2011 NORTHEASTERN NATURALIST 18(3):315–334 Use of an Artificial Nesting Mound by Wood Turtles (Glyptemys insculpta): A Tool for Turtle Conservation Kurt A. Buhlmann1,* and Colin P. Osborn2 Abstract - We constructed an artificial nesting mound for Glyptemys insculpta (Wood Turtle) in the Great Swamp National Wildlife Refuge, NJ. The original nesting site was impacted by development and invasive plants. The new nesting mound was constructed from similar soil and was 100 m distant from the original site. The new mound was 50 m from stream habitat and in an old field where it received full sun for the entire day. The mound was 18 m long, 8 m wide, and had a maximum height of 1.5 m, with gently sloping sides. We encountered nest-searching female Wood Turtles on the impacted site during late May to early June for four years, 2007–2010. We carefully hand-carried females to the new nesting area and allowed them to choose whether or not to nest on the mound. We protected all nests from predators. Seventeen of 18 nests that were deposited and left on the nesting mound produced live hatchlings. Six clutches had 100% hatching success, with only one failing completely. At least nine different female Wood Turtles nested on the mound. One female returned on her own in three subsequent years, another returned on her own in one subsequent year, several were re-shown the mound in subsequent years, and one turtle found the mound and nested on it on her own. The nesting mound has produced 142 hatchling Wood Turtles over the four years. We suggest that it is possible to entice female turtles to nest in a new area and that when nesting resources are limited, construction and protection of nesting areas can be a useful conservation action. Introduction The problems facing turtle populations worldwide are sobering. Of 305 or more recognized species, at least 42% are considered threatened (IUCN 2009). Major causes of population declines include loss of habitat through human development and habitat alteration, and collection of specimens for food and the pet trade markets (Bailey et al. 2006, Gibbons et al. 2000, Turtle Conservation Fund 2002, van Dijk et al. 2000). Conservation efforts for turtles have focused primarily on habitat protection, with turtles fortunately being included in large tracts of federal and state natural areas, parks, refuges, and wildlife management areas. However, simply setting aside areas and managing habitat for turtles may not be sufficient to sustain or achieve viable populations as unaddressed threats continue to erode population size. These factors may include mortality by automobiles on adjacent roadways (Steen et al. 2006), poaching by unscrupulous persons for the pet trade (Garber and Burger 1995, Levell 2000, Niederberger and Seidel 1999), certain agricultural practices (Saumure and Bider 1998, Saumure et 1University of Georgia, Savannah River Ecology Laboratory, Aiken, SC 29802. 2US Fish and Wildlife Service, Great Swamp National Wildlife Refuge, 241 Pleasant Plains Road, Basking Ridge, NJ 07920. *Corresponding author - kbuhlmann@earthlink.net. 316 Northeastern Naturalist Vol. 18, No. 3 al. 2007) and increased predation by subsidized meso-predators, namely Procyon lotor L. (Raccoon) and dogs (Vulpes and Canis spp.). Therefore, techniques are needed to enhance or augment turtle populations on protected sites, especially where populations exist at densities below their historic levels. Techniques that produce turtle conservation benefits for limited cost and maintenance are highly desirable. We report here on a technique that we have used since 2007 in the Great Swamp National Wildlife Refuge, NJ to provide secure nesting areas for state-threatened Glyptemys insculpta LeConte (Wood Turtle) (NJDEP 2009). Our goal has been to help compensate for the loss of historic nesting sites due to human habitat alteration on adjacent private land and perceived intense levels of turtle nest predation by Raccoons. Wood Turtles occur in the northeastern US and Canada (Ernst and Lovich 2009). They inhabit streams with gravel or cobble bottoms, deep pools, and undercut banks with overhanging trees. Hibernation and mating take place in streams. Much of the active season is spent in terrestrial habitats, notably riparian hardwood forests and old fields where these omnivores feed on slugs, earthworms, insects, and berries (Arvisais et al. 2004, Farrell and Graham 1991, Harding and Bloomer 1979). Brooks et al. (1992) found that age at maturity for Wood Turtles in Canada was 17–18 years at a minimum carapace length of 185 mm for females and 199 mm for males. Farrell and Graham (1991) suggested that Wood Turtles may mature by age 14 in a New Jersey population. Longevity of Wood Turtles in the wild has been reported to 33 years (Ross et al. 1991). One of us (K.A. Buhlmann) marked adult Wood Turtles in 1988 in Virginia, and two were found alive at the same site in 2009 (T. Akre, Longwood University, Farmville, VA, pers.comm.), and thus are +40 yrs of age. Wood Turtles in all populations studied lay only one clutch of eggs per year, and some may skip years (Walde et al. 2007). Mean clutch size was 10.6 eggs (range = 5–15) in a Québec study (Walde et al. 2007), while Farrell and Graham (1991) reported clutch means of 8.5 eggs (range = 5–11) in a New Jersey population. The largest clutch size yet documented was 20 eggs in a Quebec population (Walde and Saumure 2008). Hughes et al. (2009) reported incubation periods of 79–92 d in Ontario, and Walde et al. (2007) reported a range of 60–116 d over two years in Québec. Farrell and Graham (1991) recorded emergence of hatchlings from semi-natural nests 70–71 days after laying. Nesting occurred in early June in New Jersey (Farrell and Graham 1991) and mid-late June in Québec (Walde et al. 2007). Hatchlings emerged 23–28 August in New Jersey (Farrell and Graham 1991), 13–29 August in New Hampshire (Tuttle and Carroll 2005) and 17 August–7 October in Québec (Walde et al. 2007). Sex of the offspring is genetically determined, rather than by incubation temperature (Ewert and Nelson 1991). Nesting sites include open sloping areas consisting of welldrained gravelly or sandy soils with sparse-to-no vegetation and are associated with abandoned gravel pits, quarries, roadway embankments, agricultural fields, and natural sandy beaches (Castellano et al. 2008, Harding and Bloomer 1979, Hughes et al. 2009, Walde et al. 2007). 