2010 NORTHEASTERN NATURALIST 17(4):531–540 Use of Hair Tubes to Detect Small-Mammal Winter Activity in a Northern Forest Habitat Hollee Schwingel1,2 and Christopher Norment1,* Abstract - We used hair tubes to examine the winter activity of small mammals in relation to coarse woody debris and snow cover in a deciduous forest woodlot in western New York State during the winters of 2006 and 2007. Hairs trapped on sticky tapes in baited hair tubes were used to identify mammal species. During the winter of 2006, a higher proportion of baits was removed from hair tubes sites 2 m away from logs, relative to hair tubes adjacent to logs, while significantly more baits were removed from hair tubes adjacent to logs in 2007. A significantly greater proportion of baits was removed from log and open sites combined in 2006 than in 2007. However, in 2007, a higher proportion of hair tubes with baits removed also had mammal hairs on their tapes. The higher proportion of baits removed from hair tubes in 2006, and the increased proportion of baits removed from sites near logs in 2007, was most likely due to reduced snow cover in 2006. Hair tubes offer several advantages relative to live traps in the study of winter small-mammal activity. They are simple to use, do not require handling animals, eliminate trap-related mortality, and can be used at any spatial scale, including the landscape level. Finally, they employ low-cost materials, and can be adapted to a range of conditions and target species. Introduction Relatively little is known about small-mammal winter ecology, especially where low air temperatures and deep snow make research difficult and may decrease survival of captured animals (Formozov 1946, Halfpenny and Ozanne 1989, Pruitt 1984). In addition, some small mammals either enter torpor (Pierce and Vogt 1993) or decrease activity (Martin 1983) during bouts of cold weather. Small mammals using the subnivean space, the space between the bottom of the snowpack and the ground (Pruitt 1984), are hidden from researchers by the snow these animals travel under (Schmid 1984). In the Northeast, relatively few studies have analyzed the winter activity of small mammals, particularly in areas with long-lasting snow cover (but see Merritt 1986, Webster and Brooks 1981, Wolff and Durr 1986, Zeggers and Merritt 1988). Most studies of small-mammal winter activity in the Northeast have used live traps (e.g., Merritt 1986, Webster and Brooks 1981, Wolff and Durr 1986). However, in the winter, live traps must provide an appropriate microclimate for trapped animals. They 1Department of Environmental Science and Biology, College at Brockport, SUNY, Brockport, NY 14420. 2Current address - 35740 Highway 17 North, Coulee City, WA 99115. *Corresponding author - cnorment@brockport.edu. 532 Northeastern Naturalist Vol. 17, No. 4 must be checked frequently to prevent mortality, and it may be difficult and labor-intensive to keep captured small mammals alive during intense cold (Buech 1974, Merritt 1986, Sanecki and Green 2005, Schmid 1984). The labor-intensive aspects of live-trapping small mammals in areas with persistent snow cover also make it difficult to conduct landscape-scale studies, which require monitoring many traps (Sanecki and Green 2005). Finally, live trapping small mammals can potentially expose handlers to diseases such as hantavirus (ASM 1998). Thus, there is a need for an easy to use, efficient, and safe method for obtaining data on winter-active small mammals. One potentially useful sampling method employs hair tubes, which eliminate trap-related mortality, decrease labor, and reduce the danger of spreading diseases, because mammals are not captured (Pocock and Jennings 2006, Sanecki and Green 2005). Although hair-tube design varies, all use bait to attract target species, which leave hair samples on an adhesive substance placed inside the tube (Mortelliti and Boitani 2008, Pocock and Jennings 2006, Sanecki and Green 2005). Hair samples are identified to species, and potentially to sex and individual. Although hair tubes have been used to study small mammals in non-niveal areas (Baker et al. 2003, Dickman and Doncaster 1987, Gurnell et al. 2004, Laidlaw and Wilson 1989, Mortelliti and Boitani 2008), only recently have they been employed in areas with long-lasting