seismic hazard assessment of chennai city considering

Seismic hazard assessment of Chennai cityconsidering local site effects

A Boominathan∗, G R Dodagoudar, A Suganthi and R Uma Maheswari

Indian Institute of Technology Madras, Chennai 600 036, India.∗e-mail: [email protected]

Chennai city suffered moderate tremors during the 2001 Bhuj and Pondicherry earthquakes andthe 2004 Sumatra earthquake. After the Bhuj earthquake, Indian Standard IS: 1893 was revisedand Chennai city was upgraded from zone II to zone III which leads to a substantial increase of thedesign ground motion parameters. Therefore, a comprehensive study is carried out to assess theseismic hazard of Chennai city based on a deterministic approach. The seismicity and seismotectonicdetails within a 100 km radius of the study area have been considered. The one-dimensional groundresponse analysis was carried out for 38 representative sites by the equivalent linear method usingthe SHAKE91 program to estimate the ground motion parameters considering the local site effects.The shear wave velocity profile was inferred from the corrected blow counts and it was verifiedwith the Multichannel Analysis of Surface Wave (MASW) test performed for a representative site.The seismic hazard is represented in terms of characteristic site period and Spectral AccelerationRatio (SAR) contours for the entire city. It is found that structures with low natural periodundergo significant amplification mostly in the central and southern parts of Chennai city dueto the presence of deep soil sites with clayey or sandy deposits and the remaining parts undergomarginal amplification.

1. Introduction

Earthquakes are one of the most destructive nat-ural hazards of the world. These natural eventscan cause massive damage to structures and leadto total devastation of cities. Earthquakes, whichmanifest themselves in the form of vibrations ofthe earth, are caused by the abrupt release ofstrain that has accumulated over time. There havebeen several major earthquakes that have occurredin India over the recent years – Uttarkashi (1991;ML 6.6), Latur (1993; Mw 6.1), Jabalpur (1997;Mw 5.8), Chamoli (1999; ML 6.8) Bhuj (2001; Mw

7.6) and Kashmir (2005; Mw 7.6).Even though peninsular India has been consi-

dered as a stable continental region for years,damages caused during the 2001 Bhuj earthquake(Mw 7.6) demanded the immediate study of thepeninsular region. Earthquakes of Koyna (1967;

Mw 7.6), Latur (1993; Mw 6.1) and Jabalpur (1997;Mw 5.8) also occurred in the ‘stable’ Indian shield.A review of the historical as well as recent earth-quake activity in peninsular India indicated thatdifferent parts of the peninsular region are char-acterized by a low to moderate level of seismicactivity. But it is only in recent decades thatthe occurrence of some large and damaging earth-quakes has caused concern, which led to the studyof peninsular seismicity in greater detail (Chandra1977). There is a need to prepare a seismic haz-ard map and site specific design response spectrain accordance with the revision of Indian Stan-dard IS: 1893 (2002), which will enable urbanplanners to design earthquake resistant structuresand strengthen existing unstable structures. Thisrevision of seismic zonation of cities has broughtabout awareness of seismic microzonation of citiesparticularly in southern India. The recent studies

Keywords. Seismic hazard; SHAKE91; shear wave velocity; characteristic site period; spectral acceleration ratio.

J. Earth Syst. Sci. 117, S2, November 2008, pp. 853–863© Printed in India. 853

854 A Boominathan et al

by Sitharam et al (2006) have suggested thatBangalore needs to be upgraded from the presentseismic zone II to zone III based on the regionalseismotectonic details and hazard analysis. There-fore an attempt has been made to carry out seismichazard assessment for Chennai city considering siteeffects.