2011 K.A. Buhlmann and C.P. Osborn 317 Field-Site Description At the Great Swamp National Wildlife Refuge (NWR) in Morris County, NJ, we have studied a small population of Wood Turtles along a tributary of the Passaic River since 2006. The exact location of the study site is not published in order to help further conservation goals. The stream at our study site is typical of previously described habitat, with flowing current, gravel bottom, deep pools, and undercut banks with overhanging trees that provide suitable hibernation sites. The stream reach studied is bordered by riparian hardwood forest and old abandoned pastures that contain patches of Rubus sp. (blackberry) and invasive Rosa multiflora Thunb. ex Murr (Multiflora Rose). The Wood Turtle population has persisted presumably because it is isolated from any major roads (one road mortality has been observed in the past four years) and is protected from poaching by both private property (with supportive landowners and where trespassing would be obvious) and the federal land of the Refuge. An adjacent old farm contained a previously active spoil area where the farmer had dumped compost, manure, some trash, and occasionally excavated dirt. The spoil area provided a suitable nesting area for Chelydra serpentina L. (Snapping Turtle), Chrysemys picta Schneider (Painted Turtle), Sternotherus odoratus Latreille (Common Musk Turtle), Terrapene carolina L. (Eastern Box Turtle), and Wood Turtles. We had observed several Wood Turtles nesting on the old farm property in 2006. We also observed high levels of turtle nest destruction of co-nesting Painted and Snapping Turtles and assumed that Wood Turtle nests suffered the same fate. Nest predation was caused chiefly by Raccoons, but also by Vulpes vulpes L. (Red Fox). The active farming had ceased in 2005, the land was sold, and the nesting site began to be reclaimed by herbaceous vegetation, including Artemisia vulgaris L. (Common Mugwort), a non-native invasive plant. Simultaneously, we learned of plans to subdivide the property and build a large home on the nesting site. Thus, we began discussions about providing an alternative nesting site, how it would be constructed and located, whether nest predation could be deterred, and most importantly, whether female Wood Turtles would use it and nests hatch successfully. Methods Marking We captured Wood Turtles at our study site by hand in the stream (adults and juveniles) and on land adjacent to the stream in either the old farm nesting area (females) or in the riparian forest (adults of both sexes). All turtles were measured (carapace [CL] both maximum and midline, plastron [PL] and weight [nearest g]). All turtles were individually marked by filing marginal scutes (Cagle 1939) and using the 1-2-4-7 marking system (e.g., Buhlmann et al. 2008, Honegger 1979). We X-radiographed gravid Wood Turtles in 2006 (n = 2) and 2007 (n = 7) to determine total clutch sizes; turtles were not Xrayed in subsequent years. 318 Northeastern Naturalist Vol. 18, No. 3 Nesting mound At the beginning of the 2007 turtle nesting season, we constructed a mound of soil on Great Swamp NWR in an old field adjacent to the stream that supported the Wood Turtle population. The field was located directly across the stream from the old farm nesting area. We obtained soil that was similar in color, friability, and texture to the soil at the old farm nesting area. We used a backhoe to construct the mound (Fig. 1). The mound was 18 m in length and 8 m in width at the base, with a maximum height of 1.5 m. The sides of the mound were sloped at a 40 degree angle, and the top of the mound was flat and 2 m in width. The location of the mound in the old field was positioned with the long sides facing north and south, and it received full sun for the entire day. The mound was located approximately 25 m into the old field from the 25-m-wide riparian forest; thus, the mound was 50 m from the stream. Straight-line distance between the old farm nesting site and our nesting mound was 100 m. In May 2008, we added soil and rounded the ends of the existing structure to increase the availability of nesting areas facing east and west. The sloping front of the existing mound was given a porch-like base to minimize the length of the angled slope and reduce erosion (Fig. 2). Our goal was to make the mound visible to gravid female Wood Turtles that were in the riparian woodland or the old field. We hand-planted the slopes Figure 1. We used a backhoe to construct a mound of dirt in an open canopy area of the old field at Great Swamp National Wildlife Refuge, NJ. 2011 K.A. Buhlmann and C.P. Osborn 319 of the mound with mature clumps of translocated grasses, mostly Sorghastrum nutans (L.) Nash (Indian Grass) that we obtained from the old farm nesting site. Grass clumps were planted irregularly, but approximately 1 m distant from each other and mimicked conditions observed on the old farm nesting area, prior to Common Mugwort colonization. Grass clumps presumably provided some concealing structure for female turtles seeking a nest site, but did not shade developing nests, nor were they close enough together to create interlocking root mats that are known to cause turtle embryo mortality (Harding and Bloomer 1979, Hughes et al. 2009). We hand-weeded the mound prior to each nesting season. Grass clumps also helped prevent erosion caused by rainfall. Using the backhoe, we also constructed two small wetlands in the old field approximately 5 m x 5 m in size within 3 m of the north- and west-facing toes of the mound. Our thought was to provide an optional area for nestsearching females to recharge their bladder water or for hatchlings to escape into aquatic cover after emergence. Treatment of gravid female Wood Turtles In each of the four years of the study, we searched suitable nesting areas on the old farm property for gravid females during the day and evening from Figure 2. The mound we created at Great Swamp National Wildlife Refuge, NJ was 18 m in length and 8 m in width at the base. The maximum height was 1.5 m. The sides of the mound were sloped at a 40-degree angle and the top of the mound was flat and 2 m in width. 