SCEC Award Number 19159 View PDF
Proposal Category Workshop Proposal
Proposal Title Workshop On How Physics-Based Earthquake Simulators Might Help Improve Earthquake Forecasts
Investigator(s)
Name Organization
Edward Field United States Geological Survey
Other Participants
SCEC Priorities 5a, 2e, 5c SCEC Groups EFP, FARM, CXM
Report Due Date 07/18/2019 Date Report Submitted 07/18/2019
Project Abstract
Due to inevitable assumptions, approximations, and uncertainties, questions have persisted on the usefulness of multi-cycle physic-based earthquake simulators (e.g., RSQSim) with respect to forecasting earthquakes. Whether any model is reliable or trustworthy depends entirely on what questions we are asking of it, so one goal of this workshop is to discuss the specific inferences we would like to make (e.g., as articulated in this recent opinion piece: https://doi.org/10.1785/0220180299). The other goal of the workshop is to identify which of these inferences might be legitimate or robust with respect to current state-of-the-art simulators, as well as what scientific and computational developments will be needed to make such models even more reliable with respect to forecasting earthquakes.
Intellectual Merit Earthquake Rupture Forecasts (ERFs) represent one of two main modeling components used in a wide variety of seismic hazard and risk assessment. Multi-cycle, physics-based earthquake simulators provide one of the most promising avenues for improving the accuracy of ERFs. However, there is currently much debate on whether present capabilities are good enough to provide reliable inferences (is there enough physics in current physics-based simulators?), so the workshop took a deep dive on this question.
Broader Impacts By directly addressing and improving one of the two main modeling components used in seismic hazard and risk assessment, this activity will likely have direct societal impact around the globe, at least eventually.
Exemplary Figure Figure 2. Image from Tom Jordan's presentation illustrating a two-pronged approach for future developments.