lake Banashankari 2nd Stage20 Sarakki AgraharaLake JP Nagar 4th Phase21 Koramangala Lake National Dairy Research Institute22 Chinnagara Lake Ellpura23 Domlur Lake Domlur Second Stage24 Kodihalli Lake New Thippasandra /Government Buildings25 Banaswadi Lake Subbayapalya Extension26 Shule Tank Ashok Nagar, Football Stadium27 Hennur Lake Nagavara (HBR Layout)5.3 SYNTHETIC GROUND MOTIONS AND PEAK ACCELERATION MAPIndian peninsular shield, which was once considered to be seismically stable, has experienced many earthquakesin the recent past. Large numbers of earthquakes with different magnitudes have occurred in this region. Theseismic hazard analysis for Bangalore city by considering possible earthquake sources is presented in chapter 3.90

Local site effects and Site ResponsePotential seismic source with in a distance of 350km radius around Bangalore has been identified using theavailable data on faults, lineaments and earthquake occurrence.Location of old and encroached water bodies in Bangalore CityFigure 5.1: Location of old and encroached water bodies in Bangalore cityMandya-Chennapatana- Bangalore lineaments (L15, see Figure 5.5) is identified as vulnerable source andmoment magnitude of 5.1 has been evaluated as maximum credible earthquake (MCE). Synthetic groundmotion model and typical results for Bangalore using Boore model (1983, 2003) SMSIM program is presentedin chapter 3. The same model has been used to generate the synthetic ground motion at 170 borehole locationsand it is used as input ground motion to site response study. A typical synthetic ground motion in terms ofacceleration, velocity and displacement is shown in Figure 5.2. The most commonly used measure of amplitudeof a particular ground motion is peak91

Local site effects and Site Response90Acceleration (cm/sec 2 )60300-30-60-9029 30 31 32 33 34 35Time (sec)Velocity(cm/sec)3210-1-2-329 30 31 32 33 34 35Time (sec)0.3Displacement (cm)0.20.10.0-0.1-0.2-0.329 30 31 32 33 34 35Time (sec)Figure 5.2: Typical input ground motion used for the analysishorizontal acceleration (PHA).The PHA for a given component of motion is simply the largest (absolute) valueof horizontal acceleration obtained from the accelerogram. Here only horizontal motion has been syntheticallygenerated corresponding peak value of acceleration of synthetic ground motion is PHA at rock level. For furtherdiscussion in the site response study, the values of PHA at rock level at each borehole locations are evaluatedand a rock level PHA map has been generated as shown in Figure 3.13. From Figure 3.13 shows that theminimum PHA value of 0.061g corresponding to the boreholes in southeast corner of Bangalore where theborehole locations are far away from the vulnerable source and maximum PHA value of 0.156g correspondingto the borehole location 47 which is close to the vulnerable source (see the Figure 5.6). The duration of all the92

Local site effects and Site Responsesynthetic strong motion is about 3.5 to 5 seconds; spectral acceleration at rock level is generated for all locationsas discussed in the chapter 3.5.4 1-D GROUND RESPONSE ANALYSIS USING EQUIVALENT LINEARAPPROACHA ground response analysis consists of studying the behavior of a soil deposit subjected to an acceleration timehistory applied to a layer of the soil profile. Ground response analysis is used to predict the ground surfacemotions for evaluating the amplification potential and for developing the design response spectrum. In thepresent study, one-dimensional ground response analysis using equivalent linear model has been carried outusing SHAKE 2000 software in which motion of the object can be given in any one layer in the system andmotions can be computed in any other layer.In equivalent linear approach, the non-linearity of the shear modulus and damping is accounted for the use ofequivalent linear soil properties using an iterative procedure to obtain values for modulus and dampingcompatible with the effective strains in each layer. In this approach, first, a known time history of bedrockmotion is represented as a Fourier series, usually using the Fast Fourier Transform (FFT). Second, the TransferFunctions for the different layers are determined using the current properties of the soil profile. The transferfunctions give the amplification factor in terms of frequency for a given profile. In the third step, the Fourierspectrum is multiplied by the soil profile transfer function to obtain an amplification spectrum transferred to thespecified layer. Then, the acceleration time history is determined for that layer by the Inverse FourierTransformation in step four. With the peak acceleration from the acceleration time history obtained and with theproperties of the soil layer, the shear stress and strain time histories are determined in step five. In step six, newvalues of soil damping and shear modulus are obtained from the damping ratio and shear modulus degradationcurves corresponding to the effective strain from the strain time history. With these new soil properties, newtransfer functions are obtained and the process is repeated until the difference between the old and newproperties fit in a specified range. The basic approach of one dimensional site response study is the verticalpropagation of shear waves through soil layers lying on an elastic layer of the rock which extends to infinitedepth.Soil behavior under irregular cyclic loading is modeled using modulus reduction (G/G max ) and damping ratio(β) vs. strain curves. The non-linearity of the shear modulus and damping is accounted for by the use ofequivalent linear soil properties using an iterative procedure to obtain values for modulus and dampingcompatible with the effective strains in each layer as discussed above. The degradation curves for sand and rockused for the present work are those proposed by Seed and Idriss (1970) and Schnabel (1973) respectively. Thesecurves are shown in Figures 5.3 and 5.4 respectively. These are included in the SHAKE database and can beselected as input using option command.130Modulus Reduction (G/Gmax)0.80.60.40.2G/GmaxDamping ratio (%)252015105Damping Ratio (%)00.00001 0.00010 0.00100 0.01000 0.10000 1.00000Shear Strain (%)Figure 5.3: Shear modulus reduction and damping ratio curves considered for sand (Seed and Idriss1970)093

Local site effects and Site Response125Modulus Reduction (G/Gmax)0.80.60.40.2G/GmaxDamping ratio (%)2015105Damping Ratio (%)00.00001 0.00010 0.00100 0.01000 0.10000 1.00000Shear Strain (%)Figure 5.4: Shear modulus reduction and damping ratio curves considered for rock (Schnabel 1973)5.4.1 Geotechnical Data and Rock DepthOut of 850 borelogs 170 bore logs were carefully evaluated for the engineering bed rock depth in each borelogsfrom the rock characterization tests, selected for the site response study (bore logs locations are marked inFigure 5.5). Engineering bed rock is bed rock having the shear wave velocity of around 700m/sec which arewidely selected for placing engineering structures. From Figure 5.5, selected bore holes are distributedthroughout Bangalore Mahanagar Palike. For the 170 boreholes selected in this study the overburden thicknessvary from 1m to about 40m, and with their wide distribution in the study area, these bore holes are considered torepresent the typical features of soil profiles. The generalized soil profile in Bangalore had been described inchapter 4. To describe the ground motion parameters from this study the rock level or/ soil overburdenthickness map has been generated using 170 borehole data, which is shown in Figure 5.6. Figure 5.6 shows thatsouth eastern and central part of the study area has thick soil overburden (about more than 30m) when comparedto other areas. North and south western part of the study area has very thin overburden (less than 1m) but theaverage overburden thickness of the study area is about 10m to 15m. Typical subsurface profile informationlike unit weight, ground water level, SPT values (or shear wave velocity) are thus obtained and compiled for theabove selected bore holes, used for the shake analysis is shown in Table 5.2.LayerNo:Table 5.2: Typical input parameter for Shake analysisSoil typeDepth(m)Thickness(m)Unit Weight(kN/m 2 )0SPT ‘N’Corrected1 silty sand with clay 1.5 1.5 20.6 262 silty sand with clay 2.5 1.0 20.6 323 silty sand 3.0 0.5 20.6 324 silty sand 4.0 1.0 20.4 375 silty sand 4.5 0.5 20.4 376 silty sand 6.0 1.5 20.4 417 silty sand 6.5 0.5 20.4 418 silty sand 7.5 1.0 19.0 459 silty sand 9.0 1.5 19.0 8610 silty sand 10.5 1.5 19.0 7411 silty sand 12.0 1.5 19.0 4112 silty sand 13.5 1.5 19.0 5413 silty sand 15.0 1.5 19.0 6214 hard rock >15 19.0 10094

Local site effects and Site ResponseBoreholes selected for Site Response-1707236088Scale 1:20000121L15 105km Length77471039082122118109BoreholeFigure 5.5: Location of the selected boreholes in Bangalore city5.5 GROUND RESPONSE ANALYSIS BASED ON SPT DATAIn this study the input rock motions at bed rock were generated for each bore hole location considering thehypocentral distance calculated for each bore log to the Mandya-Channapatna-Bangalore lineament and used asinput for the corresponding borehole for site response study. The input motion for the location of the boreholenumber EW-18 is shown in Figure 5.7. The rock motion obtained from synthetic ground motion model isassigned at the bedrock level as input in SHAKE to evaluate peak acceleration values and acceleration timehistories at the top of each sub layer. Response spectra at the top of the bedrock and at ground surface andamplification spectrum between the first and last layer at a frequency step of 0.125 are obtained. Typical resultsobtained for borehole location 103, are illustrated in Figures 5.8 a-c. The variation of peak acceleration withdepth is shown in Figure 5.8-a. At this borehole, the basement is 15m deep, the top 3.5m is composed of siltysand with clay and the layers below (3.5m -15m) is made of dense silty sand. The top layer of silty sand withclay shows highest amplification values. The response spectrum of soil is shown in Figure 5.8-b, the first peakof 0.9g occurred at 5Hz and second peak of 1.1 g occurred at 15 Hz, the similar peaks are identified inamplitude spectrum which is shown in Figure 5.8-c. In the amplification spectrum, the maximum amplificationratio occurred at 5 Hz frequency, which is the predominant frequency of the soil column in that location. Similarplots, stress-strain time history and Fourier amplitude spectrum have been obtained for all the boreholelocations. These are compiled and presented in the form of maps depicting variation of different parameters andare discussed in the next section.95

Local site effects and Site Response13.04NLatitude (Degree North)13.02N13N12.98N12.96N12.94N12.92N33m31m29m25m23m20m18m15m12m10m9m8m6.5m5m3m1m77.54E 77.56E 77.58E 77.6E 77.62E 77.64E 77.66E 77.68ELongitude (Degree East)Figure 5.6: Map showing variation of rock depth or soil overburden thickness0.150.100.05Acceleration (g)0.00-0.05-0.100 20 40 60 80Time (sec)Figure 5.7: Typical input ground motion used in SHAKE2000.96

Local site effects and Site ResponseSpectral Acceleration (g)Frequency (Hz)Peak Acceleration (g)0.0 0.1 0.2 0.3 0.40Depth (m)-5-10Figure 5.8-a: Variation of peak acceleration with depth-15Figure 5.8-b: Response spectrum for 5% damping at ground surface97

Local site effects and Site Response15Amplification Ratio105Figure 5.8-c: Amplification spectrum between the bedrock and ground surface5.5.1 Spatial Variation of Local Site Effects in BangaloreThis study based on the borehole data, has analyzed the ground response and its specific variations in Bangalore.The parameters obtained from the analysis are presented as maps, which are developed using the software Surferusing natural neighbor interpolation technique to depict the variation of various parameters in the study area.The map represents different parameters of site response study and amplification potential of Bangalore. Themap shows the peak acceleration at ground surface, amplification factor, period of the soil column, peak spectralacceleration, frequency corresponding to the peak spectral acceleration and the response spectrum at the groundsurface of frequency of 1.5Hz, 3Hz, 5Hz, 8Hz and 10Hz for a 5% damping ratio. The details of parameters andresults are discussed in the following sections.5.5.2 Peak Ground AccelerationThe peak horizontal acceleration (PHA) map of Bangalore which is shown in Figure 5.9. The PHA value rangesfrom 0.088g to 0.66g. They are not evenly distributed due to variation in the soil profile at various locations.The ground surface acceleration is considerably large in the areas of tank beds, resulting from the thick layers ofsilty sand. Ground motions with high peak accelerations are usually more destructive than motions with lowerpeak accelerations thus indicating that regions in the zone having PHA greater than 0.66g are seismically moreunstable than the other regions. However, very high PHA’s that last only for a very short period of time andhave very high frequencies may cause little damage to many types of structures. Hence a better estimate of theregions of high seismic vulnerability can be made by identifying regions susceptible to higher amplification ofthe bedrock motion s. The amplification potential of the soil profile to seismic waves should thus be quantifiedas discussed in the subsequent section.5.5.3 Amplification Factor00 5 10 15 20 25Frequency (Hz)There is a need for quantifying the ground surface PHA in terms of PHA value at bedrock. The term“Amplification Factor” is hence used here to refer to the ratio of the peak horizontal acceleration at the groundsurface to the peak horizontal acceleration at the bedrock. This factor is evaluated for all the boreholes using thePHA at bedrock obtained from the synthetic acceleration time history for each borehole and the peak groundsurface acceleration obtained as a result of ground response analysis using SHAKE 2000. The amplificationfactor thus calculated ranged from 1 to 4.8. Quantitative amplification factors are obtained and these resultswere used to prepare the amplification map. Bangalore city can be divided into four zones based on the range ofamplification factors assigned to each zone as shown in Table 5.3. The amplification factor map for Bangalorecity is shown in Figure 5.10. Lower amplification values indicate lesser amplification potential and hence lesser98

Local site effects and Site ResponseFigure 5.9: Peak horizontal acceleration map at ground surface99

Local site effects and Site ResponseFigure 5.10: Amplification factor map for Bangalore city100

Local site effects and Site Responseseismic hazard. It can be observed that the amplification factor for most of Bangalore region is in the range of 2-3. This is in agreement with Sitharam et al. (2005), wherein authors have qualitatively studied amplificationsusceptibility rating based on the average shear wave velocity of 30m (Vs 30 ) depth using SPT “N” values. Theyclassified borehole soil profiles based on the Finn (1991) approach and concluded that most of Bangalore cityhas moderate amplification potential.Table 5.3: Zones and amplification factor rangeZone IVZoneAmplification Factor1 (I) 1.00-1.992 (II) 2.00-2.993 (III) 3.00-3.994 (IV) > 4.00This is a zone of high amplification potential with the amplification factor ranging from 4 to 5. A very small partof the city comes under this zone. Comparison of the variation of overburden thickness in the city map and theamplification factor map shows that the regions under this zone have an overburden thickness in the range of 5mto 10m. The high amplification potential in these regions can be attributed to factors other than the overburdenthickness. Boreholes located in the North Eastern side of the city are fall in zone IV. The amplification factor forthese soil profiles is evaluated to be very high due to presence of clay soil overburden from 2m to 15.5mthickness. This can be attributed to the fact that this region has a filled up soil of about 4m depth and a shallowwater table depth of 0.3m which considerably increases the amplification potential. The soil profile in the North-West part of the city (Borehole 121) also shows a filled up soil depth of about 1.5m causing high amplificationfactor of 4.3. Soil profiles at boreholes 118 and 122 also reveal a shallow water table depth of 1m to 2m and afilled up soil of 1.5m causing region of high amplification in the South-eastern part of the city. The borehole 1-4which is located in the north-central region has a high amplification factor of 4.2 which can be attributed to thepresence of 6m overburden of silty sand with mica.Zone IIIThe area under this zone also has a considerably large amplification potential. The bore logs located in zone IIIhave low average shear wave velocities as calculated from the SPT values when compared to the boreholes inthe other zones. Borehole number 109 which is located in the southern tip of the city has an overburden of 9mbut has a considerably high amplification factor of 3.5. This may be due to the presence of a shallow water tabledepth of 2.7m. Hence the amplification factor map shows a region of zone III in the southern part of the city.This is clearly illustrated in the amplification factor map as a patch of zone III in the northern- central part dueto shallow water table. The high amplification in the Eastern stretch of the city may be due to the presence ofdried and encroached lake beds.Zone IIMost of the area in Bangalore comes under this zone with an amplification factor3. The borehole 82 has an overburden of 11m with the water table at the ground surface. It has an amplificationfactor of 2.5 both due to high overburden and shallow water table depth. The borehole 60 which is located in thenorth-eastern part of the city has an overburden of 21m with the water table at a depth of 2m from the groundsurface. The upper 13m consists of clayey sand which is followed by weathered rock and the soil profile has anaverage shear wave velocity of 278m/s. Variation of peak acceleration with depth plots shows that much of theamplification occurs in the upper 13m as compared to that occurring in the weathered rock layer. The other partsof the city in the south-eastern and north-western region which are in this zone have an amplification factor of2.5 to 2.9. Most of the boreholes located in this region have a shallow water table depth ranging from 0.15m to5.4m. The soil profile at the borehole location of 90 shows an amplification factor of 2.7. At this location, theoverburden consisted of a 16m thick silty sand layer with a high average shear wave velocity followed by softrock. There was alsoZone IThe regions under this zone exhibit a relatively low amplification potential of 1 to 2. This zone extendspredominantly in the central region of the city and also includes the northern tip of the city. The northern tip of101

Local site effects and Site Responsethe city has a very low overburden of 1m to 2m and hence has a low amplification potential. A cluster ofboreholes (2, 3 and 7) were analyzed in this region. The water table was not encountered here and the rock depthvaried from 1m to 2m. The average shear wave velocity obtained for these boreholes locations are relativelyhigher than that obtained for other boreholes. All the above factors result in a low amplification potential in thisregion. The phenomena of soil amplification results in transfer of larger accelerations than rock levelacceleration to structures and causes destruction particularly when the resulting seismic wave frequency matcheswith the resonant frequencies of the structures. Hence, considering the amplification factor, the regions in zonesI and II can be classified as seismically more stable when compared to the zones III and IV.5.5.4 Amplification versus Overburden ThicknessMap of overburden thickness of soil column in the borehole locations has been prepared and compared with theamplification potential. Figure 4.17 shows the variation of rock depth in the city. This map has been preparedbased on the geotechnical information obtained from selected 170 boreholes in the study area of 220sq.km. Thisplot corresponds very well with the data of overburden thickness from 850 boreholes. The northern tip of thecity which has a low overburden of 1m to 5m also has a low amplification factor and lies in zone I (see Figure5.8). However, most of the regions in zone IV and zone III have an overburden thickness in the range of 5m to10m. Also, the regions of high overburden (>20m) lie in zone I or zone II thus not depicting risk of highamplification. Hence, it can be concluded that amplification of seismic waves depends not only on overburdenthickness but also on various other factors like the frequency of the input motion, average shear wave velocity ofthe soil profile and water table depth. It was also observed that the PGA value was decreasing for greaterthickness of soil overburden.5.5.5 Period of Soil ColumnsResults obtained from the site specific ground response analysis show that the natural periods are in between0.01s and 0.45s with about 85% of the locations having a period below 0.2s. Figure 5.11 shows the variation ofthe period of soil column at various locations. The major part of the study area has the soil column period lessthan 0.2s. Lower period of soil column correspond to lower amplification factor. The lower period of the soilcolumn may be resulting from the higher damping of the soil. This result is attributed to the characteristics ofthe frequency content of the ground motion generated. The frequency content of a real event can be differentfrom the frequency content of the synthetic ground motion used in this study.5.5.6 Peak Spectral ParametersThe earthquake amplitudes are represented usually by the peak ground acceleration; however for the structuraldesigns and building code the most widely used parameter is spectral acceleration and correspondingperiod/frequency. Peak spectral acceleration (PSA) and frequency corresponding to PSA of each borelog fromsite response study has been computed. Peak spectral acceleration varies from 0.2g to 2.5g which is shownFigure 5.12. North western part of the study area has larger spectral acceleration when compare to the southeastern part of the study area. Figure 5.13 shows that the frequency corresponding to the PSA varies from 3Hzto 20Hz. Major part of the study area have the frequency content of 5Hz to 15Hz. The period of the soil columndistribution map (Figure 5.11) matches with frequency distribution map at PSA.The frequency content of an earthquake motion will strongly influence the effects of ground motion and hencethe PGA value on its own cannot characterize the ground surface motion. A response spectrum is usedextensively in earthquake engineering practice to indicate the frequency content of an earthquake motion. AResponse Spectrum describes the maximum response of a single-degree-of-freedom (SDOF) system to aparticular input motion as a function of the natural frequency/period and damping ratio of the SDOF system.The combined influences of acceleration amplitudes and frequency components of the movement arerepresented in a single graph. Since the time history of the seismic excitation in a certain site is characterized bythe corresponding response spectrum, the differences among the time histories of the movements at differentplaces can be analyzed by the comparison of their response spectra. The acceleration-time histories at variousdepths are obtained as a result of ground response analysis and these motions can be characterized by thecorresponding response spectra. The ground surface response spectra for all the 170 borehole locations wereplotted with 5% critical damping value. The spectral acceleration (SA) values for all the locations at 1.5 Hz, 3Hz, 5Hz, 8 Hz and 10 Hz are computed and presented in Figures 5.14-5.18. The above frequencies from 1.5Hzto 10Hz were selected as they represent the range of natural frequencies of tall buildings to single storeybuildings (Day, 2002; Govinda Raju et al, 2004). At 1.5 Hz frequency, the SA values are very low and varied102

