Plymouth Red-bellied Turtle Carapce

How Turtles Survive the Winter in New England and Canada

North American turtles are limited in their Northern range expansion, with only nine species of turtles having Northern range limits extending across the US-Canadian border that lies predominately on the 49th parallel (Table 1) (6).  The northern range limit for all turtle species is roughly 50°N, with the painted turtle (Chrysemys picta) being found at the highest latitude in North America (7), followed closely by the snapping turtle. This commonality suggests that there is a factor associated with latitude that determines the Northern range limit for all turtle species.

Turtle species vary in their over-wintering strategy, due in part to their physiological adaptations; some species have a better supercooling capacity and anoxia-tolerance than others.  A literature search on physiological adaptations essential for over-wintering in Northern Climates, including freeze tolerance, supercooling capacity, anoxia-tolerance, and extrapulmonary methods of gas exchange, was conducted to determine if these adaptations are essential to determining Northern range limits.  Thus, the literature search focused on the hypothesis that the rigors of over-wintering comprise the limiting factors preventing North American turtles from expanding their range further north into Canada.



Table 1: Canadian Turtle Species (6)*








Snapping turtle

Chelydra serpentine


Painted turtle

Chrysemys picta


Blanding’s turtle

Emydoidea blandingii

Map turtles

Graptemys geographica.

Western pond turtle

Clemmys [Actinemys] marmorata

Spotted Turtle

Clemmys guttata

Wood turtle

Clemmys [Glyptemys] insculpta

Common musk turtle

Sternotherus odoratus


Spiny softshell   turtle

Apalone spinifera


Table 1 lists North American turtle species with a NorthernRange limit that extends into Canada.

*sea turtles have been left out due to their ability to migrate large distances in the winter


Sea turtles have been excluded from this literature review due to their instinctive ability to migrate large distances to avoid cold waters.  The three species of North American tortoises (Desert tortoise [Gopherus agassizii], Texas tortoise [Gopherus berlandieri], and Gopher tortoise [Gopherus polyphermus]) have also been excluded due to their southern range and low over-wintering mortality (6).  The remaining turtle species that are considered in this paper are listed below in Table 2.


Table 2: North American Turtle Species with Northern Range Limits in Colder Climates











Snapping turtle Chelydra serpentina Canada   (6) Chelydridae
Eastern box turtle Terrapene Carolina Michigan & New England (6) Emydidae
Ornate box turtle Terrapene ornate South Dakota   & Wisconsin    (6)
Painted turtle Chrysemys picta  Canada   (6)
Spotted Turtle Clemmys guttata Canada   (6)
Bog turtle Clemmys [Glyptemys]   muhlenbergii New York, Connecticut    (44), Massachusetts   (45)
Blanding’s turtle Emydoidea blandingii -strictly a northern species-Nebraska,   Kejimkujik    National Park   in Nova Scotia,   and Great Lakes  region of southern Ontario (1,4,6)
Map turtles Graptemys geographica. Minnesota,   Vermont,   southern Canada   (Quebec) (6)
Wood turtle Clemmys [Glyptemys]   insculpta Nova Scotia,   Canada   (6,12)
Western pond turtle Clemmys [Actinemys]   marmorata Canada   (6)
Sliders Trachemys spp.formerly: Pseudemys   scripta northern Missouri,   Illinois, Indiana  (established population Massachusetts) (6)
Red-bellied turtle Pseudemys   rubriventris mid-Atlantic coast & disjunct population segment in Massachusetts(6)
Chicken turtle Deiro reticularia North Carolina   (46)
Diamondback terrapin Malaclemys terrapin Cape Cod,    Massachusetts(6)
Common musk turtle Sternotherus   odoratus Central Michigan and Ontario, Canada   (6) Kinosternidae
Mud turtles Kinosternon spp. Iowa (6)
Spiny softshell turtle Apalone spinifera Canada   (6) Trionychidae



Table 2 lists all terrestrial and aquatic turtle species found in North America in which over-wintering poses significant physiological challenges to survival based upon temperatures associated with their Northern range limit.

Hatchlings of Some Species Over-winter in Nests:

Turtle eggs are physically incapable of enduring harsh winters typical to northern climates.  Therefore, they must complete their entire incubation period and hatch in the late summer or autumn months in order to coincide with favorable temperatures (7).  Turtle species differ significantly with respect to their over-wintering behavior exhibited by hatchlings.  Since food is scarce in aquatic environments during the late fall and early winter (1), over-wintering in the natal, subterranean hibernaculum allows hatchlings to avoid predation and expend less energy.  Conversely, species that instinctively move away from their nests to aquatic environments before the onset of winter are more vulnerable to predation (2).  Mortality by predation is considerably greater for hatchling turtles when compared to adults of the same species, which is explained by carapace length.  However, entering aquatic environments that are not subjected to winter freezing may be the only survival strategy known to species with a limited supercooling capacity.

Snapping turtle (Chelydra serpentine) and Blanding’s turtle (Emydoidea blandingii) hatchlings do not over-winter in nests, and therefore enter aquatic environments to avoid freezing (1, 2, 4, 6).  Studies indicate that they emerge from their nests in late fall and spend winter months in the unfrozen depths of aquatic environments (i.e. freshwater ponds, streams, lakes, etc.) (1,4).  These two turtle species represent those [turtles] ranging furthest north into Canada, which exhibit this behavior.  Additionally, this behavior is characteristically observed with spotted turtles (Clemmys guttata), bog turtles (Clemmys [Glyptemys] muhlenbergii), map turtles (Graptemys spp.), wood turtles (Clemmys [Glyptemys] insculpta), red-bellied cooters (Pseudemys rubriventris), common musk turtles (Sternotherus odoratus), and spiny softshell turtles (Apalone spinifera) (6).  Leaving the nest before winter is the most common strategy exhibited by hatchlings of this latitude.  This suggests that avoidance of freezing temperatures is more important to survival than predator evasion and/or minimizing energy expenditures.

