Mammalogy III

Contributed Oral

Recursive Foraging Movements of Coyotes in the Southeastern United States
Jordan Youngmann, Joseph Hinton, Nicholas Bakner, Michael J. Chamberlain, Gino D’Angelo

Coyotes (Canis latrans) colonized the southeastern United States over the last century as large predators, including the red wolf (Canis rufus) and eastern cougar (puma concolor), were extirpated from the eastern United States ecosystem. As a generalist carnivore, coyotes prey on white-tailed deer (Odocoileus virginianus) as well as a range of small mammals and plants in the Southeast. While resource selection by coyotes has been well documented at the home-range scale, little is known about their foraging behavior, which is an important factor in understanding the full impact of this predator in the Southeast. We assessed 3rd-order resource selection of foraging coyotes at sites across Alabama, Georgia, and South Carolina. Using GPS collars, we tracked 41 resident coyotes across 4 seasons during 2015-2016 and identified forage patches using recursive analysis where individuals repeatedly returned to known locations. We found that resident coyotes selected for open, early successional habitats throughout the year, while avoiding primary and secondary roads. Additionally, residents avoided forested landscapes while selecting for forest edges except from April – June where they foraged within interior forest away from edge habitat. It has been well documented that coyotes have the capacity to prey heavily on white-tailed deer fawns, and that forested landscapes and forest edge may contribute to fawn depredation. Our findings may reveal a mechanism by which coyotes seek out fawning locations during the spring. Additionally, there has been debate in the literature as to how coyotes supplement their diet year-round with deer. Our findings show that coyotes do not use roads during foraging excursions, and road-kill carcasses are unlikely to explain the presence of deer in coyote diets during the rest of the year.

Habitat Associations of Two Threatened Primates along a Conservation Corridor in Western Ecuador
Jacquelyn Tleimat, Sarah Fritts, Ryan Lynch, David Rodriguez, Shawn McCracken

The Chocó rainforest in coastal Ecuador is an internationally recognized biodiversity hotspot that has been more than 95% deforested for agricultural purposes, which has reduced wildlife habitat and isolated remaining habitat patches. In response to rapid deforestation, the Three Forest Conservation Corridor (TFCC) was established to promote connectivity among local reserves. As managers from these reserves plan to acquire land to expand the TFCC, a better understanding of threatened species’ habitat associations, such as two native threatened primates: the Ecuadorian capuchin (Cebus aequatorialis) and the Ecuadorian mantled howler (Alouatta palliata aequatorialis), is required to inform management and acquisition decisions. We recorded activity and distribution of both primate species using twenty acoustic monitoring devices and trail cameras deployed in the forest canopy across the TFCC. We assessed the influence of habitat type (agriculture and 3 forest types [cloud, dry, and wet]), vegetation structure, and landscape composition on occupancy and activity using single-season occupancy models. We also mapped total detections of both primate species to identify areas of frequent use. Models that included covariates were compared to null models using AICc and goodness-of-fit-tests. Although the 90% confidence intervals overlapped ‘0’, the model that included a positive relationship with station height was the best model for Ecuadorian mantled howlers and the model that included a positive relationship with distance from habitat edge created by human activities was the best model for Ecuadorian capuchins. The lack of significance likely was due to small sample sizes, a common issue when studying threatened species. Maps indicated that cloud forest on the northeastern edge of the TFCC were frequently used by both species. We suggest preserving intact forests with tall trees to aid in both species’ conservation. Results will guide future land acquisition and restoration for TFCC expansion.