Local site effects and Site ResponseFigure 5.11: Variation of period of the soil column at various locations103

Local site effects and Site ResponseFigure 5.12: Peak spectral acceleration for Bangalore city104

Local site effects and Site ResponseFigure 5.13: Frequency corresponding to peak spectral acceleration105

Local site effects and Site ResponseFigure 5.14: Spectral acceleration map of Bangalore at 1.5 Hz frequency106

Local site effects and Site ResponseFigure 5.15: Spectral acceleration map of Bangalore at 3 Hz frequency107

Local site effects and Site ResponseFigure 5.16: Spectral acceleration map of Bangalore at 5 Hz frequency108

Local site effects and Site ResponseFigure 5.17: Spectral acceleration map of Bangalore city at 8 Hz frequency109

Local site effects and Site ResponseFigure 5.18: Spectral acceleration map of Bangalore city at 10Hz frequency110

Local site effects and Site Responsefrom 0.01g to 0.07g for all the locations which is shown in Figure 5.14. The SA map at 1.5 Hz frequency showsthat there is not much variation of SA values and most of the area in the city has a SA value between0.02g to0.04g at this frequency. The SA values varied over a wide range from 0.02g to 0.54g at 3 Hz frequency. Figure5.15 shows the variation of these values at different locations. A major part of the city lies in the range of 0.1g to0.2g. Zones of high SA of above 0.3g are observed in the north and south-west part of the city. The central partof the city also has a SA greater than 0.2g at this frequency. The entire region can be divided into 5 groupsbased on the SA values at 5 Hz frequency as shown in Figure 5.16. The SA at this frequency varied from 0.08gto 1.14g and is unevenly distributed around the city with a major region having a SA in the range of 0.3g to0.5g. Maximum SA for this frequency was 1.14g at borehole 88 located in the northern part of the city. Figures5.17 and 5.18 shows the variations of spectral acceleration correspond to 8 Hz and 10 Hz respectively. Acomparison of these figures shows that the distribution of SA at 8 Hz and 10 Hz is considerably similar in thecity. The SA varies over a wide range between 0.17g to 2.17g at 8 Hz frequency and between 0.18g to 1.88g at10 Hz frequency. The lower spectral acceleration results from very low amplification. The borelog where in SAof 0.17g is obtained is shown in Figure 5.19. Soil layer is very thin silty sand resulting in the lower amplificationas well as lower SA. Figure 5.20 shows the borelog where in the estimated SA is more than 2.1g. The soiloverburden is considerably large in this region (10m to 15m) and also the borelog shows that thick layer ofclayey sand is present. The soil found in this region has the liquid limit of more than 32, representing cohesivenature which might have caused the higher SA values at higher frequencies. The northern central region lies inthe range above 0.8g at both the frequencies. The northern and the eastern parts of the city lie in the lower SAzone for both the frequencies. The south-west part of the city lies in the lower range of 0.1g to 0.4g at 8 Hzfrequency but lies in the higher range of 0.5g to 0.8g at 10 Hz frequency. From the SA maps, the northern andeastern parts of the city are in the low SA zone at all frequencies. The above analysis is based on the syntheticground motion generated at each location and may vary if the soil column is subjected to different inputmotions.BORE LOGDate of commencement 03.09.2002BH No 2 Date of completion 03.09.2002Ground Water Table Not encounteredDepth Soil Description Thickness Legend Details of Sampling SPTBelow of Strata Type Depth N ValueGL(m) (m) (m)SPT 1.0 25/35/400.0Reddish/Yellowish1.0 Silty Sand 1.0SPT 2.0 15/22//45Reddish/yellowishSPT 3.0 20/32//50R for2.0 sandy silt5cm PenetrationWeathered rockSPT 4.5 75R for4.5 no Penetration5.0SPT 6 75R for6.0 5no PenetrationHard RockBelow6.0mSPT 7.5 75R forno PenetrationNoteSPT Standard Penetration Test R ReboundFigure 5.19: Borelog of the corresponding borehole where in estimated spectral acceleration of 0.17g for8Hz111

Local site effects and Site ResponseBORE LOGDate of commencement 28.8.2003BH No 77 Date of completion 29.8.2003Ground Water Table 1.5mDepth Soil Description Thickness Legend Details of Sampling SPTBelow of Strata Type Depth N ValueGL(m) (m) (m)0.0 SPT 1.5 3/4//6Brownish Silty sandUDS 2.01.51.5SPT 3.0 7/10//123.04.0 SPT 4.5 12/2//246.0Brownish/greyishClayey sand4.57.0Greyish silty sand/Sandy silt with micaSPT 6.0 14/13/278.0SPT 6 20/30/339.010.0Weathered RockStarts41.5SPT 7.5 75R for 3cmPenetrationSPT 9 75R forno PenetrationNoteSPT Standard Penetration Test R ReboundUDS Undisturbed SampleFigure 5.20: Borelog of the borehole corresponding to the estimated spectral acceleration of 2.17g for 8Hz5.6 Ground Response Analysis using Shear Wave VelocityWith the development of geophysical methods, particularly SASW (spectral analysis of surface wave) andMASW are being increasingly used for the site response study and microzonation of cities world wide. Shearwave velocities (Vs) measured using geophysical method are widely used to get better results of site responsestudies than SPT data. Because, wave propagation theory shows that ground motion amplitude depends on thedensity and shear wave velocity of subsurface material (Bullen, 1965; Aki and Richards, 1980). Usually densityhas relatively little variation with depth but shear wave velocity is the logical choice for representing site112

Local site effects and Site Responseconditions. The Vs 30 (average shear wave velocity for 30m depth) parameter was proposed to overcomedifficulty of quarter-wavelength Vs parameter. Based on studies by Borcherdt and Glassmoyer (1992),Borcherdt (1994) recommended Vs 30 as a means for classifying sites for building codes, and similar sitecategories were selected for the NEHRP (National earthquake hazard research program) seismic designprovisions for new buildings (Martin, 1994).The NEHRP site classification is presented in Table 6.2 in the previous chapter. The average shear wavevelocities of 30m (Vs 30 ) depth are widely used to estimate qualitatively the amplification (Borcherdt, 1994;Joyner and Fumal, 1985; Boore et al., 1993) of site using Finn (1991) approach for the purpose of preliminarymicrozonation (Stewart et al 2001; Anbazhagan, 2004; Premalatha and Anbazhagan, 2004 and Sitharam et al,2005).Salome and Glenn (2001) carried out comprehensive study and identified soil deposits susceptible to groundmotion amplification in the Central United States using shear wave velocity and site response analysis. Brown etal (2002) used shear wave Slowness from SASW method and highlighted that slowness profiles are moresuitable for site response predictions. Baise et al (2003) studied the site response using Vs at Treasure and YerbaBuena Islands, California and compared with earlier site response evaluated using 1D equivalent linear model.Gonza´lez et al (2004) carried out site studies in Cariaco, considering the geotechnical analysis of boreholes,seismic refraction studies and microtremor measurements to determine the response of the Quaternary sedimentfill in the area. Destegül Umut (2004) used shear wave velocity from seismic and electrical methods andSHAKE2000 for seismic Microzonation for Lalitpur, Nepal. He also carried out the sensitivity analysisconsidering input parameters used in SHAKE2000, concluded that most important parameter which affectsresult are input motion, shear wave velocity, thickness of layers and unit weight. Tuladhar et al (2004) carriedout the site response analysis using SHAKE to validate the microtremor observations in greater Bangkok areafor seismic microzonation. Ansal et al (2004) carried out seismic microzonation for ground shaking as a part ofthe case study for Adapazarı region Turkey based on the average spectral accelerations and peak spectralamplifications calculated from site response analyses using SHAKE and equivalent shear wave velocities.Pankow and Pechmann (2004) measured average, frequency-dependent, low-strain site-amplification factorsand mapped for Salt Lake Valley, Utah using site response study and near-surface shear-wave velocity. Molnaret al (2004) studied site response in Victoria, B.C, showed that the peak amplification at each site could beattributed to the local geology amplifying the ground motion using shear wave velocity and SHAKE model.Westen (2005) highlighted the site response studies of Kathmandu valley using Vs and SHAKE2000. RajivRanjan (2005) carried out seismic response analysis of Dehradun city, India using shear wave velocity measuredfrom MASW and SHAKE 2000 and mapped the response parameters. Rao and Neelima Satyam (2005)characterized the sites in Delhi using geophysical testing at 118 sites and average shear wave velocity at 30mdepth (Vs 30 ). Estimation of soil amplification was carried out by using DEGTRA software and microzonationmap for amplification was generated. An attempt has been made to study the site response of study area usingmeasured shear wave velocity obtained from MASW. About locations, measurement of shear wave velocityusing MASW and velocity distribution was presented in chapter 6. Site response study using SHAKE2000 andMASW data has been presented here for Bangalore.5.6.1 Site Response Results Based on MASW DataThe peak horizontal acceleration obtained using shear wave velocity ranges from 0.188g to 0.475g. This iscomparable to peak horizontal acceleration values obtained using SPT data. The ground acceleration isconsiderably large in the areas of tank beds, resulting from the thick layers of silty sand with clay mixture. Theshape of variation of peak acceleration with depth is similar to the SPT data, typical one is shown in Figure 5.21.From Figure 5.21 shows the variation of peak acceleration of each layer for 18m soil overburden. Peakacceleration using SPT data is higher when compared to the MASW data and also these variation is large forshallow overburden (with in 12m) thickness.The amplification factor thus calculated using MASW data ranges from 0.94 to 5.3 which are comparable withthe values calculated using SPT data. However for the dense sand soil where the average shear wave velocity ismore than 350m/s, there is no amplification. Amplification factor at all the locations are less than 3.5 except atone location, where the very low shear wave velocity due to the presence loose soil deposits. Heigheramplification factor (5.33) attributed to the presence of loose deposits in that location. This can be easilyidentified from MASW seismic wave record where there was a scattering of waves (See Figure 5.22). Theresulting velocity profile at the corresponding location is shown in Figure 5.23. Figure 5.23 shows that shearwave velocity up to depth of 3.5 m is very low and for a depth 1.7m to 2.6m, Vs is about 150m/sec. Naturalperiods of soil column as obtained from MASW data is comparable (0.01s and 0.30s) with the values obtainedusing SPT data. Peak spectral acceleration resulting from MASW data (0.39g to2.4g) is comparable with SPTdata (0.3g to 2.1g).113

Local site effects and Site Response-3Peak Acceleration (g)0.0 0.1 0.2 0.3 0.40Depth (m)-6-9-12-15-18Using "N" ValueUsing "Vs" ValueFigure 5.21: Typical peak ground acceleration with soil column thicknessFigure 5.22: MASW seismic wave record114

Local site effects and Site ResponseThe frequency corresponding to the PSA using MASW data varies from 3.7Hz to 20Hz which is comparablewith the SPT data (3Hz to 20Hz). Spectral values at different frequency using MASW data does not vary muchfrom the values obtained using SPT data and are same as the lower values of SA using MASW data whencompared to SPT data. The shapes of response spectrum resulting from MASW are similar to SPT data, typicalone is shown in Figure 5.24. Peak spectral acceleration values at lower period using MASW data is lower thanSPT data, but at higher period values from both the data matches well. The shape of the amplification spectrumobtained using both data matches well, however values of amplification ratio from MASW data is lower thanthe SPT data, typical one is shown in Figure 5.25. Predominant frequencies obtained from both the methods arediscussed in the next section.VpVsSpectral Acceleration (g)1.61.41.210.80.60.40.20Figure 23: Velocity profile with depthUsing N valueUsing Vs value0 0.25 0.5 0.75 1 1.25Period (sec)1.5 1.75 2Figure 5.24: Typical response spectrum using MASW and SPT115

Local site effects and Site Response12Amplification Ratio963Using N valueUsing Vs value00 5 10 15 20 25Frequency (Hz)Figure 5.25: Typical amplification ratio using MASW and SPT5.6.2. Comparison of Predominant Frequency obtained using SPT and MASW dataPredominant frequency of soil is widely used to categorize the soil for a ground motion, is mainly dependent onthe dynamic properties of soil. The predominant frequency is defined as the frequency of vibrationcorresponding to the maximum value of Fourier amplitude. In this study predominant frequency of soil columnis obtained from Fourier spectrum estimated using SHAKE2000. Predominant frequencies of each borelog areestimated using both SPT data and MASW data. Results show that predominant frequencies are similar fromboth analyses. Predominant frequency varies from 4Hz to 12Hz based on SPT data and 3.45Hz to 12Hz basedon Vs from MASW survey. To find the variation of predominant frequencies from both the methods, the siteresponse study of 33 points using SPT and MASW (MASW testing points are very close to the SPT boreholesas described in Chapter 5, see section 5.8) are considered. Predominant frequencies corresponding to theselocations obtained from both data are comparable. Table 5.4 shows that most of the study area has higherpredominant frequency (3 Hz to 12.5 Hz) from both these methods. The predominant frequency of the soillayers has been better estimated using shear wave velocity when compared to SPT data as MASW surveydirectly measure the shear wave velocity. From the above studies and comparison, it is clear that responseparameters obtained using SPT data gives higher values when compared to the values using MASW data.Table 5.4: Predominant frequency rangesPredominant Frequency RangeNumbersSymbols(Hz)Using SPT Using MASW3.0 to 5.0 3 35.1 to 7.0 6 97.1 to 9.0 8 99.0 to 11.0 9 711.1 to 12.5 8 55.7 CORRELATION BETWEEN (N 1 ) 60CS AND G MAXIn SHAKE2000, the dynamic soil property such as shear modulus is evaluated based on the inbuilt equation(equation 13 in SHAKE2000) developed by Imai and Tonouchi (1982) which is as given below:G = (Imai and Tonouchi, 1982) (5.1)20.68max( kips / ft ) 325[( N1)60]cs116

Local site effects and Site ResponseIn response study using MASW data, shear modulus (G max ) is calculated by accounting the both density as wellas in-situ shear wave velocities, which is as given below:G ρ Vmax2s= (5.2)Where , ρ density measured from the undisturbed sample collected in the boreholesVs shear wave velocity measured using the MASW testing.Dynamic properties obtained from SPT test correspond to high strain values when compared to MASW testwhich gives properties at low shear strains. Also the factor affecting shear modulus depends on soil parameters,but in SHAKE2000 G max is calculated based on the inbuilt equation developed for completely different region.From this study it is felt that the SPT data can be effectively used for site response analysis, if regional G maxequation is developed. To fulfill this requirement an attempt has been made to correlate the measured G max(calculated from measured shear wave velocity and densities of each layer) of each borehole to the correctedSPT-N values. At about 38 locations, MASW test locations were very close to the SPT locations (see Figure6.16). From these 38 locations, about 190 data pairs of Vs and SPT corrected blow count have been used for theregression analysis. Figure 5.26 shows the actual data along with the fitted equation. The regression equationdeveloped between Vs and (N 1 ) 60cs is given below:G [( ) ] 0. 68max= 16.82 N1(5.3)60csWhere, G max –Low strain maximum shear modulus in MN/m 2 ,(N 1 ) 60cs – Corrected SPT “N” Value.Low Strain Shear modulus (MN/m 2 )450400350300250200150100500Lower boundUpper boundG =0 10 20 30 40 50 60 70 80 90SPT Corrected 'N' ValueFigure 5.26: Shear modulus versus corrected SPT “N” values[( N ) ] 0. 68max16.821 60csFigure 5.26 also shows the actual data along with the fitted equation and in addition upper bound and lowerbound curves are also shown. Regression equation corresponding to upper and lower bound values are given inequations 5.4 and 5.5 respectively.G = . N0.- Upper bound (5.4)[( ) ] 681 60[( N ) ] 0. 68max26 14csmax9.381 60csG = - Lower bound (5.5)Power regression gives the highest R squared value of 0.88. The comparison between Imai and Tonouchi (1982)equation for sandy soil (equation 5.1) with newly developed equation (5.3) are given in Figure 5.27. Fittedequation (5.3) matches up to a corrected SPT-N [(N 1 ) 60cs ] value of 40 with Imai and Tonouchi (1982) equation.Beyond the “(N 1 ) 60cs ” values of 40 fitted equation G max is higher than the Imai and Tonouchi (1982) equation.117