2. Geology of the study area

Chennai is located between 12.75◦–13.25◦N and80.0◦–80.5◦E on the southeast coast of India andin the northeast corner of Tamil Nadu. It is India’sfourth largest metropolitan city covering an areaof 1177 km2. The city stretches nearly 25 km alongthe bay coast from the southern part to the north-ern part of the city. The seacoast is flat and sandyfor about one km from the shore. The study areahas two distinct geological environments. The east-ern and southern parts of the city consist of shal-low bedrock (crystalline) while the western andnorthern areas have Gondwana deposits below thealluvium (Ballukraya and Ravi 1994). Almost theentire area is covered by Pleistocene/Recent allu-vium, deposited by the two rivers, namely, Cooumand Adyar. The thickness of this formation rangesfrom a few meters in the southern parts to as muchas 50 m in the central and northern parts, with anaverage of 20 to 25 m. This alluvium is made upof mainly clays, sands, sandy clays and occasionalboulder/gravel zones. Sandy areas are found alongthe river banks and coasts. Igneous/metamorphicrocks are found in the southern area. The marinesediments containing clay-silt sands and charnock-ite rocks are found in the eastern and northernparts. The western parts are composed of allu-vium and sedimentary rocks. A thin layer of lat-erites is also found at some places. Well-roundedpebbles and small boulders have been encounteredat several locations at varying depths. It is seenthat in general, the eastern coastal zone is pre-dominantly sandy, while the northwestern region ismostly clayey in nature.

3. Seismicity and seismotectonicsof the region

Seismic susceptibility of the area can be assessedwith the help of the regional seismicity data onthe occurrence of past earthquakes and the seismo-tectonic details that describe the tectonic featuresaround the area. Indian seismicity is characte-rized by a relatively high frequency of great earth-quakes and a relatively low frequency of moderateearthquakes. General discussions on the seismi-city of peninsular India (PI) have been previously

presented by Chandra (1977), Rao and Rao (1984)and Khattri (1992). Typical historical seismicityof PI based mainly on Gauribidanur seismic array(GBA) detections of regional earthquakes spanningtwo decades (1978–1997) is shown in figure 1 (afterGangrade and Arora 2000).

Seismological information and seismotectonicfeatures of the region were collected from the latestSeismotectonic Atlas of India (2000). It was foundthat regions with pronounced variation in sedi-ment thickness show higher seismicity as comparedto the parts with more or less uniform thickness.Historical earthquake information within a 100 kmradial distance from Chennai was obtained fromNational Earthquake Information Center (NEIC)of the US Geological Service which includes datafrom 1800 A.D. onwards. The seismological detailsgathered for establishing ground motion para-meters for the Prototype Fast Breeder Reactor(PFBR) building site at Kalpakkam, located 60 kmaway from Chennai city, were also used in thisstudy. These data were also used by Ghosh (1994)to estimate design based ground motion para-meters for the PFBR site.

Seismicity data alone are not usually sufficientfor the purpose of identifying and characteri-zing earthquake sources. Seismotectonic data arenecessary to supplement the historical and instru-mental seismicity data. A detailed investigation onthe seismotectonics has been carried out to studythe faults in and around Chennai. The fault detailsare obtained from the Seismotectonic Atlas of India(2000) and the fault studies carried out by theOil and Natural Gas Commission (Ghosh 1994).The details of known faults around 100 km of thestudy area with seismicity is given in table 1. Allthe seismic sources in the study are considered asline sources. Due to lack of sufficient informationabout seismic activity details of the faults, all thefaults are considered to be active and have beengiven equal weightage in the analysis.

4. Base map preparation

A base map is one of the important componentsof seismic microzonation studies, the preparationof which requires a special consideration. Overthe last four decades Geographical InformationSystems (GIS) have emerged as the predominantmedium for graphic representation of geospatialdata, including geotechnical, geologic and hydro-logic information. Toposheets of scale 1:50,000obtained from Survey of India were used for prepa-ring the base map. The Arc Info 8.0.1 versionin GIS platform and Google Earth software wereused for the construction of base map of the city(figure 2).

Seismic hazard assessment of Chennai city 855

Figure 1. Historical seismicity of peninsular India (Gangrade and Arora 2000).

Table 1. The details of faults and seismicity in the vicinity of Chennai city.

