Collaboratory for the Study of Earthquake Predictability

Collaboratory for the Study of Earthquake Predictability
Name	Collaboratory for the Study of Earthquake Predictability
Established	1997
Headquarters	Southern California

Collaboratory for the Study of Earthquake Predictability is an international research initiative focused on improving models of seismic forecasting through coordinated experiments, standardized data analysis, and community-driven testing protocols. The collaboratory convenes researchers from institutions such as California Institute of Technology, U.S. Geological Survey, Massachusetts Institute of Technology, Stanford University, and University of Tokyo to evaluate competing hypotheses about earthquake occurrence, statistical seismology, and operational hazard assessment. It aggregates paleoseismology, geodesy, and seismic network data and organizes prospective prediction experiments that connect efforts in Southern California Earthquake Center, Incorporated Research Institutions for Seismology, and global observatories.

History

The collaboratory traces its origins to efforts in the 1990s to formalize testing of earthquake forecasting methods developed at California Institute of Technology, U.S. Geological Survey, and University of California, Berkeley. Early workshops involved participants from Southern California Earthquake Center, Japan Agency for Marine-Earth Science and Technology, European Mediterranean Seismological Centre, and National Earthquake Information Center, which led to pilot forecasting experiments and community standards. Formalization occurred through affiliations with National Science Foundation programs and coordination with international projects led by International Seismological Centre and United Nations Educational, Scientific and Cultural Organization. Over time the collaboratory aligned with operational projects such as the Collaboratory for the Study of Earthquake Predictability-sponsored testing initiatives, integrated methodologies from Paleoseismology groups at Earthquake Research Institute, University of Tokyo, and expanded partnerships with regional agencies like California Geological Survey and British Geological Survey.

Mission and Objectives

The mission centers on establishing reproducible, prospective tests of earthquake forecasting methods by coordinating multi-institutional experiments with stakeholders such as U.S. Geological Survey, European Space Agency, and Japan Meteorological Agency. Objectives include benchmarking algorithms developed at Massachusetts Institute of Technology and Stanford University, validating statistical models used by Southern California Earthquake Center and Incorporated Research Institutions for Seismology, and translating results into operational guidance for agencies like California Office of Emergency Services and National Oceanic and Atmospheric Administration. The collaboratory aims to foster interoperability across archives maintained by International Seismological Centre and IRIS, standardize evaluation metrics used in initiatives associated with Global Seismographic Network, and accelerate methodological advances informed by work at Los Alamos National Laboratory and NASA Jet Propulsion Laboratory.

Research Programs and Methodologies

Research programs deploy statistical seismology, physics-based rupture simulation, and machine learning pipelines developed in collaborations with Princeton University, Columbia University, and Purdue University. Methodologies include retrospective catalog analyses using datasets from Southern California Seismic Network, prospective forecasting experiments mirroring protocols from the Collaboratory for the Study of Earthquake Predictability community, and multi-model ensembles inspired by climate-model intercomparison projects such as Coupled Model Intercomparison Project. Teams apply tools originating at Lawrence Berkeley National Laboratory and Los Alamos National Laboratory to test point-process models, epidemic-type aftershock sequence (ETAS) formulations advanced by researchers at University of Pisa and University of Rome Tor Vergata, and physics-based rupture scenarios informed by studies at Seismological Laboratory, Caltech and Earthquake Research Institute, University of Tokyo.

Data and Computational Resources

The collaboratory aggregates seismic catalogs from International Seismological Centre, waveform archives from Incorporated Research Institutions for Seismology, geodetic time series from GPS Earth Observation initiatives at Scripps Institution of Oceanography, and paleoseismic datasets curated by Paleoseismology Research Groups at University of California, Riverside and University of Northern Arizona. Computational resources include high-performance computing allocations at National Center for Supercomputing Applications, cloud credits from vendors used by NASA Earth Exchange, and software platforms developed in partnerships with OpenQuake contributors and teams at Carnegie Mellon University. The collaboratory emphasizes reproducible workflows using tools from GitHub-hosted projects and containerization strategies advocated by Docker, Inc. and Kubernetes adopters in academic settings.

Collaborations and Partnerships

Collaborations span academic institutions such as University of Washington, University of British Columbia, and Imperial College London, government science agencies including U.S. Geological Survey and Japan Meteorological Agency, and international bodies like European Mediterranean Seismological Centre and United Nations Office for Disaster Risk Reduction. Partnerships connect the collaboratory with operational networks—Global Seismographic Network, Alaska Earthquake Center, and California Integrated Seismic Network—and with research consortia including Southern California Earthquake Center and International Seismological Centre. The collaboratory also coordinates exercises with emergency management organizations such as California Office of Emergency Services and infrastructure stakeholders including Amtrak engineering groups and Pacific Gas and Electric Company resilience teams.

Education and Outreach

Education efforts include summer schools modeled after programs at Seismological Laboratory, Caltech and training workshops co-hosted with Incorporated Research Institutions for Seismology and Southern California Earthquake Center, offering modules on statistical seismology, geodesy, and computational reproducibility. Outreach engages the public through collaborations with museums and informal science institutions like Natural History Museum of Los Angeles County and media partnerships with National Public Radio features, while professional development interacts with policy fora at National Academies of Sciences, Engineering, and Medicine and international conferences such as American Geophysical Union and European Geosciences Union meetings.

Impact and Criticism

The collaboratory has influenced standards for prospective testing used by U.S. Geological Survey operational models and informed recommendations from National Academies of Sciences, Engineering, and Medicine, but it has faced criticism from some researchers at University of California, Berkeley and Columbia University regarding the limits of predictability and the appropriateness of evaluation metrics. Skeptics cite debates similar to those surrounding the 1990s forecasting controversies involving Parkfield earthquake experiment and argue that retrospective success does not guarantee operational utility, a critique echoed in literature originating at University of Tokyo and Seismological Society of America. Proponents point to improved reproducibility, clearer benchmark datasets from Incorporated Research Institutions for Seismology, and increased integration with operational agencies like U.S. Geological Survey as evidence of progress.

Category:Seismology