To investigate the impact of short summers and long summer solar periods at high latitudes on the behavior of a nocturnal, hibernating mammal, we recorded the phenology of Myotis lucifugus (little brown myotis) at 60^oN in the Northwest Territories (NWT), Canada. In particular we assessed the timing of spring emergence from, and autumn entry into hibernation, reproduction, and seasonal mass fluctuations. We used a combination of acoustic monitoring and capture surveys at two hibernacula and two maternity roosts during 2011 and 2012. Myotis spp. were active at the hibernacula from late April to late September/early October, suggesting that the ‘active’ season length is similar to that of populations farther south. At maternity colonies, we detected M. lucifugus activity from early May to early October, with peaks during mid-July in both years. Lactation, fledging, and weaning all occurred later in the NWT than at more southern locations, and reproductive rates were significantly lower than rates observed further south. Average mass of individuals fluctuated throughout the season, with an initial decline immediately following emergence from hibernation likely reflecting increased energy expenditure due to flight and decreased use of torpor, coupled with relatively low prey intake due to low prey abundance associated with cool temperatures. Females did not appear to have lower pre-hibernation masses than those in more southern populations, suggesting that despite the cool spring and autumn temperatures, and short summer nights, bats are able to obtain enough energy for reproduction and mass accumulation for hibernation. However, the lower reproductive rates may indicate that there are limitations to life at the northern limits of the species’ range.

Capture data - We used mist nets (Avinet Inc. Dryden, New York), to capture bats adjacent to the two Myotis lucifugus maternity colonies (Thebacha and Lady Evelyn Falls), and various foraging sites around the South Slave region, during 2011 (35 nights; May - Sep) and 2012 (23 nights; Jun - Sep). At maternity colonies, we captured bats during emergence, before they had an opportunity to feed, and at all sites we held bats in a cloth bag for ≥ 0.5 h to allow defecation prior to weighing. A small number of bats were caught post-foraging; these records have been marked with an asterix in the 'massinacuracy' column of the dataset and were omitted from the mass-related analysis. We identified individuals to species, sex, age (adult or juvenile) and reproductive condition, weighed them with a portable balance, measured forearm with calipers, fitted them with a unique band (2.9 mm; Porzana Ltd., East Sussex, United Kingdom) for future identification, and released them at their point of capture. We classified adult females into three categories: 1) Pregnant: swollen abdomen, detection of fetus via palpation of abdomen, non-enlarged nipples; 2) Lactating: enlarged, calloused nipples, milk expressed with gentle massage of mammary glands; 3) Post lactating: enlarged keratinized nipples with no milk expression (Racey, 2009). Females captured prior to 1 June were not included when calculating reproductive rates as non-reproductive and early pregnancy females were indistinguishable by external examination. All volant juveniles were identified by evenly tapered phalangeal-metacarpal joints and un-fused epiphyseal plates observed by back-illuminating the wing (Brunet-Rossinni and Wilkinson, 2009). During November 2011, after hibernation was well underway, we entered both hibernacula and recorded sex, mass and forearm length for 45 bats captured from the cave walls (Walk-in cave: n = 5; South Slave hibernaculum: n = 40). All captures followed American Society of Mammalogists guidelines (Sikes et al., 2016) and were compliant with the University of Calgary Animal Care and Use Committee protocols (Animal Care Certificate BI09R-01).

Acoustic surveys – We recorded echolocation calls using AnaBat II ultrasonic detectors with CF-ZCAIMs set to a division ratio of 16 (Titley Electronics, Ballina, Australia). Detectors were placed at five sites including two hibernacula (Walk-in Cave and South Slave Hibernacula) and two beaver ponds (WBBP and NTBP) within 3 km of each hibernaculum, and at the Thebacha maternity colony. At Walk-in Cave, we placed the detector in a weatherproof box, 3 m from the entrance with the microphone pointed away from the entrance to prevent recording bats flying within the cave. Two detectors were installed in weatherproof boxes at the South Slave hibernaculum by the Government of the Northwest Territories, approximately 10 m from the cave entrance with the microphones positioned towards the cave entrance, but the restricted cave opening prevented recording of bat activity inside the cave. At the Thebacha maternity colony, we mounted the detector directly along a bat flight path outside the main building used by the colony, 3 m off the ground and 5 m from the bat exit.

Acoustic analysis - We reviewed all AnaBat files using AnaLookW version 3.2.15. We used a filter to extract Myotis spp. calls, identified by a minimum frequency ≥ 30 kHz , and counted files recorded per night. Bats with lower-frequency echolocation calls (primarily E. fuscus) were omitted from the data. Call characteristics of M. lucifugus and M. septentrionalis overlap (Fenton and Bell 1981), often making them indistinguishable with AnaBat II detectors, therefore, we did not attempt to remove M. septentrionalis calls from the data. The composition of Myotis captures (via mist-netting) at these sites averaged 86% M. lucifugus and 14% M. septentrionalis (n = 1118; Reimer and Barclay, DRYAD dataset), therefore, we assume that the Myotis echolocation passes in our data were largely representative of M. lucifugus activity.