Dispersal genetic relationships and landscape use by colonizing wolves in the central Rocky Mountains

Boyd-Heger, Diane

Wolves were eradicated from Montana in the 1930s and the adjacent Canadian Rockies by the 1950s, followed by recolonization in the 1980s. I studied wolf recovery in and near Glacier National Park, Montana, from 1979-1997. Wolves dispersed from the protected refugia of Glacier National Park and colonized areas in northwestern Montana, Idaho, southeastern British Columbia, and southwestern Alberta. During this time period 31 of 58 tagged wolves dispersed, and 3 of those dispersed twice. Most wolves (57%) did not conduct any exploratory forays 3 months prior to permanent separation from their natal pack. Wolves usually left their natal home range quickly (median = 4 days) once they separated from the pack. Mean dispersal distance was not significantly different for males (113 km) and females (78 km). Wolves tended to disperse in a northerly direction to areas of higher wolf density. January-February and May-June were peak months for dispersal. Mean dispersal age (male = 28.7 mos, females = 38.4 mos) was not correlated with maximum pack size. Twenty percent of dispensers were > 57 months old at dispersal. Sex ratios of dispersers and captured wolves (both 71%F) were significantly different from parity. Survival for dispersers and biders did not differ. Eighty percent (n = 30) of wolf mortalities were caused by humans, with proportionately more dispersers (90%) than biders (60%) dying from human causes. Dispersers produced significantly more litters than biders. Effects of mountainous terrain on woif dispersal are discussed. Colonizing wolves had adequate genetic variation which was similar to that found in other wolf populations in North America. Sufficient gene flow occurred between the colonizing and source populations to minimize genetic divergence. Packs were founded by multiple, unrelated wolves from Canada, with no evidence of a population bottleneck. Reintroduced wolves were moderately genetically divergent from the colonizing population and have adequate genetic variation. I detected no evidence of hybridization between wolves and coyotes. Detection of non-native canids and management considerations are discussed. I analyzed landscape attributes selected by 6 colonizing wolves including elevation, slope, aspect, distance to water, distance to roads, road density, and canopy cover. Wolves selected for landscapes with relatively lower elevation, flatter terrain, and closer to water and roads than expected based on availability inside and outside of their new home range. I built a logistic regression model using the univariately significant attributes (elevation, slope, and distance to roads) to predict wolf presence in areas of potential colonization. Impacts of habitat fragmentation and management considerations are discussed.