Year:
2020
Authors:
Williams, A.*, D. Liang, and C. Rowe
Source:
Ocean Sciences Meeting, San Diego, CA
Abstract:
The diamondback terrapin (
Malaclemys terrapin), the official state reptile of Maryland, is an estuarine turtle found along the eastern and Gulf coasts of the United States. The diamondback terrapin may experience risks due to increasing global temperatures because they display temperature-dependent sex determination (TSD). Terrapins, through TSD, produce more female offspring than male offspring if their nest temperatures are relatively warm, leading to ecological issues should global temperatures get too high. This study used temperatures from several individual field, monitored nests, along with air temperatures at the nesting site, recorded in the summer of 2018, to select a predictive model with the R software package
rstanarm. The model was then used to predict the average nest temperature in the same nesting site using air temperatures that were recorded during the summer of 2019. This prediction was then compared to temperatures recorded from five nests within the nesting site in 2019 to evaluate its predictive capabilities. A regression spline mixed model (RSMM) was selected for having the lowest leave-one-out information criterion at 50905.95 LOOIC, with a standard error of 0.64 LOOIC. The prediction of average nest temperature developed by the RSMM was given a predictive interval of 95%. We found that the actual temperature of the average nest in 2019 had a prediction coverage of 92.97% from the prediction model, with a root squared mean error of 1.89
oC and a predictive mean absolute error of 2.10
oC. Benefits of the study involve the ability to predict the future diamondback terrapin nest temperatures within a specific nesting site so long as future air temperatures within that site are known.
Mentors:
Dong Liang, Ph.D.
Christopher Rowe, Ph.D.
Students:
Alan Williams, University of Arizona