Landslide in Pasaman Regency

Landslides are the second most frequent natural disaster in Indonesia, particularly in Pasaman Regency. According to the 2022 Indonesian Disaster Risk Index, Pasaman is categorized as having a high risk for landslides, with a score of 35.19. In this study, we consider a range of parameters, including slope, slope aspect, curvature, land cover, geology, proximity to rivers, roads, faults, as well as soil units. Additional parameters, such as the average shear-velocity down to 30 m (Vs30), rainfall, Peak Ground Acceleration (PGA), Topographic Ruggedness Index (TRI), and Normalized Difference Vegetation Index (NDVI) are also incorporated. Furthermore, elevation, Topographic Wetness Index (TWI), and Sediment Transport Index (STI) are included to enhance the accuracy and comprehensiveness of the Landslide Susceptibility Model (LSM). Five machine learning algorithms Frequency Ratio (FR), Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), and eXtreme Gradient Boosting (XGBoost) are applied to develop a reliable and accurate LSM model. The performance of these algorithms is then evaluated through statistical tests and compared with actual landslide occurrences in the Pasaman region.