Probabilistic Seismic Hazard Map for Bangladesh Including the Smoothed Background Seismicity and Local Site Effects
摘要
Because of its geotectonic location at the boundary between the Indian plate and the Burma subplate, Bangladesh and its surrounding regions have historically experienced numerous devastating earthquakes. Seismic hazard maps are important for mitigating seismic risk and disasters in the context of sustainable development. In this study, we used a classical probabilistic seismic hazard assessment approach to calculate the peak ground acceleration (PGA) and spectral acceleration (SA) for 10% and 2% probability of exceedance (PE) in 50 years. We utilized area seismic source and smoothed grid seismic source models, Vs30 local site effects, and seven ground motion prediction equations for the shallow crust and subduction zone. Seismicity parameters were calculated using the maximum likelihood method from the declustered and unified earthquake catalogs. The standard logic tree framework was used to integrate different source models and minimize epistemic uncertainties. The sensitivity of input parameters on hazard assessment and the uncertainty bands (fractiles) of logic tree branches were also evaluated. The result showed a similar lateral distribution of PGA or SA values followed the background seismicity. The ranges of PGA values for 10% and 2% PE were 0.15–0.65 g and 0.31–1.35 g, respectively. The SA values varied from 0.74 to 2.54 g, and from 0.39 to 2.03 g at 0.2, and 1.0 s, respectively, at 2% PE. The eastern and western regions are zones with the highest and lowest hazards, and site conditions affect these hazard values. These results support to sustainable, earthquake-resilient development in this tectonically active region, which lies between the Indian plate and the Burma subplate. To more accurately capture the spatial hazard variations, future seismic hazard assessments should incorporate region-specific Ground Motion Prediction Equations, refined Vs30 values and updated seismic source zonation within a logic-tree framework.