320 Northeastern Naturalist Vol. 18, No. 3 approximately the third week in May to the end of the second week of June. We determined gravidity by hand palpation and attached a radio transmitter to a posterior side of the carapace. After processing, females were returned to the stream, monitored, but allowed to reemerge to nest at their choosing. Processed females that were subsequently found searching for nests on the old farm were gently picked up by hand, carefully carried (horizontally with head facing away from researcher) the 100-m distance to the constructed nest mound. The researcher’s walking pace was steady with no distractions or detours, and took approximately 2 minutes. We then retreated out of sight of the turtle to a concealed location 30 m away and observed her behavior through binoculars. Treatment of individual females was highly variable and depended on time of day encountered, previous or new capture, and year (Table 1). The goal was to introduce females to the constructed mound quickly and with as little disturbance as possible. Nesting cage We installed an experimental “nesting cage” on the top of the mound (Fig. 3). The goal of the cage was to provide an opportunity for nest-searching female turtles to enter and nest under a wire mesh protective screening that made it diffi cult, if not impossible, for predators, such as Raccoons, Red Fox, and Corvus brachyrhynchos Brehm (American Crow), to access the buried nests. Our pilot nesting cage design was 3.6 m long, 1.2 m wide, and 0.6 m tall. The 7.7-cm (3-in) gap or slot was installed along the entire length of the cage and was constructed between two wood beams (one buried 2 x 4” and one above ground 2 x 2”; Fig. 3) that would allow adult Wood Turtles, but not predators, to crawl through. Gap height can be modified for select turtle species; the 7.7-cm gap also excluded large Snapping Turtles. (The list of materials needed for construction is available from the authors). Nest protection Turtle nests that were constructed on the slope of the mound were protected immediately after egg deposition by installing a 1- x 1-m piece of 0.64-cm (¼-inch) galvanized hardware cloth (Fig. 3; e.g., Baskale and Kaska 2005). The nest was centered under the hardware cloth, and the ends of the hardware cloth were secured with 6-inch-long metal landscape stakes. Approximately 2 wks prior to egg hatching, the 1- x 1-m piece of hardware cloth was replaced with a 25.5-cm-wide x 30.0-cm-long x 6.5-cm-tall cage made from the same 0.64-cm hardware cloth. At the lower portion of the cage, a 1.9-L (1/2-gal.) bucket was buried with half of the top under the cage. Moist Sphagnum sp. was added, and a wood lid was placed on the other half of the bucket top and weighted with a large rock (Fig. 4). In 2010, we also encircled the base of the mound with a 30-cm-tall aluminum drift fence, with four 19-L buckets to detect and capture hatchlings from unknown, but suspected, nests. Buckets were covered with 30- x 30-cm plywood covers and a cinder block on four 5-cm 2011 K.A. Buhlmann and C.P. Osborn 321 tall pedestals, which allowed hatchlings to fall into the bucket, but prevented predators from reaching in to access them. Thus, hatchling turtles that emerged from nests were captured in buckets and remained moist, shaded, and protected from predators until retrieved by biologists on daily inspections. All nests were exhumed post-hatching to determine the fate of unhatched eggs and match clutch sizes to X-radiographs (2007 only). We determined whether unhatched Table 1. Wood Turtle (Glyptemys insculpta) use of a constructed nesting mound at the Great Swamp National Wildlife Refuge, NJ, 2007–2010. Behavior codes: 0 = not shown mound; 1 = shown mound, nested that night; 2 = shown mound, left, returned on own, nested a subsequent night; 3= shown mound, nested elsewhere; 4 = re-shown mound in subsequent yr, nested; 5 = returned to mound on own in subsequent yr, nested; 6 = returned to mound on own in subsequent yr, inspected, but nested elsewhere; 7 = found mound on own, nested; 8 = never on mound, nested elsewhere; 9 = nested on mound on own, but unknown if turtle found it on its own or was returning in a subsequent yr after being shown it previously. Nest Behavior Clutch Partially Days Female Year date code size Hatch Infertile developed Emergence date incubated F12 2007 3 Jun 2 10 10 0 0 24 Aug 82 F25 2007 4 Jun 2 11 5 4 2 4 Sep 92 F22 2007 6 Jun 1 10 10 0 0 29 Aug 84 F24 2007 3 Jun 0 13 n/a n/a n/a n/a n/a F26 2007 13 Jun 2A 7 0B n/a n/a n/a n/a F21 2007 6 Jun 0 16 1B n/a n/a 31 Jul 55C F12 2008 4 Jun 5 9 6 2 1 22 Aug 79 F26 2008 4 Jun 5 10 5 4 1 22 Aug 79 F28D 2008 4 Jun 8 10 7 0 3 15 Aug 72 F22 2009 21 May 4 10 10 0 0 18 Aug 88 F23 2009 29 May 1 10 7E 3 0 31 Aug 94 F25 2009 31 May 4 13 8 5 0 2 Sep 94 F12 2009 1 Jun 5 9 7 2 0 3 Sep 94 F21 2009 2 Jun 1 13 13 0 0 5 Sep 95 F30 2009 2 Jun 7 9 0 5 4 n/a n/a F31 2009 7 Jun 2 12 8 4 0 8 Sep 93 F26 2009 10 Jun 4 9 9E 0 0 14 Sep 96 F26 2010 26 May 5F 9 9 0 0 10 Aug 76 F12 2010 27 May 5 8 7 1 0 6 Aug 71 F14 2010 27 May 2 11 10 1 0 7 Aug 72 F21G 2010 28 May 6 14 9 4 1 29 Jul 62 UN1 2010 n/a 9 12 8 4 0 8–9 Aug n/a UN2 2010 n/a 9 11 10 0 1 10–13 Aug, 29 SeptH n/a AShown mound twice, then nested. BClutch moved to incubator. CIncubated indoors. DNot a mound-using turtle, but nest found and protected. EHatchlings escaped cage and were not marked. FReturned on own, but left, was re-shown mound and nested. GNest protected on old farm site. H29 September is capture date, not hatch date. 322 Northeastern Naturalist Vol. 18, No. 3 eggs had been infertile (hard yellow yolks with no blood vessels) or whether embryos had died in development. During 2009, we installed Hobo Temperature dataloggers (Onset Computer Corporation, Pocasset, MA) in 4 nests on the mound, one each on north-, east-, Figure 3. We installed an experimental “nesting cage” on the top of the mound. Turtle nests that were laid on the slope or porches of the mound were protected immediately after egg deposition by installing a 1- x 1-m piece of 0.64-cm (¼-inch) galvanized hardware cloth. 