Evaluating influence of rainfall frequency and precipitation on landslide susceptibility: A case study in Lam Dong, Vietnam

Abstract

Landslide susceptibility involves random topographical, geological, hydrological morphology, and hydrometeorology factors. Recent decades have seen alarming landslide hazards in Vietnam. Potential geological hazards subjected to extreme rainfall have threatened Vietnam's mountainous regions. This study evaluated the influence of rainfall on landslide susceptibility in Lam Dong province using machine learning algorithms coporated with the Analytic Hierarchy Process. Crucial impact factors were considered and analyzed in different rainfall scenarios. Landslide susceptibility maps developed correspond to frequency scenarios of antecedent rainfall data. The key findings revealed that rainfall conditions significantly influenced the landslide susceptibility in Lam Dong. The probability of rainfall - induced landslides could differ depending on the watershed due to rainfall behavior differences. The higher the rainfall intensity, the greater the risk of landslides. Steep hillsides of over 20% could be highly susceptible landslides to heavy rainfall intensity or more. Soil zones with high - complex slope conditions containing a high water content ratio and high swell index subjected to extreme rainfall could have a very high risk of landslides. Fuzzy probability could be a feasible tool to define crucial landslide - prone conditions regarding rainfall and slope and model the likelihood of landslide occurrence.