There is some evidence to suggest that Blanding’s turtles may fail to leave their nest before the onset of cold winter temperatures (1); the probable outcome of this is mortality.  November nest temperatures in some Minnesota populations have been recorded as low as -9 °C (1).  However, survival in nests may be possible if significant snowfall results in nest temperatures above -0.7 °C, or if it is dry enough to prevent ice from penetrating the nest to allow for supercooling (1).  This over-wintering behavior is atypical of the Blanding’s turtle, but is normal behavior to other species of turtles that range even farther north.

Painted turtles (Chrysemys picta), box turtles (Terrapene carolina and Terrapene ornate), and sliders (Trachemys spp.) have a dissimilar means of winter survival strategy as their hatchlings over-winter in nests (2,4).  For Painted turtles, utilizing this strategy has resulted in very high mortality rates in the northern part of their range, such as a population that had been observed in southeastern British Columbia (latitude: 49°15’) (2).  Death in this population of hatchlings was presumed to be caused by freezing (2).  Temperature recordings in nests indicated that they fell lower in some winters than the thermal tolerance limits of the turtles, which did not survive under such circumstances (2).  This phenomenon is likely common as the habitat range of the painted turtle extends the furthest north.  In other words, painted turtles living closer to the southern range limit are more likely to survive well into adulthood as they avoid temperature stresses related to over-wintering in colder (northern) climates.

Sliders typically have a more southern range, which may be reason for hatchlings over-wintering in subterranean nests during their first year. Southern climates are likely more favorable to survival rates in nests since temperature profiles are warmer year-round and risk for freezing is reduced.  Therefore, the slider (and other southern ranging species) can survive terrestrial over-wintering without specific adaptations for tolerating freezing temperatures or supercooling.  As aforementioned, predator evasion is more plausible for hatchlings that are better protected within their nests.  Furthermore, disjunct populations of sliders have been introduced to more northern locales extending beyond their natural range (i.e. New England) due to their popularity in the pet trade and subsequent release.  Although non-indigenous sliders may threaten native species, this unintentional field experiment may afford the opportunity to conduct research relating to over-wintering adaptations and rates of survival dependent of latitudinal location.

Other turtles that have been observed over-wintering on land include diamondback terrapins (Malaclemys terrapin), western pond turtles (Clemmys [Actinemys] marmorata), and chicken turtles (Deiro reticularia).  Diamondback terrapin hatchlings have been found burrowed 30 cm in sand approximately 8 meters above the mean high-tide mark of their brackish water habitat (6).   The western pond turtle (Clemmys [Actinemys] marmorata) and chicken turtles (Deiro reticularia) have been found over-wintering in leaf litter proximate to their freshwater habitats (6).  For freshwater turtle species, in more southern locales, this strategy may be important to evading spring floods (6).

Juveniles and Adults of Most Species Over-winter Underwater: 

The Eastern Box turtle (Terrapene carolina) and Ornate Box turtle (Terrapene ornate) are the only terrestrial turtle species in North America with a northern range limit comparable to other turtle species discussed in this paper (see Table 2).  These turtles do not over-winter underwater, but instead burrow in soil (6).  The Eastern Box turtle utilizes existing leaf-filled depressions resulting in hibernacula that are less than 20 cm deep (6). One study found burrows to be 14 cm deep with only 4-5 cm between the top of the turtle carapace and the surface, and leaf litter measuring an additional 8 cm on top (6).  Reported burrow depth for the Ornate Box turtle (Terrapene ornate) ranges from 50-180 cm. Both box turtle species are capable of moving around inside their burrowing holes and digging deeper when temperatures fall during winter (6).

Some aquatic turtles with a more southern range, such as the Western pond turtle and the chicken turtle, leave aquatic environments to over-winter.  Western pond turtles in northern California may leave rivers to over-winter in leaf and pine needle litter over 200 m away from their summer habitat; this selection is advantageous to flood evasion (9). However, they may choose to over-winter underwater if there are no fitting terrestrial environments for establishing burrows. Adaptation to survival is shown in this species by their over-wintering selection based on spatial and temporal (i.e. seasonal) environmental cues.

Similarly, chicken turtles (Deiro reticularia) are characteristically aquatic, but typically over-winter on land.  Studies have observed chicken turtles burrowing approximately 5-10 cm underneath leaf litter, detritus, and/or sand (6).  However, their over-wintering behaviors are erratic and may involve frequent migrations between their aquatic and terrestrial environments during a single winter period.  This behavior suggests that these turtles are not subjected to the same winter temperature lows that have caused more northern turtles to over-winter underwater. Diamondback terrapins live in brackish water creeks, and while there have been reports of possible terrestrial hibernation of hatchlings in sand, adults are found exposed on the bottom of creeks or buried 15-20 cm into the river bank (6).

The nine Canadian species of turtles over-winter underwater as adults (1, 2, 6, 7).  Aquatic environments selected by Blanding’s turtles include lakes, marshes and steams due to their higher oxygen concentrations in winter (1).   Other turtles that select normoxic bodies of water include Map turtles (Graptemys spp, Spiny softshell turtles (Apalone spinifera), and Wood turtles (Clemmys [Glyptemys] insculpta).  Map turtles are found exposed to the bottom of streams, rivers, and large lakes that have higher levels of dissolved oxygen (6).  Spiny softshell turtles also select normoxic water in streams, large lakes and rivers and burry themselves only slightly to permit extension of their head into the water column for buccopharyngeal gas exchange (6). Wood turtles are also associated with streams and rivers, and can be found in the bottom mud, under overhanging root systems of trees (12, 13), exposed on the bottom (13), or in muskrat burrows (6).

