CAREER: Towards a comprehensive model of seismicity throughout the seismic cycle

Every day, faults worldwide produce hundreds of small earthquakes, too weak to be felt or produce any damage. Camilla Cattania and her group will be investigating whether these small earthquakes could provide clues about where and when large, destructive earthquakes might occur. Sometimes, small earthquakes turn out to be foreshocks of a larger earthquake - but we can't tell this is so until after the large earthquake has happened! Laboratory earthquake experiments (which produce "lab earthquakes" in large hunks of rock) have been done to see if there are any telltale signs from small lab earthquakes before a big lab quake occurs. Unlike real faults in the Earth, these experiments show a predictable pattern of small earthquakes before each large lab earthquake. Cattania and her team will build sophisticated computer models of faults to try to understand why the lab experiments seem to disagree with data from real faults. The models will be tested using earthquake data from California, Italy, and Iceland, and adjusted until they represent real-world fault physics as accurately as possible. Using these models, Cattania and her team will determine when laboratory experiments can be useful analogs to natural faults and when they cannot, and under what conditions foreshocks are expected to happen in the real world. This project could improve earthquake forecasts, and help researchers identify the best regions to install earthquake monitoring equipment to detect foreshocks (which may further improve earthquake forecasts). This project will explore physical processes driving microseismicity by developing a new earthquake cycle model that accounts for fault complexity and interactions between discrete faults in a three-dimensional damage zone, to simulate realistic seismicity patterns throughout the seismic cycle. The model will consist of a population of faults subject to tectonic and transient loading, governed by laboratory derived frictional laws (rate-state friction and dynamic weakening at high slip speeds) and embedded in an elastic medium. A 2.5D representation of small faults will allow modeling realistic stress interactions between a large number of faults, at a moderate computational cost. Analytical expressions based on stress transfer and fracture mechanics will be used to relate seismicity patterns (such as changes in the frequency-magnitude distribution and swarm migration) to the underlying state of stress. Finally, the project will leverage high-resolution seismicity catalogs to test these results against observations from California, Italy, and Iceland. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.