Bat activity is related to habitat structure and time since prescribed fire in managed pine barrens in New England

Introduction

Bats are generally understudied and over-feared. Their reputation as disease carriers has hindered conservation efforts (Frick, Kingston & Flanders, 2020), and their nocturnal nature makes them a difficult taxonomic group to study (Villarroya-Villalba et al., 2021). Research has recently emerged on how bats respond to land management techniques, particularly prescribed fire and timber harvesting (Gotthardt, Kelley & Schwager, 2015; Austin et al., 2018; Steel et al., 2019; Divoll et al., 2021).

Despite the fear and suspicion surrounding bats, they are an incredibly important taxonomic group. Bats constitute one-fifth of mammalian diversity, act as pollinators, and control common flying insect pests (Steel et al., 2019). Habitat loss has contributed to bat population decline as urbanization increases and demand for resources increases timber harvesting and agriculture (Frick, Kingston & Flanders, 2020). Other dangers that are increasing bat mortality include wind turbines, powerlines, and radio towers (Huzzen, Hale & Bennett, 2020). The highly migratory hoary bat (Lasiurus cinereus) makes up roughly one-third of bat-related wind turbine fatalities and is a poorly censused species that is present in New England (Cornman et al., 2021). In addition, White Nose Syndrome (WNS), caused by the fungus Pseudogymnoascus destructans, has ravaged North American bat populations (Brooks , 2011). The disease easily spreads in cold, moist environments and has caused over a 90% population decline in cave-dwelling northern long-eared bats (Myotis septentrionalis), little brown bats (Myotis lucifugus), and tricolored bats (Perimyotis subflavus; Cheng et al., 2021).

Understanding the bat ecology in an area can be accomplished through acoustic data gathered from audio surveys (Gorresen et al., 2008; Blejwas, Lausen & Rhea-Fournier, 2014; Gotthardt, Kelley & Schwager, 2015; Wordley et al., 2015; Austin et al., 2018; Fraser et al., 2020; Taillie et al., 2021). Such techniques require less labor and expertise than physical captures in addition to being non-invasive, while still providing insight into the species and activity levels of the area. Additionally, Gomes, Appel & Barber (2020) found that mist-netting in tropical areas favored the capture of bats in Family Phyllostomidae when other Families of insectivorous bats are in the majority. While we are not in a tropical study site, there is likely still a bias in which bats are caught by mist nets. Although acoustic data does not lead to exact population numbers due to our inability to determine if one or multiple bats were making calls, it can be an inherent first step to characterizing bat populations with presence or absence information.

Nine bat species (all insectivorous) are native to New Hampshire and Massachusetts (Brooks & Ford, 2005): big brown bat (Eptesicus fuscus), silver-haired bat (Lasionycteris noctivagans), hoary bat, eastern red bat (Lasiurus borealis), little brown bat, eastern small-footed bat (Myotis leibii), northern long-eared bat, Indiana bat (Myotis sodalis), and tricolored bat. Big brown bats, silver-haired bats, hoary bats, and eastern red bats prefer to roost in mature or dead trees, along with building crawl spaces. All four species are migratory (Whitaker & Hamilton, 2019). The other bat species (little brown bat, eastern small-footed bat, northern long-eared bat, Indiana bat, and tricolored bat) are all endangered in Massachusetts and New Hampshire, largely due to their exposure to WNS as they usually overwinter in caves. However, the Indiana bat has not actually been recorded in Massachusetts since 1939 (Massachusetts Division of Fisheriesildlife, 2017). During the summer, these bats roost in foliage, tree cavities, buildings, and rock crevices (Massachusetts Division of Fisheriesildlife, 2017). Big brown bats generally start foraging 20 min after sundown, mainly in riparian or open areas, and along edge habitats (Wisconsin Department of Natural Resources, 2022). Northern long-eared bats often forage in “structurally complex habitat” sometimes over water, while little brown bats primarily forage near or over water (Massachusetts Division of Fisheries and Wildlife, 2019). Three of the species—big brown bats, silver-haired bats, and hoary bats—have distinct lower-frequency calls that differ from the calls of Myotis species. This can make it difficult to distinguish the three species from each other given the few acoustic nuances that differentiate them. However, their unique calls make them easily distinguishable from most other bat species (Blejwas, Lausen & Rhea-Fournier, 2014).

