Peatlands are important ecosystems that are becoming increasingly vulnerable to climatemediated disturbances such as wildfire, which can threaten peatland hydrological, biogeochemical, and ecological function. However, the magnitude of these changes and their impacts to peatland-dependent species worldwide is a key knowledge gap. Peatlands in the eastern Georgian Bay, Ontario, region provide overwintering habitat for the eastern massasauga rattlesnake (Sistrurus c. catenatus), a species considered at-risk across its North American range. Overwintering habitat is considered suitable when peat temperature is above 0 degrees C and the water table position provides moisture without risk of prolonged flooding. This combination of suitable ecohydrological conditions, also known as the life zone or resilience zone, commonly occurs in hummocks which are raised microforms on the peatland surface. Due to a changing climate, peatlands are at risk of increased wildfire frequency and burn severity which may reduce overwintering habitat availability and suitability through changes in peat thermal and hydrological properties. In 2018, a wildfire burned over 11,000 ha of the eastern Georgian Bay landscape which supports critical habitat for the massasauga at the northern limit of the species range. To assess the potential impact of wildfire on massasauga overwintering habitat, we monitored water table position, snow depth, and peat thermal dynamics in hummocks across a burn severity gradient (unburned to severely burned) in three burned and three unburned peatlands across three winters (2019-2022). We found that hummocks were able to provide unflooded habitat above 0 degrees C regardless of peat burn severity; however, there was moderate evidence that hummock burn severity influenced mean daily resilience zone size. Overall, hummock overwintering suitability appears to be dominantly controlled by peatland surface topography and interannual winter weather. With the frequency and intensity of wildfires predicted to increase globally under a changing climate, it is critical to understand how interannual variability of winter weather conditions influences overwintering habitat suitability after wildfire, to identify peatland ecosystems that provide resilient species at risk habitat.

Turtle nesting habitat can be created as a restoration strategy to increase habitat availability or provide suitable habitat away from threats. Traditional nest habitat restoration consists of creating nesting mounds using a mix of sand and gravel. However, nesting mounds do not resemble natural turtle nesting habitat in a rock barren landscape where turtles nest in crevices and cracks in the bedrock. Therefore, our objective was to design and evaluate the success of a landscape‐appropriate design for turtle nesting habitat in a rock barren landscape. To evaluate success of the nest habitat design, we assessed the (1) survival of transplanted moss and lichen cover on created nest sites, (2) ecohydrological and physical conditions at created and natural sites, and (3) turtle egg hatching success at created and natural sites using a split‐clutch experiment. We found no difference in productivity between lichen transplants and natural sites, indicating that intact lichen transplants were successful. Moss transplant success was more variable due to moisture stress because transplants were conducted during dry conditions. In general, created nest habitat tended to have a more stable thermal and moisture regime compared to natural sites. When accounting for maternal effects, the odds of an egg hatching successfully was 6.6 times higher in a created site than a natural site. Overall, the success of our nest habitat design in the first few years suggests that this landscape‐appropriate design will be a useful restoration strategy for increasing turtle nesting habitat in rock barren landscapes.

