nena masthead
NENA Home Staff & Editors For Readers For Authors

Hibernating Bats and Abandoned Mines in the Upper Peninsula of Michigan
Allen Kurta and Steven M. Smith

Northeastern Naturalist, Volume 21, Issue 4 (2014): 587–605

Full-text pdf (Accessible only to subscribers. To subscribe click here.)

 

Access Journal Content

Open access browsing of table of contents and abstract pages. Full text pdfs available for download for subscribers.



Current Issue: Vol. 30 (3)
NENA 30(3)

Check out NENA's latest Monograph:

Monograph 22
NENA monograph 22

All Regular Issues

Monographs

Special Issues

 

submit

 

subscribe

 

JSTOR logoClarivate logoWeb of science logoBioOne logo EbscoHOST logoProQuest logo

Northeastern Naturalist Vol. 21, No. 4 A. Kurta and S.M. Smith 2014 587 2014 NORTHEASTERN NATURALIST 21(4):587–605 Hibernating Bats and Abandoned Mines in the Upper Peninsula of Michigan Allen Kurta1,* and Steven M. Smith2 Abstract - Prior to arrival of white-nose syndrome, we found bats hibernating in 82 of 119 abandoned mines in northern Michigan. Unoccupied sites typically were short (19 ± 17 m SD) and/or experienced chimney-effect airflow, which led to temperatures near or below freezing (-0.8 ± 2.9 ºC). Overall, occupied sites were more structurally complex, longer (307 ± 865 m), and warmer (5.7 ± 3.0 ºC) than unoccupied mines. Number of bats varied from 1 to > 55,000, although the median was 115. Perimyotis subflavus (Eastern Pipistrelle) and Eptesicus fuscus (Big Brown Bat) accounted for only 0.5% of the total of 244,341 bats that were observed. Ninety percent of hibernating animals were Myotis lucifugus (Little Brown Bat), and almost 10% were M. septentrionalis (Northern Bat). Relative to Little Brown Bats, Northern Bats were more common in the mines of the Upper Peninsula than in hibernacula in the East and Ohio River Valley. Maximum ambient temperature, presence of standing water, and water vapor pressure deficit were potential predictors of the number of Myotis that was present. Seventy-five percent of Northern Bats and 22% of Little Brown Bats roosted alone, rather than cluster with other bats. Little Brown Bats in Michigan were solitary much more often than in the East. Introduction In temperate areas, such as the Great Lakes region, bats provide economically important ecosystem services to humans that involve consumption of agricultural and forest pests and potential vectors of human disease (e.g., Long et al. 2013, Münzer 2008). Despite the economic importance of bats, these long-lived, slowreproducing mammals are experiencing unprecedented mortality.The burgeoning wind-power industry in the United States now kills an estimated 600,000 bats per year through collisions with spinning turbine blades and barotrauma (Hayes 2013). In addition, an introduced disease, white-nose syndrome, may have caused the death of >5.5 million hibernating bats since 2006 (US Fish and Wildlife Service 2012). Pseudogymnoascus destructans (Gargas, Trest, Christensen, Volk, and Blehart), the fungus that causes white-nose syndrome, induces frequent arousals from hibernation, which ultimately lead to the starvation of these small mammals by depleting fat reserves long before food (flying insects) is available in spring (Reeder et al. 2012). The disease spread rapidly, northward and southwestward from its epicenter in eastern New York, but it was not documented in the western basin of the Great Lakes (Wisconsin and Michigan) until spring 2014 (Whitenosesyndrome. org 2014). 1Department of Biology, Eastern Michigan University, Ypsilanti, MI 48197. 2S.M. Smith Co., 1105 Westwood Avenue, Iron Mountain, MI 49801. *Corresponding author - akurta@ emich.edu. Manuscript Editor: Hugh Broders Northeastern Naturalist 588 A. Kurta and S.M. Smith 2014 Vol. 21, No. 4 Species of bats that are affected by white-nose syndrome hibernate underground in caves and cave-like sites, such as mines and concrete tunnels, where temperatures remain cool but above freezing throughout winter. Suitable subterranean retreats are rare in many parts of the western Great Lakes region (Kurta 1995), such as the Lower Peninsula and eastern Upper Peninsula of Michigan, where only 5 hibernacula are known (Kurta 2008, Slider and Kurta 2011). The western Upper Peninsula, though, is geologically different from the rest of the state, and hundreds of underground mines were excavated there in a search for iron and copper beginning in the 1840s (Lankton and Hyde 1982, Reed 1957). Although none of these mines is operating today, and most have become flooded or are permanently sealed, some sites remain open and are used by overwintering bats (Kurta and Smith 2004). Little information exists on bats in the mines of the Upper Peninsula. Available publications do not focus on characteristics of the mines as hibernacula (e.g., Bricklin et al. 2007, Stones and Branick 1969, Stones and Oldenburg 1967, Sullivan et al. 2012), and only 1 limited attempt has been made to catalog the species of bats that are present (Stones 1981). We have examined more than 100 open mines throughout the copper and iron ranges of Michigan since 1996, searching for hibernating bats, conducting repeated censuses, and recording physical characteristics such as length, temperature, and humidity. In this paper, we summarize that data to establish a baseline for future comparison of the bats and their hibernacula following arrival of white-nose syndrome. In addition, we describe characteristics of occupied and unoccupied sites, investigate factors affecting the number of bats that are present in a mine, quantify the tendency of the most common species to form clusters during hibernation, and make pertinent comparisons with overwintering bats in the East and in the Ohio River Valley. Methods Study area Our surveys took place in the western third of the Upper Peninsula, which is bordered by Lake Superior to the north and Wisconsin to the south and west (46.79–47.43°N, 87.59–90.14°W). The area is sparsely populated by humans, with densities as low as 1.6 persons/km2 in some counties (US Census Bureau 2014). The region is largely covered by northern hardwoods and conifers and underlain by igneous and metamorphic rocks (Albert et al. 1986). Elevation varies from 184 to 604 m above sea level, but maximum elevation above Lake Superior is only 427 m. Winters are snowy, long, and cold. Although amount of snowfall varies with elevation and distance from Lake Superior, average snowfall is up to 6.1 m/year (Keen 1993). Average