Fault Hypocentral Momentlength Distance distance, R magnitude PGA

Sl. no. Name of fault L (km) (km) (km) (Mw) (g)

1 Fault 15d 40 10 14 4.0 0.066

2 Fault 24 365 10 14 4.4 0.106

3 Fault 53 137 32 34 4.1 0.029

4 Kilcheri fault 26 33 34 4.0 0.025

5 Fault 15a 105 46 47 4.5 0.032

6 Neotectonic fault 105 48 49 3.8 0.013

7 Palar fault 85 59 60 4.0 0.013

8 Tambaram fault 10 59 60 4.4 0.021

9 Fault 15 96 61 62 3.7 0.009

10 Fault 52 115 67 68 3.6 0.007

11 Fault 15e 50 68 69 4.5 0.020

12 Fault 54 129 70 71 3.8 0.009

13 Mahapalipuram fault 5 75 76 4.0 0.010

14 Kalkulam fault 36 82 83 3.6 0.005

15 Muttukadu fault 11 95 96 3.5 0.004

16 Fault 26d 160 96 97 4.5 0.013

17 Fault 56e 75 97 98 4.5 0.013

18 Fault 26 1000 98 99 4.5 0.013

More than 400 bore logs were collected fromthe reputed soil investigation agencies to establishthe soil profile and its characteristics for almost

entire Chennai. The data collected includes thick-ness of layers, type of soil, index properties suchas liquid limit, plastic limit, density and SPT


Figure 2. Base map of Chennai city.

N -value, the location of bedrock, etc. A total of38 representative sites as shown in figure 2 areidentified for the development of hazard curves forthe city.

5. Seismic hazard assessment

Seismic hazard analysis is the quantitative esti-mation of ground-shaking hazards at a particu-lar site based on the identification of all possiblesources of seismic activity, estimation of their asso-ciated seismicity and prediction of the probableconsequent ground motions. In the present study,the seismic hazard for Chennai was estimated bycarrying out deterministic seismic hazard analysis(DSHA) by adopting the procedure specified byReiter (Kramer 1996).

The fault details around the study area (smallerpart of Chennai city) are listed in table 1. The faultmap developed for Chennai city within 100 kmradial distance is shown in figure 3. The earthquakedata collected from various sources gives an over-all idea about the seismicity details associatedwith each fault source. This information is usedfor assigning the maximum magnitude for eachfault source by taking into consideration the seis-micity around that particular fault source. Themaximum magnitude for all the faults that were

obtained from the observed seismicity details isalso presented in table 1. The maximum magni-tude on a particular seismic source was taken asthe largest observed past magnitude plus 0.5 (Kijkoand Graham 1998; Sokolov et al 2001). It canbe seen from table 1 that the maximum momentmagnitude varies from 3.5 to 4.5.

After getting moment magnitude from pastearthquake history, the peak ground acceleration(PGA) at bedrock level is estimated using theattenuation equation of strong ground motionproposed for peninsular India by Iyengar andRaghukanth (2004):

ln(y) = C1 + C2(M − 6) + C3(M − 6)2

− ln(R) − C4R + ln ε, (1)

where C1 = 1.7816; C2 = 0.9205; C3 = −0.0673;C4 = 0.0035 and σ(ln ε) = 0.3136 M , y, R and εrefer to moment magnitude, Peak Ground Acceler-ation (PGA) in g, hypocentral distance (km) andstandard deviation factor respectively.

Since the peninsular shields in general, are moresusceptible to the shallow focus earthquake, a focaldepth of 10 km is assumed and the hypocentraldistance is calculated for all fault sources (Rao andRao 1984). The PGA has been calculated for a ref-erence site located at the center of the region understudy (center of the fault map). The PGA calcu-lated for all faults is presented in table 1. It can beobserved from the table that the maximum com-puted PGA value ranges from 0.004 g to 0.106 g.Thus it can be established from DSHA that a haz-ard of 0.106 g (PGA) results from the controllingearthquake of Mw 4.4 and epicentral distance of11 km from fault 24. This PGA has been used asan input acceleration after suitable scaling of inputacceleration time history and the same is used forthe ground response analysis.

6. Local site effects

Local site conditions profoundly influence mostof the important characteristics mainly the acce-leration amplitude and frequency characteristicsof ground motion during an earthquake. Theextent of this modification depends on the geo-metry of the soil profile, thickness and pro-perties of the soil profile and characteristics ofthe input motion. The local site effects on theground motion are commonly evaluated by car-rying out one-dimensional ground response analy-sis. One-dimensional ground response analysis canbe carried out by equivalent linear or non-linear


Figure 3. Fault map for Chennai city.

methods. It is preferable to carry out nonlinearground response analysis with the site specificmodulus reduction and damping curves or con-stitutive models. But due to non-availability ofsite specific modulus curves, the ground responseanalysis in the present study is carried out byequivalent linear approach using computer pro-gram SHAKE91. However, the ground responseanalysis by equivalent linear approach predicts rea-sonably well the surface ground motion for low tomoderate earthquake regions like Chennai city.