snow cover (Sanecki and Green 2005, Sanecki et al. 2006). The main objective of our study was to evaluate the suitability of hair tubes for studies of winter small-mammal activity in a region of the Northeast that may have long-lasting, but variable, snow cover. We also compared the effectiveness of hair tubes placed in two deciduous forest microhabitats, and evaluated how width of the adhesive used in hair tubes affects their ability to collect hair samples. Methods Study area We conducted our study in a deciduous forest woodlot at the College at Brockport, State University of New York, Monroe County, NY. The forest overstory was composed primarily of Acer saccharum Marshall (Sugar Maple) and Fagus grandifolia Ehrhart (American Beech), with a woody understory of Lindera benzoin (L.) Blume (Spice-bush), Ostrya virginiana (Miller) K. Koch (Hop-hornbeam), and sapling Sugar Maples. Numerous large downed logs were present. Common small mammals included Blarina brevicauda Say (Northern Short-tailed Shrew), Tamias striatus L. (Eastern Chipmunk), Sciurus carolinensis Gmelin (Eastern Gray Squirrel), Glaucomys sabrinus Shaw (Northern Flying Squirrel), and Peromyscus leucopus Rafinesque (White-footed Mouse) (C.J. Norment, unpubl. data.) 2010 H. Schwingel and C. Norment 533 Field methods Sampling occurred from 24 January 2006 to 6 March 2006 and 23 January 2007 to 6 March 2007. Hair-tube design followed Sanecki and Green (2005). Because we wanted to sample all small mammals occurring at our study site, from Northern Flying Squirrels to Northern Short-tailed Shrews, hair tubes were constructed from relatively large (42 mm diameter) PVC elbows, one end of which was blocked by a wooden plug that held the bait in place (Fig. 1). Two holes were drilled into the top of the PVC elbow, 0.5 cm from the edge, and a 20-cm-long piece of 0.05-cm-gauge wire was attached to the elbow through the holes, to allow it to be deployed and retrieved. Bait plugs were made by sawing 2-cm-long pieces of 4-cm-diameter wooden rods and boring a 1.5-cm-diameter hole into one end; a plug was placed in the end of the hair tube closest to the snow surface. Hair tubes were placed at the bottom of snow tubes, which were 7.62-cm-diameter PVC plastic tubing, 77.5 cm in length. Each snow tube was mounted vertically on a 115-cm wooden stake, and attached using plastic zip ties, leaving a 7.5-cm gap between the bottom of the PVC tubing and the ground. Each hair tube was suspended off the ground with 0.15-cm-gauge wire, 130 cm in length, hooked to the 0.05-cm-gauge wire. This allowed hair tubes to be withdrawn from snow tubes without disturbing surrounding snow. Figure 1. Assembled hair tube on left is in the position it would be in at the base of a snow tube. The disassembled hair tube on right shows the PVC elbow and wooden bait plug. 534 Northeastern Naturalist Vol. 17, No. 4 Double-sided carpet tape was placed inside the open end of the hair tube. The 2006 study used tape 1.6 cm wide and 9 cm long (Scotch 3M carpet tape), which covered the bottom half of the hair tube. The 2007 study used tape 3.18 cm wide and 13 cm long (Duck Brand light traffic carpet tape), which covered three-quarters of the top and bottom of the tube. We chose double-sided tape as the best material for hair capture because Sanecki and Green (2005) found that it maintained stickiness throughout 7-d sampling periods during a study of small-mammal distribution in relation to snow cover, even when exposed to temperatures as low as -13 ºC. Bait was made from peanut butter, honey, and oatmeal; one teaspoon-full was placed in the hole in each bait plug. In December 2005 and 2006, we placed 40 hair tubes in the woodlot, before the first snowfall of the season. Two tubes were placed at each of 20 stations, in a four by five grid of about 100 m by 100 m. At each station, one tube was placed next to a downed log ≥17 cm in diameter; the other tube was placed 2 m away, in the open and near no other logs. We conducted eight hair-tube trials, at approximately one-week intervals. During each trial (four in 2006 and four in 2007), we placed a baited and taped hair tube in each snow tube. After placing the hair tubes, we measured thickness of the snow cover and any subnivean space in 10 randomly chosen open areas. Hair tubes