Third Box Turtle Conservation Workshop

Jug Bay Wetlands Sanctuary, Maryland
November 9 - 10, 2007


Health and disease in the conservation of the Eastern Box Turtle, Terrapene carolina carolina

Matt Allender, Department of Small Animal Clinical Sciences, College of Veterinary Medicine, University of Tennessee, Knoxville, TN 37996.

Rehabilitation centers and veterinary schools that frequently encounter box turtles are potential sites for sampling and monitoring health. Routine screening for overall health function using blood, or sampling of dead individuals are key to identifying diseases or toxins that are emerging or persisting in a population. Numerous infectious and non-infectious agents have been implicated in mortalities affecting free-ranging turtles and tortoises. One infectious agent that has been increasingly reported to affect chelonians is Ranavirus, a member of the iridovirus family. Worldwide Ranavirus infections are emerging among wild and captive chelonian populations. Chelonian species diagnosed with Ranavirus infections include the Hermann’s tortoise (Testudo hermanii), Russian tortoise (Testudo horsfieldi), eastern box turtle (Terrapene carolina), soft-shelled turtle (Trionyx sinensis), and gopher tortoise (Gopherus polyphemus). Clinical manifestations of iridoviral infections in reptiles are not always present, but may include lethargy, dyspnea, ocular, nasal and oral discharges, and death. Oral lesions are usually slightly raised, white plaques that are easily seen during an oral examination. Other signs may include subcutaneous edema, hepatitis, necrotizing splenitis, conjunctivitis, and pneumonia. The duration of disease is short, and many free-ranging animals likely die prior to their presentation at wildlife rehabilitation centers or clinics. The mode of transmission of reptile iridoviral infections is unknown; however, the occurrence of viral particles in circulating blood cells suggests the possibility that the virus may be transmitted via blood-feeding parasites. Diagnosis of iridovirus is currently validated by using a polymerase chain reaction. Samples should be taken from tissues with a high viral load such as the spleen, liver, oral mucosa, and blood. Treatment of iridovirus has been successfully attempted in some cases. Successful conservation programs of chelonians require integration of numerous types of ecological and biological data, including those addressing habitat quality, nutrition, disease, reproduction, and survival.  Research programs focused on free-ranging animals should be more often complemented and triggered by clinical monitoring of animals that are presented for veterinary care at wildlife health centers, rehabilitation centers, and zoos. Cooperation among biologists, veterinarians, local, state, and national governments, and the public will be integral to the management of iridovirus infections in chelonians.Long term movement histories for headstarted juvenile, and translocated adult and juvenile Eastern Box Turtles in NW Pennsylvania sanctuaries

Long term movement histories for headstarted juvenile, and translocated adult and juvenile Eastern Box Turtles in NW Pennsylvania sanctuaries

William Belzer, Box Turtle Conservation, Oil City, Pennsylvania 16301.
Susan Seibert, AA Forestry & Wildlife Services, Inc. Utica, Pennsylvania 16326.

We present hand-drawn overhead transparency, and ArcView-plotted GPS, maps of locations recorded since 1993 for translocated adult, and 2-yr-old headstarted, eastern box turtles in our repatriation studies at two (200 vs 500 acre) sanctuaries in NW Pennsylvania.  Archived locations recorded at 1-10 d intervals for each turtle, across consecutive yrs, provide insight into movements overlooked by occasional glimpses at habitat use.

Our 20 plus representative maps (from >100 telemetered turtles) illustrate that: translocated adult box turtles, and headstarted juveniles, often move well beyond small (200-500 acres) sanctuary boundaries; most translocated adults don't develop site fidelity within sanctuary boundaries despite repeated food and mate offerings at a central site; repeatedly retrieving dispersing turtles, to return them to a sanctuary's core, probably helps little, if at all, to promote adoption of habitat inside the sanctuary; short term (e.g. 2 yrs) confinement of translocated box turtles does not guarantee their site fidelity for the locale after release; translocated adults, and headstarted juveniles, rarely disperse in directions heading back toward their origins; our first years of field data from 2-yr-old juveniles, headstarted by the Michell protocol, suggest that the male juveniles are more likely to leave a sanctuary; each passing year sees more, who had initially demonstrated apparent site fidelity, begin outbound migrations, leaving a currently unclear, fluid assessment of this strategy's promise; individuals differ greatly in movement pattern and range, and propensity to disperse ~ behavioral generalizations are hard to come by for this species; habitat locales favored by some individuals are ignored by others; 2000 acres of unfragmented habitat is probably a conservative minimum for a sanctuary to be considered as a potential box turtle repatriation site; tens of thousands of turtles could be required to even hope to rebuild density sufficient for a population to become self-sustaining; and, a century or more of sustained effort may be needed to simply discover whether it's possible to ever rebuild a self-sustaining population

Our early findings raise questions that won't find answers for many decades.  For example:  is "transience" (occasionally reported in the behavioral repertoire of wild adults) a genetic or life stage phenomenon? Is there a sexual predilection for becoming a transient?  Do transients ever reverse course and eventually revisit populations or habitat they passed through years previously? If rebuilding a self sustaining population ever proves to be possible (not at all clear from our work to-date) could such efforts that use headstarted juveniles succeed in habitats somewhat smaller than what might be required if translocated adults were used?

It's almost certain that high resident densities are needed for population sustainability (not to be confused with the sighting of {relic} individuals in ancestral habitats for many decades after their population actually fell below its capacity for self-sustainability).
Aggressive proactive protection of any remaining high-density populations is critical and should be a high priority for management agencies.  Trying to rebuild such a population may prove to be virtually impossible.

