Jabalpur earthquake: Assessing the causes and damages using remote sensing and GIS

Dr. Arun K. Saraf
Department of Earth Sciences, University of Roorkee
Roorkee – 247667
Emai: [email protected]

ABSTRACT
Jabalpur earthquake of 22nd May, 1997 caused considerable damage to life and property in Central India. Beside killing about 150, injuring about 400 people and damaging about 40 thousand houses worth billions of rupees, almost every house of Jabalpur city and surrounding village developed cracks. Jabalpur earthquake and Latur earthquake of 30th September, 1993 and several other similar events recorded in last three decades clearly demonstrate that Indian Peninsular Shield region is not so stable as was considered earlier. Recently, various studies have brought out that Indian Peninsular Shield region concealed under the Deccan Trap cover is tectonically and structurally dvisible into large number of faulted fragments which have been differentially moving relative to each other throughout Tertiary, Quaternary and Recent times. This paper attempts to explain the role of Narmada Son fault causing Jabalpur earthquake of 22nd may, 1997 (magnitude 6.1) and assessed damaged induced by this earthquake using IRS-PA pre. and post-earthquake data sets and Geographic Information System (GIS). A pseudo colour transform (PCT) approach developed for earthquake damage assessment has successfully been applied on IRS-PAN pre and post earthquake data sets to accurately identify the damages to houses and other structures. The technique developed to assess earthquake induced damage using remote sensing and GIS technologies is cost effective unbiased free from subjectivity, time saving and provides quantitative damage assessment. Using presented approach maximum relief and planning ca be provided in minimum time.

1.0 Introduction
Recent earthquake of Jabalpur (22nd May, 1997) and Latur (30 September 1993) have caused immense losses of human life and damages to house and other structures. These recent events have also shaken the long held views that the Indian Peninsualr Shield region is stable and free from such events. The recent earthquakes surprised seismologists and earth-scientists for the region which was considered generally non-seismic as very few historically significant earthquakes occurred. However, tremors were reported and felt by the people in this part of the country on several occasions in last three decades without arousing he enough concern of the scientific community and triggering scientific studies to understand the factors responsible for these rumblings and tremors (Ravi Shanker, 1995). The earthquakes at Koyna in 1967, Bhadrachalam in 1969 and Latur in 1993 showed that the Indian Peninsular Shield is not as stable as it was believed. The Jabalpur earthquake has further reinforced the view of seismic instability of the Indian Peninsular Shield.

The Indian Peninsular shield is know for moderate inraplate seismicity and several earthquakes of magnitude 5 to 6.5 have been recorded from this region. The major earthquake of magnitude 6.5 that rocked the terrain occupied Deccan Traps occurred in December, 1967 near Koyna Dam prior to Latur earthquake of 1993. This event was attributed to reservoir induced seismicity by many workers. All these events were associated with well defined tectonic surfaces/domains which were locally assigned higher seismicity hazard levels in the Seismic Zoning Map of India (Narula, 1995). The area which has struck by the present event has been described as a region of moderate seismicity (Seismic Zone-III) in the Seismic Zoning Map of India (figure 2). According to ISI Code 4326-1976, 1982 this zone could experience shocks of magnitude 6.5 occasionally by adjustments along local faults albeit with longer return periods. The Indian Peninsular Shield also known for swarm type of activity which occurs sporadically and decays with time. Such swarms occurred around this area also, for example tremors near Jablpur on 31st October, 1993 and near Khandwa in 1993 (Figure 2). It has been reported that before this earthquake a number of smaller events took place during this century in Narmada valley (Table 1).

The major burst of this earthquake effect has been borne by about 150 surrounding villages and several colonies of Jabalpur city which includes Kosamghat Village where all houses (about 150) were totally damaged and epicentre of the earthquake was located. A miximum intensity of VIII on MM scale was observed in Jabalpur and surrounding region (Gupta et al., 1997, Narayan and Kumar, 1997) (Fig. 4).

The present study explains a very important role played by Narmada-Son fault causing Jabalpur earthquake and assesses damages to houses and other structures with the integrated application of remote sensing and GIS techniques. Details of several earthquakes occured during this century in Namrata Valley

2.0 Geotectonic setup of the study area
The detailed and integraed goetectonic and geothermal studies carried out by Ravi Shanker (1995) in Central and Northern parts of the Indian Peninsular Shield region have established that the region is tectonically / structurally divisible into large number of longitudinal fault bound blocks, offset by very prominent transverse faults with periodic history of pronounced vertical and lateral movements all through post-Deccan Trap nperiod upto the Recent times. Relative vertical movement upto one kilometre and lateral movement upto thirty kilometres in Quanternary times has laready been established in 150 to 200 km wide ‘SONATA ZONE (Figure 3) (Ravi Shanker, 1995).