The ground response analysis is carried out at38 representative sites of Chennai city as shown infigure 2. In the representative sites, the depth to

bedrock varies from about 5 m in southern region toabout 35 m in north and northwestern regions.The input data include the acceleration time his-tory at bedrock, shear wave velocity or maximumshear modulus, modulus reduction and dampingcurves of soil layers. In the present study, theshear wave velocity of soil layers is estimatedfrom the corrected blow counts (N1)60. The stan-dard modulus reduction and damping curves pro-posed by Sun et al (1988), and Seed and Idriss(1970) and Idriss (1990) are used for clay andsands layers respectively. Schnabel (1973) modulusreduction and damping ratio curves are used forthe rock.


Figure 4. Acceleration time history and Fourier amplitude spectrum of input motion.

6.1 Selection of input accelerationtime history

The important characteristics of the bedrockmotion include acceleration amplitude, frequencyand duration content. The single frequency con-tent (i.e., predominant period) describes the con-centration of maximum energy of a ground motion.In this study, the frequency content has been cal-culated based on relation between moment mag-nitude and distance from causative fault proposedby Chang and Krinitzsky (1977). The predominantperiod is found to be 0.2 s. The bracketed durationof the bedrock motion is estimated as 3.0 s usingthe chart proposed by Seed et al (1969).

In places where recorded accelerograms are notavailable such as Chennai city, it is a generalpractice to select a recorded time history fromthe strong motion database or to simulate anartificial time history based on the site character-istics. In this case, a recorded acceleration timehistory is selected from the strong motion data-base and scaled to suit the requirements. The fre-quency content of the 1952 Kern Country (Taft)input motion matches with the predominant periodcalculated from the above relation. In addition tothe above, the estimated bracketed duration alsomatches with the Taft kern ground motion. Hence,the Taft Kern earthquake of PGA 0.185 g groundmotion time history has been selected and scaledusing the QUAKE/W software for the PGA of0.106 g obtained from the DSHA and the esti-mated, bracketed duration of 3 s and predominantperiod of 0.2 s. The developed scaled time his-tory and Fourier amplitude of bedrock motion areshown in figure 4. The above time history of accel-eration obtained for the reference site located atthe center of the study region is adopted as aninput motion for all 38 sites.

6.2 Determination of shear wave velocity

A large and reliable database for a number of sitesin and around Chennai city has been obtainedfrom reputed geotechnical agencies. The details

of the soil layers and their engineering proper-ties were assessed from the compiled data. TheSPT-N values obtained in the field were correctedfor various factors: overburden pressure, hammerenergy, borehole diameter, rod length and finescontent. The shear wave velocity, Vs was esti-mated from the corrected blow counts ((N1)60)using the following empirical equations (JapaneseRoad Association 1980):

Vs (m/s) = 100 (N1)1/3

60 (for clay), (2)

Vs (m/s) = 80 (N1)1/3

60 (for sand). (3)

The notation (N1)60 describes the blow countvalue corrected for field procedures and overburdenstress.

(N1)60 = N60 CN , (4)

where N60 is the blow count corrected for the fieldprocedures like energy ratio, borehole diameter,sampling method and rod length to an averageenergy ratio of 60% and CN is the overburden stresscorrection factor (Skempton 1986). The details ofrepresentative sites with the corresponding loca-tion coordinates and the weighted average shearwave velocity for 30 m obtained from the aboveequations (2) and (3) are tabulated in table 2.

In order to verify the shear wave velocityobtained from the empirical correlations, theMASW test has been conducted at selected sites.The results of MASW tests carried out at a typ-ical site near west Mogappair area located in thenorthwest region of Chennai city are describedbelow. The subsoil at the site consists of four lay-ers. The top layer of about 6 m thickness consistsof medium-stiff to stiff-silty-clay with blow countsvarying from 6 to 18. This layer is followed by aloose to medium dense sand deposit of 7m thick-ness with blow counts increasing from 4 to 16


Table 2. Details of representative sites.