AbstractEnergy absorption and flow through a nest is an important aspect of embryonic development in many reptile species including turtles. To date, few studies have explicitly attempted to quantify the energy flow through turtle nests, opting instead for the simplified approach offered by temperature index models. However, the quantification of the energy can provide an explicit abiotic link that can link biological models to biometeorological and ecohydrological processes and models. We investigated the energy flow through turtle nests occupying different bedrock morphologies within a Canadian Shield Rock Barren landscape, in Ontario, Canada. The taxons studied were Spotted Turtle (Clemmys guttata), Midland Painted Turtle (Chrysemys picta marginata), and Blanding's Turtle (Emydoidea blandingii). Nest temperature and soil moisture were measured in 2018 and 2019 using sensors placed in the soil adjacent to 12 turtle nest cavities. Three main rock morphologies were identified for each nest location, Crevice, Ledge, and Flat types, that are in order of decreasing bedrock percentage contact with the nest site. Ground heat flux and change in heat storage were determined using the calorimetric method for each nest, while the direction of energy flux between the atmosphere and the underlying rock was also determined. The Crevice nest morphology experienced the lowest ground heat flux on average (1.56 × 10−1 W m−2) and lowest cumulative heat storage (230 MJ) compared to the Flat (440 MJ) and Ledge (331 MJ) nests. However, over the diurnal cycle, large heat gains by Flat nests were mostly balanced out by nighttime heat losses. While Crevice nests saw the lowest daily heat storage gains, they experienced much lower heat losses over the evening period compared to the other nest types. Furthermore, we found that 59% of the energy is directed from the underlying bedrock into the Crevice nest, highlighting the importance of the bedrock in controlling thermal dynamics in the turtle nesting habitat. The lower variability in energy parameters for Crevice nest types can be attributed to higher amounts of nest‐to‐bedrock contact, compared to the flat nest types. Our results indicate that Crevice morphology may be ideal for turtles nesting at their northern limits because minimal heat loss during the evening can result in a more stable thermal incubation environment. Future conservation and habitat restoration efforts should consider the importance of bedrock morphology and prioritize the protection of Crevice nest sites. Furthermore, this work highlights important opportunities for potential interdisciplinary work between ecologists, climatologists, biologists, and hydrologists, specifically the integration of ecohydrological and biological models. This work also underscores the potential uncertainty of climate change impacts on turtle egg hatching success and nest sex ratios.

Natural wildfire regimes are important for ecosystem succession, but increased frequency and severity of wildfire due to climate change can negatively alter habitat characteristics. In 2018, over 11,000 ha of a granite rock barrens landscape that extends along the eastern shoreline of Georgian Bay, Lake Huron, burned in a wildfire. This landscape is a biodiversity hotspot providing habitat for many species at risk, including Emydoidea blandingii (Blanding's Turtle) and Clemmys guttata (Spotted Turtle), where turtles nest in shallow soil deposits in cracks and crevices in the bedrock. The burned, open rock-barren habitat had ∼70% fewer available nesting sites, but suitability of remaining nest habitat was unknown. To assess the impact of the wildfire on the suitability of turtle nesting habitat, we compared soil properties and in situ thermal and moisture regimes at turtle-selected nest sites, burned and unburned rock barren habitat, and burned forested uplands that may provide newly available habitat as a result of increased canopy openness following fire. Burned and unburned rock barren habitats drained quickly following rainfall, similar to turtle-selected sites, whereas burned forested uplands drained more slowly and provided wetter incubation conditions. Burned forested uplands provided a comparable thermal regime to turtle-selected nest sites and were often moss-dominated, with a relatively open canopy. Hatch success was estimated to be 20% lower at burned rock-barren habitats compared to unburned sites. Our findings suggest that severe wildfire affects soil thermal and moisture regimes, which can negatively alter the suitability of nest habitat, but burned upland forests are likely to provide suitable nesting habitat for at-risk turtles in the first years following fire.

iWetland is a community science wetland water level monitoring platform developed by the McMaster Ecohydrology Lab and tested from 2016 to 2019 in wetlands located east of Georgian Bay, Ontario, Canada. The goal of iWetland is to engage community members in wetland science while collecting data to better understand the spatiotemporal variability in water level patterns of wetlands. We installed 24 iWetland water level monitoring stations in popular hiking and camping areas where visitors can text the water level of the wetland to an online database that automatically collates the data. Here, we share our approach for developing the iWetland community science platform and its importance for monitoring all types of wetland ecosystems. From 2016 through 2019, almost 2,000 individuals recorded more than 2,600 water table measurements. The iWetland platform successfully collected accurate water table data for 24 wetlands. We discuss the successes and shortcomings of the community science platform with respect to data collection, community engagement, and participation. We found that forming mutually beneficial partnerships with community groups paired with strong outreach presence were key to the success of this community science platform. Finally, we recommend that those interested in adopting the iWetland platform in their community partner with community groups, recognize participant contributions, identify accessible sites, and host outreach activities.