Findings along the way: What we learned (and have yet to learn) after 18 years of radio tracking Box Turtles in a suburban neighborhood

I. Lehr Brisbin, Jr. and Robert A. Kennamer, Savannah River Ecology Laboratory, Aiken, South Carolina, 29802
Eric L. Peters, Chicago State University, Chicago, Illinois 60425.

We report data collected by the radio tracking of adult eastern box turtles (Terrapene carolina carolina) in an urban forest/suburban neighborhood ecotone in Aiken, South Carolina. These data span a period of more than 18 years and are still ongoing, possibly the longest continuous radio tracking data ever collected for any wild animal. This study has produced information on turtle survival characteristics in these areas and interactions of the turtles with humans and conspecifics. Information gathered during the first 15.5 years of the study, representing over 125 individual turtle-years, has been written-up in a manuscript which is currently in press and the results are summarized here: constant annual adult survival probability estimated for radiotelemetered turtles was 0.932 + 0.021 (SE), with females experiencing lower survival than males. There was a suggestion that greater time spent in suburban neighborhood habitats tended to reduce survival. An independent estimate of constant adult survival, derived from the opportunistic marking and recapture of  86 non-telemetered turtles in  the same study area was 0.954 + 0.036 (SE). Even though the population was male-dominated (male:female = 2.1:1), the encounter rate was higher for females than that for males – likely as a result of females moving into the suburban neighborhood for egg-laying in open lawns and flower beds. Thus, our data suggests that overall, adult box turtles can survive in suburban neighborhoods at comparable rates to populations in natural habitats. However these neighborhoods may be acting as “ecological traps” selectively reducing the survival of females coming into these areas to nest. Interestingly enough, deaths of telemetered turtles were not dominated by collisions with motor vehicles, which only caused two of the ten deaths observed. There was a suggestion from direct behavioral observations and inferred movement patterns, that some “experienced” turtles had developed strategies for minimizing collisions with vehicles when crossing neighborhood streets. Other causes of human-induced mortality included being run-over by powered lawn mowers, being burned-up in yard litter and drowning in backyard fish ponds. Long-term movement studies showed that (1) eleven years of continuous radio tracking is not long enough to be able to adequately assess or predict the lifelong habitat needs and movement patterns of an individual box turtle, and (2) throughout their lives, many of these turtles showed never before-reported (for this species) annual movements to and from specific winter dormancy areas which may or may not be located within their active season ranges. An uncanny homing ability was often shown, returning to specific dormancy areas which were used year after year. Dates of dormancy entrance and emergence have now been recorded for over a decade or more for individual turtles, together with concomitant temperature data collected from both inside and outside of the dormancy sites chosen. A proposal to establish a network of investigators collecting similar data across extended periods of time for turtles from northern to southern parts of the species’ range, could address interesting and important aspects of the box turtle’s reaction/adaptation to global warming and climate change.


Reducing Box Turtle mortality during landscape rehabilitation at Gettysburg National Military Park

Carolyn Davis, National Park Service, Gettysburg, Pennsylvania 17325.

Gettysburg National Military Park, located in south central Pennsylvania, is actively managing its landscape to recreate its 1863 Civil War appearance.  Part of this management includes the conversion of a 10 acre oak hickory forest into 6 acres of shrub thicket and the remaining acreage into native grassland. Eastern box turtles (Terrapene carolina) are present within the forested project area and the methods used to remove the trees (clear cutting and dragging out trees on skid trails) can result in box turtle mortality. In an effort to reduce mortality, box turtles are being temporarily relocated until the cutting is completed. Throughout late summer and fall of 2006 over 50 box turtles were removed from the project area and placed in a temporary turtle holding pen measuring ¼ acre in size located in a woodlot adjacent to the cutting.  However, due to an unseasonably warm and wet winter the cutting of the project area was delayed and the turtles were released the following spring. Of the 50 turtles released from the turtle holding pen many were later recaptured and had traveled up to 400 meters back to their original locations, demonstrating the turtles’ homing ability. In late-summer of 2007 the cutting of the project area finally began. Once again, an effort is being made to remove turtles and place them in the turtle holding pen.  Cutting in this area is expected to be completed by late August or early September 2007. At that time turtles will be released from the pen. Long term mark recapture monitoring will continue in the area to determine the success of the initiative.


Population structure and behavioral ecology of Eastern Box Turtles (Terrapene c.carolina) in a fragmented habitat in South Carolina

Mary Lang Edwards and David B. Ritland, Department of Biology, Erskine College, Due West, South Carolina 29639.

A population of eastern box turtles (Terrapene carolina carolina) inhabiting an 8 ha woodland "fragment" in the piedmont of South Carolina (N34.5o, W82 o, 260m elev.) was monitored for 9 yrs (1993-2001). An estimated 110 turtles utilize the site (density=14 turtles/ha), consisting of 57% adults (~250 g), 14% subadults (131-249 g), and 29% juveniles (~130 g). Adult sex ratio is unbiased, and adult males and females show similar recapture rates, about 3 times that of juveniles. Unlike some T. carolina populations, our adult turtles exhibit no sexual size dimorphism in carapace length, carapace width, or mass. Most turtles at this site are active from late Mar until mid-Nov, spending the winter in underground hibernacula (sometimes exhibiting high site fidelity). Foraging areas are often quite distant from hibernacula and nesting sites; dietary items range from fruits and fungi to fossorial invertebrates and carrion. Nesting occurs during June and July, with clutch size averaging 3.7 eggs. Known mortality factors include drowning, predators, and pathogens. This study contributes to the knowledge of a widespread and once-"common" species, with the goal of developing management practices that more accurately reflect regionally variable needs of declining populations.


Movements and distribution of Terrapene carolina in a large urban area, Rock Creek National Park, Washington, D.C.

