Winter diet of bats in working forests of the southeastern U.S. Coastal Plain

Working forests comprise a large proportion of forested landscapes in the southeastern United States and are important to the conservation of bats, which rely on forests for roosting and foraging. While relationships between bat ecology and forest management are well studied during summer, winter bat ecology remains understudied. Hence, we aimed to identify the diet composition of overwintering bats, compare the composition of prey consumed by bat species, and determine the potential role of forest bats as pest controllers in working forest landscapes of the southeastern U.S. Coastal Plain. During January to March 2021–2022, we captured 264 bats of eight species. We used DNA metabarcoding to obtain diet composition from 126 individuals of seven bat species identifying 22 orders and 174 families of arthropod prey. Although Coleoptera, Diptera, and Lepidoptera were the most consumed orders, we found that bats had a generalist diet but with significant differences among some species. We also documented the consumption of multiple insect pests (e.g., Rhyacionia frustrana) and disease vectors (e.g., Culex spp). Our results provide important information regarding the winter diet of bats in the southeastern U.S. Coastal Plain and their potential role in controlling economically relevant pest species and disease vectors.


Results
We captured 264 individuals of eight bat species from late-January to mid-March 2021-2022, collecting fecal samples from 209 individuals, from which we selected samples from 195 individuals.After bioinformatics processing and quality filtering, we obtained diet composition from 126 individuals of seven species (Table 1).None of the fecal samples from the two captured Dasypterus intermedius passed quality control.We obtained 2703 unique Operational Taxonomic Units (OTUs), 2127 (78.69%) of which were matched to sequences in the Barcode of Life Database v3 (BOLD) reference collection after pruning.These matches belonged to 22 orders, 174 families, and 422 genera or species (Supplementary Material 1).Within analyzed fecal samples, Coleoptera (n = 610 OTUs), Diptera (n = 684 OTUs), and Lepidoptera (n = 551 OTUs) were the most consumed orders (Supplementary Material 2).These three orders were the most consumed orders by all bat species except Lasiurus cinereus (Fig. 2; Table 1), which had a scarce representation of Coleoptera, although with a sample size of only three individual bats.For the remaining bat species, percentages varied among species such as L. borealis where 41.46% was based on Lepidoptera, to species such as Eptesicus fuscus, where 45.25% corresponded to Coleoptera, or Perimyotis subflavus with a preference for dipterans (49.21%) (Fig. 2, Table 1).

