Authors: Thomas Pingel*, Virginia Tech, Andrea Saavedra, Northeastern Illinois University, Lily Cobo, Northern Illinois University
Topics: Geographic Information Science and Systems, Remote Sensing, Water Resources and Hydrology
Keywords: Unmanned Aerial Systems, Structure from Motion, Digital Terrain Models, Groundwater Modeling
Session Type: Paper
Start / End Time: 2:00 PM / 3:40 PM
Room: Cabinet Room, Omni, West
Presentation File: No File Uploaded
The Yucatán Peninsula (YP) is considered a groundwater-dependent ecosystem, due to its reliance on aquifers for its supply of freshwater. Increasing population and use of the underlying aquifers to support the tourism industry in the YP has put pressure on its natural aquifer systems. One critical component of modeling groundwater effectively is its relation to surface elevations. In the YP, the best sources of comprehensive elevation data are Digital Elevation Models (DEMs) from the Shuttle Radar Topography Mission (SRTM) and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), but the error inherent in these products when applied to areas of dense vegetation with very low relief makes them less than ideal for modeling water flow in the YP. In contrast, Real Time Kinematic (RTK) GPS corrected surfaces derived from Unmanned Aerial Vehicle (UAV) based photogrammetry have the potential to vastly improve local measurements and provide correction parameters that can be used to locally calibrate the global elevation datasets.
This study uses UAVs equipped with cameras and precision RTK GPS units to create models of the near-ground environment, including the ground surface, trees, and exposed groundwater (cenotes). We present a novel workflow and algorithms to improve point cloud registration, water surface height estimation, and ground surface elevation using consumer grade RTK and UAV platforms, and provide error estimates for this method in low-relief, dense vegetation environments such as the YP.