NANTUCKET CONSERVATION FOUNDATION, INC. PO Box 13, Nantucket, MA 02554

Home Range Size, Movement Patterns and Habitat Preference of the Spotted Turtle (Clemmys guttata) and Implications for Management at Squam Farm, Nantucket, MA January 2012 Danielle O Dell Research Technician and Field Supervisor NANTUCKET CONSERVATION FOUNDATION, INC. PO Box 13, Nantucket, MA 02554

Home Range Size, Movement Patterns and Habitat Preference of the Spotted Turtle (Clemmys guttata) and Implications for Management at Squam Farm, Nantucket, MA Executive Summary Spotted turtles are considered threatened, endangered or vulnerable throughout most of their range. Habitat destruction and fragmentation are thought to have had the most important negative impacts on populations, with additional declines attributed to pet trade collection, road mortality and vegetation succession. Additionally, mowing of agricultural fields to curb woody vegetation growth can be a high source of mortality to turtle populations as individuals move between wetlands via upland corridors or use uplands for nesting or estivation sites. Squam Farm, located in the northeastern corner of Nantucket Island, is managed as protected open space for passive recreation through the use of mowing and sheep grazing to maintain early successional habitats. A 1993 study of spotted turtles on the Squam Farm property deemed the population healthy and possibly expanding as evidenced by apparent recruitment of hatchlings and juveniles and low observed adult mortality. Additionally, it is believed that Nantucket spotted turtle populations are likely the healthiest, most robust populations in the state, likely due to the protection of extensive patches of high quality habitat. NCF is currently developing management plans for its properties, and the perpetuation of rare species, including spotted turtles, is an important conservation goal addressed in these plans. In 2005, NCF initiated this study to assess the demographics, habitat preferences, home range size and movement patterns of spotted turtles at Squam Farm. This research sought to document the specific vegetation communities used by turtles throughout the year, and document travel corridors used to move between wetlands so that proper conservation and management of wetlands and uplands could be implemented. Research by the NCF Science and Stewardship staff has documented a large and healthy population of spotted turtles at the Squam Farm property consisting of all ages and size classes of turtles. Adult spotted turtles have been found to use all types of wetlands throughout Squam Farm, as well as early successional, grassy and wooded uplands for nesting habitat, travel corridors and dry season estivation sites. Home range size of Nantucket spotted turtles is similar to other northeastern populations, however, Nantucket populations display less variability in home range size most likely due the presence of diverse and extensive patches of habitat as turtles do not have to move as far at Squam to habitat that enables them to satisfy life history requirements. However, all turtle home ranges include areas that are regularly mowed to maintain walking trails for the public as well as upland areas managed by sheep-grazing and/or mowing. The results of this research will be used to adjust the location and timing of property management activities in an effort to eliminate potential impacts to spotted turtle populations. This research will serve as a solid baseline for long-term population trend analyses for spotted turtles at Squam Farm as well as a comparison point for spotted turtle populations on other properties on Nantucket. These data will also be used to inform future research projects and aid in the development of monitoring protocols for Squam Farm and other NCF-owned properties. 2

Introduction The spotted turtle (Clemmys guttata) is a semi-aquatic, fresh-water species of eastern North America distributed from southern Ontario and Maine south along the Atlantic Coast to Florida and west to northern Indiana (Ernst and Lovich 2009). Spotted turtles generally inhabit shallow, unpolluted, mud-bottomed, freshwater wetlands including marshes, bogs, swamps, shrub swamp forests, small slow-moving streams, drainage ditches and vernal pools (Ernst 1976, Graham 1995, Milam and Melvin 2001, Ernst and Lovich 2009). They also use grassy or wooded uplands during the nesting season and upland connecting corridors when moving between wetlands (Ernst 1976, 1982, Graham 1995, Litzgus and Mousseau 2004b). On Nantucket, spotted turtles are most commonly found in red-maple (Acer rubrum) swamps, vernal pools, streams and open fields (Creighton and Graham 1993). Spotted turtles display distinct seasonal shifts in habitat use throughout their annual activity cycle (Creighton and Graham 1993, Graham 1995, Litzgus and Mousseau 2004b). They emerge from hibernacula in the early spring to bask on sunny days and begin feeding and mating when the average water temperature reaches 13.0ºC (Ernst and Lovich 2009). Females typically nest in early to mid-june and then enter a period of semi-dormancy, or estivation, in the hotter, drier months when water temperatures exceed 30.0 C (Ernst 1982, Graham 1995, Kaye et al. 2001, Ernst and Lovich 2009). They return to wetlands in the fall and generally enter back into hibernacula by late October. Spotted turtles are considered threatened, endangered or vulnerable throughout most of their range (Ernst and Lovich 2009). Habitat destruction and fragmentation are thought to have had the most important negative impact on populations, with additional declines attributed to pet trade collection, road mortality, and vegetation succession (Ernst and Lovich 2009, Graham 1995, Litzgus and Mousseau 2004a). Frequently, individual wetlands are afforded protection, but these habitat patches are often isolated by roads and development. The greater the isolation, the farther a turtle has to move in search of habitat, thereby increasing the probability of crossing roads or densely developed areas (Burke and Gibbons 1995, Patton and Crouch 2002). Additionally, mowing of agricultural fields to curb woody vegetation growth can be a high source of mortality to turtle populations as they move through upland corridors between wetlands or use uplands during the nesting season (NHESP 2009). Squam Farm, located in the northeastern corner of Nantucket Island, is owned by the Nantucket Conservation Foundation (NCF). NCF manages the property as protected open space for passive recreation through mowing and sheep grazing to maintain early successional habitats. Prior to Foundation ownership, a 1993 study of spotted turtles on the Squam Farm property deemed the population healthy and possibly expanding as evidenced by apparent recruitment of hatchlings and juveniles and low observed adult mortality (Creighton and Graham 1993). NCF is currently developing management plans for its properties, and the perpetuation of rare species, including spotted turtles, is an important conservation goal addressed in these plans. In 2005, NCF initiated a study to assess the demographics, habitat preferences, home range size and movement patterns of spotted turtles at Squam Farm. This research sought to document the specific vegetation communities used by turtles throughout the year, and document travel corridors used to move between wetlands so that proper 3

