Year-to-year Reuse of Tree-roosts by California Bats (Myotis californicus) in Southern British Columbia

Transcription

1 Am. Midl. Nat. 146:80 85 Year-to-year Reuse of Tree-roosts by California Bats (Myotis californicus) in Southern British Columbia ROBERT M. R. BARCLAY 1 Department of Biological Sciences, University of Calgary, Calgary, AB T2N 1N4 Canada AND R. MARK BRIGHAM Department of Biology, University of Regina, Regina, SK S4S 0A2 Canada ABSTRACT. To document year-to-year reuse of roost trees by forest-dwelling bats we monitored trees in southern British Columbia that we first identified as maternity roosts of California bats (Myotis californicus) in Initially we identified roost trees by tracking radiotagged individuals. Then we revisited each tree in subsequent years up to At the start of the study the bats roosted under loose bark or in cavities in dead trees. Seven of eight trees were still standing in 2000, although all had lost bark since 1995, particularly ponderosa pines (Pinus ponderosa). In 1995, after radio-tagged bats had moved and the tags had fallen off, trees either were occupied by colonies of 5 to 52 M. californicus or they were unoccupied. In contrast, roost counts in subsequent years indicated that colonies rarely used the same trees and most observations were of one or two bats. Thus, while bats continued to use most of the trees over the 5 y period, the numbers of individuals declined and much of the use may have been by males or non-reproductive females. Although our study is preliminary, the results suggest that the suitability of roosts of tree-dwelling bats declines relatively rapidly compared to the loss of the snags themselves. More intensive studies are required given the current focus on preserving roosting habitat for forest-dwelling bats. INTRODUCTION There has been considerable recent interest in roosting behavior and roost selection of forest-dwelling bats in North America (Barclay and Brigham, 1996; Campbell et al., 1996; Vonhof and Barclay, 1996; Betts, 1998; Crampton and Barclay, 1998; Kalcounis and Brigham, 1998; Rabe et al., 1998; Waldien et al., 2000) and elsewhere (Lunney et al., 1988; Hosken, 1996; O Donnell and Sedgeley, 1999; Sedgeley and O Donnell, 1999;). This stems, in part, from the realization that natural roosts in trees offer characteristics different from those in man-made structures such as buildings and, thus, may be instructive in explaining roosting behavior. In addition, understanding the selection and dynamics of tree roosts is important from a conservation perspective in light of forest harvesting (Crampton and Barclay, 1998; Sedgeley and O Donnell, 1999; Waldien et al., 2000). Although the specifics of the roosts selected by reproductive female bats may differ between species (e.g., some prefer loose bark, others prefer cavities; e.g., Brigham et al., 1997; Kalcounis and Brigham, 1998), many features of the trees themselves are similar across species and forest types. Roost trees typically are found in open areas of older forest stands, are large, and in early to mid decay stages (e.g., Vonhof and Barclay, 1996; Betts, 1998; Crampton and Barclay, 1998; O Donnell and Sedgeley, 1999). In several species of forest-dwelling bats, individuals and entire groups frequently switch roosts (Vonhof and Barclay, 1996; Crampton and Barclay, 1998; O Donnell and Sedgeley, 1999). This is very different from the roost fidelity of bats that occupy buildings or caves 1 Corresponding author: Telephone (403) ; FAX (403) ; barclay@ucalgary.ca 80

2 2001 BARCLAY AND BRIGHAM: REUSE OF TREE ROOSTS 81 (Lewis, 1995). Various reasons have been proposed to account for such movements (Lewis, 1995), but one consequence is that conservation of roosting habitat in forests requires protection at the stand level as opposed to individual trees (Crampton and Barclay, 1998; Sedgeley and O Donnell, 1999). Although there is considerable information regarding the shorter term (weeks or months) use of specific roosts and trees by a variety of bat species, virtually nothing is known regarding longer term use (year to year) or the long-term persistence and suitability of roost trees. Lewis (1995) suggested that roost switching by tree-roosting bats is related to the relatively ephemeral nature of tree roosts compared with buildings or caves. However, O Donnell and Sedgeley (1999) found that Chalinolobus tuberculatus in New Zealand rarely reused roosts within or between years, despite using tree cavities the authors perceived to remain suitable for