Recovery Strategy for the Blanding's Turtle, Nova Scotia Population, in Canada [proposed] 2011: Species Information

Date of Assessment: May 2005

Common Name (population): Blanding's turtle (Nova Scotia population)

Scientific Name: Emydoidea blandingii

COSEWIC Status: Endangered

Reason for Designation: The three small subpopulations of this species found in central southwest Nova Scotia total fewer than 250 mature individuals. These three subpopulations are genetically distinct from each other and from other Blanding's turtles in Quebec, Ontario and the United States. Although the largest subpopulation occurs in a protected area, its numbers are still declining. The other subpopulations are also susceptible to increasing habitat degradation, mortality of adults and depredation on eggs and hatchlings.

Canadian Occurrence: Nova Scotia

COSEWIC Status History: Designated Threatened in April 1993. Status re-examined and designated Endangered in May 2005. Last assessment based on an update status report.

Globally, the current range of the species extends from extreme southern Quebec and Ontario, west to central Nebraska, and south to Illinois and Indiana (Cochran and Lyons 1986, Ernst 1973, Olson 1987). Isolated local populations occur throughout the range and the Nova Scotia population is one of the most isolated (Herman et al. 1995). In Canada, populations are restricted to southern Ontario, the southwestern corner of Quebec, and the interior of southwestern Nova Scotia. Approximately 20 percent of the global distribution exists in Canada and less than five percent of this exists within Nova Scotia.

Blanding's turtle has a Global Rank (G-Rank) of G4, and a National Rank (N-Rank) of N4 (Natureserv 2010). The Nova Scotia population, which is the focus of this recovery strategy, has a Sub-National Rank (S-Rank) of S1. This population is listed as Endangered under both Canada's Species at Risk Act (2005) and the Nova Scotia Endangered Species Act (2000).

Figure 1: Three photos of Blanding's turtles. From left to right: 1a) Photo of an adult female Blanding's turtle in the water, showing its domed top shell (carapace), which is black with yellow flecks. 1b) Photo of an adult Blanding's turtle basking in the sun. This photo shows one of the most distinctive features of the Blanding's turtle, its bright yellow chin and throat. 1c) Photo of a newly hatched Blanding's turtle. The photo shows they're uniformly grey shell

Figure 2: Photo of the bottom shell (plastron) of a juvenile Blanding's turtle. The photo shows that the plastron is yellow-beige and has irregular black patches. In juveniles, annual growth lines are visible on the lower shell and are visible in this photo.

Blanding's turtles are medium sized freshwater turtles with a semi-hinged shell. They are very long lived (80+ years) and, in Nova Scotia, slow to mature (approx. 20 years) (Congdon et al. 1993, Herman et al. 1999, McNeil 2002). One of their most distinctive features is the bright yellow chin and throat (Figure 1b). Their high-domed top shell (carapace) is grey to black with yellow flecks (Caverhill and Crowley, 2008). The flecks are typically brighter in younger turtles and most visible when the shell is wet (Figure 1a). The lower shell (plastron) is orange-yellow with irregular black patches. In juveniles, annual growth lines are visible on the lower shell (Figure 2). After turtles mature, the rings begin to wear off and the plastron eventually becomes smooth.

In Nova Scotia, adult shell length ranges from 18 to 25 cm (Nova Scotia Blanding's turtle database 2010). Adult males are typically larger than females and can be distinguished by their concave plastron, thick tail base and solid grey upper lip. Newly emerged turtles, called hatchlings, are approximately 4cm long (the size of a toonie), and have uniform grey shells (Figure 1c).

In Nova Scotia, Blanding's turtles appear to be restricted to the southwestern interior of the province, although recent findings have expanded the extent of the range within that area (Caverhill 2006; McNeil 2002) (Figure 3). Studies of turtle movements and distribution, in concert with genetic assessment, have revealed that the Nova Scotia population complex is spatially structured, containing several distinct populations (Caverhill 2003, McNeil 2002, Mockford et al. 1999, Mockford et al. 2005, Toews 2004). To date, three genetically distinguishable populations have been documented, each separated by at least 15 km (Figure 3). These populations exhibit significant biological differences in behaviour, morphology, habitat use and fecundity (Caverhill 2006, McNeil 2002, Mockford et al. 2005). Even within populations, genetic structuring among concentration sites may be evident (Toews 2004)

(1189 kB, 3296 X 1949 pixels)

The Kejimkujik population occurs on the Mersey River watershed, almost entirely within the boundaries of Kejimkujik National Park and National Historic Site (Kejimkujik). The McGowan Lake and Pleasant River populations occur in working landscapes on the Medway watershed. Two smaller concentrations have also been identified within the population complex, one in the Tobeatic Wilderness Area on the Sissaboo watershed (4 turtles) and one in Dexter Brook on the Medway watershed (3 turtles). Additionally, several isolated sightings and unconfirmed reports along the Mersey, Medway and adjacent watersheds suggest that undiscovered populations may exist.