2011 K.A. Buhlmann and C.P. Osborn 323 south-, and southwest-facing slopes in order to characterize differences in incubation temperatures. Data loggers were buried adjacent to the middle of each clutch and recorded temperature hourly. Results Use of the constructed nesting mound Gravid females were often found exploring the nesting areas during the day, and three turtles constructed nests during the day. Of those, two were nesting at 11:00 h, and one at 17:00 h. We observed 18 instances where females initiated nest construction in the evenings between 19:00–20:00 h. On 2 June 2007, we located Female #12 (F12), on the old farm nesting area at 19:20 h. She was actively searching for a nest site and was using typical nest-searching behavior such as placing her snout to the ground as if sniffing and flipping dirt with her front feet (Harding and Bloomer 1979). F12 was the first turtle we introduced to the newly constructed nesting mound, and she was carried and placed on the mound’s north-facing slope. F12 immediately resumed her nest searching behavior and within an hour had begun nest construction with her hind feet. However, at 21:35 h she abandoned the nesting attempt, slipped down off the mound, and traveled the 50 m back to the stream. At dusk, 20:05 h the following evening, we scanned the nest mound from our concealed location and were surprised to see her on the north slope Figure 4. Nest covers were modified close to time of hatching to capture the hatchlings. 324 Northeastern Naturalist Vol. 18, No. 3 of the mound (having returned to it on her own), where she was finishing a nest in which she deposited 10 eggs. We protected the nest after she finished at 20:30 h. All 10 eggs hatched on 24 August 2007 (Table 1). The use of the nest mound by F12 encouraged us to continue the study. Three other Wood Turtles were treated similarly in 2007 (n = 4). F22 nested immediately on the mound the same night, F25 behaved similarly to F12, while F26 was shown the mound two different nights and nested there on the second (Table 1). F24 was not shown the mound, but was found near the mound the next night without eggs and her nest was not located. F21 was not shown the mound, and because she subsequently moved 535 m upstream and off the Refuge, we recaptured her, induced oviposition with oxytocin and incubated her eggs in captivity. We also moved F26’s eggs to an incubator (set at 25 °C) as a hedge against total loss of that year’s reproductive effort if the mound proved to be an unsuitable site for egg incubation; however, the incubator malfunctioned, and only one egg hatched from F21’s clutch, and none from F26. In 2008, F12 was discovered nesting on the mound on her own on 4 June. She constructed her nest within 5 m of her 2007 nest. F26 also returned to the mound on her own and successfully nested (Table 1). One Painted Turtle hatchling was found at the base of the mound in April 2008, indicating that this species also nested on the mound in 2007 and that a hatchling successfully over-wintered in the nest. The mound also produced 38 Snapping Turtle hatchlings in September 2008 from an unknown number of nests, as well as another Painted Turtle hatchling. Two female Eastern Box Turtles inspected the mound, but did not nest on it; one nested 10 m away in the old field. In 2009, F12 once again returned to the mound and nested on 1 June. Seven other Wood Turtles nested on the mound in 2009 (n = 8). These included F22, F25, and F26, who were again found on the old farm nesting area, and F21, who was again found upstream as in 2007. They were transported, and three nested on the mound the night it was re-shown to them; F22 returned to the stream that night but returned to the mound on her own the next night and nested. F23 and F31 were found on the old farm site and were shown the nest mound for the first time; F23 nested that night, whereas F31 hid in the small wetland at the base of the mound for two days before nesting on the mound. F30 found the mound on her own and nested. F24 was again found near the mound, as in 2007, but she was not gravid. Thus, in 2009, one female returned to the mound on her own, one found it for the first time on her own, and all six females that were shown the mound in 2009 nested on it either that same night or within 1–2 nights. In 2010, the nesting season began early, and two Wood turtles nested on the mound without our knowledge (UN1 and UN2; Table 1), but the eggs hatched without protection. F12 again returned and nested; maximum distance between all four years’ nests was 6 m. F21 inspected the mound (displayed nest-searching behavior), but left and nested at the old farm. F26 returned to the mound on her own, inspected, but left and traveled upstream 865 m over 4 d. We decided to recover 2011 K.A. Buhlmann and C.P. Osborn 325 her and re-showed her the mound; she left, but returned on her own and nested on it the next night. F31 returned on her own, inspected the mound, but her nest was not confirmed. F14 was transported to the mound and nested there for the first time. In total, at least 9 different turtles representing 60% of our known female population (n = 15) have nested on the constructed nest mound, 2007–10. We observed Wood Turtles, as well as an Eastern Box Turtle, entering the nesting cage, but they did not nest inside. Wood Turtles preferred to nest on the 40-degree slopes of the nest mound; only three of 18 nests were at the top of the mound. One small Snapping Turtle did nest inside the cage and there were no indications that predators ever squeezed inside. Hatching success Three nests deposited on the mound in 2007 produced viable hatchlings; 25 hatchlings emerged from 31 eggs between 24 August and 4 September (Table 1). The number of eggs deposited matched the number of eggs on the X-radiographs and gave us confidence that we could accurately estimate clutch size, post hatching. Both nests deposited on the mound in 2008 produced hatchlings (11 hatchlings from 19 eggs, 22 August). In 2009, 7 of 8 nests produced hatchlings (62 hatchlings from 85 eggs, 18 August–14 September). In 2010, 5 nests on the mound produced 44 hatchlings from 51 eggs (43 hatchlings, 6–13 August). The 44th hatchling was found 29 September in a drift fence bucket. It displayed growth on its annuli, thus we presume it emerged with the others, but lived and grew in the grasses at the base of the mound before being captured in a bucket. In total (2007–10), 17 of 18 nests (186 eggs) constructed on the nest mound produced 142 hatchlings (76%), 35 infertile eggs (19%), four embryos that died in early development (no shell or skeleton discernible, 2%), and five that died in later development (3%). Six clutches had 100% hatching success, and one failed completely. In addition, one egg each from F13 in 2006 and F21 in 2007 were successfully incubated indoors, F28’s nest of 2008 (not on the mound) was protected