Inverse problems and earthquake location

• Main Author: Soh, Wei Xuan
• Other Authors: Tong Ping
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Science::Mathematics
• Online Access: https://hdl.handle.net/10356/148481
• Institution: Nanyang Technological University

An earthquake is defined as the movement of rocks which results in the shaking of the Earth. It usually occurs in the crust or mantle of the Earth and results in devastating aftereffects such as ground rupture, landslides and tsunamis (Anderson, 1997). Every year, around 7 million earthquakes occur worldwide, of which 20,000 are sizable enough to be located and about 1300 of the earthquakes are destructive (Bobrowsky, 2013). As earthquakes tend to reoccur, it is extremely important to accurately relocate earthquake hypocenter to increase confidence in the hazard modelling for the affected site. Furthermore, this provides guidelines for the local government to improve on their future urban planning to minimize the damage from the natural disaster. Specifically, in this paper, the Double Difference Method was employed to locate the earthquake hypocenter in both homogenous and heterogenous model. The Double Difference Method refers to an algorithm that accurately relocates earthquake hypocenter by minimizing the Double Difference between the observed and calculated arrival time of seismic waves travelling from one earthquake to 2 different seismic stations. In the homogenous model, seismic waves travel in a straight ray path at a constant velocity. On the other hand, in the heterogenous model, seismic waves bend when traveling from one layer to another layer of differing velocities and this behavior is described using Snell’s Law. The Double Difference Method is widely adopted by many seismologists as it minimizes the need for station corrections or highly accurate travel time for ray beyond the source region (Ma & Eaton, 2011). Furthermore, hypocenter located using this method is more accurate as the problem is linearized by the Double Difference equations. Similarly, this method could be extended to detect nuclear testing sites and underground explosion as it releases similar amount of energy as earthquake.