Exposure Database

Landslides in mountainous regions can enormously impact livingsocial, economical and infrastructural development. Landslide susceptibility maps (LSMs) are crucial for the identification of probable landslide areas and, in turn, help with sustainable urban planning but often overlook building exposure to landslides. This approach neglects the level of building exposure to landslides with respect to landslide susceptibility. Here, we introduce a framework to identify building exposure levels on the basis of based on LSMs created using Random Forest (RF), Support Vector Machine (SVM), and Multilayer Perceptron (MLP) models in Himachal Pradeshlandslide susceptibility in Himachal Pradesh. LSM was obtained via random forest (RF), support vector machine (SVM) and multilayer perceptron (MLP) methods. Building data integrated with susceptibility levels revealed that  By introducing the building data over the susceptibility, the exposure levels are used to identify potential buildings exposed to landslides. 35% of the buildings are exposed in Shimla district and 30%, 22%, and 21% in Kullu, Mandi, and Solan districts, respectively while in Kullu, Mandi and Solan districts around 30%, 22% and 21% buildings respectively are exposed tofaces landslides risk. However, Mmore than 90% buildings situated in Una, Hamirpur and Lahaul and Spiti are not exposed to landslide risk. Interestingly, frequent landslides reported many landslides event have been reported in Kangra and Kinnaur districts, only  but around 5% and 14% buildings are exposed to landslide risk indicating events landslides event happening in non-built up areas. Thus, our study aims to provide a framework for to identify the exposure of buildings to landslides, providing planners and decision makers with effective planning and mitigation strategies.