and produced 7 hatchlings, and F21’s nest of 2010 (not on the mound) was protected and produced 9 hatchlings. Thus, 160 Wood Turtle hatchlings have been released into the study site. Overall, mean clutch size for this population was 10.7 eggs (n = 23 clutches, 246 eggs, range = 7–16; Table 1). Duration of clutch incubation on the mound varied among the three years, with 2010 being the shortest (71–76 d, n = 3 clutches) < 2008 (79 d, n = 2) < 2007 (82–92 d, n = 3) < 2009 (88–96 d, n = 7). Interestingly, the overall shortest incubation observed was 62 d for F21 in 2010. Her nest was not on the mound, but on the old farm in a patch of black shredded rubber and crushed stone with natural soil substrate below. Incubation temperature mean of means was 24.0 °C for the four 2009 nests. Daily temperatures ranged as low as 13.8 °C and as high 326 Northeastern Naturalist Vol. 18, No. 3 as 37.8 °C for F22’s nest (Fig. 5). F22’s hatchlings (10 of 10) emerged on 18 August, the day after that nest experienced its highest temperature. F30’s nest on the SW portion of the mound failed completely; four embryos were in advanced development. F30’s nest experienced a maximum temperature of 39.3 °C on 15 July. Depths of the egg chambers for two nests measured in 2008 were 9.5–14.5 mm (F12) and 7.5–13.5 mm (F26). Hatchling releases Each year, the hatchlings were removed from the capture buckets that surrounded their nest cage and were measured and marked (marginal scutes clipped with nail cutters). The 2007 hatchlings were cohort marked by nest, but hatchlings in subsequent years received individual marks. All hatchlings were released within 2 days of capture. Release sites included the base of the nest mound, the edge of the main stream, the riparian woodland, and small feeder streams. The exact release location of each hatchling was recorded, and future recaptures may help elucidate which sites provide higher survivorship. The late hatchling from 29 September 2010 was fitted with a microtransmitter (0.2 g) and released at the mound on 29 October. It traveled 95 m from the mound in the old field, moving parallel to the stream. Figure 5. Maximum and minimum nest temperatures of F22’s nest deposited 21May 2009 on the east-facing portion of the mound. Ten hatchlings emerged from 10 eggs on 18 August 2009. Dataloggers were installed 15 June–26 August and temperatures recorded hourly. Lowest temperature during this period was 13.8 °C (17 June); highest was 37.8 °C (17 August) and hatchlings emerged the next day. The average temperature between 15 June and 17 August was 24.3 °C. The median temperature was 23.5 °C. 2011 K.A. Buhlmann and C.P. Osborn 327 It survived multiple nights of sub-freezing temperatures by nestling into forms in grasses but was found dead on 17 November, having never reached the stream. Discussion Understanding turtle life histories is essential for effective conservation planning. Most turtle species are long-lived, and individuals require many years to reach sexual maturity (often 15–20 years) and they live for a long time (40–70 years), depending on species (Congdon and Gibbons 1990). Once mature, they have many years over which to produce offspring. Maintaining stable turtle populations require that the annual adult survivorship is relatively high (>96%), and that the juvenile survivorship (although unknown for most species) is also high (Congdon et al. 1993, Tuberville et al. 2009). Nest and early hatchling survivorship is perceived to be low, but it is unclear if human alterations to the environment and associated increases in subsidized meso-predator abundance have not altered the true pattern. At our study site, the overall adult population size is currently known to contain 20 individuals (5 males and 15 females). We also captured 2 juveniles, not including hatchlings released. This is small relative to some other Wood Turtle populations studied. For example, approximately half of 316 animals were adults in an earlier New Jersey study (Farrell and Graham 1991); 77 turtles (21 males, 56 females) were studied in Ontario (Brooks et al. 1992), and 188 turtles (55 males, 83 females, 50 immatures) in Québec (Walde et al. 2003). Given that our population is located on protected land, the adult population seems stable, and all typical Wood Turtle habitat components are present (except for the potential impending loss of the old farm nesting site), we constructed the alternative nesting site and actively protected nests from predation in an effort to boost population recruitment. Use of the constructed nest mound Our goal of this study was to determine if female Wood Turtles would alter their nesting behavior and accept a new nesting site. Hand carrying of nestsearching gravid female Wood Turtles was considered successful as none voided their bladder water (e.g., Kinney et al. 1998) and all resumed some nest-searching behavior (12 of 12 occurrences) once placed on the nest mound, although not all completed a nest at that time. We found this particularly interesting, as Wood Turtles are known to be sensitive to disturbance prior to initiation of oviposition (Walde et al. 2007). We caution that our hand-carrying technique is unlikely to work with “high-strung” species, notably Snapping Turtles and Deirochelys reticularia Latreille (Chicken Turtle) (Buhlmann et al. 2009). The experimental nest mound was thus more successful than we initially expected. One turtle (F12) nested on it for three consecutive years after the initial introduction (total = 4 yrs). Some females returned to the old farm site in subsequent years and were reintroduced to the mound and nested there. Two females returned and investigated the mound on their own, but then left; one (F21) nested on the old farm, while F26 was re-shown the mound and 328 Northeastern Naturalist Vol. 18, No. 3 nested there. It is presumed that these turtles were not yet ready to nest, but they may also have been disturbed by us. (Walde et al. 2007). Two females that nested on the mound in 2007 were not seen there again until 2009. When radio-tracked, one of these females (F22) was found gravid, but we never saw her nest; it is possible that the other (F25) skipped a reproductive bout as suggested by Walde et al. (2007). These varied behaviors indicate that nestsite fidelity is perhaps stronger in some females than others, and whereas one female accepted the new alternative