Weighted averageThickness of shear velocity

Sites Latitude Longitude overburden (m) for 30m (m/s)

Tambaram 80.1423 12.9252 5 915

Nandambakkam 80.1987 13.0149 5 918

Saidapet 80.2114 13.0229 8 809

Guindy 80.2364 13.0263 9 788

Perungudi 80.2373 12.9561 10 747

Velachery 80.2263 12.9795 10 768

Palavakkam 80.2531 12.9603 12 723

Ramapuram 80.1817 13.0313 12 728

Balaji Nagar 80.2666 13.0707 12 731

Royapuram 80.2985 13.1016 12 756

Kolathur 80.2207 13.1177 12 769

Avadi 80.106 13.1123 12 776

Thorapakkam 80.2358 12.9511 14 680

Vepery 80.2684 13.0845 17 557

Nesapakkam 80.1855 13.0358 17 579

Perambur 80.2591 13.1103 17 637

Nungambakkam 80.2436 13.0582 18 530

Vyasarpadi 80.2574 13.1342 18 594

Nandanam 80.239 13.0237 19 496

Vadapalani 80.2174 13.0542 19 538

RA Puram 80.2609 13.0201 20 474

Korattur 80.1835 13.1018 20 484

Poes 80.2509 13.0403 20 485

Ambattur 80.157 13.0989 20 515

Ashok Nagar 80.2181 13.0315 21 431

Mandaveli 80.2677 13.0236 21 491

Tiruvottiyur 80.3054 13.1644 22 398

Manali 80.2613 13.1664 22 438

Alwarpet 80.2483 13.0354 24 430

Adyar 80.2564 13.0073 25 345

Santhome 80.2808 13.0324 25 345

T. Nagar 80.2136 13.0582 25 399

Koyambedu 80.203 13.0712 26 391

Madavaram 80.2389 13.1461 28 241

Anna Nagar 80.2149 13.0888 28 260

Aynavaram 80.2273 13.0956 28 346

Abiramapuram 80.2591 13.0292 30 251

CPR Road 80.2602 13.0357 34 271

with depth. The subsequent layer consists of a softto hard clay deposit of 14 m thickness with blowcounts varying from 5 to 78. This layer is followedby a dense sand deposit of 3 m thickness with blowcounts > 100. The subsequent layer is followed byweathered rock with the thickness of 2 m.

A geometrics 24-channel geode-based seismo-graph with single geode operating system (SGOS)software was used to carry out the MASW tests.The vertical geophones of 4.5 Hz resonant fre-quency (24 nos.) with 2m interval are used to

receive the wavefield generated by the sledgeham-mer of 5 kg. A sledgehammer was discharged atoffsets of 5, 10 and 15 m from the first geophoneto meet the requirement of different types of soil(Xu et al 2006). The acquired surface wave datawere processed using Surfseis software to developdispersion curves as shown in figure 5. The plotof theoretical and measured dispersion curves areshown in figure 6. The shear wave velocity fromthe MASW test and empirical relation along withborelog of the site is shown in figure 7. It can be


Figure 5. Typical dispersion curve.

Figure 6. Plot of theoretical and measured dispersioncurves with shear wave velocity profile.

Figure 7. Variation of shear wave velocity with borelog of the site.

observed from figure 7 that the calculated shearwave velocity values are reasonable matches withthe measured values except at deeper depths wherethe quality of the signal is poor due to reductionof signal-to-noise ratio.

6.3 Results of ground response analysis

The result of the ground response analyses arerepresented in terms of time history of surfaceacceleration and acceleration response spectrum.The amplification factor at different frequenciesobtained from GRA carried out for three typicalsites, namely, rocky, sandy and clayey sites aregiven in figure 8. It can be observed from the fig-ure that the amplification occurs at low frequency(about 1.2 Hz) for the clayey site, mid-range fre-quency (about 4 Hz) for the sandy site and higherfrequency (about 6.5 Hz) for the rocky site. Unlikeother sites, second peak for the rocky site occursat much higher frequencies. The amplification ofthe clayey site is about 1.5 times higher than therocky and sand sites due to the presence of thickdeposits with low shear wave velocity as tabulatedin table 2.

