Estimation of Site Response in Kachchh, Gujarat, India ... - Springer Link

85 downloads 6990 Views 836KB Size Report
Jan 26, 2001 - Preparation of such maps is critical for both recovery and future planning of. Gujarat. .... The maximum damage (intensity X+ on the MM Scale) occurred in an area of ... external hard disk (2 GB) and a GPS timing system.
Pure appl. geophys. 162 (2005) 2479–2504 0033–4553/05/122479–26 DOI 10.1007/s00024-005-2784-6

 Birkha¨user Verlag, Basel, 2005

Pure and Applied Geophysics

Estimation of Site Response in Kachchh, Gujarat, India, Region Using H/V Spectral Ratios of Aftershocks of the 2001 Mw 7.7 Bhuj Earthquake PRANTIK MANDAL,1 R. K. CHADHA,1 C. SATYAMURTY,1 I. P. RAJU,1 and N. KUMAR1

Abstract—Site response in the aftershock zone of 2001 Bhuj Mw 7.7 earthquake has been studied using the H/V spectral ratio method using 454 aftershocks (Mw 2.5–4.7) recorded at twelve threecomponent digital strong motion and eight three-component digital seismograph sites. The mean amplification factor obtained for soft sediment sites (Quaternary/Tertiary) varies from 0.75–6.03 times for 1–3 Hz and 0.49–3.27 times for 3–10 Hz. The mean amplification factors obtained for hard sediment sites (hard Jurassic/Mesozoic sediments) range from 0.32–3.24 times for 1–3 Hz and 0.37–2.18 times for 310 Hz. The upper bounds of the larger mean amplification factors for 1–3 Hz are found to be of the order of 3.13–6.03 at Chopadwa, Vadawa, Kavada, Vondh, Adhoi, Jahwarnagar and Gadhada, whereas, the upper bounds of the higher mean amplification factors at 3–10 Hz are estimated to be of the order of 2.00–3.27 at Tapar, Chopadwa, Adhoi, Jahwarnagar, Gandhidham and Khingarpur. The site response estimated at Bhuj suggests a typical hard-rock site behavior. Preliminary site response maps for 1–3 Hz and 310 Hz frequency ranges have been prepared for the area extending from 23–23.85 N and 69.65– 70.85E. These frequency ranges are considered on the basis of the fact that the natural frequencies of multi-story buildings (3 to 10 floor) range between 1–3 Hz, while the natural frequencies for 1 to 3 story buildings vary from 3–10 Hz. The 1–3 Hz map delineates two distinct zones of maximum site amplification (>3 times): one lying in the NW quadrant of the study area covering Jahwarnagar, Kavada and Gadadha and the other in the SE quadrant of the study area with a peak of 6.03 at Chopadwa covering an area of 70 km · 50 km. While the 3–10 Hz map shows more than 2 times site amplification value over the entire study area except, NE quadrant, two patches in the southwest corner covering Bhuj and Anjar, and one patch at the center covering Vondh, Manfara and Sikara. The zones for large site amplification values (3 times) are found at Tapar, Chopadwa, Adhoi and Chobari. The estimated site response values show a good correlation with the distribution of geological formations as well as observed ground deformation in the epicentral zone. Key words: Site response, Kachchh, intraplate seismicity, aftershocks, ground motion amplification.

1. Introduction Recent devastating Indian earthquakes suggest that about 60% of Indian lithosphere falls in an active seismic zone. Large Indian earthquakes during

1

National Geophysical Research Institute, Hyderabad, 500007, A.P, India

2480

P. Mandal et al.

Pure appl. geophys.,

1800–2001 have caused over 80,000 deaths. Most of the Indian seismic activity is confined to the Himalayan frontal arc. However, earthquakes also occur in the interior of the Indian plate, which are classified as intraplate earthquakes (JOHNSTON, 1994). The Kachchh region (Gujarat, India) is a unique intraplate seismic site that has experienced two large (Mw 7.8 and 7.7) earthquakes within a time span of 182 years (GUPTA et al., 2001). The region lies in zone V (highest seismicity and potential for M8 earthquakes) on the seismic zoning map of India (BUREAU OF INDIAN STANDARDS, 2002). Available fault plane solutions from the first motions of earthquakes suggest that both strike-slip and reverse faulting characterize the Kachchh seismic zone (CHUNG and GAO, 1995). Available campaign GPS data suggest very slow strain accumulation (SRIDEVI et al., 2001). The Bhuj earthquake of Mw 7.7 (intensity X+ on the MM Scale) shook the entire Indian region at 8:46 (IST) on 26 January, 2001 with an epicenter at 23.412oN, 70.232oE and a depth of 23 km in the Kachchh region, Gujarat, India. It caused extensive liquefaction and slope failures over an area of tens of thousands of square kilometers, but produced no surface rupture (RASTOGI et al., 2001). This earthquake caused strong shaking in many densely populated urban centers in the Gujarat state, India, amounting to 10 billion US dollars in damage. The geologic setting of Kachchh, Gujarat consists of a thick crust of Paleozoic age and is characterized by Mesozoic rifting that has been reactivated by Cenezoic tectonism (REDDY et al., 2001). Presently, construction of proper earthquakeresistant buildings in Kachchh, Gujarat is a great challenge for the Indian engineers, due to the fact that seismic hazard maps to guide how and where to build, do not exist. Preparation of such maps is critical for both recovery and future planning of Gujarat. In order to prepare seismic hazard maps for the region, it is necessary to first obtain a better understanding of the site amplification throughout the region. A correlation of geology, soil conditions and distribution of site amplification should explain observed ground deformations. Therefore, a study of site amplification using well-recorded aftershocks of the 2001 Bhuj earthquake would provide significant clues toward understanding the physical processes that brought about the widespread devastation during the Bhuj earthquake in many areas of Kachchh, Gujarat. Surface geological heterogeneities often result in ground motion amplification. This change in ground motions depending on the site geology has often been studied using the H/V ratio method, the ratio at a single site between the Fourier spectra of the horizontal and vertical components of ground motion. This method, using microtremors, was introduced by NOGOSHI and IGARASHI in 1971, but made popular by NAKAMURA (1989) who found it could give good estimates of the frequency and amplitude of the fundamental resonance peak of the local site effect. LERMO and CHAVEZ-GARCIA (1993) extended this method to estimate sediment-induced amplification by S waves. Further, this method may provide good estimates of the site response amplitude even when using single station

Vol. 162, 2005

Estimation of Site Response in India

2481

three-component recordings (KONNO and OHMACHI, 1998). A comparison between observed ground amplification caused by earthquakes and observed H/V peak amplitudes suggests that the latter is mostly smaller than the former (BARD et al., 1997). This technique has been applied on various sets of weak and strong motion data (CHAVEZ-GARCIA et al., 1996; RIEPL et al., 1998; ZARE et al., 1999). These studies suggested that this method could provide very stable and reliable estimates of the resonant frequencies as well as amplification. Others find that the H/V method compares well to the standard spectral method when determining the fundamental frequency of the sediments, but has significantly less success in estimating the amplification, especially at higher frequencies (BONILLA et al., 1997; COUTEL and MORA, 1998). In this paper we present two site response maps for the aftershock zone of the 2001 Bhuj earthquake, one for 1–3 Hz and another for 3–10 Hz. Site amplification values for a total of 20 sites were estimated using the H/V spectral ratio method with aftershock data. The correlation between geology, ground deformations and estimated site amplification values in the aftershock zone of the 2001 Bhuj earthquake is discussed in terms of seismic hazard of the region.

