Northeastern Naturalist 554 T.S. Fredericksen 22001144 NORTHEASTERN NATURALIST 2V1(o4l). :2515,4 N–5o6. 44 Thermal Regulation and Habitat Use of the Eastern Box Turtle in Southwestern Virginia Todd S. Fredericksen* Abstract - While Terrapene carolina carolina (Eastern Box Turtle) are found over a wide geographic range, they appear to select microhabitats based on the need for thermoregulation, minimization of water loss, and reproduction. Habitat selection and the activity patterns of Eastern Box Turtles in southwestern Virginia were studied in relation to shortterm weather conditions and seasonal variation. Turtles were located using telemetry 36% of the time in the interior of mature forest habitats, 23% in edge habitats, 18% in fields, 7% in a 22-year-old Pinus strobus (Eastern White Pine) plantation, 7% in forest canopy gaps, 5% in a 4-year-old clearcut, and 4% in streams. We observed a seasonal shift in habitat use, with more turtles using mature forest habitat compared to other habitat types in all months, except for May when 47% of all turtles were located in edge habitat. Also, Eastern Box Turtles often selected canopy gaps within forests during the fall, and females moved from forests to recent clearcuts during the nesting season, which accounted for a larger home-range size. As expected, turtle activity was lower during hot, dry periods in midsummer and decreased gradually during the autumn with decreasing temperature. Rainfall increased turtle activity, especially when following prolonged dry periods. Introduction Terrapene carolina carolina L. (Eastern Box Turtle) is a common terrestrial turtle found throughout the eastern United States. The geographic range closely coincides with that of the eastern deciduous forest (Dodd 2001), but Eastern Box Turtles may select microhabitats within or outside of forests for behaviors such as thermoregulation or nesting. Like most ectotherms, Eastern Box Turtles enter a state of torpor during winter months. In spring and summer, Eastern Box Turtles spend most of their time in forests, but will often venture into open habitats in search of food, and for basking and nesting (Dodd 2001). While Eastern Box Turtles are found in a wide range of macrohabitats, they appear to select microhabitats based on the need for thermoregulation and to minimize water loss (Ernst 1968, Rossell et al. 2006). As ectotherms, Eastern Box Turtles must maintain temperature homeostasis by seeking cover, such as shaded forests or streams, when body temperatures are high and by finding basking locations when body temperatures are low, in order to increase their metabolic rate for foraging and digestion. Ernst et al. (1994) noted that Eastern Box Turtles generally select microhabitats that maintain body temperatures of 29–38 ºC. Outside of this temperature range, Eastern Box Turtles may seek cover under leaf litter or vegetation, called a “form” (sensu Stickel 1950) and, in the case of high temperatures and drought, turtles may enter water bodies for thermoregulation and to prevent dehydration (Donaldson and Echternacht 2005). *Ferrum College, Ferrum, VA 24088; tfredericksen@ferrum.edu. Manuscript Editor: Thomas Maier Northeastern Naturalist Vol. 21, No. 4 T.S. Fredericksen 2014 555 Habitat selection by box turtles may vary depending on age, sex, and the relative availability of habitat types. Juvenile turtles often prefer areas with thick vegetation, presumably to avoid predation, for which they are more vulnerable than adults because of their softer shell and smaller size (Dodd 2001, Jennings 2007). Female turtles often select open sunlit areas for nesting in order to facilitate incubation (Congello 1978, Hall et al. 1999). In the eastern US, little is known about how habitat availability affects the behavior of individual Eastern Box Turtles. Presumably, home-range size should vary according to the mixture of habitat types that turtles require in order to obtain sufficient food and regulate their temperature. Eastern Box Turtles that do not have a variety of microhabitats available for thermoregulation, moisture conservation, and nesting sites (in the case of females) may require larger home ranges. Daily and seasonal weather variation is also likely to have a strong influence on habitat selection (Dodd 2001). Throughout most of their range, Eastern Box Turtle populations are threatened by urban and suburban development that may increase the likelihood of injury and mortality from motor vehicles, mowers, and edge predators, as well as the collection of turtles for pets (Budischak et al. 2006, Dodd 2001). With increasing urbanization, Eastern Box Turtles may require larger home ranges to find suitable microhabitats, thereby increasing their vulnerability to these and other threats. Understanding the variability in habitat and