Non-Contact damage detection of drainage wells during the Mj7.6 Noto Peninsula earthquake by UAV-LiDAR technique

• Point clouds of drainage wells were successfully generated by UAV-LiDAR. • Deformation of vertical stiffener and well curb is detected by differential analysis. • Proposed method enables off-site damage detection for post-earthquake assessments. Landslide prevention facilities, particularly drainage wells, serve a critical role in managing groundwater levels and pore pressure during heavy rainfall. However, these structures can be compromised during seismic events, leading to potential secondary hazards. Conventional inspection methods, including visual monitoring, routine inspections and quantitative measurements, face significant challenges such as limited expertise, labor intensity, and ambiguous decision-making processes. To address these issues, this study explores a non-contact damage detection method using UAV-LiDAR point clouds, particularly focusing on drainage wells affected by the 2024 Noto Peninsula earthquake in Japan. The UAV-LiDAR technique overcomes limitations of conventional methods by capturing point clouds of drainage wells even when obstructed by lids or trees. The study examines damage types such as deformation of vertical stiffeners and well curbs, demonstrating that UAV-LiDAR can effectively identify these issues. It includes methods for estimating the maximum inclination angle of well lids and calculating eccentricity, achieving simple and supplemental inspection for deformation detection. The findings indicate that UAV-LiDAR offers a robust, non-contact solution for quick and quantitative assessment of drainage well conditions, which is crucial for effective post-disaster evaluation and maintenance. This approach promises to enhance the efficiency of inspecting and retrofitting landslide prevention facilities, contributing to better management of disaster-related risks.