Abstract
Rainfall-triggered landslides pose a major threat to Central America and other landslide-prone regions. The development of new tools to understand and characterize landslide-triggering rainfall is a key aspect in the context of hazard assessment and risk mitigation strategies. The purpose of this study is to present a new conceptual framework in which a connection is established between rainfall thresholds and landslide susceptibility, and to suggest three potential applications of this framework: (1) an empirical estimation of the probability of landslide-triggering rainfall; (2) a threshold-based time-trend analysis of rainfall series; (3) a susceptibility-based early warning system. The applications are illustrated with cases and data from tropical climates, with particular emphasis on cases with climatological and geological settings comparable with those in Central America.