GIS-based landslide susceptibility mapping using AHP, FMEA, and Pareto systematic analysis in central Yalova, Türkiye

OpenAlex İngilizce

Yalova is a region with high landslide risk due to its climate, terrain, geographical features, and geological structure. Landslides in the region are of critical importance to produce landslide susceptibility maps and risk management because of the material and moral damage they cause. This study aimed to produce a Geographic Information System (GIS)-based landslide susceptibility map (LSM) of the Yalova central region using the Analytic Hierarchy Process (AHP), Failure Mode and Effect Analysis (FMEA), and Pareto systematic analyses. Lithology, slope, aspect, distance from roads, NDVI (Normalized Difference Vegetation Index), distance from faults, land use, rainfall, distance from rivers, and elevation factors were used in the analyses. The FMEA, Pareto, and AHP methods were used in an integrated and sequential manner to determine the LCF weights. According to the analysis, lithology was identified as the most influential factor, with the highest weight at 29.5%, while appearance had the lowest weight at 1.5%. The LSM was generated by processing the weight values in the prepared LCFs. Based on the produced landslide susceptibility map, the study area was categorized as having 29.60% very low-risk, 29.51% low-risk, 24.04% moderate-risk, 13.18% high-risk, and 3.68% very high-risk. Regional planning should be undertaken according to the landslide risk categories, and appropriate measures should be determined for each level in advance. Upon comparing the produced LSM with the existing inventory, it was determined that 95.29% of the previously occurring landslides in the region were in risky areas on the LSM. The results demonstrate the integrated applicability of the AHP, FMEA, and Pareto methods and provide a more accurate weighting of LCFs compared to a single method. The approach used in this study can be easily adapted to different regions and can be used not only for landslide analysis but also for risk assessment studies in other disciplines.