Ken Ferebee, National Park Service, Washington, D.C. 20008.
Rock Creek National Park is surrounded on all sides by the city of Washington, D.C.  At 710 hectares, it is one of the largest National Parks located exclusively within a major U.S. city.  It supports a diversity of habitats and is used extensively for human recreation.  Because of increasing traffic on interior and adjacent roadways, and a decrease in the number of live Box Turtles observed in the Park, we initiated a study in 2001 to determine the distribution and status of the Eastern Box Turtle populations.  We obtained information on the presence and distribution of Box Turtles throughout the park with visual encounter surveys of discrete sections of the park.  As of 2004, 80 turtles have been marked and 27% have been recaptured.  A 15.2 hectare study site was established in 2001 for monitoring marked and released turtles from May to October over multiple years.  After 3 years, initial estimates of population density at this study site were low, ranging from 0.97 to 1.61 turtles/ha.  To obtain information on habitat use and site fidelity, three female turtles fitted with radio transmitters were tracked during their active seasons. The home ranges for these three females, as defined by the term established activity range, was 2.0 ha (+/- 2.34 SD, range 0.4 to 4.8ha).  We include selected case studies to illustrate how these turtles use this highly urbanized park.  Today, the population study continues with additional turtles being fitted with radio transmitters and new and recaptured turtles still being recorded.  Updated data will be presented.  Given the older age structure, little to no sub adult recruitment, low productivity within Rock Creek Park, low population density, potential for high rates of road mortality particularly among adult females, the future survival of this population appears to be in jeopardy.  Our findings allowed us to explore the option of relocating park turtles within a large urban park as a management strategy for maintaining their populations.


Husbandry techniques for captive box turtles: Maximizing scientific value of natural history observations

Stephanie N. Foertmeyer, Richmond, Virginia.
Joseph C. Mitchell, Mitchell Ecological Research Service, LLC, Richmond, Virginia 23238.

Many well-meaning concerned citizens remove box turtles from roads and development sites throughout their range in an effort to “rescue” them. Turtles are released in local habitats or, as is most often the case, taken some distance away and released in natural habitat or kept in backyards. Most often, turtles are rescued individually; only rarely are entire colonies moved from sites undergoing urban development. We describe (1) the husbandry techniques developed by one of us (SF) to maintain a healthy population rescued from an urbanized site in Virginia and (2) the types of information obtained from observations of these turtles to demonstrate that detailed data can be derived from such efforts. Natural history observations on behavior, mating partners, phenology, overwintering, growth, egg laying, egg and clutch data, hatchlings and other parameters can be taken in ways to enhance their scientific value. The first requirement is to write down everything observed however trivial it may seem. Collection of eggs followed by incubation individually or by clutch by mother by year yields opportunities to ask scientific questions that could not be asked without detailed husbandry efforts. We describe the protocol used by SF to obtain such data on a single colony of rescued turtles and point out that similar efforts done in other parts of this species’ range would produce valuable scientific and conservation information.


A preliminary report on the status of the ornate box turtle (Terrapene ornata) in the sandhills of eastern Colorado

E Gangloff, AE Nash, and J Scupin of the Colorado Box Turtle Project

Despite widespread reports of declining populations, no studies have been conducted to date in Colorado on the ornate box turtle, Terrapene ornata.  The summer of 2007 was the first year of many for the Colorado Box Turtle Project, designed to begin gathering data to assess this turtle's status in the state.  On an area of approximately 50 hectares of sandhill habitat, 33 live turtles were found by visual surveys.  Initial data indicate that these turtles do not rely on standing water as a hydration source.  4 of 5 juvenile turtles were observed musking, a behavior about which little has been written.
This poster presents some preliminary findings after the first season of study and discusses future directions.


Hematology and Serum Chemistry Results from a Box Turtle Population

Jennifer Green, Glenn Olson and Paula Henry, Patuxent Wildlife Research Center, Laurel, Maryland.

Increasingly, box turtles (Terrapene c. carolina) throughout their range are having to contend with extensive loss of habitat to suburban sprawl and urban development, landscape fragmentation through the increase in roads, artificial medians, invasive plants, and changes in their microenvironment brought about by extended periods of drought and poor water quality.  As numbers decline in several states, box turtles are being listed in Wildlife Conservation plans as species of concern. With well meaning proposals to relocate box turtles “out of harm’s way” we need to be able to assess the physiological, health, and disease status of the turtles being displaced as well as the conditions of the resident populations where the relocated turtles are being introduced.  For this, we need to develop a profile of baseline ranges for what a healthy box turtle looks like. Researchers at the USGS Patuxent Wildlife Research Center (co-located on the Patuxent Research Refuge) initiated an effort to generate a blood chemistry profile on turtles found on the refuge as a way to evaluate the health status of box turtles.  For each analysis, a brief description of its function, the mean and range are presented.  Values are discussed relative to those published for other turtle species.


The ICC Box Turtle Project

Susan Hagood, The Humane Society of the United States, Washington, D.C.

The Intercounty Connector (ICC) is an 18.8-mile, six-lane highway that will connect I-270 and U.S. 1 in Montgomery and Prince George's counties north of Washington, D.C.  The selected alignment will impact a number of large county parks, many of which support populations of the eastern box turtle.  Beginning in late 2006, representatives of federal, state, county and local governments, as well as representatives of non-profit organizations, began meeting with the Maryland State Highway Administration (SHA) with the goal of reducing impacts on box turtle populations.  The Box Turtle Advisory Group (BTAG) has submitted to SHA a list of recommendations.  Principal among recommendations for immediate action was the organization of searches to find box turtles in the impact zone.  During the fall of 2007, teams of human and dogs collected over 200 turtles, most of which were taken to a secure pen to await receipt of 150 radio transmitters (turtles too small to be relocated in the pen are being overwintered indoors).  Once fitted with a transmitter, 150 were returned to their capture locations; the remaining animals will overwinter in the pen.  If  ICC construction schedules require forest clearing in areas where turtles with transmitters are known to be overwintering, individuals will be removed from dormancy sites and reburied outside the limit of disturbance.
Turtles in non-park patch habitats that will be completely destroyed by highway construction will be relocated to protected sites at some distance from their original home ranges.