Snapping turtles (Chelydra serpentine), spotted turtles (Clemmys guttata), bog turtles (Clemmys [Glyptemys] muhlenbergii ), and painted turtles (Chrysemys picta)  select winter habitats with lower oxygen concentrations.  Snapping turtles select ponds, lakes, or streams (4) and may lay exposed (6), burry themselves in mud, utilize muskrat borrows, cattail stands, or wedge themselves under banks (10).   Spotted turtles and bog turtles  over-winter in swamps and bogs, with bog turtles also being found in muskrat burrows, partially buried, under sphagnum moss, and submerged in water under Alder roots (11).

Painted turtles may not always burry themselves in the mud as frequently described in the literature (14,15,16), but instead stay exposed on the bottom of the body of water (2).  Mud burial strategies may have been assumed and in the field due to carapace and sediment temperatures being similar, and not on observational data (24). In fact, there have even been some reports of painted turtles swimming under ice (25).  Studies that evaluated winter movements have found that painted turtles move minimally during the winter, but have been found 3 m away from previously noted locations in the same winter, and the turtles do move when threatened during winter capture (2).  One study found that ice cover ranged from 3 cm to 55 cm for up to 90 days, with a water depth of 30-85cm under the ice (2).  In this environment, turtles were found from 1-6.5 m from the shore (2).  The absence of mud burial has also been observed in iced-over Massachusetts ponds containing red-bellied turtles (Pseudemys rubriventris) by scuba divers in winter (17).  The absence of mud burial means that turtles will have more access to oxygen, but may also be exposed to increased predation (6).

 Exposure of Turtles to Freezing Temperatures in Field and Lab Studies:

Winter survival for most species appears to be by avoiding freezing of bodily fluids.  Aquatic species achieve this by spending the winter on the unfrozen bottom of fresh water habitats.   Hatchlings of some species such as painted turtle and box turtle species spend their first winter buried in soil, but seem to have more freeze tolerance than other species (6).  Hatchling ornate box turtles utilize a similar strategy to the adults, but locate their burrow directly under the nest cavity below the frost line (less than 50cm) (18).

One study recorded the lowest body temperature for a population of eastern box turtles as -1.4 ° C, making freezing unlikely (19). However, other authors discuss large numbers of turtle deaths due to cold snaps after emergence (6), so although their burrows may protect turtles from surface temperatures, high mortality occasionally occurs if temperatures drop suddenly after emergence and before turtles can burrow underground to protect themselves (6).  Another study reported two ornate box turtles surviving 54 days of temperatures less than 0°C with a temperature low of -8.0 °C.  The authors were not able to conclude if the turtles froze or were supercooled. The same authors also found two turtles in June that failed to emerge two months after all other Ornate Box turtles in the area.  When the turtles were recovered, they seemed weak giving the authors the impression that they may have died if left alone (8).  This lends support to the rigors of over-wintering playing a role in turtle mortality, which may partially explain Northern range limits.

Other studies that tracked Blanding’s turtle (Emydoidea blandingii) nest temperature in early November in Minnesota, recorded temperature lows of -9°C (1), however this turtle species strategy is to over-winter in the unfrozen depths of a nearby aquatic environments (1).  Occasionally, hatchlings may fail to emerge from nests; most of the time this results in death by freezing.

Field studies that evaluate winter nest temperature and survival rate for painted turtles have varied results.  Field measurements of over-wintering painted turtle hatchlings have recorded temperatures of -6 to -8 °C (27).  A study on an Ontario population found that all turtles survived a winter nest temperature of -8°C (28).  Whereas researchers studying a population in southeastern British Columbia, that was subjected to a nest temperature of -5°C and -6°C, counted only one living hatchling out of 19 nests (1987-88) one year and 34 nests the following year (1988-89).  They also counted 52 dead hatchlings, and 25 undeveloped eggs (2).  This suggests a very high mortality for hatchlings in the northern part of their range.


Other studies of midland painted turtles (Chrysemys picta marginata) site survival at -4 °C, and death at   -10.9 °C (28). A Nebraska population of painted turtles (Chrysemys picta belli) experienced 12.6% survival when the mean nest temperature was -9.5 °C, with individual nest temperature lows ranging from -6.5°C to -15°C (2).  Difference in survival rates at similar nest temperatures may be explained by the regions snowfall, as a Michigan study, where the nest temperature was -3.3 ° C,  found that survival rates ranged from 0-80% and negatively correlated with snow cover (29).

Over-wintering underwater usually allows the turtle to avoid freezing.  A study on spotted turtles recorded an air temperature as low as -35 °C, but a water temperature range of 0.3-3.9 °C (6).  However, large winterkills are also possible for turtles that usually avoid freezing by over-wintering in aquatic environments.  Christiansen & Bickham (1989) reported an event in Iowa where a pond froze to the bottom causing the deaths of  132 painted turtles, 8 mud turtles (Kinosternon flavescens), 12 snapping turtles (Chelydra serpentine), 8 map turtles (Graptyemys pseudogeographica), and 26 spiny softshell turtles (Apalone spinifera) (3).  Death by freezing was the likely cause, but low dissolved oxygen may have also been lethal.

Survive Freezing:

The equilibrium freezing point of a turtle is the point at which bodily fluids freeze when ice is present. Blanding’s turtles have an equilibrium freezing point of -0.7 °C (1).  Not freezing at temperatures below -0.7 °C could mean that the integument prevents the ice crystals from penetrating into bodily fluid compartments (1).  This strategy could allow turtles to survive freezing temperatures by not having their bodily fluids freeze and remaining supercooled.