Pitch pine (Pinus rigida)–scrub oak (Quercus ilicifolia) barren habitat is a rare ecosystem across the New England landscape that is known to support high Lepidopteran species richness (Wagner, Nelson & Schweitzer, 2003; Grand et al., 2004) and may, in turn, support bat populations (Wickramasinghe et al., 2004; Leuenberger et al., 2016). This habitat type was formed by the recession of glaciers and resulting sand deposits, which led to a suitable environment for fire-adapted vegetation. Due to decreased wildfires from human intervention, pitch pine barrens have become overgrown, resulting in changes in ecosystem structure and function and increased wildfire risk (King et al., 2011). Pine barrens’ unique characteristics lend themselves to support rare plant and animal species, so prescribed burns and tree thinnings are used to manage the land, helping create a balance of pitch pine and scrub oak (Simmons & Hawthorne, 2014). For example, King et al. (2011) demonstrated that birds associated with shrublands and open forests, which are comparatively rare in highly forested areas like New England, were more prevalent in the managed zones, while birds associated with closed forests were more prevalent in the overgrown forests.

This study examined bat populations at two managed barren sites in New England, Montague Plains Wildlife Management Area (WMA) in western Massachusetts and Concord Pine Barrens in southern New Hampshire. Our study objectives were to (1) describe bat species most commonly found and (2) compare bat activity across habitat characteristics in these two managed pine barrens. In particular, we examined habitat type (i.e., scrub oak, mature pitch pine, treated pitch pine, hardwood forest), habitat structure (i.e., canopy closure), time since prescribed fire, and path width. We hypothesized that bats would be more active in mixed habitats, like treated pitch pine, that allow for places to roost and space to forage (Nelson & Gillam, 2017; Wisconsin Department of Natural Resources, 2022), and bats would be most active in areas that were recently burned due to elevated insect abundance (Campbell et al., 2018; Taillie et al., 2021).

Methods

Study area

Montague Plains Wildlife Management Area (WMA) is located in western Massachusetts within the town of Montague and is owned by the Massachusetts Division of Fisheries and Wildlife. It is over 1,500 acres of a mixture of habitats that resulted naturally from glacial sand deposits. The WMA is characterized by rare vegetation and animals; these include pitch pine–scrub oak barrens, the endangered spreading tick trefoil plant (Desmodium humifusum), several moth species (Lepidoptera), the eastern box turtle (Terrapene carolina carolina), and bird species like the eastern whip-poor-will (Antrostomus vociferus) and brown thrasher (Toxostoma rufum; Simmons & Hawthorne, 2014). The four main habitat types, as described in King et al. (2011), are treated pitch pine, pitch pine forest, scrub oak, and deciduous forest (Figs. 1A–1D, respectively). Montague Plains WMA is actively managed through tree thinning and prescribed burns, which help to reinforce the presence of the historically fire-adapted pitch pine–scrub oak barrens, since this successional habitat may otherwise grow into a pine/oak forest. The wildlife management area contains some undisturbed forested areas, including pine/oak and patches of deciduous forest.

The Concord Pine Barrens are managed in a similar manner to the Montague Plains Wildlife Management Area. The site is located in southern New Hampshire in the city of Concord. The site is managed by New Hampshire Fish and Game for the main purpose of providing habitat for the endangered Karner blue butterfly (Lycaeides melissa samuelis). The pine barrens are maintained with prescribed fires and mechanical disturbance. Native plants, especially lupine (Lupinus), are also introduced back into the approximately 300-acre site. In addition to the pine barrens, young deciduous forest is present in non-disturbed areas (US Fish and Wildlife Service, 2003; Holman & Fuller, 2011). Access to the study sites was provided by staff at the NH Fish and Game Department (H. Holman, S. Houghton) and MA Division of Fisheries and Wildlife (A. Vitz).