2. Geology and Tectonics Geologically, Quaternary/Tertiary sediments, Deccan volcanic rocks and Jurassic sandstones resting on Archean basement mainly characterize the Kachchh region (GUPTA et al., 2001). Rifting of Gondwanaland in the early Jurassic or late Triassic involved reactivation of Precambrian structures in the eastern as well as western parts of India, and formation of several rift basins like Kachchh, Cambay and Narmada (BISWAS, 1987). The Mesozoic rift-related extensional structures of the Kachchh basin became reactivated as strike-slip or reverse faults as a result of regional compressive stresses due to the collision of the Indian and Eurasian plates since Neogene times (BISWAS and DESHPANDE, 1970). The focal mechanisms of some earthquakes indicate reverse faulting (CHUNG and GAO, 1995). Major structural features of the Kachchh region include several E-W trending faults/folds as shown in Figure.1. The rift zone is bounded by a north-dipping Nagar Parkar fault in the north and a south-dipping Kathiawar fault in the south. Other major faults in the region are the E-W trending Allah Bund fault, Island belt fault, Kachchh mainland fault and Katrol Hill fault. In addition, several NE and NW trending small faults/lineaments are observed (BISWAS, 1987). Seismics, gravity and magneto-telluric surveys indicate undulated basement with 2–5 km deep sediments and Moho depth at 35–43 km in the southern Kachchh region (GUPTA et al., 2001; REDDY et al., 2001).

2482

P. Mandal et al.

Pure appl. geophys.,

Figure 1 A map showing twelve strong-motion accelerographs (black open diamonds) and eight seismograph stations (black solid triangles) along with the main shock epicenter (star). KMU is the Kachchh mainland uplift. Also shown are major faults (lines): ABF, Allah Bund Fault; IBF, Island belt fault; KMF, Kachchh mainland fault; KTF, Katrol Hill fault; NPK, Nagar Parkar fault; NWF, North Wagad fault. The inset in the left lower corner shows the study area with reference to Indian plate boundaries (dark lines). (modified after BISWAS, 1987).

3. Past Seismicity The Kachchh region is characterized by large and moderate but infrequent earthquakes. Based on JOHNSTON’S (1994) classification scheme, great intraplate earthquakes with M ‡ 7.7 have occurred in only two intraplate regions in the world: New Madrid, USA and Kachchh, India (BENDICK et al., 2001; BODIN et al., 2001; SCHWEIG et al., 2003). Large earthquakes are known to have been occurring in the Kachchh region since historical times (BISWAS, 1987). It has been inferred, based on the radiocarbon dating, that an earthquake occurred between A.D. 885–1035 along the Allah Bund fault (RAJENDRAN and RAJENDRAN, 1998, 2001). In 1668 a moderate earthquake occurred west of Kachchh, with an epicenter at 24oN 68oE (RASTOGI et al., 2001). The largest earthquake in the region occurred on June 6, 1819. This Mw 7.8 earthquake displaced a 100 km long ridge and created what is known as the Allah Bund (JOHNSTON, 1994; RAJENDRAN and RAJENDRAN, 2001). Between 1821–1996, 16 moderate earthquakes of magnitude varying from 4.2 to 6.1 have occurred in the

Vol. 162, 2005

Estimation of Site Response in India

2483

Figure 2 A map showing ground deformation, liquefaction sites and intensity (IX, X and X+) caused by the 2001 Bhuj main shock. Secondary features like fissures, uplift and lateral spreading are also shown.

region (RAJENDRAN and RAJENDRAN, 2001). The last damaging earthquake of Mw 6.0 (Intensity IX) prior to the most recent Mw 7.7 2001 Bhuj event occurred along the Katrol Hill fault near Anjar, Gujarat in 1956 (CHUNG and GAO, 1995). The focal mechanism determined by the USGS indicates that the devastating 2001 Bhuj earthquake occurred on the ENE-WSW trending and south-dipping reverse fault at a depth of 23 km (the India Meteorological Department issued a focal depth of 25 km from depth phases). This earthquake had a high static stress drop of 12.6 to 24.6 MPa over a small circular rupture area with a radius of 20 to 25 km for a moment of 4.5  1020 N-m (NEGISHI et al., 2001).

4. Intensity and Ground Deformation for the 2001 Bhuj Earthquake The maximum damage (intensity X+ on the MM Scale) occurred in an area of 40 km  20 km trending NNE (Fig. 2; RASTOGI et al., 2001, 2003). No house has remained standing in this area, many of which were reinforced framed structures. Surface faulting was not detected, probably due to the 23-km focal depth.

2484

P. Mandal et al.

Pure appl. geophys.,

However, intense land deformations were observed in a 40 km  20 km area including ground uplift of about a meter at Budharmora and Bhrudia (extending 200–300 m in length), ground slumping at Chobari, Khengarpar and Amarsar and deep and wide cracks (N-S at Manfara but mainly E-W at other places) (Fig. 2). Strike-slip faulting was also observed near Manfara and Sikra (RASTOGI et al., 2001). Intense liquefaction with ejection of considerable water that oozed out for months and collected in pre-existing dry channels occurred in the salt plains of the Great Rann and Banni grassland of Kachchh (Fig. 2). Important sites of liquefaction include Lodai, Umedpar, Amarsar, Chobari, between Rapar and Santalpur (23.75oN, 71.17oE), between Rapar and Dholavira and on both sides of the road for miles between Rudramata and Bhirandiala. The area between Bhirandiala and Chobari is inaccessible, nonetheless satellite images suggest evidence of liquefaction (TUTTLE et al., 2001). Thus the above-discussed observations indicate that most of the ground deformations and intense liquefaction occurred in MM intensity X and X+, located mostly on the hangingwall side of the causative thrust and some on the footwall side north of the KMF, as expected (Fig. 2). This is in accordance with the idea that the causative fault is southdipping and located north of the KMF, as will be assessed later from aftershock data. However, instances of liquefaction have occurred in the Great Rann and Little Rann of Kachchh and in coastal areas up to 200 km from the epicenter (SCHWEIG et al., 2003). These sites of liquefaction include areas around the Nagar-Parkar fault, north of Khavda and Khadir in the Great Rann, Patri (border area of Little Rann), Navlakhi, and Kandla in the coastal area (Fig. 1). Stray incident of liquefaction have occurred at distances of up to 275 km, such as one incident south of Ahmedabad and two instances near Jambusar (Bharuch District). Liquefaction at these sites is due to strong shaking in areas of intensity MM VII to IX.