Boreal peatland ecosystems are generally resilient to low severity wildfire. However, climate change may increase wildfire intensity and frequency, potentially shifting wetlands to less wildfire resistant states. Peatlands formed by infilling generally undergo a succession from open water to grounded peat, with spatially complex intermediate states which may impact wildfire resistance. We explored the relationship between wetland successional states and fire severity following a > 11,000 ha wildfire in Ontario’s Boreal Shield landscape. We digitized 144 wetlands of varying successional states from aerial imagery and assessed fire severity using the Relative differenced Normalized Burn Ratio calculated from Sentinel-2A satellite imagery. Completely peat-filled wetlands were small in area (0.7 ± 2 ha) but were most frequent on the landscape (n = 99/144) compared to a smaller number (n = 8) of large (19 ± 2 ha) spatially complex wetlands that comprised > 30% of the total wetland area. In wetlands covered completely, or nearly completely, by peat, fire severity was significantly higher compared to wetlands with interspersed patches of shallow open water. Moreover, > 90% of fire resistant wetlands with open water were associated with beaver dams. Wetlands with more complex surface cover experienced lower mean fire severity but greater variability in fire severity suggesting that variable fuel configuration (i.e., spatially heterogeneous surface cover) in complex wetlands limit wildfire propagation across the wetland surface. Our findings are important for landscape conservation and wildfire management, as spatially complex wetlands host a diverse array of habitats for at-risk species and may offer protection from severe wildfires as fire refugia.

Lichens and mosses are among the first organisms to colonize the open bedrock of eastern Georgian Bay, Ontario making them essential for primary soil formation and ecosystem succession, while also providing nesting habitat for turtle species‐at‐risk. However, the slow growing nature of lichen and moss makes them vulnerable to ecohydrological stresses caused by climate and land‐use change. In order to better understand how lichen and moss will respond to stressors, we examined which ecohydrological factors (e.g., near‐surface soil moisture and temperature) control the CO 2 exchange of lichen ( Cladonia spp.) and moss ( Polytrichum spp.) on rock barrens, and the time of year growth primarily occurs. Net ecosystem productivity (NEP) was significantly greater in the wet period of the growing season than the dry, with an estimated difference of 0.7 μmol m −2 s −1 for lichen, 2.9 μmol m −2 s −1 for moss, and 2.5 μmol m −2 s −1 for a moss and lichen mix. These findings indicate that the wet portions of the growing season are critical for growth, while lichen and moss have little to no productivity during the dry period. Our results indicate that near‐surface soil moisture is an indicator of the CO 2 exchange of lichen and moss, and this relationship varies among cover types. For the geographical regions where warm, dry conditions are expected to increase in duration and frequency with climate change, lichen and moss NEP will likely decrease, thus limiting the long‐term availability of nesting habitat for turtle species‐at‐risk.

Understanding thermal and moisture regimes in nesting habitat is an important step in management and restoration of at-risk turtle habitat because anthropogenic stressors that affect these key properties impact hatch success. In rock barrens landscapes, freshwater turtles are known to nest in unique shallow-soil deposits in depressions in the bedrock. Our objective was to provide the first comprehensive multi-scale assessment of turtle nesting habitat in a rock barrens landscape. At the nest-site scale, we characterized nesting habitat, tested for evidence of nest-site selection by the Blanding's Turtle (Emydoidea blandingii), Spotted Turtle (Clemmys guttata), and Midland Painted Turtle (Chrysemys picta marginata), and examined the effects of nest temperature and moisture on egg hatching success. At the landscape-scale, we conducted a systematic survey across 660 ha to assess the availability of nesting habitat. Although rock barrens were the second most dominant land cover type, the availability of suitable nesting habitat with deeper soil and an open canopy was limited to 11% of the rocky outcrops and <3% of the study area. We found that shallow rock barrens nesting habitat had unique soil temperature and moisture dynamics that were linked to soil properties and bedrock morphology to provide successful incubation conditions. In particular, mean daily temperature, variance of soil saturation during incubation, and their interaction had a significant effect on egg hatching success, regardless of egg mass. There was also evidence that turtles selected sites that were warmer and drained faster than paired haphazard sites despite having other similar characteristics. Given that the loss and degradation of nesting habitat is of conservation concern for the survival and recovery of at-risk turtle species, we recommend that key management strategies should include stronger protection of critical rock barrens nesting habitat and the development of landscape-appropriate strategies for restoration and creation of nesting habitat.