Surveys

We surveyed four points (>300 m apart) per habitat type at Montague Plains WMA for a total of 12 points, and one point (80–230 m apart) per habitat type at the Concord Pine Barrens for a total of four points (Fig. 2; Kay, Sadlon & Bakermans, 2023). Across the month of June 2022, surveys were repeated 3–4 times at each survey point. We performed surveys on foot using the Echo Meter Touch 2 Pro (Wildlife Acoustics) connected to a cellular device. The settings for the Echo Meter Touch 2 Pro were as follows: the audio division ratio was 1/20, nightly sessions mode was off, save noise files and real-time auto-ID were both on, auto-ID sensitivity was balanced, trigger sensitivity was medium, the trigger window was 5 s, the maximum trigger length was 15 s, the gain was medium, and the sample rate was 256 kHz. We conducted surveys for 5 min at each survey point, similar to Kotowska et al. (2020), beginning between sunset until 40 min post-sunset, and varied the survey order each night to avoid temporal bias in sampling. This shorter sampling window allowed us to gather data at more points per night with limited devices available. All of the surveys took place on paths that ran along edge habitat, which helped to reduce noise interference from any overhead leaves and increased the likelihood of bat activity as supported by research showing a positive correlation with edge habitat (Nelson & Gillam, 2017; Taillie et al., 2021). Path width was recorded at each survey point at sites, using a range finder (Bushnell Yardage Pro Sport 450). Some conditions, like rain or heavy wind, interfered with our ability to accurately capture bat calls, so we chose evenings that minimized those conditions. On survey nights, we recorded the weather conditions and wind speed.

Results

Our study captured calls from five bat species found in Massachusetts and New Hampshire: big brown bats, silver-haired bats, hoary bats, eastern red bats, and little brown bats. In total, we recorded 187 bat calls in Montague Plains and 106 bat calls in Concord Pine Barrens (Table 1) across surveys (08–28 June 2022). Of the calls, the majority were big brown bats (70.7%), followed by hoary bats (11.6%), silver-haired bats (3.1%), eastern red bats (2.7%), and little brown bats (1.4%), and 10.6% of calls remained inconclusive from unknown species. After manually analyzing randomly sampled files with bat species IDs, only 8.7% showed possible indistinction between big brown bat and silver-haired bat calls. Detections were present at each survey point with averages ranging from 0.25 calls to 14.0 calls. The average wind speed on survey nights was 5.7 mph, and the average temperature was 70°F (Kay, Sadlon & Bakermans, 2023).

In our surveys of bat activity, we found significant relationships related to the predictor variables (F_10,15 = 3.21, p = 0.055), where prescribed fire (Chi-sq = 10.9, p = 0.125) and canopy (Chi-sq = 2.8, p = 0.094; Table 2) were significant effects. Specifically, we found a significant decrease in bat activity at survey points with recent prescribed fire (7.4%, SE = 3.1, t = −2.3, p = 0.041) and never had prescribed fire (4.8%, SE = 1.9, t = −2.5, p = 0.030; Table 2) compared to survey points surrounded by distant burns. We recorded the greatest bat activity in the distant burn category (8.4 calls, SE = 1.7) compared to the very recent burn category, which recorded 3.5 (0.3 SE) average calls (Fig. 3).

On average, a survey point that had open canopy had a 2.7% decrease in bat activity compared to a point with closed canopy (Table 2). Across the survey points with an open canopy structure and closed canopy structure, we recorded 120 calls and 173 calls, respectively. Sampling points with closed canopy structures had, on average, 6.7 (1.5 SE) calls recorded while open canopy points had 3.9 (1.0 SE) calls (Fig. 4). Although habitat type was not included in the modeling (to avoid multicollinearity), we recorded 90 calls in hardwood habitat, 45 calls in treated pitch pine habitat, 107 calls in pitch pine habitat, and 51 calls in scrub oak habitat. Mean number of calls across each habitat type varied from 3.1 (0.9 SE) in treated pitch pine to 7.6 (2.5 SE) in pitch pine (Fig. 5). We recorded the highest number of calls at any one survey point in pitch pine habitat at both Montague and Concord study areas (at 9.25 and 14.0 calls, respectively).