5. Local Seismic Network and Data Aftershock activity of the 2001 Bhuj earthquake has been monitored by the National Geophysical Research Institute (NGRI), Hyderabad, since February 4, 2001 with a local digital network consisting of 5–8 digital 24-bit recorders with an external hard disk (2 GB) and a GPS timing system. Six of these were equipped with short-period seismometers (frequency range 1–40 Hz) and two were equipped with broadband seismometers (frequency range 0.01–40 Hz). The distances between stations and epicenters varied from 14 km to 90 km. Recording was continuous at 100 sps. Some stations were relocated during the study period making a total of 12 station sites. Further, we used phase data from the CERI (University of Memphis,

Vol. 162, 2005

Estimation of Site Response in India

2485

Figure 3 A map showing distribution of Geology and seismological stations (filled triangels) in the study area. Stations are (1) Adhoi (ADH), (2) Vondh (VND), (3) Manfara (MAN), (4) Chobari (CHB), (5) Sikara (SIK), (6) Dudhai (DUD), (7) Chopdwa (CHP), (8) Tapar (TAP), (9) Ramvav (RAM), (10) Kavada (KAV), (11) Bhachau (BAC), (12) Gadhada (GDH), (13) Bhuj (BHU), (14) Vadawa (VAD), (15) Fategarh (FTG), (16) Khingarpar (KNP), and (17) Gandhidham (GDM). Stations in the Quaternary/Tertiary sediments are defined as soft sediments sites, whereas, stations deployed in hard Jurrassic/Mesozoic sediments are defined as hardsediment sites.

USA) seismic network (from 13.02.2001 to 27.02.2001) and from the Hirosaki University, Japan seismic network (from 28.02.2001 to 7.03.2001) for locating aftershocks. This aftershock monitoring by NGRI with 5–8 digital seismic stations continued until July 2002. During August 2002 NGRI installed a dense digital network consisting of ten strong-motion accelerographs and three broadband seismographs within about a 30 km radius of Bhachau covering an area of 30 km  40 km, to obtain a better idea of peak ground accelerations. The area is spanned by latitudes 23.27–23.55oN and longitudes 70.12–70.50oE (Fig. 3). The accelerograph stations were equipped with 3-component Kinemetrics Episensors and ETNA recorder. In addition, five 24-bit REFTEK recorders with external hard disks (2 GB) and GPS timing systems were deployed to obtain precise hypocentral parameters of aftershocks. Out of these, three stations were equipped with three-component broadband CMG-40T sensors, while the remaining stations were equipped with

2486

P. Mandal et al.

Pure appl. geophys.,

Figure 4 (a) Epicenters of 600 selected aftershocks with M 2.0–5.3 from February 04 to March 7, 2001. These earthquakes were located using 3-D velocity tomography in which P- and S-phase data recorded at 8-18 seismic stations (solid triangles) were used. The solid star shows the epicenter for the 2001 Bhuj main shock. The inferred causative fault is shown by a dotted line and marked as NWF (North Wagad Fault). Hypocentral depth plots of selected earthquakes: in (b) E-W direction, (c) in N-S direction. The geologically validated inferred fault trace is shown by the dotted line and marked as NWF. Surface traces of Kachchh Mainland Fault (KMF) and South Wagad Fault (SWF) are marked by arrows (after Biswas, 1987). Subplot (d) is a histogram of the focal-depth distribution for the 2001 Bhuj event aftershocks.

three-component short period L4-3D sensors (Fig. 3). Azimuthal coverage of this present network is good with an azimuthal gap of less than 160o. The recording for accelerograph stations was done in triggered mode at 200 sps, while for seismograph stations recording was done in continuous mode at 100 sps. Through December 2003, about 200 strong motion records for 100 earthquakes have been recorded at 1 to 10 accelerograph stations.

Vol. 162, 2005

Estimation of Site Response in India

2487

Figure 5 Vertical component accelerograms for an Mw 3.9 aftershock (29 November 2002) recorded at seven accelerograph stations of the NGRI local seismological network.

2488

P. Mandal et al.

Pure appl. geophys.,

6. Velocity Inversions, Delineation and Characterization of the Causative Fault A new model for P- and S-wave velocities has been found by MANDAL et al. (2004a) based on a 1-D travel-time inversion of 5516 P-wave and 4061 S-wave travel times from 600 aftershocks (February to May, 2001) recorded at 8 to 18 threecomponent digital seismograph stations. The model consists of an 8-layers crustal model for the region. The model was constrained by results from a geophysical survey and controlled source seismics (KAILA et al., 1981; GUPTA et al., 2001). The detail of 1-D travel-time inversion can be found in MANDAL et al. (2004a). Recently, MANDAL et al. (2004b) relocated a total of 600 events using 3-D velocity tomography, which define an EW trending plane extending from 10 to 40 km depth and covering an area of 60  40 km2, which dips south at an angle of 45o (Figs. 4 a, b and c). They interpreted this south-dipping hidden fault as the North Wagad Fault representing the fault plane of the main shock (Figs. 1 and 4c). They also suggested that this fault appears to be a hidden fault beneath the Banni plains (comprised of soft sediments) that may be extending east as an unknown fault (ENWF) in the middle of the Wagad uplift as reported by Biswas (1987) (Fig. 1). The aftershock activity has been very intensive and is continuing with occasional occurrences of M3 and M4 aftershocks. The most recent M4.5 aftershock occurred on August 5, 2003, two and half years after the main shock. For the site response study we used 454 selected aftershocks of magnitude 2.5 to 4.7, which were recorded by the NGRI’s local seismic network between February 2001 and February 2003. The focal depths of selected aftershocks vary from 2.9 to 50 km. Examples of the recorded accelerograms (vertical component) for an M 3.9 aftershock are shown in Figure 5. They show a very heterogeneous distribution of ground acceleration in the aftershock zone of the 2001 Bhuj earthquake. As noted above, the Kachchh area is in the seismic zone V of India, and that zone has experienced two earthquakes of Mw ‡ 7.7 within a span of 182 years. In addition, we know that the region has experienced moderate size earthquakes since 1668. The last major earthquake of Mw 6.1 prior to the 2001 Bhuj earthquake occurred near Anjar in 1956. Such a high level of seismic activity calls for an urgent quantitative evaluation of seismic hazard for the region. A study of site response for the Kachchh region is potentially of immense importance to improve our understanding of the region’s seismic hazard. Previous studies produced estimates of site amplification at a few locations like Gandhidham, Kandla, and Anjar using micro-tremor data (SATO et al., 2001). This study is the first attempt to quantify the site response of Kachchh, Gujarat using the H/V spectral ratio method of S-wave part from the earthquake records.

Vol. 162, 2005

Estimation of Site Response in India

2489

Figure 6 Estimated site amplification (H/V spectral ratio) for 1–3 Hz and 3–10 Hz frequency bands for twelve strong motion sites. (a) Manfara (MAN), (b) Adhoi (ADH), (c) Dudhai (DUD), (d) Chobari (CHB), (e) Tapar (TAP), (f) Sikara (SIK), (g) Chopdwa (CHP), (h) Vondh (VND), (i) Ramvav (RAM), (j) Jahwarnagar (JAH), (k) New Dudhai (NDUD) and (l) Anjar (ANJ). See Figure 2 for the locations of the stations. The traces for all events recorded at each site were smoothed using a 5-point averaging technique and then logarithmically stacked (shown by thick line). The averaged values are used in Table 1.

