Statistical and simulation based forecasting: Applications of numerical simulations and data analysis to earthquake forecasting.

摘要

This dissertation covers various research projects related to computational analysis and prediction in complex systems, more specifically earthquake fault systems, one of the most important examples of a threshold system. The first project introduces a method of "data scoring". Simulation data is compared to historical seismicity and used to make a forecast of future seismic activity. The second project introduces the Branching Aftershock Sequence (BASS) model. The statistical distributions of b-values and largest aftershocks are given. The statistics of foreshocks are also found. The concept of record-breaking earthquakes is introduced in the third project and applied to earthquake data in the CMT catalog for the years 1977 to 2006. The final project covers the use of virtual reality and immersive visualization and its application to the task of creating and maintaining digital fault models.