site immediately and used it for four years, other females returned to the original nesting site in a subsequent year, and needed multiple introductions to the new site. Site fidelity to nesting sites, although variable, has been well-documented in a number of turtle species. Szerlag-Egger and McRobert (2007) found that nest-site selections in Malaclemys terrapin Schoepff (Diamondback Terrapin) varied greatly from approximately 4–1307 m (mean inter-nesting distance = 202.75 m), yet 39% were recaptured within 50 m of their initial nesting location. Standing et al. (1999) found that 73.3% of Emydoidea blandingii Holbrook (Blanding’s Turtle) returned to nest on the same beach. Rowe et al. (2005) found that 29.4% of Painted Turtles in their study nested between 0 and 25 m from their nests of the previous year, although some individuals nested as far as 648 m. It is unknown what factors will make a female switch her nesting site location. Under natural conditions and over the course of a turtle’s long life, certain nesting sites likely become unsuitable (i.e., due to plant succession), just as new sites would become available (i.e., tree blow-downs, fires). Therefore, turtles must have the ability to modify their choice of nesting sites. Given that Wood Turtles in our population returned to the new nesting mound in subsequent or alternate years or found it on their own initially, it seems that the turtles became aware of the new nesting site resource in their landscape. Another possible explanation for the results we observed could be that the old farm nesting site was marginal and Wood Turtles were receptive to finding a more suitable nesting site. This hypothesis might also explain the small population size and apparent dearth of juvenile animals. Protection of nests We prevented egg predation by covering each nest immediately after each female finished nesting. Returning the next day, we often found attempts by Raccoons to dig up the nest. Raccoon predation has been reported to cause 85–97% failure of sea turtle nests (Engeman et al. 2006, Ratnaswamy and Warren 1998). Congdon et al. (2000) reported that predation of Blanding’s Turtle nests averaged 78% over a 23-yr study and was 100% for nine of those years. Horne et al. (2003) reported 42–100% nest mortality during a 4-yr study of Graptemys flavimaculata Cagle (Yellow-Blotched Map Turtle). Brooks et al. (1992) reported that predators destroyed 15 of 17 nests and injured 60% of adult Wood Turtles in their study. Nest screening, along with relocation and fencing, have been demonstrated to increase hatching success of Caretta caretta L. (Loggerhead Sea Turtle) nests 2011 K.A. Buhlmann and C.P. Osborn 329 against predation and inundation (Baskale and Kaska 2005). Although two nests survived unprotected in 2010, we are relatively certain that without the use of nesting covers, most of our Wood Turtle nests would have been destroyed by predators within the first or second night after oviposition. Christiansen and Gallaway (1984) removed Raccoon predators from a study site containing Kinosternon flavescens Agassiz (Yellow Mud Turtle) and reported a coincidental increase in numbers of juvenile turtles captured 3 and 4 years later. We intend to continue this study and will document all observations of juvenile Wood Turtles, previously marked and unmarked, in future years. A potential drawback of the nesting mound is that, due to its relatively small size and its popular use by nesting turtles, a concentration of eggs may attract mammalian predators. Each year, multiple Snapping Turtles also nested on the mound, but we did not protect those nests, and those nests were regularly destroyed. However, nests were regularly destroyed on the old farm nesting area as well. Our placement of the nesting cage structure on top of the mound was an attempt to provide Wood Turtles with a suitable nesting site that predators could not access and would not require a researcher’s presence to manually protect the nest after oviposition. Similar cages have been installed on the causeway along a Georgia salt marsh for use by Diamondback Terrapins, and the nests deposited inside were not depredated (K. Buhlmann, A. Grosse, T. Norton, and B. Crawford, unpubl. data). Our future intention is to construct new nesting cages and place them on the slopes of the nesting mound, thus better correlating automatic nest protection with preferred microhabitat nest-site selection by Wood Turtles. Hatching success Our nest mound produced live hatchlings from 17 of 18 nests that incubated there. Our egg incubation period was similar to ranges reported for natural nests in New Hampshire (76 d; Tuttle and Carroll 1997) and Québec (77–86 d, Walde et al. 2007), although variable by year. Hughes et al. (2009) recorded a temperature range of 8.5–41.0 °C for five natural nests that produced live hatchlings. Our nest temperatures in 2009 were also highly variable (Table 2, Fig. 5). Our hottest nest reached 39.3 °C, but was the one nest that did not produce live hatchlings. Table 2. Four dataloggers installed in nests of the Wood Turtle (Glyptemys insculpta) on a humanconstructed nest mound at Great Swamp National Wildlife Refuge, in NJ from 15 June 2009 through each nest’s egg incubation period. To facilitate comparisons of nest temperature conditions among the four nests, we summarized the data below for 15 June–26 August, when all four dataloggers were operating at the same time. Female Location Hatching Mean nest Median nest Nest temperature nest on mound success temperature temperature range (min–max) F30 SW 0 of 9 24.1 °C 23.3 °C 13.5–39.3 °C F25 S 8 of 12 24.2 °C 23.4 °C 14.0–38.4 °C F22 E 10 of 10 24.3 °C 23.8 °C 13.8–37.8 °C F26 N 9 of 9 23.5 °C 23.3 °C 14.9–32.1 °C 330 Northeastern Naturalist Vol. 18, No. 3 Our nest mound has produced live hatchlings from nests deposited on slopes facing north, east, and south. Our only total nest failure was on the southwest-facing slope, and no nests have been deposited on the west-facing slope. Although Wood Turtles are known to have genetically determined sex, Hughes et al. (2009) found that Wood Turtles selected for warm and variable nest temperatures which led to faster embryo development. However, the construction of artificial nesting areas for other turtles with temperaturedependent sex determination will need to entail particular attention to the thermal characteristics