Landslides, a prominent geohazard, cause considerable disturbances in many natural terrains, impacting both ecosystems and human habitats. In recent years, the intervention of climatic and tectonic activities has increased the frequency of such hazards. Although numerous methodologies have been developed to analyse landslide susceptibility, there remains a pronounced gap in probabilistic slope stability techniques incorporating rainfall infiltration models on a regional scale. The study proposes a new tool TRIGRS-FOSM (Transient Rainfall Infiltration and Grid-based Regional Slope-Stability-First Order Second Moment) developed to account for the uncertainties in soil shear strength properties along with the effect of vegetative cover using probabilistic infinite slope stability analysis. This user-friendly tool is seamlessly integrated with the infiltration model of TRIGRS and adaptable with Geographic Information Systems (GIS), enabling assessment across larger regions. It is especially tailored for regions prone to rainfall-induced landslides such as the Western Ghats of India, which has been under persistent threat due to increasing rainfall. This paper aims to validate the efficacy of TRIGRS-FOSM in the Western Ghats, contrasting it with traditional methodologies and focusing on the landslide prediction accuracy, especially within the Wayanad and Idukki districts of Kerala. On verifying TRIGRS-FOSM against Monte Carlo Simulations, it was observed that TRIGRS-FOSM exhibited a lower relative error for typical ranges of variability associated with soil material properties, underlining its enhanced reliability. Furthermore, the probabilistic approach showcased improvements over the deterministic method, elevating the prediction accuracy by 10% in Wayanad and 14% in Idukki districts based on their AUROC values. Through TRIGRS-FOSM, this work intends to provide a computationally efficient method to account uncertainties of landslide susceptibility assessment, thereby making a substantial contribution to geohazard management.

National Highway G559 is the first highway in Southeast Tibet into Motuo County, which has not only greatly improved the difficult situation of local roads, but also promoted the economic development of Tibet. However, rainfall-induced shallow landslides occur frequently along the Bomi-Motuo section, which seriously affects the safe operation and construction work of the highway. Therefore, it is urgent to carry out geological disaster assessment and zoning along the highway. Based on remote-sensing interpretation and field investigation, the distribution characteristics and sliding-prone rock mass of shallow landslides along the Bomi-Motuo Highway were identified. Three-dimensional stability analysis of regional landslides along the Bomi-Motuo Highway under different rainfall scenarios was carried out based on the TRIGRS and Scoops3D coupled model (T-S model). The temporal and spatial distribution of potential rainfall landslides in this area is effectively predicted, and the reliability of the predicted results is also evaluated. The results show that: (1) The slope structure along the highway is mainly composed of loose gravel soil on the upper part and a strong weathering layer of bedrock on the lower part. The sliding surface is mostly a circular and plane type, and the main failure types are creep-tensile failure and flexural-tensile failure. (2) Based on the T-S coupling model, it is predicted that the potential landslide along the Bomi-Motuo Highway in the natural state is scattered. The distribution area of extremely unstable and unstable areas accounts for 4.92% of the total area. In the case of extreme rainfall once in a hundred years, the proportion of instability area (Fs < 1) predicted by the T-S coupling model 1 h after rainfall is 7.74%, which is 1.57 times that of the natural instability area. The instability area (Fs < 1) accounted for 43.40% of the total area after 12 h of rainfall. The potential landslides were mainly distributed in the Bangxin-Zhamu and the East Gedang section. (3) The TRIGRS and T-S coupling model is both suitable for predicting the temporal-spatial distribution of rainfall-induced shallow landslides, but the TRIGRS model has the problem of over-prediction. The instability area predicted by the T-S coupling model accounted for 43.30%, and 74% of the historical landslide disaster points in the area were correctly predicted. (4) In terms of rainfall response, the T-S coupling model shows higher sensitivity. The %LRclass (Fs < 1) index of the T-S coupling model is above 50% in different time periods, and its landslide-prediction effect (%LRclass = 78.80%) was significantly better than that of the one-dimensional TRIGRS model (%LRclass = 45.50%) under a 12 h rainfall scenario. The research results have important reference significance for risk identification and disaster reduction along the G559 Bomi-Motuo Highway.

In subtropical typhoon-prone regions, landslides are triggered by short-duration intense rainfall and prolonged periods of elevated pore-water pressure. However, fast-moving landslides pose a significant challenge for timely warning because of insufficient data on rainfall triggers and the identification of potential failure sites. Thus, our study introduces an integrated approach that combines a double-index intensity-duration (I-D) threshold, accounting for daily rainfall (R0) and 5-d effective rainfall (R5), with the MC-TRIGRS, a probabilistic physically based model, to analyze fast-moving landslide hazards at a regional scale. This approach is characterized by its innovative features: (i) it employs a double-index model to categorize rainfall events, differentiating between long-term continuous rainfall and short-term intense precipitation; (ii) it utilizes a comprehensive dataset from extensive field investigations to implement the grey wolf optimizer (GWO) -enhanced long short-term memory neural network (LSTM) to predict soil thickness distributions across the study area; and (iii) it adopts the classical Monte Carlo method to calculate failure probabilities under various rainfall scenarios, incorporating randomness in key soil parameters, such as cohesion and internal friction angle. By leveraging geotechnical data from both field and laboratory tests and integrating the accumulated knowledge, these models can be applied to the coastal mountainous basins of Eastern China, a region highly prone to landslides. Our goal was to augment the effectiveness of landslide early warning systems. Particularly, the synergistic use of rainfall empirical statistics and probabilistic physically based slope stability models is poised to bolster real-time control and risk mitigation strategies, providing a robust solution for short-term preparedness.