Local site effects and Site ResponseThis is attributed to the soil type in Bangalore. The soil type is residual in nature and it is classified as silty sand,sandy silt with clay content.450Low strain shear Modulus (MN/m 2 )400350300250200150100500DataImai and Tonouchi, 1982Present relation equation 8.110 10 20 30 40 50 60 70 80 90SPT Corrected 'N' valuesFigure 5.27: Comparison of shear modulus -SPT relation5.8 SITE RESPONSE USING MICRO TREMOR STUDIESIn assessing the seismic hazard of any urban centre, ambient noise measurements are quite popular in estimatingthe dominant frequencies. The method requires a seismic broad band station with three components. Inestimating the site response, Nakamura technique has been widely used and the resonance frequency is obtainedby evaluating the horizontal to vertical spectral ratios (Nakamura, 1989). The main consideration of thistechnique is the micro tremors which are primarily composed of Rayleigh waves, produced by local sources.These waves propagate in a surface layer over a half space, considering the motion at the interface of the surfacelayer and half space is not affected by the source effect and the horizontal and vertical motion at this interfaceare approximately equal. Site response studies mainly deal with the determination of peak frequency (f 0 ) of softsoil, amplification and the nature of response curve defines the transfer function at the site which forms animportant input for evaluating and characterizing the ground motion for seismic hazard quantification. Use ofambient noise in the determination of the above parameters has been extensively used globally, in trying toquantify the seismic hazard in a given region (Field and Jacob, 1993; Bindi et al, 2000; Parolai et al, 2001).5.8.1 Instrument and MethodologyThe instruments used in this experiment are L4-3D short period sensors available with NGRI, Hyderabad,equipped with digital acquisition system (See Figure 5.28). The survey was carried out by NGRI team led by Dr.D. Srinagesh and Dr. R. K. Chadha of NGRI jointly with IISc. The sites were selected based on the availablegeotechnical data (as presented in the chapter 6). The duration of recording was for a minimum of 3 hours and amaximum of 26 hrs. In this study Nakamura method was adopted for obtaining the transfer function at varioussites in Bangalore. The general layout of the horizontal to vertical spectral ratio technique (HVAR) is shown inFigure 5.29. The surface sources for the ambient noise generate Rayleigh waves which affect the vertical andhorizontal motion equally in the surface layer. The spectral ratio of the horizontal component by the verticalcomponent of the time series provides the transfer function at a given site. The dominant peak is well correlatedwith the fundamental resonant frequency.118

Local site effects and Site ResponseFigure 5.28: A typical configuration of the sensor and the digitizer usedHorizontal (R or T)CVertical (V) ComponentSoil SiteRTSedimentVIncident WaveA(f ) horizontalS( f ) =A(f ) VerticalTransfer Function or SpectralBed RockAmplitudeFourier Amplitude100 90 80 70605040302010Fourier AmplitudeFrequency [Hz]Frequency (Hz)Figure 5.29: Horizontal to vertical spectral ratio technique – Layout119

Local site effects and Site Response5.8.2 Criteria for Site SelectionThe locations of seismological instruments are selected based on the overburden soil distribution in Bangaloreand also close to the available boreholes. The testing points have been selected based on overburden thicknessof soil and safety requirements of the instrument. Most of the sites were located with the overburden thicknessgreater than 5 meters. At about 54 locations testing was carried out in the study area (which is shown in Figure5.30). Among these 54 locations, more than 30 seismological stations are located in important places likeschools and colleges. This work was carried out by National Geophysical Research Institute (NGRI) Hyderabad.5.8.3 Microtremor Survey ResultsFigure 5.30: Location of microtremor in the study areaThe spectra and the H/V ratios have been computed using the JSESAME program (NGRI report enclosed)About 6 stability criteria have been used to test the estimated dominant frequency and the amplitude.Dominant frequency and amplitude have been estimated at 54 sites. H/V spectral ratios at differentsites located in Bangalore are shown in Figure 5.31a-h. The predominant frequencies observed in Bangaloreranges between 1.2 Hz -11 Hz (Figure 5.32). Based on the analysis of Figures 5.32, it is clearly seen that thewestern part of study region is mainly characterized by higher frequencies compared to the eastern part ofBangalore. The lower frequencies ranging between 1 Hz to 3 Hz are mostly observed in the eastern and thesouthern part of Bangalore. Here the amplitudes of the dominant frequencies are not discussed because they aregenerally lower than other techniques and hence, do not represent the actual amplification values for a givensite. The results obtained from this study would be discussed in detail and compared with the overburdenthickness inferred from various borehole locations available in Bangalore. These bore holes are densely locatedin the central and the southern parts of the city while the north eastern part is sparsely sampled and hence thesoil thickness values for this region was not known. The thickness of the soil cover varies between 2 m to about36 m. Analysis of Figure 5.8 shows that 90% of the sites have the overburden thickness varying between 1 m to15 m. The dominant frequencies and the soil thickness more or less match well. The most notable feature of thepresent study is that the lower frequencies are closer to the lakes and erstwhile lakes.120

Local site effects and Site ResponseFigure 5.31a: Typical predominant Figure 5.31b: at National Dairy Research frequency at IISCInstitute(Black line depicts the computed spectral ratio whilst the red and blue show the standard deviation of ±1σ).Figure 5.31c: at IVRI campusFigure 5.31d: at School in Lottegonahalli121

Local site effects and Site ResponseFigure 5.31e: at Richmond TownFigure 5.31f: at VivekanandanagarFigure 5.31g: at UlsoorFigure 5.31h: at Govindapura122

Local site effects and Site Response13.0413.0214131211Latitude (Degree in North)1312.9812.9612.9410987654312.9277.52 77.54 77.56 77.58 77.6 77.62 77.64 77.66 77.68Longitude (Degree in East)21Frequency HzFigure 5.32: Contour map of dominant frequency(Yellow triangles indicate station locations, NGRI Report )5.8.4 Comparison of Predominant Frequency from Site Response Study using Vs andMicrotremorEven though the Microtremor and MASW tests were carried out separately, about 43 locations arecomparatively closer to each other. The results at these locations are further used to compare the predominantfrequency of Bangalore soil. Site response studies using SPT and MASW data shows that the predominatefrequency of Bangalore soil varies from 3Hz to 12Hz. But microtremor studies show that the predominantfrequency of Bangalore soil varies from 1.5Hz to 12Hz. The predominant frequency estimated fromMicrotremor and site response using MASW clearly shows that in most of the locations predominant frequencyfrom both the method matches well. Only at few points the low predominant frequency is not matching and stillclose to these frequencies points no site response analysis are carried out. Most of the study area haspredominant frequency of 3 Hz to 12 Hz (except at 7 locations in microtremor studies from site response usingSHAKE and microtremor studies (See Table 5.5).123

Local site effects and Site ResponseTable 5.5: Predominant frequency using site response study and microtremorPredominant Frequency RangeNumbersSymbols(Hz)Using MASW Using Microtremor3.0 to 5.0 7 165.1 to 7.0 15 107.1 to 9.0 11 59.0 to 11.0 6 211.1 to 12.5 4 31.5 to 2.9 - 75.9 NATURAL FREQUENCY AND PERIOD OF TYPICAL STRUCTUREPeak ground acceleration is used to determine the maximum horizontal forces that can be expected at base of thestructures. Many times peak acceleration often corresponds to high frequencies, which are out of range of thenatural frequencies of most of the structures. The largest amplification of the soil will occur at its fundamentalfrequency. The frequency is the number of times per second that the building will vibrate back and forth. Theperiod is the time it takes for the building to make one complete vibration. For the soil period of vibrationcorresponding to the fundamental frequency is called the characteristic site period. The characteristic site perioddepends on the soil thickness and shear wave velocity of the soil as determined in previous sections. Theresponse of a building to shaking at its base depends on the design and quality of the construction. The veryimportant factor is the height of the building. All objects or structures have a natural tendency to vibrate. Therate at which it vibrates is its fundamental period or natural frequency. High rise building (with a low naturalfrequency) reacts totally different than smaller building (with a much higher natural frequency) (see Figure5.33) When the frequency contents of the ground motion are centered around the building's natural frequency,the building and the ground motion are in resonance with one another. Resonance tends to increase or amplifythe building's response. Because of this, buildingsFigure 5.33: Tall and short building vibrations during earthquakesuffer the greatest damage from ground motion at a frequency close or equal to their own natural frequency. Ifthe frequency of the soil column is close to the natural frequency of the building (see Table 5.6), the buildingssuffer the greatest damage for ground motion. Major part of the study area (Bangalore) has building with singlestorey to 3-4 storeys, buildings which are prone to higher frequency (2Hz to 10Hz) and this is with in the naturalfrequency of soil column (3Hz to 12Hz).Table 5.6: Typical natural frequency of the different building sizes (After Day 2002 and Kramer, 1996)Type of Object or StructuresNatural Frequency (Hz)1 Storey buildings 10124

Local site effects and Site Response2 Storey buildings3-4 Storey buildingsTall buildings520.5-1.0High rise buildings 0.175.10 PROBABILISTIC SEISMIC HAZARD ANALYSIS WITH LOCAL SITEEFFECTSFrom the extensive field investigation carried out using MASW survey (at 58 locations in the study area) it hasbeen observed that the shear wave velocity for the Bangalore region falls in the range of 0.18 km/sec ≤ Vs ≤0.36 km/sec (Site Class D). Keeping this in mind surface level response spectrum has been developed based onPSHA (as described in chapter 5) considering Bangalore is as “Site Class D”.To evaluate the spectral acceleration at ground surface with probabilistic considerations, the spectralacceleration relation developed for southern India by RaghuKanth and Iyengar (2007) has been used:2ln y = c1 + c2( M − 6) + c3( M − 6) − ln R − c4R+ ln( ∈)(5.6)Where y, M, R and ∈ refer to spectral acceleration (g), moment magnitude, hypocentral distance and errorassociated with the regression respectively. The coefficients in equation (5.6), c 1 , c 2 , c 3 , and c 4 are obtained fromRaghuKanth and Iyengar (2007) for rock site (see Table 4.6).Coefficients of eq. (5.6) are modified for local sitecondition for site class D (Raghukanth and Iyengar,2007) as detailed below:ys= yF sand ln Fs= a1 y + a2+ lnδs(5.7)Where a 1 and a 2 are regression coefficients and δsis the error term corresponding to site classifications of D.These coefficients along with the standard deviation and the error were obtained from Raghukanth and Iyengar(2007) which is given in Table 5.7.Table 5.7: Coefficients for local site classification of DPeriodCoefficients(s) a 1 a 2 σ(ln δs)0.000 −2.61 0.80 0.360.010 −2.62 0.80 0.370.015 −2.62 0.69 0.370.020 −2.61 0.55 0.340.030 −2.54 0.42 0.310.040 −2.44 0.58 0.310.050 −2.34 0.65 0.290.060 −2.78 0.83 0.290.075 −2.32 0.93 0.190.090 −2.27 1.04 0.290.100 −2.25 1.12 0.190.150 −2.38 1.40 0.280.200 −2.32 1.57 0.190.300 -1.86 1.51 0.160.400 −1.28 1.43 0.160.500 −0.69 1.34 0.210.600 −0.56 1.32 0.210.700 −0.42 1.29 0.210.750 -0.36 1.28 0.190.800 −0.18 1.27 0.210.900 0.17 1.25 0.211.000 0.53 1.23 0.15125

Local site effects and Site Response1.200 0.77 1.14 0.171.500 1.13 1.01 0.172.000 0.61 0.79 0.152.500 0.37 0.68 0.153.000 0.13 0.60 0.134.000 0.12 0.44 0.15The hazard curve and spectral acceleration spectrum has been evaluated as discussed in PSHA Chapter 4. Figure5.34 shows the mean spectral acceleration mean annual rate of exceedance versus spectral acceleration for aperiod of 1s and 5% damping considering the “Site class D”. From Figures 3.28 and 5.34, it is very clear that thespectral acceleration from each source due to the local soil condition get modified (increased), resulting thatincreasing in the cumulative spectral acceleration. Figure 5.35 presents the variation of mean annual rate ofexceedance versus cumulative spectral acceleration for a period of 1s and 5% damping for rock site and soilwith site class D. Figure 5.36 presents the plot of UHRS for 10% probability of exceedance in 50 yearsconsidering local site conditions. A value of PGA (ZPA = PGA) 0.33g is obtained when local site effect is takeninto account. The observed PGA value at considering the local soil conditions is higher than bed rock level,shows an amplification of 2.73 due to local soil condition. Surface level PGA by PSHA using “site class D” isshows in Figure 5.37.This amplification factor is with in the range observed by other methods as describedearlier.Mean Annual rate of Exceedance1.E+001.E-011.E-021.E-031.E-041.E-05L15ArkF19L16L20L22Cumulative1.E-060 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6Spectral Acceleration (g)Figure 5.34: Spectral acceleration at surface level corresponding to period of 1s and 5% damping forBangalore.126

Local site effects and Site Response1.E+00Mean annual rate of exceedance1.E-011.E-021.E-031.E-041.E-051.E-061.E-07Site DBed rock1.E-080 0.1 0.2 0.3 0.4 0.5 0.6Spectral Acceleration (g)Figure 5.35: Spectral acceleration at bed rock and site D corresponding to period of 1s and 5% dampingfor Bangalore.Spectral Acceleration (g)0.60.50.40.30.20.10Bed rockSite Class D0 0.5 1 1.5 2period (s)Figure 5.36: Uniform hazard response spectrum for bed rock and site class D with 10% probability ofexceedance in 50 years (5% damping) for Bangalore.5.11 CONCLUSIONS.The synthetic bedrock motion generated for the selected 170 borehole locations and the soil profile details fromborehole information were considered as input and site specific ground response analysis was carried out for170 locations in Bangalore using SHAKE 2000. The acceleration time history at the ground surface is obtainedas output and was characterized by the PGA and the response spectra. The PGA at ground surface varied from0.088g to 0.658g. The high value of surface accelerations at some locations may be due to the frequency of the127

400m 200 0 80016002 cm to 400m2400 metersLocal site effects and Site ResponsePGA at surface for Site D at 100.3850.3900.3950.4000.4050.3800.3750.3700.4100.3650.3450.3500.3600.4150.355Figure 5.37: Surface level PGA for site class D using PSHAbase motions coinciding with that of the fundamental frequency of the soil column. The amplification factorwhich is a measure of amplification potential of the soil column was computed using this PGA and the peakacceleration at rock level. The range of amplification factor was 1 and 4.8. The high amplification factor at somelocations is due to the presence of filled up soils and silty sand deposit in the silted tank beds, shallow watertable depths and low SPT values which results in low average shear wave velocities. Thus, amplification ofseismic waves depends not only on overburden thickness but also on various other factors like the frequency ofthe input motion, average shear wave velocity of the soil profile and water table depth. The results are reportedas amplification factor map indicating zones of high vulnerability. The regions in zones I and II are seismicallymore stable than the regions in zones III and IV. Most of the area in Bangalore lies in zone II. Results obtainedfrom the ground response analysis show that the natural periods of the analyzed deposits are in between 0.01sand 0.45s with about 85% of the locations having a period below 0.2s. The response spectra for 5% damping atthe ground surface obtained for all the borehole locations clearly indicate that the range of spectral accelerationat different frequencies varied over a wide range. The peak spectral acceleration varies from 0.2g to 2.1g andfrequency corresponding PSA varies from 3Hz to 20Hz. At 5% damping, the range of SA at 1.5 Hz frequencywas 0.01 g to 0.07 g, at 3 Hz frequency it was in the range of 0.03 g to 0.65 g, while at 5 Hz frequency it was inthe range of 0.08 g to 1.14g.A similar site response study has been carried out using measured shear wave velocity from MASW. Peakacceleration at ground, peak spectral acceleration, amplification factor, period of soil column and spectralaccelerations at difference frequencies matches well with SPT site response results. However the MASW resultsare lower than SPT results. Response spectrum obtained from SHAKE2000 matches well with the shape of thespectral acceleration plots presented in IS 1893, 2002 and uniform hazard spectrum. Predominant frequenciesof each borelog are estimated using both SPT data and MASW data. The result shows that predominantfrequencies are similar from both analysis and varies from 4Hz to 12Hz using SPT data and 3.45Hz to 12Hzusing MASW data. An attempt has been made to develop the correlation between corrected SPT ‘N’ valuesversus the low strain shear modulus.Further, predominant frequencies are estimated using microtremor based on ambient noise measurement atabout 54 locations. Dominant frequencies of soil column has been estimated with a ±1σ standard deviation foreach site and presented. The microtremor studies show that the predominant frequency of Bangalore soil varies128

Local site effects and Site Responsefrom 1.5Hz to 12Hz. The predominant frequencies from both the MASW and microtremor also match well.Generally Bangalore soil has predominant frequencies of 3 Hz to 12Hz. The most notable feature is that thelower frequencies are closer to the lakes and erstwhile lakes.The spectral acceleration hazard curve and response spectrum for “site class D” has been developed. The peakground acceleration (PGA) value of 0.121g at rock level and a value of 0.33g was obtained at ground surfaceconsidering local soil conditions. The spectral acceleration value corresponding to zero period (representingPGA) obtained considering the local soil conditions is 2.73 times higher than that observed from the bed rocklevel spectral acceleration for Bangalore region. Also Uniform Hazard Response Spectrum for bed rock and“Site class D” with 10% probability of exceedance in 50 years (5% damping) for Bangalore has been presented.129

CHAPTER 6LIQUEFACTION HAZARD ASSESSMENT6.1 INTRODUCTIONThe response of soil due to seismic hazards producing a significant amount of cumulative deformation orliquefaction has been one of the major concerns for geotechnical engineers working in the seismically activeregions. Liquefaction can occur in moderate to major earthquakes, which can cause severe damage to structures.Transformation of a granular material from solid state to liquid state due to increased pore pressure and reducedeffective stress is defined as liquefaction (Marcuson 1978). When this happens, the sand grains loose itseffective shear strength and will behave more like a fluid. The grain size distribution of soil, duration ofearthquake, amplitude and frequency of shaking, distance from epicenter, location of water table, cohesion ofthe soil and permeability of the layer affects liquefaction potential of soil. The liquefaction hazards areassociated with saturated sandy and silty soils of low plasticity and density.The liquefaction potential of soil is generally estimated from laboratory tests or field tests. Among the field insitutests, the SPT test has been widely used for this purpose. Corrected ‘N’ values from large number of SPTtest data in Bangalore have been used for direct assessment of ground’s liquefaction resistance in this work. Anattempt has been made to prepare the susceptibility and liquefaction hazard maps. A simple spread sheet hasbeen generated for applying corrections to ‘N’ field values and calculation of factor of safety againstliquefaction. About 620 bore hole information are used for this purpose and the factor of safety againstliquefaction are estimated. Based on the factor of safety, the regional liquefaction hazard maps are generated forthe estimated peak ground acceleration at the ground surface. At few locations the undisturbed soil samples hasbeen collected and laboratory cyclic triaxial tests have been carried out to validate the liquefaction resistance ofsoil. Liquefaction hazard map will be very useful to identify vulnerable areas for the future microzonation andseismic risk studies.6.2 LIQUEFACTION SUSCEPTIBILITY MAPLiquefaction ‘susceptibility’ is a measure of a soil’s inherent resistance to liquefaction, and can range from notsusceptible, regardless of seismic loading, to highly susceptible, which means that very little seismic energy isrequired to induce liquefaction. Susceptibility has been evolved by comparing the properties of top soil depositsof Bangalore to the other soil deposits where liquefaction has been observed in the past (based on Seed et al,1985). Liquefaction susceptibility was 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 theborelogs. These susceptible areas have been identified by considering the approach of Pearce and Baldwin(2005). Soil is susceptible for liquefaction if (1) presence of sand layers at depths less than 20m, (2) encounterwater table depth less than 10m, and (3) SPT field “N” blow counts less than 20. By interpolation, susceptibilityof map has been prepared. To prepare the liquefaction susceptibility map, Bangalore map along with differentlayers as discussed in chapter 6 have been used. Liquefaction susceptibility map for Bangalore city has beenshown in Figure 6.1. From Figure 6.1 only susceptible areas are further considered for the evaluation of factorof safety against liquefaction.6.3 FACTOR OF SAFETY AGAINST LIQUEFACTION ASSESSMENTFactor of Safety against liquefaction of soil layer has been evaluated based on the simplified procedure (Seedand Idriss, 1971) and subsequent revisions of the simplified procedures (Seed et al., 1983, 1985; Youd et al.,2001; Cetin et al., 2004). In this study, the earthquake induced loading is expressed in terms of cyclic shearstress and this is compared with the liquefaction resistance of the soil. Liquefaction calculation or estimationrequires two variables for evaluation of liquefaction resistance of soils. Two variables are defined based oncyclic 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 represented by Cyclic Resistance Ratio (CRR).130

