A Geographical Information Systems-Multi Criteria Decision Analysis Integrated Landslide Risk Index For Yatiyanthota Divisional Secretariat Division, Sri Lanka, Using Geospatial Techniques

A landslide is one of the most destructive and frequent environmental disasters on the Earth's surface. They are a common disaster in many districts of Sri Lanka, including Kegalle, particularly during the south-west monsoon season. This research developed a landslide risk index by combining 25 criteria across six main variables (climate, geology, topography, demography, socioeconomic, and coping factors) that contribute to landslide occurrence in the Yatiyantota divisional secretariat division (DSD) utilizing Geographic Information Systems (GIS) integrated Multi-Criteria Decision Analysis (MCDA) approach. This approach incorporated the Analytic Hierarchy Process (AHP) and employed the Weighted Linear Combination (WLC). Factor and criteria weights were determined using AHP, while factor class ratings were assigned based on logical judgment informed by expert surveys. The results found that 3% (5.6 km2) of the area is prone to very high risk, while 40% (67.9 km2) falls into the high-risk category. Meanwhile, 18% (29.1 km2) was identified as very low and low-risk categories. Moderate-risk areas are also significant, accounting for 39% (66.6 km2) of the total area. The model validation results demonstrated that the resulting map closely aligns with real-world conditions, achieving 72.2% overall accuracy through Receiver Operating Characteristics (ROC) curve analysis. Accordingly, the GNDs of Mattamagoda, Pahala Garagoda, Amanawala, Berennawa Malwatta were identified to be the most vulnerable to landslide occurrences. Planning authorities should implement remedial measures to minimize landslide risks through various short-term and long-term strategies, utilizing the warning signals identified in the survey.