Recently the Maxent approach has been widely accepted for suitable habitat mapping and SDMs for several species of plants (Abolmaali et al., 2018; Dai et al., 2022; Kaky et al., 2020), reptiles (Fourcade et al., 2014; Pearson et al., 2006), birds (Baici and Bowman, 2023; Coxen et al., 2017; Hou et al., 2022; Moreno et al., 2011; Zhang et al., 2019) and mammals ((Baral et al., 2023; Deb et al., 2018; Escalante et al., 2013; Evangelista et al., 2018; Kalle et al., 213; Morán-Ordóñez et al., 2018; Zahoor et al., 2021). It has also been used to model bat species richness and spatial distribution (Gonçalves et al., 2021; Kangoyé et al., 2021) and response to changing climate (Ancillotto et al., 2018; Delgado-Jaramillo et al., 2020; Festa et al., 2023; Gonçalves et al., 2021; Smeraldo et al., 2021; True et al., 2021; Tzortzakaki et al., 2019). Recently, the impact of climate change and extreme temperature events on the future distribution of Flying Foxes in Australia has been predicted (Diengdoh et al., 2022). Despite its acknowledged limitations (Halvorsen, 2013; Merckx et al., 2011; Morales et al., 2017; Yackulic et al., 2013), Maxent is effective for presence-only data and with small sample sizes (Kaky and Gilbert, 2016; Pearson et al., 2006; Wisz et al., 2008). Also, it has been shown to produce distribution predictions of comparable accuracy to Ensemble methods (Kaky et al., 2020) and is favored for its higher efficiency over other approaches (Ahmad et al., 2020; Araújo and New, 2007; Kindt, 2018; Kaky et al., 2020; Meller et al., 2014; Ng et al., 2018). Herein we performed Maximum Entropy for the modeling of the roosts of the Indian Flying Fox achieving high predicted accuracy with an AUC of 0.933 and TSS of 0.764 ± 0.041.
Although protected areas cover about 23.39% (34419.75 km2) of Nepal’s total land (MoFSC, 2016), surprisingly the occurrence of roosting colonies of P. medius have been recorded in only two out of 20 PAs (Fig. 2). Several roosting colonies are situated in close proximity to seven PAs including Shivapuri Nagarjun National Park (SNNP), Krishnashar Conservation Area (KrCA), Parsa National Park (PNP), Annapurna Conservation Area (ACA), Chitwan National Park (CNP), Banke National Park (BaNP) and Bardia National Park (BNP). The species’ roosting colonies are notably absent from the Middle Mountain in eastern and western Nepal whereas few colonies have been reported from the Middle Mountain in the central Nepal. The presence of large valleys with similar climate in the Middle Mountain of central Nepal, which are absent in the eastern and western Middle Mountain of the country may explain such a distribution pattern. Future ecological studies should focus on this geographical discrepancy further.
A species’ habitat niche is influenced by numerous factors(Sherwin et al., 2012) with plant-animal interactions being critical for species distribution across large spatial scales. For instance, the habitat suitability of fruit bats in southern Africa depends upon distribution of 49 fig species (Ficus spp.) (Arumoogum et al., 2019). On the other hand, microclimatic variations in temperature and humidity significantly affect P. poliocephalus colonies in Australia (Snoyman and Brown, 2011). Response of bat species to global climate change are inconsistent, reflecting differences in ecological traits such as habitat preferences, foraging behaviors and reproduction strategies (Ciechanowski et al., 2007). Our predictions for the roost’s distribution of the Indian Flying Fox differ from our previous projection for two fruit bats, C. sphinx and R. leschenaultii which shows expansion and contraction, respectively (Thapa et al., 2021). This difference in distribution range is attributed to the contributing predictors: Annual Mean Temperature (bio 1) Annual Precipitation (bio 12) and Isothermality (bio 3) being the primary contributor for the latter two species. However, bio 1 (43.6%) and bio 12 (37.6%) contributes mainly for the two different species (Thapa et al., 2021). In contrast for P. medius, urban land (35.8%), water sources (27.2) and Mean Annual Temperature (bio 1) (20.7%) are the most influencing factors (Table 2). For the distribution of flying foxes in Australia, bioclimatic variables are expected to impact differently among species. For example, Precipitation of Wettest Month (bio 13) positively affects the Grey-headed Flying Fox P. poliocephalus and the Black Flying Fox P. alecto, while Precipitation of Coldest Quarter (bio 19) positively affects P. poliocephalus. Mean Temperature of Wettest Quarter (bio 8), Mean Temperature of Driest Quarter (bio 9) and Precipitation of Warmest Quarter (bio 18) positively affect P. alecto but negatively impact P. poliocephalus. Similarly, Mean Temperature of Warmest Quarter (bio 10) and bio 18 influence the distribution of the Spectacled Flying Fox P. conspicillatus in contrasting ways (Diengdoh et al., 2022). Increase in mean temperature and fluctuations in rainfall pattern are the primary drivers delineating bat distribution (Festa et al., 2023). Temperature and precipitation fluctuations influence species behaviors (foraging), physiology and interactions (phenology of plants) with extreme temperatures posing risks to flying foxes (Dey et al., 2015; O’Donnell and Ignizio, 2012; Roy et al., 2020; Sherwin et al., 2012; Welbergen et al., 2008). While heat waves have been reported globally, including in Nepal, the annual mean temperature in future climates is predicted to remain within the species' tolerance range. As different species will respond differently to global warming due to their different physiological tolerances and ecological niches (Ancillotto et al., 2018; Diengdoh et al., 2022; Smeraldo et al., 2021; Tzortzakaki et al., 2019; Welbergen et al., 2008), it is essential to further quantify their physiological tolerances, ecological niches, and dietary preferences to predict their responses to climate change more accurately. Incorporating foraging site locations into the future habitat suitability assessments for the species is recommended. Therefore, regular monitoring of roosting and foraging sites in the future is necessary. The range expansion towards the northern latitude and high elevation are observed in several bats species (LaVal, 2004; Sherwin et al., 2012) in the US and Canada (Humphries et al., 2002), Europe(Lundy et al., 2010; Rebelo et al., 2010; Sachanowicz and Ciechanowski, 2006) and South East Asia(Hughes et al., 2012) and recently in Nepal (Thapa et al., 2021). However, our study projects insignificant distribution range shifts towards the higher elevations and northern latitude for P. medius in future climate scenarios (Fig. 5).
