Enabling curiosity-driven science and digital twins for earthquake physics in the exascale era

Geohazards and risks increase worldwide rapidly due to continuing urbanization, climate change, and high-risk critical distributed infrastructure. The longest modern instrumental records of earthquakes cover less than 100 years, while recurrence intervals of large earthquakes are hundreds of years or more. Increasingly dense observations and physics-based simulations empowered by supercomputing provide pathways for overcoming the lack of data and elucidating spatiotemporal patterns that extend our knowledge beyond sporadic case studies and average statistical laws - however, are typically challenging to integrate.

Digital Twins are emerging in Solid Earth Science, allowing curiosity-driven science to test scientific hypotheses against observations over ranges of space-time scales not accessible for laboratory and field observations. The results can clarify processes leading to large earthquakes, improve our forecasting ability, and enhance the general understanding of earthquakes and faults.

In this presentation, I will highlight Geo-INQUIRE (www.geo-inquire.eu), DT-Geo (www.dt-geo.eu) and ChEESE-2P (www.cheese2.eu), European projects that aim to overcome cross-domain barriers and will exploit innovative data management techniques, modeling and simulation methods, developments in AI and big data, and extend existing data infrastructures to disseminate these resources to the wider scientific community. Specifically, we will provide and enhance access to selected key data, products, and services, enabling the dynamic processes within the geosphere to be monitored and modeled at new levels of spatial and temporal detail and precision.