The unseen effect of Box Turtle races

Alex Heeb, Chafee, Missouri.

Turtle racing is an event held at county fairs, community festivals, and other events across the United States.  For the event, wild caught box turtles (Terrapene carolina and T. ornata) are placed in a circle, with the first turtle to exit the circle being the winner.  An ongoing study was initiated in 2005 to determine how many box turtles are removed from the wild for turtle races, what effect this may be having on populations, and what husbandry techniques are used to care for the turtles.  Events holding turtle races were found through phone surveys and internet searches.  When available, data on the number of turtles entered in races were recorded.  I also attended turtle races and heard accounts from other people who had attended them. Results showed that turtles were mostly kept in unsanitary conditions that were detrimental to their needs. Turtles are usually not returned to their home ranges and are sometimes released en mass. Surveys found over 520 annual turtle races in 35 states. Based on entry data from over 50 races I estimate that over 26,000 box turtles are taken from the wild annually for these events. I discuss other issues concerning turtle races as well as a relocation study using turtles taken from races.  Due to the large number of turtles removed from the wild, these events could adversely affect box turtle populations. State wildlife agencies should consider prohibiting or regulating turtle races and entries should be closely monitored to detect any potential population declines.


Twenty-seven years with the other Box Turtle (Terrapene ornata)

John Iverson, Biology Department, Earlham College, Richmond, Indiana, 47374.

An overview of 27 years of field work on the natural history of ornate box turtles on the Crescent Lake National Wildlife Refuge in the sandhill of western Nebraska will be provided.  Discussion will include the value of long-term studies of relatively undisturbed box turtle populations in providing benchmarks for comparison with data from populations increasingly impacted by humans.  Attention will also be paid to those aspects of the natural history of box turtles that are most elusive (reproductive frequency, juvenile survivorship, population density, etc.), and the role that climate change may play in box turtle population dynamics.  Finally, the increasingly important role of individuals (not academics nor governments) in the long-term survival and success of box turtles will also be addressed.


Desert Box Turtles (Terrapene ornata lueola) in Arizona

Cristina A. Jones, Arizona Game & Fish Department, Phoenix, Arizona 85023.

The desert box turtle, Terrapene ornata luteola, occupies a variety of habitats that include Lower Madrean Evergreen Woodland, Chihuahuan Desertscrub, Semidesert Grassland, and Plains/Great Basin in southeastern Arizona, the extreme southwest of its range. The desert box turtle is listed as a Species of Greatest Conservation Need (SGCN) by the Arizona Game and Fish Department’s Comprehensive Wildlife Conservation Strategy. This inclusion was due in part to insufficient information about Arizona’s population and accounts from taxonomists and scientific experts that suspect it might be declining, but for which definitive information was unavailable. Because gaining information on this species is a priority, the Department has included a project to develop an occupancy model on our Heritage Grant Program’s Identification, Inventory, Acquisition, Protection, and Management sensitive elements list for 2007 ( Desert box turtles began receiving protection in Arizona when the Department closed their season on January 1, 2005, making it illegal to collect them from the wild. To handle or conduct research on desert box turtles, the Department now requires a Scientific Collecting Permit; and reviews and issues 3 - 4 permits annually. Recent and ongoing studies include: home range and movement, habitat and microhabitat selection, thermal ecology and reproductive ecology. Though we have limited field data and research targeting desert box turtles in Arizona, the threats that have been identified include roads, illegal collecting, and habitat loss to urban development or plant community conversion. The latter may result in a change in prey-base and increased fire intensity. Another concern is released pets which can lead to genetic swamping with ornate box turtles, Terrapene ornata ornata. Appropriate surveys and monitoring protocol need to be developed to determine the threats and status of desert box turtles in Arizona.

Radiotelemetry studies on hatchling, juvenile, resident and relocated adult, Three-toed Box Turtles (Terrapene carolina triungius) in East Texas

James F. Koukl, Dept. of Biology, The University of Texas at Tyler, Tyler, Texas,75799.
Alan Byboth, Camp Tyler Foundation, Tyler, Texas, 75799
Karen Rispin, Dept. of Biology, LeTourneau University, Longview, Texas
David Bolanowski, Dept of Anesthesiology, UTMB, Galveston, Texas
Jason Keyes, West Palm Beach Zoo, West Palm Beach, Florida

