A Method for Reconstructing Historical Destructive Earthquakes Using Bayesian Inference

In many parts of the world, earthquake-generated tsunamis cause significant risk to human populations. Assessing the seismic risk of a region relies on collecting data from past seismic events i.e. predicting the future can only happen if the past is truly understood. However, data collection with modern instruments has captured only a part of the Earth's seismic history and hence only gives us a partial understanding of the risk factors of a region. Non-instrumental records that precede modern seismic instrumentation, such as anecdotal accounts in newspapers, personal journals, or oral tradition, provide limited information on tsunamis that occurred before the modern era. Using Bayesian inference and modern scientific computing, we develop a framework for reconstructing the source events of historical earthquakes based on anecdotal accounts of the resultant tsunamis. We further extend this work to include tsunamis that were potentially generated from landslides instead of directly from uplift of an earthquake.