2490

P. Mandal et al.

Figure 6 (Continued)

Pure appl. geophys.,

Vol. 162, 2005

Estimation of Site Response in India

2491

7. Single station H/V Spectral Ratio Method to Estimate Site Amplification Since there is no closely spaced hard-rock-soft-rock station pair, and no downhole array in the Kachchh, Gujarat, we have used single-station estimates of H/V ratio for this site response study. This method is quite useful in estimating the site fundamental frequency (NOGOSHI and IGARASHI, 1971; NAKAMURA, 1989). The mathematical formulation of H/V ratio method is based on the spectral ratio ðRhv Þ between the smoothed horizontal components and the smoothed vertical component: h i 1=2 ðSNS ðfÞ 2 þ SEW ðfÞÞ 2 =2:0 Rhv ¼ ð1Þ SV ðfÞ The time window of 10.24 sec starting 0.5 second before the S arrival is used for all components. This time length was chosen to best contain most of the high amplitude direct S energy. As pointed out in BONILLA et al. (1997), using longer times results in better spectral resolution at the cost of including in the spectra scattered and reflected energy, as well as surface waves. BONILLA et al (1997) and FIELD and JACOB (1995) find no statistical variations in site response computed with spectra of different time window lengths. However, CASTRO et al. (1997) suggested that S waves could be contaminated by surface waves at larger epicentral distance, which demands the use of variable time windows for the estimation of the H/V spectral ratio using S waves. Recently, BONILLA et al. (2002) drew attention to the fact that a significant site response can be associated with the vertical component resulting from S-to-P conversions at the weathered granite boundary, and that this violates the basic assumption behind the H/V method. Nevertheless, the H/V method has been a widely used method to estimate site response of many areas. The 10.24 sec. time traces for each station were detrended, 10% cosine tapered and Fourier transformed using FFT. In order to obtain a site response for each site, a resultant horizontal component was calculated for the two horizontal components and divided by the vertical component using equation (1). The resulting H/V spectra for all events recorded at one site were smoothed using a 5-point averaging technique and then logarithmically stacked. The estimated site response values at selected sites for 1–3 Hz and 3–10 Hz frequency ranges were used to prepare site response maps for the entire region. These frequency ranges were chosen because natural frequencies of multi-story buildings (3 to 10 floor) vary from 1 to 3 Hz, while the natural frequencies for 1 to 3 story buildings vary from 3 to 10 Hz.

2492

P. Mandal et al.

Pure appl. geophys.,

8. Results The estimated H/V amplifications on ground motions of twelve accelerograph sites, two broadband and six short-period sites are listed in Table 2, and the spectra are shown in Figs. 6(a–l) and 7(a–h). The average mean amplification factor obtained for soft sediment sites (Quaternary/Tertiary) vary from 0.75–6.03 times for 1–3 Hz and from 0.49–3.27 times for 3– 10 Hz (Figs. 3, 6 and 7, Table 1). The mean amplification factor obtained for hard sediment sites (Mesozoic/Jurassic) vary from 0.32–3.24 times for 1–3 Hz and 0.37– 2.18 times for 3–10 Hz (Figs. 3, 6 and 7, Table 1). The mean amplification factors for 1–3 Hz are found to be higher at Chopdwa, Vadawa, Kavada, Sikara, Vondh, Adhoi, Jahwarnagar and Gadhada (Table 1). Thus the above-mentioned site response estimates suggest that the soft sediment sites have greater seismic hazard than hard sediment sites. The sites where more devastation occurred during the 2001 Bhuj earthquake like Ramvav, Manfara, Dudhai, New Dudhai, Tapar, Sikara, Vondh, Chopadwa, Adhoi, Jahwarnagar, Gadhada, Vadawa, Gandhidham, Khingarpar, Bhachau, Anjar, Fategarh, Bhuj and Kavada, where even reinforced buildings were damaged, show higher H/V amplification (‡2.0 times) for 1–3 Hz, which could be explained by the presence of a thick low velocity sedimentary (Quaternary/Tertiary) top layer beneath these stations (Figs. 3, 6 and 7, Table 3). This site amplification could also be due to the nonlinear behavior of the soil properties at those sites. It is worth mentioning here that liquefactions, sand blows, and formation of craters took place at these sites during the 2001 main shock. Further, because 3–10 floor buildings have natural frequencies on the order of 1–3 Hz, these sites would be more hazardous for high-rise buildings. The amplification factors at 12 sites viz. Manfara, Tapar, Dudhai, Ramvav, Anjar, New Dudhai, Gandhidham, Khingarpar, Bhachau, Bhuj and Fategarh show relatively smaller site amplification values varying from 0.40 to 2.76 times for 1–3 Hz (Figs. 6 and 7). One to three story houses are the dominant building size, therefore, the estimation of amplification factors for 3–10 Hz is more relevant for the Kachchh region. For 3–10 Hz frequency range, larger mean site amplification values (‡2.0 times) are observed at Adhoi, Tapar, Chopadwa, Jahwarnagar, Khingarpar and Gandhidham (Table 1), where 1–3 storied masonary buildings as well as RC structures were observed to be collapsed or damaged during the 2001 Bhuj main shock (MURTY et al., 2002). This could be attributed to thick low velocity Tertiary sediments beneath these sites. Further, it can be inferred that these sites are more hazardous for 1–3 story buildings (Figs. 6 and 7). The amplification factors for 3– 10 Hz for the other 14 sites are found to be relatively smaller (less than 2 times) (Vadawa, Khavda, Chobari, Manfara, Dudhai, Sikara, Vondh, Ramvav, Bhachau, Gadhada, Bhuj, New Dudhai, Anjar and Fategarh) (Fig. 6 and Table 1). This variation in site amplification values could be explained by a variable thickness of the

Vol. 162, 2005

Estimation of Site Response in India

2493

Figure 7 Estimated site amplification (H/V spectral ratio) for 1–3 Hz and 3–10 Hz frequency bands for eight seismograph sites (two broadband and six short period). (a) Gandhidham (GDM), (b) Bhuj (BHU), (c) Vadawa (VAD), (d) Fategarhh (FTG), (e) Bhachau (BAC), (f) Khingarpar (KNP), (g) Gadhada (GDH) and (h) Kavada (KAV). See Figure 2 for the locations of the stations.

2494

P. Mandal et al.

Pure appl. geophys.,

Table 1 The results of the site response study for twelve strong motion and eight seismograph sites. Station

No. of records

Site geology

Site Amplification of H/V

Manfara

19

Adhoi

13

Dudhai

20

Chobari

19

Tapar

19

Sikara

15

Chopdwa

17

Vondh

16

Ramvav

16

Jahwarnagar

25

Anjar

10

New Dudhai Gandhidha m*

30 25

Khingarpur*

25

Kavada*

15

Bhachau*

20

Gadhada*

30

Bhuj*

40

Quaternary Sediments Mesozoic Sediments Quaternary Sediments Mesozoic Sediments Tertiary Sediments Quaternary Sediments Tertiary Sediments Quaternary Sediments Mesozoic Sediments Quaternary Sediments Mesozoic Sediments Mesozoic Sediments Quaternary Sediments Mesozoic Sediments Quaternary Sediments Tertiary Sediments Tertiary Sediments Hard rocks