conservation and management of wetlands and upland travel corridors could be implemented. The results of this research will be used to adjust the location and timing of property management activities in an effort to eliminate potential impacts to spotted turtle populations. This research will serve as a solid baseline for long-term population trend analyses for spotted turtles at Squam Farm as well as a comparison point for spotted turtle populations on other properties on Nantucket. These data will also be used to inform future research projects and aid in the development of monitoring protocols for Squam Farm and other NCF-owned properties. The following specific research goals for this study have been identified to provide this information: 1. Estimate population size of spotted turtles at Squam Farm 2. Explore trends in demographics by sex, age and size 3. Determine average home range size and seasonal movement patterns 4. Assess habitat use related to specific vegetation communities 5. Evaluate current management activities and potential implications for spotted turtle populations on Nantucket Study Site The Nantucket Conservation Foundation s Squam Farm property (87 ha) is located in the northeastern portion of Nantucket Island, Massachusetts (Fig.1). The property is located within a glacial moraine and elevation ranges from 8-15 m above mean sea level. This property contains a mixture of upland and wetland habitats. The dominant soil type in upland area is the Plymouth-Evesboro complex (PcB), a loamy sand characterized by excessively drained soils, low available water capacity, and dominated by woody vegetation (USDA 1977). The dominant soil types in wetlands on this property are Medasaprists (ME), an organic muck soil, and the Berryland Variant (BE), a loamy sand, both of which are characterized by very poor drainage (USDA 1977) (Fig.2). The Squam Farm property consists of the following dominant vegetation communities (as delineated by The Nature Conservancy; Fig.3, Table 1): coastal shrubland (66 acres), mowed grassland (42 acres), wooded swamp (35 acres), mixed deciduous forest (33 acres), shrub swamp (24 acres), old field (4 acres), open scrub oak (4 acres), agricultural land (3 acres), sandplain heathland (3 acres), and developed land (3 acres), with the remaining 2 acres consisting of small patches of black pine/pitch pine woodland, marsh, and sandplain grassland. Upland areas at Squam Farm are characterized by either mowed and/or grazed grasslands or wooded areas dominated by tupelo (Nyssa sylvatica), sassafras (Sassafras albidum), winterberry (Ilex verticillata), sweet pepperbush (Clethra alnifolia), and greenbrier (Smilax spp.). The wetlands generally consist of red maple swamps, shrub swamps, lowland vernal pools, small ponds, marshes and bogs dominated by sphagnum moss (Sphagnum spp.), water willow (Decodon verticillatus), ferns, including cinnamon fern (Osmunda cinnamomea) and eastern marsh fern (Thelypteris palustris), and various sedge species Carex spp., and Scirpus spp. 4

Methods Population Size To obtain estimates of population size, we initiated a mark-recapture study of spotted turtles at Squam Farm (Davis and Winstead 1980). We set un-baited hoop traps and interruption traps in a variety of wetland habitats (Fig. 4). We chose trap locations based on availability of appropriate habitat and adequate water levels. In deeper water (0.5-1 m), we used hoop traps measuring approximately 0.75 m in diameter. In shallow water, we used 4.8-cm 2 mesh wire box interruption traps placed in suspected travel corridors in order to funnel turtles into the traps. In 2007, only two pools contained sufficient water for hoop traps because of drought. Both trap types were only partially submerged so allow access to air for breathing and covered with leaf litter to provide shade. We set traps once water temperatures reached 13 C (the temperature at which turtles begin feeding, moving and mating). We checked traps every two to three days between April and mid-july or until pools dried or water temperatures became too warm (>30 C). Additionally, we conducted visual searches for turtles in wetlands and captured individuals by hand. We concentrated our trap and visual search effort to the northern portion of the Squam Farm property because public use, trail maintenance and sheep grazing is greatest in those areas. We documented locations of capture (indicating trap or hand) with a handheld GPS unit (Trimble GeoXT) and recorded date, time of capture, vegetation community type (according to The Nature Conservancy designations described above), water temperature, behavior (basking, mating, feeding, submerged, travelling in upland, etc.) and the three dominant plant species at the point of capture. We used a 600g Pesola scale to weigh turtles and digital calipers to measure the straight-line carapace length and width, plastron length, and shell height (mm). We used the following characteristics to determine sex (Lazell 1976, Ernst et al. 1994, Graham 1995): Shape of plastron (males are concave, females are flat to convex) Position of cloacal opening (distal in males, beneath posterior marginal scutes in females) Tail (males are longer and more stout) Eye color (males tend to brown, females tend to orange) Throat color (males are dark brown, females are lighter yellow) Sex of juvenile turtles was not determined as it is difficult to distinguish with certainty in small size classes. A proxy for age was determined by counting the annuli on plastral scutes. Older individuals with worn plastrons were recorded as > 14 years (Litzgus and Brooks 1998). Turtles with a plastron length < 80 mm were considered juveniles (Ernst 1972, 1975). We photographed every turtle captured and used a triangular file to notch the marginal scutes with an individual identification number using a modified version of a standard turtle notching system (Cagle 1939) (Fig. 5). Due to the difficulty of permanently notching flexible marginal scutes of hatchlings and juveniles, we instead painted an identification number on the carapace with a white paint pen. Painted marks were visible for up to one year. After processing, we 5