microhabitat selection is important for assessing conservation threats to these turtles. The objectives of this study were to determine the variation in habitat selection and activity of Eastern Box Turtles in an area with high habitat heterogeneity. Specifically, I addressed the following questions: What proportion of the time do Eastern Box Turtles use open vs. shaded habitats? Are patchy habitat types that provide close proximity to shade and sun, such as forest canopy gaps and forest–field edges, preferentially used compared to shaded forest locations? Does habitat use and home-range size differ between male and female turtles? Are there specific weather or seasonal cues that determine when turtles move between habitat types? Field-Site Description The study area was approximately 60 ha in size, on private property located in western Franklin County, VA (36.90ºN, 80.03ºW), situated near the intersection of the Piedmont and Blue Ridge Physiographic provinces, with moderately hilly terrain and an elevation of about 400 masl (Fig. 1). The study area was selected because it contained a variety of habitat types of different successional ages, allowing evaluation of the relative use by turtles of different habitat types. The habitat types included mature mixed hardwood and pine–hardwood forests (about10 ha), a 2.4- ha mid-successional (2-year-old) Pinus strobus L. (Eastern White Pine) plantation, a 4-year-old clearcut site (about 4 ha) with natural regeneration, and grassy areas and wildflower meadows maintained by seasonal mowing or winter burning (about 3 ha). The entire site was intersected by numerous small, perennial and intermittent streams and contained an occupied residential home at its center, with a gravel Northeastern Naturalist 556 T.S. Fredericksen 2014 Vol. 21, No. 4 access road. The nearest paved road was about 800 m from the residence. The most-common tree species in the mature forests included Liriodendron tulipifera L. (Tuliptree), Acer rubrum L. (Red Maple), Eastern White Pine, Oxydendrum arboreum (L.) DC (Sourwood), and Quercus spp. (oaks). A portion of the forests had been subjected to light firewood cutting. Methods From early May to late June 2012, we opportunistically located 14 Eastern Box Turtles through searches of the study area and affixed a radio-transmitter (SOPR- 2190, Wildlife Materials Inc., Murphysboro, KY) to the carapace of each. The loss of a signal on one transmitter reduced the sample size to 13 (nine adult males, three adult females, and one 3-year-old juvenile of undetermined sex). Beginning in May and continuing until early November, we located the turtles 42 times on separate dates using radio-telemetry (about twice per week). For each location, we recorded the geographic coordinates, activity status (i.e., active on the surface, inactive under leaf litter, soil, or underwater), and the habitat type of the location (i.e., field, mature forest, forest canopy gap, forest–field edge, earlysuccessional forest, mid-successional pine plantation, stream) for each turtle. We considered edge habitats to be those within 5 m of a field–forest boundary. At the time of location, we recorded the ambient air temperature and relative humidity with a Kestrel 4000 pocket weather station (Nielsen-Kellerman, Boothwyn, PA). Rainfall was recorded in the center of the property in an open field using a standard rain gauge. Based on location data, the home-range size of each turtle was estimated using the minimum convex polygon method. Figure 1. Aerial photo of study area from Google Earth® showing heterogeneity of habitat types. The study site is located in Franklin County, VA, indicated on the state map to right. Northeastern Naturalist Vol. 21, No. 4 T.S. Fredericksen 2014 557 In July, we glued a small iButton temperature sensor (Thermochron Inc., Dallas, TX) to the top of the carapace of all adult turtles and programmed it to record air temperature at 30-minute intervals. A temperature sensor was not placed on the single juvenile found, because of its small body size. We also placed two temperature sensors in each of two different locations within both interior-forest and open-field habitats as reference points. We situated these sensors in shaded locations on top of wooden blocks that were approximately the height of an adult Eastern Box Turtle. Data were collected from 10–31 July to compare carapace sensor temperatures to those of the reference points sited in field and forest habitats . Results Of the total 305 Eastern Box Turtle locations, 36% were in the interior of mature forest habitats, 23% in edge habitats, 18% in fields, 7% in the 22-year-old Eastern White Pine plantation, 7% in forest canopy gaps, 5% in the 4-year-old clearcut, and 4% in streams. There were more turtles using mature forest habitat compared to any other