Identifying habitat features that may influence the survival and fitness of threatened species is often constrained by a lack of information about the appropriate scale for habitat conservation efforts. Canada's Species at Risk Act lists little brown bats Myotis lucifugus as Endangered and there is a need to determine the scale for delineating important summer habitat features that should be protected. We used a 19-y dataset of banded little brown bats in a 15,000-km2 area of southern Yukon, Canada, to examine fidelity to roost sites and potential foraging areas. We captured and banded 4,349 bats during 208 live-trapping sessions at maternity roosts. Adult females used multiple roosts during the maternity period, separated by up to 6.1 km, within foraging areas, to which individuals exhibited fidelity. Our fidelity rates (≤ 60.5%) are the lowest, and roost-switching rates (≤ 35.5%) the greatest, reported for little brown bats. A small percentage (14.0–20.7%) of females banded as juveniles returned to their natal roosts or foraging areas as adults. We infrequently observed long-distance (25–200 km) switching to novel foraging areas (< 1% of banded bats). We established bat houses to mitigate the loss of a cabin roost; 46.3% of the bats banded at the cabin occupied these houses. The longest documented period of roost fidelity was 18 y, by a female banded as an adult. Roost fidelity by returning adult females declined annually by 3.8–5.3% due to natural mortality, roost switching, or dispersal. Having a choice of multiple maternity roosts within a foraging area may permit little brown bats to select optimal microclimatic conditions throughout the maternity season. Given that fidelity to foraging areas may be higher than to specific roost sites for little brown bats, identification of summer habitat based on foraging areas may be a more effective conservation strategy than relying solely on roost sites.
The protection of habitat components for threatened species is often a key management action necessary for their recovery and, in some cases, a legal requirement of listing these species (e.g., Hoekstra et al. 2002; Hagan and Hodges 2006). Identifying habitat for threatened species, however, is often constrained by a limited understanding of the habitat features likely to influence survival and fitness, such as nest, den, or roost sites (Heinrichs et al. 2010; Camaclang et al. 2014). Moreover, the scale at which habitat should be identified and protected is uncertain, yet vital to ensure that protection measures are effective (Camaclang et al. 2014; Lemieux Lefebvre et al. 2018).
For several species of bats, research has focused on the importance of roost sites as a key habitat requirement of individuals or colonies (Fenton 1997; Kalcounis-Rüppell et al. 2005). Conservation-based research for bats has often been at the scale of roosting sites, particularly for those species in forest ecosystems (Kalcounis and Hecker 1996; Randall et al. 2014). While some species of bats may be highly selective of roost sites (Kuntz 1982; Kalcounis-Rüppell et al. 2005), the extent to which the availability of suitable sites limits populations or the fidelity of individuals to these sites is less well known. For species with high fidelity to roost sites, conservation of these sites may be vitally important and an appropriate scale on which to base habitat protection measures. However, where roost site fidelity is low, conservation efforts at the roost site scale may be less effective. Therefore, roost site fidelity should be a key consideration in determining the appropriate scale on which to base habitat conservation efforts for bats.
Roost fidelity is highly variable among species of bats, and appears to be greatest during the maternity period (Kuntz 1982; Lewis 1995). Bats that frequently change roosts typically remain faithful to their home ranges, using multiple roosts within them (Kuntz 1982; Lewis 1995). For instance, Indiana bats Myotis sodalis used centralized roosts within home ranges that were hierarchically dependent on social bonds (Silvis et al. 2014). Home range and foraging area appear to be synonymous for the little brown bat, where foraging areas encompass maternity roost sites (Henry et al. 2002; Broders et al. 2006; Coleman et al. 2014; Nelson and Gillam 2017). Roost switching, however, presents different advantages to maternity colonies such as site familiarity, the maintenance of social relationships (Willis and Brigham 2004; Kerth et al. 2006, 2011; Garroway and Broders 2007; Silvis et al. 2014), and selecting roosts with microclimatic conditions that optimize gestation and postnatal growth of pups (Lewis 1995). Roost switching trades some of the benefits of fidelity for other advantages such as reduced commuting costs to foraging areas, lower ectoparasite loads, and avoiding disturbance by humans or predators (Lewis 1995). Adjusting group size provides energetic benefits, as well as costs including parasite transfer and overheating (Olson and Barclay 2013). In big brown bats Eptesicus fuscus, for example, Webber et al. (2016) suggested that greater network aggregations in maternity roosts during pregnancy enhanced pathogen transmission. Despite reports that roost fidelity is a common behavior in some species of bats, researchers have rarely quantified it (Humphrey 1975).
