Spatial Analysis of Landslide Susceptibility Potential Driven by Rainfall Anomalies in the Ake Tiabo Sub-Watershed, North Halmahera

Abstract

This study aims to assess landslide susceptibility by integrating rainfall anomalies with topographic, geological, soil, and land-use factors using a spatially explicit approach. The Anomaly Percentage Method (APM) was applied to quantify deviations in rainfall from long-term climatological means, and the resulting anomaly indices were integrated into a Geographic Information System (GIS) framework through Weighted Overlay Analysis. The analysis incorporated geological maps, land-use data, soil type information, a Digital Elevation Model (DEM), and long-term rainfall records (1995–2024). Rainfall surfaces were generated using the Inverse Distance Weighting (IDW) interpolation technique. The results show that annual rainfall across the study area ranges from 2,000 to 3,000 mm, with the 2,500-3,000 mm class covering approximately 59.76% of the sub-watershed. The landscape is predominantly characterized by steep slopes (25–45%), underlain by young volcanic lithology that is highly susceptible to weathering, and largely dominated by mixed dryland agriculture. Landslide susceptibility mapping classified the area into four categories: low (16.75%), moderate (62.56%), high (19.98%), and very high (0.71%). Zones of high and very high susceptibility are primarily concentrated in areas with steep terrain, tuffaceous or alluvial lithology, and open or converted land cover. In conclusion, rainfall anomaly emerges as the most influential factor controlling landslide susceptibility in the Ake Tiabo Sub-watershed. The resulting susceptibility map provides a valuable scientific basis for spatial planning, disaster risk mitigation, and sustainable watershed management, particularly through targeted reforestation, agroforestry development, and land-use regulation in high-risk areas.