Lab experiments show that Painted turtles have a better freeze tolerance than Blanding’s turtle.  Studies conducted in environmental chambers, in which the low temperature was maintained for 24 hours, indicate that Blanding’s turtles have a very limited freeze tolerance at -2.6 °C with 2 out of 10 turtles surviving freezing, and more reasonable odds of survival at -2.4 °C with 12 of 17 frozen turtles surviving (1).  No Blanding’s turtles survived freezing at -2.8 °C and below for 24 hours (1).  This limited tolerance for freezing for hatchlings may permit survival when temperatures drop in late autumn on their way to their aquatic winter habitat (1).   The limited freeze tolerance explains why over-wintering underwater is an essential strategy for hatchling Blanding’s turtles to avoid death by freezing.  In lab experiments that involved a dry, ice-free environment, Blanding’s turtles were able to survive temperatures of -4 °C provided that they did not freeze (9 out of 12), but all froze and died at -7.5 °C (1).

The Eastern Box turtle (Terrapene carolina) has a lower freeze tolerance of -3.6°C for 3-4 days (44-55% of body water frozen) (20), but the freeze tolerance for painted turtle’s is better still at -4 °C with 52-53% of body water frozen (28).  Freezing is typically fatal, but painted turtles are more apt to endure freezing when the same laboratory conditions are produced for sliders (4).  Storey et al. (1988) claimed that increases in glycerol and glucose by up to 3 fold may provide cryoprotective properties.  They also found increased pyruvate kinase, lactate and liver fructose 2,6-biphosphate  (28).

Supercooling Capacity:

Supercooling is when temperatures are below equilibrium freezing point of an organism’s bodily fluids (4). Hatchling painted turtles (Chrysemys picta) have a much better supercooling capacity than Blanding’s turtles. Survival beneath the threshold of freeze tolerance (-4 °C) depends on the ability of the turtle to be supercooled (3).  Laboratory experiments that ensure a dry, ice-free environment have shown that painted turtles can survive at  -12 °C  (31) and perhaps even -20 °C under  ideal conditions (32), whereas the supercooling limit of Blanding’s turtles is -6 °C (1).  Turtles brought to a temperature below their supercooling limit will freeze spontaneously despite being held in an environment that is dry and ice-free (1). Provided that hatchlings do not freeze, their survival is possible at these temperatures.  The key is to prevent inoculation by ice crystals (4).  The supercooling limit for sliders is reported to be -5 °C (Lowe et al. 1988) (2).   Hatchling ornate Box (Terrapene ornate) supercool in soil to -2.4 °C for 2 days (33).



Role of Ice Nucleating Agents in Supercooling Capacity:

Heterogeneous nucleators or ice-nucleating agents (INA) are substances in the fluid compartments of an animal that promote crystallization of bodily fluid resulting in the freezing of the animal.  Laboratory experiments on supercooling Blanding’s turtles indicate that there are no heterogeneous nucleators that prevent supercooling to -6 °C in the lab (1).  However, in Painted turtles (Chrysemys picta), supercooling capacity increases with cold acclimation in laboratory experiments (3).  There seems to be a seasonal development of supercooling capacity that requires elimination of ice-nucleating agents (INA) of exogenous origin (3).  Supercooling experiments have not shown the involvement of antifreeze proteins or cryroprotectants (34).

Ice nucleating Agents (INA) affects the crystallization temperature of bodily fluids.  Experimenters studied the effects on substrate and crystallization temperature (3) in a west-central Nebraskan population of painted turtles (Chrysemys picta belli).  They found that crystallization temperature (Tc) was lower when kept on a paper substrate than on vermiculite.  Keeping turtles on a nesting soil substrate caused the crystallization temperature to be nearly 6°C higher (3).   This may indicate the presence of INA in the soil.  The authors of this study also tested turtles that had over-wintered in nests with a mean temperature of -9.5 ° C.  These turtles had a crystallization temperature of -7.5 °C, which indicates the possible presence of INA (Table 3).  The authors could not explain why the crystallization temperature was higher than the nest temperature that these turtles had just survived (3). 

Table 3: The Effect that Substrate has on Crystallization Temperature of Painted Turtles
(Chrysemys picta belli) in west-central Nebraska (3)








Recently hatched   summer turtles tested shortly after hatching (3)

22 °C

-11.1 °C


-10.5 °C


-4.6 °C

nesting soil

Tested in cold   acclimated winter turtles & elimination of ingested substrate and rarely   eggshell (3)

4 °C

(acclimated to 4 °C   for 8 weeks)

-16.7 °C


-16.2 °C


-10.4 °C

nesting soil

Over-wintering in   natural nests

(intestines   contained white, caseous material)


(mean nest temp:   -9.5 °C )

-7.5 °C

(authors perplexed   as to why nest temp is 2 °C lower)

natural environment


Anatomical/Histological Adaptations:
Hatchling Blanding’s turtles, snapping turtles and sliders have an integumentary system that bears little defense against inoculation by ice crystals (31).   In contrast, hatchling painted turtles that over-winter in nests are able to be supercooled to lower temperatures.   This resistance to ice penetration may be due to certain anatomical adaptations.   Painted turtle hatchlings (3-6 g) have a dense lipid layer in the basal portion of the α-keratin layer of the epidermis of the forelimb (4) that is not present in sliders, Blanding’s, or snapping turtles (4).  An experiment that tested the efficacy of this lipid layer in the prevention of freezing compared painted turtles to sliders when kept at -2°C for 6.5 days.  When this region of the epidermis was cut, all turtles froze in both groups, whereas only 1 of 11 painted turtles froze when the epidermis was left intact.  In contrast, 5 of 11 sliders froze despite an intact epidermis.  These experiments also show that painted turtles are more likely to survive freezing as all frozen sliders died, and 10 painted turtles out of the 13 that were frozen lived (4).

