Title: Bayesian Inference of Seismic Events from Anecdotal Historical Accounts in Indonesia
Date: Friday, February 26, 2021
Place and Time: Zoom, 3:05-3:55 pm
Abstract. Prediction of the impact and danger of future seismic hazards critically depends on an accurate understanding of past (on a geological time scale) events. As accurate seismic readings have only been available for less than 80 years, seismologists must rely on anecdotal evidence to determine the frequency and magnitude of past earthquakes. To place quantitative bounds on this anecdotal data, we apply a Bayesian framework to determine a posterior distribution on the magnitude and location of an earthquake from 1852 in the Banda Sea of Eastern Indonesia. The novelty of this application arises from the inherent noise and uncertainty which come into the likelihood distribution. Further extensions to other seismic events, improvements on the sampling method, and extensions to other settings will be considered as well.