released turtles at their capture location. We selected a subset of adult turtles to be fitted with radio transmitters, as described below. Radio Telemetry We used radio telemetry data to determine habitat use, home range size and movement distances of turtles at Squam Farm. In 2006, 2007 and 2008, we affixed up to 4 turtles per year with a 49 MHz radio transmitter (L.L. Electronics, Mahomet, Illinois, USA). We attached a 6g transmitter to the rear carapace with Oatey epoxy putty. We affixed only adult turtles with transmitters and never placed a transmitter on a turtle if it increased body weight by more than 10% (Boarman et al. 1998). The frequency of relocation efforts depended upon the seasonal level of turtle activity (Litzgus and Mousseau 2004b). From April to mid-july, when spotted turtles are most active, we tracked turtles up to three times per week. During the hotter and drier parts of the summer (generally mid-july and August) when turtles tend to estivate or move infrequently, we relocated them once or twice per week. We removed transmitters from most turtles before they entered winter hibernacula, however, we continued to track one turtle once per week during the winter of 2007-2008 to document any winter movement and the date of emergence in the spring. When relocating turtles, we tuned an FLS-40 receiver (F&L Electronics, Mahomet, Illinois, USA) to the transmitter frequency and tracked until either we obtained a visual sighting or we came within approximately 1 m of the turtle. We documented the location with a handheld GPS unit (Trimble GeoXT) and recorded the date, time, water temperature, behavior, vegetation community type, and the three dominant plant species (as described previously). Home Range and Habitat Use Home range estimates, distances traveled, and habitat preference data were calculated using radio location data collected in the active season (from April to the beginning of October when turtles entered hibernacula). Turtles that had fewer than 30 relocations were omitted from analysis because of sample sizes too low for analysis (Kenward 2001). For home range size estimates, we calculated minimum convex polygons (MCPs) which were constructed in ArcMap GIS using Hawth s Analysis Tools (Beyer 2004). The MCP method draws the smallest possible polygon around a set of points from the outermost locations with no angles exceeding 180 (Kenward 2001). All points are weighed equally which results in polygons that can sometimes encompass areas never covered by the individual. While this method may overestimate home range size (Jennrich and Turner 1969, Robertson et al. 1998, Litzgus and Mousseau 2004b), all potential habitat used by the turtle is included. Since our goal is to identify important spotted turtle habitat for inclusion in property management plans, we chose MCP as the most liberal estimate of home range. Furthermore, a study of spotted turtle home ranges in Ontario found that three different methods of home range analysis (MCP, Jennrich- Turner and Kernel analysis) yielded statistically similar results (Litzgus and Mousseau 2004b). 6

Distances traveled for each turtle were calculated by chronologically connecting location points in GIS to create paths of travel (Beyer 2004, ESRI 2004). The entire length of a line represents the minimum distance traveled for the entire survey period for that turtle. Mean distance traveled per day was calculated by summing the total distance traveled and dividing by the total number of days over which a turtle was tracked. In GIS, we plotted all locations collected for radio-tracked turtles onto the Vegetation Community Map for Nantucket (produced by The Nature Conservancy in 1998, based on 1:12,000 color, infrared stereo aerial photography taken in 1993). We used the observed plant species present at each location to correct points that fell along the border of two community types in the GIS layer in order to assign each location to the correct vegetation community. We examined the occurrence of spotted turtles in different vegetation communities during each month to determine the seasonality of habitat use, particularly wetland and upland habitats. Results Overall, we captured 112 spotted turtles over the course of the study from 2005-2008, of which 68 were females, 29 were males and 15 were either juveniles or of undetermined sex (Table 2). The observed sex ratio for spotted turtles (discounting turtles of undetermined sex) was 0.43, males to females, or approximately 30% males, 70% females. The mean age for all captured turtles was 10.7 years (±0.34 years SE). While there were few captures of juveniles in general, the presence of individuals of a very young age influenced the calculated mean age of all turtles captured. When juveniles were excluded from analysis, the mean age of adult turtles was 11.7 years (±0.27 years SE) with no difference in mean age between males and females (11.8 and 11.3 years respectively), (p=0.43) from a two-tailed t-test, t88=0.8 with a 95% CI from -0.84 1.95 years. We found hoop traps to be the most effective method for capturing spotted turtles during the early part of the season when water was available. During the hotter and drier part of the season, we found most turtles by conducting visual searches and hand capturing turtles incidentally found while radio-tracking turtles with transmitters. Interruption traps were the least effective method of capturing spotted turtles. In total, we captured 56 turtles with hoop traps, 12 with interruption traps, and 44 by hand capture. When data from interruption traps and hoop traps was combined over the time that they were employed, overall trap success was very low and there was no difference in the rates of trap capture between males and females. Adult male spotted turtles tend to be larger than adult females in overall carapace length and width, while females tend to be larger in shell height, and slightly larger in plastron length and weight (Table 3). On average, males have a carapace length 5.7 mm longer (P = 0.003 from a two-sided t-test) and a width 2.1 mm greater than females (P = 0.04) while female shell height is 2.4 mm greater than males (P = 0.02). While statistically significant, the difference between female to male plastron length (0.4 mm) and weight (0.5 g) is minimal (P = 0.01 and 0.03, respectively). We also calculated mean carapace length, width, shell height and weight for the 18 juveniles turtles captured (Table 3). In cases where sex was unable to be determined but we were able to classify the individual as an adult based on carapace length >80mm, we included 7

