Landslides Susceptibility Mapping in R Program (Case study in Lima Puluh Kota Regency)

Abstract

Landslide susceptibility zonation is necessary to be considered in land use planning at various scales., different approaches and analytical methods can be used to evaluate and zone the area and processed with GIS software. However, there are constraints in its use, such as the cost of the licenses of software and source code that cannot be accessed and evaluated by users. The recent development of open-source software that can integrate data, analysis, and graphs in a representation such as the R program, has opened up opportunities for researchers to reevaluate and modify interpretation further from available ones to address issues. In this regard, this study aims to create functions in R using the Weight of Evidence (WoE) method, a form of bivariate statistic approach to acquire the significant factors controlling landslides and generate a susceptibility map. The case study is located in Limapuluh Kota Regency, West Sumatra Province of Indonesia, a hilly and mountainous region where its districts are prone to landslides. Eight of eleven factors such as geology, landform, land cover, elevation, density of vegetation greenness, slope, rainfall intensity, and proximity to stream were regarded to control landslides which set up four classes of landslide susceptibility zone (very low, low, moderate, and high).