Based on our current knowledge of the species' range, it is believed that the entire Nova Scotia Blanding's turtle population complex contains approximately 350 adults. Estimated current population size for each of the known populations is presented in Table 1. Initial population modeling of two of the three populations suggests that both may be declining (Bourque et al. 2006). Although little is known of the history of Blanding's turtle in Nova Scotia, based on estimates of genetic diversity, it is possible that populations were once larger and more widespread than at present (Herman et al 2003). Estimates of gene flow suggest that over the last few generations the McGowan Lake population has gone from a source population to a sink population, while the reverse has occurred at Pleasant River (Howes et al 2009).

1 Kejimkujik population estimate based on two separate estimates: 1. Mean annual Jolly-Seber estimate using data from 1987- 2002 (Bourque et al. 2006). 2) Schnabel binomial estimate based on data census intervals 1969-1988 (Herman et al. 1995). Actual numbers of adults includes all adults marked from 1969-2010 and does not take into account individuals that may have died during this time period.

2 McGowan population estimate calculated using data from 1996-2002 using the Schnabel method.

Studies have shown that the degree of genetic variation in the NS population is similar to, or even higher than, that of populations in the species' main range (Mockford et al. 1999, Ruben et al. 2001). The NS population has also significantly diverged genetically from populations in the main range (Mockford et al. 1999, Mockford 2007, Ruben et al. 2001) and has been proposed as an important evolutionary unit of the species (Mockford 2007).

As a preamble to this section, there are two important considerations. First, this recovery strategy, as a requirement of SARA, adheres to the directives set out in that legislation. Second, Section 3.3 provides insight on Aboriginal perspectives on the recovery of the Blanding's Turtle as contributed by Mi'kmaw members of the recovery team, and through consultation with the Native Council.

Blanding's turtle habitat falls within Kespukwitk, one of the seven traditional districts of Mi'kma'ki. For this reason, it is important that the involvement of the Mi'kmaq living on and sharing the land is actively sought and encouraged. It is this continuum of Mi'kmaq throughout Kespukwitk, who through their sages, talks, and walks, will begin to reveal aspects of the Blanding's turtle. Mi'kmaq can make important contributions to the recovery of turtles through traditional teachings, revealing the importance of traditional practices, and sharing an Aboriginal eco-centric world view. Mi'kmaq customary use of biodiversity embodies the Mi'kmaq principle of netukulimk; a way of harvesting resources without jeopardizing the integrity, diversity, or productivity of our natural environment (Native Council of Nova Scotia 1994). A more inclusive approach to ecosystem based management may be particularly useful for Blanding's turtle recovery and also to other rare and at risk species that are tied to similar habitats in the watersheds of southern Nova Scotia. This recovery strategy cannot hope to illustrate the total knowledge of the Mi'kmaq. Recovery planning can be revised over time as involvement with the Mi'kmaq grows.

In order to integrate Aboriginal Traditional Knowledge (ATK) with other types of knowledge, it is important for recovery planners to understand how the Mi'kmaq world view may differ from other Aboriginal and scientific world views. Mi'kmaq traditional knowledge is not necessarily written, peer-reviewed, or published. ATK is a living knowledge, captured in oral language and culture, and which is highly specific to place and time – it is the ki of Mi'kma'ki. In sharing ATK, Mi'kmaq will often end by saying tan teli kji'jitu (as I know it to be), recognizing that the knowledge is living. In other words, the knowledge shaped from the land, e.g. about the Blanding's turtle, carried forward, and shared will differ between those experiencing that place at different times. It may change because of another action or because of another's perspective.

