Study on the influence of complex fault geometry on earthquake risk assessment based on physical simulation: a case study of the Zhujiangkou Fault in the Guangdong–Hong Kong–Macao Greater Bay Area

Dynamic rupture simulations offer a physics-based approach to model earthquake processes and assess seismic hazards. While these simulations rely on stress state, friction, and geometry, the influence of fault geometry on societal risk remains understudied. This study develops 3D dynamic rupture models of the Zhujiangkou Fault, incorporating geometric complexities to evaluate seismic hazards and socioeconomic impacts. The analysis integrates physical, socioeconomic, exposure, and vulnerability factors, with uncertainty assessed using empirical ground motion prediction equations (GMPEs). Results indicate that the logarithmic standard deviation of near-field peak ground velocity in most simulations is 0.3–0.4, outperforming GMPEs. Geometric complexity affects pre-stress on fault surfaces, thereby affecting rupture termination and slip behavior. Stepover zones impede rupture, while fault strikes significantly impact hazard and loss distributions. Calculated scenarios indicate a low probability of Mw ≥ 7.0 earthquakes. High-risk zones concentrate in the eastern Guangdong–Hong Kong–Macao Greater Bay Area (GGBA) coastal areas, particularly in Guangzhou, followed by Dongguan, Shenzhen, and Hong Kong. Assessing economic losses is less uncertain than fatalities because assets are more uniformly distributed than people, while concentrated populations amplify the impact of spatial variability in seismic hazards. These findings provide critical insights for seismic risk mitigation and urban resilience planning in the GGBA.