Spatial variability in terrain, vegetation, and other features affect cattle and wildlife distribution on mountainous grazing lands of the western United States. Yet we have a poor understanding of how this spatial variability influences risk of wolf-cattle encounters and associated depredation. This knowledge gap severely limits our capacity to prevent or mitigate wolf-cattle conflict. Research addressing this problem was conducted in 2009-2011 at four study areas in western Idaho to evaluate models and mapping tools for predicting spatial risk of wolf-cattle encounters. Lactating beef cows grazing these study areas were instrumented with Global Positioning System (GPS) collars and tracked at 5-min intervals throughout the summer grazing season. Resource selection function (RSF) models, based on negative binomial regression, were developed from these GPS data and used to map the relative probability of cattle use in each study area. A wolf RSF model originally developed by Ausband et al. (2010) was applied to map study-area habitat types in terms of their relative suitability as wolf rendezvous sites. Spatial relationships between cattle and wolf selectivity patterns were used to classify and map wolf-cattle encounter risk to 5 classes (very high to very low) across each study area during the wolf rendezvous period (15 June-15 August). Validation analyses using GPS-based, wolf-cattle encounter observations (n = 200) revealed 84% of observed encounters occurred in areas of high- or very high-encounter risk (class 4 or 5). About 75% of confirmed wolf depredations recorded among three of four study areas were located in areas of high or very high risk. This new predictive understanding of wolf-cattle encounter risk will greatly aid livestock producers, resource managers, and policy makers in more effectively applying husbandry practices, allocating mitigation resources, and developing conflict mitigation plans and policies applicable throughout the mountainous western United States and potentially other regions of the world where wolves and cattle come into conflict. Published by Elsevier Inc. on behalf of The Society for Range Management.

Cattle grazing lands in the mountainous western United States are rugged, complex, and extensive. Terrain, vegetation, and other landscape features vary greatly across space. Risk of wolf-cattle encounters and potential for depredation loss certainly differ spatially as consequence of this variability. Yet, our understanding of this spatial risk is quite poor and this knowledge gap severely hampers our abilities to manage wolf-livestock interactions and mitigate conflicts. During 2009-2011, a research study was conducted at four study areas (USFS cattle grazing allotments) in western Idaho to evaluate and predict risk of wolf-cattle encounters. Each year, a random sample of 10 lactating beef cows from each study area was instrumented with GPS collars that logged positions at 5-minute intervals throughout the summer grazing season. Cattle resource selection was modeled using these GPS data and negative-binomial regression. An existing model was used to classify habitats within the study areas in terms of probability of use by wolves as rendezvous sites. Efficacy of this model was confirmed using scat, telemetry, and rendezvous site data. Spatial overlaps in the predicted selectivity of wolves and cattle were assessed and study area landscapes were then classified into five encounter-risk classes (very low to very high). Concurrent wolf and cattle GPS tracking data were used to document wolf-cattle encounters and thus evaluate the accuracy of this classification. About 94% of observed wolf-cattle encounters occurred within either the high or highest encounter-risk classes. Areas classified to the highest risk class were located on smooth, relatively flat slopes in concave terrain (e.g., stream terrace meadows) but not all were associated with surface water. Having this predictive understanding of where wolf-cattle encounters are most likely to occur will allow livestock producers and wildlife managers to more effectively apply resources, husbandry practices, and mitigation techniques to reduce conflict.

Ranch management has become more complex since wolves were reintroduced into Idaho and Wyoming in 1996. In wolf areas, livestock have experienced increased death loss and greater stress. Increased livestock aggressiveness has been observed, especially toward working dogs, making handling livestock more difficult. Additionally ranchers have reported a loss of body condition, lower conception rates, increased time and expense for management. Our study was designed to investigate the effect of wolf presence on cattle behavior, landscape use patterns, and resource selection by comparing high wolf density areas against low wolf density areas. This study also generated baseline information on cattle spatial behavior before wolves were on the landscape. A Before-After/Control-Impact Paired (BACIP) experimental design was used. Control study areas in Idaho (3) have high wolf presence while Impact study areas in Oregon (3) started with no wolf presence, and are shifting to elevated wolf presence. Paired Idaho and Oregon areas have similar topography, vegetation composition, wild ungulate prey bases, and livestock management. Cows are tracked at 5-minute intervals using GPS collars (10 per area) throughout the grazing season. Wolf presence is monitored by GPS, trail cameras, and scat surveys. Ten GPS-collared cattle in an Idaho study area encountered a GPS-collared wolf 783 times at less than 500 meters during 137 days in the 2009 grazing season. At 100 meters there were 53 encounters; 52 at night. Tests of naïve and experienced cattle exposed to a simulated wolf encounter found increased excitability and fear-related physiological stress responses in cows previously exposed to wolves. This was shown through increased cortisol levels, body temperature, and temperament scores. Cattle presence near occupied houses doesn’t offer protection from wolves. Data shows wolves within 500m of occupied houses 588 times during 198 days of tracking. Many confirmed depredations on this site were also close to houses.