The seismic hazard can be represented either bypredominant period or characteristic site periodcontours. The predominant period is more relevantthan the fundamental natural period, i.e., charac-teristic site period, for the determination of vul-nerability of the structures due to the fact thatthe predominant period represents the maximumenergy concentration. However, Roca et al (2006)


Figure 8. Frequency response function of selected regionsof Chennai.

Figure 9. Characteristic site period contours.

recommend the natural period of the soil sys-tem for microzoning studies to represent local siteeffect on seismic hazard. Moreover, the result ofground response analyses shows that the predomi-nant period and characteristic site period are rela-tively close for the sites considered. In the presentstudy the seismic hazard is represented by the char-acteristic site period contour as shown in figure 9.It can be noticed from the figure that the sitesat the southern parts of the city have more shortperiods (< 0.15 s) and the central and northernparts have long periods (> 0.5 s). The character-istic site period for the typical rocky, sandy andclayey sites considered earlier is found to be 0.15 s,0.25 s and 0.83 s respectively.

The spectral acceleration (sa) normalized by thesurface PGA has been used to develop the Sa/g

Figure 10. Response spectra for the selected regions ofChennai.

Figure 11. Spectral acceleration ratio contours for 0.5 s.

response spectrum as shown in figure 10 for thethree typical sites. It can be noticed from the figurethat the spectral acceleration values for the clayeysite are high at a wide range of periods up to 0.8 sin contrast to the rocky sites where high spectralacceleration values occurs only at narrow range oflow periods.

Response spectra obtained for various sites arecompared with the response spectra obtainedfor the input bedrock. The Spectral AccelerationRatio, SAR (ratio of spectral acceleration at thesurface to the bedrock spectral acceleration) con-tours are normally plotted for typical periodsof structures with short periods in the range of0.1–0.5 s and mid period in the range of 0.4–2 s(Kramer and Stewart 2004). Figure 11 shows SAR


contours for a period of 0.5 s utilizing responsespectra obtained from 38 representative sites. Itcan be observed from figure 10 that the amplifi-cation occurs practically in all regions of the citybut significant amplification (SAR greater than 3)occurs mostly in central and southern parts ofChennai city due to the presence of deep soil siteswith clayey or sandy layers.

7. Conclusions

The seismic hazard analysis was carried out for theestablishment of PGA at bedrock level for Chennaicity by DSHA approach. The one-dimensionalground response analysis was also performed byequivalent linear method using SHAKE91 programto estimate the ground motion parameters con-sidering local site effects. Based on the resultsobtained from the above study, the followingconclusions are drawn:

The peak ground acceleration of 0.106 g hasresulted from the controlling earthquake of Mw

4.4 and epicentral distance of 11 km from fault24. The result of the ground response analysis isrepresented by characteristic site period contoursranging from 0.25–0.83 s for the city. The SpectralAcceleration Ratio (SAR) i.e., the ratio of spec-tral acceleration at the surface to the bedrock con-tours, for a typical period of 0.5 s is also plotted.It is found that at short periods, the amplifica-tion occurs practically in all regions of the citybut higher amplification (> 3) occurs mostly in thecentral and southern parts of Chennai city due tothe presence of deep soil sites with clayey or sandydeposits. Thus, it is to be noted that the seis-mic hazard assessment has to incorporate the localsite effects as realistically as possible in the analy-sis procedure in order to place reliability on theestimated peak ground acceleration.

Acknowledgements

The authors would like to thank the Departmentof Science and Technology, Government of Indiafor funding the sponsored research project enti-tled ‘Geotechnical and seismological investigationsfor development of shake maps for Chennai city’(DST No: 23(497)/SU/2004 Dt. 09/08/2005). Theauthors extend their thanks to M/s. GeotechnicalSolutions, Chennai for providing borehole data ofvarious parts of the city.

References

Ballukraya P N and Ravi R 1994 Hydrogeological environ-ment of Madras city aquifers; J. Appl. Hydrol. 87(1–4)75–82.