decades. While individual trees may stand for decades (Keen, 1955; Cline et al., 1980; Sedgwick and Knopf, 1992), the actual suitability of cavities may decline more rapidly, depending on the specific features bats are selecting. The purpose of our study was to initiate longer term monitoring of roosts initially occupied by reproductive female California bats (Myotis californicus) in southern British Columbia. Our study was relatively short (5 y) and, thus, was aimed primarily at determining whether roost trees continue to be used by maternity colonies from year to year. In addition, we monitored the persistence of roosts and roost trees over the medium-term. Myotis californicus is a small (4 6 g), insectivorous vespertilionid bat that ranges from southern Alaska to southern Mexico through much of western North America (Simpson, 1993). Females form maternity colonies which, in forested regions, occur under loose bark or in tree cavities such as those created by woodpeckers (Krutzch, 1954; Brigham et al., 1997, pers. obs.). In our study area the trees selected by reproductive females typically are snags of large ponderosa pines (Pinus ponderosa) or Douglas firs (Pseudotsuga menziesii) in intermediate stages of decay and in relatively open areas (Brigham et al., 1997). Individuals and entire colonies frequently switch roosts, even during lactation, so that single observations at a roost may not indicate whether it is used in a particular season (Brigham et al., 1997). METHODS This study took place in the West Arm Demonstration Forest, near Nelson, British Columbia (49 17 N, W). We originally located roosts of female Myotis californicus in July 1995 (Brigham et al., 1997), and subsequently observed the same roost trees during the same period in July of 1996 and No data were collected in Eight trees (four ponderosa pines, three Douglas firs and one western white pine (Pinus monticola)), found in tracts of undisturbed (for at least 5 y prior to, and during the course of the study) forest were occupied by colonies of reproductive female M. californicus. These were originally located in 1995 via radio-telemetry (see Brigham et al., 1997 for details). We made exit counts at each roost tree to determine colony size. After each radio-tagged bat had moved to another roost, or the transmitter had fallen off the bat, we conducted further exit counts at the same trees to determine the consistency of use and colony size. Only these later counts were compared to counts made in subsequent years as counts made while radiotagged bats were present would bias the data. For each roost tree we also measured or estimated various tree and site characteristics, including tree diameter and height, percent of the trunk covered in bark, distance to neighboring trees, stem density and canopy height (Brigham et al., 1997). Each year we took photographs of each roost tree to allow for a qualitative assessment of changes in tree height, bark and branches. In 2000 we each directly

3 82 THE AMERICAN MIDLAND NATURALIST 146(1) estimated the percent bark remaining and averaged the values. Throughout, we present means ( SE) and used an level of 0.05 for all statistical analyses. In July 1996, and we returned to the eight roost trees and made exit counts at dusk. All counts in all years were made in the same 2-wk period at the beginning of July, a time when Myotis californicus maternity colonies exist. We made one to four counts per tree per year and four to ten counts per tree over the four years of study. Counts began at least 30 mins before bats were expected to exit (approximately 10 mins after sunset) and continued for at least 10 mins after the last bat left, or until 1 h after sunset. Thus, we typically watched trees for at least one hour. Trees were watched in a random order, but all trees were watched at least once in a year before a subsequent count was conducted. While conducting exit counts, we were silent and as far from the tree as possible so as to minimize disturbance to the roosting bats. We visually identified the bats emerging from the roosts and used a Pettersson D-100 ultrasonic detector to monitor echolocation calls produced by the emerging bats. We tuned the detector to 45 khz, a frequency which detects the calls of Myotis spp. in the study area. Myotis californicus is the smallest species of bat in the study area and has a distinctive fluttering flight style. When bats exited from the roost trees, we identified them as M. californicus on the basis of their size, flight style and