Détails de l’article

Titre : A Multi-Criteria Geospatial Framework for Enhanced Landslide Hazard and Risk Mapping in Southern California’s Coastal Zone

Auteur.e.s : Lila Reni Bibriven, .

Résumé : Landslide events pose significant socio-economic threats, highlighting the necessity for precise risk assessments. This research introduces an innovative approach to landslide susceptibility modeling, utilizing probability distribution functions and fuzzy Multi-criteria decision analysis. By integrating landslide occurrence frequency with geomorphological attributes, the study develops a robust geophysical risk model for the coastline areas of Ventura County, California. The findings reveal the effectiveness of a multi-tiered modeling process in generating high-resolution risk assessments, which are crucial for local-scale land-use planning and settlement decision making. The initial model indicated that the gamma operator yielded optimal results, with validation data showing AUC values between 0.79 and 0.882. The highest accuracy was achieved at a gamma value of 0.9, yielding an AUC of 0.882. Notably, 60-80% of landslides occurred in high-risk zones, confirming the model’s practical applicability. The final landslide risk model categorizes susceptibility into five levels, providing a clear framework for risk assessment and disaster preparedness, thereby enhancing effective land-use planning. Keywords: Landslide Susceptibility, Risk Probability, Fuzzy Multi-criteria Analysis, Weighted Combination, Risk Management.

Numéro 2