habitat types in all months except for May, when 47% of all turtles were located in edge habitat (Table 1). Turtles spent the greatest percentage of the time in mature forests in July, the warmest month of the year, and often used stream habitats during July and August (Table 1). Field habitats were occasionally used by several turtles, but the single juvenile turtle spent nearly all of its time in the field habitat, as did another adult male. Forest canopy gaps were not used with much frequency during most of the year (2–10% of locations) until October, when 23% of all turtle locations were in gaps, the second highest habitat use after mature forest (Table 1). Three male turtles were rarely located outside of mature forests, and another male was located nearly all of the time in the mid-successional pine plantation. One adult male spent more than one week completely submersed in the mud of an intermittent stream during July. Females made similar habitat selections as males except during the nesting season, when all three females traveled to a nearby early-successional forest site (the 4-year-old clearcut), each spending a few weeks there from late May to early July, presumably to lay eggs (nesting was not confirmed). Only females used this Table 1. Monthly percentage of Eastern Box Turtle relocations by habitat type, total number of locations, mean air temperature, and mean relative humidity on days of location at study site in Franklin County, VA, during 2010. Habitat Type May June July August September October Mature forest 6 41 47 40 38 46 Mature forest gap 3 2 5 10 5 23 Early-successional forest (clearcut) 10 6 6 0 0 0 Mid-successional pine plantation 3 12 3 10 13 4 Field 21 16 15 21 22 8 Edge 47 23 13 12 19 19 Stream 0 0 11 7 3 0 Number of locations 70 49 62 61 37 26 Temperature (ºC) 24.9 26.4 28.1 28.3 20.2 23.2 Relative humidity (%) 63.6 66.0 71.1 59.4 73.7 58.6 Northeastern Naturalist 558 T.S. Fredericksen 2014 Vol. 21, No. 4 early-successional forest habitat. The movement to this habitat type by females significantly expanded the home range of females compared to males. Males had a mean home range of 1.0 ha (range = 0.3–2.2 ha), whereas females had a mean home range of 5.4 ha (range = 3.7–7.9 ha). An example of a male and female homerange map is provided in Figure 2. The single juvenile turtle had a very small home Figure 2. Examples of home ranges (as defined by area encompassing dots shown on photos, which indicate recorded locations) for a female Eastern Box Turtle (upper photo) and male Eastern Box Turtle (lower photo). The female turtle spent most of the non-nesting season in mature forest, but moved to early-successional forest during the nesting season, greatly expanding its home range. Northeastern Naturalist Vol. 21, No. 4 T.S. Fredericksen 2014 559 range (0.04 ha). The larger size of female turtle home ranges appeared attributable to their travel to the clearcut stand, as, their home ranges were similar to male turtles when their time spent at this site was excluded (mean = 1.4 ha, range = 0.8–1.8 ha). The activity of turtles changed with temperature, relative humidity, and rainfall. For example, the period spanning 20–30 June was hot and dry. Near the end of this period, 12 of the 13 turtles were either inactive in their forms or under thick vegetation in the field. The drought was broken by a 1.5-cm rainfall event on 30 June–1 July. On 2 July, only one turtle was still within a form or inactive. Generally, turtles appeared to be more active under conditions of lower temperature and higher relative humidity than at other times during the middle of summer. The prevailing ambient air conditions during periods when turtles were inactive were characterized by slightly lower relative humidity (62.4% vs. 67.2%) and a higher temperature (28.4 vs. 26.6 ºC). From 10–31 July, turtles tended to track temperatures characteristic of shaded forest conditions rather than sunlit field conditions (Fig. 3). During this time, 48% of Eastern Box Turtle locations were in mature forest habitat, whereas 15% of locations were in field habitats. Turtle carapace sensor temperatures were similar to field and forest sensor temperatures at night, but temperatures of field sensors were approximately 3 ºC higher than sensors on turtles or the sensors in the forest during the middle of the day and late afternoon (Fig. 3). Figure 3. Temperature data from sensors attached to the carapace of Eastern Box Turtles compared to sensors placed in field and forest locations showing means (bars) and standard errors (shown above bars). Data are from 10–31 July 2012. Night = 2100–0600 hrs, early morning = 0600–1000 hrs, mid-day = 1000–1400 hrs, late afternoon = 1400–1800 hrs, and evening = 1800–2100 hrs. Northeastern Naturalist 560 T.S. Fredericksen 2014 Vol. 21, No. 4 