measurements of those individuals in the mean measurements for overall adult turtles (Table 3). We radio-tracked three turtles in 2006 and four turtles in 2007 and 2008. Whenever possible, we placed transmitters on different turtles over all the years tracked with two exceptions; turtle number 4 was tracked in both 2006 and 2008 and turtle number 102 was tracked in 2006 and 2007. Turtles were tracked over a mean of 153.1 days per year with a mean of 35.6 relocations over that period. The mean home range size, as calculated by the minimum convex polygon method (MCP), was 1.7 ha (with a range from 1.5 2.0 ha) (Table 4). There was no difference in mean home range size between males and females (p = 0.49), although the sample size for males was low, with only 2 males tracked over the course of the study. Nearly all turtle home ranges exhibited some overlap in both space and time, and all home ranges incorporate portions of the Squam Farm property that are designated as mowed walking trails and/or managed upland grasslands (Figure 6). Radio-relocated and captured spotted turtles were found in eight different vegetation communities at Squam Farm (Figure 7). Throughout the entire study, turtles were most commonly found in wetland areas such as temporary wetlands, wooded and shrub swamps and vernal pools. Winter and early spring (November through April) captures and relocations were located in wetlands 100% of the time as turtles hibernated under water and emerged in the spring. Movement between wetlands began in May when water temperatures rose above 13 C and up to 25% of captures and relocations occurred in mowed grasslands, upland forests or along mowed walking trails within the property. Turtles were also incidentally captured in May and June on mowed walking trails while presumably traveling between vernal pools and shrub swamps. While we did not document the presence of turtle nests, we did relocate females in grassy uplands in June where turtles are known to nest. Spotted turtles continued to use both wetlands and uplands throughout the season, with the highest use of uplands occurring in August when up to 55% of relocations occurred in this habitat type (Table 5). Over the course of a season, turtles moved a mean total distance of 1185 m, with a mean daily distance moved of 8.0 m averaged over the entire time period. As discussed above however, the majority of these movements occurred in the early spring between May and June and again in the fall as turtles moved back to the wetlands used for winter hibernation. Very little movement was documented during the month of August when most turtles entered a brief estivation period. Other species incidentally caught in traps and during visual searches were painted turtles (Chrysemys picta) and common snapping turtles (Chelydra serpentina). We noted the capture of both species and released individuals at the location of capture. We also collected morphologic data on painted turtles and notched the marginal scutes in the same manner as spotted turtles during the 2006 and 2007 field seasons. Discussion We captured 112 spotted turtles from 2005-2008. In 2008, we captured 19 new, unmarked turtles suggesting that the actual population size could be quite a bit larger. Continuing to find 8

so many unmarked adult turtles, in addition to the presence of hatchling and juvenile turtles after two years of intensive tracking, trapping and searching lends further evidence to the existence of a large, healthy, reproducing population at Squam Farm. Additionally, during the course of this study we recaptured turtles originally captured and marked at Squam Farm by Creighton and Graham (in 1993) lending further evidence to the robustness of this population. Long-lived organisms tend to be slow to reach sexual maturity and generally have relatively low recruitment of juveniles into the population; therefore, it is critical have a high rate of adult survivorship to maintain stable populations (Congdon et al. 1993). Spotted turtles do not reach sexual maturity until 7-10 years of age and lay approximately 3-5 eggs at a time (Ernst 1972). Any threats to adult turtles can have severe impacts on populations and it is therefore of utmost importance to the Nantucket Conservation Foundation to minimize any negative impacts to spotted turtle populations at Squam Farm. The Squam Farm population sizes are comparable to other large and intact populations documented elsewhere in the United States and Canada (Ernst 1976; Litzgus 1996; Litzgus and Mousseau 2004 a and b), however, the Squam Farm population appears to be larger than populations found documented in Connecticut and Massachusetts (Graham 1995; personal communication M. Jones, MassWildlife). Patches of habitat are likely smaller and more isolated in other more densely developed parts of southern New England as compared to Nantucket. Favorable, larger patches of habitat may lead to larger observed population sizes. Freshwater turtle species commonly display nearly 1:1 sex ratios (Bury 1979) and several other studies have confirmed that spotted turtles conform to this trend (Mauger 1990; Graham 1995; Litzgus 1996). A few studies, however, have documented slightly female-skewed sex ratios in spotted turtle populations (Ernst 1976, 1982; Haxton and Berrill 1999) where the ratios ranged from 55-57% female-dominated. The ratios observed at Squam Farm however, are skewed more than those found elsewhere with approximately 30% males and 70% females. This observed ratio may be due to our trapping and search effort throughout the year. Males are most active in the early spring when they are moving between vernal pools in search of females for breeding. Most males were captured in hoop traps or by hand capture during the months of April and May. Later in the season at drier times of year, traps were no longer employed and males were encountered less frequently due to their relative inactivity whereas we continued to find active females late into the fall season. The mean home range size observed for Squam Farm spotted turtles was 1.7 ha and varied from 1.5 2.0 ha for both male and female turtles averaged over the entire study period. All turtles at Squam Farm displayed similar home range sizes and are comparable to mean home range sizes reported in the literature for spotted turtles at other locations in North America (Table 4). While the mean home range size of Squam Farm turtles is similar to that reported for other populations, the variability in home ranges displayed in other populations was not seen at Squam Farm. Ims et al. (1993) demonstrated that larger animal home range size reflects highly fragmented habitat whereas animals with smaller home range sizes were generally found in areas with more contiguous and diverse habitat. Areas that contain wetlands or other important habitat features for spotted turtles that are separated by roads, development or vast stretches of unsuitable habitat require turtles to move farther to reach breeding or feeding grounds. Conversely, a contiguous and diverse system of wetlands and uplands 9

