Daily and annual patterns of thermoregulation in painted turtles (Chrysemys picta marginata) living in a thermally variable marsh in Northern Michigan

The capacity for an ectothermic reptile to thermoregulate has implications for many components of its life history. Over two years, we studied thermoregulation in a population of Midland painted turtles (Chrysemys picta marginata) in a shallow, thermally variable wetland during summer in Northern Michigan. Mean body temperature (T_b) of free-ranging turtles was greater in 2008 (25.8 °C) than in 2010 (19.7 °C). Laboratory determined thermoregulatory set point (T_set) ranged from 25 °C (T_set-min) to 31 °C (T_set-max) and was lower during the fall (17–26 °C). Deviations of T_b distributions from field measured operative temperatures (T_e) and indices of thermoregulation indicated that C. picta marginata were capable of a limited degree of thermoregulation. Operative temperatures and thermal quality (d_e=|T_set-min?T_e| and |T_e?T_set-max|) cycled daily with maximal thermal quality occurring during late morning and late afternoon. The accuracy of thermoregulation (d_b=|T_set-min?T_b| and |T_b?T_set-max|) was maximal (d_b values were minimal) as T_b declined and traversed T_set during the late afternoon–early evening hours and was higher on cloudy days than on sunny days because relatively low T_e values decreased the number of T_b values that were above T_set. Our index of thermal exploitation (E_x=frequency of T_b observations within T_set) was 36%, slightly lower than that reported for an Ontario population of C. picta marginata. Regression of d_b (thermal accuracy) on d_e (thermal quality) indicated that turtles invested more in thermoregulation when thermal quality was low and when water levels were high than when they were low. There were no intersexual differences in mean T_b throughout the year but females had relatively high laboratory determined T_b values in the fall, perhaps reflecting the importance of maintaining ovarian development prior to winter.