Liquefaction Hazard AssessmentFigure 6.1: Liquefaction susceptibility map for Bangalore cityHere liquefaction resistance is estimated using an in-situ test based on corrected SPT ‘N’ values. The appliedcorrections for the field ‘N’ value are discussed in Chapter 6. Steps involved in the calculation of liquefactionhazard are shown in Figure 6.2 as a flow chart.Factor of safety against liquefaction: If the cyclic stress ratio caused by the earthquake is greater than thecyclic resistance ratio of in situ soil, then liquefaction could occur during an earthquake. The factor of safetyagainst liquefaction is defined as follows:FS⎛ CRR7.⎜⎝ CSR=5⎞⎟MSF⎠Here subscript 7.5 for CRR denotes that CRR values calculated for the earthquake moment magnitude of 7.5.MSF is the magnitude scaling factor. The higher factor of safety means that soil is having more resistance toliquefaction.6.3.1 Peak Ground AccelerationEstimation of factor of safety against liquefaction of soil layer requires the ground level peak acceleration due toan earthquake. In the previous chapter, ground level peak horizontal accelerations have been estimated by thesite response studies using equivalent linear response analysis software (using SHAKE2000 program). About170 SPT boreholes and 58 MASW tests have been used to study the amplification potential of Bangaloreconsidering the synthetic ground motions and ground level peak horizontal acceleration map has been generated.Ground level peak horizontal acceleration (PHA) map shown in Figure 7.10, has been used to arrive the PHA ateach borehole location using interpolations. This peak ground acceleration is further used to estimate the cyclicstress ratio (CSR).(6.1)131

Liquefaction Hazard Assessment6.3.2 Cyclic Stress Ratio (CSR)The excess pore pressure generation to initiate liquefaction depends on the amplitude and the duration of theearthquake induced cyclic loading. In the cyclic stress approach the pore pressure generation is related to thecyclic shear stresses, hence the earthquake loading is represented in terms of cyclic shear stresses. Theearthquake loading can be evaluated by using Seed and Idriss (1971) simplified approach. The earthquakeloading is evaluated in terms of uniform cyclic shear stress amplitude and it is as given below:⎛ amax⎞ ⎛ σvo⎞Cyclic stress ratio (CSR) = 0.65⎜⎟ rdg⎜⎟(6.2)⎝ ⎠ ⎝σvo' ⎠aIn this equation 0.65 maxrepresents 65 % of the peak cyclic shear stress, a maxis peak ground surfacegacceleration, g is the acceleration due gravity, σvoand σvo'are the total and effective vertical stresses and rd=stress reduction coefficient. For the calculation of stress reduction coefficient many correlation are availablewhich are discussed in detail in a 1996 NCEER workshop report ( Youd et al 2001). Youd et al (2001)recommends that for routine practice and non-critical projects, the equations given by Liao and Whitman (1986)may be used to estimate average values of r d.In this study the same has been used and it is given as below:r d= 1.0− 0. 00765z for z≤9.15 m (6.3)r d= 1.174− 0. 0267z for 9.15 m < z ≤ 23 m (6.4)6.3.3 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 are attempted. Cyclic resistance ratio (CRR) isarrived based on corrected “N” value as per Seed et al. (1985), Youd et al., (2001); Cetin et al., (2004). Seed etal. (1985) presents a plot of CRR versus corrected ‘N’ value from a large amount of laboratory and field data.However, here the corrected ‘N’ values are used to calculate the CRR for the magnitude of 7.5 earthquakesusing the equation proposed by Idriss and Boulanger (2005) as given below:234⎪⎧( N⎪⎫1)60cs⎛ ( N1)60cs⎞ ⎛ ( N1)60cs⎞ ⎛ ( N1)60cs⎞CRR = exp⎨+ ⎜ ⎟ − ⎜ ⎟ + ⎜ ⎟ − 2. 8⎬(6.5)⎪⎩14.1 ⎝ 126 ⎠ ⎝ 23.6 ⎠ ⎝ 25.4 ⎠ ⎪⎭However this estimation is proposed for a magnitude of 7.5 on the Richter scale. For the present study, for theearthquake moment magnitude of 5.1 has been considered for evaluating Magnitude Scaling Factor (MSF).6.3.4 Magnitude Scaling Factor (MSF)The CRR curves either using the SPT N values or CPT q c values or shear wave velocity Vs corresponding to anearthquake of magnitude 7.5. Seed and Idriss (1982) suggested the use of magnitude scaling factors (MSF) forearthquakes of magnitude other than 7.5. The available MSF are Seed and Idriss (1982) scaling factors, RevisedIdriss scaling factors (1995), Ambraseys (1988) scaling factors, Arango (1996) scaling factors, Andrus andStokoe (1997) scaling factors and Youd and Noble (1997) scaling factors. Detailed discussion and comparisonof these scaling factors are available in Youd et al (2001) and Bhandari et al (2003). An NCEER- 1996 and1998 NCEER/NSF workshop (Youd et al., 2001) recommends the revised Idriss scaling factors (1995) and itwas used by Yilmaz and Bagci (2006) for soil liquefaction susceptibility and hazard mapping in the residentialarea of Kutahya (Turkey). The magnitude-scaling factor used in the present study is the revised Idriss scalingfactor for the magnitude less than 7.5 and it is given as below:⎡ 2.2410 ⎤MSF = ⎢ ⎥(6.6)⎢2.56⎣ M W⎥⎦132

Liquefaction Hazard AssessmentData from Borelog and Site Response StudyCyclic Stress RatioBased on Simplified Approach⎛ a max ⎞ ⎛ σvo⎞CSR = 0.65 ⎜ ⎟ rdg⎜vo '⎟⎝ ⎠ ⎝ σ ⎠IF SAND?NoYESCyclic Resistance Ratio (CRR 7.5 )By using CRR Versus (N 1 ) 60csNoIf LL>32⎡ 10MSF = ⎢⎢⎣M2 .242 .56W⎤⎥⎥⎦Magnitude Scaling FactorYES⎡CRRFS = ⎢⎣ CSR7.5⎤⎥MSF⎦Factor of SafetyDetailed studyGroupingFactor of SafetyFS 4- SafeDS = Detailed StudyNL = Non liquefiable6.3.5 Factor of Safety CalculationFigure 6.2: Flow chart for liquefaction hazard assessmentFrom the available 850 geotechnical borelog data base, about 620 borelogs has been selected for presentcalculations. After applying necessary corrections to SPT ‘N’ values (as discussed in chapter 6) corrected “N”[(N 1 ) 60cs ] values were obtained. A simple excel spread sheet has been developed to automate these calculationsfor all the 620 borelogs with depth. The factor of safety for each layer of soil was arrived by consideringcorresponding “(N 1 ) 60cs ” values. Typical liquefaction analysis is shown in Table 6.1. It is to be noted here that,apart from Seed and Idriss (1983) recommendation, the fines content in the soil are considered usingrepresentative parameters such as liquid limit (LL) and/ Plasticity Index (PI). The soil having the liquid limit of133

Liquefaction Hazard Assessmentmore than 32 is recommended (Boulanger and Idriss 2004) for the detail study (DS), which can account thestrength loss in plastic silts or clays during cyclic or seismic loading.Table 6.1: Typical liquefaction analysis for a boreholeMagnitude, M w = 5.1 Peak Acceleration = 0.35gDepthCorrectedN valueσvoσvo'r d CSR FC Liquid Limit CRR MSF FS(m) (N 1 ) 60cs kN/m 2 kN/m 2 %1.50 4 30.00 30.00 0.99 0.22 46.2 0 0.08 2.68 0.943.20 3 64.00 47.32 0.98 0.30 40.9 0 0.08 2.68 0.694.20 21 84.00 74.19 0.97 0.25 53.3 26 0.22 2.68 2.365.20 20 104.00 94.19 0.96 0.24 53.1 31 0.21 2.68 2.357.00 44 140.00 122.34 0.95 0.25 57.1 25 19.54 2.68 NL8.50 102 170.00 155.29 0.93 0.23 59.2 27 NL 2.68 NL6.4 LIQUEFACTION HAZARD MAPLiquefaction hazard mapping has been done by many researchers using SPT data (Palmer et al, 2003; Brankmanet al, 2004; Pearce and Baldwin, 2005 and Yilmaz and Bagci, 2005). Similar approach has been followed in thispaper to map liquefaction potential. The liquefaction hazard map is prepared for the moment magnitude of 5.1.The minimum factor of safety from each bore log has been considered to represent the factor of safety againstliquefaction at that location, which are used for the mapping. These factors of safety against liquefaction havebeen grouped in to 4 as shown in Table 6.2.Table 6.2: Factor of safety and severity index of liquefactionFigure 6.3 shows the map of factor of safety against liquefaction (FS) for Bangalore city to the local magnitudeof 5.1. Figure 6.3 it is very clear that the factor of safety against liquefaction is more than unity for a localmagnitude of 5.1 in most of the area except in northwestern part (area in and around Kurubahalli) and southernpart (in and around Venkatapura, Jakasandra, Sri Narashimaraj colony, Basaveshwara nagar andKethamaranahalli) of Bangalore.6.5 RESULTS AND DISCUSSIONGroup Factor of safety range Severity Index1 3 Non liquefiableOut of 620 locations liquefaction analyses indicates that the factor of safety is less than one in only for 4.2% oftotal locations. Factor of safety of 1 to 2 and 2 to 3 each having 14.7% and 12.5% of the total locations, factor ofsafety of more than 3 is about 68% of the total locations. At 33% of total locations detailed study is neededusing laboratory tests according to Idriss and Boulanger (2005). Where these soil (silty clay having PI>12) cancause the stress reduction during earthquake. Because in these locations the liquid limit is more than 32 andplasticity index are more than 12 due to presence of silty clay soil. Most of the data points having PI>12 falls inthe area of factor of safety more than 3. Areas close to water body and streams have the factor of safety less thanunity. However, the areas having FS less than 1 is very less.Typical borelogs having a factor of safety less than 1 is shown in Figure 6.4. Figure 6.4 clearly shows that, up toa depth of about 6 m very loose silty sand with clay and sand are found in these locations which are classified asmedium to fine sand having very low field ‘N’ values. Also in this location shallow water table has been met,ground water has been found at 1.2m from the ground level. These factors may attribute to the low factor ofsafety in these locations.134

Liquefaction Hazard AssessmentFigure 6.3: Distribution of factor of safety against liquefaction135

Liquefaction Hazard AssessmentBorelogs corresponding to a factor safety 1 to 2 show that, these location having moderate field ‘N’ values. Alsoground water has been reported at about 2m from the ground level. These locations have the filled up soil, sandysoil and clayey silt up to depth of 5 m. Area covered by this range of factor safety is lager than area havingfactor of safety less than 1, i. e. in and around Venkatapura, Jakasandra, Sri Narashimaraj colony, Basaveshwaranagar and Kethamaranahalli . Typical borelog for these locations is shown in Figure 6.5.BORE LOGLocation Palace Guttahalli Road,Bangalore Date of commencement 4.12.1996BH No 2 Date of completion 5.12.1996Ground Water Table1.2mDepth Soil Description Thickness Legend Details of Sampling SPTBelow of Strata Depth Type N'GL(m) (m) (m) Value0.5 Filled Up SoilYellowish1.00.52.0 Sandy Silt1.51.5 SPT* N=1Greyish/BlackishSoft Clay 2 UDS*3 2.8 DS3.51.54 Greyish/Brownish4 SPT N=1Sand(medium to fine)5.3 DS5.0 5.5 SPT 8/9/122.3N=21Whitish Sand with Gravel6 6.6 SPT 8/8/09N=1778 SPT 10/11/228 2.2N=33Greyish/Whitish Silty ClayeySand(medium to fine)9.2 SPT 27/29/329 N=61102Borehole terminated at 10mBore hole Terminated at 10mNote* Sample not retrived SPT Standard Penetration TestR Rebound UDS Undisturbed SampleFigure 6.4: Typical bore log corresponding to location which has factor of safety less than 1136

Liquefaction Hazard AssessmentBORE LOGLocation koramangakla, BangaloreDate of commencement 23.1.1997BH No 4 Date of completion 24.1.1997Ground Water Table2mDepthbelowGroundlevel (m)Soil DescriptionFilled Up SoilThicknessDetails of SamplingLegendof Strata Depth Type(m)(m)SPT 'N'Value1.011.5 Greyish/Yellowish Sandy 1.5 SPT 5/7/093 UDS N=163.0 Clayey Silt4.5 SPT 8/11/175.0 4N=286.0 Yellowish/Greyish6 SPT 10/13/19N=32Clayey Sand 7 SPT 12/18/207.0 N=389 SPT 15/22/269.0. N=4810.55.5Note* Sample not retrived Standard Penetration TestUndisturbed SampleFigure 6.5: Typical bore log corresponding to locations which has factor of safety between 1 to 2Figure 6.6 shows the typical bore log corresponding to the factor of safety of 2 to 3. These locations have thefield ‘N’ values of around 20 with ground water level of 2.5m from the ground level. The weathered rock startfrom 4.5 m and most of the area has soil consisting of sandy clay.137

Liquefaction Hazard AssessmentBORE LOGLocation 23, Govindappa road, Basavanagudi,Bang Date of commencement 25.4.1996BH No 107-2 Date of completion 26.4.1996Ground Water Table2.5 mDepth Thickness Legend Details of SamplingSPT 'N'Below Soil Description of Strata Depth TypeValueGL(m) (m) (m)0.5 Reddish/Brownish Sandy Silt(filled up)0.51.0 YellowishAlready excavated Up to 1.5m1.5 Sandy Clay12.0 Yellowish/Greyish Sandy2 DS 8/9/10Clay with silt 2.5 SPT N=193.03 UDS 10/16/164.0 3.5 SPT N=324.32.85.0 Yellowish/Whitish/GreyishSandy Silt 24 DS 13/18/224.5 SPT N=32(weathered rock) 2.8 18/20/366.0 1.76 SPT R for 10cmpenetrationN>567 Borehole terminated at 6.40 m10.0Bore hole Terminated at 6.400mNote* Sample not retrived SPT Standard Penetration TestR Rebound UDS Undisturbed SampleFigure 6.6: Typical bore log corresponding to locations which has factor of safety between 2 to 3Analysis shows that about more than 68% of locations have the factor of safety more than 3, because theselocations have lager field ‘N’ value (more than 20) and deeper ground water level. At many locations up toabout 15m water table is not reported. The general soils found in these locations are silty sand and sand(classified as medium to dense sand with clay). Typical bore log corresponding to factor of safety of more than3 is shown in Figure 6.7. At these locations, soil is classified as non liquefiable. However many of138

Liquefaction Hazard AssessmentBORE LOGLocation Sannidhi Road Date of commencement 14.5.2001BH No 601-1 Date of completion 15.5.2001Ground Water TableNEDepth Soil Description Thickness Legend Details of SamplingSPT 'N'Below of Strata Depth TypeValueGL(m) (m) (m)Reddish Silty Sand with1.0Clay1.5 SPT 7/10/19N=292.0 2 UDS3.0 3 SPT 12/10/10N=204.04.5 SPT 14/16/184.5 N=345.0 Yellowish Silty Sand6 SPT 19/21/266.0 1.5N=4778Bore hole Terminated at 10mNote* Sample not retrived SPT Standard Penetration TestNE-Not EncounteredUDS Undisturbed SampleFigure 6.7: Typical bore log corresponding to locations which has factor of safety more than 3these locations within this area, has soil layers of silty sand and sandy silt, in presence of clay (which is plasticin nature). Typical bore log corresponding to these locations are given in Figure 6.8. The clay present has aliquid limit of more than 32 and plasticity index of more than 12, indicating soils of non liquefiable nature,however during the earthquake loading, stress reduction can occur. Typical grain size distribution and Atterbergvalues are also given in Table 6.3In general, Bangalore is safe against liquefaction, except at few locations where the ground water is veryshallow and loose silty sand deposits are found. Many parts of Bangalore has soil, rich in clay content. This maycause strength loss during the cyclic loading and these have to be studied in detailed using laboratory tests asrecommended by Idriss and Boulanger (2005).139

Liquefaction Hazard AssessmentBORE LOGLocation Govt.Boys School MalleshwaramDate of commencement 28.06.1994BH No 26-1 Date of completion 29.06.1994Ground Water TableNEDepth Soil Description Thickness Legend Details of SamplingSPT 'N'Below of Strata Depth TypeValueGL(m) (m) (m)1.0Brownish,Yellowish Sandy Clay 3.52.03.03.5 Yellowish,Reddish Clayey Silt4.3 DS 37with Sand4.0 5.5 DS 305.0 6.8 DS 566.03.36.8 Reddish Clayey Silt with Sand 0.777.58 Yellowish red clayey silt 0.8 recovered8.3Soft rock98 SPT 68 for15 cm10Bore hole Terminated at 8.7 mNote* Sample not retrived SPT Standard Penetration TestNE-Not EncounteredUDS Undisturbed SampleDS - Disturbed Sample R ReboundFigure 6.8: Typical bore log showing the presence of clay6.5.1 Cyclic Triaxial Experiments on Undisturbed Soil SamplesUndisturbed samples were collected from few locations in (south west region) Bangalore city to verify theliquefaction potential of the soil. Cyclic triaxial tests have been carried out in the laboratory on the undisturbedsoil samples collected from Boreholes locations of 482, 810 and 91. The tests have been carried out as perASTM: D 3999 (1991) in strain controlled mode. Cyclic triaxial tests are carried out with double amplitudeaxial strains of 0.5%, 1% and 2% with a frequency of 1Hz. Typical cyclic triaxial test results are presented inFigures 6.9 and 6.10. Figure 6.9 shows the variation of deviatoric stress versus strain plot for more than 120cycles of loading (axial strain = 0.25%; applied confining pressure 100 kPa, for the undisturbed samplecorresponding to a depth 3m below GL, in-situ density of the soil sample 2.0 gm/cc with in-situ moisturecontent 15%, at 3.0m depth). Figure 6.10 shows the pore pressure ratio versus number of cycles. From theseplots it is clear that even after 120 cycles, the average pore pressure ratio is about 0.94 and deviatoric stresses vs.strain plots have not become flat, indicating no liquefaction. The resistance to liquefaction of these soils is veryhigh. The calculated factor of safety against liquefaction results for this borehole is also very high indicating noliquefaction and results matches well with the lab test results.140