date of the first overnight minimum temperature of 0 °C for 14 communities in the study area is 18 September, and the last such minimum typically occurs on 1 June. The average, lowest, minimum temperature reached each winter varies from -34 to -25 °C (Albert et al. 1986). Mean annual temperature, which greatly impacts the temperature of underground sites where bats hibernate (Domínguez-Villar et al. 2013), is 4.2–5.5 °C (Albert et al. 1986). Northeastern Naturalist Vol. 21, No. 4 A. Kurta and S.M. Smith 2014 589 Mining terminology To understand where bats are usually found within the mines, definitions of basic mining terms are needed (Lankton and Hyde 1982, Reed 1957, Sherwin and Altenbach 2009). A “shaft” is an entrance to a mine that penetrates the earth at a steep angle. In Michigan, shafts may be vertical, especially at iron mines, but many shafts for copper mines descend at an angle of 45–75°, following the dip of mineral-bearing strata. An “adit” is a horizontal passage entering a mine that may end blindly or connect with shafts and drifts. A “drift” is any horizontal passage, extending from a shaft or adit, whereas a “crosscut” extends laterally from a drift. Miners excavated drifts from shafts at regular intervals, typically every 30 m, and each successive drift is considered a “level.” A “winze” is a shaft-like opening downward from a drift, usually to connect with a lower level. A “stope” is a cavity formed by removal of ore; stopes typically extend from the side of a drift and are inclined upward, following the mineral-bearing rocks. Field surveys Physical characteristics of mines. After finding a mine, we determined its latitude and longitude, using hand-held global-positioning units, and noted maximum width and height of the largest entrance. Upon entering a new site, we conducted a simple survey, recording approximate length, passage dimensions, and other significant features, such as presence of different levels, stopes, and standing water. We arbitrarily defined standing water as any pool at least 15-cm deep, 2-m long, and 1.5-m wide, which was the typical width of adits and drifts. Standing water most commonly occurred as pools in adits or drifts and occasionally in flooded shafts or winzes. We also assigned each mine a number that indicated its complexity. A score of 1 was given to mines that consisted of a single adit and/or shaft. A complexity of 2 indicated mines with drifts or crosscuts, and a value of 3 was assigned to sites with 1–2 stopes and/or a second level that was less than 25% of the length of all passages in the mine. The highest score of 4 was given to mines with 3 or more stopes and/or 2 or more levels, with the additional levels contributing more than 25% to the total length of the mine. We measured environmental variables as close to the ceiling as possible, and the number of such measurements varied with the complexity of each mine. In simple mines (complexity of 1–2), measurements occurred at the midpoint and end of adits and at the end of drifts or crosscuts. In complex sites (3–4), additional measurements were made at the top of stopes and on each level at the end of the drifts. Relative humidity was determined with a sling psychrometer, and ambient temperature was measured with various electronic instruments or the dry bulb of the psychrometer. We calculated equilibrium vapor pressure using temperature and the quadratic formula of Tabata (1973) and then determined actual vapor pressure, based on relative humidity. Evaporative water loss from an animal is proportional to the difference between vapor pressure of the body fluids (as determined by surface temperature of the body) and saturation vapor pressure (Hill 1976). However, if surface temperature of the animal is equal to ambient temperature, as is often true for a hibernating Northeastern Naturalist 590 A. Kurta and S.M. Smith 2014 Vol. 21, No. 4 bat, then vapor pressure of the body fluids is equal to the equilibrium vapor pressure (Kurta 2014). Consequently, we used the water vapor pressure deficit (i.e., the difference between equilibrium and actual vapor pressure) as an index to potential evaporative water loss that a hibernating bat might experience. Number and species of bats. In each site, we counted the bats that were present with the aid of hand counters. Bats in northern Michigan begin entering hibernation by late September and start leaving in late April, although a few are present into June (Kurta et al. 1997, Stones and Fritz 1969). Consequently, we performed counts between late December and March, although some visits unavoidably occurred outside this window. In particular, entrances to 7 occupied mines usually became completely blocked with ice by late autumn and necessarily were surveyed in late November or early December prior to icing-up. Although surveys began in 1996, most mines (93%) in which counts occurred were discovered or revisited between 2010 and 2013. The authors often simultaneously counted bats in different sections of a site to minimize time inside a hibernaculum. To compare our abilities, on 11 occasions, both authors counted the same bats in sections of mines that contained 140–1576 bats. The mean difference between counts was 3.6 ± 17 (SD) bats and was not significantly different from zero (paired t10 = 0.72; P = 0.49), indicating no consistent bias between observers. We also identified some bats to species and estimated relative abundance. Perimyotis subflavus (Cuvier) (Eastern Pipistrelle) and Eptesicus fuscus (Palisot de Beauvois) (Big Brown Bat) can be identified unambiguously from a distance, and we recorded the presence of these species whenever we encountered them. However, definitive identification of Myotis lucifugus (LeConte) (Little Brown Bat) and M. septentrionalis (Trouessart) (Northern Bat) required close inspection of a roosting animal and often handling the bats so that the ear and tragus were more visible. To minimize disturbance to bats and the attendant use of stored fat (Speakman et al. 1991, Thomas 1995), we generally did not identify each Myotis to species and simply combined all Myotis in our counts. Nevertheless, we did examine samples of Myotis, in hand, when time and safety allowed, to determine overall relative abundance of Little Brown and Northern Bats. Samples were obtained by arbitrarily delineating sections of passage, about 10 m in length, and identifying all bats that were present. Samples were obtained before and after the midpoint of simple mines