of the site. Thoughts on hatchling releases We chose a variety of sites, both terrestrial and aquatic, for release, including allowing some hatchlings to find their own way from the base of the nest mound, which was 50 m from the stream. Castellano et al. (2008) found that some hatchlings remain on land in agricultural fields from several days to weeks following emergence from the nest and that they grow during that time. Tuttle and Carroll (2005) powder-tracked 53 hatchlings and were able to document that at least 12 made it to a stream, taking 1–24 d between mid-August and early September to make the journey. The one hatchling we radio-tracked and which did not survive was probably released too late in the season to find the aquatic habitat. We suspect that the trek from nest site to stream is a high-risk period for hatchling survivorship. Accordingly, the old field habitat surrounding the nest mound is mowed only in winter, when Wood Turtles are in the stream (Saumure et al. 2007). Tuttle and Carroll (2005) also suggested that direct release into the stream may be equally risky, as hatchlings clearly feed on land and make overnight forms; those that initially survive may be those that find vegetative cover near the stream. Natural nesting sites selected by Wood Turtles are elevated and well-drained and likely represent a tradeoff between being far enough and high enough away from the stream to prevent flooding, but also close enough to allow hatchlings to find and access the water (Hughes et al. 2009). We have not yet observed released hatchlings in subsequent years. Future recaptures of these juveniles should shed some light on the best release strategies given that population recruitment is our goal. At this point, we are arguably more concerned with enhancing hatchling survivorship and recruitment and less so with assuring a link between the hatchlings and their nest site fidelity as future breeding females. However, continued population monitoring and study is needed. Conclusion It appears that female Wood Turtles can be convinced to use a new nesting area and that physically showing them the site is a successful way of making them aware of its existence. Further experimentation with automatic nestprotection devices may yet foil predators and increase turtle hatching success, as well as require less manpower and monitoring. Alternative nest sites should 2011 K.A. Buhlmann and C.P. Osborn 331 be conspicuous to nest-searching females. We suggest that construction of alternative nesting sites may help recover turtle populations where natural nesting habitat is not available or has been impacted. Thus, in densely humanpopulated areas, small protected areas may yet function to maintain viable populations if all the necessary habitat components are clustered within the protected area boundaries. Acknowledgments We thank Michael Horne for constructing the nest mound for us with the backhoe and Susi Ponce for thoughts on the initial nesting-cage design. We also thank Len Soucy and technicians of the New Jersey Raptor Trust for X-radiographing turtles, Rich Seigel and Jackie Record for providing historical information about Great Swamp Wood Turtles, and William Koch, Refuge Manager, for his support of this project. We thank the following individuals who helped to track and locate Wood Turtles in the field: Heather Barrett, George Cevera, Kean Clifford, Tom Clifford, Greg Cooper, Charlie Cotton, Ray Farrell, Andrew Ferreira, Bridget Goldsmith, Anthony Henehan, Steve Henry, Chris Hernandez, Peter Hrinewski, Georgina Jacquez, Helen Johnson, Marilyn Kitchell, Scott Kuhn, Marnie Miller-Keas, Dave Moskowitz, Laura Newgard, Adam Osborn, Hanina Osborn, Joe Pignatelli, Harry, Harriet, and Harley Spaven, Tim and Marcia Stevens, Kelly Triece, Tracey Tuberville, Peter Warny, Bob Zappalorti, and Brian Zarate. We thank Tony Cullen and Dave Miller for mowing the old field each winter. We thank Brett DeGregorio and three anonymous reviewers for greatly improving earlier drafts of the manuscript. Manuscript preparation was supported by the US Department of Energy under Award Number DE-FC-09-075R22506 to the Savannah River Ecology Laboratory, and by Buhlmann Ecological Research, LLC. Literature Cited Arvisais M., E. Levesque, J-C. Bourgeois, C. Daigle, D. Masse, and J. Jutras. 2004. Habitat selection by the Wood Turtle (Clemmys insculpta) at the northern limit of its range. Canadian Journal of Zoology 82:391–398. Bailey, M.A., J.N. Holmes, K.A. Buhlmann, and J.C. Mitchell. 2006. Habitat management guidelines for amphibians and reptiles of the southeastern United States. Partners in Amphibian and Reptile Conservation Technical Publication HMG-2, Montgomery, AL. 88 pp. Baskale, E., and Y. Kaska. 2005. Sea turtle nest conservation techniques on southwestern beaches in Turkey. Israel Journal of Zoology 51:13–26. Brooks, R.J., C.M. Shilton, G.P. Brown, and N.W.S. Quinn. 1992. Body size, age distribution, and reproduction in a northern population of Wood Turtles (Clemmys insculpta). Canadian Journal of Zoology 70:462–469. Buhlmann, K.A., T.D. Tuberville, and J.W. Gibbons. 2008. Turtles of the Southeast. University of Georgia Press, Athens, GA. 252 pp. Buhlmann, K.A., J.C. Congdon, and J.W. Gibbons. 2009. Ecology of Chicken Turtles (Deirochelys reticularia) in a seasonal wetland ecosystem: Exploiting resource and refuge environments. Herpetologica 65:39–53. Cagle, F.R. 1939. A system for marking turtles for future identification. Copeia 1939:170–173. 332 Northeastern Naturalist Vol. 18, No. 3 Castellano, C.M., J.L. Behler, and G.R. Ultsch. 2008. Terrestrial movements of hatchling Wood Turtles (Glyptemys insculpta) in agricultural fields in New Jersey. Chelonian Conservation and Biology 7:113–118. Christiansen, J.L., and B.J. Gallaway. 1984. Raccoon removal, nesting success, and hatchling emergence in Iowa turtles, with special reference to Kinosternon flavescens (Kinosternidae). The Southwestern Naturalist 29:343–348. Congdon, J.C., and J.W. Gibbons. 1990. Evolution of turtle life histories. Pp. 45–56, In J.W. Gibbons (Ed.). Life History and Ecology of the Slider Turtle. Smithsonian Institution Press, Washington, DC. Congdon, J.C., A.E. Dunham, and R.C. van Loben Sels. 1993. Delayed sexual maturity and demographics of Blanding’s Turtles (Emydoidea blandingii): Implication for conservation and management of long-lived organisms. Conservation Biology 7:826–833. Congdon, J.D., R.D. Nagle, O.M. Kinney, M. Osentoski, H. Avery, R.C. van Loben Sels, and D.W. Tinkle. 