Using unmanned aerial vehicles to sample aquatic ecosystems

Abstract

In limnology, aquatic samplings are labor-intensive, and research locations can be hindered by limited accessibility. As lightweight unmanned aerial vehicles (UAVs) can provide enhanced spatial and temporal resolutions of ecological data as well as better accessibility to study areas, we questioned the potential of UAVs as an alternative limnologist's tool to overcome the limitations of current limnologic sampling techniques. In this study, we compared the reliability and effectiveness of UAV-based in situ readings (i.e., temperature and conductivity) and water sampling with two conventional methods (i.e., manual reading using a hand-held device and grab sampling and sensor reading). We found that UAV-based in situ readings better represented the spatial and temporal variations of thermal and chemical distributions in water bodies compared to manual and sensor readings. High data density within a relatively short-sampling time demonstrate the effectiveness UAV-based in situ readings. Additionally, UAV enhanced reliability of data by minimizing connection with water (e.g., boat movement). Unlike outstanding performance of UAV for in situ readings, however, UAV collected water samples appeared to be less representative of water chemistry compared to manually collected samples. This is likely due to small water sampling volume (25 mL) of UAV compared to over 1 L of water collection by manual sampling. This study confirms that UAVs hold high potential to improve our ability to collect reliable and fine scale in situ data, but carrying capacity of UAVs needs to be improved to be fully integrated into a limnological study.