and in the stopes and on different levels of complex mines. Clustering. Many species of bats cluster while hibernating, and Langwig et al. (2012) recently suggested that clustering occurs less often after P. destructans is established. Therefore, we also gathered baseline information on clustering patterns of Myotis at 13 sites during winters 2011–2012 and 2012–2013. All tunnels for this survey were typical adits and drifts, with passages about 1.8-m high and 1.5-m wide, which allowed the senior author to inspect each roosting bat, identify it to species, and note the number of individuals with which it was in contact. If a survey yielded less than 10 individuals of a species, then data for that species from that mine were discarded. Clusters containing more than 1 species were uncommon and not included in the analyses. Northeastern Naturalist Vol. 21, No. 4 A. Kurta and S.M. Smith 2014 591 Analysis of data Species of Myotis were the most common (>99%), and to investigate factors affecting number of Myotis present in a site, we performed negative binomial regression, using a loge-link function and loge of the length of the mine as an offset variable (Coxe et al. 2009, Gardner et al. 1995). Although many factors may affect the number of bats, we restricted our analysis to 5 variables to avoid overfitting and spurious results (Anderson 2008). Based on our knowledge of bats and the literature, we hypothesized that potentially important independent variables were maximum ambient temperature, which affects energetic expenditure (Boyles et al. 2007); water vapor pressure deficit, which should reflect evaporative water loss (Kurta 2014); presence of standing water, which would provide a source of drinking water (Taylor and Tuttle 2007); and size of the largest entrance, which could impact a bat’s ability to access the site. Because maximum temperature in an underground site typically is limited by surface temperature (Slagstad et al. 2008), we always paired maximum temperature squared with temperature, to reflect the expected nonlinear relationship (Hayes et al. 2011). Height and width of the entrance were highly correlated, so we used principal components analysis to create a single variable reflecting overall maximum size of the entrance. We analyzed 15 models that represented the null model (intercept only) and all possible combinations of the variables of entrance size, water vapor pressure deficit, presence of standing water, and the pair of temperature variables, as main effects (see Results). We evaluated competing models using an informationtheoretic approach and by calculating Akaike’s information criteria (AICc), model probabilities, and evidence ratios (Anderson 2008). Only occupied mines (by bats of any species) that had been completely surveyed and for which all variables had been measured were used in this analysis. In addition to the negative binomial regression, we examined some simple relationships between variables using Pearson’s correlations. Comparisons between means were analyzed with appropriate t-tests. All statistics were performed using IBM Statistics 21 (IBM Corporation 2012). Means are presented with ± 1 standard deviation, and alpha was set to 0.05. Results We examined 119 subterranean sites in the western Upper Peninsula, including 91 copper mines, 26 iron mines, 1 dolomite mine, and 1 putative gold mine. Mines were not randomly distributed across the landscape, but their location was highly dependent on local geology (Fig. 1). In particular, most copper mines were restricted to a narrow outcropping of igneous rock called the Portage Lake Volcanics (Butler and Burbank 1929), and most iron mines were located in the southeastern part of the study area, in the Menominee Iron Range (Reed 1957). Physical characteristics of mines Unoccupied sites. Thirty-seven sites (31%) contained no bats of any species. At 1 mine, a modern building had been erected over the shaft, effectively excluding Northeastern Naturalist 592 A. Kurta and S.M. Smith 2014 Vol. 21, No. 4 bats. The 36 other sites without bats were typically short (19 ± 17 m; Table 1), cold (-0.8 ± 3.1 ºC; Table 1), and simple, with 97% receiving a complexity score of 1. Most of these (84%) were blind-ending shafts (14 sites) or adits (17). Although 5 additional mines possessed 2–3 entrances, all underground passage was located between the openings, leading to noticeable chimney-effect airflow and temperatures near freezing (Sherwin and Altenbach 2009). Occupied sites. Eighty-two of the 119 mines (69%) contained at least 1 hibernating bat. Total length of passage for the occupied mines varied from 10 m to 6.4 km, with an average of 307 ± 865 m (Table 1; Fig. 2). Number of bats of all species present was correlated with length of the mine (loge-transformed values: r = 0.50; P < 0.0001). Although most occupied sites (57%) also received a complexity score of 1, many were given values of 2 (23%), 3 (9%), or 4 (11%). Complexity and logetransformed length were correlated (r = 0.79; P < 0.0001; n = 81). Maximum ambient temperature in occupied sites was 5.7 ± 3.0 ºC (Table 1), and most (55%) mines used by bats were between 5 and 9 ºC (Fig. 3). Mean vapor pressure deficit was (0.075 ± 0.071 kPa). Pools of standing water occurred in 52 (63%) of the 82 mines. Figure 1. Map showing the location of 82 mines used for hibernation by bats in the western Upper Peninsula of Michigan. Background shading represents different types of bedrock; PLV = Portage Lake Volcanics. Northeastern Naturalist Vol. 21, No. 4 A. Kurta and S.M. Smith 2014 593 Table 1. Characteristics of mines that were unoccupied or occupied by different groups of bats in the western Upper Peninsula of Michigan. Listed values are mean ± SD (median), range, n. Maximum ambient Water vapor pressure Type of site Length (m) Number of bats temperature (°C) Water vapor pressure (kPa) deficit (kPa) Unoccupied mines 19 ± 17 (14), -0.8 ± 2.9 (-0.3), 0.697 ± 0.112 (0.747), 0.111 ± 0.112 (0.075), 2–84, -8.7–4.0, 0.509–0.817, 0–0.355, 36 22 14 14 All occupied mines 307 ± 865 (52), 2980 ± 8494 (115), 5.7 ± 3.0 (6.4), 0.889 ± 0.134 (0.915), 0.075 ± 0.071 (0.070), 10–6402, 1–55,685, -1.3–10.8, 0.484–1.111, 0–0.260, 79 82 81 72 72 Mines with Eastern Pipistrelles 374 ± 589 (145) 3.3 ± 2.3 (3), 7.3 ± 2.4 (8.0), 0.963 ± 0.103 (0.988), 0.043 ± 0.063 (0.005), 24–2692, 1–10, 1.5–10.8, 0.709–1.111, 0–0.198, 23 23 23 22 22 Mines with Big Brown Bats 241 ± 525 (72), 24 ± 62 (4), 5.5 ± 2.8 (6.0), 0.878 ± 0.136 (0.910), 0.079 ± 0.075 (0.068), 14–2692, 1–338, -1.3–10.8, 0.484–1.111, 0–0.026, 46 46 46 43 43 Mines with Myotis 314 ± 886 (48), 3173 ± 8733 (132), 6.0 ± 2.9 (6.5), 0.898 ± 0.126 (0.924), 0.071 ± 0.069 (0.068), 10–6402, 1–55,685, -0.1–10.8, 0.533–1.111, 0–0.251, 77 77 78 72 72 Northeastern Naturalist 594 A. Kurta and S.M. Smith 2014 Vol. 21, No. 4 Over the years, many entrances had become partly blocked by dirt and rocks, often leaving an irregular space barely large enough for a human to pass. Maximum height or width of the largest entrance was less than 1 m in 25 occupied mines (32%), although average height and width was 1.6 ± 1.5 and 2.1 ± 1.9 m, respectively. Sixty-nine mines (84%) had only a single entrance. In sites with more than 1 opening, few, if any, bats roosted in the passage between entrances; virtually all bats (>95%) were concentrated instead in stopes above the tunnels that connected the entrances, in blind-ending drifts or cross-cuts, or on different levels. Number and species of bats Eastern Pipistrelles. We counted 244,341bats representing 4 species—the Little Brown Bat, Northern Bat, Big Brown Bat, and Eastern Pipistrelle, the latter of which was the least common species. Only 75 Eastern Pipistrelles were discovered, and the species was found at 23 (28%) of the 82 occupied sites and constituted 0.03% of all bats hibernating in the mines. Median number of Eastern Pipistrelles, when present, was 3, and the maximum was 10 (Table 1). Eastern Pipistrelles were Figure 2. Top: Percent of all occupied mines of different length and percent of the total number of bats associated with mines of different length. Bottom: Percent of all occupied mines of different complexity and percent of the total number of bats associated with mines of differing levels of complexity. Northeastern Naturalist Vol. 21, No. 4 A. Kurta and S.M. Smith 2014 595 never the only species present in a mine; all 23 sites were shared with Myotis, and 17 (77%) also contained Big Brown Bats. Hibernacula with Eastern Pipistrelles were warm (7.3 ± 2.4 °C) and had low water vapor pressure deficits (0.043 ± 0.063 kPa; Table 1). Big Brown Bats. A total of 1106 Big Brown Bats roosted in 45 (55%) of the 82 occupied mines, and the species represented about 0.5% of all bats overwintering in the mines. Median number present was 5. Thirteen mines contained only 1 Big Brown Bat, 19 had 2–10 individuals, 9 sites sheltered 11–40 Big Brown Bats, and 4 sites had more than 100 individuals, with a maximum of 338 Big Brown Bats in any mine. Forty-one hibernacula (89%) occupied by Big Brown Bats also contained at least 1 Myotis. Maximum air temperature in mines with Big Brown Bats was 5.5 ± 2.8 °C, and the water vapor pressure deficit was 0.079 ± 0.075 kPa (Table 1). Little Brown and Northern Bats. Species of Myotis occurred in 77 of 82 (94%) occupied mines and represented over 99% of all bats counted. Median number of Myotis in mines that contained at least 1 Myotis was 132, but the 39 hibernacula Figure 3. Top: Percent of all occupied mines having different maximum ambient temperature and percent of the total number of bats associated with different maximum ambient temperatures. Bottom: Percent of all occupied mines having different water vapor pressure deficits and percent of the total number of bats associated with different water vapor pressure deficits. Northeastern Naturalist 596 A. Kurta and S.M. Smith 2014 Vol. 21, No. 4 with the number of Myotis at or below the median accounted for only 1052 bats or 0.4% of total Myotis. Over half the Myotis (56%) were concentrated in 4 large, complex sites. Almost all Myotis (99%) hibernated in mines with a maximum ambient temperature of 4–10 °C, and 59% of Myotis occupied hibernacula with a temperature between 7 and 9 °C. Mean water vapor pressure deficit was 0.071 ± 0.069 kPa (Table 1). Samples containing as many as 875 Myotis were handled and unequivocally identified to species in 48 sites, with a strong correlation between number of bats in a sample and number of Myotis in the mine (loge-transformed values: r = 0.88; P less than 0.0001). Little Brown Bats constituted 90% of identified animals, and the other 10% were Northern Bats. Although relative abundance was not estimated in all sites, our notes indicated that Northern Bats and Little Brown Bats co-occurred in 92% of the 77 mines occupied by Myotis, and the few exceptions contained only 5 or fewer Myotis of either species. For the negative binomial regression, a complete set of data was available for 66 mines. The best model (lowest AICc) for predicting number of Myotis contained temperature, temperature squared, and presence of standing water, whereas the second best-supported model included the same variables along with water vapor pressure deficit (Table 2). The initial third- and fourth-place models involved addition of entrance size to the first and second models. However, entrance size appeared to be a “pretending variable” (sensu Anderson 2008) in both instances, because ΔAICc was approximately 2 when compared with the less complex model, Table 2. Candidate models and associated parameters for predicting number of Myotis in mines, based on negative binomial regression, with a loge-link function and loge of the length of a mine as an offset variable. Models were ranked using Akaike’s information criterion (AICc), model weights, and evidence ratios (Anderson 2008). Variables are: ta = ambient temperature; ta2 = ambient temperature squared; water = presence of standing water; wvpd = water vapor pressure deficit; and entrance = size of entrance, as indicated by scores extracted from principal components analysis of maximum height and width of entrance. Two models containing the variable entrance were discarded because entrance appeared to be a “pretending variable” (Anderson 2008); these excluded models were ta ta2 water entrance and ta ta2 water wvpd entrance. Model AICc ΔAICc Weight Evidence ratio ta ta2 water 807.71 0.00 0.652 1.00 ta ta2 water wvpd 809.22 1.51 0.307 2.12 water 814.93 7.22 0.018 36.98 water entrance 815.96 8.25 0.011 61.74 water wvpd 817.04 9.33 0.006 106.11 water wvpd entrance 817.41 9.70 0.005 127.55 ta ta2 entrance 821.29 13.58 0.001 888.03 ta ta2 822.03 14.32 0.001 1284.98 ta ta2 wvpd entrance 822.90 15.19 <0.001 1987.23 ta ta2 wvpd 824.27 16.55 <0.001 3934.35 intercept 829.95 22.24 <0.001 67,541.67 wvpd 830.44 22.73 <0.001 86,120.26 entrance 832.14 24.43 <0.001 20,1692.59 wvpd entrance 832.46 24.75 <0.001 23,6925.26 Northeastern Naturalist Vol. 21, No. 