The use of amino acid racemization (AAR) for estimating ages of Quaternary fossils usually requires a combination of kinetic and effective temperature modeling or independent age calibration of analyzed samples. Because of limited availability of calibration samples, age estimates are often based on model extrapolations from single calibration points over wide ranges of D/L values. Here we present paired AAR and ⁸⁷Sr/⁸⁶Sr results for Pleistocene mollusks from the North Carolina Coastal Plain, USA. ⁸⁷Sr/⁸⁶Sr age estimates, derived from the lookup table of McArthur et al. [McArthur, J.M., Howarth, R.J., Bailey, T.R., 2001. Strontium isotopic stratigraphy: LOWESS version 3: best fit to the marine Sr-isotopic curve for 0–509 Ma and accompanying Look-up table for deriving numerical age. Journal of Geology 109, 155–169], provide independent age calibration over the full range of amino acid D/L values, thereby allowing comparisons of alternative kinetic models for seven amino acids. The often-used parabolic kinetic model is found to be insufficient to explain the pattern of racemization, although the kinetic pathways for valine racemization and isoleucine epimerization can be closely approximated with this function. Logarithmic and power law regressions more accurately represent the racemization pathways for all amino acids. The reliability of a non-linear model for leucine racemization, developed and refined over the past 20 years, is confirmed by the ⁸⁷Sr/⁸⁶Sr age results. This age model indicates that the subsurface record (up to 80 m thick) of the North Carolina Coastal Plain spans the entire Quaternary, back to ∼2.5 Ma. The calibrated kinetics derived from this age model yield an estimate of the effective temperature for the study region of 11 ± 2 °C., from which we estimate full glacial (Last Glacial Maximum – LGM) temperatures for the region on the order of 7–10 °C cooler than present. These temperatures compare favorably with independent paleoclimate information for the region.