Identifying ecosystems resilient to climate and land-use changes is recognized as essential for conservation strategies. However, wetland ecosystems may respond differently to stressors depending on their successional state and the strength of ecohydrological feedbacks resulting in fluctuations in habitat availability and suitability. Long-term habitat suitability is necessary for the persistence of wetland-dependent species and a key characteristic of climatic refugia. In the present article, we review and synthesize biogeochemical, thermal, ecological, and hydrological feedbacks and interactions that operate within wetlands and, consequently, regulate overwintering suitability for many freshwater turtles and snakes. We propose that understanding the breadth and interconnected nature of processes controlling temperature, dissolved oxygen, and water table position are vital for the conservation of northern reptile populations that depend on wetlands to survive winter conditions. Finally, we suggest that our integrated framework can guide future research and the management of wetland ecosystems in an era of unprecedented change.

The eastern massasauga rattlesnake ( Sistrurus catenatus ) relies on small-scale differences in peatland surface elevation to survive harsh overwintering conditions at the northern limit of its range. We characterized the spatial heterogeneity of surface topography in peatlands within the eastern Georgian Bay rock barrens landscape of Ontario, Canada, to assess available snake overwintering habitat. At six peatlands, we used a differential global positioning system to collect surface-elevation data. We created spatially-explicit surface models to map peatland surface topography, quantify habitat that was likely to remain unflooded during the overwintering period, and identify key characteristics associated with greater habitat availability. While surface elevations were spatially heterogeneous within and among sites, larger peatlands were associated with greater surface spatial variability relative to the lowest elevation measured within each site. However, even peatlands with very little spatial heterogeneity (average of 0.24 m above lowest elevation), provided unflooded overwintering habitat. Inter-annual weather conditions and peatland and watershed characteristics likely control the availability and distribution of unflooded overwintering habitat. We found that trees, specifically white pine ( Pinus strobus ) and maple ( Acer spp. ), were spatially associated with higher surface elevations and could be used to identify areas of unflooded winter habitat. Our findings are useful for landscape-scale assessments of available overwintering habitat to prioritize conservation and management efforts.

Peatlands typically act as carbon sinks, however, increasing wildfire severity and annual area burned may challenge this carbon sink status. Whilst most peat resistance to wildfire and drought research is based on deep peatlands that rarely lose their water table below the peat profile, shallow peatlands and peat deposits may be most vulnerable to high peat burn severity and extensive carbon loss. To examine the role of pre-fire peat depth on peat burn severity, we measured the depth of burn (DOB) in peat of varying depths (0.1–1.6 m) within a rock barrens landscape. We found that DOB (0–0.4 m) decreased with increasing pre-fire peat depth, and that there was a strong correlation between the percent of the profile that burned and pre-fire peat depth. Breakpoint analysis indicates a threshold depth of 0.66 m where deeper peat deposits experienced little impact of wildfire, whereas shallower peat typically experienced high peat burn severity (median percent burned = 2.2 and 65.1, respectively). This threshold also corresponded to the loss of the water table in some nearby unburned peatlands, where water table drawdown rates were greater in shallower peat. We suggest that peat depth may control peat burn severity through feedbacks that regulate water table drawdown. As such, we argue that the identification of a critical peat depth threshold could have important implications for wildfire management and peatland restoration aiming to protect vulnerable carbon stores.

ABSTRACT Natural wildfire regimes are important for ecosystem succession but can have negative ecological effects depending on fire characteristics. A portion of a granite rock barrens landscape that extends along the eastern shoreline of Georgian Bay, Lake Huron to eastern Ontario, Canada, burned in 2018 during a wildfire that affected >11,000 ha. This landscape is a biodiversity hotspot providing habitat for many species at risk where freshwater turtles nest in soil deposits in cracks and crevices in the bedrock dominated by moss ( Polytrichum spp.) and lichen ( Cladonia spp.) cover. To assess the initial effect of wildfire on freshwater turtle nesting habitat, we measured soil depths and estimated moss, lichen, and vascular plant cover at 2 morphology types (crevice, flat) in burned and unburned areas of the landscape. The probability that burned flat plots supported soil was near zero; the burned flat plots had 98% less soil volume compared to unburned flat plots. Although crevices were more resistant to soil loss, burned crevices still had a 15% lower probability of having soil and 35% less soil volume compared to unburned crevice plots. We estimated nest site availability by calculating the number of locations with shallow (5–10 cm), intermediate (10–20 cm), and deep (>20 cm) soils required for a small (5 cm × 5 cm) or medium (10 cm × 10 cm) nest chamber. Overall, the burned open rock barrens had 71–73% fewer sites with suitable soil depth and volume for a nest chamber of either size. Furthermore, burned plots had almost no lichen and moss cover but were dominated by bare soil, forbs, and jack pine ( Pinus banksiana ) seedlings. Although the loss of tree cover in previously forested areas may increase nest site availability for freshwater turtles in newly open areas, we suggest that organic soil combustion and soil erosion may require restoration activities in the post‐fire landscape to support successful nesting of at‐risk turtles. © 2020 The Wildlife Society.