The other anatomical adaptation that allows turtles to prevent inoculation by ice is the ability to withdraw into the shell.  This is important for painted turtles, whose hatchlings over-winter in nests, since their neck, axillary and inguinal regions do not contain the protective lipid layer (4).  This also may help box turtles evade inoculation by ice and allow them to supercool slightly.

Range Limitations due to Oxygen Requirements:
Studies that evaluated turtle species for anoxia-intolerance found that painted turtles and snapping turtles are capable of surviving anoxia at 3 °C for over 118 days while musk turtles, spiny soft-shell turtles, and common map turtles can only survive for 45 days (Table 4) (6).

Table 4: Oxygen Requirements during Over-wintering (6)


 Studies Indicate   Anoxia-tolerant (>118 days @ 3°C) (6) Painted turtle Chrysemys picta
Snapping turtle Chelydra serpentine
 Winter Habitat   Suggests Anoxia-tolerance (6) Spotted Turtle Clemmys guttata
Bog turtle Clemmys [Glyptemys]   muhlenbergii
Blanding’s turtle Emydoidea blandingii
 Studies Indicate   Anoxia-intolerance (6) Spiny softshell turtle Apalone spinifera
Common Map turtles Graptemys   geographica
Common musk turtle Sternotherus   odoratus
Winter habitat   suggests anoxia- intolerance (6) Wood turtle Clemmys [Glyptemys]   insculpta

Table 4 illustrates the varying ability of turtles to withstand low-oxygen environments.  Predictions of anoxic-tolerance based on winter habitat are also included.

Oxygen levels appear to be more important in habitat selection for the anoxia-intolerant group listed in the table above that choose rivers, lakes and streams, whereas painted turtles and snapping turtles do not have to be as selective due to their anoxia-tolerance, and can potentially live in any body of water that does not freeze to the bottom in the winter (6).  Some anoxia-intolerant species, such as the musk turtle, selects habitats with higher oxygen levels in the northern part of their range, but are less picky further south where there is less ice cover over bodies of water that would otherwise produce very low winter oxygen levels if located further north (6).

It has not been established that hypoxia is limiting to over-wintering painted turtles (2). Oxygen levels fall with ice cover, but turtles remain exposed on the mud bottom in the more oxygen-rich water column as discussed above (2). Studies on a British Columbia population found that dissolved oxygen in Hidden Lake, B.C was 8.6-9.4 parts per million in December, 1988, and 5.5-6.2 parts per million at the end of March 1989 (2).   Painted turtles did not seem to be effected by this drop in oxygen concentration (2). Painted turtles are also aided by two hemoglobin variants (the S and D form), with the S form dominating painted turtles at higher latitudes.  This is advantageous because this hemoglobin variant has a higher oxygen affinity, which is useful at low oxygen concentrations (7).

 The ability to tolerate anoxic conditions must be critical in surviving at high latitudes where there is significant ice cover during the winter, as the anoxic-tolerant group (painted turtles and snapping turtles) has the most northern range limit.  However, anoxic water has not been proven as the limiting factor that determines the northern range limit.

Metabolic Depression and Energy Storage for Hibernation:
In comparison to aerobic metabolism at 20°C, the metabolic rate of over-wintering turtles can be depressed by as much as 99.4% due to anoxia caused by being submerged under water, fasting, and temperatures nearing freezing (35). Fasting can account for 69% of this depression after 19 days (36).

This lowered metabolic rate can be observed as a depressed heart rate and blood pressure.  The heart rate of painted turtles conducted at 3 °C was recorded at 5-6 beats/hour, with a blood pressure of 10.5/1.6 (35).  This metabolic depression allows glycolysis to act as the main ATP producer with limited metabolic acidosis.  Enough ATP is produced to power the Na+/K+ pump of cell membranes, preventing changes in Na+ and K+, and ensuring the maintenance of homeostasis.

Despite having a lowered metabolic rate, turtles still require stored energy in macromolecules to survive over-wintering.  Although there are reports of turtles moving under the ice, no studies have observed any form of feeding under ice cover.  Turtles do not have fat bodies for lipid storage as other ectotherms, such as frogs, do (37).  However, glycogen storage by cardiac tissue is much higher in turtles than it is in mammals (38).

Studies that evaluate lipid, protein, and glycogen stores before and after hibernation differ in their results.  One study reported a decrease in neutral lipids of 32.7%, proteins of 9.9%, and liver glycogen of 38.4% (7), while other studies found no decrease in total body fat or liver lipids (7).  Cardiac glycogen stores seem to be exhausted first, then liver glycogen, which acts as the main storage site, increasing to 50% of the liver dry weight in map turtles (Graptemys pseudogeographica) just before over-wintering , then muscle glycogen (39).

 Extrapulmonary methods of gas exchange:
All turtles are obligate air breathers during their active period in warmer months, but studies indicate that turtles are also capable of extrapulmonary respiration, that is, exchanging respiratory gases with water (7).  The integument (21) and heavily vascularized villiform processes in the buccal/pharyngeal cavity (Figure 1) (23) have both been described. The later has been termed buccopharyngeal gas exchange.  Many turtles that over-winter slightly covered by bottom mud of aquatic habitats extend their neck into the water column, presumably to facilitate increased buccopharyngeal pumping in water with a higher oxygen concentration than the mud (6).  This buccopharyngeal pumping has been observed by researchers in the field (22).
This method of extrapulmonary respiration ensures that turtles are able to obtain oxygen for cellular respiration; this allows turtles to avoid producing lactic acid via anaerobic respiration.  Having oxygen available, to go beyond glycolysis, increases metabolic efficiency, and ensures that more ATP can be generated for each glucose molecule arising from glycogen storage sites in the liver and cardiac tissue.  When oxygen is limiting, the lowered metabolic rate of over-wintering turtles ensures that glycolysis alone is sufficient at providing ATP levels required for maintenance of homeostasis, with minimal lactate build up.  Extrapulmonary methods of gas exchange may supplement air breathing in warmer months, but is essential to over-wintering.

