2012 SOUTHEASTERN NATURALIST 11(4):689–698 Utilization of Woody Debris by Peromyscus leucopus in a Fragmented Urban Forest Calley G. Jones1,* and Erin S. Lindquist2 Abstract - Small nocturnal mammals, such as Peromyscus leucopus (White-footed Mouse), tend to avoid open spaces due to the threat of predation. Previous studies have shown that Peromyscus and other small-mammal species are captured at higher frequencies at mature forest sites with higher densities of woody debris. Reported trapping frequencies along forest edges relative to continuous forest have varied in previous literature, possibly due to regional differences in forest composition and Peromyscus distribution. We hypothesized that mice in an urban forest setting would be captured at a higher frequency in trapping sites with higher volumes of woody debris and capture rates would be lower at the edge than the interior. We trapped mice in 100 to 121 Sherman live traps in a permanent 1-ha plot in an urban, fragmented forest on the Meredith College campus in Raleigh, NC over a two-year period. We also measured volume of woody debris at each trapping site in one (2007) of the two years. Between the two years (2007 and 2008), trapping rates of P. leucopus were lower in 2008 than in 2007, but we estimated a higher population size in 2008 than in 2007. We found no correlation between volume of woody debris and number of P. leucopus captured in 2007 and that capture rates did not vary with distance from the forest edge in both years. Our results support previous findings that P. leucopus are nonspecific users of microhabitat, but are contrary to other research that found a positive correlation between amount of woody debris and abundance of Peromyscus. Introduction As urban habitat encroaches further upon forested areas, there are an increasing number of fragmented urban forests (Rytwinski and Fahrig 2007). The number of occupying species generally decreases as the area of forest decreases (Nupp and Swihart 2000). In small forest fragments in the eastern US, the mammal population is almost exclusively made up of small mammals including Peromyscus leucopus Rafinesque (White-footed Mouse; Nupp and Swihart 2000). Because P. leucopus is a major reservoir for Lyme borreliosis, it is important to document what factors affect the distribution of P. leucopus populations. Nearly all individuals in a natural population of P. leucopus in the North Carolina Piedmont were found to be infected with Lyme borreliosis (Bunikis et al. 2004). Lyme borreliosis is the most common vector-borne zoonosis in the US and incidents are increasing (Bunikis et al. 2004). Small mammals are found where there is an abundance of food, but their foraging habits are greatly influenced by habitat configuration and risk of predation 1School of Veterinary Medicine, North Carolina State University, 4700 Hillsborough Street, Raleigh, NC 27607-1428. 2Department of Biological Sciences, Meredith College, 3800 Hillsborough Street, Raleigh, NC 27607-5298. Corresponding author - cgjones8@ncsu.edu. 690 Southeastern Naturalist Vol. 11, No. 4 (Brinkerhoff et al. 2005). P. leucopus are an integral part of many food webs because they are prey to many species of birds, snakes, and carnivorous mammals, and they are potential agents of seed dispersal as well (Pearson and Ortega 2001, Webster et al. 1985). Small mammals have been found to respond not only to the size of fragmented forests, but to other urban features as well. Urban forests are often split or bordered by paved roads, and P. leucopus generally will not cross roads (Rytwinski and Fahrig 2007). Despite this fragmenting effect, Rytwinski and Fahrig (2007) found a positive correlation between P. leucopus and road density. Structures such as roads and cleared fields create an edge between habitat types. Orrock and Danielson (2005) found decreased activity of Peromyscus polionotus Wagner (Oldfield Mouse) near edges of experimental landscapes. Wolf and Batzli (2001, 2002, 2004) found that forest edge is a lower quality habitat for P. leucopus than the interior because of the increased risk of predation as well as higher rates of Cuterebra fontinella Clark (Bot Fly) infestations. Other research found no difference in