Vadawa*

40

Fategarh*

40

0.95–2.27 times for 1–3 Hz 0.53–1.17 times for 3–10 Hz 1.80–3.24 times for 1–3 Hz 0.90–2.18 times for 3–10 Hz 1.45–2.13 times for 1–3 Hz 0.49–1.45 times for 3–10 Hz 0.32–1.00 times for 1–3 Hz 0.37–1.00 times for 3–10 Hz 1.34 – 2.76 times for 1–3 Hz 0.85 – 3.27 times for 3–10 Hz 1.17–4.16 times for 1–3 Hz 0.71–1.60 times for 3–10 Hz 3.07–6.03 times for 1–3 Hz 0.70–3.07 times for 3–10 Hz 1.16–4.19 times for 1–3 Hz 0.72–1.33 times for 3–10 Hz 1.27–2.42 times for 1–3 Hz 0.69–1.50 times for 3–10 Hz 1.00–3.13 times for 1–3 Hz 1.12 – 2.00 times for 3–10 Hz 0.40–2.29 times for 1–3 Hz 0.48–1.59 times for 3–10 Hz 1.15–2.09 times for 1–3 Hz 0.99–1.89 times for 3–10 Hz 0.75–2.30 times for 1–3 Hz 0.64–2.36 times for 3–10 Hz 1.17–2.34 times for 1–3 Hz 1.12–2.00 times for 3 –10 Hz 1.18–4.47 times for 1–3 Hz 0.48–1.59 times for 3–10 Hz 1.15–2.08 times for 1–3 Hz 0.86–1.22 times for 3–10 Hz 1.49–3.67 times for 1–3 Hz 1.30–1.71 times for 3–10 Hz 1.00 – 2.00 times for 1–3 Hz 0.70–1.40 times for 3–10 Hz 0.90–4.59 times for 1–3 Hz 1.09–1.95 times for 3–10 Hz 1.15–2.39 times for 1–3 Hz 0.86–1.76 times for 3–10 Hz

Quaternary Sediments Mesozoic Sediments

*Seismograph sites

top sedimentary (Quaternary/Tertiary) layer as well as the nonlinear behavior of soil properties beneath the whole study area. Interestingly, H/V amplification at Chobari shows a maximum value of 1.01 for the whole frequency range of interest (1–10 Hz). However, the entire Chobari village

Vol. 162, 2005

Estimation of Site Response in India

2495

was devastated during the 2001 Bhuj earthquake, perhaps due to poor building construction. The influence of deeper focal depths of selected events might also have contributed to the dispersion in the site response estimates at different stations (Table 1). Several high-rise reinforced concrete (RC) buildings (4 – 6 storied RC structures) were damaged or collapsed at Bhuj, Gandhidham, Adipur and Bhachau during the 2001 Bhuj main shock (PAREEK et al., 2002). Further, a detailed microtremor study to investigate reinforced concrete buildings damaged in the 2001 Bhuj main shock was carried out by KONO and TANAKA (2001). They estimated the fundamental periods of all five buildings by microtremor measurements, which were done at the top of the buildings using an accelerograph with a 20.48 sec. natural period. The recorded accelerograms were then Fourier transformed and smoothed using a Parzen window with a 0.5 Hz bandwidth. Finally, three spectra were averaged and the fundamental period was obtained from the average spectra. This procedure was carried out on both longitudinal and the transverse directions independently. The estimated fundamental periods are listed in Table 2. The estimated fundamental periods for three (4, 5 and 6 story) investigated RC buildings of the Bhuj area are estimated to be 0.17 s (5.9 Hz), 0.41 s (2.4 Hz), and 0.53 s (1.9 Hz), respectively. Finally, KONO and TANAKA (2001), based on the H/V spectra from a microtremor study, suggested that the soil condition at Gandhidham could not account for the concentration of building damage. Nevertheless, it is interesting to note that our site amplification values at Bhuj are found to be 1.00– 1.90 for 1–3 Hz and 0.70–1.40 for 3–10 Hz. These site response values perhaps indicate that the Bhuj site nearly shows the behavior of a hard-rock site. However, the investigated five-story RC building at Gandhidham implies a fundamental period of 0.38 s (2.6 Hz) during which estimated site amplification value is found to be 0.75–2.30 for 1–3 Hz. Therefore, our estimated H/V spectra suggest amplification at approximately the fundamental frequencies of damaged RC buildings at Bhuj as well as Gandhidham. It is relevant to discuss here that a total death toll of 18,403 was reported from Kachchh alone. The death toll distribution (in %) for the most affected talukas (in India, each state is divided into several districts and each district is further subdivided into subdivisions or talukas) is listed in Table 3 along with the percentage of collapsed Pucca (made of bricks/stones and cements) and Kuchcha (made up of stones and mud/mortar) houses during the 2001 Bhuj earthquake (MURAKAMI, 2001). Figures. 8 (a–c) as well as Table 3 suggest a maximum death toll for Bhachau Taluka (6.47%) where our estimated upper bound for the site response values at Sikara, Chopadwa, Vondh, Adhoi, Chobari, Manfara and Bhachau are found to vary from 1.01–6.03 for 1–3 Hz and 1.00–3.07 for 3–10 Hz. The second highest death toll (2.93%) was recorded in Anjar Taluka where our estimated upper bound for the site response values at Anjar, Jawaharnagar, New Dudhai, Dudhai, Vadawa and

2496

P. Mandal et al.

Pure appl. geophys.,

Table 2 Results of microtremor study of investigated RC buildings (KONO and TANAKA, 2001) Building name

Location

No. of stories

Height (m)

Fundamental Periods (Frequency) Longitudinal

Prince Hotel Limdiwaca terrace NK Tower Pooja Flat Classic Complex

Bhuj Bhuj Bhuj Anjar Gandhidham

4 5 6 6 5

10.0 15.4 21.6 18.0 14.7

0.17 0.29 0.54 0.59 0.38

s s s s s

(5.9 (3.5 (1.9 (1.7 (2.6

Hz) Hz) Hz) Hz) Hz)

Transverse 0.17 0.41 0.53 0.44 0.36

s s s s s

(5.9 (2.4 (1.9 (2.3 (2.8

Hz) Hz) Hz) Hz) Hz)

Tapar was found to be 2.09–4.59 for 1–3 Hz and 1.45–3.27 for 3–10 Hz. The third highest death toll (1.62%) was in Bhuj Taluka where our estimated upper bound for the site response values at Bhuj and Khingarpur ranged from 2.00–2.34 for 1–3 Hz and 1.40–2.00 for 3–10 Hz. The death toll in Gandhidham Taluka was 0.82% where our site amplification values at Gandhidham were estimated to be 0.75–2.30 for 1–3 Hz and 0.64–2.36 for 3–10 Hz. However, the death toll in Rapar taluka was reported to be 0.49% where the upper bound of the estimated site amplification values at Ramvav are found to be 2.42 for 1–3 Hz and 1.50 for 3–10 Hz. Figure 8 shows a good correlation between taluka-wise death toll and the upper bound of the estimated mean site amplification for 1–3 Hz as well as 3–10 Hz, except for the Gandhidham and Rapar talukas where estimated site response values are greater but the percentage of dead is smaller (0.82% and 0.49%). This could be explained in terms of better construction in Gandhidham taluka/subdivision and hard rocks/ sediments of Wagad uplift in Rapar taluka/subdivision. Further, it is important to note that in the above-discussed talukas in the Kachchh district most of the collapsed Pukka buildings were 2–4 stories (natural frequencies 2.5–5 Hz) and collapsed Kuchchh buildings were 1–2 stories (natural frequencies 5–10 Hz).These taluka-wise percentages of death toll have also been shown in the Figures. 9 (a and b), which suggests the maximum death toll and collapse of houses in a zone of maximum site amplification. Thus, it can be inferred that the amplification of ground motion in

Table 3 Taluka/subdivison-wise death toll in the most affected areas of Kachchh Taluka (Subdivison) Name

1991 Population

Number of dead

Dead %

Collapsed Pucca %

Collapsed Kuchcha %

Bhuj Rapar Bhachau Anjar Gandhidham

277215 150517 114759 160640 104585

4503 732 7424 4702 861

1.62 0.49 6.47 2.93 0.82

57 69 95 77 45

67 91 95 81 54

Vol. 162, 2005

Estimation of Site Response in India

2497

Figure 8 A histrogram plot showing the correlation between the taluka-wise dead percentage due to the 2001 Bhuj main shock and the upper bound of the estimated site response values for 1–3 Hz and for 3–10 Hz in four talukas of Kachchh, Gujarat, India (a) taluka-wise dead percentage, (b) taluka-wise upper bound of the estimated site amplification values at 1–3 Hz and (c) taluka-wise upper bound of the estimated site amplification values at 3–10 Hz.