Blood Lactate Levels During Over-wintering Turtles:

Turtles are unsurpassed in their ability to withstand hypoxia and lactic acidosis among vertebrates (2). Despite physiological adaptations allowing oxygen uptake from the water, and a lowered metabolic rate due to cold temperatures, lactic acid build up is inevitable as oxygen concentrations decrease under thicker ice cover.  In painted turtles, lactate acidosis of over-wintering adults in aquatic environments appears to not be limiting (2), although they do emerge lethargic and weakened after not breathing air for 4-6 months under ice cover (7).  Their success at preventing higher lactate levels is due in part to a lowered metabolic rate during cold winter temperatures. Extrapulmonary respiration may play a role with studies showing that many habitats selected seem to contain enough oxygen in the water (2).

Lactate levels observed in the field are below maximum tolerable levels seen in lab experiments, and although oxygen levels may fall due to ice cover, it has not been established that this is limiting for painted turtles (2).  Field observations in British   Columbia show lactate levels of 20-29 mmol/L.  This is after more than 90 days of ice cover at temperatures of 1-3 ° C.  This is much lower than levels shown to be tolerable in lab experiments, where lactate levels of 90 mmol/L have been seen (Table 5) (2).  

Table 5: Lactic acid levels








(40) 62 mmol/L 67 d in outdoor tank at 0-8 °C (mean 3.7°C)
(41,42) 90 mmol/L 90 d @ 3 °C under severe hypoxia
(41, 42) 30-85 mmol/L oxygenated water (9.4 ppm) @ 3 °C
(2) 20-29 mmol/L >90 d ice cover in shallow lakes of Southeastern   B.C.  @1-3 °C  (may have lower lactate due to lowered   metabolism at lower temp)

Table 5 shows that Painted turtles observed in the field under ice cover have much lower lactate levels than lactate levels shown to be tolerable during lab experiments.

Along with metabolic depression, a heavily vascularized shell, and the ability to utilize buffers from the skeleton to sequester lactate, helps prevent excessive metabolic acidosis (6). This allows the three northern subspecies of painted turtles to survive for 150 days in anoxic water (6).   The turtle with the poorest lactate buffering capacity is the spiny softshell turtle (Apalone spinifera) that has limited ossification of shell (43).  This is further evidence of the shell acting as an important mechanism to prevent lactate build-up. 

Freezing tolerance and supercooling capacity does not appear to be limiting to the northern range expansion for turtles that over-winter in aquatic environments as they are able to remain unfrozen under ice cover in environments with sufficient oxygen.  It is evident that certain species of turtles have an innate ability to survive under unfavorable winter conditions, such as anoxic water at low temperatures.  These same species are optimally adapted to buffer lactic acid, enabling them to maintain acid balance under anoxic conditions.  Furthermore, studies show that lactate levels seen in the field near the northern range limits are considerably lower than limits shown to be tolerable for survival in laboratory experiments, at least for painted turtles.  This suggests anoxia alone is not restricting the northern expansion of turtle species.

Turtle hatchlings that over-winter in nests are severely limited in their northern range expansion due to higher percentages of mortality; this includes painted turtles.  Although they have a better supercooling capacity and freeze tolerance, the over-wintering strategy observed for these hatchlings ensures that they will not be able to expand their range to latitudes with temperatures lower than their typical supercooling capacity seen in the field.  Reproductive limitations, such as nesting sites that provide suitable temperatures for incubation, may play a larger role in this function.

Further study is needed to identify more-northern habitats that may exist that may provide favorable incubation conditions that are essential to embryonic development and subsequent hatchling survival.  To discern if more northern locales provide proper incubation temperatures, a study should be conducted in which recently laid turtle eggs are uncovered and placed under similar conditions at higher latitudes above the northern range limit.  Successful hatching of these eggs would help determine if there are reproductive barriers to a northern range expansion.  However, this study would have limitations as eggs may not be laid by the same date at higher latitudes.  Studies that look for correlations between latitude and the date that eggs are laid by each species may be useful here.

At higher latitudes, the active warm-weather season is shortened.  Turtle hatchlings for most species do not emerge until September (7).  The time required for reproduction and egg incubation is fairly homogeneous among turtle species discussed in this paper.   Studies that evaluate mortality of hatchlings over-wintering underwater, and temperature profiles of nests during the incubation period near the northern range limits of the Canadian species need to be conducted to further evaluate reasons for the homogenous northern range limit (9).

Future studies should utilize radio telemetry and release both adult turtles and hatchlings into aquatic environments for over-wintering above their northern range limit.  Mortality rates should be tracked closely, and if survival is high it would help rule out over-wintering as the limiting factor in expanding the northern range limit.  This should be tracked over a number of seasons to eliminate seasonal variation in temperature as a source of error.  Canadian study sites would have to be located where there would be minimal impact on native species.

The majority of studies to date are on painted turtles, or compare painted turtles a few other species.  Although many studies evaluate similar adaptations, such as supercooling capacity, freeze tolerance and anoxia tolerance, all studies differ slightly in their methodology, which leaves room for error when comparing physiological limits across species.  Studies that utilize the same equipment and methods for all species listed, that also take specimens from the same latitude would be insightful.

Implications of this research would not be to encourage a more northern range of turtles, but to understand the reasons for the northern range limit.  Because the northern range limit is roughly the same around the globe at 50°N, changes to this upper range limit may lend support to global warming altering the range of specific species if there was future evidence of turtles occurring further north.  Understanding which adaptations are most limiting to a range expansion will also help determine what direction conservation efforts should be taken to maintain current range boundaries of turtle species. Current evidence supports the supposition that over-wintering plays a role in preventing North American turtles from expanding their range further north into Canada, but does not rule out other reproductive factors from also determing the northern range limit.


