Development and Testing of a 3D Velocity Model for Improved Event Location: A Case Study for the India-Pakistan Region

Abstract

3D velocity models will play an important role in improving location accuracies of small events within the context of a comprehensive nuclear-test-monitoring treaty. Commonly, such models are developed using tomographic methods that require the initial velocity model to be a reasonable approximation of the true Earth structure. In this article, we present a method for the development of an a priori model for the crust and upper mantle in the India-Pakistan region and techniques for evaluating its suitability for use as an initial model in tomographic studies. We developed the Weston India-Pakistan 3D velocity model (WINPAK3D) by integrating the results of more than 60 previous studies related to crustal velocity structure in the region. A well-located event that occurred near Nilore, Pakistan, on 14 February 1977 provided a basis for testing the model via several different approaches. Evaluation of the 3D model included (1) cross-validation analysis, (2) comparison of model-determined hypocenters with a known event location, and (3) comparison of model-based and empirically derived travel-time correction surfaces generated for an event within 5 km of the seismic station at Nilore. The 3D model is demonstrated to improve the regional location of the 1977 event as compared to 1D models. The epicenter mislocation (using the first-arrival phase at 23 regional stations) determined by the 3D model was only 6.2 km, compared to 13.1 km and 15.4 km for a regional 1D model and the IASP91 model, respectively. Results of extensive testing demonstrate that WINPAK3D will be a particularly good initial model for tomographic studies, as it improves regional location accuracy, specifically for sparse networks. These results demonstrate that 3D velocity models are essential for improving location accuracy in regions that have complicated crustal geology and structures. Such 3D models may be a prerequisite for achieving location accuracies as called for by the Comprehensive Nuclear-Test-Ban Treaty, particularly for small events that are not recorded teleseismically.