Over a seven-year period the behavior of hatchling, juvenile, and adult three-toed box turtles (Terrapene carolina triungius) has been examined via a series of studies.  For adults, both resident and translocated turtles were monitored.  During daily radio-tracking of each age group we collected data on specific movements, habitat preference, home range, and predation rates.  Turtles were radio-tracked on a 153-hectare mixed forest area in East Texas.  During radio-tracking and organized searches, 98 adult box turtles were captured, marked, and released with 12 recaptures recorded.  One hundred percent of 22 hatchlings and 16 juveniles radio-tracked were predated over a 4-month and 16 month period, respectively.  None of the adult turtles were predated.  Hatchling and juvenile turtles were cryptic in behavior and did not exhibit a tendency for dispersal, mainly remaining hidden under dense cover.  Resident adult box turtles exhibited a circular movement pattern compared to a more linear movement pattern for translocated turtles. Both resident and translocated radio-tracked turtles remained on the study site with the exception of one translocated turtle that swam across Lake Tyler (800m) in one day. After being returned to the study site this turtle traveled in the opposite direction from Lake Tyler and has remained for 7 years and behaves similarly to resident turtles.  In a more empirical study, we compared two different groups of radio-tracked adult, translocated, three-toed box turtles that were provided different acclimation periods prior to release: hard-release (2 weeks acclimation, n = 5) vs. soft-release (88 weeks acclimation, n = 5).  Minimum convex polygon analysis showed no statistical difference in home range between hard-released and soft-released groups.  Population status of Terrapene at the study site may be difficult to assess due to the larger numbers of older, adult individuals and inadequate techniques for locating cryptic hatchling and juvenile age classes.  These characteristics may allow for overestimation of population size, which may be detrimental to the long-term survival of populations and the species.  It appears that box turtle conservation will become a larger problem in the future as older, adult populations diminish in size and are slowly lost, especially given our data on recruitment. Preservation of these populations will depend on addressing core causes of decline such as habitat fragmentation and over-predation.  Data presented here will give better insight as to the behavior of age classes, predation rates, and possible population dynamics in forest fragments in East Texas.  In addition, data from our repatriation and translocation observations will also be discussed as a viable technique for supplementation and reestablishment of box turtle populations.


Population survey of Eastern Box Turtles prior to a silviculture clear-cut

Leah Lavoie, Erin Burns, Charlie Burns, Katie Sloop, John Rucker, Whitney Bell, and Tiffany Bell, Clinch River Environmental Studies Organization.

In 2006 researchers at the Clinch River Environmental Studies Organization (CRESO) began a population study of the eastern box turtle (Terrapene carolina carolina) in order to assess the impact of a silvicultural clear-cut.  The study site, located in Oak Ridge, TN, is owned and managed by the University of Tennessee Forest Resource Research and Education Center.  We systematically searched the pre-logging area in 50 square meter plots and beginning in July of 2006 dogs (Boykin Spaniels) were used to hunt for turtles.  The above combination of techniques has resulted in a total of 367 captures (342 by dog and 25 by people) and the use of dogs has enabled us to obtain a better estimate of population size and age structure in our sample.  Radiotelemetry was used to determine home-ranges, hibernation sites, reproductive behavior and resource use.  Radiotelemetry has also provided information to help define the boundaries of our post-logging search effort outside of the clear-cut area.  Observations of turtle behavior in our study suggest that certain trees affected with wet-wood disease, a bacteria infection affecting the cambium, are sometimes used as an energetic resource by turtles; an observation previously undocumented in the literature.


Terrapene: Laws and Regulations Regarding Possession and Commercial Sale

Evelyn Lee, Southport, CT

Box turtles (Terrapene spp.) have a long history of being collected, which is believed to have had a profound influence on their native populations. Recognizing several serious negative impacts of collection, including its unsustainability in a species with delayed mating and long lifespan, high mortality of collected specimens, and potentially devastating spread of disease from turtles which are collected and later released to native populations, state wildlife agencies have implemented regulations which prohibit commercial take in all but two of the thirty-nine states in which Terrapene naturally occur and have imposed limits on take for personal use in all but one. The aggregate of personal limits still allows for considerable loss from native populations. In addition, it appears that state departments of agriculture must tighten regulations on importation and possession of non-native species and subspecies of Terrapene (over which wildlife agencies have little or no jurisdiction) to honor the protective intent of wildlife agencies.  With increasingly restricted commercial take permitted, greater pressure for captive breeding programs will certainly arise as demand for turtles by hobbyists remains high. States must be prepared to address this issue, or, alternatively, expand education on the detrimental impacts of Terrapene collection and ownership so as to reduce demand.


What do search experiments show about how to assess Box Turtle habitats and populations?

Eugene R Meyer, Maryland Natural History Society, Baltimore, Maryland.

Search experiments can show how detectable animals are. An example is an experiment finding box turtle shells at Jug Bay. We asked volunteers to find shells at randomized locations. The shells were either entirely above the leaf litter or half-buried by leaves. Findings: 93% of the completely exposed shells were found by each search; they were in areas easy to walk thru. The find rates drop to about half, for shells half covered by leaves. The find rate dropped to zero for a shell that was placed randomly in a brushy area searchable easily only if a person walked on a narrow trail thru dense bushes. It was never found even when five people, some with years of field experience, looked for it.  Most of my search experiments use a variety of lifelike plastic replicas of snakes and amphibians in addition to turtles. The advantages of using several species include comparing taxa that intuitively seem camouflaged in different ways, and piggybacking work on species with populations that are not well documented, with species known to be restricted or imperiled.  Setting up search experiments is a learning experience as much as the results. Deciding the kinds of places to put shells or replicas leads to seeing the choices that a species makes in choosing habitat and cover. Detectability is the subject of much recent work on salamander populations, and a good article for an aquatic turtle population, and summarizing those articles shows that choosing habitats requires knowing the breadth and time of habitat use. Fortunately, knowledge of box turtle movements and habitat uses is now much clearer from Jug Bay’s recent telemetry and census surveys.  Training: search experiments gave me and other searchers (1) visual training for what the species look like partially concealed on actual backgrounds, and (2) walking training, meaning useful ways to walk thru a particular cover. This clarified how we look around, and how we walk around, visual screens. I will show how to set up a brief search experiment. First key detail: briefing searchers before an experiment is useful. We always start by picking up a sample so that we know what it looks like in a habitat. My instructions always state that we will find how an animal’s camouflage works in this habitat. It is definitely not an evaluation of individual searchers. It is a test of that species camouflage and location choices. Second key detail: at the end of a search I offer a debrief to show any that were missed. This can be a much-appreciated learning opportunity for each searcher. This presentation is intended to aid the participants’ conservation interests. The solutions to problems of detectability can be as simple as measuring them in ways that teach us while setting up an experiment that volunteers will enjoy searching, and that show afresh how we walk and look for species that can be remarkably camouflaged.


