The assessment of seismic hazard involves specifying the likelihood, magnitude, location, and nature of earthquakes that might have damaging effects at the site of construction; and estimating the intensity, duration, and frequency characteristics of the induced ground shaking. This project is applying statistical techniques to investigate the adequacy of the Poisson-Independent-Exponential (PIE) model of earthquake occurrence used in most seismic analyses in the United States. However it is recognized that the PIE model is not consistent with the physical process of crustal strain accumulation and release associated with earthquakes. Therefore the research is also exploring alternative models of seismic hazard and its uncertainty, especially those models which include the effects of non-stationarity and memory patterns in earthquake occurrence. The research focuses on the inadequate understanding of seismicity of continental interiors, specifically within the central and eastern United States. This is a region where the short historical earthquake record available cannot be trusted to be representative of future hazards. The record of earthquakes in China is the only one that can be used to assess "memory models" of seismicity in intraplate regions. The project involves a collaboration with the Institute of Geophysics in the PRC State Seismological Bureau: the long catalog of earthquakes in China is being used to explore and calibrate models of seismicity with memory. For intraplate regions, many engineering decisions related to earthquake hazard mitigation involve great uncertainties. As a result, this research project is expected to contribute significantly to the assessment of earthquake hazard in the central and eastern United States.
