MakeQuake: Developing Validation and Model Intercomparison Tools for Earthquake Simulators

Bruce E. Shaw

Published September 8, 2024, SCEC Contribution #13681, 2024 SCEC Annual Meeting Poster #121

A fundamental goal of earthquake simulators is to develop models that can produce synthetic data that can pass the "Turing Test" of being indistinguishable from real earthquakes. Matching not just one type of data, but a whole suite of different kinds of data is critical to obtain the deepest correspondence. We are developing tools that can assist in this varied data validation part. It contains multiple types of data and python tools to quantify comparisons. The comparisons are against collections of events and are statistically based. We are developing a Jupyter Notebook that will automatically calculate data comparisons and other measures of model behaviors. Models receive individual measure scores, and cumulative scores across measures, with an option to put either equal or different weights on individual measures for the cumulative score. For models reaching sufficient cumulative scores additional model intercomparison measures are being developed as well.

Key Words
Earthquake simulators, Validation tools, Intercomparison tools

Citation
Shaw, B. E. (2024, 09). MakeQuake: Developing Validation and Model Intercomparison Tools for Earthquake Simulators. Poster Presentation at 2024 SCEC Annual Meeting.

Related Projects & Working Groups
Fault and Rupture Mechanics (FARM)