EVALUATION OF RACEMIZATION-DERIVED AGE ESTIMATES USING STRONTIUM 87/86 VALUES IN MOLLUSKS FROM EARLY AND MIDDLE PLEISTOCENE UNITS, NORTH CAROLINA COASTAL PLAIN

Amino acid racemization (AAR) is a useful tool for relative age estimation and correlation of Quaternary units in the Atlantic coastal plain. Numerical age estimates based on AAR are dependent on factors such as genus, temperature history, and kinetic modeling. The effect of these variables can be reduced by use of a single taxon for analysis and by using samples from limited geographic regions where effective temperatures can be considered identical for all analyzed samples. Two models for racemization kinetics, usually identified as “parabolic” and “non-linear” models, are tested for the amino acid leucine (frequently used in aminostratigraphic studies) by paired AAR and Sr 87/86 analysis of Mercenaria samples from early and middle Pleistocene strata. All samples for this study are from a series of Rotasonic cores obtained by the NC Geological Survey between 2002 and 2006 at NC barrier island and mainland sites of the lower coastal plain. Stratigraphic control is provided by superposition, seismic stratigraphic correlation, biostratigraphy and paleoenvironmental analysis (e.g., Culver et al., 2008; Mallinson et al., 2009). Four major clusters of D/L values (aminozones), identified as AZ3+, AZ3.5, AZ4, and AZ4+, are found in the older part of the stratigraphic section. All are statistically distinct and found in superposition (Wehmiller et al., 2009). Mean Sr 87/86 age estimates (MacArthur et al., 2001) for shells from these four aminozones are 0.3, 0.6, 1.2, and 1.5 Ma, respectively. These four aminozones have non-linear model age estimates of 0.36, 0.65, 1.2 and 1.53 Ma, respectively. The parabolic model ages are much younger: 0.20, 0.29, 0.40, and 0.46 Ma, respectively. Although both AAR and Sr- age estimates have large uncertainties (ranging up to 50%), the paired AAR-Sr 87/86 analysis of individual shells leads to the important conclusion that the parabolic model does not appear to be applicable to the interpretation of leucine D/L values in Quaternary Mercenaria specimens from the coastal plain. It is likely that different kinetic models are required for the derivation of ages based on other amino acids commonly found in Quaternary mollusks.