Seismic ground motion modelling

Within-event ground motion residuals for the M7.8, 2023 Kahramanmaras earthquake, considering spectral ordinates at 2s against rupture distance (Weatherill, 2024). (Residuals are computed for both empirical and simulated data using the ground motion model implemented in the European Hazard Model ESHM20. Computations are performed using eGSIM for the DT-Geo project.)
Within-event ground motion residuals for the M7.8, 2023 Kahramanmaras earthquake, considering spectral ordinates at 2s against rupture distance (Weatherill, 2024). Residuals are computed for both empirical and simulated data using the ground motion model implemented in the European Hazard Model ESHM20. Computations are performed using eGSIM for the DT-Geo project. (Section 2.6)

Background

Seismic ground motion modeling is essential for assessing earthquake hazard and risk, providing critical information for infrastructure design, emergency planning, and seismic resilience. Ground motion is influenced by multiple factors, including source, path, and site effects, each contributing to its complexity and variability. Source effects refer to the earthquake rupture process, where parameters such as magnitude, stress drop,  and rupture directivity control the generated seismic waves. Path effects involve the attenuation and scattering of seismic waves as they travel through the Earth. This is influenced by factors such as geometrical spreading, anelastic attenuation, and crustal heterogeneities. The way attenuation is modeled can significantly impact ground motion predictions, highlighting the need for robust attenuation corrections. Site effects describe how local geological conditions modify seismic waves and the distribution of seismic energy across frequencies. 

Understanding and reducing ground motion variability is a key challenge in seismic hazard assessment. The observed variability in ground motion records arises from both physical processes (e.g., stress drop differences, path-dependent attenuation, and site response) and statistical uncertainty due to limited data. Ultimately, improved ground motion models enhance the reliability of probabilistic seismic hazard assessments (PSHA), leading to more effective risk mitigation strategies and better-informed seismic design codes. Ongoing research in Section 2.6 aims to integrate machine learning, large-scale simulations, and new observational datasets to refine these models further.

Scientific key questions

  • How can we improve empirical ground motion models to better capture regional effects?
  • How can physics-based simulations contribute to the refinement of empirical ground motion models?
  • What are the dominant physical sources of aleatory variability in ground motion, and can they be reduced through improved modeling?
  • How do site conditions (e.g., nonlinear soil response) and rupture geometries (e.g. directivity effects) influence ground motion predictions across different frequency ranges?

Related projects

Data, Products, Services

back to top of main content