available for turtles to use should concentrate the population within a more restricted area and require less movement. Squam Farm has extensive wetlands systems interspersed with grassy uplands which are used for nesting, albeit for short periods of time. The reduction in home range size variability seen at Squam Farm compared to other studies such as Ims et al. (1993) is likely due to the absence of roads and infrastructure and connectivity to wetlands within the Squam Farm property as well as extensive networks of wetlands on bordering conservation and private properties. While spotted turtles at Squam Farm were found or relocated in wetland areas 75% of the time in all months except August, uplands are still a critical component of spotted turtle habitat. It is clear that the turtles traveled through upland pastures and grasslands and over mowed walking trails to reach these wetlands (Figure 6). Uplands are important to nesting females during the months of June and July as evidenced by up to 25% of captures and relocations occurring in uplands. Clearly, uplands are critical to both males and females in August where up to 55% of captures and relocations occurred. August tends to be hot and dry and turtles enter a period of inactivity (estivation) at this time. They move slowly and infrequently during estivation which puts them at greater risk. These trails and uplands are regularly maintained by the Nantucket Conservation Foundation by mowing and/or sheep grazing. While there is little evidence that grazing livestock will physically harm spotted turtles (Tesauro 2001; Tesauro and Ehrenfeld 2007), mortality due to mowing has been well documented elsewhere and identified as a threat to turtle populations (Saumure and Bider 1998; Saumure et al. 2006; MNHESP 2009, Erb and Jones 2011). Management Implications The Nantucket Conservation Foundation recognizes that it is imperative to our organizational mission to maintain a balance between passive recreational use of our properties by the public and to protect sensitive plant and animal species found on its properties. Past management activities at Squam Farm focused on mowing upland fields to prevent overgrowth by woody vegetation and to maintain walking paths by mowing trail edges. NCF seeks alternative management that can achieve these goals while also decreasing or eliminating unintended impacts to spotted turtles and other reptiles and amphibians on the property. Currently, NCF maintains a flock of sheep at Squam Farm. Rotational and strategic grazing by the sheep has been implemented as a means of maintaining early successional and open grasslands at Squam Farm. During the initiation of this grazing project, consideration was given to the potential for detrimental interactions between spotted turtles and sheep during turtle nesting season as females often nest in sandy, grassy areas. On several occasions, we did incidentally observe turtles using upland areas where sheep were either slated to graze, were currently grazing or had recently grazed. These observations were made throughout the season from May through September when turtles were using uplands as nesting and estivation sites or while moving between wetlands. We did not document a single incidence of mortality or injury directly related to the presence of sheep; however we recognize that this study was not designed to document the impacts that sheep grazing may have on turtle populations. Research on the interactions of bog turtles (Glyptemys muhlenbergii) with goats and cows in New York and New Jersey has shown no negative effects on the turtles and grazing by both served to 10

