*Refat B M Faisal1, Yuichi S HAYAKAWA 2
(1.Graduate School of Environmental Science, Hokkaido University, 2.Faculty of Environmental Earth Science, Hokkaido University)
Keywords:Unmanned aerial vehicle (UAV) , Structure from motion (SfM), NASA-DEM, DEM of difference (DoD), Cross-profiles, volumetric change estimate
Topographical changes in riverine floodplains need to be measured for assessing geomorphological dynamics and protecting floodplain areas, although topographic evaluation in the disaster-prone floodplain in South Asia has been limited due to the lack of multitemporal, high-definition digital elevation models (DEMs) derived from modern techniques including airborne laser scanning, structure-from-motion (SfM) photogrammetry accompanied with Unmanned Aerial Vehicle (UAV), and field-based mapping approaches. Here we carried out a preliminary study at two locations of the Teesta river (Brahmaputra’s tributary) in Bangladesh using the UAV-SfM techniques and generated high-resolution DEMs. The selected locations represent dynamic changes of sediment and water on the floodplain over years, and the UAV-SfM approach can be an effective method for monitoring those changes, but an archive of the past data has been unavailable. Here we evaluated the topographic changes by comparing the UAV-SfM-derived DEMs of 2022 with global DEM products (NASADEM of 1999), which are often the only available choice of DEMs in this river floodplain. The elevation differences of these two sets of DEMs were in the range of -5.23 to -84.66 m, and volumetric changes of -4.11±0.15 to -86.25±0.20 million m3, likely dominated by erosional processes where the elevation errors are supposed to be several meters for UAV-DEM and ca. 5–12 m for NASADEM. Although it is not easy to accurately evaluate the absolute values of the changes, these changes may be associated with the upper basin’s hydrology, sediment flux, and anthropogenic stressors along the floodplain. The considerable changes in topography, including cross-sectional profiles, elevation, and sediment volume, indicate the importance of continuous river topographic monitoring that can be facilitated from the present workflow of generating high-resolution DEMs and orthophotos. This study successfully demonstrates the effective applicability of DEM products (global and UAV-SfM derived) for evaluating the terrain changes and morphological dynamics of disaster-prone floodplains, which can optimize river management strategies in South Asia, particularly Bangladesh.