capture rates between the edge and interior, but found higher recaptures on the edge during a wet year (Anderson et al. 2006). Wilder and Meikle (2006) found no difference in reproduction of P. leucopus between the edge and interior during the spring, but found higher mouse density, litter production, and reproductive effort on the edge during the autumn, suggesting seasonal patterns of edge use. The variability in response to edge habitat by Peromyscus documented in previous studies may be due to regional differences or other unidentified factors. Another forest feature affecting the behavior of P. leucopus is the availability of cover. Small mammals use cover such as logs or dense brush to avoid detection by predators and as nesting sites (Drickamer 1990, Greenberg 2002). Generally, P. leucopus prefers forest sites with dense coarse woody debris (Laerm and Castleberry 2007). Research in the northeastern US found that available cover affected capture of Peromyscus maniculatus (Wagner) (Deer Mouse) and P. leucopus, and found higher than expected capture rates along large logs (Drickamer 1990). Traveling along logs may reduce the noise made by small mammals and reduce risk of predation (Roche et al. 1999). In the Pacific Northwest, Carey and Harrington (2001) found a positive correlation between coarse woody debris and Peromyscus keeni (Rhoads) (Northwestern Deer Mouse), but not P. maniculatus, captures. Peromyscus gossypinus Le Conte (Cotton Mouse), uses woody debris almost exclusively as daytime refuges (Hinkleman and Loeb 2007, McCay 2000) and prefer long logs and stumps (McCay 2000). In southern Appalachian hardwood forests, capture rates of P. leucopus (Greenberg 2002) and P. maniculatus (Menzel et al. 1999) were higher in traps adjacent to coarse woody debris. However, Menzel et al. (1999) found that P. leucopus captures were not correlated with coarse woody debris. Furthermore, micro- and macrohabitat use by P. leucopus varies according to general region (Bowman et al. 2000, Drickamer 1990, Loeb 1999). It is apparent that findings on the spatial response of Peromyscus to variable woody debris differ and data are lacking in the southeastern Piedmont region. Our objective was to examine microhabitat use of P. leucopus in an urban forest in the southeastern Piedmont. We hypothesized lower capture rates at the 2012 C.G. Jones and E.S. Lindquist 691 edge versus the forest interior. We also hypothesized higher P. leucopus captures at trap sites with large woody debris volumes. Methods Field-site description Our research was conducted using a permanent 1-ha plot located in the urban, fragmented forest on the Meredith College campus, Raleigh NC (35o48'22.05"N, 78o41'31.39"W; Powell and Lindquist 2011). The forest is approximately 22.3 ha and is bordered by Meredith College to the south, a highway (I-440) to the west, and residential neighborhoods to the north and east. The southwest side of the plot extends approximately 10 m into a grassy utility right-of-way, creating an edge between the field and the forest. Dominant tree species are Acer rubrum L. (Red Maple), Oxydendrum arboreum L. (Sourwood), and Quercus alba L. (White Oak) (Powell and Lindquist 2011). Trapping Trapping was conducted during summer and fall of 2007 (1 July–29 Oct.), and the winter, spring, and summer of 2008 (27 Jan.–20 July). During summer, traps were open at night during 4 consecutive days per week (May–July 2007 and 2008); for winter, spring, and fall dates, traps were open at night during 2 consecutive days per week. Traps were set every afternoon between 1600 and 1800 and checked every morning between 0700 and 1000. In 2007, 100 non-folding Sherman live traps (7.6 cm x 8.9 cm x 30.5 cm) were set in a 10 x 10 grid, each 10 m apart. In 2008, two trapping lines to the north and east edges of the plot were added for a total of 121 traps. Traps were baited with a mixture of sunflower seeds and vanilla extract. During the months of September through March, a handful of synthetic cotton was placed in each trap to protect mice from low temperatures. All captured small mammals, including P. leucopus, were weighed, measured, and given a unique mark using a numbering system on the ventral area using permanent