Monitoring debris flow dynamics: insights from 4D-LiDAR observations in Ohya landslide, Central Japan

Abstract

Debris flows, characterized by their destructive potential, rapid velocities, and extensive runout distances, pose significant hazards. Recent technological advancements have enabled the monitoring of debris flows using 4D (3D + time)-LiDAR (Light Detection And Ranging) from automotive-grade LiDAR for serial production, which is essential given the rapid temporal and spatial changes associated with debris flows. However, the data acquisition remains insufficient, particularly regarding the runout and deposition characteristics of debris flows traversing natural channels, which have yet to be fully elucidated. In this study, we installed 4D-LiDAR alongside video cameras at two locations; upstream and downstream, in the Ohya landslide scar in central Japan. This setup aimed to capture the three-dimensional surface morphology of the debris flows during its runouts, development, and deposition phases. As a result, we observed differences in longitudinal and cross-sectional profiles, surface morphology, and sediment deposition between fully and partly saturated debris flows. The cross-sectional profile showed a convex shape at the front of partly saturated debris flow. The surface morphology of debris flow, as indicated by the mean and standard deviation value of slope gradient and roughness, revealed that the front and middle sections of the surges exhibited similar tendencies, while the tail section displayed a different trend. This difference suggests that the standard deviation of slope gradient and roughness of flow surface were influenced by flow turbulence, particularly in saturated debris flows. Additionally, particle size also affects the standard deviation of slope gradient in partly saturated debris flows that are fully covered by boulders. However, the movements of boulders during the 0.1 s analysis interval of the LiDAR data may have resulted in poor correlations between the particle size determined from video images. The deposition of the surge front formed inverse slope topography relative to the original channel bed. These reverse gradients were consistently present during debris flow deposition, indicating that the topography decreased the mobility of subsequent flows, leading to the backfilling of sediments in the debris flow channel. We conclude that 4D-LiDAR monitoring is effective in enhancing our understanding of the runout characteristics of debris flows.