remained in the snow tubes for one week before being removed. The tape from each hair tube was removed; we also recorded whether the bait had been removed from the plug. Hair tubes were then rebaited, retaped, and placed in snow tubes for another sampling period. Hair identification We identified small-mammal hairs to species, using the method of Martin et al. (2001). When present, two to three hairs from each piece of carpet tape were removed with forceps and placed on a glass microscope slide. A plastic cover slip was placed over the hairs, and another glass microscope slide placed over the cover slip. The glass microscope slides and plastic cover slip were held over a Bunsen burner flame with crucible tongs. The slides and slip remained over the flame for about 10 s, or until the edges of the cover slip had melted. The slides and cover slip were cooled for 1 min; the plastic cover slip then was removed from between the glass microscope slides. The plastic cover slip was placed on a new microscope slide and the cuticle impressions left by the hairs examined with a compound microscope. Impressions were compared to reference slides of cuticle impressions of guard hairs and underhairs of species in the study area. Length and diameter of hairs, and a key to hairs of Michigan mammals (Mathiak 1938), aided in identification of samples (see also Mayer 1952, Moore 1988, Teerink 1991). Although cuticle impressions and hair measurements may not always be sufficient to distinguish between closely related species (Harris and Yalden 2004, Sanecki and Green 2005), no congeneric species were present in our 2010 H. Schwingel and C. Norment 535 study area, which simplified hair identification. Two people independently examined each hair sample; only those samples for which agreement occurred were considered to have been identified to species. We used χ2 tests to compare frequency distributions of baits removed from hair tubes adjacent to downed logs and those 2 m away from logs, and the proportion of tapes with hair samples in tubes from which from the bait had been removed during the two study seasons. Statistical significance was accepted at P ≤ 0.05. Results In 2006, mean snow depth during the study period was 3.2 cm; maximum recorded depth was 12.7 cm. Seventeen out of 43 days were without snow cover, and a subnivean space never developed. In 2007, average snow depth was 14.7 cm; maximum recorded depth was 27.9 cm, with a minimum depth of 7.6 cm. There were no days without snow cover; a thin subnivean space of ca. 2 cm was maintained throughout the study period, although snow-free spaces up to 15 cm thick were present beneath the shelter of downed logs. Hairs from four small-mammal species were collected from hair tubes at the study site: Northern Short-tailed Shrew, Eastern Chipmunk, Northern Flying Squirrel, and White-footed Mouse; the most frequently detected species was the White-footed Mouse. More hair tubes yielded samples in 2006 than in 2007 (Table 1). Five hair samples were unidentified in 2006, either because the independent examiners could not reach an agreement, or because the material was inadequate for identification. No hair samples were unidentified in 2007 (Table 1). In 2006, the proportion of baits removed from hair tubes was significantly greater in sites 2 m away from logs (0.925; n = 80) than in sites adjacent to logs (0.81; n = 79) (χ2 = 4.57, df = 1, P = 0.03). However, in 2007, the proportion of baits removed from sites adjacent to logs (0.162, n = 80) was significantly greater than the proportion removed from sites 2 m away from logs (0.012, n = 80) (χ2 = 14.46, P < 0.001). The proportion of baits removed from log and open sites combined was significantly greater in 2006 (0.868, n = 159) than in 2007 (0.088, n = 160) (χ2 = 194.7, df = 1, P < 0.001). In 2006, 0.87 of the hair tubes had their bait removed, while only 0.18 of those Table 1. Number of hair samples identified from hair tubes, Brockport, NY, 2006–2007. Species 2006 2007 Blarina brevicauda (Northern Short-tailed Shrew) 3 3 Tamias striatus (Eastern Chipmunk) 5 0 Glaucomys sabrinus (Northern Flying Squirrel) 0 4 Peromyscus leucopus (White-footed Mouse) 12 4 Unidentified 5 0 Total 25 11 536 Northeastern Naturalist Vol. 17, No. 4 hair tubes (n = 138) had hair on the tape. In 