echolocation calls. It is possible, however, that some of the bats may have belonged to other species of Myotis. RESULTS Of the eight roost trees identified in 1995, seven remained standing and were approximately the same height in July 2000, although some had lost a few major branches. One Douglas fir fell between July 1996 and July In 1995 this tree had been occupied by the largest colony we observed (52 individuals), although no bats emerged from it during three emergence counts in Myotis californicus roosted primarily under loose bark in ponderosa pine and Douglas fir when the colonies were first located in 1995 (Brigham et al., 1997). Bats roosted under loose bark on six of the eight trees and we could not determine the exact roost location for the other two. In one instance an old woodpecker hole was used in addition to loose bark. By 2000 all seven remaining roost trees had less bark than in This loss averaged 27.6% ( 9.6 SE; range, %) of the bark present in Four ponderosa pines lost % (mean 42.9%), whereas less was lost from two Douglas firs (5 and 5.6%). By 2000 the ponderosa pines had between 5 and 68% of their bark remaining, whereas one western white pine (Pinus monticola) and the two Douglas firs had 80 95% left. On four occasions we observed one to three big brown bats (Eptesicus fuscus) emerging from a roost tree. On two of those occasions, bats we identified as Myotis californicus also emerged from the tree, but from a different specific roost. We excluded big brown bats from our analyses. Fewer Myotis californicus used the roost trees in the years after the initial colonies were located (Fig. 1). In 1995 counts conducted on days after the radio-tagged bats had left a roost revealed either no bats, or colonies of 5 34 individuals. We never observed fewer than five individuals if bats were present (n 7 counts). In subsequent years the proportion of exit counts that revealed no bats (21 of 45 counts at 8 trees, 46.7%) was no different than that in 1995 (4 of 11 counts at 6 trees, 36.4%; Yates corrected , P 0.7), but one or two individuals were commonly observed (15 of 24 counts with bats). The same was true if we included only the six trees for which we had data in all years (Fig. 1). Mean colony size for trees, when bats were present, was significantly less in (x , n 18 counts) than in 1995 (x , n 7; t-test on log-transformed data, t 3.79,

4 2001 BARCLAY AND BRIGHAM: REUSE OF TREE ROOSTS 83 FIG. 1. Comparison of the number of Myotis californicus occupying six roost trees in 1995 and in subsequent years. n is the number of counts df 23, P 0.001). In one ponderosa pine, colony size in 1996 (8) and 1999 (5) was approximately the same as it had been in 1995 (6). In another ponderosa pine, colony size was 27 in 1995, 25 in 1996, but no larger than 10 after that. This roost always had bats in it after 1995 (x 9.6, n 5 counts), and the lowest number of bats (1) occurred when a northern flying squirrel (Glaucomys sabrinus) was also observed exiting from the tree at dusk. Two other ponderosa pines rarely had any bats in them (e.g., 0 of 4 counts, 4 of 10 counts). There was no obvious relationship between the decline in bat numbers and loss of bark, although we had insufficient sample size to test this statistically. DISCUSSION Most of the snags used by maternity colonies of Myotis californicus in 1995 survived through July 2000, which is not surprising given the relatively long lifespan of snags of the two main tree species used by the bats. Ponderosa pine snags can last for over 40 y (Keen, 1955) and Douglas fir snags also can remain in a state potentially suitable for bats for over 40 y (Cline et al., 1980). A similar proportion of the roost trees were occupied by bats in the initial year compared to subsequent years. The difference was that whereas colonies of 4 to 52 bats occupied the trees in the first year, one to two individuals were most common in subsequent years and larger colonies were rare. We suggest that these single bats and pairs were likely males or nonreproductive females. All of the reproductive females we tagged in 1995 roosted in colonies of at least five individuals and other studies of Myotis californicus have found the same (Krutzsch, 1954; Simpson, 1993). In contrast, in our study area two radio-tagged males roosted alone (pers. obs.) Males and nonreproductive females of other colonial species also tend to roost solitarily or in small groups (e.g., Hamilton and Barclay, 1994; Vonhof and Barclay, 1997; O Donnell and Sedgeley, 1999). The difference in roosting behavior is likely related to differences in the use, and costs and benefits of torpor (Hamilton and Barclay, 1994). There are several possible explanations for reduced use of the roost trees by maternity groups of Myotis californicus. The most plausible explanation is that the suitability of the roosts declined over the study. This seems likely, as the colonies originally roosted primarily

