A Review of Rainfall Thresholds for Landslide Occurrence in the Hindu Kush Himalaya (HKH)

Abstract

The Hindu Kush Himalaya (HKH) region is one of the world's most landslide-prone areas because of its rugged topography, complex geology, and intense monsoonal rainfall. Rainfall thresholds are widely used to forecast rainfall-induced landslides and form a key component of landslide early warning systems. This study presents a comprehensive review of rainfall-threshold research across HKH countries, including Nepal, India, China, Bangladesh, Bhutan, Afghanistan, Pakistan, and Myanmar. A bibliometric analysis using the Scopus database was complemented by a systematic review of peer-reviewed literature from Scopus, Web of Science, and Google Scholar. A total of 54 studies defining 61 rainfall thresholds were identified and evaluated based on publication trends, dataset characteristics, threshold formulations, derivation methods, validation procedures, and operational applications. The results show a substantial increase in rainfall-threshold research during the past three years, with most studies conducted in India and China. Empirical and statistical approaches dominate, whereas physically based models remain limited because of their high data requirements. Most thresholds are derived from 5–25 years of daily rainfall records, with Intensity–Duration (I–D) and Event–Duration (E–D) relationships being the most common formulations. The review identifies major gaps in data reporting, threshold derivation, and validation, while only a few studies demonstrate operational implementation in early warning systems. No region-wide rainfall threshold currently exists for the HKH, highlighting the need for standardized methodologies, robust validation, long-term monitoring datasets, and stronger integration into operational landslide early warning systems.