by Nicolette L. Cagle, Ph.D., May 21, 2026
In the last post, I explored Eastern Box Turtle movement and home ranges, the remembered routes, seasonal movements, and complex spatial lives of turtles often dismissed as simply slow. In this post, we move inward, into box turtle habitats themselves. Here, we review what the Western scientific literature has found about Eastern Box Turtle habitat use, microhabitat selection, and overwintering.
The Eastern Box Turtle (Terrapene carolina carolina) is called a woodland turtle for good reason, but the literature suggests that “woodland” is only the beginning of the story. These turtles use rich and varied habitat mosaics. A single turtle’s year may involve mature forest, early successional openings, wetland edges, stream corridors, nesting sites, shaded summer refuges, and upland hibernacula. In other words, habitat is not just where a turtle is found; it is the shifting set of places that allows a turtle to meet its needs across a lifetime.
Habitat as a Mosaic
The wooded center of Eastern Box Turtle habitat is supported by research. For example, Grant (2024) found that Eastern Box Turtles selected hardwood forests most strongly, followed by pine plantations, grasslands, and roadsides; no individual traveled more than 20 m from the forest edge. Greenspan et al. (2015) found landscape-scale selection for pine-hardwood forest and hardwood forest, while agricultural land, aquatic habitat, developed areas, pine forest, pine plantations, and shrub/scrub occurred in home ranges at lower proportions than expected from availability.
Yet that forest habitat template varies regionally. Roe et al. (2018) found that turtles at the fire-maintained Weymouth Woods sandhills site preferred hardwoods, non-Longleaf Pine, mixed hardwood–Longleaf Pine forest, bottomlands, and areas near streams, while turtles at the nearby Lumber River site preferred hardwoods and uplands and used streams randomly. In West Virginia, Weiss (2009) concluded that optimal habitat at Beech Fork included mixed hardwood forests with nearby fields and first-order streams.
This suggests that conservation cannot simply preserve any forest and assume the turtle’s needs are met. Eastern Box Turtles use mosaics: mature forest, edges, moist lowlands, fields, openings, pine-hardwood mixtures, stream corridors, and structurally complex ground layers.
The Small-scale World
Microhabitat is the immediate, fine-scale environment an animal uses, e.g., the leaf litter it rests under, the log beside it, the humidity around its body, the depth of litter under its shell, or the bramble patch that offers both cover and shade. For Eastern Box Turtles, microhabitat can matter as much as broad habitat type.
Harris et al. (2020) found that turtles selected microsites with greater bramble cover, coarse woody debris, litter depth, 10-hour and 100-hour fuel abundance, and visual obstruction at 0–0.25 m, while large-scale vegetation type and time since fire did not strongly explain selection. Roe et al. (2018) found selection for locations closer to logs, with deeper litter, more leaf litter, more coarse woody debris, more vine and fern cover, and higher woody stem density.
Rossell et al. (2006) showed the same pattern from a thermal and hydric perspective in a North Carolina mountain wetland. Turtles used habitats with lower surface temperatures and higher humidity than the surrounding environment, especially during hot and dry periods, and constructed “forms” in more humid areas. A form is a shallow resting depression or sheltered spot, often in leaves, soil, or debris, where a turtle settles into cover.
Parlin et al. (2017) found that turtles selected shaded habitats, especially deciduous forest, evergreen forest, and herbaceous grassland; their observations included 54.2% in deciduous forest versus 47.5% available, 22.8% in evergreen forest versus 6.0% available, and 7.4% in grassland/herbaceous habitat versus 1.1% available.
At this scale, the turtle’s world becomes textured: leaf litter, logs, brambles, vines, ferns, shade, humidity, and cover. To a box turtle, these nuances may make the difference between overheating and surviving, drying out and staying moist, being exposed and being hidden.
A Terrestrial Turtle Seeking Moisture
Eastern Box Turtles are terrestrial, but many studies show that moisture and aquatic refuges are important, especially seasonally. Donaldson and Echternacht (2005) found 131 turtles in two small temporary ponds from June through August; individuals remained in ponds for up to 23 consecutive days, and as many as 32 turtles were found at one time in one pond. In the same study, seven turtles on nine occasions made abrupt linear movements toward a temporary pond from as far as 105 m away, and the mean greatest distance from the pond among ten trailed turtles was 219 m, with a range of 153–358 m.
