Roost site selection by lesser short‐tailed bats (Mystacina tuberculata) in mixed podocarp‐hardwood forest, Whenua Hou/Codfish Island, New Zealand

Abstract

Roosting ecology of the lesser short‐tailed bats (Mystacina tuberculata) was studied in mixed podocarp‐hardwood forest on Whenua Hou/Codfish Island, New Zealand during winter 1996, and winter to summer 1998–99. The structural characteristics of roosts used by communal groups and by solitary bats are described, and roost selection is examined at three scales: forest stand, tree, and cavity. Differences in structural characteristics and microclimate of roosts used in winter and in summer are compared. Forty‐three bats were radio‐tracked and 122 tree roosts located. Communal groups occupied 47.5% of roosts and solitary bats occupied 52.5%. Bats selected roosts in Hall's totara (Podocarpus hallii) trees more than expected based on the availability of this tree species. Roost trees had significantly larger stem diameters and greater numbers of cavities than random trees. Communal bats demonstrated a higher degree of roost selectivity than solitary bats. Communal roosts had significantly less vegetation clutter surrounding roost entrances, larger stem diameters at cavity height, larger internal cavity heights, and thicker cavity walls than potentially available cavities. Communal roost cavities were consistently larger than solitary roosts. Solitary roosts had less open space around them than communal roosts and random cavities. Communal roosts were well insulated and exhibited stable internal temperatures. Structural and thermal characteristics did not differ between roosts used in winter and in summer. Differences between characteristics of communal and solitary roosts, and selection of well‐insulated cavities in some of the largest available trees, appears consistent among different forest types used by lesser short‐tailed bats in New Zealand, and with behaviour of many other tree‐dwelling bat species in temperate climates. Descriptive statistics of trees provide guidance to forest managers, especially those practicing silviculture, for identifying potential bat roosting habitat. Regression analyses of cavity characteristics provide insight into factors explaining the rarity of favoured cavities. Intraspecific differences in roost selection by bats should be considered when undertaking conservation management actions such as identifying trees for retention through logging cycles, or designating and designing reserves.