Development of a Regional Fence Model with Implications for Wildlife Management

Scientific Disciplines
Biological Sciences - Terrestrial
Wildlife conservation
world wildlife fund
wildlife management
Geographic information system
Faculty of Environmental Design
modeled fence locations
fence model
Volume 18, No. 1-4

Development of a regional fence model with 
Implications for wildlife management
Andrew F. Jakes*, Faculty of Environmental Design, University of Calgary, Calgary, Alberta T2N 
1N4 Canada
Erin E. Poor, Conservation Science Program, World Wildlife Fund, Washington D.C. 20037
Colby Loucks, Conservation Science Program, World Wildlife Fund, Washington D.C. 20037
Michael J. Suitor, Faculty of Environmental Design, University of Calgary, Calgary, Alberta T2N 
1N4 Canada
Barbed and woven wire fence are ubiquitous features across much of western North 
America, yet their effects on wildlife have received less attention than those of other 
anthropogenic features. At this time,  no geospatial fencing data is available at broad level 
scales; potentially making wildlife modeling of vagile species less accurate and conservation  
planning less reliable at various scales. Here, we model fence density across 13 counties in 
Montana‚Äôs Hi-Line region, based on publicly available GIS data and assumptions created 
from local, expert knowledge. The resulting fence location and density GIS layers are based 
on assumptions about where fence locations occur in association to different types of land 
tenure, land cover and roads. Locations of fences were collected via GPS along random 3.2 
km long road transects (= 738) to assess overall model accuracy. Using a confusion matrix 
to determine variation between field and modeled fence locations, the total accuracy of the 

model was 73% and Kappa was .40. Although we found inaccuracies associated with large 
parcels (>3 contiguous sections) of cultivated agriculture, our model is a promising step 
towards delineating fencing across the west. These general rules may be used and refined 
in the other areas based on the regional historical context. This new data may advance both 
wildlife research and management/mitigation activities. Using the relative density of fences 
across a region can prioritize conservation efforts at this broad scale. In addition, modeled 
fence locations provide useful and accurate information at a local scale.