Marchand et al. (2004) found a similar pattern at Jug Bay Wetlands Sanctuary in Maryland, where Eastern Box Turtles used wetlands extensively when those habitats were available. In a 45-ha study area along the tidal Patuxent River, three radio-tracked turtles had home ranges of 0.48 ha, 0.84 ha, and 2.25 ha, with a mean of 1.19 ha. Although upland forest dominated the study area, 52% of radio-telemetry sightings occurred in or within 10 m of wetland habitat, and individual turtles used wetlands in 48%, 45%, and 64% of sightings. Combined habitat use was 40% forest, 36% non-tidal floodplain/wetland, 16% tidal wetland, and 8% meadow. They also found that turtles were observed in non-tidal wetlands at higher ambient temperatures than in forest, meadow, or tidal wetland habitats, suggesting that shaded wetlands may serve as thermal and hydric refuges during hot weather (Marchand et al., 2004).
Quinn’s Connecticut thesis adds a northern-range version of the same story. At West Rock Ridge, where open- and closed-canopy wetlands were available, turtles used open-canopy wetlands most in spring (38.1%) and closed-canopy wetlands most in early summer (34.28%), before shifting toward mature uplands in late summer and fall. At Mountain Road, where wetland availability was much lower, turtles instead used early successional uplands most heavily in spring (61.46%) and early summer (41.10%), then shifted strongly to mature uplands in late summer (64.69%) and fall (97.23%) (Quinn, 2008).
This is a useful comparison because it suggests that box turtles may meet seasonal needs through different local habitat mosaics. Where wetlands are available, they may be used heavily in spring and summer; where wetlands are scarce, early successional uplands may partially fill similar seasonal roles, perhaps by offering warmth, cover, or foraging opportunities (Quinn, 2008).
Roe et al. (2018) found that hydrologic habitat preferences patterns differed by site. Weymouth Woods turtles preferred bottomlands and stream proximity, with turtle locations averaging 50.5 ± 5.0 m from streams, whereas Lumber River turtles used streams randomly and averaged 127.2 ± 16.4 m from streams. Williamson (2013) similarly found that creek-associated habitat influenced home-range and within-home-range selection, and estimated the local population at 332 turtles and 6 turtles/ha.
This is one of the paradoxes of the Eastern Box Turtle: it is not aquatic despite being relatively closely related to water turtles (think Emydidae family), but it is deeply tied to moisture. Rain, humidity, wet leaf litter, seeps, streams, lowlands, and temporary ponds may all become part of the turtle’s seasonal map.
The Thermal Life of Turtles
As ectotherms, turtles rely on external conditions to regulate body temperature. But the thermal story doesn’t mean that “warmer is better.” Weiss (2009) found that box turtles in West Virginia were more likely to be active when air temperature decreased and substrate temperature increased; the strongest activity model included air temperature, substrate temperature, and canopy cover. Rossell et al. (2006) found turtles using cooler, more humid microhabitats during hot, dry periods, suggesting that thermoregulation and avoidance of water loss were major drivers of microhabitat selection.
Parlin et al. (2017) complicated the story further. In southwest Ohio, turtles selected shaded habitats and conformed to ambient temperatures, even though those habitats were cooler than the species’ estimated thermal optima; movement was not correlated with internal body temperature. In other words, turtles were not always selecting the warmest thermal environments. They may also have been balancing moisture, shelter, safety, and other habitat needs.
Gordon et al. (2024) added an especially interesting disease-related angle. They studied myiasis, a parasitic condition in which fly larvae infest living tissue. In this case, Eastern Box Turtles at Camp Edwards, Massachusetts, were infected by the flesh fly Dexosarcophaga cistudinis. Gordon et al. (2024) found that infected turtles had higher shell temperatures than noninfected turtles, 27.92 ± 5.28°C versus 26.77 ± 5.64°C, but myiasis did not affect body condition, habitat use, or average daily distance moved. The authors suggest infected turtles may have been exhibiting behavioral fever, a response in which an ectotherm moves to warmer microhabitats to raise body temperature in response to infection.
Sites to Sleep through the Winter and Awakening in Spring
Quinn (2008) strengthens the overwintering piece of the story. A hibernaculum is the place an animal uses for winter dormancy; for Eastern Box Turtles, these are often shallow, self-excavated or naturally sheltered sites in soil and leaf litter. Although Quinn initially considered whether turtles might overwinter in wetlands, all radio-tracked turtles at both Connecticut sites hibernated in mature upland forest. They entered hibernacula between October 13 and 26 and used areas with loose soils, thick leaf litter, little understory, and no herbaceous cover.
This finding helps clarify that wetlands and moist habitats may be essential during the active season, while mature upland forests may become critical as overwintering habitat. In other words, a turtle’s “home” is not one habitat type, but a seasonal sequence of habitats that must remain connected.
Other studies show that hibernacula are not simply random winter holes. In West Virginia, Koester (2016) found that Eastern Box Turtles selected hibernation sites with softer, less compacted soil and deeper cover, especially mixed deciduous leaf litter. In North Carolina, Roe and Bayles (2021) found that overwintering turtles selected hardwood forests and mesic habitats and avoided upland pine forests; the turtles were deepest in late February, only a few centimeters below the soil-litter interface, and winter refuges could buffer them from lethal temperatures during dormant-season prescribed fire. Taken together, these studies make the hibernaculum less like an isolated burrow and more like a collaboration between turtle and site conditions that create a winter “box” around the animal.