The presence of gray wolves (Canis lupus) can directly and indirectly affect beef cattle (Bos taurus) production on rangelands of the Northern Rocky Mountains. While fairly extensive knowledge exists for the direct effects of wolf predation threat (e.g., cattle death and injury losses, elevated stress), our understanding of wolf-caused changes in cattle behavior and the associated cascade of potential indirect effects on cattle resource selection, diet quality, activity budgets, and energetic relationships is still largely in its infancy. We investigated whether wolf presence affected the daily travel distance of Global Positioning System (GPS)−collared cattle under a replicated, Impact-Control study conducted in western Idaho and northeastern Oregon during 2008−2012. Cattle in three Control (Oregon) study areas, where wolf presence was consistently low, traveled farther per day (13.7 ± 0.396 SE km day−1) than those in three Impact (Idaho) study areas (11.4 ± 0.396 SE km day−1) with moderate to high wolf presence. At Control study areas, cattle traveled farthest per day in July (13.2 ± 0.355 SE km day−1) and were least mobile in October (11.8 ± 0.365 SE km day−1), but daily travel distances were similar across all months for cattle in Impact study areas. This observational study provides evidence suggesting cattle in mountainous grazing areas alter their spatial behavior in response to gray wolf presence. These behavioral changes have energetic consequences that could potentially impact cattle productivity and ranch economics. Additional research into the activity budget and resource selection responses of these collared cattle is required to better understand the specific mechanisms behind these daily travel distance results.

The spatial occupancy patterns and activity of cattle grazing three riparian pastures was investigated in northeastern Oregon using Global Positioning System (GPS) collars logging at 1-sec intervals. Cattle consistently selected plant communities as grazing areas that had forage in sufficient volume to meet their requirements and favored communities as resting areas that were dry and open. Cattle were stationary for more than 50% of the time in each pasture and consistently rested between dark and 04:00 hours. Interaction with stream channels was found to be 1-2% of total occupancy time and occurred on less than 10% of channel length. Cattle were indifferent or avoided channel areas relative to their area and, when in this zone, they spent most of their time moving not resting. Cattle did not prefer the stream bank zone and spent only 2% of their time in that zone. When occupied, the stream bank zone was used as a travel corridor to gain access to water or cross the channel to access other pasture areas. These results are in contrast with the general belief that cattle are a primary occupant of the stream bank/channel area; additional research is needed to define factors influencing cattle occupancy.

Potential for large prairie remnants to provide habitat for grassland-obligate wildlife may be compromised by nonsustainable range-management practices. In 1979-1980, high nesting densities of 3 species of hawks in the genus Buteo-Ferruginous Hawk (Buteo regalis), Red-tailed Hawk (B. jamaicensis), and Swainson's Hawk (B. swainsoni) were documented on the Zumwalt Prairie and surrounding agricultural areas (34,361 ha) in northeastern Oregon, USA. This area has been managed primarily as livestock summer range since it was homesteaded. Unlike in other prairie remnants, land management on the Zumwalt Prairie was consistent over the past several decades; thus, we predicted that territory occupancy of these 3 species would be stable. We also predicted that territory occupancy would be positively related to local availability of nesting structures within territories. We evaluated these hypotheses using a historical dataset, current survey and habitat data, and occupancy models. In support of our predictions, territory occupancy of all 3 species has not changed over the study period of similar to 25 yr, which suggests that local range-management practices are not negatively affecting these taxa. Probability of Ferruginous Hawk occupancy increased with increasing area of aspen, an important nest structure for this species in grasslands. Probability of Swainson's Hawk occupancy increased with increasing area of large shrubs, and probability of Red-tailed Hawk occupancy was weakly associated with area of conifers. In the study area, large shrubs and conifers are commonly used as nesting structures by Swainson's Hawks and Red-tailed Hawks, respectively. Availability of these woody species is changing (increases in conifers and large shrubs, and decline in aspen) throughout the west, and these changes may result in declines in Ferruginous Hawk occupancy and increases in Swainson's Hawk and Red-tailed Hawk occupancy in the future.