hair dye. Marks were reapplied each time an individual was captured. Individuals were released at the capture site. Woody debris Woody debris measurements were taken at all trapping locations (n = 100) on 6–13 July 2007. All woody debris was measured within a 3-m circle centered on the trap site. Downed woody material was included in measurements if it touched the ground at more than one place (did not include stumps), was >0.5 m long, and >5 cm in diameter at the middle. Woody debris was categorized as blowdown (fallen trees with exposed roots that appeared to have been blown over), log (tree stems that had been cut down), or limb (all other naturally fallen debris, such as tree limbs). Length, diameter at top, diameter at middle, and diameter at base were measured for all woody debris. We calculated volume for each piece of debris using Harmon and Sexton’s (1996) formula: (length[diameter at base + 4*diameter at middle + diameter 692 Southeastern Naturalist Vol. 11, No. 4 at top])/6. We obtained total volume by summing volumes of all debris within the 3-m circle at each trapping site. Data analysis We calculated daily trapping rates by dividing the number of trap nights (number of open traps; 100 in 2007 and 121 in 2008) for each trapping day by the number of P. leucopus captured on the same day and multiplied this proportion by 100. We calculated mean trapping rates for each season in 2007 and 2008. We calculated population estimates for 2007 and 2008 using the pseudo-removal method as described by Sutherland (2006). We examined how distance from the edge, year (2007 and 2008), and the distance*year interaction predicted total number of mice captures with a general linear model (GLM; JMP 9.0). We also analyzed how distance from the edge, amount of woody debris in 2007, and the distance*woody debris interaction predicted total number of mice captures in 2007 with a second GLM (JMP 9.0). We constructed a third GLM to test how type of woody debris and volume of woody debris in 2007 explained variation in number of captures at each trapping location in 2007 (JMP 9.0). Results P. leucopus were captured 391 times in 8429 trapping nights with 86 individual mice identified. In addition to P. leucopus, we captured 12 Glaucomys volans L. (Southern Flying Squirrel), 25 Sigmodon hispidus Say and Ord (Hispid Cotton Rat), and 13 Tamias striatus L. (Eastern Chipmunk). Highest and lowest daily trapping rates for P. leucopus occurred in the spring of 2008 (7.23% ± 3.24) and the fall of 2007 (3.27% ± 2.61), respectively, with a mean of 4.62% ± 2.96 across all seasons. We estimated the P. leucopus population on the grid at 40 and 45 individuals in 2007 and 2008, respectively. Numbers of P. leucopus captures at individual trapping sites appeared to be randomly distributed across the plot in both years (Fig. 1). However, number of captures varied in response to the combination of distance from the edge, year, and all possible interactions between these factors (F = 23.847, d.f. = 220, P less than 0.0001). When we assessed individual factor effects within the GLM model, number of captures changed relative to year (F = 68.99, P < 0.0001) but not distance from edge (Fig. 2; F = 0.7221, P = 0.3964); the year*distance from edge interaction was not significant. Number of P. leucopus captures in 2007 did not vary with amount of woody debris (F = 0.8285, P = 0.3650) or trapping location relative to the plot edge (F = 1.187, P = 0.2787; whole GLM model F = 0.7751, d.f. = 99, P = 0.5107; Fig. 3); the interaction between amount of woody debris and trapping location was not significant (F = 1.560, P = 0.2147). Volume (cm3) of woody debris in 2007 was highest in the form of limbs (mean ± SD = 3857 ± 4640), followed by blowdowns (1342 ± 4566) and logs (236 ± 960). High volumes of woody debris were randomly distributed across the plot (Fig. 4). However, the distribution of woody debris types varied across the plot; blowdowns were found primarily in the interior of the forest, and logs were found exclusively at the 2012 C.G. Jones and E.S. Lindquist 693 forest edge. Number of P. leucopus captures was not predicted by amount of woody debris in 2007 (F = 1.19, P = 0.276), woody debris type (blowdowns, log, and