Mammalogy IV

Contributed Oral

Mesocarnivore Space Use and Resource Selection in Illinois
Nicole Gorman, Michael Eichholz, Peter Schlichting, Dan Skinner, Guillaume Bastille-Rousseau

Movement is an essential behavioral link to understanding how predators interact with their ecosystems, including the top-down controls they exert on prey species, as well as the bottom-up influences predators experience. Research on predator movement has mainly focused on large predator species, but mesocarnivores are the top predators throughout much of the eastern and midwestern United States today. In these regions, the wide variety of natural and human-altered environments suggest that variation in habitat may translate to variation in mesocarnivore spatial behavior. We fit bobcats (Lynx rufus) and coyotes (Canis latrans) with GPS collars in two study sites, one patchy area dominated by corn and soybean agriculture in central Illinois and one more contiguous oak-hickory forested area in southern Illinois. We evaluated variation in home range size and resource selection of these two species, explicitly focusing on interspecific differences within sites and intraspecific differences between sites. We investigated the temporal and spatial drivers of home range size and resource selection by measuring functional responses in resource selection to landscape features and measuring drivers of individual variation in mesocarnivore spatial behavior. We found that landscape features affected mesocarnivore home range size, with individuals having larger home ranges in agricultural habitat than in forested habitat. By exploring habitat selection on spatial, temporal, and individual axes, we found that mesocarnivores also differed in their response to landcover categories, displaying variation in resource selection behavior, even within species. Individual variation in behavior was driven by functional responses in resource selection, as well as individual attributes such as sex and species. Overall, our results indicate that the spatial behavior of mesocarnivores might be more variable than top predators due to mesocarnivores’ inherent generalist and flexible behavior.

Comparing Cost-Efficiency of Game Cameras and Physical Capture Methods to Assess Raccoon Abundance
Cady Sartini, Wesley Dixon, Guha Dharmarajan, Amy Davis

Game cameras are often cited as a noninvasive monitoring technique that can allow for some cost-savings compared to physical capture-recapture methods.  However, few claims of cost-efficiency compare the complete costs of camera use, including the time to analyze sometimes tens or hundreds of thousands of photos, with the cost of physical capture methods.  To directly compare the costs and benefits of using cameras and physical capture, we deployed a grid of Reconyx game cameras in the same location as Tomahawk box traps as a part of ongoing efforts in the National Rabies Monitoring Program from January-April of 2019 and 2020.  We directly compare the costs of labor and equipment for monitoring with both methods and will make recommendations about camera settings and photo processing methods that can help project managers better determine if cameras might be a more cost-effective option for monitoring wildlife populations.

Comparison of Two Non-Invasive Survey Techniques to Concurrently Assess Site Occupancy By Semiaquatic Mammals
Adam Ahlers, Max Larreur, Benjamin Matykiewicz, Bryce Olson, Reid Plumb, Steve Windels

Non-invasive survey methods are often used to establish site occupancy by semiaquatic mammals including muskrat (Ondatra zibethicus), North American river otter (Lontra canadensis, hereafter ‘otter’), and American mink (Neovison vison, hereafter ‘mink’). The effectiveness of different survey techniques for these sympatric species, however, has not been simultaneously evaluated. We assessed the relative effectiveness of two survey methods, walking surveys and floating raft surveys, at concurrently detecting semiaquatic mammals in lacustrine wetlands. We sampled 40-71 randomly selected wetland sites in Voyageurs National Park, Minnesota, USA, during five summers (2016-2020) for sign of muskrat, otter, and mink using both survey techniques. We used stacked, single-season occupancy models to control for survey-specific covariates (prior rainfall, day of the year) while also assessing the effectiveness of our survey methods. Muskrat detection was influenced by survey technique (higher detection rates with floating rafts) and related negatively to the day of year. Detection of otter was also influenced by survey technique but with higher detection probabilities associated with walking surveys. Mink detection was influenced positively by day of year and negatively to the amount of precipitation prior to surveys, and there was a modest effect associated with our survey techniques (higher with waling surveys). Species-specific unique detections (a species was detected by using only one survey technique) varied, suggesting multiple concurrent survey techniques are necessary to enhance detection probabilities.

Home Range and Habitat Selection of Gray Wolves on Red Lake Indian Reservation and Surrounding Areas
June Levin, Jay Huseby, Dave Price, Sarah Ruffing, Bob Newman

For species of conservation concern, identification and protection of critical habitat are crucial steps to recovery. Although recently declared recovered and no longer listed as federally endangered under the Endangered Species Act, the gray wolf (Canis lupus) is living on a human dominated landscape and understanding home range and habitat needs remains critical to their conservation. The Red Lake Department of Natural Resources started GPS tracking gray wolves on Red Lake Reservation in 2012 with the intent to estimate wolf population and we analyzed the GPS data to identify gray wolf home range and habitat selection. We used minimum convex polygon (MCP) and autocorrelated kernel density estimation (AKDE), which reveals if an animal is using a well-defined home range and better captures the autocorrelative nature of GPS relocation data. We used a kernel density estimator (KDE) and Brownian bridge (BB) for comparison. We tested for habitat selection using multinomial models. We found that the average home range of Red Lake wolves was 1716km2 using MCP and 291km2 using AKDE. Only 7 of 16 total wolves displayed a defined home range. For wolves with a defined home range, KDE, AKDE, and BB provided similar results, whereas MCP provided significantly lower results. For wolves without a defined home range, MCP and KDE provided similar results, whereas BB provided significantly lower results. Because home range is typically estimated to inform managers of minimum required area for a species, estimator selection should be considered carefully due to the possible underestimation of home range area. We detected an overall selection of woody wetlands and an avoidance of developed areas by gray wolves, and managers will use this information to identify gray wolf habitat that can support sustainable populations on Red Lake Reservation.

