A highlight of the event was researcher Rut Blanco from the Barcelona Supercomputing Center (BSC) receiving an award for her outstanding work, “Machine Learning-based Estimator for Ground Shaking Maps Workflow Applied to New Zealand.” Her presentation was part of the special session “Earthquake Source and Ground Motion Modeling for Seismic Hazard and Risk Assessment.”
The study, a collaboration with Marisol Monterrubio-Velasco, Brendon Bradley, Claudio Schill, and Josep de la Puente, introduced the Machine Learning Estimator for Ground Shaking Maps (MLESmap). This tool, originally designed for Iceland, was applied to New Zealand using data and simulations provided by the University of Canterbury team.
During her talk, Rut explained how MLESmap uses three machine learning algorithms to estimate ground acceleration after earthquakes. She also shared the promising results obtained from these simulations, emphasizing the potential of this approach to improve seismic hazard assessment and emergency response efforts.
Advancing Urgent Computing and Early Warning Systems
ChEESE made a strong impression in Track 14, a special session focusing on Urgent Computing and Early Warning Systems for Geohazards. The session highlighted the importance of integrating HPC with early warning technologies to reduce the impact of geohazards. Some key contributions included:
- Enabling Urgent Computing Services for Geohazards at the EuroHPC Supercomputing Infrastructure
Presented by Arnau Folch, Josep de la Puente, and colleagues, this talk explored how EuroHPC resources are used to deliver real-time computing solutions for earthquakes and tsunamis. - Advances and Challenges in Seismic and Tsunami Hazard Assessment in Urgent Computing Contexts
This presentation, led by Natalia Zamora and team, addressed the use of HPC workflows to improve tsunami and seismic hazard evaluations. - Meteotsunami-HySEA: Advanced GPU-Accelerated HPC Code for Simulating Meteotsunamis
Alejandro González del Pino and his co-authors demonstrated the efficiency of GPU-accelerated simulations for modeling meteotsunamis on realistic bathymetries. - High-Resolution Numerical Simulations for Tsunami Preparedness on the Andalusian Coast
Carlos Sánchez-Linares and collaborators presented a study using high-resolution simulations to assess tsunami risks along the Andalusian coastline. - Assessing Exposure and Potential Impacts of Seismically-Induced Tsunamis in Northeastern Italy
This research, led by Hazem Badreldin and team, analyzed the potential impact of tsunamis triggered by seismic events in northeastern Italy. - Using HPC for Probabilistic Tsunami Inundation Maps at Stromboli
Juan Francisco Rodríguez Gálvez and colleagues (Paper ID: 901) shared their work on probabilistic tsunami inundation maps for Stromboli, using HPC to enhance risk assessment accuracy. - HPC and AI Workflows: A Dual Urgent Computing Strategy for Earthquakes
In this presentation, Marisol Monterrubio-Velasco, Rut Blanco, and others introduced a dual strategy combining HPC and AI to address earthquake emergencies effectively.
