Integrating social vulnerability and landslide susceptibility for spatial risk assessment in the Langat basin, Malaysia

Landslides are among the most significant natural hazards in Malaysia, threatening human lives, infrastructure, and socioeconomic development. This study developed an integrated landslide risk assessment framework by combining landslide susceptibility modelling and social vulnerability analysis in the Langat Basin, Malaysia. Landslide susceptibility was modelled using the Frequency Ratio (FR) approach based on six conditioning factors: elevation, slope, aspect, geology, land use, and distance to river. Social vulnerability was quantified using eight demographic and socioeconomic indicators to construct a Social Vulnerability Index (SoVI) at the level of the 14 spatial units. The SoVI results revealed considerable spatial variation, with Lenggeng (0.597), Nilai (0.586), and Putrajaya (0.475) exhibiting the highest vulnerability, whereas Morib (0.202), Sungai Pelek (0.206), and Sijangkang (0.216) showed the lowest. ROC analysis yielded an Area Under the Curve (AUC) of 0.800 for distinguishing the sampled landslide locations from randomly generated background points. Because the same landslide inventory was used to calculate the Frequency Ratio weights, this result should not be interpreted as independent validation of predictive performance. Integrating the SoVI and susceptibility maps in a Geographic Information System (GIS) produced a relative Landslide Risk Index (LRI) map, in which 38.91% of the basin was classified as Very Low risk and 13.64% as High to Very High risk. High-risk zones were concentrated where elevated landslide susceptibility coincided with socially vulnerable communities. The proposed framework provides a practical approach for supporting disaster risk reduction, sustainable land use planning, and community resilience in line with the Sendai Framework and Sustainable Development Goal 11.