Liquefaction Hazard AssessmentTable 6.3: Typical laboratories reports corresponds to high plastic indexBHNo.Depth(m)Bulk Density(g/cc)Grain Size Distribution (%) Atterberg Limits (%)Water ContentSandFines Liquid Limit Plastic Limit(%) Gravel Coarse Med Fine Silt & Clay (LL)(PL)Plastic Index(PI)26-1 4.3 2.1 14 2.1 6.3 12.6 20 59 65.2 26.4 395.5 2.1 18 3.5 2.4 8.2 11.8 74.1 74.5 35.2 396.5 2.1 19 0.6 1.2 7.6 13.5 77.1 70.1 35.6 357.5 1.2 5.5 13.2 80.1 74 39 358 1.9 20 0.9 1.9 5.8 18.1 73.3 75 41.8 3326-2 3.6 1 2 21 39 37 39.2 18.1 214.5 2 18 1.5 3.1 20.9 26 48.5 37.4 17.9 205.5 2.1 17.1 0.9 0.9 7.2 42.8 48.2 39.5 18.8 216 2 17.2 1.3 1.2 11.9 31.1 54.5 43.2 24 197 1.9 17.3 0.5 1.1 8.9 26.9 62.6 44.8 22.4 22141

Liquefaction Hazard Assessment4030Deviatoric Stress (kPa)20100-0.003 -0.002 -0.001-100 0.001 0.002 0.003-20-30-40Axial StrainFigure 6.9: Typical hysteresis loops from cyclic triaxial testsPore Pressure Ratio1.000.980.960.940.920.900.880.860.840 20 40 60 80 100 120 140No of CyclesFigure 6.10: Typical pore pressure ratio plots with no of cycles6.6 CONCLUDING REMARKSThe liquefaction hazard study has been carried out and liquefaction susceptibility map has been prepared basedon the soil layer properties, water table depth and field ‘N’ values. The factor of safety against liquefaction hasbeen calculated for the susceptible areas in Bangalore city. About 620 borelogs have been used for calculationof factor of safety against liquefaction. Simple spread sheets have been developed using Seed and Idriss (1971)simplified procedure. The results are grouped in to four groups for mapping and presented in the form of 2-dimensional maps. About 85% of Bangalore have higher factor of safety and non liquefiable. However, at manylocations the liquid limit is more than 32 and plasticity index is more than 12. At these locations, it is suggestedthat a detailed study need to be carried out using laboratory cyclic triaxial tests to evaluate strength loss. Thisstudy shows that Bangalore is safe against liquefaction except at few locations where the overburden is sandysilt with shallow water table. Strain controlled cyclic triaxial tests on undisturbed soil samples collected fromsouth west region of Bangalore city indicates that soils are non liquefiable.142

CHAPTER 7INTEGRATION OF HAZARD MAPS ON GIS PLATFORM7.1 INTRODUCTIONSeismic microzonation is the generic term for subdividing a region into smaller areas having different potentialfor hazardous earthquake effects, defining their specific seismic behavior for engineering design and land-useplanning. Usually seismic microzonation includes approaches for assessing local ground response, slopeinstability and liquefaction. Microzonation studies involve experimental techniques together with theoreticalapproaches involving ground motion modeling. In this chapter an integration of all the developed maps isattempted based on weights and ranks. A final hazard index map for BMP area is developed using AnalyticHierarchy Process (AHP) on GIS (Geographical Information System) platform. Application of GIS formicrozonation mapping is amply demonstrated by many researchers all over the world. Nath (2004) used GIS asintegration tool to map seismic ground motion hazard for Sikkim Himalaya in India. In this study, similarapproach of Nath (2004) is used to develop a hazard index map where in the seismic hazard parameters areintegrated and coupled with ground information. The hazard index maps are prepared using both deterministicand probabilistic approaches.7.2 GEOGRAPHIC INFORMATION SYSTEM (GIS)Geographical Information System (GIS) provides a perfect environment for accomplishing comprehensiveregional information including seismic damage assessment. GIS has the capability to store, manipulate, analyzeand display a large amount of required spatial and tabular data. One of the most important features of ageographic information system is the data analyses of both spatial (graphic) and tabular (non-graphic) data. Theprocedures for data analysis typically found in most GIS programs are as follows:Map overlay procedures, including arithmetic, weighted average, comparison, and correlation functions.Spatial connectivity procedures, including proximity functions, optimum route selection and network analysis.Spatial neighborhood statistics, such as slope, aspect ratio, profile and clustering.Measurements of line and arc lengths, point-to-point distances, polygon perimeters, areas and volumes.Statistical analysis, including histograms or frequency counts, regressions, correlations and cross-tabulation.Report generation, including maps, charts, graphs, tables and other user-defined information.Figure 7.1 shows the fully integration GIS as a flow chart. Numerous application-specific functions for theanalysis of geotechnical data and hazard mapping exist in a GIS platform. Most systems include some sort ofbuilt-in programming capability usually in the form of a software-specific macro language. This allows the userto develop a set of functions or analysis procedures that can be stored in a user-defined library. Often, the GISmacro language is very simplified and doesn’t have to handle very high-level computational features such asrecursion, numerous simulations, subscripted variables, and subroutines. For this reason, most GIS programshave the ability to communicate with external analysis and modeling programs. A system can typically outputdata in various formats to be used in various external programs such as spreadsheets, word processing, graphics,and other user-specified executable programs. The results of an external analysis can then be used by GIS asboth graphic and non-graphic data for further interpretation and analysis, or for final report and map generation.With these wide areas of application, GIS play a unique role for hazard preparedness and management. In thisstudy Arc GIS 9.2 has been used to integrate all themes.143

Integration of hazard maps on GIS PlatformAutomatedMappingSystemMAPInformationSystemData BaseManagementSystemSpatialModelingSystemFully integratedGIS SystemNon-SpatialModelingSystemAnalysis andModeling System7.2.1 GIS Integration LogicFigure 7.1: The information system composing a fully integrated GISThe representation and interpretation of uncertainty related to the classification of individual locations providedby the fuzzy logic based on location attribute values. Fuzzy logic implements classes or groupings of data withboundaries. The central idea of fuzzy sets is aided by the Analytic Hierarchy Process (AHP). AHP is a multicriteriadecision method that uses hierarchical structures to represent a problem and then develop priorities forthe alternatives based on the judgment of the user (Saaty, 1980). The idea of multi-criteria decision-making wasbased on the concept of McHarg (1968). McHarg (1968) introduced a systematic land use planning by using theconcept of compatibility of multiple land uses. He mentioned that the factors affecting land and its relativevalues are different and, therefore, it is difficult to think of optimizing them for a single use. It can be optimizedfor multiple compatible uses. He introduced simple matrix system for determining the degree of compatibility.Saaty (1968) has shown that weighting activities in multi-criteria decision-making can be effectively dealt withhierarchical structuring and pair-wise comparisons. Pair-wise comparisons are based on forming judgmentsbetween two particular elements rather than attempting to prioritize an entire list of elements (Saaty, 1980). Formulti-criteria evaluation, Saaty's Analytical Hierarchy Process (AHP) is used to determine the weights of eachindividual criterion (Saaty, 1990). AHP is a mathematical method to determine priority of criteria in the decisionmaking process. It is a popular tool used by decision makers in the multi-attribute decisions.Saaty's Analytical Hierarchy process constructs a matrix of pair-wise comparisons (ratios) between the factorsof earthquake hazard parameters (EHP). The constructed matrix shows the relative importance of the EHP basedon their weights. If 9 earthquake hazard parameters are scaled as 1 to 9, 1 meaning that the two factors areequally important, and 9 indicating that one factor is more important than the other. Reciprocals of 1 to 9 (i.e.,1/1 to 1/9) show that one is less important than others. The allocation of weights for the identical EHP dependson the relative importance of factors and participatory group of decision makers. Then the individual normalizedweights of each EHP are derived from the matrix developed by pair-wise comparisons between the factors ofEHP. This operation is performed by calculating the principal Eigen vector of the matrix. The results are in therange of 0 to 1 and their sum adds up to '1' in each column. The weights for each attribute can be calculated byaveraging the values in each row of the matrix. These weights will also sum to '1' and can be used in derivingthe weighted sums of rating or scores for each region of cells or polygon of the mapped layers (Jones, 1997).Since EHP vary significantly and depends on several factors, they need to be classified into various ranges ortypes, which are known as the features of a layer. Hence each EHP features are rated or scored within EHP andthen 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 normalizedusing the relation,144

Integration of hazard maps on GIS PlatformXR − Ri mini= (7.1)Rmax− RminWhere, R i is the rating assigned for features with single EHP, R min and R max is minimum and maximum rate ofparticular EHP.7.3 EARTHQUAKE HAZARD PARAMETERSSeismic microzonation is subdividing a region into smaller areas having different potential for hazardousearthquake effects. The earthquake effects depend on ground geomorphological attributes consisting ofgeological, geomorphology and geotechnical information. The parameters of geology and geomorphology, soilcoverage/thickness, and rock outcrop/depth are some of the important geomorphological attributes. Otherattributes are the earthquake parameters, which are estimated by hazard analysis and effects of local soil for ahazard (local site response for an earthquake). The Peak Ground Acceleration (PGA) [from deterministic orprobabilistic approach], amplification/ site response, predominant frequency, liquefaction and land slide due toearthquakes are some of the important seismological attributes. Weight of the attributes depends on the regionand decision maker, for example flat terrain has weight of “0” value for land slide and deep soil terrain hashighest weight for site response or liquefaction. Different attributes considered for Bangalore microzonation arepresented below;7.3.1 Geomorphological AttributesThe geomorphological attributes (here after called as themes) considered in this study are the geology andgeomorphology (GG), rock depth/ soil thickness (RD/ST), soil type and strength (represented in terms ofaverage shear wave velocity) (SS), drainage pattern (DP) and elevation level (EL).Geology and Geomorphology (GG)As the study area is densely covered by buildings, it is very difficult to obtain the detailedgeological/geomorphological maps. A simple geological and geomorphology map of Bangalore is prepared andpresented in Figure 7.2 for the study area. Bangalore lies on top of the south Karnataka Plateau (MysorePlateau) and its topology is almost flat with the highest point being at Doddabettahalli (954 m above Mean SeaLevel) in the direction of a NNE-SSW trending ridge lies east of the Vrishabhavathi river. The study area fallsunder the expanse of the Peninsular Gneissic Complex. The main rock types in the regions are Gneissic countryrock and as well as intrusions of Granites and Migmatites. Bangalore city lies over a hard and moderately denseGneissic basement dated back to the Archean era (2500-3500mya). A large granitic intrusion in the southcentralpart of the city extends from the Golf Course in the north central to Vasantpur VV Nagar in the south ofthe city (almost 13 km in length) and on an average 4 km from east to west along the way. A magmatiteintrusion formed within the granitic one extends for approximately 7.3 kms running parallel with KrishnaRajendra Road/ Kanakpura Road from Puttanna Chetty Road inChamrajpet till Bikaspura Road in the south. A2.25km Quatrzite formation is found in Jalahalli East (see Figure 7.2).Dike swarms are seen around the western outskirts of the city (west of the Outer Ring Road) majority strikingapproximately oriented on N15 o E. However random east west trending ones are also seen. They appear to strikeparallel to the strike of the vertical foliation of the country rock. These basic intrusions dated back to the close ofthe Archean era (Lower Proterozoic; 1600-2500 mya) mainly constitute of hard massive rocks such as Gabbro,Dolerite, Norite and Pyroxenite.Bangalore city is subjected to a moderate annual soil erosion rate of 10 Mg/ha. The basic geomorphology of thecity comprises of a central Denudational Plateau and Pediment (towards the west) with flat valleys that areformed by the present drainage patterns. The central Denudational Plateau is almost void of any topology andthe erosion and transportation of sediments carried out by the drainage network gives rise to the lateritic clayeyalluvium seen throughout the central area of the city. The pediment/pediplain is a low relief area that abruptlyjoins the plateau.Rock depth/ soil thickness (RD/ST)Another important theme is the overburden thickness of soil; it can be represented as rock depth or soilthickness. The overburden thickness of Bangalore is estimated using drilled boreholes information at selected170 locations from 850 borehole data. The overburden thickness varies from 1 m to 33 m in the study area and it145

Integration of hazard maps on GIS PlatformFigure 7.2: Geology and Geomorphology of the study area146

Integration of hazard maps on GIS Platformis shown in Figure 7.3. The south central part and north eastern part has largest overburden thickness whencompared to other areas. An average, Bangalore has the overburden thickness of less than 5m on western sideand about 15m in rest of the places. These overburden thickness obtained from boreholes doesnot representthickness of soil from the true engineering rock (shear wave velocity is 700m/s) level. They correspond tothickness of overburden above the weathered rock. Hence the overburden thickness of Bangalore is representedin the form of engineering rock level using MASW results. Figure 7.4 shows the overburdenthickness/engineering rock depth using MASW. Engineering rock depth varies from 5 m to 55 m, engineeringrock are of shallow depth on the western part, moderate depth on the eastern part and deeper depth on the southeast to north east. An average engineering rock depth is found to be about 15m from the original ground level.Soil type and strength (SS)Seismic response and liquefaction analyses require the mechanical and geometrical parameters of theoverburden soil above the engineering rock depth. Mechanical and geometrical parameters is nothing butproperties of overburden soil, which is usually represented in term of geology, average shear wave velocity ofFigure 7.3: Soil thickness using borehole datathe medium and standard penetration test “N” values. The site/area can be classified/ characterized for siteresponse or liquefaction behavior based on the above three parameters. During site characterization, it isnecessary to determine the variations in soil stratification and engineering properties of soil and rock layersencountered at the site preferably based on in-situ tests and laboratory tests conducted on the samples obtainedduring the soil exploration.147

Integration of hazard maps on GIS PlatformInitially the site classification is measured based on characteristics of geologic units taking into considerationthe possible variations in each unit. But in this method the deviations from the mean values obtained for eachgeological unit may exceed the permissible limits; hence it is needed to justify the geological use for assessingthe effects of local soil conditions (Ansal, 2004).Seismic microzonation study at Silivri, Turkey demonstratedthat the existing geological units are not homogenous and significant changes in their properties could beobserved from one point to another, even in the same formation. Therefore, considering the geological units asthe only criteria in seismic microzonation is not appropriate (Ansal, 2004). But the geology map may beregarded as the basic information to plan the detailed site investigations and to control the reliability of theresults obtained by site characterizations and site response analyses.Later, Wills and Silva (1998) suggested that average shear wave velocity in the upper 30 m can be used as oneparameter to characterise the geological units. Also they have encountered significant variations in theequivalent shear wave velocities especially in the case of alluvium deposits. Finally they recommended that theshear wave velocity is for classifying site conditions rather than as geological units, even though thedetermination of shear wave velocities requires extensive field investigations. Equivalent (average) shear wavevelocity became popular criterion for site characterization. Equivalent (average) shear wave velocity is definedas the weighted average of shear wave velocities of soil and rock layers in the top 30 meters. Equivalent shearwave velocities are being used in earthquake codes for the purpose of evaluating the design earthquakeFigure 7.4: Engineering rock depth using MASWcharacteristics on the ground surface (Borchert, 1994). In addition, it is also possible to use empiricalrelationships to estimate spectral amplifications based on equivalent shear wave velocity. Equivalent shear wavevelocity can be calculated by conducting in-situ seismic wave velocity measurements or by using correlationsdeveloped in terms of SPT-standard penetration or CPT-cone penetration tests. Equivalent (average) shear wavevelocity of the study area is calculated based on in-situ measured shear wave velocity using MASW (details arepresented in Chapter 6). Figure 7.5 shows the 30m average shear wave velocity of Bangalore.148

Integration of hazard maps on GIS PlatformFigure 7.5: 30m Average shear wave velocityBut this map (Figure 7.5) does not show the average shear wave velocity of soil because of the wide variation inthe soil overburden/ rock level. Vs 30 is usually followed where overburden is considerably high (more than100m), recently average shear wave velocity above engineering rock (Vs>700m/s) depth has also beenconsidered for classification. In Bangalore the hard rock (having Vs more than700m/s) is found with in 30m,hence the site classification is attempted considering soil overburden velocity in addition to Vs 30 . It is found thatthe average of 30m velocity shows the major part of the BMP area can be classified as “site class D”, and “siteclass C” and a smaller part in and around Lalbagh Park is classified as “site class B”. The average shear wavevelocity of soil has been calculated based on engineering rock depth obtained from MASW testing. The averageshear wave velocity for soil overburden in the study area is shown in Figure 7.6. Figure 7.6 shows that wholestudy area has a medium to dense soil with an average velocity range of 180m/s to 360m/s falling in to “siteclass D” as per classification chart. These two average shear wave velocities maps are considered as separatethemes for GIS integration.Drainage pattern (DP) and Elevation Level (EL)The geotechnical attributes presented above are based on large number of geotechnical data and experiments.But the geological and geomorphologic information is presented as one map (Figure 7.2) based on availableinformation. This map does not have sufficient information and account for other factors such as impedancecontrast, 3-dimensional basin and topographical effects. Hence to account the above parameters, the otherimportant parameters of drainage pattern (DP) and elevation level (EL) are considered as separate themes basedon the recent available information. Figure 7.7 shows the drainage pattern of study area with water bodies andFigure 7.8 shows the elevation levels in the form of contours at 10m intervals.149

Integration of hazard maps on GIS Platform7.3.2 Seismological AttributesThe seismological thematic maps have been generated based on detailed studies of seismic hazard analysis, siteresponse studies and liquefaction analysis. From these studies different earthquake hazard parameters aremapped. But for final Index map preparation and GIS integration only selected maps are considered as themes:Peak ground acceleration (PGA) at rock level based on synthetic ground motions from MCE based onDSHA.PGA at rock level at 10 % probability in 50 years exceedance based on PSHA.Amplification factor based on ground response analysis using SHAKE2000.Predominant frequency based on site response and experimental studies.Factor of safety against Liquefaction potentialPGA map for MCE from DSHAFrom detailed deterministic seismic hazard analysis presented in chapter 3, maximum credible earthquake for Bangalore isMw of 5.1. A synthetic ground motion model hasFigure 7.6: Soil Overburden average shear wave velocity150

Integration of hazard maps on GIS PlatformFigure 7.7: Tanks and drainage Features in the study areabeen generated for MCE considering the Mandya-Channapatna-Bangalore lineament (L15) as the source. Thesynthetic ground motions are generated at rock level at 653 borehole locations using rock depth informationobtained from geotechnical data. The peak ground acceleration at each borehole locations obtained fromsynthetic ground motions. PGA at rock level from ground motions using DSHA is as shown in Figure 7.9.PGA map from PSHA.A detailed probabilistic seismic hazard analysis has been carried out using six seismogenic sources identified inDSHA (see Chapter 3). PSHA analyses have been carried out using a MATLAB program which has beendeveloped for this purpose. The hazard curves and UHRS 10% probability exceedance in 50 years are calculatedfor about 1400 grid points in the study area having the size of 0.5 km× 0.5 km. Further to define the seismichazard at rock level for the study area, PGA at each grid point has been estimated. These values are used toprepare PGA maps for 10% probability exceedance in 50 years, which corresponds to return periods of 475years. Rock level PGA map for Bangalore is shown in Figure 7.10. Figure 7.10 shows that the PGA values varyfrom 0.17g to 0.25g, which is considered as theme in GIS for probabilistic hazard index map.Amplification factorThe basic intention of the site response analysis is to estimate the effect of local site conditions in assessing thesite amplification with respect to ground shaking. Amplification factor is used as theme to represent the groundbehavior during the earthquake. In this study the term “Amplification Factor” is referred to the ratio of the peakhorizontal acceleration at the ground surface to the peak horizontal acceleration at the bedrock. This factor is151