The Committee on the Status of Endangered Wildlife in Canada, following an emergency assessment in 2012, recommended listing the little brown bat Myotis lucifugus as Endangered under Canada's Species at Risk Act due to the impact of white-nose syndrome (COSEWIC 2013). The Committee reexamined and confirmed this status in November 2013 (COSEWIC 2013) and the little brown bat was subsequent listed on Schedule 1 of the Species at Risk Act (2002). White-nose syndrome, caused by the fungus Pseudogymnoascus destructans, affects bats during hibernation in cool, humid caves (Weller et al. 2018). It has caused significant little brown bat population declines in eastern North America (Frick et al. 2010a), and has recently been detected in Washington State (Weller et al. 2018). A national recovery strategy for the species points to the need for field studies to determine the appropriate scale for delineating critical habitat for maternity roosts of remnant and recovering populations (Environment and Climate Change Canada 2018).
Adult female little brown bats exhibit relatively high annual fidelity and natal philopatry to maternity roosts (Humphrey and Cope 1976; Reynolds 1999; Frick et al. 2010b). The longest reported use of a roosting site by a little brown bat was at least 23 y by an adult female (Florko et al. 2017). These roosts rarely contain adult males, which roost apart from females (Fenton and Barclay 1980; Randall et al. 2014). In the American Midwest, little brown bat maternity colonies may be distinct social units since researchers recaptured only 1.3% of females away from the maternity colonies where they were originally banded (Humphrey and Cope 1976). Conversely, in Saskatchewan little brown bats that switched roosts likely remained faithful to particular foraging areas (Kalcounis and Hecker 1996). Perry (2011) has correlated long-term roost fidelity with foraging area fidelity of several bat species (but not little brown bats) in Arkansas. Researchers have not reported roost switching beyond the home range or foraging area for female little brown bats. Cavity-roosting little brown bats exhibit a fission–fusion pattern of roosting behavior, involving roost switching by adult females and pups among maternity groups occupying roost networks (Olson and Barclay 2013). Moreover, long-term site fidelity may become more or less important for little brown bats under a changing climate that may result in reductions of suitable trees as roost sites (e.g., Randall et al. 2011; Jung 2020). Climate change may affect foraging success (i.e., insect abundance and activity) and roosting site characteristics (i.e., thermal conditions), through changes in temperature or precipitation that affect the selection of optimal sites (Sherwin et al. 2012).
The occupancy of replacement maternity roosts (e.g., bat houses) and fidelity to foraging areas by little brown bat maternity colonies excluded by humans—often as a pest control measure—is also reportedly uncommon (Humphrey and Cope 1976; Neilson and Fenton 1994). For instance, after exclusion by humans, Humphrey and Cope (1976) found no evidence of successful or attempted colony reestablishment at a new roost after exclusion. Less than 10% of evicted bats occupied other nearby roosts or replacement bat houses (Neilson and Fenton 1994). Conversely, Brittingham and Williams (2000) found that little brown bats occupied bat houses that provided suitable microclimates and were placed nearby. Wildlife managers need further data on the use of bat houses by evicted little brown bat maternity colonies to help assess their efficacy as a mitigation measure.
In this study, we used a 19-y mark–recapture dataset to examine fidelity and movements of banded little brown bats from 13 maternity roosts across southern Yukon, Canada, to determine rates of fidelity and switching within and between distinct foraging areas. We also report foraging area fidelity by a maternity colony of little brown bats that were evicted from a building. A lack of longitudinal data from marked bats, and the number and distribution of maternity roosts within and among foraging areas, have limited previous studies of roost fidelity and movements by female little brown bats. Our intent is that these analyses may be useful for developing a better understanding of site fidelity by little brown bats, and to aid in identifying the appropriate scale for identifying summer habitat for remnant or recovering populations.
We captured little brown bats at 13 maternity roosts in a 15,000-km2 area of southern Yukon between 1997 and 2015 (Table 1; Figure 1). We directed our efforts at roosts in or near the City of Whitehorse (60°42′N, −135°6′W), where we monitored eight roosts. Five roosts were within two presumed foraging areas, including the Chadburn Lake foraging area (three roosts separated by 3.0–6.1 km; Figure 1) and the Squanga Lake foraging area (two roosts separated by 2.1. km). Distances of 27–272 km separated the other maternity roosts. These roosts and maternity colonies were in rock crevices, occupied and unoccupied buildings, or bat houses (Slough and Jung 2008; Table 1).
