Deriving bathymetries from unmanned aerial vehicles: a case study of a small intermittent reservoir

Small reservoirs have traditionally been used worldwide to store surface water, but they are complex to manage. Accounting for a water resource scattered across thousands of water bodies remains challenging. Images acquired using unmanned aerial vehicles (UAVs), which are already used in different ways in hydrology, could provide additional support. We evaluated the use of UAVs as an alternative way to derive bathymetric data from the topography of a small reservoir. Digital elevation models (DEMs, 58) were produced from UAV imagery and in-situ differential global positioning system (DGPS) points using standard structure from motion (SfM) and interpolation methods. Image processing was able to produce DEMs with 10 times the vertical accuracy of DGPS-point interpolation methods for similar accuracy in determining the relationships between the water height, surface area, and reservoir volume. These results suggest that the SfM photogrammetry can be used to derive bathymetry data from UAV images.