2498

P. Mandal et al.

Pure appl. geophys.,

Bhachau, Anjar and Bhuj might have contributed significantly to the damage observed during the 2001 Bhuj earthquake. However, a smaller percentage of dead in spite of larger site amplification could be attributed to better construction in Gandhidham taluka and to hard rocks/sediments related to the Wagad uplift in Rapar taluka. Preliminary site response maps of Kachchh, Gujarat have been prepared for two frequency ranges 1–3 Hz and 3–10 Hz (Figs. 9a, b). Figure. 9(a) shows two zones of large site amplification (more than 3 times) for 1–3 Hz. Zone I is in the NW part of the study area covering the worst affected areas such as Jahwarnagar, Kavada, Khingarpur and Gadhada (Fig. 9a). Further, zone II is in the SE part of the study area covering highly damaged areas like Sikara, Chopadwa, Bhachau, Vondh and Adhoi. The large site amplification in these zones I and II could be attributed to the thick (2–3 km) low velocity Quaternary/Mesozoic sediments. It is relevant to mention here that these two zones I and II of large amplification show a good correlation with the highest intensity zone of X+ where uplift, liquefaction, lateral spreading and ground fissures were observed at Budharmora, Amarsar, Lunwa, Chopadwa and Sikara during the occurrence of the main shock (Fig. 9a). However, H/V amplification contours suggest site amplification less than three times in the NE and SW parts of the study area that covers areas Chobari, Manfara, Ramvav, Fategarh, Tapar, Dudahai, Anjar, Bhuj and Gandhidham (Fig. 9a). This distribution of site amplification can explain the relatively greater damage to the engineered structures in the SW, SE and NW parts of the study area in comparison to the NE region. Contours of estimated site amplification values for 3–10 Hz reveal two patches of large site amplification (more than 2) as shown in Figure. 9(b). Patch -I covers most of the western section of the study area (covering Jahwarnagar, Kavada, Dudhai, Chobari, Tapar, Gandhidham and Chopadwa) except small areas around Bhuj and Anjar. Patch II covers most of the SE portion of the study area covering Sikara, Bhachau, and Adhoi. It is important to note that both patches lie within the zone of intensity X where the majority of two to three story buildings collapsed. Further, the liquefaction, lateral spreading and ground fissures were observed at Kadol, Chobari, Bhurudia, falling in the above-mentioned patch-I. Nevertheless, site amplification contours for 3–10 Hz also show three patches of relatively smaller amplification (less than 2 times); two of them are small patches near Bhuj and Anjar (Fig. 9b). The third is a big patch in the NE part of the study area covering Manfara, Ramvav, Fategarh and Vondh. The variation in site response at these stations can be explained in terms of the variable thickness of top low velocity Quaternary/Mesozoic sediments. The smaller site amplification values (£3) for 1–3 Hz are estimated at Fategarh, Gadhada, Ramvav, Manfara, and Chobari, which could be attributed to the smaller low velocity Quaternary/Tertiary sediment thicknesses. Thus, it can be inferred that the local large amplification for 3–10 Hz could be explained in terms of the thick sedimentary top layer beneath those sites.

Vol. 162, 2005

Estimation of Site Response in India

2499

Figure 9 Estimated site response maps (a) for 1–3 Hz and (b) 3–10 Hz. Elliptical zones marked by dotted lines represent intensity X and X+ areas. Liquefaction sites characterized by sand blows caused by the 2001 Bhuj earthquake are shown. Taluka-wise percentages of deaths are also marked. Seismological stations are marked by plus symbols.

Sand blows were the most common liquefaction features observed in the meizoseismal area of the 2001 Bhuj earthquake. Sand blows are constructional cones of mostly sand vented with water to the ground surface through ground cracks. Sand blows resulting from the 2001 Bhuj earthquake range from tens of centimeters to tens of meters in length and up to tens of centimeters in thickness are shown in Figures. 9 (a) and (b). Figure. 9(a) shows only four liquefaction sites characterized by sand blows in the zone of larger amplification (‡3 times) for 1–3 Hz, whereas all liquefaction sites characterized by sand blows are confined to the zones of large amplification (‡2 times) for 3–10 Hz. Thus, the correlation between the confinement

2500

P. Mandal et al.

Pure appl. geophys.,

of liquefaction sites with sand blows and the large amplification for 3–10 Hz could be attributed to the presence of a top sedimentary (Quaternary/Tertiary) layer beneath the zone. We compare our estimated site amplification values with other studies which are listed in Table 4. A comparison between different site response studies suggests that our estimates of site amplification values favorably agree with the estimates of the San Fransisco Bay, USA (BORCHERT and GIBBS, 1976). Table 4 suggests that the site response values at 1 Hz estimated for the (Bay mud, alluvium) of San Francisco Bay, USA (BORCHERT and GIBBS, 1976) and (Quaternary, Holocene) sediments of California, USA (BONILLA et al., 1997) are (11, 3.9) and (3.8, 4.22), respectively, which are in good agreement with our estimates for Quaternary/ Tertiary sediments of Banni area, Kachchh, Gujarat show better correlation with our estimates for the Bhuj area, Gujarat, India. However, the site response estimates for Kanto and Tokyo, Japan show no correlation with our estimates (MIDORIKAWA, 1987).

