Roost site selection by forest-dwelling male Myotis in central Ontario, Canada

We used radiotelemetry and random exit counts to determine roost site selection by male northern long-eared bats (Myotis septentrionalis) and unidentified Myotis, either northern long-eared or little brown bats (M. lucifugus) in a conifer-dominated, mixedwood forest landscape in central Ontario, Canada. We compared the characteristics of snags used as roosts (n = 26), with randomly located snags and random points (n = 52 and 50, respectively), at three spatial scales: focal tree, surrounding forest, and landscape. Snags used as roost sites by these bats differed from random snags for 19 of the 23 variables measured (P < 0.05). Logistic regression models were derived which suggested that the bats selected large snags, in open canopies, of intermediate stages of decay, that were located in upland areas, but away from water bodies. Bats may have selected roost sites for their thermal advantage in the mornings, as most used cavities or bark on the south and east sides of snags. To ensure that snags of appropriate characteristics persist for these two species of Myotis, forest managers in the northern Great Lakes forest region should retain, and manage for, large white pine (Pinus strobus) and trembling aspen (Populus trentuloides) snags at sites in forests that meet these criteria. Such a regime is possible under a selection harvest that is normally practiced in these mixedwood forest types. (C) 2004 Elsevier B.V. All rights reserved.