2000. Nesting ecology and embryo mortality: Implications for the demography of Blanding’s Turtles (Emydoidea blandingii). Chelonian Conservation and Biology 3:569–579. Engeman, R.M., R.E. Martin, H.T. Smith, J. Woolard, C.K. Crady, B. Constantin, M. Stahl, and N.P. Groninger. 2006. Impact on predation of sea turtle nests when predator control was removed midway through the nesting season. Wildlife Research 33:187–192. Ernst, C.H., and J.E. Lovich. 2009. Turtles of the United States and Canada (2nd Edition). The John Hopkins University Press, Baltimore, MD. Ewert, M.A., and C.E. Nelson. 1991. Sex determination in turtles: Diverse patterns and some possible adaptive values. Copeia 1991:50–59. Farrell, R.F., and T.E. Graham. 1991. Ecological notes on the turtle Clemmys insculpta in northwestern New Jersey. Journal of Herpetology 25:1–9. Garber, S.D., and J. Burger. 1995. A 20-yr study documenting the relationship between turtle decline and human recreation. Ecological Applications 5:1151–1162. Gibbons, J.W., D.E. Scott, T.J. Ryan, K.A. Buhlmann, T.D. Tuberville, B.S. Metts, J.L. Greene, T. Mills, Y. Leiden, S. Poppy, and C.T. Winne. 2000. The global decline of reptiles, deja vu amphibians. Bioscience 50:653–666. Harding, J.H., and T.J. Bloomer. 1979. The Wood Turtle, Clemmys insculpta: A natural history. Bulletin of the New York Herpetological Society 15:9–26. Honegger, R. 1979. Marking amphibians and reptiles for future identification. International Zoo Yearbook 19:14–22. Horne, B.D., R.J. Brauman, M.J.C. Moore, and R.A. Seigel. 2003. Reproductive and nesting ecology of the Yellow-Blotched Map Turtle (Graptemys flavimaculata): Implications for conservation and management. Copeia 2003:729–738. Hughes, G.N., W.F. Greaves, and J.D. Litzgus. 2009. Nest-site selection by Wood Turtles (Glyptemys insculpta) in a thermally limited environment. Northeastern Naturalist 16:321–338. IUCN. 2009. The IUCN Red List of Threatened Species. Available online at http://www. iucnredlist.org. Accessed 27 December 2010. Kinney, O.M., R.D. Nagle, and J.D. Congdon. 1998. Water transport by nesting Painted Turtles (Chrysemys picta marginata) in Michigan. Chelonian Conservation and Biology 3:71–76. 2011 K.A. Buhlmann and C.P. Osborn 333 Levell, J.P. 2000. Commercial exploitation of Blanding’s Turtle, Emydoidea blandingii, and the Wood Turtle, Clemmys insculpta, for the live animal trade. Chelonian Conservation and Biology 3:665–674. New Jersey Department of Environmental Protection (NJDEP). 2009. Endangered and threatened species lists. Available online at http://www.state.nj.us/dep/fgw/tandespp. htm. Accessed 28 December 2010. Niederberger, A.J., and M.E. Seidel. 1999. Ecology and status of a Wood Turtle (Clemmys insculpta) population in West Virginia. Chelonian Conservation and Biology 3:414–418. Ratnaswamy, M.J., and R.J. Warren. 1998. Removing Raccoons to protect sea turtle nests: Are there implications for ecosystem management? Wildlife Society Bulletin 26:846–850. Ross, D.A., K.N. Brewster, R.K. Anderson, N. Ratner, and C.M. Brewster. 1991. Aspects of the ecology of Wood Turtles, Clemmys insculpta, in Wisconsin. Canadian Field Naturalist 105:363–367. Rowe, J.W., K.A. Coval, and M.R. Dugan. 2005. Nest placement, nest-site fidelity, and nesting movements in Midland Painted Turtles (Chrysemys picta marginata). American Midland Naturalist 154:383–397. Saumure, R.A., and J.R. Bider. 1998. Impact of agricultural development on a population of Wood Turtles (Clemmys insculpta) in southern Québec, Canada. Chelonian Conservation and Biology 3:37–45. Saumure, R.A., T.B. Herman, and R.D. Titman. 2007. Effects of haying and agricultural practices on a declining species: The North American Wood Turtle, Glyptemys insculpta. Biological Conservation 135:581–591. Standing, K.L., T.B. Herman, and I.P. Morrison. 1999. Nesting ecology of Blanding’s Turtle (Emydoidea blandingii) in Nova Scotia, the northeastern limit of the species’ range. Canadian Journal of Zoology 77:1609–1614. Steen, D.A., M.J. Aresco, S.G. Beilke, B.W. Compton, E.P. Condon, C.K. Dodd, Jr., H. Forrester, J.W. Gibbons, J.L. Greene, G. Johnson, T.A. Langen, M.J. Oldham, D.N. Oxier, R.A. Saumure, F.W. Schueler, J. Sleeman, L.L. Smith, J.K. Tucker, and J.P. Gibbs. 2006. Relative vulnerability of female turtles to road mortality. Animal Conservation 9:269–273. Szerlag-Egger, S., and S.P. McRobert. 2007. Northern Diamondback Terrapin occurrence, movement, and nesting activity along a salt marsh access road. Chelonian Conservation and Biology 6:295–301. Tuberville, T.D., J.W. Gibbons, and H.E. Balbach. 2009. Estimating viability of Gopher Tortoise populations. Report ERDC/CERL TR-09-2 to US Army Corps of Engineers, Construction Engineering Research Laboratory, Champaign, IL. 58 pp. Turtle Conservation Fund. 2002. A global action plan for the conservation of tortoises and freshwater turtles: Strategy and funding prospectus 2002–2007. In K.A. Buhlmann, R. Hudson, and A.G.J. Rhodin (Eds.). Conservation International and Chelonian Research Foundation, Washington, DC. 30 pp. Tuttle, S.E., and D.M. Carroll. 1997. Ecology and natural history of the Wood Turtle (Clemmys insculpta) in southern New Hampshire. Linnaeus Fund Research Report, Chelonian Conservation and Biology 2:447–449. Tuttle, S.E., and D.M. Carroll. 2005. Movements and behavior of hatchling Wood Turtles. Northeastern Naturalist 12:331–348. 334 Northeastern Naturalist Vol. 18, No. 3 van Dijk, P.P., B.L. Stuart, and A.G.J. Rhodin (Eds.). 2000. Asian Turtle Trade: Proceedings of a Workshop on Conservation and Trade of Freshwater Turtles and Tortoises in Asia. Chelonian Research Monographs 2. Chelonian Research Foundation, Lunenburg, ME. 164 pp. Walde, A.D., and R.A. Saumure. 2008. Glyptemys insculpta (Wood Turtle) maximum clutch size. Herpetological Review 39:82. Walde, A.D., J.R. Bider, C. Daigle, D. Masse, J.-C. Bourgeois, J. Jutras, and R.D. Titman. 2003. Ecological aspects of a Wood Turtle, Glyptemys insculpta, population at the northern limit of its range in Québec. Canadian Field-Naturalist 117:377–388. Walde, A.D., J.R. Bider, D. Masse, R.A. Saumure, and R. Titman. 2007. Nesting ecology and hatching success of the Wood Turtle, Glyptemys insculpta, in Québec. Herpetological Conservation and Biology 2:49–60.