Search

Connectivity is important for population persistence and can reduce the potential for inbreeding depression. Connectivity between populations can also facilitate disease transmission; respiratory diseases are one of the most important factors affecting populations of bighorn sheep (Ovis canadensis). The mechanisms of connectivity in populations of bighorn sheep likely have implications for spread of disease, but the behaviors leading to connectivity between bighorn sheep groups are not well understood. From 2007–2012, we radio-collared and monitored 56 bighorn sheep in the Salmon River canyon in central Idaho. We used cluster analysis to define social groups of bighorn sheep and then estimated connectivity between these groups using a multi-state mark-recapture model. Social groups of bighorn sheep were spatially segregated and linearly distributed along the Salmon River canyon. Monthly probabilities of movement between adjacent male and female groups ranged from 0.08 (±0.004 SE) to 0.76 (±0.068) for males and 0.05 (±0.132) to 0.24 (±0.034) for females. Movements of males were extensive and probabilities of movement were considerably higher during the rut. Probabilities of movement for females were typically smaller than those of males and did not change seasonally. Whereas adjacent groups of bighorn sheep along the Salmon River canyon were well connected, connectivity between groups north and south of the Salmon River was limited. The novel application of a multi-state model to a population of bighorn sheep allowed us to estimate the probability of movement between adjacent social groups and approximate the level of connectivity across the population. Our results suggest high movement rates of males during the rut are the most likely to result in transmission of pathogens among both male and female groups. Potential for disease spread among female groups was smaller but non-trivial. Land managers can plan grazing of domestic sheep for spring and summer months when males are relatively inactive. Removal or quarantine of social groups may reduce probability of disease transmission in populations of bighorn sheep consisting of linearly distributed social groups. © 2016 The Wildlife Society.

Many animals, including gray wolves (Canis lupus), live in social groups. Genetic techniques can help reveal the structure and composition of social groups, providing valuable information about group and population dynamics. We evaluated the effectiveness of using noninvasive genetic sampling (NGS) of fecal and hair samples at wolf rendezvous sites combined with spatial and genetic assignment criteria for assigning individuals to packs, detecting dispersers and lone wolves, determining the number of packs in an area, and obtaining group metrics. We applied this approach in 4 study areas covering 13,182 km2 in Idaho, USA while concurrently monitoring wolves using telemetry techniques. We assigned pack affiliation to 78–97% of individuals across study areas and identified 12 potential dispersers. We detected a successful gene flow event by reconstructing a breeding male's genotype and tracing it back to a pack of origin using genetic assignment techniques. Average pack size was consistent between our NGS- and telemetry-based counts (x¯ = 10 for both), and both methods detected similar age composition within groups (31% pups and 69% adults for NGS and 33% pups and 67% adults for telemetry). Our NGS approach has the advantage of providing pack metrics including sex ratio, inferred breeders, and intra-pack relatedness that telemetry and observational techniques alone cannot. This NGS field sampling strategy combined with our pack assignment method was successful and provides an approach for characterizing functional social groups in the absence of previously acquired NGS, telemetry, or other observational data that may not be available when sampling new areas. © 2016 The Wildlife Society.

Estimating coyote (Canis latrans) density and other demographic parameters is difficult, particularly for populations that exist at low density. This is the situation for recently established coyote populations in the eastern United States where populations may be below carrying capacity and growth unregulated. We used non-invasive fecal DNA collected from 5 scat sampling sessions over 2.5 years to estimate population parameters (i.e., density, apparent survival, recruitment, and population growth) for coyotes at 2 different sites in the Ridge and Valley region of the central Appalachians in Virginia, USA. We identified individuals using microsatellite genotypes and estimated apparent survival for the local population at both sites across the 5 sessions in a single Cormack–Jolly–Seber model. We estimated density for each site and session separately using single session spatial replicates of 0.5-km transect segments as traps in a spatial capture–recapture model. Finally, we derived estimates of recruitment and population growth using an ad hoc robust design approach. We were able to estimate population parameters, even though coyote densities at both sites were low. Generally, derived recruitment and apparent survival were inversely related across sites, however, precision in estimates was poor. Thus, although there appeared to be some differences in demographic estimates for local coyote populations, uncertainty in parameters was too great to detect changes in demographic rates over short periods of time using ad hoc robust design. However, the non-invasive genetic sampling and spatial capture–recapture approach provides a useful methodology and framework for future research intended to estimate population dynamics for coyotes. This method will also be useful for other species that occur at low densities, over large spatial scales, and lack distinguishing marks for camera-trap surveys. Finally, we believe this method will allow for detection of population trends over greater periods of time, and we consider alternate sampling strategies and modeling approaches that may improve the ability to estimate demographic rates of change for coyote populations using noninvasive genetics and spatial capture–recapture. © 2016 The Wildlife Society.

