Landslide susceptibility mapping along the Solan-Shimla highway using the analytic hierarchy process, Himachal Pradesh, India

Every year, landslides create a major risk to the economy as well as human life in Himachal Pradesh, India. Frequent landslides in the state especially on the major highways cause serious problem for trade, connection, massive disruption to traffic and public safety. Hence, assessing the landslide susceptibility mapping (LSM) along the highway routes can significantly help in safeguarding the people and property. The primary objective of this research is to prepare LSM on the basis of eleven causative factors to identify very high, high, moderate, low and very low susceptibility. For this, Remote Sensing and Geographical Information System (GIS) based on Analytic Hierarchy Process (AHP) approach is practiced in this study to prepare the LSM along National Highway-5 (NH-5) from Solan to Shimla, in the state of Himachal Pradesh (H.P.), India. All these factors are further sub-classified and weightages are given as per AHP technique. A landslide susceptibility map is prepared from the combined weighted raster thematic maps of each factor based on the assigned values and rating. LSM was categorized into five categories such as: (1.18–2.00) very low, (2.00–2.45) low, (2.45–2.90) moderate, (2.90–3.40) high and (3.40–4.50) very high by natural break classifier in ArcGIS environment. The model demonstrates good predictive efficacy, evidenced by a ROC-AUC of 0.835, a Precision-Recall AUC of 0.943, and an F1-score of 0.923 at a threshold of 2.664. Furthermore, 97.92% of identified landslides transpire within high and very high susceptibility categories. The generated LSM of this study highlight landslide area and assist local authorities and decision-makers to enhance preparedness and strengthen contingency planning.