Description
A medium-sized turtle with a low, broad, gray to brown, usually keeled carapace that is intricately sculptured with concentric growth layers. Plastron is yellow, each scute having an irregular dark lateral blotch. Adults have orange on neck and limbs and usually are 14-20 cm in carapace length, rarely to 23 cm (Smith and Brodie 1982, Conant and Collins 1991). Hatchlings average 26.6-34.0 mm carapace length (CL) (Harding and Bloomer 1979, Lovich et al. 1990) and have a tail that may be as long as the carapace.
Diagnostic Characteristics
Differs from box turtles and Blanding's turtle in lacking a hinged plastron. Differs from diamondback terrapin in habitat and having orange neck and leg skin in adults and a plain colored (vs. patterned) head in young.
Habitat
Wood turtles live along permanent streams during much of each year but in summer may roam widely overland and can be found in a variety of terrestrial habitats adjacent to streams, including deciduous woods, cultivated fields, woodland bogs, and marshy pastures (Ernst and Lovich 2009).
Preferred stream substrates are hard, such as gravel or sand, but streams with clay and, rarely, soft muck are used occasionally (Buech et al. 1990, Harding 1990, Buech et al. 1991, Ernst and Lovich 2009). Western populations are closely associated with water year round, and eastern populations tend to be more terrestrial in the summer (Ernst and Lovich 2009). According to Harding and Bloomer, Michigan wood turtles were never found more than 152 m (500 ft) from water and had leeches (evidence of aquatic habits) at all times of the year. New Jersey wood turtles were found farther from water and were free of leeches during summer months. Hatchlings and small juveniles are much more closely associated with water than adults.
Wood turtles are often associated with the margins of woods. Harding (1990) reported that wood turtles are usually found where openings in the streamside canopy allow growth of herbaceous plants. These openings provide both food and basking sites. In Wisconsin, wood turtles used wet mesic forest in river bottom and riparian shrub/forest ecotones. Most captures were in ecotones between alder thickets and grassy openings (Ross et al. 1991). In western Maine, within activity areas, wood turtles selected non-forested locations close to water with low canopy cover. Within a watershed, they selected activity areas close to streams with moderate forest cover and little open water. Overall, they appeared to select forest edges to balance thermoregulatory and feeding needs (Compton et al. 2002). Some agricultural operations may be locally beneficial by providing a mixture of different food and cover types near wooded streams (Kaufmann 1992). Use of woodland bogs and marshy fields is most common in the northern part of the range.
As with other turtles, nesting wood turtles require loose substrate on fully exposed (unshaded) sites, such as sandy banks or sand gravel bars in streams. When natural openings are unavailable, they may use such manmade disturbances as road grades, railroad grades, sand pits, or plowed fields (Ernst and Lovich 2009). In Minnesota, Buech et al. (1990, 1991) found that nesting habitat and stream substrate are the most important habitat determinants.
Overwintering occurs in bottoms or banks of streams where water flows all winter, including pools underneath a layer of ice. Underwater muskrat burrows, beaver lodges, or over bank root systems also may be used as winter hibernation (brumation) sites (Ernst 1986).
Ecology
Solitary from late spring to summer. May aggregate in or near hibernation sites. Not territorial (Kaufmann 1992, which see for a detailed study of social behavior in central Pennsylvania).
In New Jersey, wood turtles grow to 165 mm (6.5 inches) in 7 or 8 years. In Michigan, growth rates are slower, and it may take as many as 12 years to attain a 169 mm CL (Harding, 1990). Growth rates for males and females are constant until secondary sexual differences begin to appear, when males begin to grow faster, and ultimately become larger than females (Lovich et al. 1990). Harding (1990) found that average CL of females was 182 mm (n = 105), and average CL of males was 200 mm (n = 86). After and early growth spurt, growth of both sexes slows considerably, until by 20 years of age, growth rates are so slow that annual growth rings on the shell no longer yield accurate age data (Harding 1990).