Integration of hazard maps on GIS PlatformFigure 7.8: Terrain slope based on the elevation contourevaluated for all the boreholes using the PGA at bedrock obtained from the synthetic acceleration time historyfor each borehole and the peak ground surface acceleration obtained as a result of ground response analysisusing SHAKE 2000. Figure 7.11 shows the theme of amplification factor for GIS integration.Predominant FrequencyEven though amplification is a major concern in the ground response analysis, building failure depends onperiod of ground motion at which the resonance can occur (i.e predominant period/frequency of soil column).Hence it is necessary to add a predominant period/frequency of soil column as a theme to represent hazard indexof study area. From detailed study presented in Chapter 5, it is very clear that most of the study area haspredominant frequency ranges from 3 Hz to 12 Hz. Figure 7.12 shows the theme of predominant frequency forthe study area.Factor of safety against LiquefactionFactor of safety against liquefaction (FS) is added as a theme to represent the liquefaction behavior of soil in thestudy area. At about 620 borehole locations, FS has been calculated using amplified PGA values at the groundlevel considering simplified procedure by Seed and Idriss, (1971) with subsequent revisions after Seed et al.,(1983, 1985); Youd et al., (2001); Cetin et al., (2004). More details are presented in Chapter 6. Figure 9.13shows the theme of factor of safety against liquefaction.152

Integration of hazard maps on GIS PlatformFigure 7.9: Peak ground acceleration using DSHA7.4 INTEGRATION OF DIFFERENT LAYERS (THEMES)For seismic microzonation and hazard delineation the different themes as presented above, considering bothgeomorphological and seismological are integrated to generate seismic microzonation maps. The finalmicrozonation maps can be represented in three forms, 1) hazard map, 2) vulnerability map, and 3) risk map.Because earthquake loss not only depends on the hazard caused by earthquakes, but also on exposure (socialwealth) and its vulnerability. Usually hazard map gives the hazard index (HI) based on hazard calculation andsite conditions. Vulnerability map gives us the expected degree of losses within a defined area resulting from theoccurrence of earthquakes and often expressed on a scale from 0 (no damage) to 1 (full damage). Vulnerabilitystudy includes all the exposure such as man-made facilities that may be impacted in an earthquake. It includesall residential, commercial, and industrial buildings, schools, hospitals, roads and railroads, bridges, pipelines,power plants, communication systems, and so on. Risk map will be combination of hazard classes andvulnerability classes an output risk classes. At present only hazard maps have been prepared and presented forBMP area.Hazard index is the integrated factor, depends on weights and ranks of the seismological and geomorphologicalthemes. Theme weight can be assigned based on their contribution to the seismic hazard. Rank can be assignedwith in theme based on their values closer to hazards. Usually higher rank will be assigned to values, which ismore hazards in nature, for example larger PGA will have the higher rank. The contributing themes and theirweights are listed below in Table 7.1.153

Integration of hazard maps on GIS PlatformFigure 7.10: Peak ground acceleration at 10% probability in 50 years exceedanceTable 7.1: Themes and its weights for GIS integrationIndex Themes WeightsPGA Rock level PGA using DSHA-DPGA 9Rock level PGA using PSHA-PPGA 9AF Amplification factor 8ST Soil Thickness using MASW 7Soil Thickness using borehole 7SS Equivalent Shear wave velocity for Soil 6Equivalent Shear wave velocity for 30 depth 6FS Factor of safety against liquefaction 5PF Predominant period / frequency 4EL Elevation levels 3DR Drainage pattern 2GG Geology and geomorphology 1154

Integration of hazard maps on GIS PlatformFigure 7.11: Amplification factor mapOnce the identical weights are assigned then normalized weights can be calculated based on the pair-wisecomparison matrix. Some of the attributes (like PGA and Vs) has two values for the same theme, hence both aregiven same weights with different percentage. The normalized weights are calculated using Saaty's AnalyticalHierarchy Process (Nath, 2004). In this method, a matrix of pair-wise comparisons (ratio) between the factors isbuilt, which is used to derive the individual normalized weights of each factor. The pair-wise comparison isperformed by calculating the principal Eigen vector of the matrix and the elements of the matrix are in the rangeof 0 to 1 summing to '1' in each column. The weights for each theme can be calculated by averaging the valuesin each row of the matrix. These weights will also sum to '1' and can be used in deriving the weighted sum ofrating or scores of each region of cells or polygon of the mapped layers. Since the values within each thematicmap/layer vary significantly, those are classified into various ranges or types known as the features of a layer.Table 9.2 shows the pair-wise comparison matrix of themes and the calculated of normalized weights.With in individual theme a grouping has been made according to their values. Then rank is assigned based onthe values. Usually these ranks varies from 1 to 10, highest rank is assigned for values more hazard in nature.These rank are normalized to 0 -1 using the equation 7.1. The assigned ranks with normalized values are givenin Table 7.3.Based on above attributes, two types of hazard index map are generated. One is deterministic seismicmicrozonation map (DSM), which is basically deterministic hazard index map using PGA from deterministicapproach and other themes. Another map is the probabilistic seismic microzonation map (PSM). Probabilistichazard index are calculated similar to DSM but PGA is obtained from probabilistic seismic hazard analysis.155

Integration of hazard maps on GIS PlatformFigure 7.12: Predominant Frequency mapTable 7.2: Pair-wise comparison matrix of Themes and their normalized weightsTheme PGA AF ST Vs FS PF EL DR GG WeightsPGA 1 9/8 9/7 9/6 9/5 9/4 9/3 9/2 9/1 0.200AF 8/9 1 8/7 8/6 8/5 8/4 8/3 8/2 8/1 0.178ST 7/9 7/8 1 7/6 7/5 7/4 7/3 7/2 7/1 0.156Vs 6/9 6/8 6/7 1 6/5 6/4 63 6/2 6/1 0.133FS 5/9 5/8 5/7 5/6 1 5/4 5/3 5/2 5/1 0.111PF 4/9 4/8 4/7 4/6 4/5 1 4/3 4/2 4/1 0.089EL 3/9 3/8 3/7 3/6 3/5 3/4 1 3/2 3/1 0.067DR 2/9 2/8 2/7 2/6 2/5 2/4 2/3 1 2/1 0.044GG 1/9 1/8 1/7 1/6 1/5 1/4 1/3 1/2 1 0.022156

Integration of hazard maps on GIS PlatformDeterministic seismic microzonation mapFigure 7.13: Factor safety against liquefactionDeterministic seismic microzonation map is hazard index map for worst scenario earthquake. Important factorof PGA (weight is 9) is estimated from synthetic ground motions, which are generated based on MCE of 5.1 inmoment magnitude for closest vulnerable source of Mandya- Channapatna- Bangalore lineament. Hazard indexvalues are estimated based on normalized weights and ranks through the integration of all themes using thefollowing equation:⎛DPGAWDPGAr+ AFWAFr+ STWSTr+ SSWSSr+ ⎞DSM = ⎜⎟ / ∑ W7.2⎝ FSWFSr+ PFWPFr+ ELWELr+ DRwDRr+ GGwGGr⎠Using estimated values deterministic seismic microzonation map has been generated. Figure 7.14 shows thedeterministic seismic microzonation map for Bangalore. Integrated GIS map shows that hazard index valuesvary from 0.10 to 0.66. These values are groped into four groups,

Integration of hazard maps on GIS PlatformTable 7.3: Normalized ranks of the themesThemes Values Weight Ranks Normalized Ranks44 11-5 1 05-10 2 0.25ST (m)10-15 0.1563 0.515-20 4 0.7520-255 1

Integration of hazard maps on GIS Platformattached to the seismic hazard index greater than 0.6 at south western part of Bangalore. Lower part (south) ofBangalore is identified as moderate to maximum hazard when compare to the northern part.Figure 7.14: Deterministic seismic microzonation map159

Integration of hazard maps on GIS PlatformFigure 7.15: Probabilistic seismic microzonation map7.5 SUMMARYIn this chapter, the final hazard maps preparation using GIS based Fuzzy logic sets aided by the AnalyticHierarchy Process (AHP) is described. Two different seismic hazard maps have been generated, one usingdeterministic seismic microzonation map based on PGA from DSHA and another is the probabilistic seismicmicrozonation map based on PGA from PSHA. Maximum hazard covered by DSM is larger when compared toPSM. In DSM, western part of city is having maximum hazard and in PSM, southern part of city is havingmaximum hazard. Maximum hazard at western part of city in DSM may be attributed to the location of seismicsource (Mandya- Channapatna- Bangalore lineament) and larger PGA in that area. PSM shows that themaximum hazard is at south western part, because the maximum number of seismogenenic sources is located inthat direction. These seismic microzonation hazard maps contain important information for the city and regionalplanning, considering different earthquake hazards. However for important structures a detailed site specificstudy need to be performed at each site during the design stage to evaluate the local site conditions. On the otherhand, site specific studies, including in-situ and laboratory tests, must be obligatory in the assessment of therequired parameters for the structures with higher importance levels.160

CHAPTER 8SUMMARY AND CONCLUSIONS8.1 SUMMARYIn the present study an attempt has been made to investigate the seismic hazard analysis considering thelocal site effects and to develop microzonation maps for Bangalore. Seismic hazard analysis and microzonationof Bangalore is addressed in this study in three parts: In the first part, estimation of seismic hazard usingseismotectonic and geological information. All the earthquake sources and seismicity has been consideredwithin a radius of 350 km from the Bangalore city for the study. Second part deals about site characterizationusing geotechnical and shallow geophysical techniques. An area of 220 sq.km, encompassing BangaloreMunicipal Corporation has been chosen as the study area in this part of the investigation. There were over 150lakes, though most of them are dried up due to erosion and encroachments leaving only 64, present in an area of220 sq km. emphasizing the need to study local site effects. In the last part, local site effects are assessed bycarrying out one-dimensional (1-D) ground response analysis (using the program SHAKE 2000) using bothborehole SPT data and shear wave velocity survey data within an area of 220 sq.km. Further, field experimentsusing microtremor studies have also been presented (jointly carried out with NGRI) for evaluation ofpredominant frequency of the soil columns. The same has been assessed using 1-D ground response analysis andcompared with microtremor results. Further, Seed and Idriss simplified approach has been adopted to evaluatethe liquefaction susceptibility and liquefaction resistance assessment (Idriss and Boulanger, 2005).Microzonation maps have been prepared for Bangalore city covering 220 sq. km area on a scale of 1:20000. Thefollowing sections list the major conclusions drawn in each chapter:8.2 DETERMINISTIC SEISMIC HAZARD ANALYSIS1. Deterministic Seismic Hazard Analysis (DSHA) for Bangalore, India has been carried out byconsidering the past earthquakes, assumed subsurface fault rupture lengths and point source syntheticground motion model by considering the regional seismotectonic activity in circular area with about350 km radius around Bangalore city centre.2. The seismotectonic map has been prepared by considering the faults, lineaments, shear zones in thearea and past earthquake events.3. The Peak Ground Acceleration (PGA) is calculated for the different sources considering pastearthquake events expressed in moment magnitude and a regional attenuation relation for peninsularIndia. The maximum PGA at rock level estimated from this method is about 0.146g.4. Based on Wells and Coppersmith (1994) relationship, subsurface fault rupture length of about 3.8% oftotal length of the fault shown to be matching with past earthquake events in the area. The PGA valueof 0.146g at rock level is obtained for subsurface length of 3.8% of total length of fault.5. To simulate synthetic ground motions, Boore (1983, 2003) SMSIM programs have been used and thePGA for different sources is evaluated. The maximum PGA of 0.136g is obtained for Bangalore.6. From this study, it is very clear that Bangalore area can be described as seismically moderate activeregion. It is also recommended that southern part of Karnataka in particular Bangalore, Mandya andKolar, need to be upgraded from current Indian Seismic Zone II to Seismic Zone III.7. The rock level PGA map is generated for Bangalore using synthetic ground motion generated. Thismap is useful for the purpose of seismic microzonation, ground response analysis and design ofimportant structures.8.3 PROBABILISTIC SEISMIC HAZARD ANALYSIS1. Seismic data completeness has been studied using the seismic data collected. The seismic data ishomogenous for the last four decades irrespective of the magnitude.2. Seismic parameters were then evaluated using completed and incomplete data using Gutenberg-Richterrecurrence relation. The ‘b’ value of 0.89 is obtained for complete part of data. For 200 years data ‘b’value of 0.87 is obtained for the magnitude interval of 3.5 to 6.2 and 0.92 is obtained for the magnitudeinterval of 4.0 to 6.2.161

Summary and Conclusions3. Computer programme HN2, Release 2.10, 2005 developed by Kijko and Sellevoll (1989, 1992) hasbeen used to assess seismic hazard parameters considering the magnitude uncertainty. A similar ‘b’value of 0.87±0.03 and maximum expected magnitude is (M max ) 6.0± 0.54 is obtained for the region.4. The ‘b’ values determined in this study are comparable with the earlier reported seismic parameters forsouth India. These ‘b’ values are slightly higher than previous studies indicating neotectonic activity ofthe region.5. The probabilities of distance, magnitude and peak ground acceleration have been evaluated for the sixmost vulnerable sources. The mean annual rate of exceedance has been calculated for all the six sourcesat the rock level. The cumulative probabilistic hazard curve has been generated at the bedrock level forpeak ground acceleration and spectral acceleration.6. The spectral acceleration calculation corresponding to a period of 1sec and 5% damping are alsoevaluated. For the design of structures, uniform hazard response spectrum (UHRS) at rock level isdeveloped for the 5% damping corresponding to 10% probability of exceedance in 50 years.7. The peak ground acceleration (PGA) values corresponding to 10% probability of exceedance in 50years are comparable to the PGA values obtained in deterministic seismic hazard analysis (DSHA) forthe same area and are higher than Global Seismic Hazard Assessment Program (GSHAP) maps ofBhatia et.al (1997) for the Indian shield area.8. The quantified hazard values in terms of the rock level peak ground acceleration (PGA) are mapped for2% and 10 % probability of exceedance in 50 years on a grid size of 0.5 km 0.5 km.8.4 SITE CHARACTERIZATION USING GEOTECHNICAL BOREHOLE DATABase map of Bangalore on a scale of 1:20000 with about 12 layers of information has been prepared andpresented.1. Geotechnical data for 850 boreholes upto a depth of about 40m has been collated and attached to thelocations in the map using ARC View GIS 8.1 with 3D analyst package.2. A 3-D subsurface geotechnical model has been generated based on large amount of geotechnicalborehole data and Geographical Information System on a scale of 1:20000 for the Bangalore city.3. Reduced levels of rock depth at unknown locations are predicted using Artificial Neural Network(ANN) model. The soil overburden thicknesses/ rock depths have been assessed and a map of the sameis presented.4. For field SPT ‘N’ values corrections are applied by using the necessary corrections as recommended byNCEER-1998.8.5 SITE CHARACTERIZATION USING MULTICHANNEL ANALYSIS OFSURFACE WAVE (MASW) SURVEY1. MASW one-dimensional survey at 58 locations and two-dimensional surveys at 20 locations have beencarried out with in 220sq km area of Bangalore.2. The shear wave velocity profiles (Vs versus depth-1-dimensional), spatial variability of shear wavevelocity (Vs versus depth and length, 2- dimensional) are evaluated and presented.3. The average shear wave velocity of the study area has been estimated for 5m, 10m, 15m, 20m, 25m and30m depth and presented.4. Average shear wave velocity for the soil depth, which is estimated based on overburden thickness fromSPT data is presented.5. Site soil classification has been carried out based on the NEHRP and IBC using the estimated averageshear wave velocity for depth of 30m (Vs 30 ) and soil overburden thickness. Bangalore soil is classifiedas “Site class D”.6. Among total 58 locations of MASW survey carried out, 34 locations were very close to the SPTborehole locations and these are used to generate correlation between Vs and corrected “N” values.The equation between Vs and (N 1 ) 60cs with upper and lower bound values are presented. This regressionequation can be effectively used to find out the shear modulus for ground shaking response studies atsimilar soil sites in Peninsular India.7. The correlated relation lies between the JRA equations for sandy and clay equations for wide range of“(N 1 ) 60cs ” values, because the soil overburden in Bangalore has sand and silt with some percentage ofclay content.162

Summary and Conclusions8.6 LOCAL SITE EFFECTS AND SITE RESPONSE1. The amplification potential at various locations in Bangalore has been estimated using theoretical onedimensionalground response equivalent linear model and experimental study based on microtremor.2. The synthesized bedrock motion and 170 borehole locations soil profile information were consideredas input and site specific ground response analysis was carried out using SHAKE 2000.3. Peak horizontal acceleration at each layer, the acceleration time history at the ground surface,amplification spectrum and response spectrum are obtained.4. Amplification factor is obtained by dividing the peak horizontal acceleration at ground by peakhorizontal acceleration at rock. The amplification varies from 1 to 4.7, based on this Bangalore isdivided as four zones. The regions in zones I and II are seismically more stable than the regions inzones III and IV. Most of the area in Bangalore lies in zone II.5. The ground response analysis shows that the natural periods of the analyzed deposits are in between0.01s and 0.45s with about 85% of the locations having a period below 0.2s.6. The peak spectral acceleration varies from 0.2g to 2.1g and frequency corresponding PSA varies from3Hz to 20Hz.7. The response spectra for 5% damping at the ground surface obtained for all the borehole locationsclearly indicate that the range of spectral acceleration at different frequencies varied over a wide range.At 5% damping, the range of SA at 1.5 Hz frequency was 0.01 g to 0.07 g, at 3 Hz frequency it was inthe range of 0.03 g to 0.65 g, while at 5 Hz frequency it was in the range of 0.08 g to 1.14g.8. Measured shear wave velocity at 58 locations using MASW also used for site response study usingSHAKE2000. Ground response parameters obtained are comparable with the response parametersobtained using SPT data.9. The ground repose parameters obtained using MASW data is slightly lower than the results obtainedusing SPT data.10. Predominant frequencies at each borehole locations are estimated using both SPT data and MASWdata. Predominant frequencies varies from 4Hz to 12Hz using SPT data and 3.45Hz to 12Hz usingMASW data11. From 38 locations about 190 data pairs of Vs and SPT corrected blow count have been used for theregression analysis. Correlation between corrected SPT ‘N’ values and low strain shear modulus withupper and lower bound values have also been developed and presented.12. Ambient noise survey has been carried out using duration of recording minimum of 3 hours and amaximum of 26 hrs at 54 locations to evaluate horizontal to vertical spectral ratio and predominantfrequency of soil column.13. Predominant frequency determined using microtremor study ranges from between 1.2 Hz -11 Hz.14. Generally, Bangalore soil has predominant frequencies in the range from 3 Hz to 12Hz.15. Spectral acceleration hazard curve and Uniform Hazard Response Spectrum for “Site Class D” with10% probability of exceedance in 50 years (5% damping) for Bangalore are presented.16. A value of PGA (ZPA = PGA) 0.33g is obtained when local site effect is taken into account.Amplification factor of 2.73 times higher than rock level PGA (ZPA = PGA) 0.121g has been obtained.17. The amplification factor obtained using SHAKE2000 and probabilistic approaches with local siteeffects are comparable.8.7 LIQUEFACTION HAZARD ASSESSMENT1. Corrected SPT data and amplified peak acceleration at 620 locations are used for evaluation ofliquefaction hazard in BMP area of 220 sq.km for Bangalore.2. Liquefaction hazard has been calculated based on factor of safety against liquefaction using Seed andIdriss (1971) simplified approach.3. A simple spreadsheet was developed to carryout all the calculation for each bore log. Arrived factor ofsafety against liquefaction is grouped together and classified the Bangalore city in to four groups forliquefaction hazard mapping. Factor of safety against liquefaction are presented in the form of 2-dimensional maps.4. About 85% of Bangalore area has high factor of safety and they are non-liquefiable.5. Bangalore city is safe against liquefaction except at few locations where shallow water table is metalong with sandy silt and filled up soils have been found.6. From this study it is found that at many locations (having FS>3) the liquid limit is more than 32 andplasticity index is more than 12. At these locations a detailed study based on laboratory cyclic triaxialtests is suggested to evaluate strength loss.163

