SCEC Award Number 11049 View PDF
Proposal Category Collaborative Proposal (Integration and Theory)
Proposal Title A Collaborative Project: Comparison, Verification, and Validation of Earthquake Simulators
Investigator(s)
Name Organization
Terry Tullis Brown University Michael Barall Invisible Software, Inc. Keith Richards-Dinger University of California, Riverside Steven Ward University of California, Santa Cruz John Rundle University of California, Davis Louise Kellogg University of California, Davis Olaf Zielke Columbia University Nadia Lapusta Deutsches GeoForschungsZentrum Potsdam (Germany) Bruce Shaw Columbia University James Dieterich University of California, Riverside Fred Pollitz United States Geological Survey Donald Turcotte University of California, Davis Nick Beeler United States Geological Survey Russell Robinson GNS Science (New Zealand)
Other Participants
SCEC Priorities A6, A10, A9 SCEC Groups FARM, WGCEP, Simulators
Report Due Date 02/29/2012 Date Report Submitted N/A
Project Abstract
 Simulators help us understand the mechanics of earthquakes. They help us learn whether aspects of earthquakes may be predictable and, if so, how those predictions might be done. This project involves a concerted effort toward the comparison and evaluation of a variety of earthquake simulators to gain a better understanding of which features are common to them all and which features depend strongly upon variable details of input and assumption. The efforts of the Earthquake Simulator Group lie in three areas: benchmark test problems, full statewide simulator runs, and comparison of results. We have made progress in all three area. In particular the group has developed standard output formats such that long catalogs from any of the simulators can be run through software tools placed in an open and developing “toolbox’ housed on SCEC servers. These tools help us explore where simulators might hold predictive information. They also help us identify outliers in behavior and search for explanations.
Intellectual Merit Simulators help us understand the mechanics of earthquakes. They help us learn whether aspects of earthquakes may be predictable and, if so, how those predictions might be done. For earthquake simulators to be useful however, they must behave in a manner that has elements of reality. Determining how realistic simulator results might be is a difficult task. This project involves a concerted effort toward the comparison and evaluation of a variety of earthquake simulators to gain a better understanding of which features are common to them all and which features depend strongly upon variable details of input and assumption.
Broader Impacts The over-arching goal of my efforts is SYSTEM LEVEL SCIENCE unifying geodesy, geology, simulators and seismicity. Geodesy deforms the region and stresses a realistic fault system in a manner consistent with Geology. Simulators, like these, release those on-fault stresses in believable earthquake sequences. If this goal is achieved, earthquake forecasts will attain the same status and political/economic impact as Global Climate Models now have.
Exemplary Figure Figure 4. Various consistency tests for ALLCAL2 output. (A) Magnitude versus Fault Area, (B) Slip versus Fault Area, (C) Slip versus Rupture Length, (D) Magnitude versus Static Stress Drop. Plots like these are part of the “toolbox” being developed by the simulator group to compare and validate simulator results.