limbs; F = 0.477, P = 0.621), or the interaction between woody debris amount and type (F = 0.841, P = 0.433). Figure 1. Comparison of the distribution of P. leucopus individual captures across the plot in 2007 (A) and 2008 (B). The forest edge was located along the N–S edge of the plot and was 0–10 m from the edge of the plot (1–2 along W–E edge of plot in [A] and [B]). Figure 2. Mean (±SD) number of captured P. leucopus individuals at each trapping location relative to the edge of the plot (for each location: n = 10 in 2007, n = 11 in 2008). The forest edge was located 0–10 m from the edge of the plot. Distance from the forest edge (x) relative to distance from edge of plot (1–10 on x-axis) is therefore the following: (1) x ≤ 0 m; (2) 0 m ≤ x ≤ 10 m; (3) 10 m ≤ x ≤ 20 m; (4) 20 m ≤ x ≤ 30 m; (5) 30 m ≤ x ≤ 40 m; (6) 40 m ≤ x ≤ 50 m; (7) 50 m ≤ x ≤ 60 m; (8) 60 m ≤ x ≤ 70 m; (9) 70 m ≤ x ≤ 80 m; (10) 80 m ≤ x ≤ 90 m; (11) 90 m ≤ x ≤ 100 m. *This value is zero because an additional trapping line was added in 2008, increasing the total width of the plot. 694 Southeastern Naturalist Vol. 11, No. 4 Figure 3. Total number of P. leucopus captures as a function of woody debris volume in 2007. Each observation (n = 100) is one trapping location within the trapping plot. Figure 4. Three-dimensional comparison of the distribution of volume of woody debris across the plot in 2007. The edge of the forest was located along the N–S edge of the plot and was 0–10 m from the edge of the plot (1–2 along W–E edge of plot). 2012 C.G. Jones and E.S. Lindquist 695 Discussion Trapping occurred over two consecutive years, but 2007 and 2008 were treated as separate populations because P. leucopus has a maximum lifespan of approximately one year in the wild (Webster et al. 1985). None of the mice marked during the 2007 trapping period were recaptured in 2008, supporting the reported lifespan of less than one year. Our population estimates for 2007 and 2008 are similar to those reported from Illinois with populations of greater than 30 mice/ha in areas surrounded by a high amount of urban habitat (Barko et al. 2003). Large numbers of recaptures suggest that mice are residents with little or no emigration from the study area. The differences in estimated population size and capture rates we documented between years could be due to changes in abiotic and/or biotic factors at the individual sites from year to year, such as availability of food sources, presence of predators, amount of leaf litter, or amount of vegetative cover. For example, Clotfelter et al. (2007) found a strong positive correlation between Peromyscus spp. and the acorn mast crop from the previous year. We captured mice across the plot; P. leucopus did not favor the forest edge. Our findings contrast with those of Wolf and Batzli (2004), who found a negative correlation between forest edge and mouse densities. The difference in findings may be due to the disturbance level and forest size of the study sites. Wolf and Batzli (2001, 2004) worked in 4.8–610-ha deciduous forests with low levels of disturbance, whereas our study site was a 22.3-ha urban forest bordering a utility right-of-way and six-lane highway. Furthermore, Rytwinski and Fahrig (2007) found a positive impact of road density on P. leucopus abundances, and Cummings and Vessey (1994) observed higher P. leucopus densities along the edge of a 1-ha woodlot. These previous findings, along with ours, suggest that when compared to more-interior forest habitat, P. leucopus abundances are higher along edges of small, highly disturbed forest fragments, no different along edges of intermediate-sized and moderately disturbed forests, and lower along edges of large relatively undisturbed forest fragments. Although some studies have suggested that forest edges tend to be poor quality habitats for P. leucopus, the typical increase of understory at an edge increases the available food supply (Wolf and Batlzi 2004). Because the edge of our study plot extended out into the grassy track under power lines, it was regularly maintained by mowing (Powell and Lindquist 2011). Mowing encourages rapid understory growth and may provide more cover and food sources for mice, balancing out the potential