We aimed to assess the importance of anthropogenic roosts for bats hibernating in the Roztocze National Park (south-east Poland), based on data collected from 2009 to 2021. We recorded 310 bats from nine species hibernating in 27 artificial underground roosts (root cellars and basements of buildings). The most abundant and constantly recorded species were Plecotus auritus (60.2%), Barbastella barbastellus (20.6%) and Myotis nattereri (14.2%), while the remaining species can be considered of secondary importance; Plecotus austriacus (1.3%), Myotis myotis (1.9%), Myotis bechsteinii (1.3%), M. daubentonii (0.3%), Myotis mystacinus sensu lato (0.3%) and Eptesicus serotinus (1.3%). An estimate of the Shannon diversity index gave a mean H = 0.947 (SD = 0.247, range 0.377-1.352), while the Buzas and Gibson´s evenness index gave values of E = 0.695 (SD = 0.125, range 0.551-0.940). The Shannon index was positively correlated with the number of recorded bats and varied substantially among years.
Introduction
Insectivorous bats have high metabolic rates while euthermic and, therefore, have substantial nutritional requirements (Kunz & Fenton 2003). This feature of their biology translates directly into the consumption of large quantities of insects, including pest species that cause significant losses in agriculture and forestry (Lacki et al. 2007, Puig-Montserrat et al. 2015), and they thereby provide significant ecosystem services (Boyles et al. 2011). Bats have relatively low reproduction rates (Barclay et al. 2004), thus have low rates of increase (Voigt & Kingston 2016). Unfortunately, in modern human-dominated landscapes they are endangered by disease, development of harmful human infrastructure and direct mortality, as well as loss and degradation of natural habitats (Frick et al. 2019).
The majority of bat species are primarily associated with forests (Kunz & Fenton 2003, Lacki et al. 2007). In European primeval temperate forests, bats mainly use hollows and spaces under protruding bark as roosts (Ruczyński & Bogdanowicz 2008, Gottfried et al. 2015, Dietz et al. 2018). A decrease in forest cover, increase in forest fragmentation, changes in tree species composition, simplification of stand age structure and finally, the scarcity of dead trees caused by forest management, has decreased the availability of natural roosts for bats (Jung et al. 2012, Russo et al. 2016, Treitler et al. 2016). Consequently, many bat species inhabiting forests utilize a variety of human-made structures as alternative roosts (Mazurska & Ruczyński 2008, Ruczyński et al. 2010, Kurek et al. 2020) and anthropogenic structures can, in some circumstances, be superior than natural roosts (Johnson et al. 2019).
In temperate latitudes winter is a critical season for bats, when lack of food and low temperatures force either a seasonal migration or a search for hibernation roosts that can ensure a stable temperature and high humidity. Bats hibernate in diverse natural shelters, both above ground (e.g. hollow trees) and below ground (e.g. animal burrows, caves). However, they also use various artificial structures, such as tunnels, fortifications, basements of buildings, storage cellars, sewage systems and wells (Kunz & Fenton 2003).
In this paper we assessed the importance of anthropogenic roosts for bats hibernating in Roztocze National Park (south-east Poland), based on data collected over 13 years. Combining this information with evidence from other parts of Europe (Lesiński et al. 2004, Lesiński 2009, Vintulis & Pētersons 2014), we hypothesise that artificial roosts serve as important sites for bat hibernation in this large forest tract.
Material and Methods
Study area
Roztocze National Park (50° 36′ 11.9″ N, 23° 01′ 31.0″ E; 84.8 km2), established in 1974, is situated in the southern Polish uplands. It protects fragments of mainly natural deciduous and mixed forests dominated by stands of Scots pine Pinus sylvestris (35%), European beech Fagus sylvatica (22%) and silver fir Abies alba (16%) (Tittenbrun & Radliński 2019). The climate is a transition between Atlantic and Continental with a mean temperature in January of –2.4 °C, a mean temperature in June of 19 °C, and an annual precipitation ca. 730 mm.
The park is inhabited by at least 16 bat species (Jurczyszyn 1994, Kasprzyk et al. 2004, Bashta et al. 2011, Stolarz & Lesiński 2017, Stachyra et al. 2019), all of which are strictly protected in Poland. Additionally, the national park is also protected as a Natura 2000 site “Roztocze Środkowe” (PLH060017) dedicated to the protection of four bat species: Barbastella barbastellus, Myotis myotis, Myotis bechsteinii and Myotis dasycneme (Standard Data Form 2021).
Table 1.
Characteristics of bat assemblages wintering in Roztocze National Park, 2009-2021.
Table 2.
Species composition of bat assemblages wintering in Roztocze National Park, 2009-2021.
Data collection
We conducted bat monitoring in both root crop cellars and basements of buildings in Roztocze National Park from 2009 to 2021. Field work was conducted mainly during February each year. The number of structures checked for bats varied from 15 to 27 per year (mean = 24, SD = 3.2), depending on the identification of new roosts and their accessibility during winter (Table 1). Species identification was based on external features (Dietz et al. 2009) and as identification of cryptic species from the Myotis mystacinus group (M. mystacinus, M. brandtii, M. alcathoe) during hibernation may lead to some doubts (von Helversen et al. 2001, Bashta et al. 2011), we counted all individuals of these species together and named them Myotis mystacinus sensu lato.
Data analysis and statistics
We described assemblages of bats hibernating in Roztocze National Park using the following parameters:
(1) Dominance D% according to the formula:
where: ni = number of individuals from species i, N = number of individuals from all species.
(2) Constancy C% according to the formula:
where: qi = number of years when species i was recorded, Q = number of years when monitoring was conducted.
