NOAA/WDS Paleoclimatology - Gulf of Maine bulk and amino acid nitrogen isotope data from 1783-2010 CE

Developing high resolution, well-dated marine proxies of environmental, climatic, and oceanographic conditions is critical in order to advance our understanding of the ocean's role in the global climate system. While some work has investigated bulk and compound specific stable nitrogen isotopes (d15N) in bivalve shells as proxies for environmental variability, the small concentrations of nitrogen found in the organic matrix of the shell calcium carbonate (CaCO3) makes developing high resolution records challenging. This study investigates the potential of using the bulk and amino acid d15N of bivalve periostracum, the protein layer on the outside of the shell, as a proxy archive of nitrogen cycling processes and water source variability. Bulk d15N values were measured on the periostracum, aragonitic CaCO3, and adductor muscle of Arctica islandica shells collected in the Gulf of Maine. Increased variability of isotopic values across growth lines compared to along growth lines support mechanistic reasoning based on growth processes that periostracum is recording changes in d15N over the course of the clam's lifetime (up to 500 years). In addition, the statistically significant relationship between periostracum d15N and contemporaneous carbonate d15N of the same shell (r= 0.82, p<.0001, n=40) suggests that periostracum preserves a similar d15N signal to that preserved in the carbonate. This finding, coupled with the fact that source amino acid d15N values of periostracum are similar to that of the adductor muscle and the particulate organic matter (POM) consumed by the clam, suggests that periostracum bulk d15N reflect the d15N of the clam's food source. The isotopic offsets between periostracum, carbonate, and adductor muscle d15N values are primarily caused by differences in amino acid composition of the different tissue types, as evidenced by isotope mass balance calculations, although may also be related to differences in d15N values of the individual amino acids of the different tissue types, especially the trophic amino acids. Compound specific d15N analyses of the periostracum of A. islandica shells were used to determine that the calculated trophic position of the clams in this study (1.4 +/- 0.4) did not change significantly between 1783 and 1997. Phenylalanine d15N values over the last 70 years show similar trends to that of the bulk record, suggesting that changes in bulk d15N of that time period are related to changes in baseline d15N. Periostracum d15N values from shells collected in the western Gulf of Maine have decreased by ~1‰ since the mid-1920s. This trend (-0.008‰/year) is not statistically different from the trend of previously published d15N values of deep-sea corals from the entrance to the Gulf of Maine over the same time period. This coral record has been shown to indicate a shift in water mass source to the region and therefore the similarity between the two records suggest that changes in periostracum d15N values are reflecting broader North Atlantic hydrographic changes. Our study introduces a new, high-resolution and absolutely dated paleoceanographic proxy of baseline d15N, presenting the opportunity for future reconstructions of aspects of nitrogen cycling and water source changes in the global oceans.