Date of Award
December 2024
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Aviation
First Advisor
Mark Dusenbury
Second Advisor
Robert Newman
Abstract
For decades, wildlife researchers have taken advantage of an aerial perspective for performing population counts because of challenges associated with ground observations. Due to advances in aviation technology and regulations, unoccupied aerial systems (UAS) have become increasingly available for wildlife population monitoring in addition to manned aircraft, with success largely dependent on the species, their habitat, and UAS utilized. This study compares a currently accepted method of counting North American elk (ground/spotting scope counts) to a novel method (UAS) to evaluate if a commercial-off-the-shelf quadcopter (COTS) UAS equipped with zoom-capable RGB cameras would be an acceptable tool for counting elk in Boulder County, Colorado. In addition, this study explored the operational challenges of utilizing a COTS UAS, such as disturbance to the elk, weather, and regulatory limitations. The primary data collected for this study were numbers of elk coincidentally counted by ground and aerial methods. By calculating and analyzing the differences in count numbers between the two methods, we sought to reveal if counts numbers varied depending on the collection method. Twenty-one comparative observations were analyzed with herd sizes ranging from six to 181. On average, 28 more elk were found with the UAS than by the ground method with the differences ranging from zero to 129 elk. Significance was found in the linear relationship between the aerial count and ground count, indicating the results were not random (p < .001). In addition to the method of collection, insight into the effects of field condition factors and how they impact the counting method was also explored through a generalized linear model selection process. Herd size and instances where terrain blocked the view of the herd for the ground observer were found to be significant factors in the count differences. A content analysis was performed to further evaluate the link between the conditional factors and the count differences. It was found that herd bunching and a level field of view for the ground observer were other sources of major variability in the coincident counts.Additionally, the effects of the drone on elk behavior were analyzed and were found to be minimal, even when directly overhead. This general lack of response is likely a result of these herds’ acclimation to human activity due to the suburban nature of the study area. Overall, the aerial perspective and agility of the UAS proved to be highly useful for obtaining high-quality images of elk, especially when the ground view was not ideal. This study has shown that a COTS UAS with zoom capability should be implemented as a standard tool for ongoing elk population counts in Boulder County, with use expanded to additional elk herds.
Recommended Citation
Wussow, Michael, "Evaluating UAS As A Tool For Improving Elk Population Counts In A Suburban Forested Area Of North Central Colorado" (2024). Theses and Dissertations. 6584.
https://commons.und.edu/theses/6584