SYNOPSIS   

Landslide Early Warning Systems (LEWS) must bridge geotechnical mechanisms and real-world decisions. This talk links rainfall infiltration, suction loss, groundwater rise, progressive failure, and post-failure debris-flow transformation with evidence from element tests, real-scale slope-failure experiments, debrisflow flume tests, and field monitoring. We then connect observation technologies—IoT sensors (moisture, pore pressure, deformation), UAV/3D pointcloud change detection, InSAR and satellite rainfall—with hybrid physics-based, AI models and digital-twin dashboards. The focus is how to design actionable thresholds, uncertainty-aware alerts, and operational workflows for roads, cut slopes, and communities. Finally, we discuss ASEAN’s rising landslide risk under intensifying extreme rainfall and land-use change, and discuss a staged implementation roadmap from minimum viable monitoring to scalable LEWS.


BIODATA OF SPEAKER

Prof. Naoki Sakai is a principal senior researcher in geotechnical and landslide risk at Japan’s National Research Institute for Earth Science and Disaster Resilience (NIED) and a professor of University of Tsukuba. He leads rainfall-induced slope failure experiments using a large-scale rainfall simulator and develops end-to-end landslide early warning concepts integrating lab testing, field monitoring, UAV/3D point clouds, satellite data, IoT sensing, and hybrid AIphysics modeling. He collaborates internationally across ASEAN and Europe on multi-hazard risk reduction and practical deployment of monitoring and decision-support systems.