How do stand characteristics and crown heterogeneity influence bat activity in forests?

Abstract

In forest ecosystems, changes in three-dimensional structure that may be induced by forest management influence the presence and abundance of plant and animal species. This is particularly true for bats as they make extensive use of the three-dimensional habitat space in forests for foraging and commuting purposes. Bats show different responses to vegetation structure depending on their foraging strategies and ecomorphological and acoustic traits such as wing morphology and echolocation call design. For instance, species with low wing loading and short-range echolocation are more likely to be found foraging within the forest where the vegetation is relatively dense whereas species with high wing loading and long-range echolocation will be found flying above the canopy. Despite the extensive literature on the effects of different forest structural characteristics on bat activity at the stand scale, the influence of canopy architecture has received very little attention. This may be principally due to difficulties associated with both bat sampling and vegetation survey at the canopy level. Using passive acoustic sampling to record bats coupled with airborne Light Detection and Ranging (LiDAR) and terrestrial field surveys for measuring vegetation structure, the aim of the study was to determine key structural forest variables that influence bat activity in managed forests. We deployed bat detectors both at ground and canopy levels in 32 mixed and deciduous forest stands located in Switzerland. Each stand was surveyed between 6 to 12 nights. Bat echolocation calls were identified using a semi-automatic approach. Bats were grouped into different guilds reflecting their echolocation range (namely, short-, mid-, and long-range echolocators; SRE, MRE, LRE). From the LiDAR point clouds, we computed nine variables that describe forest vegetation structure such as mean canopy height, outer canopy ruggedness, and foliage height diversity. Seven variables were measured during field surveys including tree density, shrub vegetation cover, and leaf area index. We tested the effects of forest structure on bat guild activity by fitting a series of generalized linear mixed models. Our results suggest that the activity of the MRE and LRE guilds were clearly associated with canopy characteristics. LRE activity increased with increasing mean canopy height while MRE bats responded positively to increasing outer canopy ruggedness. Species that constitute the MRE guild are edge specialists and may therefore use the external canopy surface as a surrogate of edges. SRE activity was mainly determined by vegetation clutter as we found that foliage height diversity and tree density had a negative influence on activity. We highlight that (i) crown heterogeneity alongside other stand characteristics strongly influence bat guild activity; and (ii) airborne LiDAR brings a top-down perspective that enables accurate characterisation of the forest canopy architecture.