The suitability of overwintering habitat in response to temporal fluctuations in winter temperatures and water table position, and variability in timing and intensity of snowfall and rainfall is critical to informing management strategies for climate-sensitive species such as the Eastern Massasauga rattlesnake (Sistrurus catenatus). Important subterranean conditions contributing to the suitability of overwintering habitat for S. catenatus includes the availability of an unflooded space and temperatures above 0 °C which is conceptualized as the resilience zone. Here, we show that the timing and duration of resilience zone loss varies between years and among sites in response to variable weather conditions. Peatland habitat occupied by snakes lost the resilience zone fewer times and for shorter durations than unconfirmed habitat. In some cases occupied habitat continuously provided a resilience zone, suggesting that snakes may only be able to tolerate short periods of unsuitable conditions. As expected, larger precipitation events resulted in greater reduction of the resilience zone size. In particular, rain on snow events resulted in the greatest change in resilience zone size and could increase the risk of mortality events. With extreme weather-related events predicted to increase with climate change, long-term conservation of climate-sensitive species should focus on protecting habitats that are naturally resilient to changing and variable weather conditions and identifying vulnerable habitats that would benefit from additional research on adaptive management strategies.

In central Ontario, Canadian Shield rock barrens are a dominant geographic feature supporting at‐risk reptiles near their northern range limit. To better understand the characteristics of the organic soil that make Canadian Shield rock barrens suitable turtle nesting habitat, we measured moisture retention and evaporative potential and calculated the thermal properties of lichen (Cladonia) mats and moss (Sphagnum and Polytrichum) cushions, as well as their underlying mineral–organic soils. The upper soil profile consisted almost entirely of low density (14–49 kg m−3), high porosity (72–98%) organic matter (loss on ignition [LOI] of 84–99%), which transitioned rapidly to comparatively high density (304–815 kg m−3) mineral–organic soil (LOI of 10–85%). In contrast to Sphagnum and Cladonia, under laboratory conditions, Polytrichum was able to maintain an evaporation rate well above the open‐water potential for several days during a drying experiment. Overall, contrasts in composition and water retention between soil layers are likely to dampen diurnal temperature fluctuations. However, differences in potential water loss between species will have a direct impact on soil thermal dynamics, particularly if substantial water loss occurs in the mineral–organic layer. Because soil depth and temperature regulation by moisture content and soil composition are an important component of nesting habitat, this research provides evidence for the need to conserve moss/lichen‐dominated habitats within turtle species' home ranges. Understanding the ecohydrological controls and limits to how these key moss/lichen species develop and influence primary peat formation represents a critical research need for habitat restoration strategies.

Roads are one of the most widespread human-caused habitat modifications that can increase wildlife mortality rates and alter behavior. Roads can act as barriers with variable permeability to movement and can increase distances wildlife travel to access habitats. Movement is energetically costly, and avoidance of roads could therefore impact an animal's energy budget. We tested whether reptiles avoid roads or road crossings and explored whether the energetic consequences of road avoidance decreased individual fitness. Using telemetry data from Blanding's turtles (Emydoidea blandingii; 11,658 locations of 286 turtles from 15 sites) and eastern massasaugas (Sistrurus catenatus; 1,868 locations of 49 snakes from 3 sites), we compared frequency of observed road crossings and use of road-adjacent habitat by reptiles to expected frequencies based on simulated correlated random walks. Turtles and snakes did not avoid habitats near roads, but both species avoided road crossings. Compared with simulations, turtles made fewer crossings of paved roads with low speed limits and more crossings of paved roads with high speed limits. Snakes made fewer crossings of all road types than expected based on simulated paths. Turtles traveled longer daily distances when their home range contained roads, but the predicted energetic cost was negligible: substantially less than the cost of producing one egg. Snakes with roads in their home range did not travel further per day than snakes without roads in their home range. We found that turtles and snakes avoided crossing roads, but road avoidance is unlikely to impact fitness through energetic expenditures. Therefore, mortality from vehicle strikes remains the most significant impact of roads on reptile populations.

