Microclimate conditions of bat boxes occupied by New Zealand's native long-tailed bat (Chalinolobus tuberculatus) in Hamilton City

Abstract

Globally, land clearance has resulted in the destruction of bat habitat features such as tree roost cavities. Installation of artificial roosts, also known as bat boxes, is intended to mitigate such losses. However, bat boxes are often installed with limited understanding of conditions such as temperature stability and relative humidity required by the target bat species. In New Zealand, long-tailed bat/pekapeka-tou-roa (Chalinolobus tuberculatus) populations largely favour tree cavities for roosting, and populations have been documented across a range of habitats including native forest, agricultural land, and within urban areas such as Hamilton City. Temperature and relative humidity have been identified as important roost selection factors for C. tuberculatus, but these properties have never been assessed in New Zealand bat boxes.

To understand microclimate conditions within bat boxes occupied by C. tuberculatus in relation to natural roosts, environmental loggers were installed in Kent and Schwegler type roost boxes in Hamilton City, and natural roost cavities near Hamilton and Pukekohe. For further comparison, loggers were also installed near the top and bottom of Grand Canyon Cave, a known roost for C. tuberculatus. Fortnightly surveys of Hamilton City bat boxes were conducted using an endoscopic camera mounted on a 15 m telescopic carbon fibre pole or evening watches of tree roosts. This provided box occupancy rates and allowed comparisons to be made between temperature and relative humidity within bat boxes and natural roosts in relation to occupancy.

Twenty-nine (63 %) of the 46 monitored bat boxes had confirmed occupation, with generalised linear mixed modelling (GLMM) indicating that sheltered Kent boxes were more likely than exposed Kent boxes to be occupied. Modelling also indicated that there was no significant effect of box chamber on occupation. However, microclimate conditions of Kent boxes were less stable than natural roosts, particularly exposed boxes, although the differences may not have been substantive enough to be physiologically relevant to C. tuberculatus. On average, bat boxes were generally warmer or similar to ambient conditions on a diel cycle, while natural roosts were warmer than ambient at night and cooler during the day. Although heat tolerance of C. tuberculatus is unknown, criteria used for similar species indicated that two exposed Kent boxes experienced a limited number of potential overheating events (>40 °C). The microclimate conditions in Grand Canyon Cave indicated differences between the top and bottom of the cave, with the top of the cave having similar thermal properties to tree cavity roosts. Generalised linear mixed modelling indicated that there was no significant effect of box thermal conditions on occupation rates.

Although microclimate differences between the inner and outer chambers of Kent type boxes were potentially too small to be physiologically relevant to C. tuberculatus, microclimate stability and box exposure should be considered during installation. Providing a range of internal microclimates is likely to support the range of thermoregulatory requirements of the population. Future research should consider the development of boxes with improved microclimate stability, and designs which facilitate the thermoregulatory requirements of C. tuberculatus during different reproductive phases.