Summary and Conclusions8.8 INTEGRATION OF HAZARD MAPS ON GIS PLATFORM1. Seismic microzonation maps are prepared using GIS based Fuzzy logic sets aided by the AnalyticHierarchy Process (AHP).2. Two different seismic hazard maps have been generated, one using deterministic seismicmicrozonation map based on PGA from DSHA and another is the probabilistic seismic microzonationmap based on PGA from PSHA.3. Maximum hazard covered by deterministic seismic microzonation map is larger when compared toprobabilistic seismic microzonation map.4. Maximum hazard is at western part of city in deterministic seismic microzonation and it may beattributed to the location of seismic source (Mandya- Channapatna- Bangalore lineament) and largerPGA in that area.5. Probabilistic seismic microzonation shows that the maximum hazard is at south western part, becausemaximum number of the seismogenic sources is located in that direction.8.9 CONCLUSIONSThis study shows that, expected peak ground acceleration (PGA) at rock level for Bangalore is about 0.15gusing DSHA (0.136g using synthetic ground motion model). Seismic parameter ‘b’ value is estimated as 0.87,which is slightly higher than the published values which may be due to increase in seismotectonic activity of theregion. PSHA used to quantify the uncertainty involved in the hazard analysis, which also gives similar peakground acceleration of 0.121g. Generally PSHA estimates a lower PGA values compared to that of PGA valuesobtained from DSHA. Mean annual rate of exceedance for particular acceleration is obtained for both PGA andspectral accelerations. Uniform hazard spectra at rock level and return period have been evaluated. Sitecharacterization using SPT data has been carried out and 3-D subsurface model has been generated using GIS.Field SPT ‘N’ values are corrected by applying necessary corrections for further use in engineering applications.Site characterization also carried out using measured shear wave velocity using MASW and average shear wavevelocity at each 5m interval up to a depth of 30m was evaluated and presented. Based on soil average shearwave velocity and 30m average shear wave velocity, as per NEHRP and IBC, Bangalore can be classified as“Site class D”. Correlation between corrected SPT ‘N’ values and measured shear wave velocity has beendeveloped and it is represented by the following equation:Vs = (10.1)0.4078[(N1 )60cs]Theoretical 1-D site response study shows that the amplification factor is in the range of 1 to 4.7 andpredominant frequency varies from 2 Hz to 12Hz. The results of site response studies using SPT data andMASW data are comparable. Ground response parameters evaluated using MASW data are slightly lowervalues when compared to the parameters obtained using SPT data. Correlation between corrected SPT ‘N’values and low strain shear modulus has been developed and is given by:G [( ) ] 0. 68max= 16.82 N1(10.2)60csField study microtremor survey also shows similar values of predominant frequencies for these sites.Predominant frequency obtained from all the three methods matches very well. Liquefaction study shows thatBangalore is safe against liquefaction except at few locations where the overburden is sandy silt with presenceof shallow water table.In this study seismic hazard analysis by deterministic and probabilistic approach, site characterization using SPTand MASW and site response and local site effects studies are carried out for microzonation of Bangalore.Probabilistic hazard curves for peak ground acceleration and spectral acceleration and uniform hazard responsespectrum (UHRS) at rock level and ground level is developed for the 5% damping corresponding to 10%probability of exceedance in 50 years. Two types of seismic microzonation map are generated. One isdeterministic seismic microzonation map (DSM), which is basically deterministic hazard index map using PGAfrom deterministic approach and other themes. Another map is the probabilistic seismic microzonation map(PSM). Probabilistic hazard index are calculated similar to DSM but PGA is obtained from probabilistic seismichazard analysis.164

Summary and Conclusions8.10 RECOMMENDATIONS FOR FURTHER STUDY1. This study was carried out for Bangalore in an area covering Bangalore Mahanagar Palike (220km 2 ).This study can be extended to greater Bangalore (696km 2 ) area.2. A site specific attenuation relation needs to be developed for the study area.3. The developed maps can be also combined with appropriate weighting factor for site response,liquefaction, soil overburden thickness, etc, and importance of the same. A single hazard map based onprobability parameters zonation can be developed.4. It is recommended to carryout Risk and vulnerability studies for Bangalore and prepare Risk map inaddition to hazard maps.165

REFERENCESAki, K. (1969), “Analysis of the seismic coda of local earthquakes as scattered waves” J. Geophys. Res., 74, pp 615-631.Aki, K., and Richards, P.G. (1980), “Quantitative seismology”, W.H. Freeman, San, Vol. 1.Algermissen, S. T. and Perkins, D. M. (1976), “A probabilistic estimate of maximum acceleration in rock in thecontiguous United States,” USGS Open File Report 76-416, 45pp.Al-Hunaidi, M.O. (1992), “Difficulties with phase spectrum unwrapping in spectral analysis of surface waves nondestructivetesting of pavements,” Can. Geotech. J. 29, pp 506–511.Ambraseys, N. N. (1988), ‘‘Engineering seismology’’ Earthquake Engrg. and Struct. Dynamics, Vol. 17, pp 1–105.Anbazhagan, P. (2004), “Microzonation for Amplification, attenuation and liquefaction of Chennai city”, M.E.Thesis, Anna University, Chennai.Andrus, R. D., and Stokoe, K. H., II (1997). ‘‘Liquefaction resistance based on shear wave velocity.’’ Proc.,NCEER Workshop on Evaluation of Liquefaction Resistance of Soils, Nat. Ctr. for Earthquake Engrg. Res., StateUniv. of New York at Buffalo, 89–128.Ansal, A (ed) (2004), “Recent Advances in Earthquake Geotechnical Engineering and Microzonation,” KluwerAcademic Publishers. Printed in the Netherlands.Arango, I. (1996), ‘‘Magnitude scaling factors for soil liquefaction evaluations.’’ J. Geotech. Engrg., ASCE,Vol.122(11), pp 929–936.Arora, S. K. (1971),” A study of the Earth's crust near Gauribidanur in southern India”, Bull. Seism. Soc. Am., Vol.61, pp 671 –683.Atkinson, G. M. and Boore, D. M. (1995), “Ground-motion Relations for Eastern North America”, Bulletin of theSeismological Society of America, Vol. 85(1), pp 17-30.Baise,L.G., Glaser,S.D., and Dreger,D. (2003), “Site Response at Treasure and Yerba Buena Islands, California”Journal of Geotechnical and Geoenvironmental Engineering, Vol. 129, No. 6, pp 415–426.Bansal, B.K. and Gupta, S. (1998), “A Glance through the Seismicity of Peninsular India”, Jour. Geol. Soc. India,Vol..52, pp 67-80.Bell, F.G., Cripps, J.C., Culshaw, M.G. and O'Hara, M. (1987) “Aspects of geology in planning. In: M.G.Culshaw, F.G. Bell, J.C. Cripps and M. O'Hara, Editors, Planning and Engineering Geology,” Geological SocietyEngineering Geology Special Publication no. 4 , PP. 1–38.Bhandari, R.K., Rajarathnam, S., Janardhanan,G. and Ganapathy,G.P. (2003), “Seismic Microzonation” ATraining progarmme guide, published by Centre for Disaster Mitigation and Management, Anna University,Chennai, India, P-140.Bhatia, S. C., Ravi Kumar, M., and Gupta, H.K. (1997),” A Probabilistic Seismic Hazard Map of India andAdjoining Regions”, http://www.seismo.ethz.ch/gshap/ict/india. html, pp 1-12.Bilham, R. (2004), “Earthquakes in India and the Himalaya”, tectonics, geodesy and history, Annals of Geophysics,Vol. 47(2), pp 839-858.Bindi, D., Parolai,S., Spallarossa,D. and Cattaneo,M. (2000), “Site effects by H/V ratio: comparison of twodifferent procedures”, J. Earth. Eng., Vol. 4, pp 97-113.Boore, D. M. (1983), “Stochastic simulation of high-frequency ground motions based on seismological models ofthe radiated spectra”, Bull. Seism. Soc. Am., Vol. 73, 1865-1894.Boore, D. M. (2003), “Simulation of ground motion using the stochastic method”, Pure and Applied Geophysics,Vol. 160, pp 635-675.166

ReferencesBoore, D. M., Joyner, W. B., and Fumal, T. E. (1993), “Estimation of response spectra and peak accelerationsfrom western North American earthquakes,” An interim report, United States Geological Survey Open- File Report93-509, Menlo Park, California, 72 pp.Borcherdt, R. D., and Glassmoyer, G., (1992), “On the characteristics of local geology and their influence onground motions generated by the Loma Prieta earthquake in the San Francisco Bay region, California”, Bull. Seimol.Soc. Amer., v. 82, p 603-641.Borcherdt, R.D. (1994), “Estimates of site-dependent response spectra for design (methodology and justification),”Earthquake Spectra, 10(4), 617-653.Brankman, C. M., Baise, L. G., Higgins, R. B. and Dawson, K. M. (2004), “Liquefaction hazard mapping inBoston, Massachusetts: Collaborative Research with William Lettis &Associates, Inc., and Tudts University”,Published in web page, http://earthquake.usgs.gov/research/external/reports/02hqgr0040.pdfBrown, L.T., Boore, D.M. and Stokoe, K.H. II, (2002), “Comparison of shear-wave slowness profiles at 10strong-motion sites from noninvasive SASW measurements and measurements made in boreholes”, Bull. Seismol.Soc. Am., Vol. 92, pp 3116–3133.Brune, J. (1970), “Tectonic stress and the spectra of seismic shear waves from earthquakes”, J. Geophys. Res. 75,pp 4997-5009.Bullen, K.E. (1965), “An introduction to the theory of seismology”, Cambridge Univ. Press, Cambridge, U.K.Cetin, K.O, Seed, R.B., Kiureghian, A.D., Tokimatsu, K., Harder, L.F. Jr., Kayen, R.E., and Moss, R.E.S.(2004), “Standard penetration test-based probabilistic and deterministic assessment of seismic soil liquefactionpotential”, Journal of Geotechnical and Geoenvironmental Engineering, Vol. 12, pp 1314-1340.Chandra, U. (1977), “Earthquakes of Peninsular India” a seismotectonic study,” Bull. Seism. Soc. Am., Vol. 65, pp1387-1413.Chen. W.F and Scawthorn, C. (2003), “Earthquake Engineering Handbook” published by CRC Press LLC, 2000N.W. Corporate Blvd., Boca Raton, Florida 33431.Cornell, C. A., and Van Marke, E. H. (1969), “The Major Influences on Seismic Risk,” Proceedings of the ThirdWorld Conference on Earthquake Engineering, Santiago, Chile, Vol. A-1, pp 69-93.Cornell, C.A. (1968), “Engineering seismic risk analysis”, Bulletin of the Seismological Society of America , Vol.58 , pp 1583–1606.Dai, F.C., Lee C.F. and Zhang, X.H. (2001) “GIS-based geo-environmental evaluation for urban land-useplanning: a case study.” Engineering Geology, 61, pp. 257–271.Dai, F.C., Liu Y. and Wang, S. (1994) “Urban geology: a case study of Tongchuan City, Shaanxi Province,China.” Engineering Geology, 38 PP. 165–175.Day, R. W. (2002), “Geotechnical Earthquake Engineering Handbook”, McGraw-Hill, Two Penn Plaza, New York.Destegul Umut, (2004), “Sensitivity Analysis of Soil Site Response Modelling in Seismic Microzonation forLalitpur, Nepal,” M.Sc Thesis, International Institute of Geo-Information Science and Earth Observations,Enschede, Netherlands.Dong, W. M., Bao, A. B.and Shah, H. C. (1984). Use of maximum entropy principle in earthquake recurrencerelationships, Bull. Seism. Soc. Am. 74, 725-737.Dong, W. M., Shah H. C., and Bao A. B. (1984b). “Utilization of geophysical information in Bayesian seismichazard model,” Soil Dynamics and Earthquake Engineering 3, 103-111.DRM (2003); “Project Microzonation for Earthquake Risk Mitigation (MERM) - Microzonation of Pilot Areas:Adapazari, Gölcük, Ihsaniye and Degirmendere, Part 2C -CaseStudies,”http://geotechnics.ch/main.htm?http://geotechnics.ch/www/search/PublicationDisplay.asp?publication=722;publField E.H. and Jacob, K.H. (1993), “The Theoretical Response of Sedimentary Layers to Ambient Seismic Noise”,Geophys. Res. Lett. 20, pp 2925-2928.167

ReferencesFinn, W. D. Liam (1991), “Geotechnical Engineering Aspects of Microzonation”, Proceeding of the FourthInternational Conference of Seismic Zonation”, Stanford, California, Vol. I, pp.199-259.Ganesha Raj, K. (2001), “Major Lineaments of Karnataka and their significance with respect to Seismicity –Remote sensing based analysis”, Ph.D. Thesis, Gulbarga University, Karnataka, India.Ganesha Raj. K., and Nijagunappa, R. (2004),” Major Lineaments of Karnataka State and their Relation toSeismicity” Remote Sensing Based Analysis, Jr.Geol.Soc.India, Vol.63, pp.430-439.Ganji, V., Gukunski, N., Maher, A. (1997), “Detection of underground obstacles by SASW method—Numericalaspects”. J. Geotech. Geoenviron. Eng., 123(3) ASCE, pp 212– 219.González, J., Schmitz, M., Audemard, F.A., Contreras, R., Mocquet, A., Delgado, J., De Santis, F. (2004),”Site and induced effects associated with the 1997 Cariaco earthquake”, Engineering Geology, Vol. 72(1-2), pp 143-177.GovindaRaju, L., Ramana, G.V., HanumanthaRao, C., and Sitharam, T.G. (2004),”Site specific groundresponse analysis, Special section”, Geotechnics and Earthquake Hazards, Current Science, Vol. 87, NO.10, pp.1354-1362.Gutenberg, B. (1945), “Magnitude determination for deep-focus earthquakes”, Bulletin of the Seismological Societyof America, Jul, Vol. 35, pp 117 – 130.Gutenberg, B., and Richter, C. F. (1956), “Earthquake magnitude, intensity, energy, and acceleration: (Secondpaper)”, Bulletin of the Seismological Society of America, Apr , Vol. 46, pp 105 - 145.Hake, S.S. (1987) “A review of engineering geological and geotechnical aspects of town and country planning withparticular reference to minerals and the extractive processes. In: M.G. Culshaw, F.G. Bell, J.C. Cripps and M.O'Hara, Editors, Planning and Engineering Geology” Geological Society Engineering Geology Special Publicationno. 4, PP. 69–74.Heaton, Tajima, T. F., and Mori, A.W. (1986), “Estimating Ground motions using recorded Accelerograms”,Surveys in Geophysics, Vol. 8, pp 25-83.Hwang, H., and Huo, J. R. (1997),” Attenuation relations of ground motion for rock and soil sites in eastern UnitedStates”, Soil Dynamics and Earthquake Engineering, Vol. 16, pp 363-372.IBC, (2000), “International Building Code,” International Code Council: Inc. 5th Edition, Falls Church, VA.Idriss, I. M., and Boulanger, R. W. (2004), "Semiempirical procedures for evaluating liquefaction potential duringearthquakes." Proc., 11th International Conference on Soil Dynamics and Earthquake Engineering, and 3rdInternational Conference on Earthquake Geotechnical Engineering, D. Doolin et al., eds., Stallion Press, Vol. 1, pp32-56.Idriss, I. M., and Boulanger, R. W. (2005), “Evaluation of Liquefaction Potential, Consequences and Mitigation”Invited Expert Lectures, Proc. Indian Geotechnical Conference-2005, 17-19 December 2005, Ahmedabad, INDIA,pp 3-25.Imai, T., and Tonouchi, K. (1982), “Correlation of N-value with S-wave velocity and shear modulus”,Proceedings, 2nd European Symposium on Penetration Testing, Amsterdam, pp 57-72.Ivanov, J., Park, C.B., Miller, R.D., and Xia, J. (2005), “Analyzing and filtering surface-wave energy by mutingshot gathers,” Journal of Environmental and Engineering Geophysics, 10 (3), 307-321.Iyengar, R. N. and Ghosh, S. (2004), “Microzonation of earthquake hazard in greater Delhi area”, Curr. Sci., Vol.87, pp 1193–1202.Iyengar, R.N. and Raghukanth, S. T. G. (2004), “Attenuation of Strong Ground Motion in Peninsular India”,Seismological Research Letters, Vol. 75, No. 4, pp 530-540.Jaiswal, K. and Sinha, R., (2006), “Probabilistic Modeling of Earthquake Hazard in Stable Continental Shield ofthe Indian Peninsula,” ISET Journal of Earthquake Technology. Vol. 43, No, 3. pp 49-64.Japan Road Association (JRA), (1980), “Specification for highway bridges” Part V, Earthquake Resistant Design.168