There are several legends and stories of Mikjikj, the Mi'kmaq name for turtles. These range from creation stories to legends of how the turtle got its shell. Some describe how the turtle was used by the Mi'kmaq and other stories of how and why specific turtles were named (Speck, 1923). The vast array of stories and legends signify a deep spiritual connection between the Mi'kmaq and turtles, suggesting there is much we can learn from their collective knowledge.

It is not known if Mi'kmaq people hold specific knowledge of Blanding's turtles but they likely hold knowledge of the turtle's habitat and that information has not yet been accessed. For example, ATK may be able to inform changes that have occurred in turtles' range and provide insight as to whether changes in distribution may be related to changes in habitat. In addition, the Mi'kmaw world view of planning for 7 generations into the future would be beneficial for understanding and advancing the necessity for planning for long periods that are biologically relevant to the long lived Blanding's turtle (e.g.10 generations for Blanding's turtle is 400 years).

This Recovery Strategy recognizes and incorporates Mi'kmaq knowledge in the recovery of Blanding's turtles and their habitats.

• First, by recognizing the significance that the Mi'kmaq People have shared a long history with the Blanding's turtle in Kespukwitk.
• Second, by inviting Mi'kmaq individuals and communities to sit with others, as peers, to discuss, draft, and implement recovery strategies and action plans.
• Third, by recognizing and acknowledging that it is important to gather and understand the collective knowledge of the Mi'kmaq about Blanding's turtles and their habitats.
• Fourth, by recognizing the need to include Mi'kmaw worldviews to ensure the value of netukulimk for the long term sustainability of Kespukwitk.
• Finally, by recognizing the valuable contribution of the Mi'kmaq and their traditional knowledge as an important step towards reconciling Aboriginal and scientific world views.

Blanding's turtles use a variety of habitats during their life cycle, and because of their longevity, individuals may have to shift these areas in response to changes in habitat over their lifetime. Limiting features of habitat are not currently known at this time. The following summary of known habitats used is presented to inform the identification of critical habitat.

Summer habitats: Blanding's turtles in Nova Scotia are found in a variety of habitats including fens, shallow lake coves, vernal ponds, and slow flowing brooks and rivers. They tend to use these habitats from April to late September. Within these habitats, Blanding's turtles tend to frequent shallow water (<2 m deep) containing abundant vegetation, often with deep, organic sediments (Ernst and Barbour 1972, Gilhen 1984, Graham and Doyle 1979, McMaster and Herman 2000, Power 1989, Ross 1989, Ross and Anderson 1990). Sites where turtles are present tend to have sweetgale (Myrica gale) or sedge (Carex spp.) overhanging the banks, few rocks (Bourque 2006), a mid range of water colour with peaty soils (Bourque 2006, Power 1989) and living sphagnum mats (McMaster and Herman 2000). Extensive beaver activity is also apparent at most known Blanding's turtle sites in NS. Juveniles typically occur in the same overall habitats as adults, although they may use different microhabitats within the larger wetland (McMaster and Herman 2000, McNeil 2002). Summer habitat for hatchlings and very young juveniles may vary. While some have been encountered in traditional turtle summer habitats, others have been radio tracked to wet areas in the woods or in small pools prior to moving to the larger wetlands (unpublished data).

Winter habitats: Blanding's turtle adults and juveniles often densely aggregate at their aquatic overwintering sites, and return to the same sites year after year (Caverhill 2006, McNeil 2002, Power 1989). They typically arrive at these sites in September and October and leave shortly after the ice melts in March and April. The characteristics and location of overwintering sites vary, including wooded pools or channels, railway trenches, and specific sections of streams or wetlands (McNeil 2002, Newton and Herman 2009, Power 1989). Sites typically have a deep organic bottom and undercut banks (Newton and Herman 2009). Temperature profile, dissolved oxygen level, water flow, water depth and vegetation composition vary considerably among sites (Newton and Herman 2009). Mating activity is often observed at overwintering sites, particularly during the fall (McNeil 2002) and basking activity occurs in nearby areas particularly in spring. Recent tracking studies of hatchlings show that their overwintering habitat is different from adults, as they spend the winter in a variety of habitats such as vegetated water or buried in moist substrates on land (Camaclang 2007).