Turtle activity varied monthly. Turtles were most active during May, based on the percentage of locations when turtles were observed moving or resting on the ground surface as compared to those inactive in forms or idle under thick cover (Table 2). The percentage of active turtles decreased in June and July, but then increased again in August, despite higher average ambient temperatures and lower average relative humidity. In September and October, turtle activity decreased sharply (Table 2). Discussion Eastern Box Turtles’ frequent use of mature hardwood forest, the predominant habitat type in the study area, was expected given such habitat’s importance for temperature regulation, particularly during the warmest months of the year. The frequent use of edge habitats was also expected because Eastern Box Turtles are likely to move between forests and open habitats for thermoregulation. Interestingly, Eastern Box Turtles also frequented the mid-successional Eastern White Pine plantation, although only a small portion of the study area, likely because of the dense canopy overstory and protection it provided. Perhaps most interesting was the seasonal shift in habitat use. In the study area, Eastern Box Turtles have typically emerged from their forested overwintering locations under the leaf litter in early May (Ellington et al. 2007). In the present study, however, edge habitat during May represented about 50% of turtle locations compared to 13–23% in other months, indicating that many turtles move to edge habitats after overwintering in the forest. High daily temperature fluctuation during this time may make edge habitats desirable for thermal regulation because turtles can easily move between sunlit and shaded locations. Diel temperature fluctuations are also high in the fall, but Eastern Box Turtles used edge habitats only 19% of the time during September and October, with a strong relative increase in the use of canopy gaps in October, suggesting that this habitat type provides a more suitable thermoregulation area for turtles near the end of the active season, as they move further into the forest to prepare for overwintering (Reagan 1974). The use of shaded forest habits during the mid-summer suggests a reduced need for basking locations during this time due to higher ambient air temperatures. Nighttime temperatures during that time remain relatively high, not dropping below the lower threshold for active temperature for Eastern Box Turtles (~10 °C in the Table 2. Percentage of locations where turtles were active on the surface or inactive (in forms, underneath leaf litter, or submerged in streams) at the study site in Franklin County, VA, from May–October 2010., Status May June July August September October Active 91 62 71 75 59 38 Inactive 9 38 29 25 41 62 Number of locations 70 49 62 61 37 26 Temperature (ºC) 24.9 26.4 28.1 28.3 20.2 23.2 Relative humidity (%) 63.6 66.0 71.1 59.4 73.7 58.6 Northeastern Naturalist Vol. 21, No. 4 T.S. Fredericksen 2014 561 study area; Ellington et al. 2007). Eastern Box Turtle carapace temperatures in July closely followed ambient ground-level temperatures in the forest, indicating that they were mostly located in forest habitats (Fig. 3). Field-habitat use was relatively consistent (15–22% of locations) during most of the study period, but decreased in October (8%) as turtles spent more time within forests. Similar to observations by Donaldson and Echternacht (2005), Eastern Box Turtles in the present study occasionally entered streams during the warmest part of the year, sometimes spending several days there, likely in attempt to lower body temperatures and reduce desiccation. Males and females were similar with respect to habitat use, except for the use of early-successional forest habitat by females during the nesting season. The use of the clearcut site by all three female turtles during the nesting season was unexpected, given the proximity of field habitat closer to their core non-nesting season home ranges. These females moved about 300–500 m to this habitat type and back during the late spring and early summer. Stickel (1989) similarly found that movement to nesting sites extended the home range of female Eastern Box Turtles by 400–700 m. The clearcut habitat was covered by a dense layer of hardwood sprouts and saplings about 5 m tall. Although shaded at ground level, the relatively short stature of the vegetation may have provided sufficiently high temperatures for incubation and a sparser ground-vegetation layer compared to the grassy fields, which may be preferable for nest excavation, although I have observed female turtles nesting in the field habitat during other years. Only one juvenile turtle was included in this study, and it spent nearly the entire study period within a 100-m2 area of dense field