Fence types influence pronghorn movement responses

Impediments that constrain animal movements across spatiotemporally heterogeneous landscapes can result in reduced or complete loss of access to critical resources. Across their range in North America, pronghorn (Antilocapra americana) are exposed to fences that can affect their ability to permeate the landscape, access critical resources, and respond to climatic variations. Understanding pronghorn movement responses to fences is essential for improving landscape permeability; however, prior studies provide only limited insight due to lack of information on fence characteristics and small sample sizes. Our study used hourly collar locations from adult female pronghorn in six herds in Montana, USA, and identified encounters with mapped fences to evaluate three movement responses (i.e., probability of an unaltered initial response, probability of crossing following an altered initial response, and passage time following an altered initial response) as a function of fence and landscape attributes. Based on 5581 encounters identified from movement pathways of 265 collared pronghorn and 979 km of mapped fences, we found that variability in pronghorn fence response was correlated with fence type. Woven wire fences substantially reduced unaltered initial and crossing responses and increased passage times as compared with low (i.e., average lowest wire height <41 cm) or high (i.e., average lowest wire height ≥41 cm) strand fences. Both low and high strand fences elicited similar responses of being relatively permeable at the initial encounter with reduced permeability thereafter. Fence crossing probabilities following altered initial responses increased through time modestly for strand fences but only negligibly for woven wire fences, with passage times averaging approximately 14 h. Pronghorn knowledge of and fidelity to specific permeable locations along fences, which may be due to inconsistent fence and landscape characteristics along the fence stretch, likely allow some woven wire fences and most strand fences, regardless of the average lowest wire height, to be permeable. Improving landscape permeability for pronghorn should focus on removing woven wire fences, replacing woven wire fences with strand fences, and incorporating variation in the lowest wire heights into new fence designs or modifications of existing fences.