Advancements in pyCSEP: Enhancing Earthquake Forecast Evaluation with New Features and Regional Applications

Kenny Graham, José A. Bayona, Khawaja M. Asim, Pablo C. Iturrieta, Francesco Serafini, Emanuele Biondini, David A. Rhoades, William H. Savran, Philip J. Maechling, Matthew C. Gerstenberger, Fabio Silva, & Maximilian J. Werner

Published September 8, 2024, SCEC Contribution #14015, 2024 SCEC Annual Meeting Poster #213

The Collaboratory for the Study of Earthquake Predictability (CSEP) is a global initiative dedicated to enhancing earthquake predictability by rigorously testing probabilistic earthquake forecast models and prediction algorithms. A key tool in this effort is pyCSEP, an open-source software toolkit designed to evaluate earthquake forecasts. PyCSEP includes modules for accessing and processing earthquake catalogs, visualizing forecast models, and performing statistical evaluations. Recent updates have expanded pyCSEP's capabilities, including support for authoritative earthquake catalogs from Italy (Bollettino Seismico Italiano), New Zealand (GeoNet), and globally (the Global Centroid Moment Tensor catalog). New features enable the creation of multi-resolution spatial forecast grids, the implementation of non-Poissonian testing methods, the evaluation of alarm-based models, and the application of a global seismicity reference model to specific regions.

Here, we highlight recent advancements and their application in regional studies, with a particular focus on New Zealand. These updates build on the initial introduction of pyCSEP by incorporating new tests and extending the toolkit’s capabilities to serve the earthquake forecasting community better. Contributions from the CSEP community have significantly enriched pyCSEP’s capabilities. PyCSEP exemplifies successful collaborative research and community-driven development and promotes transparent scientific practice.

Key Words
Earthquake Forecasting, CSEP, Evaluation

Citation
Graham, K., Bayona, J. A., Asim, K. M., Iturrieta, P. C., Serafini, F., Biondini, E., Rhoades, D. A., Savran, W. H., Maechling, P. J., Gerstenberger, M. C., Silva, F., & Werner, M. J. (2024, 09). Advancements in pyCSEP: Enhancing Earthquake Forecast Evaluation with New Features and Regional Applications. Poster Presentation at 2024 SCEC Annual Meeting.

Related Projects & Working Groups
Earthquake Forecasting and Predictability (EFP)