Integrating GIS-based and AHP-enhanced approach for identifying earthquake zones in Northwest Vietnam
摘要
This study presents an Analytic Hierarchy Process (AHP)-enhanced approach by integrating Garson’s algorithm with an 8-18-1 Multi-Layer Perceptron (MLP) neural network to improve the identification of seismic sources. In this study, geological, geomorphological, and geophysical factors are categorized into two distinct groups including (1) indicators of subsurface fault existence, including surface fault density, Bouguer gravity gradient, magnetic anomaly gradient, and crust thickness gradient and (2) indicators of fault activity, comprising vertical movement gradient, drainage density, mountain front sinuosity, and landslide/hot mineral water density. The proposed method demonstrates effectiveness in identifying active subsurface fault zones and calculating the Seismic Susceptibility Index (SSI), thereby providing a robust framework for seismic source mapping. The results show 11 seismic source zones in Northwest Vietnam including SZ1 – Muong Nhe, MOb.Max = 5.3; SZ2 – Muong Te, MOb.Max = 4.9; SZ3 – Lai Chau - Dien Bien, MOb.Max = 5.2; SZ4 – Dien Bien – Sam Nua, MOb.Max = 6.8; SZ5 – Ma River, MOb.Max = 4.8; SZ6 – Son La, MOb.Max = 6.7; SZ7 – Tuan Giao, MOb.Max = 5.1; SZ8 – Da River, MOb.Max = 5.3; SZ9 – Phong Tho - Muong La - Bac Yen, MOb.Max = 6.4; SZ10 – Nam Tong, MOb.Max = 4.7; SZ11 – Mu Cang Chai, MOb.Max = 4.9.