ReferencesJoyner, W. B. and Fumal T. E. (1985), “Predictive mapping of earthquake ground motion, in EvaluatingEarthquake Hazards in the Los Angeles Region” – An Earth-Science Perspective, J. E. Ziony (Editor), U.S. Geol.Surv. Profess. Pap. 1360, pp 203-220.Kaila., K. L., Gaur V. K., and Hari Narain, (1972), “Quantitative seismicity maps of India”, Bull. Seism. Soc.Am. Vol. 62, pp 1119-1131.Kanamori, H. (1977), “The Energy Release in Great Earthquakes”, Journal of Geophysical Research, Vol. 82, pp2921-2987.Kanli, A.I., Tildy, P., Pronay, Z., Pinar, A. and Hemann, L. (2006), “Vs 30 mapping and soil classification forseismic site effect evaluation in Dinar region , SW Turkey”, Geophysics J. Int, Vol. 165, pp 223-235.Kayen, R. E., Mitchell, J. K., Seed, R. B., Lodge, A., Nishio, S., and Coutinho, R. (1992), ‘‘Evaluation of SPT-,CPT-, and shear wave-based methods for liquefaction potential assessment using Loma Prieta data,’’ Proc., 4thJapan-U.S. Workshop on Earthquake-Resistant Des. of Lifeline Fac. and Countermeasures for Soil Liquefaction,Vol. 1, pp 177–204.Khattri, K.N. (1992), “Seismic Hazard in Indian region,” Current science, Vol. 62 No 1&2, pp 109-116.Kijko, A. and Sellevoll, M. A. (1989), “Estimation of earthquake hazard parameters from incomplete data files.Part I Utilization of extreme and complete catalogs with different threshold magnitudes” Bull. Seismol. Soc. Am.,Vol. 79, pp 645–654.Kijko, A. and Sellevoll, M. A. (1992), “Estimation of earthquake hazard parameters from incomplete data files.Part II. Incorporation of magnitude heterogeneity,” Bulletin of the Seismological Society of America, Feb 1992,Vol.82, pp 120 - 134.Kiureghian, D. A., and Ang., A. H. S. (1977), “A fault-rupture model for seismic risk analysis“, Bull. Seism. Soc.Am., Vol. 67(4), pp 1173-1194.Knopoff. L and Kagan. Y. (1977), “Analysis of theory of external values as Applied to Earthquake Problems” J.Geophys. Res, Vol 82, pp 5647-5657.Kramer, S. L. (1996), “Geotechnical Earthquake Engineering “, Published by Pearson Education Ptd.Ltd,Reprinted 2003, Delhi, India.Krinitzsky, E. (2005),“Discussion on Problems in the application of the SSHAC probability method for assessingearthquake hazards at Swiss nuclear power plants”, Eng. Geol. Vol. 78, pp. 285–307, Engineering Geology, Vol. 82,pp 66–68.Krishna, V. G., and Ramesh, D. S. (2000), “Propagation of crustal waveguide trapped Pg and seismic velocitystructure in the south Indian shield,” Bull. Seismol. Soc. Am., 90, pp 1281–1296.Legget, R.F. (1987) “The value of geology in planning. In: M.G. Culshaw, F.G. Bell, J.C. Cripps and M. O'Hara,Editors, Planning and Engineering Geology,” Geological Society Engineering Geology Special Publication no. 4,PP. 53–58.Liao, S. S. C., and Whitman, R. V. (1986), ‘‘Catalogue of liquefaction and non-liquefaction occurrences duringearthquakes.’’ Res. Rep., Dept. of Civ. Engrg., Massachusetts Institute of Technology, Cambridge, Mass.Marcuson, W. F., III (1978), ‘‘Definition of terms related to liquefaction.’’ J. Geotech. Engrg. Div.,ASCE,Vol.104(9),pp 1197–1200.Mark, R.K. (1977), “Application of linear statistical model of earthquake magnitude versus fault length inestimating maximum expectable earthquakes”, Geology 5, pp 464– 466.Martin, G. R. Editor, (1994), “Proc. of the NCEER/SEAOC/BSSC, Workshop on Site Response duringEarthquakes and Seismic Code Provisions,” University of .Calof.Los Angeles.McGuire, R. K. (1976), “FORTRAN computer program for seismic risk analysis,” U.S. Geol. Surv., Open File Rep.No. 76-67.McGuire, R.K. (1978), “FRISK — A Computer Program for Seismic Risk Analysis,” U.S. Department of Interior,Geological Survey: Open-File Report 78-1007.169

ReferencesMcHarg (1968)Miller, R.D., Xia, J., Park, C.B., and Ivanov, J. (1999a), “Using MASW to map bedrock in Olathe, Kansas [Exp.Abs.]”, Soc. Explor. Geophys., pp 433-436.Molnar, S., Cassidy, J.F. and Dosso, S.E. (2004), “Site Response Studies in Victoria, B.C., Analysis of Mw 6.8Nisqually Earthquake Recordings and Shake Modelling”, 13th World Conference on Earthquake EngineeringVancouver, B.C., Canada, August 1-6, 2004, Paper No. 2121.Nakamura, Y. (1989), "Earthquake Alarm System for Japan Railways," Japanese Railway Engineering , Vol.28(4), pp 3-7.Nath, S. K. [2004] “Seismic Hazard mapping and Microzonation in the Sikkim Himalaya through GIS integration ofsite effects and strong motions attributes”, Journal of Natural hazards, 31, 319-342.Nazarian, S., Stokoe II, K.H., Hudson, W.R. (1983), “Use of spectral analysis of surface waves method fordetermination of moduli and thicknesses of pavement systems”, Transp. Res. Rec. 930, pp 38– 45.O’Leary, D.W., Driedman, J.D., and Pohn, H.A. (1976), “Lineaments, linear, lineation: Some proposed newstandards for old terms”, Geol. Soc. Am. Bull. 87, pp 1463-1469.Palmer, P. S, Magsion, L. S, Niggemann, A. R, Bilderback, L. E, Folger, S. D And Poelstra, L. J, (2003),“Preliminary liquefaction Susceptibility Map of Island County, Washington”,http://www.islandcounty.net/gsa/des/documents/island_liquefaction.pdf .Pankow, K.L and Pechmann, J. C. (2004), “Determination of Low-Strain Site-Amplification Factors in the SaltLake Valley, Utah, Using ANSS Data”, Basin and Range Province Seismic Hazards Summit II-2004, p-15.Park, C.B., Miller, R.D., and Miura, H. (2002), “Optimum field parameters of an MASW survey [Exp. Abs.]”,SEG-J, Tokyo, May , pp 22-23.Park, C.B., Miller, R.D., and Xia, J. (1999), “Multi-channel analysis of surface waves,” Geophysics, Vol. 64(3),pp 800-808.Park.C.B., Miller.R.D., Xia. J., and Ivanov .J. (2005a), “Multichannel Seismic Surface-Wave Methods forGeotechnical Applications,” http://www.kgs.ku.edu/ Geophysics2/Pubs/Pubs PAR-00-03.pdf.Parolai, S., Bormann, P and Milkert, C. (2001), “Assessment of the natural frequency of the sedimentary cover inthe Cologne area (Germany) using noise measurements,” Journal of Earthquake Engineering , Vol. 5, pp 541–564.Parvez, A.I., Franco Vaccari and Panza October, G. F. (2003),” A deterministic seismic hazard map of India andadjacent areas”, Geophysical Journal International, Vol. 155, pp 489–508.Pearce, J.T. and Baldwin, J.N. (2005), “ Liquefaction Susceptibility Mapping ST. Louis, Missouri and Illinois-Final Technical Report”, published in web.er.usgs.gov/reports/ abstract/2003/cu/03HQGR0029.pdf.Premalatha, K. and Anbazhagan, P. (2004), “Microzonation of Liquefaction and Amplification Potential ofChennai City,” International Workshop on Risk Assessment in Site Characterization and Geotechnical Design,Bangalore, INDIA. GEORISK-2004, November 26, 27, pp 249-255.Project Vasundhara: (1994), “Geo scientific Analysis, Database creation and Development of GIS for parts ofsouth Indian Peninsular Shield”. ISSN O254–0436.Purnachandra Rao, N. (1999), “Single station moment tensor inversion for focal mechanisms of Indian intra-plateearthquakes”, Current Science, Vol. 77, pp 1184-1189.Radhakrishnan, B. P and Vaidyanathan. R. (1997), “Geology of Karnataka” Geological Society of India,Bangalore.RaghuKanth, S. T. G. and Iyengar, R.N. (2006),”Seismic hazard estimation for Mumbai city”, Current Science,Vol. 91, No. 11, pp 1486-1494.RaghuKanth, S. T. G. and Iyengar, R.N. (2007), “Estimation of seismic spectral acceleration in Peninsular India,”J. Earth Syst. Sci. 116, No. 3, pp. 199–214.170

ReferencesRaghuKanth, S.T, (2005), “Engineering Seismic source Models and Stroung Ground motion,” PhD, Thesis,Department of Civil Engineering, Indian Institute of Science Bangalore, India.Rajendran, K. and Rajendran, C. P. (1999),” Seismogenesis in the stable continental interiors: an appraisal basedon two examples from India”, Tectonophysics, Vol. 305, pp 355–37.Rajiv Ranjan, (2005), “Seismic Response Analysis of Dehradun city, India,” M.Sc Thesis, International Institutefor Geo-Information Science and Earth Observations -Enschede, Netherlands, P-86.Ram, A. and Rathor. H. S. (1970), “On frequency magnitude and energy of significant Indian earthquakes,” PureAppl. Geophys. 79, 26-32.Ramalingeswara Rao, B. (2000), “Historical Seismicity and deformation rates in the Indian Peninsular Shield“,Journal of Seismology, Vol. 4, pp 247-258.Ramalingeswara Rao, B. and Sitapathi Rao, P. (1984), “Historical Seismicity of Peninsular India”, Bull.Seismological Soc. America,Vol.74, pp 2519 -2533.Rao, K.S and Neelima Satyam D, (2005), “Seismic Microzonation Studies for Delhi Region,” Symposium onSeismic Hazard Analysis and Microzonation, September 23-24, Roorkee, pp 213-234.Rao, R., Seshamma, C. V. and Mandal, P. (1998), “Estimation of Coda Qc and spectral characteristics of somemoderate earthquakes of southern Indian peninsula”, Unpublished Report.Rau, J.L. (1994) “Urban and environmental issues in East and Southeast Asian coastal lowlands,” EngineeringGeology, 37,PP 25–29.Robertson, P. K. and Wride, C. E. (1998), ‘‘Evaluating cyclic liquefaction potential using the cone penetrationtest.’’ Can. Geotech. J., Ottawa, Vol. 35(3), pp 442–459.Saaty, T.L. (1968) Mathematical models of arms control and disarmament. In: Application of MathematicalStructures in Politics. John Wiley, N.Y.Saaty, T.L. (1980), The Analytic Hierarchy Process, (McGraw-Hill, New York, NY).Saaty, T.L. (1990) Multicriteria Decision Making: The Analytic Hierarchy Process, (RWS Publications, Pittsburgh,PA).Salome M. R, and Glenn J. R, (2001), “Ground Motion Amplification of Soils in the Upper MississippiEmbayment”, National Science Foundation Mid America Earthquake Center, Report No. GIT-CEE/GEO-01-1.Schnabel, P. B., (1973), “Effects of Local Geology and Distance from Source on Earthquake Ground Motion,”Ph.D. Thesis, University of California, Berkeley, California.Seed, H. B., and I. M. Idriss, (1971), “Simplified procedure for evaluating soil liquefaction potential,” Journal ofthe Soil Mechanics and Foundation Division, ASCE, Vol.97, No.9, pp 1249-1274.Seed, H. B., and Idriss, I. M., (1970), “Soil Moduli and Damping Factors for Dynamic Response Analyses,”Earthquake Engineering Research Center, University of California, Berkeley, California, Rep. No. EERC-70/10.Seed, H. B., Idriss, I. M., and Arango, I. (1983), “Evaluation of Liquefaction potential Using Field PerformanceData,” Journal of Geotechnical Engineering, Vol. 109, No. 3, March, pp 458-482.Seed, H. B., Tokimatsu, K., Harder, L. F., and Chung, R. M. (1985), “The influence of SPT procedures in soilliquefaction resistance evaluations,” J. Geotech. Engrg., ASCE, 111(12), 1425–1445.Seed, H.B, Mori, K. and Chan, C.K. (1975), “Influence of Seismic History on the Liquefaction Characteristics ofSands,” Report No. EERC 75-25, Earthquake Engineering Research Center, University of California, Berkeley.Seed, H.B., and I.M., Idriss, (1982), “Ground Motions and Soil Liquefaction during Earthquakes, Monogr.5”,Earthquake Engineering Research Institute, University of California, Berkeley.SEISAT, (2000), “Seismotectonic Atlas of India,” published by Geological Survey of India.171

ReferencesShankar, D and Sharma, M.L. (1997), “Statistical Analysis of Completeness of seismicity data of the Himalayaand its effects on Earthquake Hazard Determination,” ISET Journal of Earthquake Technology , 34, No.3,pp 159-170.Singh, S. K., Ordaz, M., Dattatrayam, R. S. and Gupta, H. K. (1999), “A Spectral Analysis of the 21 May 1997,Jabalpur, India, Earthquake (Mw = 5.8) and Estimation of Ground Motion from Future Earthquakes in the IndianShield Region”, Bulletin of the Seismological Society of America, 89(6), pp 1620-1630.Sitharam, T. G. and Anbazhagan, P. (2007), “Seismic Hazard Analysis for the Bangalore Region, Nat.Hazards”,Vol. 40, pp 261–278.Sitharam, T. G., Anbazhagan, P., and Ganesha Raj, K. (2006), “Use of remote sensing and seismotectonicparameters for seismic hazard analysis of Bangalore”, Nat. Hazards Earth Syst. Sci., Vol. 6, pp 927–939.Sitharam, T. G., Anbazhagan, P., Mahesh, G. U., Bharathi, K., and Nischala Reddy, P. (2005), “SeismicHazard Studies using Geotechnical Borehole Data and GIS,” Symposium on Seismic Hazard Analysis andMicrozonation, September 23-24, Roorkee, pp 341-358.Sitharam, T.G., Srinivasa Murthy, B.R., and Aravind Kolge, (2001), ’’A Post- Mortem of the Collapse ofStructures in Ahmedabad during the Bhuj Earthquake“, Proceedings, Indian Geotechnical Conference, Vol.I, pp344-347.Skempton, A. W. (1986), “Standard Penetration Test Procedures.” Geotechnique, Vol.36, No.3, pp 425-447.Sridevi Jade,(2004), “ Estimates of plate velocity and crustal deformation in the Indian subcontinent using GPSgeodesy“, Curr. Sci., Vol. 86, pp 1443–1448.Srinivasan, R. and Sreenivas, B. L. (1977); “Some new geological features from the Landsat imagery ofKarnataka”, Jour. Geol. Soc. India, Vol. 18, pp 589-597.Stepp, J. C. (1972), “Analysis of the completeness of the earthquake sample in the Puget Sound area and its effectson statistical estimates of earthquakes hazard”, Proc. Inter. Conf. Microzonation for safer construct. Res. Appl.Seattle, Washington, Vol. 64, No. 4, pp 1189-1207.Stewart, J.P., Liu, A.H., Choi, Y., and Baturay, M.B.(2001), “Amplification factors for spectral acceleration inactive regions”, Pacific Earthquake Engineering Research Center, PEER Report 2001/10.Stokoe II, K.H., Wright, G.W., James, A.B., Jose, M.R. (1994), “Characterization of geotechnical sites by SASWmethod”, In: Woods, R.D. (Ed.), Geophysical Characterization of Sites: ISSMFE Technical Committee #10, OxfordPublishers, New Delhi.Subrahmanya, K. R. (1996), “Active intraplate deformation in south India, Tectonophysics”, Vol.262, pp 231-241.Subrahmanya, K. R. (2002), “Deformation related lineaments in the Indian Peninsula near 13 0” , Journal ofGeophysics, XXIII-2, pp 59-68.Tandon, A.N. (1992), “Seismology in India –an Overview up to 1970”, Current science, Vol. 62 No 1&2, pp 9-1.Tokimatsu, K., (1995), “Geotechnical site characterization using surface waves,” Proc. 1st Int. Conf. on Earth.Geotechn. Eng., IS-Tokyo, p-36.Tripathi, J. N, and Ugalde, A.(2004), “Regional estimation of Q from seismic coda observations by theGauribidanur seismic array (southern India),” Physics of the Earth and Planetary Interiors, 145, pp 115-126.Tseng T.L. and Chen, W.P,(2006), “Probing the Southern Indian Shield with P-Wave Receiver-Function Profiles”Bulletin of the Seismological Society of America, Short note, 96: pp 328 – 333.Tuladhar, R., Yamazaki, F., Warnitchai, P., and Saita, J. (2004), “Seismic microzonation of the greater Bangkokarea using microtremor observations”, Earthquake Engng Struct. Dyn. 33:211–225.USGS (2006), “Susceptibility Map of the San Francisco Bay Area”, open File Report 00-444, URL:http://geomaps.wr.usgs.gov/sfgeo/.172

ReferencesValdiya, K.S. (1998), “Late Quaternary Movements and Landscape Rejuvenation in South-eastern Karnataka andAdjoining Tamil Nadu in South India Shield”, Journal of Geological Society of India, India, Feb.1998, Vol. 51, pp139-166.Van Rooy, J. L. and Stiff, J.S. (2001) “Guidelines for urban engineering geological investigations in South Africa”Bulletin of Engineering and Geological Environment, 59, PP, 285–295.Wang, Z. (2005), “Discussion on Problems in the application of the SSHAC probability method for assessingearthquake hazards at Swiss nuclear power plants”, Eng. Geol., Vol. 78, pp 285–307, Engineering Geology, Vol. 82,pp 86–88.Wells, D.L., Coppersmith, K.J. (1994), “New empirical relationships among magnitude, rupture length, rupturewidth, rupture area, and surface displacement”, Bull. Seismol. Soc. Am. 4 (84), 975– 1002.Westen, C.V. (2005), “Strengthening Local Authorities in Risk Management SLARIM”availableonline”,http://www.itc.nl/research/policy/spearhead3/vwesten/PaperSLARIM.PDF.Xia, J., Miller, R.D., and Park, C.B. (1999), “Estimation of near-surface shear-wave velocity by inversion ofRayleigh wave,” Geophysics,Vol.64(3), pp 691-700.Xu, Y., Xia, J., and Miller, R.D. (2006), “Quantitative estimation of minimum offset for multichannel surfacewavesurvey with actively exciting source,” Journal of Applied Geophysics, 59, No. 2,117-125.Yilmaz , I., and Bagci, A.(2006), “Soil liquefaction susceptibility and hazard mapping in the residential area ofKutahya (Turkey)”, Environ Geol, Vol. 49, pp 708–719.Youd, T. L., and Noble, S. K. (1997), ‘‘Magnitude scaling factors.’’ Proc., NCEER Workshop on Evaluation ofLiquefaction Resistance of Soils, Nat. Ctr. for Earthquake Engrg. Res., State Univ. of New York at Buffalo, pp 149–165.Youd, T.L., Idriss, I.M., Andrus, R.D., Arango, I., Castro, G., Christian, J.T., Dobry, R., Liam Finn, W.D.,Harder Jr., L.H., Hynes, M.E., Ishihara, K., Koester, J.P., Liao, S.S.C, Marcuson, W.F., Marting, G.R.,Mitchell, J.K., Moriwaki, Y., Power, M.S., Robertson, P.K., Seed, R.B, And Stokoe, K.H. (2001), “LiquefactionResistance of Soils: Summary from the 1996 NCEER and 1998. NCEER/NSF Workshops on Evaluation ofLiquefaction Resistance of Soils”, Journal of Geotechnical and Geoenvironmental Engineering, pp 817 – 833.Zhang, S.X., L.S. Chan, and Xia, J (2004), “The selection of field acquisition parameters for dispersion imagesfrom multichannel surface wave data.” Pure and Applied Geophysics, 161, pp 185-201.173