Spring emergence adds another layer to this seasonal story. Woodley (2013), working with a northern population at Fort Custer in Michigan, distinguished between “surfacing” and full “emergence.” Surfacing occurred when turtles moved upward toward the entrance of overwintering burrows and sometimes became partly visible, while emergence occurred later, when turtles left the burrow and resumed regular aboveground movement. Soil temperature and accumulated growing degree days predicted these events better than calendar date alone. Woodley found that turtles could surface during an early warm spell, then wait days or weeks before fully emerging if cold returned; early-surfacing turtles often had longer delays before full emergence. This means a turtle may be vulnerable before it is fully active or easy to see. Spring, for a box turtle, is not simply a date on a calendar.
The North Carolina Story
North Carolina studies are especially useful because they show how much Eastern Box Turtle habitat use can vary across landscapes within a single state. In the Piedmont, Hall (1987) studied Eastern Box Turtles near Chapel Hill in southern Orange County, working from UNC’s Mason Farm Biological Reserve and adjacent land owned by William Lanier Hunt. Hall described the site as representative of the prevailing Piedmont Biological Province around Chapel Hill: an upland ridge about 400 ft in elevation, drained by moderately steep watersheds, with surface water limited mostly to seeps, springs, and a temporary vernal pool.
Hall’s site also carried the history of the Piedmont itself: former farms, old wagon tracks, eroded soils, abandoned fields, and a patchwork of oak-hickory hardwoods mixed with successional pine stands. In that landscape, he estimated a relatively sparse resident density of 0.5 turtles/acre, or 1.2 turtles/ha, much lower than Stickel’s Maryland bottomland estimate of 4.6 residents/acre. Hall also found an unusually high proportion of apparent transients — about 37% by Stickel’s definition — with many individuals moving straight out of the study area after being encountered (Hall, 1987).
Hall’s contribution is less about simple home-range size and more about the structure of movement. He distinguished residents, transients, and experimentally displaced turtles, and used spool-trailing to record continuous routes rather than isolated location points. The resident turtles showed recognizable home-range behavior, but their routes differed markedly among individuals: some used their ranges fairly evenly, while others moved through intensively used patches connected by straighter passages. Several females made long excursions that Hall interpreted as likely nesting trips, and at least one turtle made a route to a spring (Hall, 1987).
Later North Carolina studies add habitat and regional context to Hall’s movement-focused work. In the Piedmont, Kapfer et al. (2013) tracked turtles at three sites in Alamance and Randolph counties and found an average 100% MCP home range of 2.68 ha; turtles were located most often in lowland and upland deciduous hardwood forests, though compositional analysis did not detect strong selection for a particular habitat type. In the mountains, Rossell et al. (2006) studied microhabitat selection in a North Carolina mountain wetland and found that turtles used cooler, more humid sites than nearby random locations, particularly during hot and dry periods. In south-central North Carolina, Roe et al. (2018) compared turtles in fire-maintained sandhills at Weymouth Woods with turtles in unburned coastal plain forest at Lumber River. The Weymouth Woods turtles selected bottomlands and areas near streams, while Lumber River turtles preferred uplands and used streams randomly.
Together, the North Carolina literature shows that there is no single North Carolina “box turtle habitat.” Piedmont turtles may move through old farm-forest mosaics of oak-hickory woods, pine stands, seeps, springs, and temporary pools; mountain wetland turtles may respond strongly to humidity and cool microhabitats; sandhills turtles may depend on scarce bottomlands and stream corridors; and coastal plain turtles may use uplands in ways that differ from nearby sandhills populations. The common thread is not one vegetation type, but access to a connected mosaic of shelter, moisture, thermal refuge, nesting space, and seasonal movement routes.
Home is Many Places
The Eastern Box Turtle may carry a shell that closes like a box, but its ecological life is open, and spread across many places. It needs forest, but not just forest. It needs shade, but also sun. It needs damp hollows, but also dry uplands. It may use wetlands in summer and mature forest in winter. It may rest beneath leaves, bury into loose soil, travel to water, bask near an edge, and return to a familiar overwintering site.
The lesson of this habitat literature is that a box turtle’s home is not a single habitat type. It is a connected seasonal mosaic. To care for Eastern Box Turtles, we need to protect the full variety of tesselae that creates that mosaic: woods, wetlands, openings, edges, logs, leaf litter, brambles, seeps, streams, nesting sites, and hibernacula. The Eastern Box Turtle may live in a small bony box, but box of habitat it needs is large and varied.