We used harp traps to capture bats during 1–8 sessions each year (Palmeirim and Rodriques 1993; Figure 2). We conducted capture–mark–recapture sessions annually between late April and early September, with most occurring during the maternity season in late May through early August (Slough and Jung 2008). We distinguished juvenile bats from adults by unossified epiphyses of the metacarpal–phalangeal joints (Anthony 1988). Ossification of cartilage within the joints was typically complete by mid-August, but juveniles retained dark, glossy fur and waxy-appearing membranes. We attached a unique alphanumeric aluminum band (4.0 × 0.38 mm, 2.9 mm internal diameter, Lambournes Ltd., Birmingham, UK, or Porzana Ltd., Icklesham, UK; or 5.0 × 0.64 mm, 3.0 mm internal diameter, Gey Band and Tag Co. Inc., Norristown, PA) to the forearm of bats to mark individuals.
In 2007, a maternity roost in an abandoned cabin that we had monitored for 7 y (2001–2007; Jung and Slough 2005) was demolished. We replaced it with two large pole-mounted bat houses at the site in mid-May 2008 (Figure 2). Bats typically began inhabiting the cabin roost in mid-April prior to its destruction. We erected an additional pole-mounted bat house 2.1 km away in June 2008 to provide an alternate roost for bats inhabiting a campground picnic shelter. We monitored the response of the maternity colony to the loss of the cabin roost and to the replacement bat houses through recaptures of banded individuals at the bat houses in subsequent years.
We calculated fidelity on a cumulative basis, where we assumed recaptured individuals to have been present in all years between initial and final capture, if we did not recapture them at other roosts. This method did not account for roost switching between recapture events; therefore, it may overestimate roost fidelity more so than foraging area fidelity. We defined foraging area fidelity as the recurrent movement of individuals between roosts that were proximate (separated by ≤ 6.1 km in our study) to each other, and where banded individuals moved between roosts within the apparent foraging area, both within and among years. Long-distance switching was rare, and occurred between roosts separated by 27 km or more. We obtained geographic coordinates for each site, and used Google Earth Pro 220.127.116.1176 (accessed March 20, 2019) to measure the distances between roost pairs.
We conducted 208 trapping sessions at 13 little brown bat maternity roosts between 1997 and 2015 (Table 1; Data S1, Supplemental Material). We monitored individual roosts for 1–19 y (median = 9) with 1–53 trapping sessions each (median = 11; Table 1). We focused most of our sampling effort (163 of 208 trapping sessions; 78.4%) at maternity roosts within two putative foraging areas: the Chadburn Lake and Squanga Lake foraging areas. We monitored roosts in the Chadburn Lake foraging area (n = 3) for 8–18 y with 15–53 trapping sessions, and the two Squanga Lake foraging area roosts for 8 and 15 y with 21 and 39 trapping sessions, respectively.
We captured and banded 4,349 individual little brown bats. Most were adult females (81.9%), 8.6% were juvenile females, and 6.7% were juvenile males (Table 1). We rarely captured adult males at maternity colonies (0.87%, n = 38). We recaptured 1,274 of the 4,349 marked individuals (29.3%) at least once. In total, there were 2,568 recaptures of 1,226 adult females, with individuals being recaptured up to nine times (Figure 3).
Maternity roost site and foraging area fidelity
We calculated roost and foraging area fidelity for the Chadburn Lake foraging area (three roosts; 568 adult females banded), the Squanga Lake foraging area (two roosts; 1,467 adult females banded), and the Little Atlin Lake roost (952 adult females banded; Table 2). Foraging area fidelity for adult females the year after initial capture was 34.9% at the Chadburn Lake area and 48.8% at Squanga Lake area (Table 2). Bats began to occupy replacement bat houses in the Squanga Lake foraging area in the first year after eviction; in small numbers by July 8, 2008 (eight bats observed), and were resident in large numbers (> 100) by July 22, 2008. We recaptured at least 124 of 268 (46.3%) adult females banded at the previous cabin roost at the replacement bat house roosts within 7 y of the cabin being demolished, demonstrating foraging area fidelity and eventual group reestablishment. Roost fidelity was 18.8% at Little Atlin Lake. These rates declined annually by means of 3.7% (Chadburn), 5.3% (Squanga), and 3.5% (Little Atlin) at these foraging areas (Figure 3). The longest observed fidelity was 18 y by a female banded as an adult at the Chadburn Lake roost. Nine other females demonstrated roost fidelity for 10 to 18 y.
