Examining the Impact of Landslide Hazards in the PeriUrban Areas by Global and Local Modelling Techniques

Abstract

Landslides can be considered one of the most severe natural hazards globally, and their management has a key role to inman safety. Landslide susceptibility maps can help the sustainable management of peri-urban areas such as determining the target areas of projects to develop landslide resistant areas, forest planning, infrastructure planning, and land use zoning. This study aims to analyse the relationship between landslide occurrence and its determinants in a peri-urban region, namely Taşlıdere Basin located in Güneysu District of Rize, Turkey, through providing both global and local regression models. To consider spatial non-stationarity, geographically weighted regression was used as a local model. It was found that the local model outperformed the global model in the estimation of landslide susceptibility determinants. The spatial analysis results indicated that almost all variables had heterogeneous variation over the study area. Therefore, this study provides a methodology for understanding the local dynamics of landslide occurrence.