Nesting habitats: Blanding's turtles nest in a variety of habitats, both natural and human-influenced, including lakeshore cobble beaches, slate outcrops, gravel roadsides, borrow pits, vegetable gardens, mine tailings, and recreational tracks (Caverhill 2006, McNeil 2002, Power 1989, Standing et al. 2000). Sites must be exposed to adequate solar heat gain to allow for incubation of the eggs, and are often facing southwest (Power 1989, Standing 1997). Nest sites may be close to water or considerable distances inland. Females may nest densely in one area or singly. Many, but not all, females show high fidelity to specific nesting areas (Standing et al 2000). However, they have also demonstrated the ability to shift sites when necessary. Nesting typically occurs in evenings in June-early July and females may spend several days at or near the site prior to nesting. Females lay only one nest per year, and may not nest every year. Hatchling emergence from the nest typically occurs from late August to mid October. During emergence, hatchlings may remain in the vicinity of the nest for several days before dispersing (Camaclang 2007, McNeil et al. 2000, Smith 2004, Standing et al. 1997). In rare cases, hatchlings have been observed to overwinter in the nest cavity (Nova Scotia Blanding's turtle database 2010).

Travel routes: Blanding's turtles are capable of travelling considerable distances both overland and along waterways. They move regularly among seasonal habitats and may make occasional long distance migrations (Kydd 2010, Power 1989). Turtles travelling to and from nesting sites have been shown to use the same travel route in multiple years (Kydd 2010). Newly emerged hatchlings often make extensive treks on land as they leave their nest site and seek an appropriate overwintering location (Camaclang 2007, McNeil et al. 2000, Smith 2004, Standing et al 1997).

Blanding's turtles in Nova Scotia are limited in their ability to respond to threats and habitat changes by their long generation time, physiology and small population size and geographic isolation from the species' main range.

As a long lived species with substantially delayed maturation (20 years), Blanding's turtles are especially vulnerable to increases in adult mortality. Even slight increases in adult mortality can drive populations to local extinction (Congdon et al. 1993, Heppel 1998). Because of their longevity and slow maturation, the species has a very long generation time (approx. 40 years), which can delay their ability to respond quickly to threats and result in significant time lags in recovery. These time lags also limit our ability to detect changes in the population until long after an event has occurred.

Restricted heat units and low productivity conditions significantly constrain Blanding's turtles in Nova Scotia. Sufficient heat units are required for successful incubation of eggs and emergence of hatchlings (Gutze and Packard 1987). In colder years, poor hatchling success, high rates of developmental abnormalities and lethargy in hatchlings are often apparent (Standing et al. 2000). Low productivity habitats may limit population density and in conjunction with short growing seasons, inhibit individual growth rates. As a result, maturation appears to be substantially delayed; in fact, Blanding's turtles in Nova Scotia have the highest age of maturity reported for the species (McNeil 2002).

As a small, isolated population complex, Nova Scotia Blanding's turtles are vulnerable to stochastic (chance) events (COSEWIC 2005). Local disturbances such as unusually high predation, extreme weather, or disease, which would have little effect on a large population, could be devastating. The species' tendency to densely aggregate at seasonal locations, such as nesting and overwintering sites exacerbates this risk.

Despite the long term research, knowledge gaps remain. As a result of their longevity and late maturation, accurately assessing the status of the Blanding's turtle population is a long term process. All of the research and recovery efforts listed below in section 6.1 aid in our ability to understand the population but the majority of this work has taken place in less than half of a turtle generation, which is estimated to be approximately 40 years (Herman et al. 2003).

Some important remaining questions include:

• What is the overall distribution in the province? Is this distribution currently shrinking, expanding and/or stable? How has it changed historically?
• Are the Kejimkujik and McGowan populations currently declining as indicated by the PVA? If so, why? What is the state of the Pleasant River population?
• What are the natural long term variations in survivorship and fecundity rates and how do they compare with elsewhere in the species range?
• How much habitat is necessary to sustain the population over the long term (allowing for shifts in habitat use)?
• What is the effective population size? Is the proportion of males successfully reproducing unusually low compared to elsewhere in the geographic range? If so, why is it low and does it reflect a recent change and/ or is it an indicator of a yet-to-be identified threat?
• What are the long term effects of ³headstarting on fitness?
• If laboratory incubation and headstarting is to continue to be an important component of recovery, what is the most appropriate incubation temperature to produce healthy males and females?
• What will be the long term effect of climate change?
• What is the long term effect of cumulative habitat loss / alteration and disturbance?

3 Headstarting is the process by which hatchling turtles are raised in captivity for a period of up to 2 years.