INTRODUCTION
In temperate zones of North America, hibernating animals, includingMyotis lucifugus (little brown myotis), bridge resource poor winters through energetic budgeting and behavioural changes (Hock 1951, Wang 1978, Speakman and Rowland 1999, Ruf and Geiser 2015). Survival over winter hibernation depends upon three main facets: 1) the amount of energy stored, primarily in body fat, 2) energetic expenditure (rate of metabolic consumption), and 3) the duration of hibernation (Humphries et al. 2002). Hibernation is composed of bouts of torpor, during which body temperature drops to near ambient temperature to limit heat loss and results in reduced metabolism to restrict the consumption of finite metabolic resources. Torpor is periodically interrupted by energy intensive periods of arousal during which hibernators return to euthermic body temperature (Hayman et al. 2010). Hibernators arouse for a variety of proposed reasons (for a review see Carey [2019] and citations within), including the need to eliminate metabolic waste, regain water balance, or mate. While arousals represent a small fraction of the total time spent in hibernation, they account for the majority of energy consumed, with a single arousal costing as much as 5% of total overwinter energetic costs (Thomas et al. 1990).
Microclimate selection is critical for hibernators (Boyles et al. 2007). By allowing body temperature to drop during torpor, M. lucifugusconsumes roughly 80-fold less energy per unit time due to the relationship between metabolic rate and temperature (Hock 1951, Speakman and Thomas 2003). To maximize the utility of these metabolic reductions, bats seek out caves, mines, scree slopes, or other locations generally referred to as ”hibernacula” where they can overwinter (Speakman and Thomas 2003). Relative to the surface landscape, subterranean locations can provide suitable low temperature (i.e., 0 –10 ˚C) habitats for hibernators (Thomas and Cloutier 1992). Microclimate selection will vary greatly based on species-specific preferences (Haase et al. 2021), and roost selection within the larger hibernaculum critically affects both the frequency of arousals and efficiency of torpor (Humphries et al. 2002, Czenze et al. 2013, Haase et al. 2019). Species such as M. lucifugus appear to choose roost locations with stable, nearly saturated environments and low temperatures to ameliorate their relatively high rates of evaporative water loss, while other bat species are capable of using more arid, and less thermally-stable, roosts as hibernacula (Klüg-Baerwald and Brigham 2017, Klüg-Baerwald et al. 2017). Even so, within a cave or mine system, roost conditions may not remain stationary throughout the duration of hibernation and some bats will relocate within the hibernaculum to seek specific microclimate conditions as their body condition changes (Hayman et al. 2010).
The duration of winter hibernation is another critical determinant to the survival of hibernators. The stimuli that drive immergence (entrance) to and emergence (exit) from hibernation, and their geographic variation, are under-described (Norquay and Willis in press, Lane et al. 2012, Czenze et al. 2013). The duration of winter hibernation presents a strong selective pressure, as longer winters result in shorter growing seasons and less time available for pre-hibernation fattening (Kunz et al. 1998). While animals that hibernate at more southern latitudes may be able to capitalize on breaks in the winter weather to opportunistically feed (Thomas et al. 1990), this is not always an option for other latitudes. Entrances to hibernacula may be blocked by snow, preventing foraging even if breaks in the weather did allow for the re-emergence of prey species (CLL,unpublished data ). Because North American hibernating bats feed on insects, researchers have estimated effective hibernation duration based upon the number of freezing days, with the assumption that freezing temperatures prevent insect availability until spring (Humphries et al. 2002, Hayman et al. 2016). However, some populations of bats may emerge from hibernation while freezing temperatures are still present, suggesting that there may be more complex determinants of emergence times (Johnson et al. 2017). Additionally, site-level differences, such as slope aspect, foliage cover, and proximity to water may influence the density of prey insects and their ability to persist on the landscape.
Beyond the normal challenges of hibernation, the epizootic white-nose syndrome (WNS), caused by the psychrophilic fungusPseudogymnoascus destructans , has increased energetic demands for hibernating bats (Blehert et al. 2009). The fungal pathogen responsible for the disease has spread rapidly through North America since 2006 and has killed millions of hibernating bats (Frick et al. 2015). Bats can be exposed to the fungus during the swarming period or over hibernation in the hibernacula. Once infected, the fungus grows during hibernation when bat skin temperatures are reduced and immune function is suppressed (Verant et al. 2012, Langwig et al. 2015, 2016). While there is still discuss the ultimate cause of mortality in WNS-impacted bats, it has been linked to the increased frequency of arousals in infected bats, ultimately resulting in the depletion of fat stores prior to the end of the hibernation period and subsequent starvation (Warnecke et al. 2012, Lilley et al. 2016).
There has been much research on energy consumption over hibernation in multiple bat species (Thomas et al. 1990, Cryan and Wolf 2003, Willis et al. 2006, Jonasson and Willis 2012, McGuire et al. 2014, Haase et al. 2019) and across the distribution of single bat species (Humphries et al. 2002, Hayman et al. 2016), but we have yet to determine how the spatial variation in the other two critical parameters governs overwinter survival for hibernators: duration of winter hibernation and amount of fat stores taken into hibernation. Previous models (Hayman et al. 2016) allowed for spatial variation in winter duration; however, the definition of winter duration was made a priori and based solely upon the number of nights with an average temperature below 0°C (Humphries et al. 2002). Similarly, the amount of body fat has generally been fixed as 25-30 % of total body mass in most studies (Humphries et al. 2002, Hayman et al. 2016, Haase et al. 2021). Fat resources are a major determinant of survival (Haase et al. 2019) and thus this assumption of proportion of body fat warrants review. Here we used generalized linear and linear models to: 1) estimate hibernation duration, 2) relate body mass to pre-hibernation fat stores, and 3) predict pre-hibernation body mass and fat across the distribution ofM. lucifugus . We focused on M. lucifugus due to the high impact of WNS on M. lucifugus populations, its widespread distribution, and the availability of published data. We predicted that spatial variation in overwinter duration would drive variation of body mass and fat to account for different energy requirements leading to spatial variation in WNS disease outcomes. Finally, we used a mechanistic model of hibernation energetics (Haase et al. 2019) to estimate the total metabolic costs of hibernation with and without the impacts of WNS across the species’ distribution of M. lucifugus .