References
Donaldson, B. M., & Echternacht, A. C. (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(2), 278–284. https://doi.org/10.1670/0022-1511(2005)039%5B0278:AHURTH%5D2.0.CO;2
Gordon, A. B., Jr., Drummey, D., Tur, A., Curtis, A. E., McCumber, J. C., Jones, M. T., Andersen, J. C., & DiRenzo, G. V. (2024). The effect of myiasis on Eastern Box Turtle (Terrapene carolina carolina) body condition, movement, and habitat use at Camp Edwards in Massachusetts. Northeastern Naturalist, 31(Special Issue 12), T55–T76. https://doi.org/10.1656/045.031.s1220
Grant, S. (2024). Effects of different land-use types on the space use of the Woodland Box Turtle [Master’s thesis, Winthrop University].
Greenspan, S. E., Condon, E. P., & Smith, L. L. (2015). Home range and habitat selection in the Eastern Box Turtle (Terrapene carolina carolina) in a longleaf pine (Pinus palustris) reserve. Herpetological Conservation and Biology, 10(1), 99–111.
Hall, S. P. (1987). The movement patterns of free-ranging animals: New theory and methods of study illustrated by comparisons of the routes of travel associated with residency, transiency and experimental displacement in a population of eastern box turtles (Terrapene c. carolina) [Doctoral dissertation, University of North Carolina at Chapel Hill]. ProQuest Dissertations & Theses.
Harris, K. A., Clark, J. D., Elmore, R. D., & Harper, C. A. (2020). Spatial ecology and resource selection of Eastern Box Turtles. The Journal of Wildlife Management, 84(8), 1590–1600. https://doi.org/10.1002/jwmg.21945
Kapfer, J. M., Muñoz, D. J., Groves, J. D., & Kirk, R. W. (2013). Home range and habitat preferences of Eastern Box Turtles (Terrapene carolina Linnaeus, 1758) in the Piedmont Ecological Province of North Carolina (USA). Herpetology Notes, 6, 251–260.
Koester KR. 2016. Hibernacula site selection of Eastern Box Turtles (Terrapene carolina carolina) in a West Virginia population [master’s thesis]. Huntington (WV): Marshall University.
Marchand, M. N., Quinlan, M. M., & Swarth, C. W. (2004). Movement patterns and habitat use of Eastern Box Turtles at the Jug Bay Wetlands Sanctuary, Maryland. In C. W. Swarth, W. M. Roosenburg, & E. Kiviat (Eds.), Conservation and ecology of turtles of the Mid-Atlantic Region (pp. 55–62). Bibliomania.
Parlin, A. F., do Amaral, J. P. S., Dougherty, J. K., Stevens, M. H. H., & Schaeffer, P. J. (2017). Thermoregulatory performance and habitat selection of the Eastern Box Turtle (Terrapene carolina carolina). Conservation Physiology, 5(1), cox070. https://doi.org/10.1093/conphys/cox070
Quinn, D. P. (2008). A radio-telemetric study of the Eastern Box Turtle (Terrapene carolina carolina): Home-range, habitat use, and hibernacula selection in Connecticut [Master’s thesis, Central Connecticut State University].
Roe, J. H., Wild, K. H., & Lunn, Z. R. (2018). Inter- and intra-population variation in habitat selection for a forest-dwelling terrestrial turtle, Terrapene carolina carolina. Herpetological Conservation and Biology, 13(3), 711–725.
Roe JH, Bayles Z. 2021. Overwintering behavior reduces mortality for a terrestrial turtle in forests managed with prescribed fire. Forest Ecology and Management. 486:118990. https://doi.org/10.1016/j.foreco.2021.118990
Walden MA. 2017. Hibernal phenology of the Eastern Box Turtle, Terrapene carolina carolina [master’s thesis]. Kingston (RI): University of Rhode Island.
Rossell, C. R., Jr., Rossell, I. M., & Patch, S. (2006). Microhabitat selection by Eastern Box Turtles (Terrapene c. carolina) in a North Carolina mountain wetland. Journal of Herpetology, 40(2), 280–284. https://doi.org/10.1670/236-05N.1
Weiss, J. A. (2009). Demographics, activity, and habitat selection of the Eastern Box Turtle (Terrapene c. carolina) in West Virginia [Master’s thesis, Marshall University].
Williamson, B. A. (2013). Examining habitat selection and home range behavior at multiple scales in a population of Eastern Box Turtles (Terrapene c. carolina), with notes on demographic changes after 17 years [Master’s thesis, Marshall University].
Woodley CL. 2013. Predicting spring emergence in a northern population of the Eastern Box Turtle (Terrapene c. carolina) [master’s thesis]. Fort Wayne (IN): Purdue University.
Sense of Place 
Leave a comment