Search

We examined intra- and interspecific patterns of roost-site selection by females of three species of forest-dwelling bats (silver-haired, Lasionycteris noctivagans; big brown, Eptesicus fuscus; and California, Myotis californicus) in the Pend d’Oreille Valley in southern British Columbia. Roost-sites were located by outfitting reproductive and non-reproductive female bats with radio-transmitters and tracking them during the daytime. The tree and site characteristics of roost trees used by each bat species were then compared to those of randomly available trees with cavities, as well as to each other, using logistic regression. All three bat species selected roosts that were taller than cavity trees, and the majority of roost trees were emergent or canopy trees. In addition, tree species and the type of cavity were important predictors of whether trees were used as roosts in both intra- and interspecific analyses. The majority of silver-haired and big brown bat roost trees were in hollows in live trembling aspen trees (Populus tremuloides), but both species used small numbers of roosts on conifer snags. In contrast, the majority of California bat roosts were beneath loose bark on conifer snags, particularly Douglas-fir (Pseudotsuga menziesii), in intermediate stages of decay. In general, site characteristics were only weak predictors of differences between roost and random trees, and roost trees used by the three bat species, but roost trees were found in patches with lower percent canopy closure and greater density of available trees. These results suggest that while all three species of forest-dwelling bats use large trees, within this restricted range of trees there is considerable variation among species in the types of cavities and the tree species used. Maintaining stands containing mature trembling aspen is necessary to provide silver-haired and big brown bats with suitable roosting opportunities, but to meet the needs of California bats, areas rich in conifer snags must be protected. Thus, unique management strategies or silvicultural practices may be required to maintain suitable habitat for all bat species in a given area.

Mule deer (Odocoileus hemionus) populations throughout the west appear to be declining, whereas white-tailed deer (Odocoileus virginianus) populations are increasing. We compared abundance, number of fetuses per female (maternity rate), recruitment, and cause-specific adult ([Formula: see text]1 year old) mortality rate for sympatric mule deer and white-tailed deer in south-central British Columbia to assess population growth for each species. White-tailed deer were three times more abundant (908 ± 152) than mule deer (336 ± 122) (mean ± 1 SE). Fetal rates of white-tailed deer (1.83) were similar to those of mule deer (1.78). There was no statistically significant difference in recruitment of white-tailed deer (56 fawns : 100 does) and mule deer (38 fawns : 100 does). The annual survival rate for adult white-tailed deer (SWT = 0.81) was significantly higher than that for mule deer (SMD = 0.72). The main cause of mortality in both populations was cougar predation. The lower mule deer survival rate could be directly linked to a higher predation rate (0.17) than for white-tailed deer (0.09). The finite growth rate (λ) was 0.88 for mule deer and 1.02 for white-tailed deer. The disparate survival and predation rates are consistent with the apparent-competition hypothesis.

We followed nine radiotagged female California bats ( Myotis californicus ) to 19 roosts in trees at two study sites in southcentral British Columbia. Bats regularly switched roosts and the number of bats emerging from known roosts fluctuated widely. Logistic-regression analysis showed that reproductively active females preferred trees further away from other trees of the same height or greater, and closer to neighboring trees, relative to available trees in the immediate vicinity of the roost. Diameter at breast height and distance to the nearest tree of the same or greater height explained significant proportions of the variation between roost and available trees found in other areas of the same forest stand. Roost trees had significantly larger diameters and were further away from trees of the same or greater height. Percentage of canopy closure also explained a significant proportion of the variation between roost and available trees, such that roost trees were situated in areas with lower canopy closure than available trees in other areas of the same stand. Roost and available trees were classified correctly >70% of the time based on the above tree characteristics. However, roost trees were only correctly classified 39% of the time in the analysis of site characteristics. Our results, combined with those from other recent studies, lead to the general hypothesis that forest-roosting bats require a number of large dead trees of specific species, in specific stages of decay, and that project above the canopy in relatively open areas. For management and conservation reasons, there should be a strong incentive to evaluate this hypothesis for a variety of species in a variety of locations to test its generality.