SEISMIC SOURCE STUDIES IN THE HIMALAYA USING DIGITAL SEISMOGRAMS

Abstract

It has been well accepted that the Plate Tectonics theory has provided abasis to explain the occurrence of earthquakes at the boundaries of the six large and many other small plates which cover the surface of the globe. The Alpide - Himalaya belt is a boundary between two large lithospheric plates, namely, the Indian and the Eurasian plates, where comparatively high seismicity has been observed. A number of investigations carried out in the region have allowed construction of models of the tectonics as well as geodynamic processes occurring in the region. A finer understanding of the relationship between these processes and the seismicity of the region will allow refinement of the existing models which will be useful also in seismic hazard reduction research to a considerable extent. Incisive seismological studies became possible with the availability of high quality data from the World wide Standardized Seismograph Network (UUSSN) in the early sixties. The quality of the seismological data further improved with the deployment of the Global Digital Seismograph Networks (GDSN) in the early seventies. A better idea of the active tectonics of a region can be obtained from the fuller use of the seismograms for providing reliable parameters for specifying the earthquake sources.

The depths of foci for earthquakes of the Himalaya reported in seismological bulletins are relatively less precisely constrained because of various factors which include the distribution of the recording station, regional and local perturbation of the velocity structure and the trade off between origin time and the depth of focus in routine location procedures. Furthermore, because of the scarcity of local networks in and around the Himalaya, the investigations of the focal depth of the earthquakes there are mostly based on the teleseismic data. Thus one finds a considerable range of the depth of focus ( 10 km to 70 km) obtained from such studies. The source mechanism studies based on the first motion data for earthquakes in the Himalaya have been beset with considerable ambiguity. Many events may be interpreted as having either a thrust or a strike slip type source mechanism. Such an ambiguity of interpretation may be reduced by making fuller use of the information carried in the seismograms. The modelling of earthquake sources using synthetic seismograms of teleseismic long period P and S waves has been demonstrated to be a powerful technique to estimate both the source mechanism and focal depth of earthquakes reliably. In the present study we use this technique to determine the source parameters and the depth of focus of two moderate sized earthquakes of the central ( Nepal,Sikkim ) Himalaya. For this study the seismograms digitally recorded by the GDSN were used. To the best of our knowledge this is the first seismological study made in India using GDSN seismograms. -lv- The synthetic seismograms have been generated by representing the source by a displacement dislocation which is specified by three orientation angles (viz., strike, dip and slip) and scalar seismic moment. A trapezoidal source time function has been used. The medium response which includes the source crust transfer function, the receiver crust transfer function and the mantle transfer function constitutes one of the fundamental elements in generating the synthetic seismograms.The source and the receiver crusts are represented by a layered medium overlying a half space.The source and receiver crust transfer functions have been evaluated using propagator matrix method ( e.g.,Hudson,1969a,b). Mantle transfer function has been represented by using the constant Q model of attenuation. The attenuation parameter given by Langston and Helmberger (1975) has been used. The synthetic seismograms have been generated by multiplying the source functions with the medium response, and the instrument transfer function in the frequency domain and converting them into time domain using the fast Fourier transform. Digital seismograms recorded by the GDSN were collected from the United States Geological Survey (USGS). Four events of our interest were extracted from the Network-day tapes that contain data with other necessary information about epicenters and stations of a particular day recorded at these observatories. -v- The quality of the observed seismograms in the case of two of these earthquakes were found to be poor and could not be used for the estimation of their source parameters. A trial and error procedure was adopted for estimating the source parameters and the depth of focus by comparing the synthetic and the observed seismograms. Before analysing the seismograms using the above procedure a study was made to investigate the trade off of the influence of the source time function and the depth of focus on the apparent period of the body-wave waveforms. An apparent linear relationship has been obtained between the duration of the source time function and the apparent period of the synthesized P wave. Similarly a positive correlation was also obtained between the depth of focus and the apparent period of the synthetic P waveform. In order to resolve this ambiguity recourse was taken to constrain the source time function on the basis of the magnitude of the earthquake. In order to estimate the depth of focus of the earthquakes the synthetic seismograms were generated for different source depths and each of them was compared with the corresponding observed seismograms visually until best fit was observed. In this way the depths of focus for the two events analysed here have been estimated to be about 20 - 22 km. The fault plane solution of the earthquake which occured in eastern part of the central Himalaya (HIM1) shows strike slip faulting where as the one that has occurred in the western section (HIM2) admits thrust faulting. This difference can be -viexplained in terms of local tectonic environments of the localities of the earthquakes. The fault plane parameters of the two earthquakes estimated are as follows. Event HIM1 HIM2 Nodal Plane 1 Dip = 74° 26* Strike = 119° 110* Dip direction = S29'U N20*E Nodal Plane 2 Dip = 64* 64' Strike = 38' 290° Dip direction = N52*U S20#U In the case of earthquake HIM1 our solution matches closely with the fault plane solutions determined by Nl and Barazangi (1984) and Das Gupta et al.(1987). However, while selecting the fault plane among the nodal planes,we are guided by the geology of the region and have favoured right lateral strike slip faulting for HIM1. The earthquake HIM2 has been investigated by Kanamori and Given(1982),Ni and Barazangi (1984), Singh (1985), Sipkin(1987), Das Gupta et al.(1987) and Molnar and Lyon-Caen (1989) using various techniques.The fault plane solution determined in this study matches with those obtained by Ni and

Barazangi (1984), Singh (1985), Sipkin (1987) and Molnar and Lyon-Caen (1989) to a considerable extent. The fault plane chosen corresponds to the plane dipping towards the north east at 26* and reflects the thrusting occurring in the region.