We observed a small percentage (14.0–20.7%) of females banded as juveniles returning to their natal foraging areas as adults the next year (Table 2). The fate of the rest of the banded juvenile females was unknown. Interestingly, at least 4 of 294 (1.4%) juvenile males returned to their natal roosts for up to 4 y and we recaptured 1 of 38 (2.6%) males banded as an adult at the same roost. Three of these males returned to their natal roosts only after the maternity period, from mid-August onward, when the maternity colonies had largely dispersed for the season.
During the maternity period, we observed roost switching during multiple live-capture sessions. Adult females frequently switched roosts separated by 2.1 km (35.5% of banded bats), 3.0 km (20.9%), and 4.4 km (11.8%; Table 3). Two adult females (1.7%) switched between roosts separated by 6.1 km. Of the 540 observed roost switches, approximately 37.8% of them occurred within the same summer season. We captured three females carrying pups in their uropatagia, indicating that they were switching roosts prior to their pups becoming volant.
Few adult female little brown bats (n = 20; < 1% of banded bats) dispersed between 27 and 200 km to new roost sites within different foraging areas during different years (Table 3). A female moving 27 km returned to the original roost. We observed nine long-distance roost switches between roosts separated by 75.1–77.2 km, and two of these females made return movements.
Evaluating fidelity of bat species to individual maternity roost sites can help inform the scale at which wildlife managers should identify and protect summer habitat. However, doing so requires large datasets of roost use by known individuals. The large number of little brown bats that we banded at multiple study sessions across a long temporal scale and large spatial scale allowed us to estimate rates of roost fidelity, switching within foraging areas, and long-distance switching. Our study provides the lowest fidelity rates and the highest roost-switching rates reported for little brown bats in the literature. Foraging area fidelity rates are likely higher than we reported, since we did not know all roosts within foraging areas, some individuals may have evaded recapture, and some band loss may have occurred. The lower fidelity rate of 18.8% at Little Atlin Lake is most likely attributable to banded individuals occupying additional unknown roosts within the foraging area. This supports our hypothesis that little brown bats in our study area were more faithful to foraging areas than to specific maternity roosts.
Our findings for little brown bats in northwestern Canada varied from trends reported from central or eastern North America, suggesting geographical variability. In Manitoba and northwestern Ontario, Norquay et al. (2013) reported a 12% (n = 41) switching rate among maternity roosts separated by 25 to 464 km. This is much greater than the 0.4% relocation rate observed by Humphrey and Cope (1976) in Indiana and Kentucky. In New York, Neilson and Fenton (1994) recaptured only 3 of 5,647 banded bats in colonies up to 70 km from the colony of original capture. Dixon (2011) inferred a high natal dispersal rate of females between maternity colonies in Minnesota, with some degree of natal philopatry, using population genetics. Norquay et al. (2013) speculated that unpredictable or unfavorable weather patterns near the northern limit of the little brown bats' distribution was associated with the higher rate of relocation that they observed. Little brown bats in our study area are also near the northern limit of their range, where roost sites may be a limiting factor (Thomas and Jung 2019), and they may select areas that offer suitable roosting opportunities, as well as good foraging habitat. In this light, selection of suitable foraging areas would include a number of potential roosting sites within close proximity (e.g., < 6 km; Randall et al. 2014) that bats can use when needed.
Researchers have reported mean summer home ranges and foraging areas of female little brown bats of between 30 and 994 ha in an agricultural landscape (Henry et al. 2002; Broders et al. 2006; Bergeson et al. 2013; Coleman et al. 2014). Henry et al. (2002) reported foraging distances within these ranges to be 2.6 ± 0.6 km from the roost during pregnancy to 1.7 ± 0.6 during lactation. Hypothetical roost switching within the radius of these foraging distances (i.e., ≤ 5.2 km) would permit females to continue foraging in the same area. Little brown bats occupying roosts up to 6.1 km apart in our study were using the same foraging area, based on both our recaptures between roost sites, and radiotelemetry conducted on bats in our study area (Randall et al. 2014). In southwestern Yukon, Randall et al. (2014) found nightly movements of radio-tagged adult female little brown bats of up to 6 km from their roost. Longer-distance movements between roosts would likely place bats in new foraging areas. Other researchers have reported that roost switching occurs to nearby areas (Lewis 1995; Kalcounis and Hecker 1996; Broders and Forbes 2004).