Flying foxes have been documented roosting in a diverse array of plant species (30 species) throughout Nepal (Katuwal et al., 2019; Manandhar et al., 2018; Neupane et al., 2016; Prajapati et al., 2020; Sharma et al., 2018). Over 57% of the known roosting colonies are located in agricultural land (Fig. 4). However, agricultural land contributing only 0.6% was not identified as an influencing predictor for the roost distribution suitability in the current study. Approximately 32 % of the known roosting colonies are situated within or in the close proximity of urban and suburban areas. Urbaization is known to affect bat communities resulting in low diversity and domination by synurbic species (Tzortzakaki et al., 2019). On the other hand, neither urbanization(Ancillotto et al., 2018) nor vegetation cover(Tzortzakaki et al., 2019) impacts synurbic and non-synurbic bats’ distribution and activity. Our observations also indicate that P. medius is not affected either by urbanization or vegetation change. The species has been observed to roost in urban areas, exploiting suburban orchards, gardens and farmlands as well as distant vegetation and forests and farmlands for foraging. On the other hand, the species is known to select various trees for roosting and has been observed abandoning the roosts and migrating to new ones, suggesting that roost collapse and loss of plant diversity may not have detrimental effect on the species. However, continuous loss of winter habitats has been linked to population decline in P. alecto (Baranowski and Bharti, 2023). Colony size in south India has been correlated with the diameter at breast height (dbh) and canopy size of the roosting tree (M. Pandian and S. Suresh, 2021). Similarly, distribution of fruit bats is influenced by the availability of figs in Africa(Arumoogum et al., 2019) and eucalyptus dominated vegetation in Australia (MacDonald et al., 2021). The specific relationship between particular roosting trees and the roosting ecology of the bat in Nepal warrants further documentation.
Fruit bats, including flying foxes, are capable of thermoregulation, allowing them to maintain body temperatures despite fluctuations in air temperatures (Humphries, 2009). Temperature and humidity fluctuations in Nepal have not been as extreme as those recorded from Australia and India (Dey et al., 2015; Welbergen et al., 2008; MacDonald et al., 2021). Our predictions indicate a slight contraction (less than 1%) in the species' future climate distribution range compared to the current potential range, suggesting that while human disturbances, roost collapse, hunting and persecution may affect the species on a site-specific basis, the overall impact of climate change is likely to be insignificant.
Approximately, 32% of the total known roosts are located in the built-up areas including urban and sub-urban areas. Since the species is known to be reservoir for several viruses, regular surveillance and screening from a One Health perspective are essential, particularly in the context of pandemics. The data and distribution predictions from this study can inform hotspots. We have observed bats occupying new roosts and new areas within settlements and in agricultural landscapes, while also observing reduction of roosts and roosting colonies in built up areas such as downtown Kathmandu. Urban land and water sources are the two most influential predictors for the distribution of the Flying Fox’s roosts in this study. Habitat suitability for the species decreases with the increase in urban land and with decrease in water sources. The lack of foraging areas and water sources within the urban areas could explain the decline of flying foxes. Therefore, plantation of tall and broad canopy native broad-leaved trees with in and around the urban areas and conservation of water sources including through rain water harvesting could help to maintain the species in its current roosting sites. This will also be an early preparedness for protecting flying foxes for the extreme temperatures in the future. Recently, water sprinkler has been used to cool flying foxes during the extreme temperatures in Australia. Tall and broad canopy plantations will keep the flying foxes cool and water availability due to water sources conservation can be utilized to sprinkle to keep flying foxes cool during the extreme temperatures in the future.