Date of Award




Document Type

Master's Thesis

Degree Name

Master of Science (MS)


Department of Geography and Planning

Content Description

1 online resource (viii, 88 pages) : color illustrations, color maps.

Dissertation/Thesis Chair

Alexander Buyantuev

Committee Members

Steven Campbell, Daniel Bogan


Drones, Unmanned Aerial Systems, Unmanned Aerial Vehicles, White-tailed deer, Wildlife Conservation, Wildlife Management, Drone aircraft in remote sensing

Subject Categories

Natural Resources and Conservation | Natural Resources Management and Policy | Remote Sensing


Unmanned aerial systems (UAS) have seen recent advancements in technology that gave rise to their increasing use in recreational and commercial application, including wildlife conservation. Adaptive management is a must for wildlife conservation, with the goal of learning from management decisions to improve future management strategies, especially in the face of growing human related stressors such as climate change and habitat loss. Monitoring is a critical step for adaptive management, as it allows the manger to learn about the ecology of the natural system and quantify the impacts of management strategies. Species and habitats are frequently monitored for wildlife conservation purposes. The demand for more detailed and real-time monitoring along with the limits of conventional monitoring techniques gave rise to the growing use of UAS in wildlife monitoring. UAS have been used to monitor the abundance and habitats of a variety of different species across the globe. One species that has been extensively managed and monitored in North America is the white-tailed deer (Odocoileus virginianus). There have been many methods to assess white-tailed deer abundance and the relationships with their habitats, however the overall effectiveness of these methods is debated due to issues with accuracy and cost. A new method using UAS equipped with thermal-infrared (TIR) cameras has recently been explored and shown to improve upon detection accuracy and lower the overall cost of monitoring. Very few studies have explored this approach and have only touched the surface for how UAS can be used to monitor deer ecology. The major goal of my study was to show how TIR UAS can be used in a real-world application to collect and assess abundance data in a small study area in unison with a UAS-derived habitat assessment to assess habitat use. I performed 34 total TIR UAS surveys from March 6th to May 28th, 2020 and detected 413 total deer, with an overall average abundance of 12.14 individuals per survey and a 95% CI range of 9.8-15.1 deer. The observed abundance in the study area ranged from 1 to 30 deer over the entire course of the study with a median of 11.5 deer, an interquartile range of 8-16 deer, and a standard deviation of 7.12 deer. I found no significant relationships between the abundance observed during a survey and the ambient temperature (p=.0375) or wind-speed (p=0.638) during the survey. I also found no significant difference in the abundance measured in March, April, and May (p=0.09). I assessed the observed variation in abundance based on sampling intensity using a bootstrap analysis and determined that 20 surveys would have been adequate to capture the variation of abundance on the landscape. My habitat assessment using UAS imagery showed that deer had an selection for the pitch pine dominant forests and pitch pine scrub oak barrens/grassland in the study area over the entire study duration, while pitch pine dominant forests were selected for when the temperature was less than 4oC and pitch pine scrub oak barrens/grasslands were selected for when temperatures was higher than 4oC. Selection for habitats also varied by month, suggesting that seasonality also plays a role in habitat use. Overall, my study tested a novel use for UAS-based white-tailed deer monitoring and demonstrated high potential for the method as a monitoring tool. I then use the experience I gained while developing my TIR UAS survey approach along with findings from relevant literature to discuss the considerations for developing a TIR UAS survey and make recommendations for others than plan to develop their own TIR UAS survey approach. Recommendations include the use of rotor-based UAVs rather than fixed wing UAVs for slower flight speed, using a visible camera simultaneously with a TIR camera to aid in species identification, using additional observers to maintain line of sight with the UAS and increase the size of the flight plan and coverage of the survey, and performing enough surveys in the same area to adequately assess the natural variation of deer on the landscape to prevent bias from under-sampling.