9. Summary and Conclusions A preliminary site response study using the H/V spectral ratio method has been estimated using twelve strong motion sites and eight seismograph sites. A heterogeneous distribution of mean H/V amplification in the aftershock zone is observed. The mean amplification factors obtained for soft sediment sites (Quaternary/Tertiary) vary from 0.75–6.03 times for 1–3 Hz and 0.49–3.27 times for 3– 10 Hz. The mean amplification factors obtained for hard sediment sites (hard Jurassic/Mesozoic sediments) range from 0.32–3.24 times for 1–3 Hz and 0.37–2.18 times for 3–10 Hz. The upper bounds of the larger mean amplification factors for 1– 3 Hz are found to be of the order of 3.13–6.03 at Chopadwa, Vadawa, Kavada, Vondh, Adhoi, Jahwarnagar and Gadhada. For 3–10 Hz frequency range, larger mean site amplification values (‡2.0 times) are observed at Adhoi (0.90–2.18 times), Tapar (1.85–3.27 times), Chopadwa (0.70–3.07 times), Jahwarnagar (1.22–2.00 times), Khingarpar (1.12–2.00 times) and Gandhidham (0.64–2.36 times). The upper bounds of the higher mean amplification factors for 3–10 Hz are estimated to be of the order of 2.00–3.27 at Tapar, Chopadwa, Adhoi, Jahwarnagar, Gandhidham and Khingarpar. These larger site amplification values for 3–10 Hz could be attributed to the large thickness of the top sedimentary (Quaternary/Tertiary) layer as well as the nonlinear behavior of soil properties at these sites. Preliminary site response maps for 1–3 Hz and 3–10 Hz frequency ranges have been prepared. The 1–3 Hz map delineats two distinct zones of maximum site amplification (>3 times): one lying in the NW quadrant of the study area covering Jahwarnagar, Kavada and Gadadha and the other in the SE quadrant of the study area with a peak of 6.03 at Chopadwa covering an area of 70 km  50 km. These two zones suffered maximum damage to

Vol. 162, 2005

Estimation of Site Response in India

2501

Table 4 Correlations between different surface geology and relative amplification factor Geological Unit

Relative amplification factor

Methods / data used

BORCHERT and GIBBS (1976):

0.25–3.0 Hz

Average horizontal spectral amplifications of ground motion recordings (with reference to a bedrock site) by nuclear explosion and earthquake in Neveda, USA

Bay mud Alluvium Santa Clara Formation Great Valley sequence Franciscan Formation

11.2 3.9 2.7 2.3 1.6

Granite

1.0

MIDORIKAWA (1987)

1.0–4.0 Hz

Holocene sediments Pleistocene sediments Quaternary volcanic rocks Miocene sediments Pre-Tertiary rocks

3.0 2.1 1.6 1.5 1.0

BONILLA et al. (1997) Mesozoic tertiary Pleistocene Alluvium Quaternary Holocene

1 Hz 3 Hz 10 Hz 1.01 1.00 1.01 2.27 2.28 2.09 2.42 2.00 1.59 3.80 2.23 1.54 4.22 2.51 1.77

Site effect estimation based using direct spectral ratio, inversion of S-wave spectra, coda waves and receiver transfer function analysis reference hard rock site using aftershock data of 1994 Northridge, California earthquake

This Study:

0.75–6.03 for 1–3 Hz and

H/V spectral ratio of S-wave part of 2001 Bhuj aftershock recordings.

Quaternary/ Tertiary Mesozoic/Jurrassic

0.49–3.27 for 3–10 Hz 0.32–3.24 for 1–3 Hz and 0.37–2.18 for 3–10 Hz

Spectral amplification of microtremor recordings of Tokyo, Japan

reinforced multi-storied buildings during the 2001 Bhuj earthquake. While the 3– 10 Hz map shows more than two times site amplification value over the entire study area except the NE quadrant, two patches in the southwest corner covering Bhuj and Anjar, and one patch at the center covering Vondh, Manfara and Sikara. The zones

2502

P. Mandal et al.

Pure appl. geophys.,

for large site amplification values (>3 times) are found at Tapar, Chopadwa, Adhoi and Chobari, where maximum damage to one and two story buildings occurred in the 2001 Bhuj earthquake. Consequently, it can be inferred that our estimated site response distributions are in good agreement with the geology of the area. The variation in site response at these stations is probably correlated with the geologic and soil formation on which the seismic stations are located. We would further refine the ground amplification map by using H/V amplification estimates at more sites. The site response estimated at Bhuj suggests a typical hard-rock site behavior. The Bhuj station was in operation for a period of a few months, and during that time no other station operated in the region. Hence it was not popular to use the reference-station method to estimate the site response (e.g., BONILLA et al., 1997). Having additional future stations in the Bhuj area will allow us to obtain better estimates of site response using the reference site method.

Acknowledgements The authors are thankful to Dr. V. P. Dimri, Director, NGRI for his encouragement and kind permission to publish this work. This study was supported by the Department of Science and Technology, New Delhi. We are grateful to CERI, Memphis, USA and Hirosaki University of Japan for providing their aftershock data.

REFERENCES BARD, P. Y., DUVAL. A. M., LEBURN, B., LACHET, C., RIEPL, J. and HARTZFELD, D. (1997), Reliability of the H/V technique for site effects measurement: An experimental assessment. In Seventh International Conference on Soil Dynamics and Earthquake Engineering, July 19–24, Istanbul, Turkey. BENDICK, R., BILHAM, R., FIELDING, E., GAUR, V. K., HOUGH, S. E., KIER, G., KULKARNI, M. N., MARTIN, S., MUELLER, K. and MUKUL, M. (2001), The 26 January 2001 ‘‘Republic Day’’ Earthquake, India, Seismol. Res. Letts. 72, 328–335. BISWAS, S. K. (1987), Regional Framework, Structure and Evolution of the Western Marginal Basins of India Tectonophysics 135, 302–327. BISWAS, S.K. and DESHPANDE, S.V. (1970), Geological and Tectonic Maps of Kachchh, Bull. Oil and Natural Gas Comm. 7, 115–116. BODIN, P., HORTON, S., JOHNSTON, A. C., WITHERS, M., LANGSTON, C., CHIU, J. M., BUDHBATTI, K. and GOMBERG, J. (2001), Aftershocks of the Gujarat, India Republic Day Earthquake, Seismol. Res. Letts. 72, 397. BONILLA, L. F., STEIDL, J. H., LINDLEY, G. T., TUMARKIN, A.G. and ARCHULETA, R. J. (1997), Site Amplification in the San Fernando Valley, California; Variability of Site-effect Estimation Using the S-wave, Coda, and H/V Methods, Bull. Seismol. Soc. Am. 87(3), 710–730. BONILLA, L. F., STEIDL, J. H., LINDLEY, G. T., TUMARKIN, A.G. and ARCHULETA, R. J. (2002), Borehole Response Studies at the Garner valley downhole Array, Southern California, Bull. Seismol. Soc. Am. 92(8), 3165–3179. BORCHERDT, R. D. and GIBBS, J. F. (1976), Effects of Local Geological Conditions in the San Francisco Bay Region on Ground Motions and the Iintensities of the 1906 Earthquake, Bull. Seismol. Soc. Am. 66, 467– 500.