increase available habitat and decrease cover of invasive species such as purple loosestrife (Lythra salicaria) and common reed (Phragmites australis) (Tesauro 2001; Tesauro and Ehrenfeld 2007). Trail edge maintenance by mowing during the growing season is an important means of reducing exposure to ticks and improving visibility by the visiting public. Unfortunately, public use of this property is highest at times when spotted turtles are most active. It is clear that mowing has negative impacts on turtle populations but is a necessary component to trail maintenance at Squam Farm. In order to minimize unintended harm to spotted turtles and other reptiles on this property while maximizing the safety and enjoyment by the public, we highly recommend minimizing or avoiding mowing activities from mid-april through early fall, based on the results of this study. We have documented on an annual basis the presence of spotted turtles along the mowed walking trails and grassy uplands during the months of May through September (Figure 6). NCF maintenance staff have found turtles in fields while mowing and we have also documented deformities and injuries to the carapace of several individuals that are very likely caused by interactions with mowers or other vehicles. If mowing absolutely cannot be avoided as a means of trail maintenance during these months, we recommend it be conducted at very low speeds, during the hottest part of the day and with the use of a walker/spotter in front of mowers in order to remove turtles and other fauna from the trails. We recommend that whenever possible, sheep grazing be utilized as a preferred means of maintaining early successional fields and reducing woody vegetation cover at Squam Farm during the months of April through October. Erb and Jones (2011) have documented near 100% mortality rates to wood turtles and eastern box turtles elsewhere in Massachusetts due to tractor tires alone. Various blade types and height settings add additional risks to turtles. Erb and Jones recommend that in cases where mowing in early successional fields during periods of high turtle activity absolutely cannot be avoided, it is best conducted during the hottest part of the day, beginning mowing in the center of the field and rotating out towards the edges, using very low speed, maintaining a strip of unmowed vegetation at the edge of the field and maintaining blade heights >15 cm. A large and healthy population of spotted turtles and an abundance of interconnected, high quality habitat is yet another attribute that sets Nantucket and the properties owned and maintained by the Nantucket Conservation Foundation apart. Conservation biologists and managers from other parts of the state of Massachusetts have expressed surprise and envy at the size of Nantucket spotted turtle populations and the extent of quality habitat. Given that spotted turtles are long-lived and slow to reproduce, they are very susceptible to disturbance and the loss of even a few adult individuals can be damaging to the population. Utmost consideration should be given to maintaining this population in its current state by altering management regimes to avoid consequences to spotted turtles. 11

Literature Cited Beyer, H. L. 2004. Hawth's Analysis Tools for ArcGIS. http://www.spatialecology.com/htools. Boarman, W. I., T. Goodlett, G. Goodlett, and P. Hamilton. Review of radio transmitter attachment techniques for turtle research and recommendations for improvement. Herpetological Review 29:26-33. Burke, V. J., and J. W. Gibbons. 1995. Terrestrial buffer zones and wetland conservation: a case study of freshwater turtles in a Carolina bay. Conservation Biology 9:1365-1369. Bury, R. B. 1979. Population ecology of freshwater turtles. Pp. 571-602, In M. Harless and H. Morlock (Eds.). Turtles: Perspectives and Research. John Wiley and Sons, New York, NY. Cagle, F. R. 1939. A system of marking turtles for future identification. Copeia 1939:170-173. Congdon, J. D., A. E. Dunham, and R. C. Van Loben Sels. 1993. Delayed sexual maturity and demographics of Blanding s turtles (Emydoidea blandingii): Implications for conservation and management of long-lived organisms. Conservation Biology 7:826-833. Creighton, C. A., and T. E. Graham. 1993. Population status and habitat use patterns of the spotted turtle in Squam Swamp, Nantucket Island, Massachusetts. Summary of research prepared for the Nantucket Land Council. 14 pp. Davis, D. E. and R. L. Winstead. 1980. Estimating the numbers of wildlife populations. pp. 221-245. In S.D. Schemnitz (Ed.). Wildlife Management Techniques Manual. The Wildlife Society. Washington, DC. Erb, L., and M. T. Jones. 2011. Can turtle mortality be reduced in managed fields? Northeastern Naturalist 18:489-496. Ernst, C. H. 1970. Home range of the spotted turtle, Clemmys guttata. Copeia 1970:391-393. Ernst, C. H. 1975. Growth of the spotted turtle, Clemmys guttata. Journal of Herpetology 9:313-318. Ernst, C. H. 1976. Ecology of the spotted turtle, Clemmys guttata (Reptilia, Testudines, Testudinidae), in Southeastern Pennsylvania. Journal of Herpetology 10:25-33. Ernst, C. H. 1982. Environmental temperatures and activities in wild spotted turtles, Clemmys guttata. Journal of Herpetology 16:112-120. Ernst, C. H., and J. E. Lovich. 2009. Turtles of the United States and Canada. pp. 212-221. The Johns Hopkins University Press, Baltimore, MD. 12

ESRI 2004. Environmental Systems Research Institute. ArcView 9.0 software. Redlands, CA. Graham, T. E. 1995. Habitat use and population parameters of the spotted turtle, Clemmys guttata, a species of special concern in Massachusetts. Chelonian Conservation and Biology 1:207-214. Haxton, T., and M. Berrill. 1999. Habitat selectivity of Clemmys guttata in central Ontario. Canadian Journal of Zoology 77:593-599. Ims, R. A., J. Rolstad, and P. Wegge. 1993. Predicting space use responses to habitat fragmentation: can voles Microtus oeconomus serve as an experimental model system (EMS) for Capercaillie Grouse Tetrao urogallus in boreal forest? Biological Conservation 63:261-268. Jennrich, R. I., and F. B. Turner. 1969. Measurement of non-circular home range. Journal of Theoretical Biology 22:227-237. Kaye, D. R. J., K. M. Walsh, and C. M. Ross. 2001. Seasonal movements and habitat preferences for the spotted turtle and eastern box turtle in Massachusetts. Report from ICOET 2001 Proceedings. pp. 66-82. Kenward, R. E. 2001. A Manual for Wildlife Radio Tagging. Academic Press, New York. 311 pp. Lazell, J. D., Jr. 1976. This Broken Archipelago. Quadrangle/The New York Times Book Co., 260 pp. Litzgus, J. D., and R. J. Brooks. 1998. Testing the validity of counts of plastral scute rings in spotted turtles, Clemmys guttata. Copeia 1998:222-225. Litzgus, J. D. 1996. Life history and demography of a northern population of spotted turtles, Clemmys guttata. Unpublished Masters Thesis, University of Guelph, Guelph, ON, Canada. Litzgus, J. D., and T. A. Mousseau. 2004. Demography of a southern population of the spotted turtle (Clemmys guttata). Southeastern Naturalist 3:391-400. Litzgus, J. D., and T. A. Mousseau. 2004. Home range and seasonal activity of southern spotted turtles (Clemmys guttata): Implications for management. Copeia 2004:804-817. Massachusetts Natural Heritage and Endangered Species Program. 2009. Mowing advisory guidelines in turtle habitat: Pastures, successional fields, and hayfields. http://www.mass.gov/dfwele/dfw/nhesp/conservation/pdf/mowing_guidelines.pdf 13