Long-term demography of the Eastern Box Turtle: What can 47 years of data from a protected site in southwestern Pennsylvania tell us?

Tricia A. Miller, Powdermill Nature Reserve, Carnegie Museum of Natural History, Rector, Pennsylvania 15677.
Walter Meshaka, State Museum of Pennsylvania, Harrisburg, Pennsylvania  17120
Andrew Mack, Powdermill Nature Reserve, Carnegie Museum of Natural History, Rector, Pennsylvania 15677

We are examining a 47-year mark-recapture data set of the eastern box turtle (Terrapene carolina carolina) from Powdermill Nature Reserve, an 890 ha field research station in the Laurel Highlands region of the Allegheny Mountains in Southwestern Pennsylvania. During 1960-2007, 82 turtles were marked of which 44% were recaptured at least once. Six percent of all new captures were juveniles (<100mm CL), 49% males (141 ± 10.8 mm CL), and 54% were females (136.7 ± 9.1 CL). Females were nearly three times more likely to be recaptured than males. Turtles have been observed from late April to early October. However, 97% of observations were from mid-May to mid-September.  Nesting mainly occurs during the month of June. The oldest male turtle first captured as an adult and subsequently recaptured was at least 65 years old, while the oldest female was at least 59. The maximum number of years between encounters for a single individual was 37 years. We are also exploring temporal changes in population size and structure within the context of successional changes and the implications for future land management decisions.


Size, structure and habitat of a population of Ornate Box Turtles in fragmented grasslands in Nebraska

Mark M. Peyton, Senior District Biologist, The Central Nebraska Public Power & Irrigation District, Gothenburg, Nebraska  69138.

The Central Nebraska Public Power and Irrigation District (Central) purchased 1,821 ha along the Platte River in Nebraska to be enhanced as habitat for endangered species.  The area, known as the Jeffrey Island Habitat Area, is fragmented grassland bordered by channels of the Platte River and riparian cottonwood/ash forest. Beginning in 2001 Central cleared 121 ha of the riparian forest in an attempt to develop a riparian grassland interspersed with temporary and permanent wetlands.  The purpose of the work was to develop and enhance habitat for migrating birds, specifically the federally endangered whooping crane (Grus americana) and interior least tern (Sternula antillarum) and the threatened piping plover (Charadrius melodius).  In addition other groups of animals, specifically small mammals, reptiles, and amphibians have been monitored to determine if they also utilized the developed areas.  Of the reptiles and amphibians present in nearby native riparian grasslands all have been collected in the newly developed areas with the sole exception of the ornate box turtle (Terrapene o. ornata). In 2006 ornate box turtle surveys were initiated in the nearby grasslands to determine range of occupation in these grasslands, size of the population, age structure of the population, and to determine the various components of the habitat utilized by the turtles. To date the surveys and mark-and-recapture work have shown a much larger and multi-structured population than was first suspected. While work is on-going, at this time it appears the habitat of the turtles in the native grasslands of the Jeffrey Island Habitat Area is confined to the transitional area between the riparian forest and grasslands.  Habitat components that appear to be the most likely indicators of box turtle presence are mulberry trees and the burrows of kangaroo rats (Dipodomys ordi) and/or pocket gophers (Geomys bursarius).  Work to be completed in 2008 is to utilize the data collected in the native grasslands during 2007 to evaluate the newly cleared area as potential habitat for ornate box turtles.  Because of barriers to immigration, box turtles may never colonize the newly cleared area, even if suitable habitat is available. Therefore, the final task of this study will be to evaluate the feasibility of transplanting individual box turtles into the cleared area in order to establish a self-sustaining population.


Corridors and habitat selection of the Eastern Box Turtle, Terrapene carolina, at Gifford Pinchot State Park, York County, PA

Sally Ray, Western Pennsylvania Conservancy, Wellsville, Pennsylvania 17365.

A population of eastern box turtles has been monitored at Gifford Pinchot State Park since 2001.  The purpose of this study has been to establish baseline data for the number of turtles present in the park and to learn more about their habitat use within the park.  Currently 173 box turtles have been identified.  Morphometric data collected on all turtles includes carapace length and width, hinge width and mass.  Each turtle has been photographed and marked with a unique identifying number by notching the marginal scales.  In addition to monitoring the population within the park, several aspects of their biology have been investigated: In 2001, eight box turtles (4 males, 3 females and 1 juvenile) were monitored by means of radio transmitters and tracked on a weekly basis from the beginning of June through the end of December.  The location of each turtle was recorded each time it was tracked with a GPS unit. These points were used to create maps showing the movements of the turtles and the size of their home ranges.  In 2006, another group of five females were outfitted with transmitters and tracked in an attempt to locate nesting and over-wintering areas as well as the corridors used to access these areas.  A map showing the locations of all turtle road crossings in and out of the park has been maintained since 2001 and is updated each year.  During the past five years over 30 box turtles have been killed on roads around the park.  This number does not reflect the true number of casualties as the evidence quickly disappears.


Collaboration for research and education

Ann Berry Somers, Biology Department, University of North Carolina at Greensboro, Greensboro, North Carolina 27402-6170.