4 A. Kurta and S.M. Smith 2014 597 the deviance of the model changed minutely after addition of entrance size, and the 95% confidence intervals of the coefficient for entrance size included zero. Therefore, these 2 models were removed from the analysis. None of the remaining models was supported (all ΔAICc > 7). Clustering At 13 sites, we determined cluster size for 85–2585 Little Brown Bats (total = 9640 bats) that roosted in 22–1169 groups per mine (total = 4390 groups, with a solitary animal considered a group of 1). Solitary bats represented 52 ± 12% of the groups and 22 ± 11% of all Little Brown Bats examined in each mine, whereas clusters containing more than 10 bats were 4 ± 5% of the groups and involved 17 ± 16% of the individuals (Fig. 4). Average maximum number of Little Brown Bats in a cluster was 21 ± 14, with a single highest value of 58. Actual ambient temperature in these 13 mines at the time and location of the survey was 7.5 ± 2.3 °C, and the percentage of Little Brown Bats that was solitary was positively correlated with temperature (r = 0.67; P = 0.01; Fig. 5), i.e., Little Brown Bats were more likely to cluster at lower temperatures. Figure 4. Top: Average percent of groups of varying size and average percent of the total number of bats found in clusters of varying size for Little Brown Bats at 13 mines; overall, 9640 Little Brown Bats in 4390 groups were surveyed (a solitary animal was considered a group of 1). Bottom: Average percent of groups of varying size and average percent of the total number of bats found in clusters of varying size for Northern Bats at 9 mines; overall, 214 Northern Bats in 186 groups were surveyed. Northeastern Naturalist 598 A. Kurta and S.M. Smith 2014 Vol. 21, No. 4 For Northern Bats, cluster size was determined at 9 sites, containing 10–79 Northern Bats (total = 214 bats) in 8–72 groups per site (total = 186 groups). Solitary animals represented 87 ± 12% of all groups and 75 ± 21% of all Northern Bats (Fig. 4). Only 2 clusters, 1 with 4 and the other with 6 bats, contained more than 3 Northern Bats. Ambient temperature in these 9 mines was 7.3 ± 1.5 °C, but unlike Little Brown Bats, the percentage of Northern Bats that was solitary in each mine was not significantly correlated with temperature (r = 0.36; P = 0.34; Fig. 5). The proportion of Northern Bats that roosted alone was greater than for Little Brown Bats (paired t8 = 7.99; P < 0.0001) in the 9 sites in which both species were evaluated with a sufficient sample. Discussion Much of the literature on hibernating bats emphasizes the effect that the structure of an underground site has on creating an internal temperature suitable for torpor during hibernation. The oft-stated dogma is that caves and mines are too warm internally and must be cooled for successful hibernation through cold-air traps or chimney-effect airflow (McAney 1999, Nagel and Nagel 1991, Prather and Briggler 2002, Sherwin and Altenbach 2009, Tuttle and Kennedy 2002, USFWS Figure 5. Top: Percent of Little Brown Bats that was solitary in each of 13 mines versus ambient temperature. Bottom: Percent of Northern Bats that was solitary in each of 9 mines versus ambient temperature. Northeastern Naturalist Vol. 21, No. 4 A. Kurta and S.M. Smith 2014 599 2013). This pervasive view, however, is a southern perspective, applicable only to warm regions with shorter winters and higher mean annual surface temperatures than found in the Upper Peninsula. For example, chimney-effect airflow throughout simple mines up to 100 m in length in the Upper Peninsula causes air temperature to be near or below freezing, making such sites unsuitable for overwintering bats. Even in mines over 500 m in length, passages subjected to chimney-effect airflow are cold, drafty, and generally devoid of bats, and it is only in complex sites, with drifts, crosscuts, stopes, or multiple levels that chimney-effect airflow and large numbers of hibernating bats can co-occur in the North. Similarly, blind-ending shafts are cold-air traps that can contain ice into June in northern Michigan (Kurta and Smith 2009) and thus are not used by hibernating bats. Indeed, more than 85% of all bats in the mines overwinter in sites with a maximum air temperature (Fig. 3) greater than the mean annual surface temperature of the region (4.2–5.5 °C; Albert et al. 1986), suggesting that presence of warm-air traps, not cold-air traps, may be a factor in selection of hibernacula by bats in the Upper Peninsula. Number and species of bats Eastern pipistrelles. Eastern Pipistrelles were detected at only 28% of 82 occupied mines in Michigan, which contrasted with southern portions of the Midwest, where this species occurred in a greater percentage of hibernacula than any other species (Brack et al. 2003, Dixon 2011, Hall 1962). Throughout their range, Eastern Pipistrelles most frequently hibernated in sites with ambient temperature between 8 and 12 °C (Brack 2007, McNab 1974, Rabinowitz 1981, Raesly and Gates 1987), but temperatures of 8 °C or higher, occurred in only 1/3 of the mines in Michigan (Fig. 3). In more southern locations, the Eastern Pipistrelle often was the only species found in short (less than 50 m) hibernacula, which did not cool sufficiently at those latitudes for use by other species (Brack 2007, Dixon 2011). Most short mines in the Upper Peninsula, however, appeared too cold and variable in temperature (Table 1) for successful hibernation by this species, and the Eastern Pipistrelle was never the only species found in any mine in Michigan. The western Great Lakes region has few natural caves (Culver 1999), and the Eastern Pipistrelle likely was not present before mines were dug and abandoned, thereby providing the sites needed for hibernation (Kurta 1995, Kurta et al. 2007, Stones and Haber 1965). A recent arrival and cool available temperatures may explain the low proportion of occupied sites and the small number of Eastern Pipistrelles per mine, compared to other states in the core of the range of the species (Brack et al. 2003, Turner et al. 2011). Big Brown Bats. The Big Brown Bat roosted in small numbers in many hibernacula, most of which were shared with Myotis. These bats accounted for less than 1% of the bats that were counted, although they likely were more abundant than underground surveys indicated because most Big Brown Bats probably do not migrate to the mines. Many Big Brown Bats hibernate in the walls or attics of heated buildings (Whitaker and Gummer 2002), and some likely