Mauger, D. 1990. A resurvey of the spotted turtle (Clemmys guttata) population at Lockport Prairie Nature Preserve, Will County, Illinois. Unpublished report. 12 pp. Paton, P. W., and W. B. Crouch III. 2002. Using the phenology of pond-breeding amphibians to develop conservation strategies. Conservation Biology 16:194-204. Robertson, P. A., N. J. Aebischer, R. E. Kenward, I. K. Hanski, N. P. Williams. 1998. Simulation and jack-knifing assessment of home range indices based on underlying trajectories. Journal of Applied Ecology 35:928-940. Tesauro, J. 2001. Restoring wetland habitats with cows and other livestock. Conservation Biology in Practice 2:26-30. Tesauro, J., and D. Ehrenfeld. 2007. The effects of livestock grazing on the bog turtle {Glyptemys (=Clemmys) muhlenbergii]. Herpetologica 63:293-300. The Nature Conservancy. 1998. Vegetation Community Map of Nantucket, GIS Data Layer. Treanor, S. A. and E. Molden. 2006. A study of spotted turtle (Clemmys guttata) demographics in two local wetland habitats on Nantucket Island, Massachusetts. Unpublished report. The Nantucket Conservation Foundation, Inc. Wilson, T. P. 1994. Ecology of the spotted turtle, Clemmys guttata, at the western range limit. Unpublished Master s Thesis. Eastern Illinois University, Charleston. United States Department of Agriculture. 1977. Soil survey of Nantucket County, MA. 14

Figure 1: Aerial photograph indicating the location of the Nantucket Conservation Foundation s Squam Farm Property, Nantucket, MA.

Figure 2: Soil types surveyed at Squam Farm, Nantucket, MA and delineated in the Soil Survey of Nantucket County (USDA 1977). 16

Figure 3: Vegetation communities at Squam Farm, Nantucket, MA, delineated by the Nature Conservancy based on 1993 aerial photography (TNC 1998). 17

Figure 4: Aerial photograph of the Squam Farm study site with wetlands and trap locations, 2005-2008, Nantucket, MA. NOTE: Figure 4 has been removed from the website version of this report to protect the locations of rare and sensitive species. To request more detailed information than what is contained here, please email the author: dodell@nantucketconservation.org. 18

Figure 5: Diagram of the modification of the Cagle notching system (Cagle 1939) used to mark and identify spotted turtles at Squam Farm, Nantucket, MA. 19

Figure 6: Minimum convex polygons (MCP) for all spotted turtles tracked at Squam Farm from 2006-2008 and mowed walking trails and upland grasslands managed by mowing and sheep-grazing, Nantucket, MA. Figure 7: Wetland boundaries and all spotted turtle capture and relocation points from 2006-2008, Squam Farm, Nantucket, MA. NOTE: Figures 6 and 7 have been removed from the website version of this report to protect the locations of rare and sensitive species. To request more detailed information than what is contained here, please email the author: dodell@nantucketconservation.org. 20

Table 1: Definitions of The Nature Conservancy vegetation community types found on Squam Farm, Nantucket, MA (TNC 1998). Squam Farm Vegetation Communities Total Vegetation Community Type Acres Agricultural Land 3.0 Black Pine/Pitch Pine Woodland 0.4 Coastal Shrubland 66.2 Deep Marsh 1.3 Developed Land 2.5 Marsh 0.4 Mixed Deciduous Forest 32.9 Mowed Grassland 42.2 Old Field 3.6 Open Scrub Oak 3.6 Potential Vernal Pool 0.5 Red Cedar Woodland 0.0 Sandplain Grassland 0.1 Sandplain Heathland 2.9 Shrub Swamp 23.5 Wooded Swamp 35.2 TOTAL 218.6 Vegetation communities at the site were categorized using a GIS map distinguishing 36 distinct communities on Nantucket that was produced by The Nature Conservancy in 1998, based on 1:12,000 color, infrared stereo aerial photography taken in 1993. The vegetation community types listed above are defined as follows: Agricultural Land: includes tilled fields, nurseries, orchards and pastures. Black Pine/Pitch Pine Woodland: stands of black pine (Pinus thunbergii) and/or pitch pine (Pinus rigida) that cannot be distinguished via aerial photo. Coastal Shrubland: tall (>0.5m), usually dense, shrubland on upland. Common species include Viburnum dentatum, Gaylussacia baccata, Prunus maritima, Amelanchier spp., Morella, Rosa spp., Vaccinium spp., and Toxicodendron radicans. Deep Marsh: deep freshwater wetland community dominated by herbaceous vegetation, but may have some shrubs, including Decodon verticillatus. Developed Land: includes residential, commercial, industrial and recreational land uses. Also includes large mowed lawns often found in residential and developed areas, sand and gravel pits, and recently bulldozed or excavated areas that appear to be undergoing development. Marsh: herbaceous freshwater wetland. Mixed Deciduous Forest: forest dominated by Quercus spp., Sassafras albidum, Nyssa