vegetation. While I encountered many other turtles besides the ones with transmitters during the course of the study, I never found another juvenile turtle, illustrating the difficulty of detecting them due to their secretive behavior (Stickel 1950, Wilson and Ernst 2005). The only significant movement of the juvenile was a 50-m transit across a stream into a lowland hardwood forest in early November, presumably to seek an overwintering site. An interesting observation from this study is the highly individualistic behavior of Eastern Box Turtles with respect to habitat selection. Three male turtles were relocated mostly in forest habitat, whereas two other males were located most of the time in overgrown field habitat. Another male was found most of the time in the mid-successional Eastern White Pine forest. Also, another male turtle was found most of the time in edge habitats. Eastern Box Turtle behavior can be highly plastic with respect to habitat selection. For example, when confronted by a sudden loss of forest cover on a site due to clearcutting and chipping of residual slash, Fredericksen and Bernard (2010) observed a female Eastern Box Turtle using open habitat for several weeks in July without seeking shelter in the forest, and it also spent over one month under sparse cover in an un-shaded clearcut during very hot (>32 ºC) daytime temperatures and dry conditions. The activity of Eastern Box Turtles varied by season and with changes in temperature and precipitation. Turtles tended to be very active following emergence from hibernation (May) and had a greater tendency to use edge habitats during this Northeastern Naturalist 562 T.S. Fredericksen 2014 Vol. 21, No. 4 time. Most Eastern Box Turtles became inactive during very warm periods and droughts, but periods of rain and/or lower temperatures often caused a sudden increase in activity. Strass et al. (1982) found no relation between activity of Eastern Box Turtles and soil and air temperatures. Dodd et al. (1994) attributed the use of resting forms by Eastern Box Turtles as a means to prevent desiccation, and Rossell et al. (2006) found that relative humidity was the principal environmental factor that differed between form locations and the surrounding environment. In the present study, an increase in the percentage of locations of inactive Eastern Box Turtles occurred in the fall. October was the only month when there were more observations of inactive turtles than active ones. Cooler nighttime temperatures during this period may explain this pattern as turtles prepare for overwintering (Stickel 1950). In late October and early November, two turtles were buried into the soil, perhaps indicating that they had already entered dormancy. Most other turtles at this time were formed within leaf litter. Seasonal differences in habitat use observed in this study have implications for Eastern Box Turtle conservation. Their use of edge habitats during the late spring may make Eastern Box Turtles more vulnerable to injury or mortality from road traffic, agricultural activities, or lawn mowers at this time of year. Additionally, the larger home range of female turtles during their nesting season makes them more likely to cross roads or frequent field habitats where mowing occurs. Gibbs and Steen (2004) found a trend for an increasing proportion of males in populations of both freshwater and terrestrial turtles in the United States and cited increased road mortality during nesting as a potential major factor in the decline of breeding females, among other threats. Hall et al. (1999) also found an increasing proportion of males in a long-term study of a population of Eastern Box Turtles in Maryland and suggested that movements during the nesting season make females more vulnerable to mortality. Such mortality is likely to increase in more urbanized landscapes (Budischak et al. 2006). The larger number of males turtles in the present study could have been due to mere chance or higher mortality risks to the more mobile females. In the study area, turtles were subjected to a relatively low probability of road mortality, given the one unpaved road that had little traffic. On the other hand, mowing is carried out within some of the fields in the study area, which could have subjected females to higher mortality, given their use of these fields for nesting in previous years. Interestingly, all three females in the year of this study used a recent early-successional forest for nesting rather than field habitat. Although only one juvenile turtle was observed in this study, it spent nearly the entire summer in the field habitat. If this habitat use is typical of juveniles, it could substantially increase their risk of mortality from mowing. The female Eastern Box Turtles’ preference for the use of clearcuts during