|
" Predicting Spatial Risk of Wolf-Cattle Encounters on Rugged and Extensive Grazing Lands "
Clark, Patrick E.; Chigbrow, Joe; Johnson, Douglas E.; Williams, John; Larson, Larry L.; Roland, Tyanne
| Document Type
|
:
|
AL
|
| Record Number
|
:
|
944069
|
| Doc. No
|
:
|
LA1s6431rf
|
| Language of Document
|
:
|
English
|
| Main Entry
|
:
|
Clark, Patrick E.; Chigbrow, Joe; Johnson, Douglas E.; Williams, John; Larson, Larry L.; Roland, Tyanne
|
| Title & Author
|
:
|
Predicting Spatial Risk of Wolf-Cattle Encounters on Rugged and Extensive Grazing Lands [Article]\ Clark, Patrick E.; Chigbrow, Joe; Johnson, Douglas E.; Williams, John; Larson, Larry L.; Roland, Tyanne
|
| Title of Periodical
|
:
|
Proceedings of the Vertebrate Pest Conference
|
| Volume/ Issue Number
|
:
|
28
|
| Date
|
:
|
2018
|
| Abstract
|
:
|
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.
|
| |