overwinter in deep rock crevices (Neubaum et al. 2006). Northeastern Naturalist 600 A. Kurta and S.M. Smith 2014 Vol. 21, No. 4 Little Brown and Northern Bats. Bats in the genus Myotis accounted for more than 99% of bats hibernating in the mines and were widely distributed, with a few individuals in many sites and most animals concentrated in a few, long, complex mines. Our analysis, using negative binomial regression, indicated that maximum ambient temperature and presence of standing water likely had a positive effect on the number of Myotis that was present. Temperature greatly affects energetic expenditure, and its importance for hibernating bats in the field has been implicated before (e.g., Gates et al. 1984). We used maximum recorded temperature in a mine, because all lower temperatures presumably are available in other parts of the site. Bats in varying physical condition may choose different temperatures at which to hibernate (Boyles et al. 2007), and the energetic cost of different metabolic activities during hibernation, such as maintenance during torpor and functioning of the immune system during an arousal, may be minimized at different temperatures (Boyles and McKechnie 2010). Presence of standing water has a number of potential benefits. During periodic arousals, bats often drink (Boyles et al. 2006), and they typically do so on the wing (Taylor and Tuttle 2007). Hence, a pool of water potentially enhances the ability of a bat to maintain water balance during the long winter. Standing water also contributes to higher ambient water vapor pressures, which reduce evaporative water loss, and depending on location within a mine, the water may act as a deterrent to would-be predators and disturbance-causing humans. Our study is the first to indicate water vapor pressure deficit as a factor affecting number of Myotis in a hibernaculum (Table 1). This variable is mathematically proportional to the amount of evaporation that occurs when surface temperature of an animal approximates ambient temperature (Hill 1976). Although earlier studies (e.g., Bogdanowicz and Urbańczyk 1983, Kim et al. 2009, Raesly and Gates 1987) suggested that relative humidity is a factor in determining the number of hibernating bats, such conclusions may be spurious, because relative humidity does not reliably predict evaporative loss in environments that vary in air temperature (Kurta 2014). Nevertheless, our results provide statistical support for the concept that ambient moisture is a factor in determining the number of hibernating bats in a mine. Hibernating bats in the Upper Peninsula were dominated by 2 species of Myotis that co-occurred in most mines. During our study, Northern Bats that were identified in hand accounted for 10% of Myotis, which was similar to the 9.2% estimated by Stones (1981), who examined over 18,000 Myotis in 16 discrete sites. The Northern Bat, which is proposed for listing as an endangered species because of white-nose syndrome (USFWS 2013), appeared more common, relative to Little Brown Bats, in Michigan than in the East or the Ohio River Valley (Brack et al. 2003, Trombulak et al. 2001). In New York, for example, Northern Bats represented only 0.2% of the combined number of Northern and Little Brown Bats that were counted before arrival of white-nose syndrome (Turner et al. 2011). Clustering Although both Little Brown and Northern Bats frequently clustered with other bats of their species, the most common “group” for both species contained only Northeastern Naturalist Vol. 21, No. 4 A. Kurta and S.M. Smith 2014 601 1 bat (Fig. 4). For Northern Bats, solitary animals constituted 87% of the groups and 75% of the individuals, and maximum cluster size was only 6. A solitary nature during hibernation seems typical of this bat (e.g., Barbour and Davis 1969, Brack 2007), although the frequency of clustered versus single individuals apparently has not been quantified elsewhere. For Little Brown Bats, single animals represented 52% of groups and 22% of individuals, and the proportion of Little Brown Bats that was solitary was positively correlated with temperature. The large number of solitary Little Brown Bats and small clusters during 2011–2013 in Michigan matches our subjective impression at most locations over 17 winters. For example, the original description (Kurta 1996:21) of the Mead Mine, which contains more than 15,000 bats, indicated that “bats were spread evenly throughout the [mine], with no large concentrations, and clusters were consistently small, rarely exceeding 10–15 individuals.” The percentage of Little Brown Bats roosting alone in Michigan is much higher than the 1% reported for New York prior to white-nose syndrome (Langwig et al. 2012), but the reason for such a large difference is not clear. Air temperatures for the hibernacula in New York are not published, but it is doubtful whether temperature can explain the difference in clustering behavior between states. Even at our coldest site (2 °C) during the survey of clusters, 10% of Little Brown Bats were solitary, and the lowest percentage at any of the 13 mines was 7% (Fig. 5). Future impacts of white-nose syndrome The late arrival of P. destructans in Michigan, compared to other eastern states (Whitenoseyndrome.org 2014), likely is attributable to the obstacles presented by the Great Lakes to the east–west movement of bats and isolation of the copper and iron ranges from areas of natural caves (hibernacula) in the East and the Ohio River Valley (Culver 1999). Unfortunately, many factors suggest that the disease will be devastating in the next few years. For example, high levels of ambient moisture and temperatures of 4–10 °C in most hibernacula (Fig. 3) are conducive to growth of the fungus (Verant et al. 2012), and the severe and prolonged winters make it unlikely that afflicted animals can leave the hibernaculum and find food. Large colonies may be at particular risk from this disease (Wilder et al. 2011), and although hibernating bats are found in at least 82 sites in the western Upper Peninsula, most bats are concentrated in a few mines (Fig. 2). Degree of sociality within sites also may impact the severity of the infection and the ability of a species to persist after introduction of the fungus (Langwig et al. 2012). In New York, Little Brown Bats reduced their propensity to cluster after introduction of P. destructans, with single animals representing 1% of hibernating animals before the disease and 45% afterwards (Langwig et al. 2012). The