1. Packard, G.C., Packard, M.J., Lang, J.W. (2000) Why Hatchling Blanding’s Turtles Don’t Overwinter inside Their Nest. Herpetologica, Vol. 56, No. 3 (Sep., 2000), pp. 367-374

2. St. Clair, R.C., Gregory, P.T. (1990) Factors Affecting the Northern Range Limit of Painted Turtles (Chrysemys picta): Winter Acidosis or freezing? Copeia, Vol. 1990, No. 4 (Dec. 31, 1990), pp. 1083-1089

3. Costanzo, J.P., Baker, P.J., Dinkelacker, S.A., Lee, R.E. (2003)  Endogenous and exogenous ice-nucleating agents constrain supercooling in the hatchling painted turtle. Journal of Experimental Biology  Feb;206(Pt 3):477-85.

4. Willard, R., Packard, G.C., Packard, M.J., Tucker, J.K. (2000) The role of the integument as a barrier to penetration of ice into overwintering hatchlings of the Painted Turtle (Chrysemys picta) Journal of Morphology, 2000 Volume 246 Issue 2, Pages 150 – 159

5. Packard, G.C. (1997) Temperatures during Winter in Nests with Hatchling Painted Turtles (Chrysemys picta) Herpetologica, Vol. 53, No. 1 (Mar., 1997), pp. 89-95

6. Ultsch, G.R. (2006) The ecology of overwintering among turtles: where turtles overwinter and its consequences Biological Reviews,81, pp. 339-367.

7. Ultsch, G.R.  (1989) Ecology and physiology of hibernation and overwintering among freshwater fishes, turtles, and snakes – Biological Reviews 64, pp.435-516

8. Bernstein, N.P., Black, R.W. (2005) Thermal environment of overwintering ornate box turtles, Terrapene ornata ornata, in Iowa. The American Midland Naturalist, 2005

9. Reese, D.A., Welsh, H.H.. (1997) Use of terrestrial habitat by western pond turtles, Clemmys marmorata: implications for management  Proceedings: Conservation, Restoration, and Management of Tortoises and Turtles. An International Conference, pp. 352-357 held 1997 by the New York Turtle and Tortoise Society

10. Lagler, K.F. (1943)   Food Habits and Economic Relations of the Turtles of Michigan with Special Reference to Fish Management. American Midland Naturalist, Vol. 29, No. 2 (Mar., 1943), pp. 257-312

11. Ernst, C.H., Zappalorti, R.T., Lovich, J.E.. Overwintering sites and thermal relations of hibernating bog turtles, Clemmys muhlenbergii Copeia, Vol. 1989, No. 3 (Aug. 8, 1989), pp. 761-764

12. Farrell, R.F., Graham, T.E.. Ecological Notes on the Turtle Clemmys insculpta in Northwestern New Jersey  Raymond F. Farrell and Terry E. Graham Journal of Herpetology, Vol. 25, No. 1 (Mar., 1991), pp. 1-9

13. Ernst, C.H. (1986) Environmental temperatures and activities in the wood turtle, Clemmys insculpta Journal of Herpetology, Vol. 20, No. 2 (Jun., 1986), pp. 222-229

14. Cagle, F.R. (1942) Turtle populations in southern Illinois– Copeia, 1942, pp. 155-162

15. Copper, J.E. (1956) A Maryland Hibernation Site for Herptiles. Herpetologica, Vol. 12, No. 3 (Sep. 1, 1956), p. 238

16. Ernst, C.H. (1972) Temperature-Activity Relationship in the Painted Turtle, Chrysemys picta . Copeia 1971, 217-222.

17. Graham, T.E., Guimond, R.W. (1995)  Aquatic Oxygen Consumption by Wintering Red-Bellied Turtles  Journal of Herpetology, Vol. 29, No. 3 (Sep., 1995), pp. 471-474

18. Doroff, A.M., Keith, L.B. (1990) Demography and Ecology of an Ornate Box Turtle (Terrapene ornata) Population in South- Central Wisconsin  Copeia, Vol. 1990, No. 2 (Jun. 27, 1990), pp. 387-399

19. Claussen, D.L., Daniel, P.M., Jiang, S., Adams, N.A.(1991) Hibernation in the Eastern Box Turtle, Terrapene c. carolina  Journal of Herpetology, Vol. 25, No. 3 (Sep., 1991), pp. 334-341

20. Costanzo, J.P., Claussen, D.L. (1990) Natural Freeze Tolerance In The Terrestrial Turtle, Terrapene Carolina The Journal Of Experimental Zoology 254:228-232 (1990)

21. Jackson, D.C., Rauer, E.M., Feldman, R.A., Reese, S.A. (2004) Avenues of extrapulmonary oxygen uptake in western painted turtles (Chrysemys picta belli) at 10 °C. Comparative Biochemistry and Physiology, Part A 139 (2004) 221– 227

22. Brown, G.P., Brooks, R.J. (1994) Characteristics of and Fidelity to Hibernacula in a Northern Population of Snapping Turtles, Chelydra serpentina Copeia, Vol. 1994, No. 1 (Feb. 1, 1994), pp. 222-226

23. Yokosuka, H.,  Murakami, T., Ishiyama, M., Yoshie,  S., Fujita, T. (2000) The Vascular Supply Of The Villiform Processes In The Pharynx Of The Soft-Shelled Turtle, Trionyx Sinensis Japonicus. A Scanning Electron Microscopy Study Of Corrosion Casts. Archives Of Histology And Cytology 63, 193-198

24. Taylor, G.M., Hanley, R.W., Bauman, T.R. (1989). Movements and hibernation sites of overwintering painted turtles in southern Ontario.  Can. J. Zool. 67:1877-1881

25. Gibbons, J.W. (1967) Variation in Growth Rates in Three Populations of the Painted Turtle, Chrysemys picta. Herpetologica, Vol. 23, No. 4 (Dec. 30, 1967), pp. 296-303

26. Variation in Growth Rates in Three Populations of the Painted Turtle, Chrysemys picta Author(s): J. Whitfield Gibbons Source: Herpetologica, Vol. 23, No. 4 (Dec. 30, 1967), pp. 296-303

27. Packargd., C., Packardm,. J., Mcdanielp,. L., And Mcdaniell,. L. (1989) Tolerance Of Hatchling Painted Turtles To Subzero Temperatures. Can. J. Zool. 67: 828 -830.

