3D CRUSTAL STRUCTURE MODELLING OF GARHWAL HIMALAYAS

Abstract

Determination of seismic velocity structure of crust and upper mantle of the earth is one of the major objectives of seismology. Accurate velocity information is necessary for a variety of purposes, including the location of earthquakes, the determination of the composition and origin of theouter layers of the earth, andtheinterpretation of large-scale tectonics. While lot of advancement has taken place in instrumentation and digital data acquisition has the capabilities to mark the arrivals with requisite accuracies, the velocity modeling has been lacking in the past for accurate estimation of locations and to make use of relatively more accurate phase arrival data. Taking advantage of the high rate of occurrence of microearthquakes, an attempt has been made in the present study to estimate the crustal velocity structure in Garhwal Himalaya. The main objective of the present study has been to revisit the current velocity structures available for the Garhwal Himalaya and propose a crustal velocity structure based on the earthquake data acquired from this region. Local Earthquake Tomography (LET) technique has been used in the present study which uses simultaneous inversion for hypocenter and seismic velocity parameters to minimize the data misfit. The earth structure has been represented in 3-D by velocity at discrete points, and velocity at any intervening point is determined by linear interpolation among the surrounding eight grid points. The data acquired through a local seismological network in the Garhwal Himalaya deployed byDepartment of Earthquake Engineering, IIT Roorkee has been used in the present study. 1-D analyses has been carried out to estimate minimal 1-D models using VELEST by considering three ofthe 1-D velocity models from past studies carried out by in various workers for Himalayas. Based on the output of the 1-D analysis an average 1-D crustal velocity model has beenproposed for Garhwal Himalaya. The three 1-D crustal velocity models thus estimated using 1-D analyses and the proposed 1-D crustal velocity model are then used separately as minimum 1-D velocity models for estimation of 3-D crustal velocity structure for Garhwal Himalaya. The 3-D crustal velocity structure has been estimated using linearized inversion using LET. The grid point locations with respect to the center of the coordinate system have been chosen iteratively based on the earthquake locations and tectonic features in the area. The resolution matrix, derivative weighted sum (proportional to number of rays that sample each cell) and hit counts (number ofrays that traverse a specific cell) shows that the profiles falling inthe center ofthe grid and in the upper layers are well resolved. Further, the whole grid was rotated by 45° to get the grids perpendicular to the predominant direction ofthe seismogenic features in the Garhwal Himalaya. The comparison of resolution matrix, derivative weighted sum and hit counts shows better results in later case. An endeavor has been made to validate the results by comparing the velocity distribution patterns observed in the estimated 3-D crustal velocity structures with the surface manifestations of the tectonic features. Velocity profiles along the grid lines have been used for matching the geological and tectonic features in the Garhwal Himalaya. Further, the traverses reported by Valdiya (1980) elucidating the lithotectonic settings have been also compared with the depth profiles obtained through 3-D crustal velocity structure analyses. The comparison reveals the depth wise behavior ofsurface manifestations ofthe geological features namely, Basul Thrust, Srinagar Thrust, Dunda Thrust, Main Boundary Thrust, Main IV Central Thrust and Dunda Fault. Further, out of sequence thrust which is Srinagar thrust is found to be dipping southerly in many of the profiles. The 3-D velocity structure reveals a variation in velocities at a depth of about 10-20 km in almost all the profiles which can be attributed to the detachment (decollement) surface as reported in earlier studies throughout the Himalayas. The transition zone at detachment surface is found to be about 10 km thick with a velocity contrast of about 5.6 km/sec at the top and about 6.0 km/sec at the bottom. The undulating pattern observed in the vertical profiles reveals that detachment surface is not an even surface and possibly participating in the tectonic process. Further, the depth sections of contemporary seismicity have been overlaid on the vertical profiles which reveal that most of the features are seismically active. Based on the matching of the 3-D crustal velocity structure with the existing geological and tectonic features and relatively higher values of resolving parameters (KHIT, DWS and RESOLUTION) the 3-D Model IV estimated using the average 1-D velocity model (Model IV) has been proposed for Garhwal Himalaya. The 3-D crustal velocity structure estimated in the present study will be useful for accurately locating the earthquakes in the area for better seismotectonic modeling which in turn is useful in seismic hazard assessment and mitigation.