ABSTRACT

Subduction zones, where tectonic plates are recycled back into the mantle as in the Cascadia margin of the Pacific Northwest of the United States, host the largest earthquakes and give rise to significant hazard through ground shaking, landslides, and tsunami. This project seeks to better utilize existing geophysical and geological observations from important ?natural laboratories? (Cascadia, Japan and New Zealand) by merging them more fully into new, comparative computer models of system behavior. Developing new modeling software and integrating constraints is expected to lead to new insights into the physics of subduction zone earthquakes, what observations imply for future earthquakes, and, importantly, which observations are needed to improve our understanding of subduction zone hazards and how to reduce uncertainties about system behavior. The project will involve international collaborations, leverage past investments, and will contribute to defining future, optimal observational strategies. An interdisciplinary workforce of students and post-docs will be trained through research and educational efforts, and all project software, tutorials and ?cookbooks? for subduction earthquake modeling will be shared with the community, contributing to advancing computational geoscience approaches in general. A program for precollege, undergrad, and early grad students will be developed to emphasize computational geoscience as an avenue to enhance diversity in the geosciences.

This collaborative effort seeks to integrate seismological, geodetic, experimental, and geological constraints for the Japan, New Zealand and Japan natural subduction zone laboratories into numerical models to advance our understanding of megathrust earthquakes. Forward models and a new numerical modeling framework for data assimilation will be deployed to get closer to versatile tools for data-driven, physics-based hazard assessment. The focus is on the evolution of fault stress and strength over a range of spatio-temporal scales, quantifying uncertainties and sensitivity to parameters. This will allow formulating best strategies for inferring relevant parameters from data in the presence of ambiguous physics, including optimal observational design within the ongoing SZ4D community effort. All code will be made publicly available along with cookbooks and tutorials, and a networked effort will establish new, quantitative links and leverage individual efforts greatly. FRES funding will support a growing community of solid Earth geodynamicists who want to deploy their models in a hazard and monitoring context. A focus will be on training and sharing material for interdisciplinary computational geoscience efforts, from undergraduate to post-doc and practitioner level. Project participants will develop sustainable pathways for participation and work to enhance representation and inclusion in the geosciences by providing new pathways of entry based on modeling and remote sensing to complement field-based approaches.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Please report errors in award information by writing to: awardsearch@nsf.gov.