The effectiveness of an unmanned aerial vehicle in controlled fecal pellet surveys

Authors: Israel D. Parker, Aaron N. Facka, Andrea E. Montalvo, Ian T. Gates, Brian L. Pierce, Roel R. Lopez

The declining costs and increasing capabilities of unmanned aerial vehicles (UAVs) have led to their expanded use in natural resources research and management. Generally, UAV-based data collection involves larger (i.e., more visible) components (e.g., large mammals, blocks of forest) that are more easily observed by UAV cameras. Little research has focused on UAV effectiveness in researching and monitoring relatively small and less visible objects. Fecal surveys are broadly applied methods for determining wildlife occupancy, population abundance and trends, and land use. Potentially, UAVs could improve, or augment, fecal surveys by reducing time and effort expenditures, expense, and impacts on focal species behavior. Yet, their effectiveness and ability to produce accurate and precise estimates have not yet been evaluated. We compared UAV surveys at multiple observation altitudes to traditional in-person on-the-ground surveys to test relative UAV effectiveness. We created artificial survey plots with a randomly assigned number of cereal pellets that mimicked the morphology of rabbit pellets. UAVs provided similar data to in-person counts for presence-absence inference. Additionally, raw counts were similar in pattern to in-person observations for pellets across a range of cover classes but were biased low in most circumstances. Heavy cover negatively affected both methods but resulted in higher undercounting with the UAV. The density of vegetation cover impacts pellet detection for both in-person and UAV-based surveys. Our research demonstrates that UAV-based fecal surveys are viable strategies. Further research in different conditions and fecal shapes is required for full implementation.

Suggested Citation