2007, only 0.088 of the hair tubes had their bait removed, but 0.79 of those tubes (n = 14) had hair on the tape; this difference was significant (χ2 = 25.7, df = 1, P < 0.001). Discussion The most frequently detected species in our study was the White-footed Mouse, the most abundant small mammal during a 16-yr live trapping study at the same site (C.J. Norment, unpubl. data). All species detected with hair tubes remain surface-active during the winter (Northern Short-tailed Shrew [Merritt 1986], Eastern Chipmunk [French 2000], Northern Flying Squirrel [Whitaker and Hamilton 1998], and White-footed Mouse [Pierce and Vogt 1993]), although some individuals may enter torpor during the coldest part of the winter (e.g., Pierce and Vogt 1993). The number of hair samples obtained with hair tubes, and the number with baits removed, was greater in 2006 than in 2007; the only species more frequently detected in 2007 was the Northern Flying Squirrel. The disparity between the proportion of baits removed from hair tubes in 2006 and 2007, and the relatively high proportion of baits removed from hair tubes placed adjacent to logs in 2007, as compared to those 2 m away from logs, can be attributed to different snowfall patterns during 2006 and 2007. From late January through early March of 2006, there was no snow cover, or only a temporary, thin layer, present at our study area; most likely, snow did not serve as a barrier to small-mammal movement. However, from late January through early March of 2007, snow cover was continuous and never decreased to less than 7.6 cm. Thus, small mammals could access hair tubes, both those adjacent to logs and those in open areas, more easily in 2006. The deeper, long-lasting snow cover in 2007 most likely restricted access to hair tubes, especially those in open areas, hence the increased rate of bait removal from hair tubes adjacent to logs, which supported a larger snow-free space. Researchers elsewhere have found that coarse woody debris is important in structuring the subnivean space and providing connectivity between subnivean and supranivean environments (Sanecki et al. 2006, Sherburne and Bissonette 1994). Even though a thin subnivean space of ca. 2 cm formed in 2007, this space was probably too small for most winter-active small mammals to use (Sanecki et al. 2006). An alternative explanation for the greater number of baits removed from hair tubes in 2006 could be that small-mammal populations were higher during this sampling period. We have no data to test this explanation directly. However, trap success (number of mice captured/ number of traps set) for White-footed Mouse at the study site were similar in May 2006 (0.120) and May 2007 (0.134), suggesting that, at least for the species most frequently detected by hair tubes, population size did not differ substantially between winters (C. Norment, unpubl. data). Also, 2010 H. Schwingel and C. Norment 537 differences in small-mammal population size between years would not have affected within-year differences in bait removal from sites adjacent to and distant from logs. Greater tape effectiveness during 2007 (0.79) was likely due to tape width and length, and tape placement. The wider and longer tape used during the 2007 winter season, which covered a greater proportion of the tube, probably was more effective at removing hairs from small mammals that entered the hair tubes than tape used in 2006. However, hair tubes in our study were less effective than in the study by Sanecki and Green (2005), who used the same hair tube design to examine small-mammal distribution in relation to snow cover in the Snowy Mountains, Australia. The reason for the lower effectiveness in our study is unclear, as Sanecki and Green (2005) placed double-sided tape only in the “upper inside surface” of the opening. One possible reason is that the common mammal species studied by Sanecki and Green (2005) were larger (average mass range = 35 g to 125 g) than the two common species in our study, which range from about 23 g (Whitefooted Mouse) to 15 g (Northern Short-tailed Shrew) (C. Norment, unpubl. data). Consequently, small mammals in the study by Sanecki and Green (2005) may have contacted the tape more frequently. Although their hairtube design was different, Pocock and Jennings (2006) speculated that large hair tubes may undersample Sorex minutus L. (Pygmy Shrew), which weigh between 2–5 g (McDevitt and Andrews 1997). Even though hair tubes offer advantages over traditional trapping methods for studying small-mammal winter ecology, they also possess several limitations. First, as in our study, some small mammals may remove baits from tubes without depositing hair samples. Selection of a tube size appropriate to the target species (Lindenmayer et al. 1999, Pocock and Jennings 2006, Sanecki and Green 2005) and using pieces of double-sided tape large enough to cover most of the inside diameter of the hair tube should increase the proportion of hair samples obtained from hair tubes visited by small mammals. Second, hair tubes may not provide accurate abundance indices, because they cannot distinguish the number of individuals of the same species visiting tubes (Lindenmayer et al. 1994; Sanecki and Green 2005). However, a comparative study based on live trapping and hair-tube sampling found a significant correlation between number of captures and hair-tube index for two of three shrew species in Britain (Pocock and Jennings 2006), and hair tubes have been used to determine abundance indices for Sciurus vulgaris L. (European Red Squirrel; Mortelliti and Boitani 2008). Also, DNA samples from hair tubes can be used to identify individual animals (Foran et al. 1997). Finally, although impressions of hair cuticles, and hair length and width, are commonly used to identify small-mammal hairs, they may not be sufficient for distinguishing between closely related species (Harris and Yalden 2004, Sanecki and Green 2005). 538 Northeastern Naturalist Vol. 17, No. 4 In summary, despite several drawbacks, hair tubes offer important advantages for the study of small-mammal ecology, especially during periods with low temperatures and long-lasting snow cover. They are simple to use, do not require handling the animals, eliminate trap-related mortality, and can be used at any spatial scale, including the landscape level. Use of appropriately sized hair tubes may also offer an inexpensive way of determining when smaller-sized hibernators emerge during the spring. Finally, they employ low-cost materials, and can be adapted to a range of conditions and target species (Sanecki and Green 2005). Acknowledgments We thank Levi Atwater, Jesse Batz, Norman Gervais, Adam Lotyczewski, Alex Nies, Casey Pealo, Sarah Stio, and Allison Vegh for help with this project. Materials were provided by the Department of Environmental Science and Biology, College at Brockport, State University of New York. Dr. Ken Green and two anonymous reviewers offered helpful comments on drafts of this paper. Jim Dusen, College at Brockport, supplied the photograph of the hair tubes. The model for our hair tubes was provided by the late Dr. Glenn Sanecki, to whom we dedicate this paper. Literature Cited American Society of Mammalogists (ASM). 1998. Guidelines for the capture, handling, and care of mammals as approved by the American Society of Mammalogists. Journal of Mammalogy 79:1416–1431. Baker, P.J., R.J. Ansell, P.A.A. Dodds, C.E. Webber, and S. Harris. 2003. Factors affecting the distribution of small mammals in an urban area. Mammal Review 33:95–100. Buech, R.R. 1974. A new live-trap and techniques for winter trapping small mammals. Canadian Field-Naturalist 88:317–321. Dickman, C.R., and C.P. Doncaster. 1987. The ecology of small mammals in urban habitats. 1. Populations in a patchy environment. Journal of Animal Ecology 56:629–640. Foran, D.R., S.C. Minta, and K.S. Heinemeyer. 1997. DNA-based analysis of hair to identify species and individuals for population research and monitoring. Wildlife Society Bulletin 25:840–847. Formozov, A. 1946. Snow cover as in integral factor of the environment and its importance in the ecology of mammals and birds. Boreal Institute, Edmonton, AB, Canada. French, A. 2000. Interdependency of stored food and changes in hibernation of the Eastern Chipmunk (Tamias striatus). Journal of Mammalogy 81:979–985. Gurnell, J., P.W.W. Lurz, M.D.F. Shirley, S. Cartmel, P.J. Garson, L. Magris, and J. Steel. 2004. Monitoring Red Squirrel Sciurus vulgaris and Grey Squirrel (Sciurus carolinensis) in Britain. Mammal Review 34:51–74. Halfpenny, J., and Ozanne, R. 1989. Winter: An Ecological Handbook. Johnson Books, Boulder, CO. 273 pp. Harris, S., and D.W. Yalden. 