sylvatica, Acer rubrum or other deciduous trees. Mowed Grassland: grasslands that are predominately Schizachyrium scoparium, and show evidence of mowing. May contain scattered trees or shrubs (<30%). Usually located near developed areas. Old Field: Juniperus virginiana as shrub or tree overstory (30% or less), usually with a grass and forb understory, sometimes with shrubs. May also contain Pinus rigida or thunbergii. In its earlier stages, this type is an overgrown field, without cedar or other trees. Open Scrub Oak: thickets of Quercus ilicifolia approximately 30-75% total cover, of which at least 50% is scrub oak, with Arctostaphylos uva-ursi, Gaylussacia baccata or herbaceous understory. Potential Vernal Pool: basins under ¼ acre in size that appear to have the hydrologic and vegetative characteristics common to vernal pools. Red Cedar Woodland: 30% or greater overstory of Juniperus virginiana, with a variety of understories including Gaylussacia baccata, Arctostaphylos uvi-ursi or herbaceous vegetation. Sandplain Grassland: grassland dominated by Schizachyrium scoparium, Carex pensylvanica, and other grasses, with <50% Arctostaphylos uva-ursi, Gaylussacia baccata, and other heathland shrubs. Sandplain Heathland: dwarf shrubland dominated by Arctostaphylos uva-ursi, Hudsonia ericoides or Gaylussacia baccata, with lesser amounts of Morella pennsylvanica, Viburnum dentatum and Quercus ilicifolia. Often contains a matrix of grassland. Shrub Swamp: variable mix of Rhododendrum viscosum, Vaccinium spp., Clethra alnifolia and Ilex verticillata, usually in a basin. Wooded Swamp: deciduous forested wetland, usually in a basin, and dominated by Acer rubrum or Nyssa sylvatica. 22

Table 2: Summary data of total number of spotted turtles found at Squam Farm, Nantucket, MA from 2005-2008. Turtles of unknown sex include both juveniles and adults based on carapace length of <80 mm (Ernst 1972, 1975). 2005 2006 2007 2008 Overall Males 3 5 12 9 29 Females 1 17 17 33 68 Unknown 1 5 7 2 15 Total 5 27 36 44 112 Table 3: Mean (±1 SE) size of male, female and juvenile spotted turtles at Squam Farm, Nantucket, MA, summed overall years. Overall adult size includes adult turtles of undetermined sex. CL CW SH PL WT Male 108.1 (2.3) 81.2 (1.5) 37.0 (0.8) 93.7 (2.0) 184.4 (11.0) Female 101.4 (1.9) 78.3 (1.2) 38.8 (0.8) 92.7 (1.7) 181.5 (8.7) Juvenile 64.2 (2.9) 54.6 (2.2) 25.3 (1.4) 57.5 (2.9) 52.6 (9.7) Overall Adult 105.5 (1.2) 87.2 (6.7) 38.9 (0.5) 95.0 (1.1) 187.6 (6.4) *CL = carapace length, CW = carapace width, SH = shell height, PL = plastron length, all in millimeters, WT = weight (g). Table 4: Estimated home range sizes for various studied populations of spotted turtles. All home range estimates were calculated using the Minimum Convex Polygon Method except for data from Cedar Swamp, MA (Graham 1995). The Nantucket, MA home range size reported is for both males and females *. This table is reproduced from Litzgus and Mousseau (2004b). Mean Home range size (ha) Location Females Males Source Nantucket, MA 1.70* current study Beidler Forest, SC 19.06 5.15 Litzgus and Mousseau 2004b Lancaster, PA 0.53 0.53 Ernst 1970 Lockport, IL 1.75 0.72 Wilson 1994 Romeoville, IL 0.58 0.99 McGee et al. unpubl. Cedar Swamp, MA 0.56 0.84 Graham 1995 Victoria Co., ON 4.70 2.00 Haxton and Berrill 1999 Georgian Bay, ON 3.22 3.58 Litzgus 1996 *Nantucket home range was combined for males and females due to low sample size of males. 23

Table 5: Percentage of spotted turtle captures and relocations found in wetlands and uplands by month from 2006-2008 at Squam Farm, Nantucket, MA. April May June July August September October November December 2006 Wetlands 100% 74% 76% 93% 73% 85% 100% ---------- ---------- Uplands 0% 26% 24% 7% 27% 15% 0% ---------- ---------- 2007 Wetlands 100% 78% 75% 73% 45% 80% 91% 100% 100% Uplands 0% 22% 25% 27% 55% 20% 9% 0% 0% 2008 Wetlands ---------- 75% 88% 75% 57% 56% 100% ---------- ---------- Uplands ---------- 25% 12% 25% 43% 44% 0% ---------- ---------- 24