negative impact of poor-quality habitat. We also recaptured the same individuals in various trapping locations on the edge and up to 60 m in the interior of the forest. Their movement between the edge and interior forest is significant because it shows that small-scale edge effects (within 20 m of a forest edge) may not be the primary driver of P. leucopus abundance. Our findings suggest that neither the volume nor type of woody debris at a capture site affect P. leucopus captures. This result is contrary to several previous studies (Carey and Harrington 2001, Drickamer 1990, Greenberg 2002), but not completely unexpected as several researchers have found that habitat 696 Southeastern Naturalist Vol. 11, No. 4 use by P. leucopus can vary greatly between regions and forest types (Bowman et al. 2000, Drickamer 1990, Loeb 1999). For example, Greenberg (2002) examined P. leucopus habitat use in hardwood forests at a much higher elevation (700 to 1070 m) containing no documented occurrences of edges within the study area. Furthermore, many of the other studies (Bowman et al. 2000, Carey and Harrington 2001, Greenberg 2002) did not measure the quantity of course woody debris, but instead measured number of logs or percent cover of course woody debris. Finally, because we documented high quantities of woody debris throughout the trapping plot relative to these other studies, P. leucopus may not be limited by woody debris availability at any of our trapping locations. Given that course woody debris dominates the understory in our plot due to low shrub, vine, and herb cover, we believe P. leucopus were responding to the relatively high woody debris abundance throughout. Our data provides insight into habitat use by P. leucopus in an urban forest setting with a high volume of coarse woody debris. Humans are increasingly encroaching into forested areas, creating forest edges and fragmented forests completely surrounded by urban habitat. It is therefore critical that we understand how P. leucopus and other small-mammal species respond to forest edges and other human disturbances such as log cutting and clearing. Additionally, understanding this species in urban settings is important because it is linked to the spread of Lyme disease in the human population (Logiudice et al. 2008). Our study illustrates that P. leucopus can be abundant in fragmented, urban mixed hardwood forests in the southeastern United States, but may not respond directly to the quantity of course woody debris or smallscale edge effects. Acknowledgments Funding for the field research was provided by the Undergraduate Research Program and the Department of Biological Sciences at Meredith College. The publication of the manuscript was supported by a grant from the Margaret A. Cargill Foundation to Meredith College. We thank Sara Roberson and Brittany Carr Beattie for their help in data collection. Special thanks go to Ben Hess at the North Carolina Museum of Natural Sciences for his assistance in identifying the Peromyscus species. Live mammal trapping on the Meredith College campus was approved and licensed by the North Carolina Wildlife Resources Commission. Literature Cited Anderson, C.S., D.B, Meikle, A.B. Cady, and R.L. Schaefer. 2006. Annual variation in habitat use by White-footed Mice, Peromyscus leucopus: The effects of forest patch size, edge, and surrounding vegetation type. Canadian Field-Naturalist 120:192–198. Barko, V.A., Feldhamer G.A., Nicholson M.C., and D.K. Davie. 2003. Urban habitat: A determinant of White-footed Mouse (Peromyscus leucopus) abundance in southern Illinois. Southeastern Naturalist 2:369–376. Bowman, J.C., D. Sleep, G.J. Forbes, and M. Edwards. 2000. The association of small mammals with coarse woody debris at log and stand scales. Forest Ecology and Management 129:119–124. 2012 C.G. Jones and E.S. Lindquist 697 Brinkerhoff, R.J., N.M. Haddad, and J.L. Orrock. 2005. Corridors and olfactory predator cues affect small-mammal behavior. Journal of Mammalogy 86:662–669. Bunikis, J., J. Tsao, C.J. Luke, Luna, D. Fish, and A.G. Barbour. 2004. Borrelia burgdorferi infection in a natural population of Peromyscus leucopus mice: A longitudinal study in an area where Lyme borreliosis is highly endemic. Journal of Infectious Disease 189:1515–1523. Carey, A.B., and C.A. Harrington. 