The combination of late maturity, low reproductive success, and long-lived adults results in a population structure skewed heavily toward adults. Harding's study populations consisted of 80 to 85% adults. Farrell and Graham (1991) reported 3% juveniles (1 to 8 years), 53% subadults (9 to 13 years), and 34% adults (over 13 years) in one New Jersey population; almost half of the population comprised individuals over 14 cm in plastron length These characteristics combine to delay the detection of population declines, and to reduce the ability of small, declining populations to recover. A population studied in West Virginia included 46% juveniles (Niederberger and Seidel 1999).
New Jersey populations averaged 12.5 adults/ha, but the turtles were usually concentrated around basking areas or favorite food patches, rather than spread evenly across an area. In New Jersey, population density over several years averaged 10.7/ha of suitable habitat (Farrell and Graham 1991). In Michigan, the populations seem to be more scattered, and density is likely considerably lower. In southern Quebec, density was estimated at 1.2 turtles per 100 m of river (Daigle 1997). In West Virginia, estimated density was 19.1 individuals per hectare of total habitat (287-337 individuals along a 1.7 km length of river) (Niederberger and Seidel 1999). In Pennsylvania, density for 240 ha of available habitat was 0.66 turtles/ha, whereas density for available riparian habitat where most turtles occurred was 4.42 turtles/ha (Ernst 2001).
Adults may live for many years, with maximum ages of 32 years (wild caught) and 58 years (captive) reported by Harding and Bloomer (1979). In Pennsylvania, several known-age turtles marked as juveniles were found to live at least 30 to 42 years (Ernst, 1992, personal communication). Given the difficulty of aging turtles over 20 years, the wild caught age is likely conservative.
Reproduction
In Pennsylvania, secondary sexual characteristics began to appear at 5-9 years of age, at a size of 160 to 180 mm (Lovich et al. 1990). However, there is usually a delay of several years between sexual differentiation and sexual maturity. Maturity is apparently not attained until 12 to 15 years of age (Lovich et al. 1990, Farrell and Graham 1991, Harding 1990). In a long term study in Michigan, Harding (1990) reported that the smallest female found laying eggs was 158 mm carapace length and had twelve growth rings, indicating she was at least 12 years old. In New Jersey, attained maturity in 14th year (Farrell and Graham 1991). In Wisconsin, the youngest gravid female was 14 years old; the smallest male observed copulating was 20 years old (Ross et al. 1991). In Ontario age at maturity was 17-18 years (Brooks et al. 1992).
Reproductive activity (courtship, copulation) is aquatic (Ernst 1986). Copulation can occur in spring or fall (Niederberger and Seidel 1999, Ernst 2001, Ernst and Lovich 2009). Copulation occurs more often in spring in northern populations and more often during fall in southern populations (Farrell and Graham 1991, Kaufmann 1992, Ernst and Lovich 2009).
Depending on local climate, eggs can be laid anytime from mid-May to early July (Ernst and Lovich 2009. In New Jersey, Virginia, and Pennsylvania (Ernst 2001), a single clutch generally is laid in June. Clutch size usually is 3-20 (Ernst and Lovich 2009). Clutch size averaged 11 in Wisconsin (Ross et al. 1991), 8.5 in New Jersey (Farrell and Graham 1991), about 9 in Ontario (Brooks et al. 1992), and 10 in Michigan (Harding and Bloomer 1979). Eggs hatch after 70-80 days, August-October (after about 70 days, generally in late August, in New Jersey) (Ernst and Lovich 2009). Sex is genetically determined, and sex ratios are approximately 1:1 at birth (Ewert and Nelson 1991).
Nesting success generally is very low, with egg predators taking a heavy toll. One report conservatively estimated egg and hatchling mortality at 98% (Harding 1990). An Ontario population incurred a high rate of predation on nests and adults (Brooks et al. 1992). Reproductive success depends on a high rate of adult survival, long-lived adults that reproduce many times during their lifetime, and the occasional good season when a nest survives (Harding, pers. comm. 1992).