Crustal heterogeneity and fault parametrization effects on seismic hazards assessment: southeastern margin of the Tibetan Plateau case study

X Rui,C A Williams,D S Stamps,L H Fang
DOI: https://doi.org/10.1093/gji/ggae418
IF: 3.352
2024-12-14
Geophysical Journal International
Abstract:SUMMARYAssumptions about lithospheric composition and fault discretization schemes are crucial for estimating seismic hazards. This study aims to examine the impact of these two variables in the Southeastern Margin of the Tibetan Plateau, with a specific focus on the Daliangshan mountain area. This area, located along the central segment of the Xianshuihe–Xiaojiang fault system, has been considered in the last few decades to host the highest seismic hazards in mainland China given its historical recurrence intervals of large magnitude earthquakes. Previous studies have constrained the kinematics of the region, including estimates of fault slip and accumulated moment magnitudes, assuming (1) the crust is compositionally homogeneous and isotropic and (2) the discretized fault patches (nodes) are coupling independent. In this work, we use constraint from an updated GNSS velocity solution comprised of 287 sites to test the influence of a compositionally heterogeneous crust and two fault patch (node) parametrizations to assess predicted fault slip rates and several seismic hazard quantities. The two fault node parametrizations are: (1) node depth-dependent (NDD; locking coefficients of nodes along a depth profile are dependent) and (2) node depth-independent (NDI; locking coefficients of all nodes are independent). Statistical tests (F-tests) indicate that the NDD model fits GNSS data significantly better than the NDI model even with less than ∼50 per cent of to-be-estimated parameters. We further examine the impact of lithospheric composition on regional kinematics by incorporating data from high-resolution seismic tomography. Comparisons between homogeneous and heterogeneous models for our preferred NDD fault locking parametrization scheme suggest that the material heterogeneity has minimal influence on the predicted recurrence interval estimates of large (e.g. M w = 7) earthquakes but some influence on discerning the details of the fault coupling distribution. These results show that the crustal medium and parametrization scheme have an impact on the description of fault kinematics, which should be considered as one of the sources of uncertainty in hazard assessment.
geochemistry & geophysics
What problem does this paper attempt to address?