Detecting Earthquake Epicenter Temporal Patterns via Dirichlet Process

• Activity Number: 119 - SPEED: Bayesian Methods Student Awards
• Type: Contributed
• Date/Time: Monday, July 31, 2017 : 8:30 AM to 10:20 AM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Meredith Ray* and Dale Bowman and Roy Von Arsdale and Hongmei Zhang
• Companies: University of Memphis and University of Memphis and University of Memphis and University of Memphis
• Keywords: Bayesian inference ; Dirichlet Process ; Clustering ; Earthquake epicenters

Abstract

Earthquakes can be one of the deadliest natural disasters known to man. Our study focuses on earthquakes occurring around intraplate fault lines, such as the New Madrid seismic zone (NMSZ) in the middle of the United States. Detecting general temporal patterns of earthquake epicenters will significantly improve the ability of earthquake prediction and is in great need. Based on magnitude and geographic locations of each earthquake, including latitude, longitude, and kilometers below sea level, occurring in the NMSZ from 1996-2013, as the first attempt, we developed a Bayesian clustering method to group earthquake epicenters and infer their temporal patterns. Dirichlet process is utilized to detect the centers and estimate the patterns. Simulations were conducted to assess the sensitivity, specificity, and accuracy of the proposed method and to compare to other commonly used clustering methods such as Kmean, Kmedian, and PAM. We then applied the developed method to tease out temporal patterns in grouped earthquake epicenters.