(3) Shannon diversity index (H) according to the formula:
where: ni = number of individuals from species i; n = number of individuals from all species (Shannon 1948).
(4) Buzas and Gibson´s Evenness index (E) according to the formula:
where: H = Shannon diversity index; e = base of the natural logarithm; S = number of species (Buzas & Gibson 1969).
We applied the Whittaker index (Whittaker 1960, Wilson & Shmida 1984) to assess differences between diversity indexes among years, following the formula:
where: S = number of species recorded during our study, 
= mean number of species in given year.
Diversity indices were calculated in the Past 3.0 software (Hammer et al. 2001), while correlations were calculated in Statistica 13.3 (TIBCO Software, USA).
Results
We recorded 310 bats from nine species hibernating in artificial underground roosts in Roztocze National Park from 2009 to 2021. On average, we recorded 24 individuals (SD = 12.4, range 8-49) from four species (SD = 1.1, range 2-6) each year (Tables 1 and 2). Neither the number of individuals nor the number of species was correlated with the number of shelters examined in a given year. The number of recorded species was, however, correlated with number of individuals (y = 0.077x + 2.097, R2 = 0.78, P < 0.001).
Table 3.
Differences between species diversity indices of winter assemblages of bats in Roztocze National Park, 2009-2021, characterised by the Whittaker index.
Among bats recorded in winter roosts, Plecotus auritus was by far the most common (D = 60.2%) and was recorded during all years (C = 100%). A substantial proportion of the bat assemblages also contained B. barbastellus (D = 20.6%) and Myotis nattereri (D = 14.2%), which were recorded almost every year (C = 91.7% for both species). The remaining species can be considered of secondary importance, both because of their small contribution to the assemblages and their low constancy – Plecotus austriacus (D = 1.3%, C = 15.4%), M. myotis (D = 1.9%, C = 38.5%), M. bechsteinii (D = 1.3%, C = 15.4%), Myotis daubentonii (D = 0.3%, C = 7.7%), M. mystacinus sensu lato (D = 0.3%, C = 7.7) and Eptesicus serotinus (D = 1.3, C = 23.1%) (Table 2).
An estimate of the Shannon diversity index gave a mean of H = 0.947 (SD = 0.247, range 0.377-1.352), while the Buzas and Gibson´s Evenness index gave values of E = 0.695 (SD = 0.125, range 0.551-0.940). The Shannon diversity index was positively correlated (y = 0.015x + 0.596, R2 = 0.54, P < 0.001) with the number of bats recorded in winter roosts (Table 1), but it did not correlate with evenness indices. Diversity indices were not associated with either the mean temperature in February or the number of days with temperatures below zero in February. Pairwise comparison of diversity indices varied greatly among years from βw = 0 to βw = 0.600, though on average this parameter was moderate βw = 0.266 (SD = 0.132) (Table 3).
Discussion
In line with our expectations, small anthropogenic roosts, such as storage cellars and basements, were used for hibernation by several species of bats wintering in the Roztocze National Park. The most numerous species recorded wintering in the park was P. auritus, which is also one of the most common bats in Poland (Sachanowicz et al. 2006). Furthermore, this species is the most abundant bat hibernating in cellars and basements in northern and western Poland (Lesiński et al. 2004, Lesiński 2009) and Latvia (Vintulis & Pētersons 2014).
Among bat species hibernating in anthropogenic roosts in Roztocze National Park there are three listed as conservation targets for the Natura 2000 site “Roztocze Środkowe”, i.e. B. barbastellus, M. myotis and M. bechsteinii (Standard Data Form 2021). Moreover, the abundance of B. barbastellus in cellars and basements in the national park was higher than in similar winter roosts in other parts of Poland (Lesiński et al. 2004, Lesiński 2009), underlining the importance of such shelters for local populations of this species.
Barbastella barbastellus prefers to hibernate at lower temperatures than M. myotis and M. bechsteinii (Bogdanowicz & Urbańczyk 1983, Webb et al. 1996). Recent studies have indicated that a warming climate will likely hamper the use of well-insulated winter roosts by species preferring colder conditions, such as B. barbastellus (Gottfried et al. 2020). Thus conservation action regarding management of underground bat roosts (Mitchell-Jones et al. 2007) should take into consideration environmental changes (Rydell et al. 2018, 2020). De Bruyn et al. (2021) predicted that under climate change, when winter conditions are expected to be milder, bats such as B. barbastellus may begin hibernating in trees, therefore the availability of old-growth forests will be essential for their long-term survival. Roztocze National Park offers patches of old beech forests, mainly in the strict protection zone, so the reduction of old tree extraction in the remaining zones, where active forest management is allowed, would seem to be crucial for bat conservation in the future.
Our data confirmed earlier observations that protection of anthropogenic roosts is important for bat conservation in forest tracts, including protected woodlands (Lacki et al. 2007). The composition of bat assemblages hibernating in anthropogenic roosts varied considerably between years. Therefore, we suggest that conservation action should be based on long-term data on the number and species composition of hibernating bats rather than single surveys of roosts (Mitchell-Jones et al. 2007).
Acknowledgements
We are grateful to Paweł Marczakowski, Bogusław Radliński and Andrzej Tittenbrun from the Roztocze National Park for the assistance. We are grateful to Dr. Chris Young for linguistic advice, and an anonymous reviewer for comments that improved our manuscript. Author contributions: P. Stachyra – conceptualization, methodology, investigation, writing original draft; M. Piskorski – investigation; M. Tchórzewski – investigation; K. Łopuszańska-Stachyra – investigation; R.W. Mysłajek – conceptualization, methodology, investigation, formal analysis, writing original draft, review and editing.