The maternity period for little brown bats at high latitudes (e.g., ≥ 60°N) is brief and temperatures can be low (Slough and Jung 2008). The maternity season lasts from about mid-May to early September in southern Yukon with an average daily high temperature above 13°C. The daily maximum temperature was 26°C in July, with an average maximum of 20°C and an average minimum of 8°C (Environment and Climate Change Canada 2019). How the unique environmental constraints placed on little brown bats at high latitudes affects their behavior and life history is largely unknown (Jung et al. 2014). Selecting roosts with optimal microclimatic conditions is likely of particular importance to little brown bats at high latitudes, and may have contributed to the high level of roost switching observed in this study. Female little brown bats likely select roost sites that provide thermal conditions that allow them to avoid torpor and its negative effects on pregnancy, lactation, and juvenile development. Dzal and Brigham (2013) have shown pregnant little brown bats used torpor less than lactating and postlactating bats. Olson and Barclay (2013) found larger groups of tree-roosting little brown bats near the start of parturition, when pups would benefit from thermoregulation. Big brown bats have a greater social cohesion during pregnancy than during lactation, possibly due to lower ambient temperatures during gestation (Webber et al. 2016). Social cohesion to minimize energetic costs could explain roost switching (and replacement roost occupancy) by little brown bats.
After exclusion, Humphrey and Cope (1976) found no evidence of successful or even attempted group establishment at a replacement roost and Neilson and Fenton (1994) recaptured less than 1% of bats banded in buildings at other roosts following eviction. After exclusion, a colony reoccupied a building, and included 0.1–2% of bats from four other nearby colonies. None of 43 bat houses installed around the area of eviction in their study attracted populations of roosting bats, nor did the excluded colonies move to existing maternity colonies. The evicted maternity colony in our study reoccupied bat houses within a foraging area largely as a cohesive group. Little brown bats are known to use bat houses, particularly if conditions such as temperature, predation risk, and accessibility are suitable (White 2004) and if they are replacements for bats excluded from buildings and are placed near the former roost (Brittingham and Williams 2000; White 2004).
In conclusion, our study provides data on fidelity to roost sites and foraging areas by little brown bats in high-latitude boreal forests, based on an extensive dataset of marked and recaptured individuals. Although not previously reported for this species, our findings are consistent with species of tree-cavity–roosting bats that exhibit a fission–fusion pattern of behavior, which involves regular roost switching by adult females and pups (Olson and Barclay 2013). Given that fidelity to foraging areas may be higher than fidelity to specific roost sites for little brown bats, identification of summer habitat based on foraging areas may be a more effective conservation strategy than relying solely on maternity roost sites. That is, little brown bat conservation in the northwestern boreal forest is contingent on management at the scale of the foraging area, which includes the provision of multiple roosts for maternity colonies. Roosts capable of supporting larger groups of bats may be more important in northern latitudes where the maternity season is brief and social thermoregulation is more important (Olson and Barclay 2013). Further research aimed at improving our knowledge of the summer habitat requirements of little brown bats should focus on better understanding habitat composition, as well as the number and characteristics of maternity roost sites within their foraging areas. However, it is also important to consider that climate change may significantly alter habitats available to little brown bats in the boreal forest (Randall et al. 2011; Jung et al. 2014; Thomas et al. 2019; Jung 2020). As such, there will likely be a need to ensure that a diversity of foraging areas and roost sites are available for bats in the future.
Please note: The Journal of Fish and Wildlife Management is not responsible for the content or functionality of any supplemental material. Queries should be directed to the corresponding author.
Data S1. Little brown bat Myotis lucifugus capture–mark–recapture data from Yukon, Canada, 1997–2015. Recapture: Y = yes, the bat was captured previously; N = no, initial capture. Band ID: unique prefix/number on each band. Location: capture–mark–recapture location.
Found at DOI: http://doi.org/10.3996/052019-JFWM-039.S1 (1.03 MB XLSX).
We thank the many field assistants that helped capture and band bats over the years. We especially thank P.M. Kukka for organizing many of the banding sessions and for capture data management. Financial support provided by the Government of Yukon, the Government of Canada (Habitat Stewardship Program), and the Northern Research Endowment Fund Grants from the Yukon Research Centre, Yukon College. We kindly thank three anonymous reviewers and the Associate Editor of the journal for comments that improved this manuscript.
Any use of trade, product, website, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.
Citation: Slough BG, Jung TS. 2020. Little brown bats utilize multiple maternity roosts within foraging areas: implications for identifying summer habitat. Journal of Fish and Wildlife Management 11(1):311–320; e1944-687X. https://doi.org/10.3996/052019-JFWM-039
The findings and conclusions in this article are those of the author(s) and do not necessarily represent the views of the U.S. Fish and Wildlife Service.