Identifying Essential Habitat Features for Bat Conservation in a Mixed-Use Region
Kelly Russo-Petrick, Karen Root
Bat populations are decreasing, with anthropogenic land use the biggest threat worldwide, so it is necessary to determine the habitat factors most important for their survival. The objective of this study was to determine impacts of habitat variables, especially those associated with human land use, on bat activity and species richness at multiple spatial scales. Research was conducted in the Oak Openings Region of Northwest Ohio, which is a biodiverse mixed-use region. Bat activity was measured May-September 2019-2020 at stationary sites in parks and on road transects with Anabat acoustic monitors. Local-scale habitat variables were measured in the field and landscape-level variables were measured in ArcGIS and FRAGSTATS. AICc tables from stepwise regression were used to compile the most important variables for overall activity and richness for stationary points and transects. Across both years, the best final activity model for stationary points included a negative association with clutter at 0-3 m, percent floodplain forest, and patch richness and a positive association with percent sand barrens. There was a positive association of bat species richness at stationary points with percent sand barrens and upland prairie and a negative association between species richness and upland conifer and swamp forest. Activity along transects was higher with more natural habitat, clutter at 0-3 m, canopy height, and total forest. Activity along transects was lower with higher understory height and percent cropland. Species richness along transects was higher with more natural habitat and habitat types and lower with more percent cropland and urban. Results differed across the 8 species, survey types, and scales. Bat activity and richness can be improved by increasing natural forest habitat along roads, decreasing clutter and increasing habitat connectivity in parks, and preventing urban and agricultural development in important bat habitat.
Camera Trapping Survey for Plains Spotted Skunk in Eastern Oklahoma
Kalynn @kalynnbranham, Vicki Jackson

Once known as a regularly harvested fur bearer, Spilogale putorius (eastern spotted skunks) have experienced drastic population decline and are now considered a species of conservation concern throughout much of their range. In an effort to contribute to distribution and habitat presence data in Oklahoma, we surveyed 95 locations using camera traps during the months of October and April (2018-2019 and 2019-2020) attempting to detect S. p. interrupta (plains spotted skunk) subspecies. We began surveying the Ouachita National Forest in eastern Oklahoma, and broadened our survey to include areas of recently confirmed sightings at the Cookson Wildlife Management Area. We only detected eastern spotted skunks in 5 of the 95 (0.053% locations that were surveyed, during 12 of the 5,287 (0.002%) trap nights, and 13 of the 2,085 (0.006%) mammal detections. Although we had low trap success, we believe our results are useful in demonstrating the rarity of this species. Additionally, the information resulting from this study should be useful in establishing a basis for continued sampling in Oklahoma and providing information to improve sampling technique and design when researching this species.

Temporal Interactions Among Forest Carnivores and Their Prey at Varying Resolutions of Time
Austin Smith, Nichole Bjornlie, John Squires, Joseph Holbrook

Camera traps are an effective survey tool to assess temporal interactions among species. Few studies have examined multiple temporal resolutions when asking questions about species interactions. Here, we evaluated how activity patterns among species shifted during winter, as well as assessed time between detections within a camera station. Our species of interest included forest carnivores and their prey: Pacific marten (Martes caurina), montane red fox (Vulpes vulpes macroura), coyote (Canis latrans), red squirrels (Tamiasciurus hudsonicus), and snowshoe hares (Lepus americanus). We sampled carnivores and their prey across an extensive network of camera traps (n = 107) during the 2015-16, 2016-17, and 2017-18 winter seasons in the Greater Yellowstone Ecosystem of Wyoming. We generated kernel density plots and calculated coefficients of overlap among our predators and prey during early and late winter. Furthermore, we calculated the time between interactions (i.e., hours) during instances of prey detections followed by a predator detection within a camera station. We discovered that Pacific marten exhibit activity that evenly captures peaks in activity of two primary prey species. Overlap between coyotes and red fox was less in early winter, whereas coyotes modified activity in late winter such that they more closely aligned with red fox. This may have consequences on competitive interactions between these two canids during resource constraints. This work provides a new lens into species interactions across varying timeframes.

Doing More with Less: Assessing Novel Approaches to Estimating Jaguar Space-Use from Camera-Trap Data
Robert Nipko, Kelly Burdette, Alice Stitzer, Marcella Kelly