much greater occurrence of solitary Little Brown Bats in Michigan prior to arrival of the fungus (22%) may slow the spread of the disease, but ultimately, it seems inevitable that white-nose syndrome will envelope the Great Lakes region and the continent (USFWS 2013). Northeastern Naturalist 602 A. Kurta and S.M. Smith 2014 Vol. 21, No. 4 Acknowledgments We thank B. Scullon, Michigan Department of Natural Resources, for consistent logistical support and for bringing an awareness of flying mammals to a traditional management agency. B. Turin, mine inspector for Ontonagon County, provided leads on the location of many sites. M. Kurta, C. Rockey, and R. Slider offered field assistance at multiple sites, and D. Derrick and D. Sabel often supplied technical support at vertical entrances. We are grateful to the numerous owners and managers of property and mineral rights, especially R. Whiteman, who allowed access to mines. The Michigan Department of Natural Resources provided primary funding, with additional support from the Ottawa National Forest, Bat Conservation International, and our own pockets. The senior author thanks A. Rodríguez-Durán for access to the field station at Mata de Plátano, Arecibo, Puerto Rico, where many ideas were formulated and analyses completed, while on sabbatical from Eastern Michigan University. Literature Cited Albert, D.A., S.R. Denton, and B.V. Barnes. 1986. Regional landscape ecosystems of Michigan. University of Michigan, School of Natural Resources, Ann Arbor, MI. 32 pp. Anderson, D.R. 2008. Model Based Inference in the Life Sciences: A Primer on Evidence. Springer, New York, NY. 184 pp. Barbour, R.W., and W.H. Davis. 1969. Bats of America. University Press of Kentucky, Lexington, KY. 286 pp. Bogdanowicz, W, and Z. Urbańczyk. 1983. Some ecological aspects of bats hibernating in the city of Poznań. Acta Theriologica 28:371–385. Boyles, J.G., and A.E. McKechnie. 2010. Energy conservation in hibernating endotherms: Why “suboptimal” temperatures are optimal. Ecological Modeling 221:1644–1647. Boyles, J.G., M.B. Dunbar, and J.O. Whitaker, Jr. 2006. Activity following arousal in winter in North American vespertilionid bats. Mammal Review 36:267–280. Boyles, J.G., M.B. Dunbar, J.J. Storm, and V. Brack, Jr. 2007. Energy availability influences microclimate selection of hibernating bats. Journal of Experimental Biology 210:4345–4350. Brack, V., Jr. 2007. Temperatures and locations used by hibernating bats, including Myotis sodalis (Indiana Bat), in a limestone mine: Implications for conservation and management. Environmental Management 40:739–746. Brack, V., Jr., S.A. Johnson, and R.K. Dunlap. 2003. Wintering populations of bats in Indiana, with emphasis on the endangered Indiana Myotis, Myotis sodalis. Proceedings of the Indiana Academy of Science 112:61–74. Bricklin, R.B., A. Kurta, S.M. Smith, and W.H. Scullon. 2007. Long-term philopatry of a hibernating Eastern Pipistrelle (Pipistrellus subflavus). Michigan Birds and Natural History 14:183–185. Butler, B.S., and W.S. Burbank. 1929. Copper deposits of Michigan. US Geological Survey Professional Publication 124:1–238. Coxe, S., S.G. West, and L.S. Aiken. 2009. The analysis of count data: A gentle introduction to Poisson regression and its alternatives. Journal of Personality Assessment 91:121–136. Culver, D.C. 1999. Distribution map of caves and cave animals in the United States. Journal of Cave and Karst Studies 61:139–140. Dixon, J.W. 2011. The role of small caves as bat hibernacula in Iowa. Journal of Cave and Karst Studies 73:21–27. Northeastern Naturalist Vol. 21, No. 4 A. Kurta and S.M. Smith 2014 603 Domínguez-Villar, D., I.J. Fairchild, A. Baker, R.M. Carrasco, and J. Pedraza. 2013. Reconstruction of cave air temperature based on surface-atmosphere temperature and vegetation changes: Implications for speleothem palaeoclimate records. Earth and Planetary Science Letters 369–370:158–168. Gardner, W., E. P. Mulvey, and E. C. Shaw. 1995. Regression analyses of counts and rates: Poisson, overdispersed Poisson, and negative binomial models. Psychological Bulletin 1198:392–404. Gates, J.E., G.A. Feldhamer, L.A. Griffith, and R.L. Raesly. 1984. Status of cave-dwelling bats in Maryland: Importance of marginal habitats. Wildlife Society Bulletin 12:162–169. Hall, J.S. 1962. A life-history and taxonomic study of the Indiana Bat, Myotis sodalis. Reading Public Museum and Art Gallery Publications 12:1–58. Hayes, M.A. 2013. Bats killed in large numbers at United States wind-energy facilities. BioScience 63:975–979. Hayes, M.A., R.A. Schorr, and K.W. Navo. 2011. Hibernacula selection by Townsend’s Big-Eared Bat in southwestern Colorado. Journal of Wildlife Management 75:137–143. Hill, R.W. 1976. Comparative Physiology of Animals: An Environmental Approach. Harper and Row, New York, NY. 656 pp. IBM Corporation 2012. IBM SPSS Statistics 21 brief guide. Available online at ftp://public. dhe.ibm.com/software/analytics/spss/documentation/statistics/21.0/en/client/Manuals/ IBM_SPSS_Statistics_Brief_Guide.pdf. Accessed 15 January 2014. Keen, R.A. 1993. Michigan Weather. American and World Geographic Publishers, Helena, MT. 142 pp. Kim, S., Y. Choi, B. Kim, and J. Yoo. 2009. The current distribution and habitat preferences of hibernating Myotis formosus in Korea. Journal of Ecology and Field Biology 32:191–195. Kurta, A. 1995. Mammals of the Great Lakes Region. University of Michigan Press, Ann Arbor, MI. 376 pp. Kurta, A. 1996. Preliminary survey of bats hibernating in mines of the Upper Peninsula. Unpubl. Report. Michigan Department of Natural Resources, Nongame Program, Lansing, MI. Kurta, A. 2008. Bats of Michigan. Indiana State University Center for North American Bat Research and Conservation, Terre Haute, IN. 72 pp. Kurta, A. 2014. The misuse of relative humidity in ecological studies of hibernating bats. Acta Chiropterologica 16:249–254. Kurta, A., and S.M. Smith. 2004. New life for old mines: Bat hibernacula in Michigan’s Upper Peninsula. Pp. 47–51, In T. Rea (Ed.). 2004 NSS Convention Guidebook. National Speleological Society, Huntsville, AL. 106 pp. Kurta, A., and S.M. Smith. 2009. Potential habitat for bats in mines of the Norwich Escarpment. Unpubl. Report. Ottawa National Forest, Bessemer, MI. Kurta, A., J. Caryl, and T. Lipps. 1997. Bats and Tippy Dam: Species composition, seasonal use, and environmental parameters. Michigan Academician 29:473–490. Kurta, A., L. Winhold, J. O. Whitaker, Jr., and R. Foster. 