The effects of fire and environmental variables on the ecophysiology, ecology and behaviour of insectivorous bats

Abstract

Bats are unique among mammals in being the only mammal capable of true flight and the second largest order of mammals in the world. Because of their small body size and ability to fly, they occupy a wide range of habitats and have developed varying strategies to cope with the constraints associated with their environments. Insectivorous bats have a high surface area to volume ratio, and therefore they must compensate for high rates of heat loss with high rates of energy expenditure. One way bats deal with this limiting energetic conundrum is the use of torpor, whereby metabolic rate (MR) and body temperature (Tb) are markedly reduced. Previous work from the Northern Hemisphere has shown that bats will hibernate in response to a reduction in ambient temperature (Ta) and food availability. However, recent studies from the Southern Hemisphere have shown that torpor is expressed by bats not only in harsh climates or when bats are energetically constrained, but also warm or mild climates, when food availability is scant or in abundance, and is even used by females when pregnant and during the reproductive period. An increase in global temperature has amplified the need to understand how heterothermic endotherms respond to ecological constraints in their environment. Global warming can affect the physiology and ecology of bats not only via a long-term increase in Ta, but also by encouraging fire and other extreme weather events. How bats respond physiologically to fire on a long-term scale remains, to my knowledge, uninvestigated. Temperature effects may confound the response of bats to habitat modification due to fire, or bats may alternatively use more or less torpor depending on roost choice and food availability. Additionally, research on how smoke and fire affects bats in the immediate term and whether it encourages arousal from torpor and, consequently, escape and survival, is scant. Fire may also burn roosting trees, darkening bark and creating differing roost microclimates, in turn affecting the physiological patterns of bats. Additionally, bat boxes made to provide roosting habitat for bats are often painted without consideration to how colour will affect torpor expression; roost choice can have implications for the duration of torpor and thus duration of activity as well energy expenditure. For example, bats may choose darker and/or more thermally labile roosts to allow for midday passive rewarming or, alternatively, may choose cooler and/or more stable roost types to reduce energy expenditure during torpor to a minimum. How bats physiologically respond to darker coloured roosts remains, to my knowledge, unknown. Not only are bats potentially affected by fire and roost type, but also by other mild or extreme variations in climate. Thus, it is unsurprising that nightly bat activity and the number of species present is linked to environmental variables. However, how bat activity and the number of species present respond to a temporal increase from the end of the hibernation period and into the reproductive cycle in spring remains relatively unknown in the Australian context. It is possible that despite fluctuations in environmental variables, reproductive females are pressured to increase foraging effort to increase body mass prior to parturition. Therefore, I aimed to untangle the complicated web of relationships between habitat modification, acute environmental stressors, roost choice, as well as environmental and temporal variables, such as Julian date, Ta, rain, lunar cycle and wind speed to determine how they affect the expression of torpor and/or activity on bat assemblages. My Ph.D. focuses on the Gould's long-eared bat ('Nyctophilus gouldi') and the lesser long-eared bat ('Nyctophilus geoffroyi') because of their appropriateness to answer my research questions. My study shows torpor expression is affected by habitat modification and food availability after wildfire. Bats used less torpor immediately after a fire, when presumably insect abundance had increased locally, compared to two years later when vegetation had begun to regrow. Additionally, they used more passive rewarming four months after the fire compared to two years later, potentially because habitat was largely decluttered after the fire and likely allowed for more radiant heat penetration, in turn warming roosts. The results of this study are opposite to what has been found recently in terrestrial mammals. Although hibernators have the option to use more torpor in the face of disasters, they are able to opportunistically capitalize on a post-fire increase in pyrophilous insect abundance and choose to maintain high, normothermic Tb even in a severely burnt habitat. Not only do bats respond physiologically to habitat modification and food availability due to fire, but also to an introduced, nontactile stimulus, smoke. Bats in a separate study that were exposed to a smoke stimulus responded by increasing respiratory rate (RR). Importantly, bats at low Ta/Tb (< 15°C) took longer to respond to smoke. Additionally, bats at Ta < 15°C took longer to reach their peak respiratory rate during the arousal process and took longer to express movement. Perhaps unsurprisingly, following smoke exposure, bats at Ta < 15°C re-entered torpor more quickly than bats at high Ta. However, bats at Ta < 15°C never returned to heart rate values consistent with steady-state torpor, whereas all bats at Ta ≥ 15°C did. Bats at low Ta may have been reluctant to re-enter deeper torpor only to again be exposed to smoke, whereas bats at high Ta could energetically "afford" to re-enter torpor as it would take less time and energy to rewarm from a high torpor Tb. My study also provides the first evidence that Australian tree-roosting bats show a clear preference for dark-coloured roost boxes in winter. Additionally, their expression of heterothermy is significantly affected by both roost colour and feeding regime. Bats expressed longer and deeper bouts of torpor when food-restricted (given food on every 4th day) and when roosting in white-coloured boxes than when fed ad libitum and roosting in black boxes. Furthermore, when roosting in black boxes, bats were able to passively rewarm to a higher box temperature, reducing the normothermia-torpor skin temperature differential and thus expending comparatively less energy during the active arousal process compared to bats roosting in white boxes. Surprisingly, bats showed similar roost colour choice responses both when fed 'ad libitum' and when food-restricted. This is likely because the study species shows a high proclivity for passive midday rewarming, even during winter. My work has shown that, as bats move out of the winter hibernation period and into the reproductive cycle in spring, nightly activity increases. I used a step-wise elimination procedure to determine which variables best predicted bat activity from the end of the hibernation period and into the reproductive cycle; I found that an increase in Julian date was the only consistent predictor of nightly activity amongst overall and species-specific models. Further, the number of species present largely tracked patterns of Ta and increased on warmer nights, while the effects of lunar cycle on nightly activity were complex and appeared to be species-dependent. Additionally, I showed that a single, common bat species emerged on the night of a severe storm, indicating that some bats may be active during and immediately following extreme weather events. In conclusion, my study is the first to show that a heterotherm capable of hibernation show a positive physiological response to fire by reducing torpor and increasing normothermia to capitalize on an influx of insects. I have provided the first evidence that bats respond to smoke by increasing RR and, at low Ta, are more reluctant to re-enter steady-state torpor following smoke exposure. I have also shown that bats will choose darker coloured roosts over lighter ones when given the option, likely as a mechanism to facilitate normothermia and reduce physiological demands during the active rewarming process. Finally, I assessed how nightly activity and the number of species present, measured on a continuous, nightly scale, respond to Julian date and other environmental variables from the end of hibernation.