Furthermore, model results indicated a non-significant decrease in bat activity for every 1% increase in path width (0.04%, SE = 0.06, t = −0.77, p = 0.457; Table 2). Lastly, logistic regression analyses revealed there were no significant differences in the presence or absence of each bat species by habitat type or structure (all p > 0.0142) except for silver-haired bat where habitat type was significantly related to that species’ presence (Chi-sq = 7.64, df = 3, p = 0.054). Silver-haired bats were detected at 100% (n = 4) of the points in hardwood forests, 50% of the points in scrub oak, and 25% of the points in pitch pine and treated pine.

Discussion

Few studies have examined bat presence and activity at managed pitch pine–scrub oak barrens, such as the sites in Montague Plains and Concord Barrens (Jackson & Schwager, 2012; Gotthardt, Kelley & Schwager, 2015; Taillie et al., 2021). These pine barrens offer a unique habitat that hosts a number of at-risk species, like spreading tick trefoil plant and lupine (Simmons & Hawthorne, 2014), as well as a wide array of Lepidopteran species, such as the Karner blue butterfly (Webb, 2010; Simmons & Hawthorne, 2014; Leuenberger et al., 2016). As a baseline study, our goal was to perform repeat surveys to ensure the audio capture of a range of species present and characterize their activity (Fraser et al., 2020). Any confusion in the auto-ID of these recordings was between silver-haired bats and big brown bats, but after manually checking 25% of our data, the margin of error was deemed to be acceptable (>90% were correct). The abundance of big brown bats is consistent with other literature, such as Nocera et al. (2019). As can be seen by our recordings of little brown bats, which are endangered across New England, the repeat survey method in addition to using stationary surveys with multiple points per location likely aided our observation of this rare species (Braun De Torrez et al., 2017; Deeley et al., 2021). Research shows that little brown bats often select edge and open habitats, which is common throughout managed pine barrens where tree thinnings and prescribed fires take place (Brooks, 2009; Nelson & Gillam, 2017). Additionally, Brooks (2009) found that little brown bats heavily prefer areas over water, something our survey points did not include. Though the use of prescribed fires creates desirable edge habitats for little brown bats, there is evidence to suggest it must be used cautiously as little brown bats avoid heavily burned areas (Jung, 2019).

Overall, our study found that bat activity increased with distant burns and decreased in areas with very recent burns or no burns. These results are at odds with Braun de Torrez, Ober & McCleery (2018) who found that recent burns increased bat activity in Florida, though our study had limited data. Additionally, given the small sample size of little brown bats that were detected and our aggregation of data between species, we cannot make any recommendations about the use of prescribed fire for little brown bat conservation, and we recommend more research be done in this area. We were also able to record each of the four migratory species that are more common (big brown bats, hoary bats, silver-haired bats, and eastern red bats), similar to Gotthardt, Kelley & Schwager (2015), Silvis, Gehrt & Williams (2016), and Austin et al. (2018).

Bat activity was related to forest structure, where activity was increased in closed canopy points (hardwood and pitch pine) compared to open canopy points (scrub oak and treated pitch pine). Divoll et al. (2021) found that the Indiana bat and northern long-eared bat were positively correlated with closed canopy habitats. They tended to roost in open canopy spaces, but had high incidences in forested areas likely for foraging, so their ideal habitat has a combination of both open and closed canopy spaces (Divoll et al., 2021). While different bat species were studied in Divoll et al. (2021), their results still concur with our findings that the studied bat species were detected more in closed canopy areas as they foraged. The bat species observed in each study exhibit similar foraging behavior; however, they tend to differ in their hibernation and roosting habits, aspects not studied in our work (Massachusetts Division of Fisheriesildlife, 2017). In addition to further study on the aforementioned aspects, a study of how bats use different canopy structures throughout the night, similar to Gomes, Appel & Barber (2020), could be an interesting extension to fully examine the impact of the different habits in managed pine barrens on bat activity.

