Detection and modeling of bat species occupancy at multiple scales across a forested landscape in southeastern Missouri

Abstract

[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT REQUEST OF AUTHOR.] Populations of forest dwelling bats have been in decline in recent years, and have therefore become increasing foci in forestry research efforts. Difficulty in bat species detection complicates bat - habitat studies. Findings from this study reveal that acoustic methods provided greater detectability of species presence at a site than mist netting. Species presence was detected using two acoustic detections, and while individuals were equally detectable, detector placement influenced detectability. Incorporating differences in detectability, a priori habitat models from three spatial scales were compared to determine occupancy of five bat species. No habitat models adequately described the occupancy of gray bats. Red bat occupancy was inversely related to basal area (BA), directly related to canopy closure and varied among ecological subsections within the study area. Eastern pipistrelle bat occupancy was inversely related to BA and directly related to canopy closure. Northern long-eared bat occupancy was directly related to amount of hard edge in the landscape and was greater in shortleaf pine forests and woodlands. Indiana bat occupancy was directly related to BA of large diameter snags, directly related to the amount of hard edge and proportion of non-forested land cover in the landscape.