5 84 THE AMERICAN MIDLAND NATURALIST 146(1) under bark and all of the trees lost bark during the study. This would not only reduce the availability of potential roosts under bark, but also may have altered the thermal characteristics of other cavities in the trunk itself by removing insulation (Nicolai, 1986). Thermal properties of roosts are likely one of the main characteristics influencing the suitability and selection of roosts by bats (Kurta et al., 1993; Hamilton and Barclay, 1994; Vonhof and Barclay, 1997; Chruszcz, 1999). Another possible explanation for the decline in use by colonies is that switching roosts is beneficial not only within years, but also between years, and that suitable roosts are plentiful. An abundance of roosts was suggested as the reason why Chalinolobus tuberculatus in New Zealand rarely reused roosts (Sedgeley and O Donnell, 1999), and other studies have found low rates of roost reuse between years (e.g., Betts, 1998). However, these studies followed radio-tagged bats rather than revisiting previously used trees, and nontagged or previously tagged bats may well have reused trees between years. Reproductive bats of various species are selective in the roosts they choose (e.g., Vonhof and Barclay, 1996; Betts, 1998; Kalcounis and Brigham, 1998; Sedgeley and O Donnell, 1999), and reusing trees from year to year, as individuals that roost in buildings or caves do, has benefits (e.g., reduced search costs; Lewis, 1995). However, if suitable roosts are abundant, and benefits of switching (e.g., reduced predation risk or ectoparasitism; Lewis, 1995) apply between years as well as within, then reuse between years may not be favored. It could be that the overall population of Myotis californicus declined in the study area over the course of the study. We have no direct evidence of this, and to the contrary, throughout the study we observed many bats, including what we identified as M. californicus, at dusk in the vicinity of the roost trees. Finally, it could be that the roost trees we observed were used by colonies of Myotis californicus, but not on the days we observed them. While this may explain the lack of bats at a given roost on a particular day, we monitored most roosts more than once each year and we do not believe it explains the overall shift from use by large colonies to single individuals and small groups. The proportion of observations when no bats were present did not change and if we had simply missed seeing the colonies, we would expect a similar distribution of group sizes to that observed in Our confidence in this conclusion is bolstered by the fact that the data were collected in the same way during the baseline (1995) year as for the other study years. However, our sample size (both number of trees and nights) was relatively small. This coupled with the fact that radio-tagged M. californicus switch roosts relatively often, means that we cannot completely rule out that we missed seeing larger groups of bats use the trees. If our hypothesis is correct, and roosts used by maternity colonies of Myotis californicus are ephemeral and last for considerably shorter times than do the snags themselves, this has important implications for the management of bat populations in forests. It means that large numbers of snags must be maintained and a continual supply of new snags must be recruited. Clearly, more long-term intensive studies on various species of bats are required to adequately test our hypothesis and determine whether it applies generally to the roosts of other species in different forest types. Acknowledgments. We thank the many people who braved the bears and mosquitoes to watch trees which often had no bats in them. The following were particularly helpful: G. Barclay, T. Brigham, L. Barclay, K. Brigham, R. Barclay, A. Brigham, M. Anderson, H. Pinnell, M. Vonhof, J. Morissette and M. Firman. The comments of B. Betts, D. Nagorsen and an anonymous reviewer improved earlier drafts of the manuscript. Funding was provided by research grants from the Natural Sciences and Engineering Research Council of Canada to RMRB and RMB.