Another study by Taillie et al. (2021) in south Florida showed that bat activity was more closely related to vegetation cover than prescribed fire activity. Smaller bats tended to be more active in closed and cluttered canopy areas than larger bats due to their increased mobility and compact size (Taillie et al., 2021). We also observed in our study that smaller bats (eastern red bats, little brown bats, and silver-haired bats) appeared to be most active in closed canopy habitats. Additionally, the logistic regression analyses we ran with species vs. habitat type showed that only silver-haired bats had a significant relationship to habitat type and were more active in hardwood forest, while a study from Arkansas found more bat activity in pine and mixed pine/hardwood forests (Perry, Saugey & Crump, 2010). Other studies have found smaller bat activity to be higher in more cluttered habitats specifically during foraging, compared to bigger bats, likely due to their size (Patriquin & Barclay, 2003; Sleep & Brigham, 2003). There is also evidence to suggest that bats use different adaptations to overcome competitive exclusion. Species adapted to faster and farther flight may utilize feeding grounds farther away from roosting sites while more agile fliers can outcompete other species in closer areas (Roeleke, Johannsen & Voigt, 2018). Overall, Taillie et al. (2021) found that the most bat activity was recorded in an intermediate canopy cover with greater woody understory. Closed canopy seemed to be too cluttered for bats to perform their hunting maneuvers and open canopy lacked abundant prey and cover from predators (Taillie et al., 2021). Finally, a study conducted by Wordley et al. (2015) showed that bats in Indian tea plantations performed better when forest fragments or shade coffee plants were present.

The management techniques performed at these sites could affect bat presence and activity for several reasons. Prescribed fires could affect the insect community and, therefore, insectivorous bat activity (Leuenberger et al., 2016; Nelson & Gillam, 2017). Insect activity often increases after burns; however, there is little long-term research on the effects of prescribed burns on the nocturnal insect species bats feed upon (Taillie et al., 2021). The fires could also create preferred roosting sites by creating snags and removing unstable trees (Braun de Torrez, Ober & McCleery, 2018). The tree thinning and clearing that takes place at these sites could provide more suitable edge habitat for the bats to forage (Taillie et al., 2021). Our results showed that prescribed fires did not positively affect bat activity, as areas that were burned 10+ years ago or were never burned hosted the most bat activity, while more recent burns (within 10 years) hosted less activity, which is supported by other research (Loeb & Blakey, 2021). However, it is important to note that we cannot definitively state that prescribed fire dampens bat activity as we did not have an even representation of burn categories or a large sample size.

Conclusions

This study demonstrated that restored pine barrens in New England can support an array of bat species. Research performed in managed pine barrens is important because there is currently a strong push to restore New England pine barrens due to their rare yet vital habitat structure and composition (Robertson et al., 2015). Our results are based on acoustic bat call IDs representing an index of bat activity, not the number of bats in the area. It is possible that IDs came from a few bats or the same bat calling multiple times. This is important to note as the results show how bat activity is related to habitat types, not bat population numbers related to habitat type. Stationary recording devices were out of the scope of our project, so we implemented sampling multiple times at the same sites in the research design. Manual audio recording using a smaller, mobile device has its advantages as it provides an opportunity in its accessibility (size, mobility, cost) for citizen science projects, which benefits the field of ecology as it expands the physical range of projects and complements professional field research (Dickinson, Zuckerberg & Bonter, 2010). This methodology also allows for real-time observations of weather and other environmental conditions that may not be made using a stationary device left overnight. Future studies looking at the relationship between bat species, especially little brown bats, and prescribed fires are recommended using similar methodologies with an increased number of surveys and a focus on prescribed fire.