Abstract. Road networks threaten biodiversity and particularly herpetofauna, including common snapping turtles (Chelydra serpentina), which have an especially slow life history that prevents rapid recovery of populations subjected to road mortality. Cootes Drive is a 2.5-km 4-lane highway that bisects wetland habitat used for nesting and overwintering by snapping turtles. We hypothesized that turtle mortality from collisions with vehicles on Cootes Drive has caused a male bias and a decline in the population as turtles attempt to access habitat on both sides of the road. Capture–mark–recapture studies confirmed a dramatic decline in the turtle population from 941 individuals in 1985 to 177 individuals in 2002, a loss of 764 individuals in only 17 yrs. Using the same data, we also determined that the population has been significantly male-biased since 1985. Using 2009–2016 road mortality data obtained from the Dundas Turtle Watch (a citizen-science program), we completed a population viability analysis using the 2002 population size estimate to isolate the impact of road mortality. We found that this population is at risk of extirpation due to road mortality. The population range overlapped with the Cootes Drive and 7 of the 10 tracked turtles had individual home ranges that overlapped with the road. Our findings support the hypothesis that road mortality has contributed to the dramatic decline in the snapping turtle population in Cootes Paradise Marsh. This population is in jeopardy of extirpation; therefore, exclusion fencing must be installed for an extended distance along both sides of surrounding roads to prevent turtles from crossing the road and to promote their use of existing aquatic culverts.

In many jurisdictions, rare species and their habitats can receive protection if species are assessed as being at risk of declining. The assessment process requires data on habitat occupancy as well as identification of threats to a species critical habitat, both of which are difficult to obtain when the species occurs across large spatial scales. Such is the case for Eastern Musk Turtles (Sternotherus odoratus), which are obligate coastal wetland species in the Laurentian Great Lakes. We use data collected between 2003 and 2015 to map occupancy and conditional occupancy for musk turtles in coastal wetlands of eastern Georgian Bay (Lake Huron) to identify threats to resident wetland habitat. Data collected from a synoptic survey of 58 coastal wetlands were used to create occupancy models, estimate detection probability, and to conduct a sensitivity analysis to determine model robustness. We had a 64-71% probability of detecting musk turtles, whenever present in the wetland, and an area under curve value of 0.82 confirmed high model accuracy. Coastal wetlands that supported musk turtles were associated with higher proportions of forest cover, lower densities of roads, buildings, and docks within 1 km of the wetland, and more-variable bathymetric slopes. High conditional occupancy across the majority of our study area indicates that, at present, habitat in eastern Georgian Bay is in good condition; however, land-use alterations and development should be limited to ensure the persistence of this population of musk turtles.

At the northern limit of the Eastern Massasauga Rattlesnake’s (Sistrurus catenatus (Rafinesque, 1818)) range, individuals spend up to half the year overwintering. In hummock hibernacula found in peatlands, it is likely that subsurface temperature and water table position are contributing factors dictating habitat suitability. As a step towards assessing the vulnerability of hibernacula to anthropogenic changes, we combined subsurface temperature and water table dynamics to assess the likelihood that unflooded and unfrozen conditions were present in hummock hibernacula. Our results indicate that taller hummocks are more resilient to an advancing frost line and fluctuating water table by providing a larger area and duration of unfrozen and unflooded conditions, and a critical overwintering depth that is farther from the hummock surface. In two study sites along eastern Georgian Bay, an unflooded and unfrozen zone was present for over 90% of the overwintering period for hummocks taller than 25–27 cm. Our findings highlight the vulnerability of peatland hummocks to variability of winter weather where deep freezing and (or) water table rise may nonlinearly reduce resilience. This suggests that height is not the only component affecting the suitability of hummock hibernacula and that further research should examine the structure and spatial arrangement of hummocks within a peatland.