The arrival of white-nose syndrome (WNS) to North America in 2006, and the subsequent decline in populations of cave-hibernating bats have potential long-term implications for communities of forest-dwelling bats in affected regions. Severe declines in wintering populations of bats should lead to concomitant shifts in the composition and relative abundance of species during the staging, maternity, and swarming seasons in nearby forested habitats. We examined capture rates of bats collected in mist nets from 2009 to 2016 to evaluate summer patterns in abundance of species pre-and post-arrival of WNS to Mammoth Cave National Park, KY. The data demonstrated a significant change in overall relative abundances. Myotis septentrionalis (Northern Long-eared Myotis) was the most commonly captured species pre-WNS but declined to 18.5% of its original abundance. Nycticeius humeralis (Evening Bat), uncommonly caught in mist nets pre-WNS, demonstrated the largest increase in capture success following arrival of WNS to the Park, followed by Eptesicus fuscus (Big Brown Bat) and Lasiurus borealis (Eastern Red Bat). These data suggest that losses of cave-hibernating bats to WNS may be leading to a restructuring of foraging bat assemblages in nearby forested habitats, with species less affected by WNS potentially exploiting niche space vacated by bats succumbing to infection with WNS.

Spring emergence and summer maternity seasons are critical to long‐term survival of cave‐hibernating bats inhabiting geographic regions affected by white‐nose syndrome (WNS). White‐nose syndrome reached cave‐hibernating populations of bats in Mammoth Cave National Park (MCNP), Kentucky, USA, in 2013, with significant declines in several species of bats that hibernate in MCNP during winter 2014–2015, including the northern long‐eared bat (Myotis septentrionalis). To assess behavior and roost selection of individuals surviving to spring emergence and the summer maternity season, we radio‐tracked 26 northern long‐eared bats captured at various locations with differing habitat conditions in MCNP during 2015 and 2016. We tracked bats daily to identify roosting sites and patterns in use of roosts, describe habitat conditions associated with roosting sites, and quantify variation in size of colonies occupying roost trees. The average number of bats emerging from roost trees (3.58 ± 0.57 [SE]) was low compared with estimates for the species in studies completed pre‐WNS. Adult female bats selected trees within 6 roost areas with no overlap in use of roost areas between non‐reproductive females in spring and pregnant and lactating females in summer. Variation in roost tree and surrounding forest characteristics occurred across sex and reproductive condition classes, with males and pregnant and lactating females roosting at higher landscape positions than non‐reproductive females. Comparisons of roosting habitat models demonstrated the torpor model (decay class, stem diameter, canopy cover) to be parsimonious, regardless of sex and reproductive condition class, suggesting importance for choosing roost trees that facilitate use of daytime torpor. Such a strategy would promote energy savings and be consistent with behaviors anticipated for bats vulnerable to WNS effects during winter hibernation because these bats are potentially compromised in health and physiologic condition upon emergence from hibernation in spring. We encourage land managers responsible for roost trees of northern long‐eared bats to consider seasonal and landscape‐level variation in roosting habitat needs of these bats. © 2017 The Wildlife Society.

Context The roosting habits of many temperate zone bats are well documented at microhabitat scales, but fewer studies have included multi-scale assessments of landscape patterns in bat roost site selection. Objectives To identify and assess at the landscapescale the location of spring and early season maternity roosts of female northern long-eared bats (Myotis septentrionalis) from 2015 to 2016 at Mammoth Cave National Park (MACA), Kentucky, USA. Methods We used mist-nets and radiotelemetry to catch and track bats to roost trees across the landscape of MACA. Data on roosting sites were evaluated using spatial point pattern analysis to examine distributional trends of roosts. A variety of spatial covariates were used to model the effect of landscape pattern, including: forest type, elevation, and proximity to hibernacula, water, and road corridors. Results Data indicate that roost locations of female northern long-eared bats in MACA were typically situated within 2000 m of known winter hibernacula, occurring more often at higher elevations in mesic upland deciduous forests, and in close proximity to water sources and roads. We present hypotheses to account for the patterns observed in relation to landscape features and habitat resources in the Park. Conclusions Our data indicate that a more comprehensive understanding of habitat requirements which includes empirically-based, landscape-scale patterns, and not solely considerations at stand or local levels, could lead to better informed management policies targeting conservation of maternity habitat of forestdwelling bats, including the northern long-eared bat, a species in decline throughout much of its distribution in North America. Keywords Bats Á Landscape patterns Á Maternity season Á Myotis septentrionalis Á Roosting habitat Á Spatial point pattern Á Spring staging Á Winter hibernacula

No abstract