Estimating densities of cougar (Puma concolor) is important for managing cougars and their prey but remains challenging because of cougar's elusive and solitary behavior. To evaluate a non-invasive, genetic capture–recapture method to estimate cougar population size and density, we surveyed a 220-km2 area using conservation detection dogs trained to locate scat over a 4-week sampling period in northeast Oregon. We collected 272 scat samples and conducted DNA analysis on 249 samples from which we determined individual identification from 73 samples that represented 21 cougars (9 males and 12 females). We evaluated 4 models to estimate cougar densities: Huggins closed population capture–recapture (Huggins), CAPWIRE, multiple detections with Poisson (MDP), and spatially explicit capture–recapture (SECR). Population estimates for cougars using our study area were 26 (95% CI = 22–35, 9 males and 17 females) from Huggins models, 24 (95% CI = 21–30, 9 males and 15 females) from CAPWIRE, and 27 (95% CI = 24–42, 9 males and 18 females) from the MDP model. We accounted for the edge effect in density estimates caused by individuals whose home ranges included only a portion of the survey grid by buffering the study area using the mean home range radius of 8 cougars equipped with global positioning system collars on or near the study area. We estimated densities of 4.6 cougars/100 km2 (95% CI = 3.8–8.3) for the Huggins model, 4.8 cougars/100 km2 (95% CI = 4.2–7.8) for the MDP model, 4.2 cougars/100 km2 (95% CI = 3.3–5.3) for the CAPWIRE model, and 5.0 cougars/100 km2 (95% CI = 3.2–7.7) for the SECR model. Our results suggested estimating cougar densities using scat detection dogs could be feasible at a broader scale with less effort than other methods currently being used. © 2014 The Wildlife Society.

Various monitoring methods have been developed for large carnivores, but not all are practical or sufficiently accurate for long-term monitoring over large spatial scales. From 2009 to 2010, we used a predictive habitat model to locate gray wolf rendezvous sites in 4 study areas in Idaho, USA and conducted noninvasive genetic sampling (NGS) of scat and hair found at the sites. We evaluated species and individual identification PCR success rates across the study areas, and estimated population size with a single-session population estimator using 2 different recapture-coding methods. We then compared NGS population estimates to estimates generated concurrently from telemetry data. We collected 1,937 scat and 166 hair samples and identified 193 unique individuals over 2 years. For fecal DNA samples, species identification success rates were consistently high (>92%) across areas. Individual identification success rates ranged from 78% to 80% in the drier study areas and dropped to 50% in the wettest study area. The degree of agreement between NGS- and telemetry-derived population estimates varied by recapture-coding method with considerable variability in 95% confidence intervals. Population estimates derived from NGS methods were most influenced by the average number of detections per individual. We demonstrate how changes in field effort and recapture-coding method can affect population estimates in a widely used single-session population estimation model. Our study highlights the need to further develop reliable population estimation tools for single-session NGS data, especially those with large differences in capture frequencies among individuals stemming from severe capture heterogeneity (i.e., overdispersion). © 2014 The Wildlife Society.

The behavioral patterns and large territories of large carnivores make them challenging to monitor. Occupancy modeling provides a framework for monitoring population dynamics and distribution of territorial carnivores. We combined data from hunter surveys, howling and sign surveys conducted at predicted wolf rendezvous sites, and locations of radiocollared wolves to model occupancy and estimate the number of gray wolf (Canis lupus) packs and individuals in Idaho during 2009 and 2010. We explicitly accounted for potential misidentification of occupied cells (i.e., false positives) using an extension of the multi-state occupancy framework. We found agreement between model predictions and distribution and estimates of number of wolf packs and individual wolves reported by Idaho Department of Fish and Game and Nez Perce Tribe from intensive radiotelemetry-based monitoring. Estimates of individual wolves from occupancy models that excluded data from radiocollared wolves were within an average of 12.0% (SD = 6.0) of existing statewide minimum counts. Models using only hunter survey data generally estimated the lowest abundance, whereas models using all data generally provided the highest estimates of abundance, although only marginally higher. Precision across approaches ranged from 14% to 28% of mean estimates and models that used all data streams generally provided the most precise estimates. We demonstrated that an occupancy model based on different survey methods can yield estimates of the number and distribution of wolf packs and individual wolf abundance with reasonable measures of precision. Assumptions of the approach including that average territory size is known, average pack size is known, and territories do not overlap, must be evaluated periodically using independent field data to ensure occupancy estimates remain reliable. Use of multiple survey methods helps to ensure that occupancy estimates are robust to weaknesses or changes in any 1 survey method. Occupancy modeling may be useful for standardizing estimates across large landscapes, even if survey methods differ across regions, allowing for inferences about broad-scale population dynamics of wolves. © 2014 The Wildlife Society.