the nesting season compared to grass-dominated fields in the present study merits further investigation. There are few data on the use of clearcuts by Eastern Box Turtles, yet such sites often provide dense, low cover, which could reduce detection by nest predators. Additionally, temperatures may be high enough to facilitate Northeastern Naturalist Vol. 21, No. 4 T.S. Fredericksen 2014 563 incubation in clearcuts, given the lack of overstory canopy. Partially logged forests may also be useful to Eastern Box Turtles, given the formation of canopy gaps, which may provide small basking areas within forests. In this study, canopy gaps were most often used in the fall. There are few studies of the effects of partial logging on Eastern Box Turtles, although Fredericksen et al. (2006) found an equal number of Eastern Box Turtles on logged compared to unlogged sites over three counties in southwestern Virginia, which may indicate that partially logged forests are used at levels similar to mature forests. More studies are needed examining habitat changes that are brought about from logging and other land-use, and how they may change the behavior of Eastern Box Turtles. Acknowledgments I thank Dr. Leslie Lambert, Dr. Gail Summer, the Ferrum College Professional Development Committee, and Ferrum College Freshman Scholars Program for funding this research. I also thank Brent Weiss, Trevor Leach, Savanna Hartmann, Neil Fredericksen, and Lily Fredericksen for field assistance. I also thank two anonymous reviewers and Thomas J. Maier, Manuscript Editor, for their helpful suggestions that improved this manuscript. Literature Cited Budischak, S.A, J.M. Hester, S.J. Price, and M.E. Dorcas. 2006. Natural history of Terrapene carolina (Box Turtles) in an urbanized landscape. Southeastern Naturalist 5:191–204. Congello, K. 1978. Nesting and egg-laying behavior in Terrapene carolina. Proceedings of the Pennsylvania Academy of Sciences 52:51–56. Dodd, C.K., Jr. 2001. North American Box Turtles: A Natural History. University of Oklahoma Press, Norman, OK. 231 pp. Dodd, C.K., Jr., R.F. Franz, and L.L. Smith. 1994. Activity patterns and habitat use of Box Turtles (Terrapene carolina bauri) on a Florida island, with recommendations for management. Chelonian Conservation and Biology 1:97–106. Donaldson, B.M., and A.C. Echternacht. 2005. Aquatic habitat use relative to home range and seasonal movement of Eastern Box Turtles (Terrapene carolina carolina: Emydidae) in eastern Tennessee. Journal of Herpetology 39:278–284. Ellington, D., K. Ingram, T. Walker, and T.S. Fredericksen. 2007. Hibernation behavior of the Eastern Box Turtle in the Virginia Piedmont. Catesbeiana 27:63–71. Ernst, C.H. 1968. Evaporative water-loss relationships of turtles. Journal of Herpetology 2:159–161. Ernst, C.H., J.E. Lovich, and R.W. Barbour. 1994. Turtles of the United States and Canada. Smithsonian Institute Press, Washington, DC. 578 pp. Fredericksen, T.S., and J.L. Bernard. 2010. Observations on the response of four Eastern Box Turtles (Terrapene carolina carolina) to clearcut logging and chipping in southern Virginia. Catesbeiana 30:3–9. Fredericksen, T.S., K. Graves, and T. Pohlad-Thomas. 2006. Herpetofauna in logged and unlogged forest stands in south-central Virginia. Catesbeiana 26:52–63. Gibbs, J.P, and D.A. Steen. 2004. Trends in sex ratios of turtles in the United States: Implications of road mortality. Conservation Biology 19:552–556. Hall, R.J., P.F.P. Henry, and C.M. Bunck. 1999. Fifty-year trends in a box turtle population in Maryland. Biological Conservation 88:165–172. Northeastern Naturalist 564 T.S. Fredericksen 2014 Vol. 21, No. 4 Jennings, A.H. 2007. Use of habitats and microenvironments by juvenile Florida Box Turtles, Terrapene carolina bauri, on Egmont Key. Herpetologica 63:1–10. Reagan, D.P. 1974. Habitat selection of the Three-toed Box Turtle, Terrapene carolina triunguis. Copeia 1974:512–527. Rossell, C.R. Jr, I.M. Rossell, and S. Patch. 2006. Microhabitat selection by Eastern Box Turtles (Terrapene c. carolina) in a North Carolina mountain wetland. Journal of Herpetology 40:280–284. Stickel, L.F. 1950. Populations and home-range relationships of the Box Turtle, Terrapene c. carolina (Linnaeus). Ecological Monographs 20:353–378. Stickel, L.F. 1989. Home-range behavior among Box Turtles (Terrapene c. carolina) of a bottomland forest in Maryland. Journal of Herpetology 23:40–44. Strass, P.K., K.J. Miles, B.S., Jr., McDonald, and I.K. Brisbin. 1982. An assessment of factors associated with the daytime use of resting forms by Eastern Box Turtles (Terrapene carolina carolina). Journal of Herpetology 16:320–322. Wilson, G.L., and C.H. Ernst. 2005. Reproductive ecology of the Terrapene carolina carolina (Eastern Box Turtle) in Central Virginia. Southeastern Naturalist 4:689–702.