Main content area

Transdisciplinary habitat models for elk and cattle as a proxy for bovine tuberculosis transmission risk

Brook, Ryan K., McLachlan, Stéphane M.
Preventive veterinary medicine 2009 v.91 no.2-4 pp. 197-208
elks, wildlife diseases, cattle, cattle diseases, Mycobacterium bovis, bovine tuberculosis, disease transmission, risk factors, habitats, statistical models, global positioning systems, ecological processes and phenomena, pastures, grazing, water use, temporal variation, spatial distribution, Canada
Zoonotic diseases such as bovine tuberculosis (TB) that infect wildlife and livestock are particularly difficult to eradicate where wild animals make extensive use of agricultural landscapes. Transmission of TB between cattle (Bos taurus) and wild elk (Cervus elaphus) in southwestern Manitoba, Canada remains poorly understood but there is a risk when comingling occurs on summer pasture. Elk use of cattle summer pastures was assessed using ecological data (187 VHF and 25 GPS collared elk monitored over four years representing 8% of the elk population). Local knowledge was documented by conducting interviews and participatory mapping exercises with 86 cattle producers (98% of those within the study area). Of the 294 cattle pastures mapped by farmers, 13% were used by radio-collared elk, 38% were reported by farmers as being used by elk, and 42% were identified as used by elk when both when all datasets were combined. Cattle pastures that had been used by elk and those that had no elk were compared using binary logistic regression based on each of the three datasets (i.e. farmer observations, radio-collared elk on pasture, and combined dataset). For all three datasets, distance to protected area and proportion of forest cover on the cattle pasture were identified as the most and second most important predictor variables, respectively. There was strong agreement among the relative probabilities of elk occurrence on each pasture derived from the resource selection function (RSF) models developed using farmer interviews and elk collaring data. The farmer interview and collar datasets were then combined to generate a final integrated RSF map summarizing the probability of elk-cattle comingling and were contrasted over each of four cattle grazing seasons (spring, early summer, late summer, and autumn). These predictive maps indicate that use of cattle pastures by elk is extensive, particularly in spring and early summer. Farmer observations indicate that elk and cattle share water sources and livestock mineral supplements on pasture. Local knowledge and conventional ecological data complement and validate one another and help us better understand the temporospatial aspects of shared space use among wildlife and livestock and more generally the risks of disease transmission in agricultural landscapes.