Ranking uncertainties in EGF near-fault strong motion predictions: Application to the Middle Durance Fault (France)
Year: 2026
Type: preprint
Abstract: This study addresses the need for reliable near-fault ground-motion predictions in low-to moderate seismicity regions, focusing on a Mw 6 scenario on the Middle Durance Fault. The objective is to identify and rank the main sources of uncertainty controlling amplitudes and durations in Empirical Green’s Function (EGF) simulations. Ground motions are computed at two nearby stations—one located on rock and the other within a small sedimentary basin—using heterogeneous kinematic rupture models combined with a single EGF event. Results show that the overall variability of the simulated ground motions exceeds the typical total sigma reported by ergodic ground-motion prediction equations, highlighting significant epistemic uncertainties associated with EGF-based approaches. Using an ensemble machine-learning framework, we identify the EGF source parameters as the dominant contributors to variability. In particular, the EGF seismic moment and stress drop primarily control the ground motion motions, while the nucleation position of the main event has a secondary influence. Additional analyses combining three-dimensional numerical simulations of the main rupture with point-source simulations mimicking EGFs indicate that the impact of the configuration of the EGF dataset has a relatively small impact, mainly at frequencies below ~ 1 Hz. In this case, the spatial distribution of EGFs across the rupture area proves more important than their number. Notably, a single EGF located as close as possible to the station can already provide reliable predictions. Finally, comparison with the widely used Irikura-recipe EGF approach shows that methodological differences are small compared with the variability induced by uncertainties in the input parameters.
Source: Research Square
Institutions: Électricité de France (France), Centre National de la Recherche Scientifique, Université Gustave Eiffel, Institut de Recherche pour le Développement, Université Savoie Mont Blanc, King Abdullah University of Science and Technology, Commissariat à l'Énergie Atomique et aux Énergies Alternatives, CEA Cadarache
Language: English
Subfield: Civil and Structural Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
Funder: Agence Nationale de la Recherche
OpenAlex ID: W7204940968