A simple model for the evolution of temperature-dependent sex determination explains the temperature sensitivity of embryonic mortality in imperiled reptiles

Reptiles are declining globally, and turtles in particular are among the most imperiled group of vertebrates on earth. Common turtle conservation strategies include artificial incubation and head-starting to rear healthy juvenile turtles for release into wild populations. Temperature-dependent sex determination (TSD) in turtles and other reptiles allows conservation managers to bias sex ratios towards females by using high incubation temperatures, allowing the introduction of more egg-bearing individuals into populations. However, theory and data suggest high incubation temperature may potentially degrade phenotypic quality, and therefore investigation into incubation temperature and fitness outcomes is warranted for species with TSD. We performed a meta-analysis to explore the relationship between hatchling mortality, constant temperature incubation, and pivotal temperature (TPiv, the temperature where the sex ratio is 1:1). We found models incorporating species-specific TPivs better explained hatchling mortality data than models without species-specific information, suggesting TPivs exhibit a correlated evolution with embryonic thermal tolerance. Moreover, we found embryo mortality followed a parabolic distribution centered near the pivotal temperature, and critically, depreciation of fitness was relatively rapid above vs below the pivotal temperature. These results suggest artificial incubation regimes may benefit from avoiding high-temperature incubation. We suggest that fluctuating temperature regimes are easily implemented and are representative of the environmental context in which turtles evolved over millions of years. Further, recent models allow the prediction of sex ratio under fluctuating regimes, thereby allowing managers to maintain sex ratio bias of wild releases without compromising phenotypic quality.