28. Storey, K.B., Storey, J.M., Brooks, S.P., Churchill, T.A., Brooks, R.J. (1988) Hatchling turtles survive freezing during winter hibernation. PNAS November 1, 1988 vol. 85 no. 21 8350-8354

29. Breitenbach, G.L., Congdon, J.D., van Loben Sels, R.C. (1984) Winter Temperatures of Chrysemys picta Nests in Michigan: Effects on Hatchling Survival.  Herpetologica, Vol. 40, No. 1 (Mar., 1984), pp. 76-81

30. Christiansen, J.L., Bickham, J.W. (1989) Possible Historic Effects of Pond Drying and Winterkill on the Behavior of Kinosternon flavescens and Chrysemys picta. Journal of Herpetology, Vol. 23, No. 1 (Mar., 1989), pp. 91-94

31. Packard, G. C., Lang, J. W., Lohmiller, L. D. and Packard, M. J. (1997). Cold tolerance in hatchling painted turtles (Chrysemys picta): supercooling or freeze tolerance? Physiol. Zool. 70, 670-678.

32. Lee, R. E. and Costanzo, J. P. (1998). Biological ice nucleation and ice distribution in cold-hardy ectothermic animals. Annu. Rev. Physiol. 60, 55-72.

33. Dinkelacker, S. A., Costanzo, J. P., Iverson, J. B. & Lee, R. E. (2004). Cold-hardiness and dehydration resistance of hatchling Blanding’s turtles (Emydoidea blandingii) : implications for overwintering in a terrestrial habitat. Canadian Journal of Zoology 82, 594–600.

34. Costanzo, J. P., Litzgus, J. D., Iverson, J. B. and Lee, R. E. (2000b). Seasonal changes in physiology and development of cold hardiness in the hatchling painted turtle, Chrysemys picta. J. Exp. Biol. 203, 3459-3470.

35. Herbert, C.V., Jackson, D.C. (1985a). Temperature effects on the responses to prolonged submergence in the turtle Chrysemys picta bellii. I. Blood acid-base and ionic changes during and following anoxic submergence. Physiological Zoology 58, 655-669.

36. Sievert, L.M., Sievert, G.A., Cupp, P.V. (1988). Metabolic rate of feeding and fasting juvenile midland painted turtles, Chrysemys picta marginata. Comparative Biochemistry and Physiology go A, I 57-1 59.

37. McPherson, R.J., Marion, K.R. (1982). Seasonal changes of total lipids in the turtle Sternotherus odoratus. Comparative Biochemistry and Physiology 71 A, 93-98.

38. Beall, R.J., Privitera, C.A. (1973). Effects of cold exposure on cardiac metabolism of the turtlc Pseudemys (Chrysemys) picta. American Journal of Physiology 224, 435-441.

39. EMERSOND, N. (1967).P reliminary study on seasonal liver lipids and glycogen, and blood sugar levels in the turtle Graptemys pseudogeographica (Gray) from South   Dakota. Herpetologica 23, 68-70.


40. Gatten, R.E. (1981) Anaerobic metabolism in freely diving painted turtles (Chrysemys picta). J. Exp. Biol. 216:377-385.

41. Ultsch, G.R., Jackson, D.C. (1982a). Long-term submergence at 3°C of the turtle, Chrysemys picta bellii, in normoxic and severely hypoxic water: I. Survival, gas exchange, and acid base status. J. Exp/ Biol. 96: 11-28


42.  Ultsch, G.R., Jackson, D.C. (1982a). Long-term submergence at 3°C of the turtle, Chrysemys picta bellii, in normoxic and severely hypoxic water: III. Effects of changes in ambient PO2 and subsequent air breathing. Ibid. 96: 87-99

43. JACKSON, D. C. (2000). Living without oxygen: lessons from the freshwater turtle. Comparative Biochemistry and Physiology A 125, 299–315.

44. Chase, J.D., Dixon, K.R., Gates, J.E., Jacobs, D., Taylor, G.J. Habitat Characteristics, Population Size, and Home Range of the Bog Turtle, Clemmys muhlenbergii, in Maryland.  Journal of Herpetology, Vol. 23, No. 4 (Dec., 1989), pp. 356-362

45. Rosenbaum, P.A., Robertson, J.M., Zamudio, K.R. (2007) Unexpectedly low genetic divergences among populations of the threatened bog turtle (Glyptemys muhlenbergii) Conservation Genetics 8:331–342 2007

46. Buhlmann, K.A., Gibbons, J.W., and Jackson, D.R. 2008. Deirochelys reticularia (Latreille 1801)- chicken turtle. In Rhodin, A.G.J., Pritchard, P.C.H., can Dijk, P.P., Saumure, R.A., Buhlmann, K.A., and Iverson, J.B. (Eds.). Conservation Biology of Freshwater Turtles and Tortoises: A complication Project of the IUCN/SSC Tortoise and Freshwater Turtle Specialist Group. Chelonian Research Monographs No. 5. pp. 014.1-014.6

Leave a Reply

Your email address will not be published. Required fields are marked *