6 2001 BARCLAY AND BRIGHAM: REUSE OF TREE ROOSTS 85 LITERATURE CITED BARCLAY, R. M. R. AND R. M. BRIGHAM (EDS.) Bats and forests symposium, October 19 21, 1995, Victoria, British Columbia, Canada. Research Br., B. C. Min. For., Victoria, B. C. Working Paper 23/ p. BETTS, B. J Roosts used by maternity colonies of silver-haired bats in northeastern Oregon. J. Mammal., 79: BRIGHAM, R. M., M. J. VONHOF, R. M. R. BARCLAY AND J. C. GWILLIAM Roosting behavior and roost-site preferences of forest-dwelling California bats (Myotis californicus). J. Mammal., 78: CAMPBELL, L. A., J. G. HALLETT AND M. A. O CONNELL Conservation of bats in managed forests: use of roosts by Lasionycteris noctivagans. J. Mammal., 77: CLINE, S. P., A. B. BERG AND H. M. WIGHT Snag characteristics and dynamics in Douglas fir forests, western Oregon. J. Wildl. Manage., 44: CRAMPTON, L.H.AND R. M. R. BARCLAY Selection of roosting and foraging habitat by bats in different-aged aspen mixedwood stands. Cons. Bio., 12: CHRUSZCZ, B Foraging and thermoregulatory behaviour of the long-eared bat (Myotis evotis) roosting in natural habitat. Unpublished M.Sc. Thesis, University of Calgary. 100 p. HAMILTON, I. M. AND R. M. R. BARCLAY Patterns of daily torpor and day-roost selection by male and female big brown bats (Eptesicus fuscus). Can. J. Zool., 72: HOSKEN, D. J Roost selection by the lesser long-eared bat, Nyctophilus geoffroyi, and the greater long-eared bat, N. major (Chiroptera: Vespertilionidae) in Banksia woodlands. J. Royal Soc. W. Australia, 79: KALCOUNIS, M. C. AND R. M. BRIGHAM Secondary use of aspen cavities by tree-roosting big brown bats. J. Wildl. Manage., 62: KEEN, F. P The rate of natural falling of beetle-killed ponderosa pine snags. J. Forestry, 53: KRUTZCH, P. H Notes on the habits of the bat, Myotis californicus. J. Mammal., 35: KURTA, A., J. KATH, E.L.SMITH, R.FOSTER, M.ORICK AND R. ROSS A maternity roost of the endangered Indiana bat (Myotis sodalis) in an unshaded, hollow sycamore tree (Platanus occidentalis). Am. Midl. Nat., 130: LEWIS, S. E Roost fidelity of bats: a review. J. Mammal., 76: LUNNEY, D., J. BARKER, D. PRIDDEL AND M. O CONNELL Roost selection by Gould s long-eared bat, Nyctophilus gouldi Tomes (Chiroptera: Vespertilionidae), in logged forest on the south coast of New South Wales. Australian Wildl. Res., 15: NICOLAI, V The bark of trees: thermal properties, microclimate and fauna. Oecologia, 69: O DONNELL, C. F. J. AND J. A. SEDGELEY Use of roosts by the long-tailed bat, Chalinolobus tuberculatus, in temperate rainforest in New Zealand. J. Mammal., 80: RABE, M. J., T. E. MORRELL, H. GREEN, J. C. DEVOS, JR. AND C. R. MILLER Characteristics of ponderosa pine snag roosts used by reproductive bats in northern Arizona. J. Wildl. Manage., 62: SEDGELEY, J. A. AND C. F. J. O DONNELL Roost selection by the long-tailed bat, Chalinolobus tuberculatus, in temperate New Zealand rainforest and its implications for the conservation of bats in managed forests. Biol. Cons., 88: SEDGWICK, J. A. AND F. L. KNOPF Cavity turnover and equilibrium cavity densities in a cottonwood bottomland. J. Wildl. Manage., 56: SIMPSON, M. R Myotis californicus. Mamm. Sp., 428: 1 4. VONHOF, M. J. AND R. M. R. BARCLAY Roost-site selection and roosting ecology of forest-dwelling bats in southern British Columbia. Can. J. Zool., 74: , AND Use of tree stumps as roosts by the western long-eared bat. J. Wildl. Manage., 61: WALDIEN, D. L., J. P. HAYES AND E. B. ARNETT Day-roosts of female long-eared Myotis in western Oregon. J. Wildl. Manage., 64: SUBMITTED 5 SEPTEMBER 2000 ACCEPTED 21 FEBRUARY 2001