Discussion
Our results show a great variability (22 arthropod orders) in diet across bat species, highlighting the consumption mainly of Coleoptera, Diptera, and Lepidoptera.As expected, diet composition differed among bat species with different foraging strategies, but surprisingly also among species in similar foraging guilds.Our findings complement previous work conducted during summer indicating that bat diets vary seasonally 33,41,45 , which may depend on insect phenologies and weather conditions.Specifically in winter, seasonal prey limitations may lead to shifts towards more generalist behavior in several bat species, with changes in dietary composition and diversity compared to other times of the year and life stages of bats.Further, our results confirm the role of overwintering bat communities as consumers of agricultural and forest pests and potential arthropod vectors of human and animal diseases.
Traditionally, dietary preferences of insectivorous bats have been explained based on differences in their ecomorphologies and morphometric characteristics, with larger species feeding on larger insects or insects with more resistant exoskeletons 43,46 .The energetic cost-benefit of feeding on smaller insects compared to larger insects or insects with more resistant exoskeletons would lead to dietary selection based on the morphological characteristics of each bat species 47 .For example, it is often questioned whether species, especially large-body bats, can meet energy demands consuming small soft-body insects such as flies and mosquitoes 48 .However, availability and temporal variation of prey may lead to shifts in preferences towards more generalist diets.Eptesicus fuscus, the second largest of the seven species captured, is considered a coleopteran specialist e.g., 30,49,50 .Recently, this assumption has been questioned, placing E. fuscus instead as generalist consumers in summer with preferences for Coleoptera when available 33,36,38 .In our study, a large portion of their diet was Coleoptera, but we found high dietary diversity, including many dipterans, possibly attributed to more dipterans in winter relative to other insect orders 39 .Flexible hunting strategies may allow bat species to adapt to different food availabilities by consuming prey that is abundant at the time, although of non-optimal sizes or other characteristics 51 .In contrast, the diet of L. cinereus, the largest species in our study and one of the largest species in North America, was comprised primarily of Diptera and Lepidoptera.Although our results should be interpreted with caution because of the small sample size (n = 3 individual bats), previous studies suggest that L. cinereus select large, soft-bodied insects (e.g., Lepidoptera and Neuroptera) and avoid small or hard-bodied insects (e.g., Coleoptera, Diptera, and Hemiptera) [52][53][54] .Most of the dipterans we documented in the diet were large crane fly species such as Nephrotoma ferruginea (Table 2), which supports a preference for large, soft-bodied prey.
The remaining bat species in our study are smaller and adapted to foraging along forest edges or within forests e.g., 22,55,56 .Lasiurus borealis and L. seminolus share similar ecomorphologies, to the point that it is difficult to separate them by the characteristics of their echolocation calls or external morphology 57,58 .Both species have robust dentition like other Coleoptera specialists 47 .However, both ours and previous dietary analyses indicate that they consume a wide range of soft-bodied prey such as Diptera, Lepidoptera, and Neuroptera e.g., 12,40,59 .The dietary differences identified between N. humeralis with L. borealis and L. seminolus could be due to the partitioning of selected prey within the same spaces and slight differences in ecomorphology and general external morphologies.The morphometrics and dentition of N. humeralis together with previous summer dietary analyses show flexibility in its diet, which allows it to eat a wide range of arthropods, from coleopterans to soft-bodied prey 30,47,60 .Our results confirm similar preferences in the diet during winter, where we observed high dietary diversity, distinguishing N. humeralis from other species.These findings are supported by the presence of OTUs from all 22 identified orders.www.nature.com/scientificreports/Previous works indicate that M. austroriparius and P. subflavus consume primarily soft body prey 30,59 .Using morphological dietary analyses, Feldhamer et al. 30 found that both species consumed mainly trichopterans, suggesting a diet of soft-bodied species found predominantly above water.However, we observed numerous Coleoptera OTUs present in the diet of M. austroriparius, which highlights its dietary plasticity, consuming hard-bodied insects in winter.Differences between M. austroriparius with diets of L. seminolus and N. humeralis suggest a tendency towards a more specialized diet likely influenced by its forest-interior foraging strategies 22 .Perimyotis subflavus is among the smallest bats in North America 61 .Previous studies have noted that P. subflavus shows an opportunistic approach when foraging, exhibiting one of the most diverse diets in eastern North American bat species 59,62 .However, we found that P. subflavus consumed the second lowest number of orders, but a large proportion of dipterans, which concurs with previous research that documented frequent consumption of dipterans by P. subflavus in winter 39 .Disproportionate consumption of dipterans in winter compared to other seasons could be a consequence of a selection for small soft-bodied prey and a higher abundance of Diptera relative to other orders.
To our knowledge, our study is the first to document the consumption of agricultural and forest pests by winter bat communities in the southeastern U.S. Coastal Plain where intensive pine management and agriculture dominate the landscape.Among the most common forest pest species we documented in bat diets, R. frustrana, is an economically important pest of young pines, especially for loblolly pine (Pinus taeda), the preferred host species 63 .Our study coincided with the time period when R. frustrana typically emerges 64 , highlighting the importance of this moth to most bat species when availability is high.Additionally, H. pales was also widely consumed by most bat species in our study.Hylobius pales causes damage to young pine seedlings and is a vector of commercially damaging Ophiostomatalean "blue-stain" fungi such as Leptographium spp., which discolor and degrade the value of colonized wood 65,66 .Our research also reveals the consumption of various agricultural pests by bats, such as the moths C. peritana and H. scabra, which likely inhabit agricultural areas embedded within the working forest landscapes.While H. scabra was not the most frequently consumed pest nor found in large numbers, it was present in the winter diet of five bat species, including migratory species like L. borealis and L. cinereus 20 .Hypena scabra is a migratory moth, with most populations overwintering south of the midwestern U.S. Corn Belt 36,67 .Although it is generally of minor economic importance, this moth is one of the most common defoliating insects in alfalfa and soybean fields 68 .Consumption of overwintering populations of H. scabra in this ecoregion may provide a yet undocumented ecosystem service in controlling populations outside of the growing season and outside the major crop-producing areas of the Corn Belt.Overall, our findings suggest that consumption of agricultural and forest pests by bats in late winter and early spring could play a crucial role in minimizing damage during the subsequent growing season, highlighting the potential significance of bats as natural pest controllers in agricultural and forested landscapes.
Finally, we identified several species of flies and mosquitoes (Diptera) in winter diets that are recognized as threats to human health.Global concern about mosquitoes (Family Culicidae) stems from their significant impact on public health, attributed to their role as disease vectors.This impact extends to the transmission of multiple diseases (e.g., West Nile virus 69 , malaria 70 , dengue 71 , dog (Canis lupus familiaris) heartworm 72 , myxomatosis 73 , or avian malaria 74 ) with far-reaching consequences for human societies, wildlife, and ecosystems.Our results reveal a diverse array of mosquito vectors, including species of the genera Aedes and Culex, common vectors of diseases such as West Nile virus.In addition, we identified malaria vectors, such as Anopheles mosquitoes, and specific cases of non-native mosquitoes, such as Aedes japonicus, implicated in the transmission and/or maintenance of arboviruses, both endemic to the region (e.g., West Nile virus) and exotic (e.g., Zika, dengue, and chikungunya) 75 .Although little known to date 42 , our results also demonstrated consumption of other dipterans that may pose a threat to wildlife, livestock, and poultry.For example, we confirmed consumption of Diptera such as black flies (Family Simuliidae), which are capable of transmitting pathogens, including protozoa and nematode worms to vertebrates, and are thus a veterinary concern, even if none of them cause disease in humans in North America 76 .
Identifying diet composition in overwintering bats and recognizing differences in prey consumption among species contribute valuable insights into the ecological role of bats in working forest landscapes.As these forests are crucial for remnant populations affected by WNS and migratory species affected by wind energy development, understanding winter bat foraging ecology becomes paramount.The potential role of forest bats as pest controllers during winter underscores the importance of managing working forests in ways that support the diverse dietary needs of the bat community.Our findings have a direct connection to economics and timber quality; for example, R. frustrana is known to have a drastic impact on pine growth, both in tree height and diameter 63 .Hence, proactive forest management practices that improve bat habitat conditions 77 , such as retention of hardwoods, trees with exfoliating bark, and cavity trees (live and dead) also increase their economic benefits.Additionally, our results show the role of bat communities outside forest boundaries consuming agricultural pests and other potential arthropod vectors of disease.Conservation efforts thus may consider ecological services provided by bats, including their ability to contribute to control of agricultural and forest pests and potentially limit the spread of disease vectors.We also emphasize the role of private lands conservation in promoting bat habitat and their consequent ecosystem services.