2001. Small mammals in young forests: Implications for management for sustainability. Forest Ecology and Management 154:289–309. Clotfelter, E.D., A.B. Pedersen, J.A. Cranford, N. Ram, E.A. Snajdr, V. Nolan, and E.D. Ketterson. 2007. Acorn mast drives long-term dynamics of rodent and songbird populations. Oecologia 154:493–503. Cummings, J.R., and S.H. Vessey. 1994. Agricultural influences on movement patterns of White-footed Mice (Peromyscus leucopus). American Midland Naturalist 132:209–218. Drickamer, L.C. 1990. Microhabitat preferences of two species of deermice Peromyscus in a northeastern United States deciduous hardwood forest. Acta Theriologica 35:241–252. Greenberg, C.H. 2002. Response of White-footed Mice (Peromyscus leucopus) to coarse woody debris and microsite use in southern Appalachian treefall gaps. Forest Ecology and Management 164:57–66. Harmon, M.E., and J. Sexton. 1996. Guidelines for measurements of woody debris in forest ecosystems. Publication No. 20. US LTER Network Office, University of Washington, Seattle, WA. 73 pp. Hinkelman, T.M., and S.C. Loeb. 2007. Effect of woody debris abundance on daytime refuge use by Cotton Mice. Southeastern Naturalist 6:393–406. Laerm, J., and S.B. Castleberry. 2007. White-footed Mouse, Peromyscus leucopus. Pp. 332–336, In M.K. Trani, W.M. Ford, and B.R. Chapman (Eds.). The Land Manager’s Guide to Mammals of the South. The Nature Conservancy, Durham, NC. 546 pp. Loeb, S.C. 1999. Responses of small mammals to coarse woody debris in a southeastern pine forest. Journal of Mammalogy 80:460–471. Logiudice, K., S.T.K. Duerr, M.J. Newhouse, K.A. Schmidt, M.E. Killilea, and R.S. Ostfeld. 2008. Impact of host community composition on Lyme disease risk. Ecology 89:2841–2849. McCay, T.S. 2000. Use of woody debris by Cotton Mice (Peromyscus gossypinus) in a southeastern pine forest. Journal of Mammalogy 81:527–535. Menzel, M.A., W.M. Ford, J. Laerm, and D. Krishon. 1999. Forest to wildlife opening: Habitat gradient analysis among small mammals in the southern Appalachians. Forest Ecology and Management 114:227–232. Nupp, T.E., and R.K. Swihart. 2000. Landscape-level correlates of small-mammal assemblages in forest fragments of farmland. Journal of Mammalogy 81:512–526. Orrock, J.L., and B.J. Danielson. 2005. Patch shape, connectivity, and foraging by Oldfield Mice (Peromyscus polionotus). Journal of Mammalogy 86:569–575. Pearson, D.E., and Y.K. Ortega. 2001. Evidence of an indirect dispersal pathway for Spotted Knapweed, Centaurea maculosa, seeds via Deer Mice, Peromyscus maniculatus, and Great Horned Owls, Bubo virginianus. Canadian Field-Naturalist 115:354. Powell, A.S., and E.S. Lindquist. 2011. Effects of power-line maintenance on forest structure in a fragmented urban forest, Raleigh, NC. Southeastern Naturalist 10:25–38. Roche, B.E., A.I. Schulte-Hostedde, and R.J. Brooks. 1999. Route choice by Deer Mice (Peromyscus maniculatus): Reducing the risk of auditory detection by predators. American Midland Naturalist 142:194–197. 698 Southeastern Naturalist Vol. 11, No. 4 Rytwinski, T., and L. Fahrig. 2007. Effect of road density on abundance of White-footed Mice. Landscape Ecology 22:1501–1512. Sutherland, W.J. (Editor). 2006. Ecological Census Techniques, 2nd Edition. Cambridge University Press, New York, NY. 432 pp. Webster, W.M., J.F. Parnell, and W.C. Biggs, Jr. 1985. Mammals of the Carolinas, Virginia, and Maryland. The University of North Carolina Press, Chapel Hill, NC. 255 pp. Wilder, S.M., and D.B. Meikle. 2006. Variation in effects of fragmentation on the Whitefooted Mouse (Peromyscus leucopus) during the breeding season. Journal of Mammalogy 87:117–123. Wolf, M., and G.O. Batzli. 2001. Increased prevalence of Bot Flies (Cuterebra fontinella) on White-footed Mice (Peromyscus leucopus) near forest edges. Canadian Journal of Zoology-Revue Canadienne De Zoologie 79:106–109. Wolf, M., and G.O. Batzli. 2002. Effects of forest edge on populations of White-footed Mice Peromyscus leucopus. Ecography 25:193–199. Wolf, M., and G.O. Batzli. 2004. Forest edge—high or low quality habitat for Whitefooted Mice (Peromyscus leucopus)? Ecology 85:756–769.