Vol. 162, 2005

Estimation of Site Response in India

2503

BUREAU OF INDIAN STANDARDS, Criteria for Earthquake Resistant Design of Structures (Fifth Revision 2002), pp. 39. CASTRO, R. R., MUCCIARELLI, M., PACOR, F. and PETRUNGARO, C. (1997), S-wave Site Response Using Horizontal –to- vertical Spectral Ratios, Bull. Seismol. Soc. Am., 87(1), 256–260. CHAVEZ-GARCIA, F . J., SANCHEZ, L. R. and HATZFELD, D. (1996), Topographic Site Effects and HVSR: A Comparison between Observations and Theory, Bull. Seism. Soc. Am., 86(5), 1559–1573. CHUNG, W-Y. and GAO, H. (1995), Source Parameters of the Anjar Earthquake of July 21,1956, India and its Seismotectonic Implications for the Kachchh rift Basin, Tectonophysics 242, 281–292. COUTEL, F. and MORA, P. (1998), Simulation–based Comparison of four-site Response Estimation Techniques, B.S.S.A, 87(1), 30–42. FIELD, E. H. and JACOB, K. (1995), A Comparison and Test of Various Site Response Estimation Techniques, Including Three that are Non-reference-site Dependent, Bull. Seismol. Soc. Am., 85, 1127–1143. GUPTA, H. K., T. HARINARAYANA, M. KOUSALYA, D.C. MISHRA, I. MOHAN, N. PURNACHANDRA RAO, P. S. RAJU, B. K. RASTOGI, P. R. REDDY and DIPANKAR SARKAR (2001), Bhuj earthquake of 26 January, J. Geolog. Soc. In., 57, 275–278. JOHNSTON, A. C. Seismotectonic interpretations and conclusions from the stable continental regions, In The Earthquakes of Stable Continental Regions: Assessment of Large Earthquake Potential Electric Power and Research Institute, Palo Alto, (1994) Report TR 10261, ch. 3. KAILA, K. L., TEWARI, H. C. and SARMA, P. L .N. Crustal Structure from deep seismic sounding studies along Navibandar - Amreli profile in Saurashtra, India. In Deccan Volcanism and Related Basalt Provinces in other Parts of the World. Memoir.3, Geol. Soc. India., (1981) pp. 218–232. KONNO, K. and OHMACHI, T. (1998), Ground-motion Characteristics Estimated from Spectral Ration between Horizontal and Vertical Components of Microtremor, Bull. Seismol. Soc. Am., 88(1), 228– 241. KONO, S. and TANAKA, H.(2001) Damage of reinforced concrete structures. In A Comprehensive Survey of the 26 January 2001 Earthquake (Mw 7.7) in the State of Gujarat, India, Research Report on Natural Disasters, Report submitted by Japanese Team, Sato et al., pp. 75–83. LERMO, J. and CHAVEZ-GARCIA, F.J. (1993), Site Effect Evaluation Using Spectral Ratios with only one Station, Bull. Seismol. Soc. Am. 83, 1574–1594. MANDAL, P., RASTOGI, B.K., SATYNARAYANA, H.V.S., KOUSALYA, M., VIJAYRAGHAVAN, R., SATYAMURTY, C., RAJU, I. P., SARMA, A. N. S. and KUMAR, N. (2004a), Characterization of the Causative FaultSsystem for the 2001 Bhuj Earthquake of Mw 7.7, 378, 105–121. MANDAL, PRANTIK, RASTOGI, B. K., SATYANARAYANA, H. V. S. and KOUSALYA, M (2004b), Results from Local Earthquake Velocity Tomography: Implications toward the Source Process Involved in Generating the 2001 Bhuj Earthquake in the Lower Crust beneath Kachchh (India), Bull. Seismol. Soc. Am. 94(2), 633–649. MIDORIKAWA, S. (1987), Prediction of Isoseismal Map in the Kanto Plain due to Hypothetical Earthquake, J. Struc. Eng., 33B, 43–48. MURAKAMI, H. (2001), Outline of damage survey. In A Comprehensive Survey of the 26 January 2001 Earthquake (Mw 7.7) in the State of Gujarat, India, Research Report on Natural Disasters, Report submitted by Japanese Team, Sato et al., pp. 51–55. MURTY, C.V.R., GOEL, R.K., gOYAL, A., JAIN, S. K., SINHA, R., RAI, D.C., ARLEKAR, J.N. and METZGER, R. (2002), Reinforced concrete structures. In, Bhuj, India earthquake of January 26, 2001 Reconnaissance Report, (eds) by S.K. jain et al., 18, 149–185. NAKAMURA, Y. (1989), A Method for Dynamic Characteristics Estimation of Subsurface Using Microtremor on the Ground Surface. QR of R.T.R., 30–1. NEGISHI, H., MORI, J., Sato, H., Singh, R.P., and Kumar, S. (2001), Aftershocks and slip distribution of Mainshock, A comprehensive Survey of the 26 January 2001 earthquake (Mw 7.7) in the state of Gujarat, India, Research Report on Natural Disasters, December 2001, pp. 33–45. NOGOSHI, M. and IGARASHI, T. (1971), On the Amplitude Characteristics of Microtremor (Part 2), J. Seismol. Soc. Japan, 24, 26–40 (in Japanese with English abstract). PAREEK, S., HAYASHI, Y., and SAWADA, S. (2002), Building damage in Gandhidham during the Bhuj earthquake of 2001. In: Earthquake Engineering, 12 SEE – 2002, (eds. Paul et al.,) V. II, 727–744.

2504

P. Mandal et al.

Pure appl. geophys.,

RAJENDRAN, C.P., and RAJENDRAN, K. (2001), Character of Deformation and Past Seismicity Associated with the 1819 Kachchh Earthquake, Northwestern India, Bull. Seismol. Soc. Am. 91(3), 407–426. RASTOGI, B. K., GUPTA H. K., MANDAL P., SATYANARAYANA H.V.S., KOUSALYA M., RAGHAVAN R.V., RICHA JAIN, SARMA ANS , KUMAR N., and SATYAMURTHY C. (2001), The Deadliest Stable Continental Region Earthquake that Occurred near Bhuj on 26 January 2001, J. Seism. 5, 609–615. RASTOGI, B. K., PRANTIK MANDAL and C. SATYAMURTY (2003), Intensity and aftershock studies of the 2001 Bhuj earthquake of Mw 7.7, Proc. 12 SEE, Roorkee, Vol.I, pp. 3–11. REDDY, P. R., SARKAR, D., SAIN, K. and MOONEY, W. D. (2001), A Report on a Collaborative Scientific Study at USGS, Menlo Park, USA (30 October – 32 Decemebr, 2001), 19pp. RIEPL, J., BARD, P.-Y., HATZFELD, D., PAPAIOANNOU, C. and NECHTSCHEIN, S. (1998), Detailed Evaluation of Site Response Estimation Methods across and along the Sedimentary valley of Volvi (EUROSEISTEST), Bull. Seismol. Soc. Am., 88(2), 488–502. *SCHWEIG, E., GOMBERG, J., PETERSEN, M., ELLIS, M., BODIN, P., MAYROSE, L. and RASTOGI, B. K. (2003), The M 7.7 Bhuj Earthquake: Global Lessons for Earthquake Hazard in Intra-plate Regions, J. Geol. Soc. In. 61(3), 277–282. SRIDEVI, J. MUKUL, M., PARVEZ, I .A., ANANDA, M. B., KUMAR, P. D. and GAUR, V. K. (2001), Estimates of Co-seismic Displacement andPost-seismic Deformation Using Global Positioning System Geodesy for the Bhuj Earthquake of 26 January, 2001, Current Science, 82, 748–752. TUTTLE, M. P., JOHNSTON, A., RAJENDRAN, C.P., RAJENDRAN, K. and THAKKAR, M. (2001), Liquefaction Features Induced by the Republic Day Earthquake, and Comparison with Features Related to the 1811–12 New Madrid Earthquakes, Am. Geophys. Union, Transactions, EOS, Vol. 82. ZARE, M., BARD, P.-Y., GHAFORY-ASHTIANY, M. (1999), Site Categorization for the Iranian Strong Motion Network, Soil Dyn. and Earthq. Eng. 18, 101–123. (Received March 22, 2004; accepted December 21, 2004) Published Online First: October 7, 2005

To access this journal online: http://www.birkhauser.ch