Validation of the use of doubly labeled water for estimating metabolic rate in the green turtle (Chelonia mydas L.): a word of caution

Marine turtles often have extremely high water turnover accompanied by a low field metabolic rate (FMR), a combination that can contraindicate the use of doubly labelled water (DLW). Therefore, we conducted a validation study to assess the suitability of the DLW technique for determining FMR of marine turtles. Six green turtles (22.42 +/- 3.13 kg) were injected with DLW and placed in a tank of seawater with a respirometer for continuous monitoring of oxygen consumption (MR) over a 5-day period. Trials were conducted for turtles in both fed and fasted states. Respiratory exchange ratio (RER) was determined in a dry respirometer and used to calculate energy expenditure. For fed and fasted turtles, total body water (TBW) was 66.67 +/- 3.37% and 58.70 +/- 7.63% of body mass, and water flux rates were 9.57 +/- 1.33% and 6.14 +/- 0.65% TBW day(-1), respectively. Water turnover in fasted turtles was 36% lower than that of fed turtles but MR (from oxygen consumption) of fasted turtles (13.77 +/- 1.49 kJ kg(-1) day(-1)) was 52% lower than in fed turtles (28.66 +/- 5.31 kJ kg(-1) day(-1)). Deuterium to oxygen-18 turnover rate (k(d):k(o)) ratios averaged 0.91 +/- 0.02 for fed turtles and 1.07 +/- 0.16 for fasted turtles. Fed turtles had a mean group difference of 8% and a mean individual difference of 53% between DLW and respirometry. The DLW method gave negative MR values in fasted turtles and could not be compared with respirometry data. Researchers should use caution when applying the DLW method in marine reptiles, especially when high water flux causes >90% of the labeled oxygen turnover to be due to water exchange.