2004. An integrated monitoring programme for terrestrial mammals in Britain. Mammal Review 34:157–167. 2010 H. Schwingel and C. Norment 539 Laidlaw, W.S., and B.A. Wilson. 1989. Distribution and habitat preference of small mammals in the eastern section of the Angahook-Lorne State Park. Victorian Naturalist 106:224–236. Lindemayer, D.B., R.B. Cunningham, C.F. Donnelly, B.E. Triggs, and M. Belvedere. 1994. Factors influencing the occurrence of small mammals in retained linear strips (wildlife corridors) and contiguous stands of montane ash forest in the central highlands of Victoria, southeastern Australia. Forest Ecology and Management 67:113–133. Lindemayer, D.B., R.B. Cunningham, and M.L. Pope. 1999. A large scale “experiment” to examine the effects of landscape context and habitat fragmentation on mammals. Biological Conservation 88:387–403. Martin, I. 1983. Daily activity of Short-tailed Shrews (Blarina brevicauda) in simulated natural conditions. American Midland Naturalist 109:136–144. Martin, R.E., R.H. Pine, and A.F. DeBlase. 2001. A Manual of Mammalogy with Keys to the Mammals of the World. Third Edition. McGraw Hill, New York, NY. 333 pp. Mathiak, H. 1938. A key to hairs of mammals of southern Michigan. Journal of Wildlife Management 2:251–268. Mayer, M.V. 1952. The hair of Californian mammals, with keys to the dorsal guard hairs of Californian mammals. American Midland Naturalist 48:480–512. McDevitt, R.M., and J.F. Andrews. 1997. Seasonal variation in brown adipose tissue mass and lipid droplet size of Sorex minutus, the Pygmy Shrew: The relationship between morphology and metabolic rate. Journal of Thermal Biology 22:127–135. Merritt, J.F. 1986. Winter survival adaptations of the Short-tailed Shrew (Blarina brevicauda) in an Appalachian montane forest. Journal of Mammalogy 67:450–464. Moore, J.E. 1988. A key for the identification of animal hairs. Journal of the Forensic Science Society 28:2335–339. Mortelliti, A., and L. Boitani. 2008. Inferring Red Squirrel (Sciurus vulgaris) absence with hair tubes surveys: A sampling protocol. European Journal of Wildlife Research 54:353–356. Pierce, S.S., and F.D. Vogt. 1993. Winter acclimatization in Peromyscus maniculatus gracilis, P. leucopus noveboracensis, and P. l. leucopus. Journal of Mammalogy 74:665–677. Pocock, M.J.O., and N. Jennings. 2006. Use of hair tubes to survey for shrews: New methods for identification and quantification of abundance. Mammal Review 36:299–308. Pruitt, W.O., Jr. 1984. Snow and small animals. Pp. 1–8, In J. F. Merritt (Ed.). Winter Ecology of Small Mammals. Special Publication of Carnegie Museum of Natural History 10. Pittsburgh, PA. Sanecki, G.M., and K. Green. 2005. A technique for using hair tubes beneath the snowpack to detect winter-active mammals in the subnivean space. European Journal of Wildlife Research 51:41–47. Sanecki, G.M., A. Cowling, K. Green, H. Wood, and D. Lindenmayer. 2006. Winter distribution of small mammals in relation to snow cover in the subalpine zone, Australia. Journal of Zoology 269:99–110. 540 Northeastern Naturalist Vol. 17, No. 4 Schmid, W.D. 1984. Materials and methods of subnivean sampling. Pp. 25–32, In J.F. Merritt (Ed.). Winter Ecology of Small Mammals. Special Publication of Carnegie Museum of Natural History 10. Pittsburgh, PA Sherburne, S.S., and J.A. Bissonette. 1994. Marten subdivision of the subnivean access point use: Response to subnivean prey levels. Journal of Wildlife Management 58:400–405. Teernik, B.J. 1991. Hair of West-European Mammals. Atlas and Identification Key. Cambridge University Press, Cambridge, UK. Webster, A.B., and R.J. Brooks. 1981. Social behavior of Microtus pennsylvanicus in relation to seasonal changes in demography. Journal of Mammalogy 62:738–751. Whittaker, J.O., Jr., and W.J. Hamilton, Jr. 1998. Mammals of the Eastern United States. Third Edition. Comstock Publishing Associates, Ithaca, NY. 583 pp. Wolff, J.O., and D.S. Durr. 1986. Winter nesting behavior of Peromyscus leucopus and Peromyscus maniculatus. Journal of Mammalogy 67:409–412. Zeggers, D.A., and J.F. Merritt. 1988. Adaptations of Peromyscus for winter survival in an Appalachian montane forest. Journal of Mammalogy 69:516–523.