Study area
We conducted our study on private, working forest landscapes in late-January through mid-March, 2021-2022 in four states (Georgia, Louisiana, Mississippi, and North Carolina) (Fig. 1).Our study areas were characterized by a mosaic of forested landscapes with crop fields and areas with varying degrees of development 78 .We selected study areas > 3000 ha that consisted primarily of planted loblolly pine stands interspersed with riparian management areas (predominantly mature hardwood stands), roads, and wildlife openings.Management activities were typical of commercial forestry operations in the region, including clear-cutting at 20-35 years, mechanical and/ or chemical site preparation, and planting 182-283 pine trees ha − 1 79 .Competing vegetation was temporarily suppressed through herbicide applications, prescribed fire, or mechanically, with most stands being thinned at least once.We defined January-March as the winter sampling season, as mean nighttime temperatures are lowest (typically < 10 °C) during this time in most of the Coastal Plain region 22  www.nature.com/scientificreports/sometimes be affected by non-homogeneity of dispersion for unbalanced sampling schemes, we also performed a permutational dispersion test 94 .Lastly, we performed post-hoc pairwise multilevel comparisons using the pairwise Adonis package with Bonferroni adjustment (version 0.4.1 95 ) to determine differences among species.

Figure 1 .
Figure 1.Location of study sites (circles) in the southeastern United States Coastal Plain where bat sampling was conducted from late-January to mid-March 2021-2022.Landscape cover types derived from a reclassification of The National Land Cover Database 2021.

Figure 2 .
Figure 2. Winter diet including order, family, and genus of seven bat species in private, working forests of the southeastern U.S. Coastal Plain from late-January to mid-March 2021-2022.Colors represent number of samples and width of nodes represent number of Operational Taxonomic Unit (OTU) counts for each taxonomic level.

Table 1 .
44ts captured, number of fecal samples collected, number of samples analyzed, number of Operational Taxonomic Units (OTUs) for each bat species within insect orders, and bat species foraging strategies in private, working forest landscapes across four states (Georgia, Louisiana, Mississippi, and North Carolina) of the southeastern U.S. Coastal Plain from late-January to mid-March 2021-2022.Foraging strategies follow Norberg and Rayner43and Denzinger and Schnitzler44.