Human demand for food, fiber, and space is accelerating the rate of change of land cover and land use. Much of the world now consists of a matrix of natural forests, managed forests, agricultural cropland, and urbanized plots. Expansion of domestic energy production efforts in the United States is one driver predicted to influence future land-use and land management practices across large spatial scales. Favorable growing conditions make the southeastern United States an ideal location for producing a large portion of the country’s renewable bioenergy. We investigated patterns of bat occurrence in two bioenergy feedstocks commonly grown in this region (corn, Zea mays, and pine, Pinus taeda and P. elliottii). We also evaluated potential impacts of the three major pathways of woody biomass extraction (residue removal following clearcut harvest, short-rotation energy plantations, and mid-rotation forest thinning) to bat occurrence through a priori land-use contrasts. We acoustically sampled bat vocalizations at 84 sites in the Southeastern Plains and Southern Coastal Plains of the southeastern United States across three years. We found that mid-rotation thinning resulted in positive effects on bat occurrence, and potential conversion of unmanaged (reference) forest to managed forest for timber and/or bioenergy harvest resulted in negative effects on bat occurrence when effects were averaged across all species. The effects of short-rotation energy plantations, removal of logging residues from plantation clearcuts, and corn were equivocal for all bat species examined. Our results suggest that accelerated production of biomass for energy production through either corn or intensively managed pine forests is not likely to have an adverse effect on bat communities, so long as existing older unmanaged forests are not converted to managed bioenergy or timber plantations. Beyond bioenergy crop production, mid-rotation thinning of even-aged pine stands intended for timber production, increases to the duration of plantation rotations to promote older forest stands, arranging forest stands and crop fields to maximize edge habitat, and maintaining unmanaged forests could benefit bat communities by augmenting roosting and foraging opportunities.

As part of a large investigation on bioenergy and biodiversity, we collected bat occurrence and activity data from 84 study sites stratified across three broad geographic regions in the states of Alabama, Florida, and Georgia, USA. We selected sites so that they occurred in eight different land use types: clearcuts (0-3 years post-harvest) with logging residues removed, clearcuts of similar age with residues retained, young plantations (8-10 years), mature plantations (20-32 years), recently thinned stands (12-16 years), stands of similar age (12-16 years) that were not thinned, corn fields, and natural (unplanted) mature pine stands. The latter were natural forests of longleaf pine (Pinus palustris) representative of the historical conditions that prevailed across the study region prior to the advent of modern agricultural and forestry production. All other forest surveys occurred in plantations of slash pine (P. elliottii) and loblolly pine (P. taeda), the two most common species grown for timber in the region.

We surveyed bat activity by recording bat echolocation calls. We placed four Anabat II echolocation detectors coupled with Zero Crossing Analysis Interface Modules (ZCAIMS, Titley Electronics, Ballina, NSW, Australia) simultaneously at each study site on 336 nights between late March and late July during 2013-2015. Two detectors were placed at the center of each site and two near the edge. We attempted to survey each site four times per field season. Detectors were programmed to turn on 30 min prior to sunset and off 30 min after sunrise.

We used Kaleidoscope Pro 3.14B (Wildlife Acoustics, Inc.) with the ‘Bats of Florida Classifier’ (3.1.0) to identify and analyze calls. We used the “balanced” sensitivity/accuracy setting in Kaleidoscope Pro (which gives parity to both sensitivity and accuracy) and assessed all files with characteristics of more than one species, retaining only those in which >50% of the calls matched the species identification assigned by the software.