Curtailment and Acoustic Deterrents Reduce Bat Mortality at Wind Farms
Rhett Good, Gonielia Iskali, John Lombardi, Trent Mcdonald, Karl Dubridge, Michael Azeka

The impacts of wind-energy on bat populations is a growing concern because wind turbine blades can strike and kill bats, and wind turbine development is increasing. We tested the effectiveness of an acoustic bat deterrent developed by NRG Systems combined with curtailing turbine blades when wind speeds were below 5.0 m/sec at 2 adjacent wind-energy facilities in northeast Illinois during the fall migration period (1 Aug–15 Oct) of 2018, compared to turbines that operated at manufacturer cut-in speed and lacked acoustic deterrents. We measured the effectiveness of the acoustic deterrent and curtailment with carcass searches of cleared plots below turbines, and field trials that estimated searcher efficiency and carcass persistence. The effectiveness of each management action was estimated using a generalized linear mixed-effects model with several potential co-variates. Overall bat fatality rates were 66.9% (90% confidence interval [CI]: 54.5–75.9%) lower at curtailed turbines with acoustic deterrents compared to turbines that operated at manufacturer cut-in speed. Curtailment and the deterrent reduced bat mortality to varying degrees between species, ranging from 58.1% (90% CI: 41.1–70.3%) for eastern red bats to 94.4 (68.8–99.0%) for big brown bats. Hoary and silver-haired bat mortality was reduced by 71.4% (90% CI: 55.8–81.5%) and 71.6% (90% CI: 56.1–81.7%), respectively. Curtailment alone reduced overall bat mortality by 42.5% (90% CI: 16.1–60.6%), but also varied by species. Curtailment alone effectively reduced hoary bat and eastern red bat mortality, but not silver-haired bat mortality. We estimated the additional reduction in mortality resulting from the using the acoustic deterrent and curtailment under the assumption that curtailment and the acoustic deterrent would have reduced mortality by the same percentage at adjacent wind-energy facilities. Acoustic deterrents resulted in 31.6%, 17.4%, and 66.7% in additional reductions of bat mortality compared to curtailment alone for eastern red bat, hoary bat, and silver-haired bat, respectively. 

Drivers of Home Range Size and Daily Movement Rates of Mountain Lions in the Olympic Peninsula
Maximilian Allen, Morgan Farmer, Alexandra Avrin, Bethany Ackerman, Kristen Phillips, Daniel Ravenel, Shannon Murphie, L. Mark Elbroch

An individual’s space use, including their home range and resource use within their home range, is often a balance between maximizing fitness through foraging strategy selection and minimizing risk and intraspecific competition. Identifying the relative importance of these factors can inform effective conservation. Our objectives were to determine which demographic, environmental, or anthropogenic factors influence the home range size and daily movement rates of mountain lions. Between 2011 and 2020, we captured 62 mountain lions using trained hounds and box traps. After chemical immobilization, we sexed, aged, and fit each mountain lion with a GPS collar. We calculated 95% annual home range sizes (km2) using Fixed Kernel Density estimation (FKDE), the average percent of intersexual and intrasexual neighbor overlap, and the mean daily movement rate (km/day) for all adults (≥ 18 months). Following a systematic literature review, we selected a suite of a-priori covariates including sex, age, reproductive status, average intersexual and intrasexual overlap, elevation, precipitation, habitat type, average terrain roughness index, road density, and ungulate suitability. We then modeled annual home range size and mean daily movement rate as a function of our a-priori covariates. Annual 95% FKDE home ranges varied from 21.7 km2 to 9,290.6 km2, with an average of 596.8 km2 (±166.5 SE). Female mountain lions had significantly higher intersexual overlap (13.05% ± 1.34) than males (3.96% ± 0.66, p < 0.0001); however, males had significantly higher intrasexual overlap (8.66% ± 1.13) than females (3.18% ± 0.36, p < 0.001). Mountain lions moved on average 4.12 km per day (± 0.37), with males moving significantly farther per day than females (males 5.46 km/day ± 0.91, females 3.39 km/day ± 0.25, p = 0.03). Determining drivers of individual space use can help guide management and conservation by identifying critical habitat and potential limiting factors for a population.

