Purdue University Graduate School
DiGirolamo Thesis Final_v1.pdf (1.85 MB)
Download file


Download (1.85 MB)
posted on 2022-07-29, 02:49 authored by Eleanor L Di GirolamoEleanor L Di Girolamo


Semi-aquatic mammals, such as American beavers (Castor canadensis), muskrats (Ondatra zibethicus), North American river otters (Lontra canadensis), and American mink (Neogale vison), often play important roles in their ecosystem. Beavers and muskrats can manipulate plant community structure through the use of woody debris and forbs. As mesocarnivores, North American river otters and American mink can also drive community structure through the predation. Traditionally, these species are monitored using sign surveys (i.e., walking transects and visually identifying scat, tracks, and latrines). Camera trapping has also been used to survey semi-aquatic species occupancy to a lesser extent. However, due to their almost exclusive use of edge habitat, they may be ideal species to camera trap. Another more recently employed survey method is environmental DNA (eDNA), which involves the extraction of DNA from environmental samples (such as soil, water, air, and snow) to determine species occupancy. In this study, I evaluate environmental DNA and camera trapping as survey methods for detecting semi-aquatic mammals around northeastern Indiana. In the first chapter, I used eDNA sampling and camera trapping to monitor seven sites for three weeks during March – May 2021 in order to determine the presence of American mink. I found that the naïve occupancy for each site was 0.86. Although the detection probability of eDNA was lower than that of camera trapping (0.25 and 0.36, respectively), the occupancy models created suggest that there was no difference in detection probability between the two methods. I also compared the cost and time spent per sample and found that both were 20% lower for eDNA than camera trapping. The results of my study suggest eDNA may be a cost- and time-effective method for surveying for American mink occupancy. The objective of my second chapter was to determine the number of camera traps required to obtain reliable data for detecting semi-aquatic mammals. A minimum requirement for number of camera traps would be useful knowledge for wildlife managers in terms of budgeting and resource management and could also help to refine current camera trapping methodologies. I camera trapped four ponds for four weeks during June – July 2021, varying the number of camera traps (1 – 5) used at each pond each week. I collected a total of 66,543 photos and detected one semi-aquatic mammal throughout the study period (Neogale vison). Due to the lack of semi-aquatic mammals detected, I could not perform any analyses.


Degree Type

  • Master of Science


  • Biological Sciences

Campus location

  • Fort Wayne

Advisor/Supervisor/Committee Chair

Scott Bergeson

Additional Committee Member 2

Mark A. Jordan

Additional Committee Member 3

Bruce A. Kingsbury