Fichtner, Andreas

The EU Center of Excellence for Exascale in Solid Earth (ChEESE) develops exascale transition capabilities in the domain of Solid Earth, an area of geophysics rich in computational challenges embracing different approaches to exascale (capability, capacity, and urgent computing). The first implementation phase of the project (ChEESE-1P; 2018–2022) addressed scientific and technical computational challenges in seismology, tsunami science, volcanology, and magnetohydrodynamics, in order to understand the phenomena, anticipate the impact of natural disasters, and contribute to risk management. The project initiated the optimisation of 10 community flagship codes for the upcoming exascale systems and implemented 12 Pilot Demonstrators that combine the flagship codes with dedicated workflows in order to address the underlying capability and capacity computational challenges. Pilot Demonstrators reaching more mature Technology Readiness Levels (TRLs) were further enabled in operational service environments on critical aspects of geohazards such as long-term and short-term probabilistic hazard assessment, urgent computing, and early warning and probabilistic forecasting. Partnership and service co-design with members of the project Industry and User Board (IUB) leveraged the uptake of results across multiple research institutions, academia, industry, and public governance bodies (e.g. civil protection agencies). This article summarises the implementation strategy and the results from ChEESE-1P, outlining also the underpinning concepts and the roadmap for the on-going second project implementation phase (ChEESE-2P; 2023–2026).

A challenge in earthquake seismology is the provision of realistic and accurate ground-shaking scenarios for seismic hazard and risk assessment. Owing to the exponential growth of HPC resources, the application of physics-based methods to ground motion characterization has become increasingly popular. However, physics-based techniques could break down in the simulation of high-frequency ground motion, particularly at frequencies higher than 1 Hz, which are the most relevant for earthquake engineers. A main issue is represented by the limited resolution of seismic wave speed models. Currently, one of the best approaches to retrieve high-resolution wave speed models relies on the combination of the Spectral Element Method (SEM) with adjoint-state methods within the framework of (3D) Full Waveform Inversion (FWI). Here we use the Salvus software to resolve the velocity structure of central Italy, which is exposed to high seismic hazard. We invert a dataset including 248 events (occurred between 2005 and 2019, with a magnitude ranging between 2.8 and 5.5) recorded by the Italian seismic network and provided by INGV. We implement a multi-scale approach to 3D-FWI by progressively incorporating the high-frequency content of waveform recording. In the first cycle of iterations, we retrieve the long-wavelength structure of the model. At this stage, we also perform several tests, e.g. different source/receiver cutouts, to properly tune the FWI configuration parameters to our problem. Then we use the best fitting model as the initial model of a second cycle of iterations, combining it with Vs30 and topographic data. By accounting for site effects, we maximize the level of detail in the starting model, thus preventing the FWI algorithm from being trapped in local minima. The retrieved high-resolution wave speed model will shed light on the velocity structure of central Italy, thus improving the seismic hazard assessment of this region.