A Box Turtle Research Group has formed in North Carolina to examine the feasibility of starting a widespread box turtle research effort that will engage trained citizen scientists in data collection.  The group consists of university researchers, state parks personnel, a NC Zoological Park curator, wildlife agency personnel, and a representative of the state museum of natural sciences.  We are analyzing the merits and pitfalls of conducting a multi-tiered project that will allow for different levels of involvement and commitment. The discussion of project goals and potential funding will continue into the fall with a pilot project planned for spring 2008.


Differences in the home range of male and female Eastern Box Turtles

Chris Swarth and Mike Quinlan, Jug Bay Wetlands Sanctuary, Lothian, Maryland 20711.

From 2000 through 2005, we used radio telemetry to study the home range characteristics of 50 box turtles (28 females, 15 males, and 7 juveniles) inhabiting a deciduous forest adjacent to a freshwater wetland.  Our 53 ha study site within the Jug Bay Wetlands Sanctuary (76.7o Long; 38.8o Lat) includes forests, managed meadows, wetlands and streams along the Patuxent River in central Maryland. This site is also a component of Maryland’s Chesapeake Bay National Estuarine Research Reserve. Mean female box turtle home range size was 6.2 ha (sd =7.0), whereas the mean size of male home ranges was only 1.2 ha (sd=0.61). Sixteen of 31 female home ranges exceeded 4 ha, and seven female home ranges exceeded 10 ha in size. By tracking the same five females over several seasons we obtained a more accurate home range estimate because individual size can vary considerably from one year to the next. Most females made extensive use of tidal wetlands and managed meadows, however  males restricted their activities primarily to forests. Meadows were used in June for nesting. Females used wetlands for feeding, thermoregulation and probably for restoring moisture lost during nesting forays in dry, exposed habitats. The differences in home range size is related mostly to female’s need to travel to meadows for nesting and to wetlands where they rehydrate and foraging. An appreciation of gender-based differences in home range can aid conservation efforts aimed at protecting all the various habitats that are required by box turtles.


Using plot surveys to assess Eastern Box Turtle (Terrapene carolina carolina) density and site fidelity

Chris Swarth and Mike Quinlan, Jug Bay Wetlands Sanctuary, Lothian, Maryland 20711.

Determining the density or merely just the presence of box turtles in an area can be a challenge for resource agencies that might need to make population assessments. Box turtles are generally spatially dispersed; they are cryptically colored; and they can remain hidden and immobile for days or even weeks beneath leaf litter and branches, or within dense vegetation. As a result, multiple site surveys may be necessary to accurately document their occurrence and abundance. In order to measure the density and site fidelity in a population of individually-marked eastern box turtles, we conducted 175 plot surveys from 2000 to 2007. Plots measured 100m X 100m (N=11) and were placed in a mixed-deciduous forest adjacent to a freshwater tidal wetland at the Jug Bay Wetlands Sanctuary, Patuxent River, Maryland. Surveys in a single plot were made every 7 to 14 days, extending from early May to mid-October. Each survey was conducted by 4 to 9 searchers and took about 45 to 60 minutes to complete. Mean single-survey density varied from 1.1 to 4.5 turtles/ha; these are minimal estimates because not all turtles in a plot are observed. High single counts yielded 8 to 9 turtles/ha. About 15% of surveys yielded no turtles. As many as 26 to 29 different turtles were recorded using a single one hectare plot over the course of several years. A high degree of site fidelity was confirmed for males and females: 41% of turtles were observed 3 or more times, and 3 turtles were observed 25 or more times on surveys. Based on this study,  8 to 10 surveys within the same plot over a season may be required to obtain a reasonable estimate of density. Plots adjacent to wetlands supported significantly more turtles and had higher densities than plots that were several hundred meters from wetlands. The proximity of water and wetlands appear to be important on a local scale for determining  turtle habitat selection and home range size. Repeat surveys within one hectare plots are an effective method for measuring box turtle density and could be a useful tool for assessing population status in selected areas throughout the species’ range. 


Demography and morphology of the Eastern Box Turtle (Terrapene c. carolina) in a partly agriculturally disturbed habitat, Letterkenny Army Depot, South-Central Pennsylvania

Kristina Tarasan and Pablo R. Delis Shippensburg University, Shippensburg, Pennsylvania 17257.

Worldwide declines in herpetofaunal populations, as part of a general biodiversity crisis, are deeply troubling for the scientific community.  At least anecdotal reports suggest that reptilian species such as the eastern box turtle, Terrapene carolina carolina, have been undergoing widespread population reductions. Unsustainable use and habitat loss and degradation are two of the main reasons supposedly responsible for these population declines. During 2007, we studied the demography and morphology of the population of T. c. carolina in Letterkenny Army Depot (LEAD) Zone 2, South-Central Pennsylvania, as part of an ongoing long term effort to compile data on the status of its general herpetofauna.  The site studied contains two contiguous sub areas (approximately 1,000 ha each); one of relatively pristine northeastern natural habitats and another of habitats intensively disturbed by agriculture.  To determine the effects of anthropogenically modified habitat on this population, we surveyed turtles using on foot transects and road cruising throughout the main axes of LEAD Zone 2. These techniques were used for approximately similar amounts of time and effort in the two sub areas: Mountain Side Zone 2 (NAT), a relatively undisturbed habitat, and Buds Lake Zone 2 (AGR), a highly disturbed agricultural habitat. During the first half of the 2007 season, morphological characteristics, frequency of injuries, and age and sex ratios have been preliminarily analyzed. To date, no turtles have been found in the agricultural sub area, while forty turtles have been found in the natural sub area.  We found 16 males, 21 females, and 3 juveniles, with the youngest being two years of age. More than half of the turtles found appear to be over 20 years old. Eight turtles were between the age of 15 and 20 and three in the age class of 10 to 15 years old.  Most common injuries seem to be chipping and scrapping of the carapace and plastron- with 12 turtles having shells that were not fully intact. Three turtles had one missing leg, one turtle had a missing foot, and four turtles had toes and/or claws missing or injured. There was one incidence of a dermal cephalic tumor.  During our study, we have documented three deaths: one caused by road mortality, the second apparently by respiratory infection and the third was of unknown causes as the turtle was found dead and intact in the field.  Preliminary findings suggest that box turtles do not utilize the agriculturally impacted zones within LEAD.  Routine use of heavy machinery, the dominance of monocultures, the exposed nature of the soils, and other consequences of agricultural practices might deter individuals from inhabiting this area.  On the other hand, the relatively pristine habitats seem to have a significant number of individuals, including juveniles, which hints a viable population dynamic.  These data and additional information to be collected in future seasons, will advance our understanding of long-term population trends of T. c. carolina in LEAD and aid in the conservation and management of this important species in Pennsylvania.


