Rainfall-induced landslides are significant threats to human lives and assets globally. In the absence of a localised community-centred early warning system (EWS) in many areas of Asia, the proposed project aims to develop effective approaches, i.e. tools and methodologies in establishing a “Landslide Early Warning (AppLEW) System” by partially utilising AI/machine learning technology. The local warning capacity includes two fundamental pillars: (1) a set of localised monitored instruments installed at critical sites to provide real-time hazard data, and (2) trained local warning teams responsible for detecting early signs of coming hazards and initiating warning and responses. The implementation and operation of the AppLEW system is supported by scientific and technical components, e.g., hazard monitoring and detection, risk assessment, and capacity development for self-warning and self-response.

The AppLEW project will be conducted in collaboration between Japan, Vietnam and Malaysia. The novel localised community-centred AppLEW System will demonstrate effective use of emerging technologies that are accessible and operable by local stakeholders (e.g. local residents and warning teams). Its science-based recommendations, as well as self-help approaches or “self-warning and self-response” capacity-building programs, could empower the stakeholders to proactively detect and recognise early signs; issue self-warnings; and respond to future hazards autonomously.