Laboratory Application of Real-Time LiDAR Technology for Assessing Breach Morphology during the Failure of Earth Dams and Fluvial Dikes

Embankments such as fluvial dikes or dams may fail due to breaching caused by overtopping or internal erosion. The mechanisms associated with the temporal evolution of embankment breaching are still not fully understood. A deeper knowledge can be attained if novel cost-effective experimental techniques are applied in laboratory studies. Contributing to this general aim, this study presents the application of light detection and ranging (LiDAR) technology in medium-scale experimental models of earth dams and fluvial dikes failures. This technology enables the complete 3D digital reproduction of the embankment structures at various stages of failure using point clouds generated by a sensor with an integrated depth camera operating in the near infrared spectrum and using the time-of-flight (ToF) principle. This study is aimed at discussing the errors associated with the ToF principle, through a benchmark analysis involving structure from motion (SfM), and at discussing its application to real-time laboratory-scale embankment breaching. Emphasis is placed on the challenges associated with mapping the water surface and refractive index correction when the structures are submerged. The applications show the potential of ToF-based real-time monitoring to quantify the volumes of material eroded by hydraulic erosion and mass detachments. Two different laboratory embankment overtopping experiments were conducted to verify the ToF principle. In the first, the flow was stopped at different instants, and the reservoir was drained to allow scanning for 3D digital reconstitution. In the second, a dike breach evolution was continuously scanned at five-point clouds per second to enable near-continuous 3D reconstitutions. This study shows that SfM yields the highest point density per measurement, but that ToF has the same measurement error range. Unlike SfM, ToF enables realistic real-time characterization and quantification of mass detachments, allows for reconstruction of submerged 3D structures, and enables precise identification of the water-free surface position.