Unmanned Aerial Vehicle (UAV) photogrammetry as a tool in aquatic ecosystem mapping, assessment and planning

Abstract: The understanding of aquatic ecosystems such as wetlands requires that they be examined and understood from a wide range of perspectives. Unmanned Aerial Vehicle (UAV) photogrammetry using inexpensive digital cameras has recently become a powerful tool that offers a viable alternative to traditional remote sensing systems, particularly for applications covering relatively small spatial extents. UAV photography has a high spatial accuracy needed by scientists and proof to be a valuable tool to enhance our understanding of aquatic ecosystems. Accurate planning derived from this technological advancement allows for more effective management and conservation of wetland areas. This thesis presents results of a study that aimed at investigating the use of UAV photogrammetry as a tool to enhance the assessment of wetland ecosystems. A baseline wetland delineation, classification and WET-Health assessment (baseline wetland assessment) was conducted first. Twenty ground control points (GCPs) were then positioned across the site to achieve geometrical precision and georeferencing accuracy. The XYZ location of each GCP was recorded using a Trimble SPS985 GNSS GPS. The UAV images were collected during a single flight within 2½ hours over a 100 ha area at the Kameelzynkraal farm, Gauteng Province, South Africa. An AKS Y-6 MKII multi-rotor UAV and a NIKON D3200 (28 mm) digital camera on a motion compensated gimbal mount were utilised for the survey. Structure from Motion (SfM) computer vision techniques were used to reconstruct the camera positions, terrain features and to derive ultra-high resolution point clouds, orthophotos and 3D models from the multiview photos using Agisoft Photoscan Professional Version 1.1 software. A spatial resolution of up to 0.018 m was achieved. The results of the geometric accuracy of the data based on the 20 GCP’s were 0.018 m for the overall, 0.0025 m for the vertical root mean squared error (RMSE) and an overall root mean square reprojection error of 0.18 pixel. The UAV products were then edited using Photoscan and Quick Terrain Modeller (QTM) 805. The edited products were subsequently analysed, interpreted and key attributes extracted using a selection of tools/ software applications including QGIS 2.2.0 and 2.12.0 application, Google Earth, QTM 805 and Global Mapper v17. The UAV products were then applied to conduct a wetland delineation and WET-Health assessment (UAV wetland assessment). The baseline wetland assessment results were then compared with the UAV assessment results to assess if the latter enhanced wetland delineation and WET-Health. The UAV products provided a valuable enhancement to the wetland delineation and classification which would have been difficult to achieve using field studies alone. UAV photogrammetry was successfully applied to determine the landscape setting (terrain and geomorphic), obtain precise slope profiles, assisted with the identification of areas of saturation and water accumulation, mapping of hydrophilic vegetation including surface water sources and surface hydrodynamic analysis. UAV photogrammetry further enhanced the WETHealth assessment allowing wetland practitioners to better understand the degradation of the study area where all the wetland indictors were not that apparent by providing accurate data that can assist with decision making.M.Sc. (Aquatic Health