Chandra U 1977 Earthquakes of peninsular India – A seis-motectonic study; Bull. Seismol. Soc. Am. 67 1387–1413.

Chang F K and Krinitzsky E L 1977 Duration, spectralcontent and predominant period of strong motion earth-quake records from Western United States, Miscellaneouspaper 5-73-1, U.S. Army Corps Engineers WaterwaysExperiment Station Vicksburg, Missipppi.

Gangrade B K and Arora S K 2000 Seismicity of the Indianpeninsular shield from regional earthquake data; PureAppl. Geophys. 157 1683–1705.

Ghosh A K 1994 Design basis ground motion parametersfor PFBR site Kalpakkam; BARC, Mumbai.

IS 1893 (Part 1): 2002 Indian Standard Criteria for Earth-quake Resistant Design of Structures: Part 1 GeneralProvisions and Buildings (Fifth Revision).

Idriss I M 1990 Response of soft soil site during earthquakes;In: Proceedings of H. Bolton Seed Memorial Symposium(ed.) Duncan J M, University of California, Berkeley 2273–289.

Iyengar R N and Raghukanth S T G 2004 Attenuation ofstrong ground motion in peninsular India; Seismol. Res.Lett. 75 530–540.

Japan Road Association (JRA) 1980 Specification and inter-pretation of bridge design for highway – Part V: Resilientdesign, 14–15.

Khattri K N 1992 Seismic hazard in Indian region; Curr.Sci. 62(1–2) 109–116.

Kijko A and Graham G 1998 Parametric-historic procedurefor probabilistic seismic hazard analysis, Part I: Estima-tion of maximum regional magnitude Mmax; Pure Appl.Geophys. 152 413–442.

Kramer S L 1996 Geotechnical earthquake engineering;Prentice-Hall, Inc.

Kramer S L and Stewart J P 2004 Geotechnical aspects ofseismic hazards; In: Earthquake Engineering from Engi-neering Seismology to Performance Based Engineering,Chapter 4.

Mark R K 1977 Application of linear statistical modelof earthquake magnitude versus fault length in esti-mating maximum expectable earthquakes; Geology 5464–466.

Rao B R and Rao P S 1984 Historical seismicity of Penin-sular India; Bull. Seismol. Soc. Am. 74(6) 2519–2533.

Roca A, Oliveira C S, Ansal A and Figueras 2006 Local siteeffects and microzonation; In: Assessing and ManagingEarthquake Risk (eds) Oliveria C S, Roca A and Goula X(Netherlands: Springer Publishers) Chapter 4.

Seed H B, Idriss I M and Kiefer F W 1969 Characteris-tics of rock motions during earthquakes; Journal of theSoil Mechanics and Foundations Divisions, ASCE 95(5)1199–1218.

Seed H B and Idriss I M 1970 Soil moduli and dampingfactors for dynamic response analyses; EarthquakeEngineering Research Center, University of California,Berkeley, EERC 70/10.

Seismotectonic Atlas of India 2000 Geological Survey ofIndia, New Delhi.

Sitharam T G, Anbazhagan P and Ganesha Raj K 2006 Useof remote sensing and seismotectonic parameters for seis-mic hazard analysis of Bangalore; Natural Hazards EarthSystem Science 6 927–939.

Skempton A W 1986 Standard penetration test proceduresand the effects in sands of overburden pressure, rela-tive density, particle size, aging and overconsolidation,Geotechnique 36 425–447.

Sokolov Y V, Loh C H and Wen K L 2001 Empiricalmodels for site- and region-dependent ground motionparameters in the Taepei area: A unified approach; Earth-quake Spectra 17(2) 313–332.


Sun J I, Golesorkhi R and Seed H B 1998 Dynamicmoduli and damping ratios for cohesive soils; EarthquakeEngineering Research Center, University of California,Berkeley, California, EERC-88/15.

Xu Y, Xia J and Miller R D 2006 Quantitative estimationof minimum offset for multichannel surface wave surveywith actively exciting source; J. Appl. Geophys. 59(2)117–125.

MS received 30 September 2007; revised 8 April 2008; accepted 10 April 2008

seismic hazard assessment of chennai city considering

Documents