Characteristics of an isolated population of Terrapene ornata in Illinois

Shannon. L. Thol and Fredric. J. Janzen, Department of Ecology, Evolution, and Organismal Biology, Iowa State University, Ames, Iowa 50011.,

The long-term persistence of ornate box turtles (Terrapene ornata ornata) will require effective conservation and management strategies informed by scientific investigations.  Given the extent to which preferred ornate box turtle habitat has declined and become fragmented, population-level conservation decisions will be critical.  In Illinois, where ornate box turtles are threatened, an isolated population resides in a relict sand prairie (Carroll and Whiteside Counties, IL) used highly by humans and managed through controlled burns and vegetation removal.  Previous analyses of mark-recapture data for this population from 1996 to 2003 estimated a recruitment rate (l) of 1.02 (±0.06).  Genetic diversity was also previously analyzed and found to be similar to that of a larger reference population in Nebraska, despite evidence of a recent bottleneck.  Preliminary results indicate that individuals use the habitat non-randomly, with preferential use of woody trees and shrubs during hot mid-summer days.  Information from further investigations of habitat use, along with previous demographic and genetic data will be used to develop a suggested management plan for this population.


Observations from a long-term study of the desert box turtle in Arizona

Marty Tuegel, Arizona Ecological Services Office, USFWS, Tucson, Arizona.

The western box turtle is composed of two subspecies: the ornate box turtle (T. ornata ornata) and the desert box turtle (T. o. luteola).  The latter is the westernmost subspecies, ranging from the Pecos River in New Mexico west to the Altar Valley in Pima County, Arizona.  While many studies of the ornate box turtle have been published, relatively few studies have been conducted on the life history of the desert box turtle.  Data were collected over a period of 12 years on a population in the San Pedro River Valley of Cochise County, Arizona.  The study site is located on Fort Huachuca at the ecotone of semi-arid grasslands and Madrean evergreen woodland communities.  The site has no permanent water, but is transected by an ephemeral stream and associated xeroriparian community.  Observations were made through a combination of radio telemetry (1997-2001) and broad area searches.  Seasonal activity patterns were tied to summer monsoon rains, with individuals generally exhibiting crepuscular daily activity patterns.   Mean home range size is estimated at 4.76 hectares (ha) using 95% minimum convex polygons, with no differences between males and females observed.  Adult population size is estimated at 99 individuals, with a density of approximately 0.5 adults per hectare.


Ecology of box turtles in central Massachusetts

Lisabeth L. Willey, Organismic and Evolutionary Biology, University of Massachusetts, Amherst, Massachusetts 01003.
Paul R. Sievert, U.S. Geological Survey, Massachusetts Cooperative Fish & Wildlife Research Unit, and Department of Natural Resources Conservation, University of Massachusetts, Amherst, Massachusetts 01003

Eastern box turtles (Terrapene c. carolina) are rare throughout central New England and are a Species of Special Concern in Massachusetts. In 2005 we initiated a four-year telemetry / mark-recapture study to examine the ecology, demographics, and home range structure of eastern box turtles in the Connecticut River Valley in Massachusetts, at the northern limit of the species’ range. Over the first two field seasons, we tracked 62 individuals at six distinct sites for 1-2 years. Turtles were located by radio-telemetry twice a week during the active season (May-October), and over 5,000 locations were recorded. Total straight line movement distance over one season ranged from 135 to 2,200 meters. Home range size did not differ significantly between the sexes, but males tended to move consistently throughout the active season, while female movements were concentrated before and after nesting. Females, males, and juveniles utilized open areas such as residential yards, abandoned gravel pits, fields, and power-line corridors in spring and early summer, and deciduous and pine-oak forests for over-wintering. A single clutch was deposited by females during June into early July, and nest sites were limited to anthropogenic disturbed areas. Clutch sizes ranged from 3 to 10 eggs (mean=5.9, n=31). Nest depredation pressure varied across sites. Nests that were protected from predators had a 53% success rate (n=31). Straight carapace length, movement distance, and clutch size are generally larger than reported throughout the rest of subspecies’ range, consistent with latitudinal size increases observed in some related species. We will continue to monitor these populations through 2008 and draft a conservation management plan for the species in Massachusetts. Road and mowing mortality, collection, nest depredation, prescribed burning, and disturbance of nest sites by off-road vehicles could lead to population declines even at protected sites throughout the region. Appropriately timed management of habitats, education regarding collection, and human use restrictions on nest sites could help mitigate the continued loss and fragmentation of habitat across the state.



The Humane Society of the United States

Jug Bay Wetlands Sanctuary

North Carolina Zoological Park

Whalen Properties

James and Ann Robinson Foundation

Friends of Jug Bay

Gulf Coast Turtle and Tortoise Society

Mid-Atlantic Turtle & Tortoise Society