3.0 Data used and methodology
IRS-PAN pre- and post-earthquake remote sensing digital data sets of two different dates have been used to study the Jabalpur earthquake induced changes (Table 2). Pre-earthquake IRS-LISS-III digital were also used to understand structural framework of the study area. Structure contour map at the bottom of Deccan trap, which depicts fragmented nature of part of Deccan shield prepared by Ravi Shanker (1995) has been used along with the USGS Moment Tensor Solution obtained from Internet to identify the causes of Jabalpur earthquake. Various GIS softwares (Arc View 3.0 and IL WIS 3.1) wee also used to prepare various overlays and epicentre data (obtained from USGS) plots. Digital image processing of IRS-PAN and LISS-III datasets were performed mainly in MEG Advance Imager. Two graphical packages (Corel Draw 7.0 and Photo Styler 2.0) were also employed during the preparation of final figures.

The conventional Pseudo Colour Transformation (PCT) techniques facilitates to combines various bands in different colour schemes, such as RGB, HIS & CMY. However, instead of using the different bands of same date data, a new approach has been adopted in the present study to detect the earthquake induced changes in IRS-PAN pre and post-earthquake data sets. Both pre- and post-earthquake IRS-PAN scenes were georeferenced using image-to-image rectification technique. Since IRS-PAN sensor provides only one band information hence post-earthquake scene kept in red channel and pre-earthquake scene is green and blue channels. This has provided a change index depicting red colour for positive change in terms of reflectance between two dates and blue & green colours for negative change, whereas black & white colours for no change 

4.0 The causes of recent earthquake of 22nd May, 1997
The differential and complex zig-zag movement of these block in the fragmented Indian Peninsuarl Shield during Quaternary and Recent times have been producing compressions and tensions at shallow crustal levels in different segments. Such stress accumulations are the prime causes of observable evidences of neotectonism and contemporary tectonism. The eventual release of these accumulated compressive and tensional stresses produce earthquakes of varying magnitude, factual depths and focal mechanism in different segments (Ravei Shanker, 1995).

Jabalpur is located in Narmada-son fault (F11 in figure 3) and surrounded by various faults (F8, F9 and F10) (numbering scheme of all faults mentioned here has been kept same as given by Ravi Shanker, 1995). The recent movement along NE-SW trending Narmada-Son fault is the prime cause of triggering the Jabalpur earthquake and subsequent propagation of energy. The NE-SW trends of isoseismals obtained during the recent post-earthquake surveys (Narayan and Kumar, 1997) are primarily controlled by the propagation of energy through NE-Sw fault system. Figure 3 shows the location of various subsurface faults (after Ravi Shanker, 1995) which have been plotted by directly projecting on the surface.

USGS (May 1997) estimates yield the coordinates of the epicentre as : 80.021 0E; 23.1770 N and the depth of the focus as 32 km. The location lies about 20 km NNW of the surface trace of the Narmada-Son fault (Figure 7), which dips southward (710 from north) at

5.0 Conclusions
It is true, that earthquakes can and will continue to cause widespread social and economic destruction. Earthquakes, cannot be prevented. We cannot completely eliminate losses from earthquakes. We cannot promise that earthquakes can be predicted. We can, however, assure that the continuation of efforts will improve our understanding of earthquakes, their mode of occurrence and the motions they produce (Simpson, 1997). Present study successfully demonstrated that integrated remote sensing and GIS technique can provide an useful tool in earthquake studies. Remote sensing based change detection technique can assess damages to houses and other structures accurately nd speedily. The study also identified the important role played by the Narmada-son fault causing Jabalpur earthquake of 22nd May, 1997.

The technique developed to assess earthquake induced damage using remote sensing and GIS technologies is cost effective, unbiased, free from subjectivity, time saving and provides quantitative damage assessment. Using presented approach maximum relief nd planning can be provided in minimum time.

References:

• Ravi Shanker (1995), Fragmented Indian Shield and Recent Earthquakes, Geolgical Survey of India Special Publication No. 27 (A collection of papers presented at the workshop on “30th September, 1993 Killari Earthquake, Maharastra” held at Hyderabad on 24th December, 1993, Hyderabad), pp 41-48
• Gupta, R.P., A. K. Saraf, P. Saxena and R. Chander (1994), IRS Detection of surface effects of the Uttarkashi earthquake of 20 October,1991, Himalaya, International Journal of Remote Sensing Vol. 15 No. 11, pp 2153-2156
• Gupta, H. K., R. K. chadha, M. N. Rao, B. I. Narayana, P. Mandal, M. Ravi Kumar and N. Kumar, (1997), the Jabalpur Earthquake of May 22, 1997, Journal Geological Society of India vol 50, July 1997, pp85-91
• Narayan, J. P. and Ashwni Kumar (1997), field observations and intensity distribution related to Jabalpur earthquake of May 22, 1997, Proceedings of Workshop on Earthquake Disaster Preparednes, October 13-14, 1997, University of Roorkee, Roorkee, pp 219-228
• Narula (1995), Geotechnical Investigations of Killari (Latur) Earthquake of 30th September, 1993 – An overview, Geological Survey of India Special Publication no. 27 (A collection of papers presented at the workshop on ” 30th September, 1993 Killari Earthquake, Maharastra” held at Hyuderabad on 24th December, 1993, Hyderabd), pp 7-16
• Simpson, D. W. (1997), National Earthquake Hazards Reduction program, IRIs Newsletter, 1997, Vol. XVI pp9-10 (also available on the web –