Two fundamental questions in ecology are how many animals are in a population and how individuals use the space around them. Historically, space-use studies are based on telemetry, but these methods are labor-intensive and often include relatively few individuals, thus limiting the strength of population level inferences. Non-invasive methods like camera-trapping are widespread for studying population dynamics, especially for elusive, wide-ranging species like jaguars (Panthera onca). If georeferenced, such data could be an alternative, underutilized source of information about animals’ space-use. These data often include a larger swathe of the population, albeit with fewer observations per individual, potentially allowing stronger inferences about population-level variation in movement patterns. We assessed two approaches for examining space-use solely from camera-trapping data, using photo-captures of 73 individual jaguars (27 females and 46 males) collected in northwestern Belize from 2010-2018. We use the term Area of Use (AoU) because these simplified representations likely do not correspond to true home ranges. First, we defined minimum convex polygons (MCP) for each jaguar photo-capture locations. Second, we used spatially explicit capture-recapture (SECR) models to probabilistically estimate each AoU. The observation process in SECR models is typically used to account for imperfect detection but can also be considered a simplified movement model. For the MCP method, female AoUs ranged from 1.2 – 99.3 km2 and males from 2.6 – 550.7 km2. For the SECR method, females ranged from 6.9 – 5,087.3 km2 and males from 4.3 – 11,922.2 km2. The mean female AoU from the SECR method (319.1 km2) was 12.5 times the MCP method (25.4 km2) and mean male SECR AoU (913.3 km2) was 8.0 times the MCP approach (113.5 km2). For each method, we also calculated and compared individual overlap by sex and annual change in AoU location. Finally, we describe future directions for improving these approaches.

Analysis of Jaguar Space-Use Via a Novel Application of Spatially Explicit Capture-Recapture
Alice Stitzer, Kelly Burdette, Robert Nipko, Marcella Kelly

Assessment of home range size and space-use patterns for species of concern allows managers to address conservation needs and efficiently design wildlife reserves or movement corridors. For low-density carnivores like jaguars, (Panthera onca), telemetry is expensive and labor-intensive. Camera trapping is widely used to study population dynamics for such species, and, if georeferenced, provides an alternative source of space-use data. Although telemetry produces more observations per individual, fewer individuals may not offer a clear picture of space-use variation across populations. We paired camera trap data with a spatially explicit capture-recapture (SECR) model to estimate population level space-use dynamics of 73 individual jaguars, 27 females and 46 males, over a nine-year period (2010-2018) in northwestern Belize. SECR models are typically used to study population dynamics, but because they explicitly incorporate animal movements, we adapted the models to estimate home ranges. Resulting areas were labeled ‘Area of Use’ (AoU) rather than home ranges as fixed locations may not fully encompass an individual’s movement. We found the average AoU across 2010-2018 to be 287.9 km2 for females and 937.4 km2 for males. We calculated annual turnover, meaning whether or not an individual was detected again the following year, for each sex. Throughout 2010-2018, male turnover ranged 20-100%, and female turnover ranged 20-75%. Finally, we calculated overlap between individuals on a sex basis to evaluate exclusivity of space-use. This novel application of SECR models suggests that jaguar AoUs are larger and more variable than found in previous telemetry-based studies. There has been little research on regional variation in jaguar ranging behavior, so current understanding may not account for variation in ecology in differing tropical ecosystems. With the robust population level inferences this technique provides, managers may better understand localized space-use and movement patterns vital to conservation of jaguars.

Landscape Composition and Proximity to Water Structure American Badger Distributions in Shortgrass Prairies
Colleen Piper, Adam Ahlers, Ty Werdel, Andrew Ricketts, Matt Peek, Daniel Sullins

Widespread land-use changes in the Great Plains have resulted in a patchy mosaic of prairie embedded within human-modified landscapes (e.g., agriculture and energy development). The distribution of prairie-obligate species in this region may be constrained by these alternate land-use types, though many carnivore-specific examples are unknown. We used three years (2018-2020) of presence/absence data collected from camera-trap sites (n = 381) distributed across western Kansas to assess multiscale effects of landscape change on the distribution of American badgers (badger, Taxidea taxus), an important predator and ecosystem engineer throughout its range. We predicted initial occupancy probabilities and colonization rates would be positively associated with the amount of native prairie and Conservation Reserve Program (CRP) hectares surrounding sites. We also predicted occupancy and colonization would be negatively associated with the amount of agriculture and density of wind towers surrounding sites along with distance to permanent water sources. We found habitat use by badgers was insensitive to the amount of prairie and CRP surrounding sites at local and landscape scales. However, badgers were more likely to occupy and colonize sites with greater amounts of row crop agriculture at both landscape and local scales. Additionally, badgers were less likely to occupy sites farther from permanent water sources. Our study suggests that badgers, although considered prairie-obligate carnivores, may be exploiting agricultural areas because of increased prey densities or suitable burrowing/digging substrates. Moreover, our research highlights the importance of permanent water resources to badgers in arid regions within the Great Plains.

Contributed Oral
Location: Virtual Date: November 5, 2021 Time: 4:00 pm - 5:00 pm