UCIS4EQ applied to the South Iceland region

Urgent Computing (UC) refers to the use of High-Performance Computing (HPC) and High-Performance Data Analytics (HPDA) during or immediately after emergency situations. It typically combines complex edge-to-end workflows with capacity computing, where multiple model realizations are required (to account for input and model uncertainties) under strict time-to-solution constraints. Enabling urgent HPC for emergency scenarios, such as earthquakes, can prove valuable towards resilience and relief. The temporal horizon for UC typically ranges from minutes to a few hours.

A novel  HPC-based urgent seismic simulation workflow, the Urgent Computing Integrated Services for Earthquakes (UCIS4EQ),  focuses on short-time reports of the consequences of moderate to large earthquakes. UCIS4EQ automatically prepares and manages sets of physics-based deterministic simulations to rapidly obtain synthetic results. Based on pre-computed and on-the-fly simulations, UCIS4EQ delivers estimates of relevant ground motion parameters, such as peak ground velocity, peak ground acceleration, or shaking duration, with very high spatial resolution.  The physics-based engine includes pre-trained Machine Learning (ML) models fed with pre-computed simulation databases and r full 3D simulations on demand, providing results in minutes and hours, respectively.  The combined results, when well-calibrated, could complement GMPEs for rapid hazard assessment

To demonstrate the potential of UC in seismology, we show the capability of the UCIS4EQ workflow both for the ML predictions and for deterministic simulations in the South Iceland Seismic Zone (SISZ) and the Reykjanes Peninsula Oblique Rift (RPOR). The largest historic earthquakes in Iceland have occurred within these zones and have exceeded magnitude 7. The study region  (63.5°-64.5°N, 20°-22°W) is where the largely sinistral East-West transform motion across the tectonic margin is taken up by a complex array of near-vertical and parallel North-South oriented dextral transform faults in SISZ-RPOR. The high seismic activity in the area widely affects the capital Reykjavik, the most populous city in Iceland. 

This work has been supported by the ChEESE and eFlows4HPC projects.