2007. Range expansion and changing abundance of the Eastern Pipistrelle (Chiroptera: Vespertilionidae) in the central Great Lakes region. American Midland Naturalist 157:404–411. Langwig, K.E., W.F. Frick, J.T. Bried, A.C. Hicks, T.H. Kunz, and A.M. Kilpatrick. 2012. Sociality, density-dependence, and microclimates determine the persistence of populations suffering from the novel fungal disease, white-nose syndrome. Ecology Letters 15:1050–1057. Northeastern Naturalist 604 A. Kurta and S.M. Smith 2014 Vol. 21, No. 4 Lankton, L.D., and C.K. Hyde. 1982. Old Reliable: An Illustrated History of the Quincy Mining Company. Quincy Mine Hoist Association, Hancock, MI. 159 pp. Long, B.L., A. Kurta, and D.L. Clemans. 2013. Analysis of DNA from feces to identify prey of Big Brown Bats (Eptesicus fuscus) caught in apple orchards. American Midland Naturalist 170:287–297. McAney, K. 1999. Mines as roosting sites for bats: Their potential and protection. Proceedings of the Royal Irish Academy 99B:63–65. McNab, B.K. 1974. The behavior of temperate cave bats in a subtropical environment. Ecology 55:943–958. Münzer, O.M. 2008. Ecology of the Evening Bat, Nycticeius humeralis, in Michigan. Unpubl. M.Sc. Thesis. Eastern Michigan University, Ypsilanti, MI. Neubaum, D.J., T.J. O’Shea, and K.R. Wilson. 2006. Autumn migration and selection of rock crevices as hibernacula by Big Brown Bats in Colorado. Journal of Mammalogy 87:470–479. Nagel, A., and R. Nagel. 1991. How do bats choose optimal temperatures for hibernation? Comparative Biochemistry and Physiology Part A: Physiology 99:323–326. Prather, J.W., and J.T. Briggler. 2002. Use of small caves by Ozark Big-eared Bats (Corynorhinus tonsendii ingens) in Arkansas. American Midland Naturalist 148:193–197. Rabinowitz, A. 1981. Thermal preference of the Eastern Pipistrelle Bat (Pipistrellus subflavus) during hibernation. Journal of the Tennessee Academy of Science 56:113–114. Raesly, R.L., and J.E. Gates. 1987. Winter habitat selection by north temperate cave bats. American Midland Naturalist 118:15–31. Reed, R.C. 1957. Michigan iron mines. State of Michigan, Department of Conservation, Geological Survey Division, Lansing, MI. 12 pp. Reeder, D.M., C.L. Frank, G.G. Turner, A. Kurta, E.R. Britzke, S.R. Darling, C.W. Stihler, A.C. Hicks, C.U. Meteyer, R. Jacob, L.A. Grieneisen, S.A. Brownlee, M.E. Vodzak, L.K. Muller, and D.S. Blehert. 2012. Frequent arousal from hibernation linked to severity of infection and mortality in bats with white-nose syndrome. PLoS ONE 7(6):e38920. doi:10.1371/journal.pone.0038920. Sherwin, R.E., and J.S. Altenbach. 2009. Managing Abandoned Mines for Bats. Bat Conservation International, Austin, TX. 113 pp. Slagstad, T., K. Midttømme, R.K. Pamstad, and D. Slagstad. 2008. Factors influencing shallow (less than 1000 m depth) temperatures and their significance for extraction of ground-source heat. Geological Survey of Norway, Special Publication 11:99–109. Slider, R.M., and A. Kurta. 2011. Surge tunnels in quarries as potential hibernacula for bats. Northeastern Naturalist 18:378–381. Speakman, J.R., P.I. Webb, and P.A. Racey. 1991. Effects of disturbance on the energy expenditure of hibernating bats. Journal of Applied Ecology 28:1087–1104. Stones, R.C. 1981. Survey of winter bat populations in search of the Indiana Bat in the western Upper Peninsula of Michigan. Unpubl. Report. Michigan Department of Natural Resources, Wildlife Division, Lansing, MI Stones, R.C., and L.P. Branick. 1969. Use of hearing in homing by two species of Myotis bats. Journal of Mammalogy 50:157–160. Stones, R.C., and W. Fritz. 1969. Bat studies in Upper Michigan's copper-mining district. Michigan Academician 2:77–85. Stones, R.C., and G.C. Haber. 1965. Eastern Pipistrelle in Michigan. Journal of Mammalogy 46:608. Stones, R.C., and T. Oldenburg. 1967. Occurrence of torpid Myotis lucifugus in a cold mine in summer. Journal of Mammalogy 49:123. Northeastern Naturalist Vol. 21, No. 4 A. Kurta and S.M. Smith 2014 605 Sullivan, A.R., J.K. Bump. L.A. Kruger, and R.O. Peterson. 2012. Bat-cave catchment areas: Using stable isotopes (δD) to determine the probable origins of hibernating bats. Ecological Applications 22:1428–1434. Tabata, S. 1973. A simple but accurate formula for the saturation vapor pressure over liquid water. Journal of Applied Meteorology 12:1410–1411. Taylor, D.A.R., and M.D. Tuttle. 2007. Water for Wildlife: A Handbook for Ranchers and Range Managers. Bat Conservation International, Austin, TX. 18 pp. Thomas, D.W. 1995. Hibernating bats are sensitive to nontactile human disturbance. Journal of Mammalogy 76:940–946. Trombulak, S.C., P.E., Higuera, and M. DesMeules. 2001. Population trends of wintering bats in Vermont. Northeastern Naturalist 8:51–62. Turner, G.G., D.M. Reeder, and J.T.H. Coleman. 2011. A five-year assessment of mortality and geographic spread of white-nose syndrome in North American bats and a look to the future. Bat Research News 52:13–27. Tuttle, M.D., and J. Kennedy. 2002. Thermal requirements during hibernation. Pp. 68–78, In A. Kurta and J. Kennedy (Eds.). The Indiana Bat: Biology and Management of an Endangered Species. Bat Conservation International, Austin, TX. 253 pp. US Census Bureau. 2014. State and county QuickFacts: Michigan. Available online at http://quickfacts.census.gov/qfd/states/26000.html. Accessed 15 June 2014. US Fish and Wildlife Service (USFWS). 2012. North American bat death toll exceeds 5.5 million from white-nose syndrome. Available online at http://www.fws.gov/ whitenosesyndrome/pdf/wns_mortality_2012_nr_final.pdf. Accessed 15 June 2014. USFWS. 2013. US Fish and Wildlife Service proposes endangered status for the Northern Long-eared Bat; Listing not warranted for eastern Small-footed Bat. Available online at http://www.fws.gov/midwest/endangered/mammals/nlba/pdf/ FRpropListNLBA2Oct2013.pdf. Accessed 15 June 2014. Verant, M.L., J.G. Boyles, W. Waldrep, Jr., G. Wibbelt, and D.S. Blehert. 2012. Temperature- dependent growth of Geomyces destructans, the fungus that causes bat white-nose syndrome. PLoS ONE 7:e46280. Whitaker, J.O., Jr., and S.L. Gummer. 1992. Hibernation of the Big Brown Bat, Eptesicus fuscus, in buildings. Journal of Mammalogy 73:312–316. Wilder, A.P., W.F. Frick, K.E. Langwig, and T.H. Kunz. 2011. Risk factors associated with mortality from white-nose syndrome among hibernating bat colonies. Biology Letters 7:950–953. White-nosesyndrome.org. 2014. Where is it now? Available online at: http://www.whitenosesyndrome. org/about/where-is-it-now. Accessed 15 June 2014.