AUTHORS: Derek Whipkey, Cooperative Wildlife Research Lab, Southern Illinois University; Charlotte Narr, Department of Zoology, Southern Illinois University; Brent Pease, School of Forestry and Horticulture, Southern Illinois University; Guillaume Bastille-Rousseau, Cooperative Wildlife Research Lab, Southern Illinois University
ABSTRACT: Semi-aquatic mammals play important roles in aquatic ecosystems, particularly in the transfer of nutrients between aquatic and terrestrial environments. However, detecting and estimating the occurrence of these species can be difficult due to their elusive nature. Additionally, most occupancy studies have focused on a single type of waterbody (e.g. lentic or lotic systems), limiting a broader understanding of factors impacting these species distribution. To estimate occupancy of semi-aquatic mammals, we surveyed 67 sites across four counties in southern Illinois from March – May 2023 for American beaver (Castor canadensis), muskrat (Ondatra zibethicus), river otter (Lontra canadensis), and American mink (Neovison vison). Sites were distributed evenly between waterbody type and size. Given the elusive nature of these species, we combined two detection methods, sign surveys and camera traps, to increase detection. We applied a Bayesian multi-method occupancy model that incorporates both detection methods to estimate a single occupancy probability for each target species. To evaluate the relative importance of aquatic and terrestrial factors on occupancy, we built candidate models of aquatic and terrestrial covariates separately to identify the top model of each category. Aquatic covariates were consistently more important in predicting occupancy for all species. However, a combined additive model of the top aquatic and terrestrial models provided the best overall predictions. Beaver, otter, and mink occupancy showed positive associations with large waterbodies, while muskrat occupancy was positively linked to lotic systems. Additionally, muskrat and mink occupancy were positively related to increasing distance from roads. Our results suggest that while aquatic structure is more influential for predicting semi-aquatic mammal occupancy, considering both aquatic and terrestrial factors yields the most accurate results. All four semi-aquatic mammal species we studied were impacted by aquatic type, highlighting the importance of considering both waterbody types to better understand their distribution.