How to Avoid the Camera TRAP: Recommendations for Effective Use of Remotely Triggered Cameras
Jody Tucker, Jessie Golding

Camera traps have become a ubiquitous survey tool for wildlife research. While many studies have evaluated design parameters such as camera density, duration, or sampling season, far less evaluation has been done of the influence of the camera themselves and the trade-offs that different settings have on metrics such as detection probability or site occupancy, as well as post-processing data management and storage. In particular, there is little guidance on how two camera settings, trigger-delay interval and number of photos per trigger (photo burst), affect camera performance. As these settings determine the number of photos a camera generates, they influence both species detectability and the time, labor, and storage cost of processing the resulting photos. Non-optimal trigger speed or burst settings that are too low can result in inability to detect species of interest but conversely, settings that are too high may generate extraneous photos that that do not significantly improve detectability but significantly increase labor costs of photo processing. Additionally, there is an assumption that different camera units are equivalent, particularly within a single brand and model.  We analyzed camera trap data sets across multiple study areas in the western U.S. to evaluate the influence of camera trigger-delay and burst settings on detection probability and site occupancy for an array of mammalian species ranging from common to rare. We also tested camera brands and models for their ability to detect the same species. Our results suggest that camera settings can significantly influence detection probability and post-processing time. In addition, we found significant variation in species detections between camera models, even within single camera brands. We provide guidance on optimizing camera settings to maximize detectability while minimizing extraneous photos, as well as guidance on short pilot studies that can be conducted prior to large-scale deployment to address these issues.

Performance of Automated Echolocation Classifiers for Northeastern Bat Species
Donald Solick, Christopher Nations, John Chenger

Identification of bat echolocation calls to species is increasingly important in the United States, particularly for assessing the presence of protected bat species and the status of species impacted by wind-energy development and by White-Nose Syndrome. The programs Bat Call Identification (BCID), Kaleidoscope Pro (KPro), and Sonobat offer automated classification of echolocation calls, but the accuracy and performance of these systems is not well understood. We used 1,500 files containing known echolocation calls from nine northeastern bat species, including threatened and endangered species, to create confusion matrices and calculate general accuracy measures (overall accuracy, no information rate [NIR], negative and positive predictive values [NPV and PPV], sensitivity, and specificity) for each classifier. The classifiers performed well in terms of overall accuracy relative to the NIR (> 80%, 95%, 96% for BCID, KPro, Sonobat), indicating that each classifier was significantly more accurate than expected by chance. Each classifier also had high NPV (>80%, 94%, 97%) and specificity (>81%, 98%, 97%) across species reflecting low rates for both false positives and false negatives. However, PPV and sensitivity varied among species within each classifier, indicating differential rates for true positives. For example, sensitivity for Myotis septentrionalis was 10%, 61%, 90% for BCID, KPro, and Sonobat, while sensitivity for Perimyotis subflavus was 99%, 97%, and 98%. Following USFWS guidance, classifiers calculate the Maximum Likelihood Estimate (MLE) to determine whether a species is present or absent on a given night. To test whether differences in program accuracy affect determinations of nightly presence and probable absence calculated from MLEs, we ran 2000 nights of simulated bat data through the programs. We will discuss the results of this comparison and the implications for managers of a more liberal or more conservative approach to automated classification for bat acoustic surveys.

Tracking Red Bat Movement Response to Ultrasonic Deterrent in a Flight Cage
Brittany Stamps, Emma Guest, Sara Weaver, Amanda Hale, David Rodriguez, Brogan Morton, Janine Crane, Mark Chaffee, John Ugland, Cris Hein, Sarah Fritts

Wind energy is an essential aspect of combating climate change, but it is not without risks. Bat fatalities due to wind turbine blade strikes are an unintended consequence of this renewable energy source. To date, strategies to reduce bat fatalities have been inconsistent among species or too expensive to implement. Ultrasonic acoustic deterrents (UADs) are designed to create a disorienting airspace around wind turbines by emitting sounds that jam bat echolocation calls; however, field tests of UADs have had variable success among species. Variability may be due to higher-frequency sounds attenuating over shorter distances or, relatedly, due to bats not encountering their characteristic frequency range at greater distances from the UAD. Our objective was to compare effects of sound emissions from an NRG Systems-manufactured UAD at deterring eastern red bats (Lasiurus borealis), a high-frequency echolocating species commonly reported at wind energy facilities. Existing data indicates UADs are not successful in reducing fatalities for this species. During July-October 2020 and March-May 2021, we individually released eastern red bats (n = 33) into a 60x10x4.5-m (LxWxH) outdoor flight cage located in San Marcos, Texas, USA. We tracked flight using thermal video cameras during three, four-minute trials of three frequency ranges: 20–32 kHz (low), 38–50 kHz (high), and 20–50 kHz (combined), interspersed by four-minute control periods. We assessed differences in distance eastern red bats flew from UADs and compared between each treatment and the control using compositional and Kolmogorov-Smirnov analyses. Results indicated the high treatment deterred bats farthest; however, bats flew farther from the UAD during all treatments than controls. These results indicate low-frequency ultrasound may provide a viable option to deter eastern red bats, as the sound does not attenuate as quickly; thus, may be more effective at greater distances as wind turbine blade size increases.

Contributed Oral
Location: Virtual Date: November 4, 2021 Time: 12:00 pm - 1:00 pm