Mg/Ca in foraminifera from plankton tows: evaluation of proxy controls and comparison with core tops

Calibrations and validations of the Mg/Ca paleothermometer in planktic foraminifera have traditionally been performed by means of core tops, sediment trap samples and culture experiments. In this study, Mg/Ca ratios have been measured in 8 species of planktic foraminifera (non-globorotaliids Globigerina bulloides, Neogloboquadrina incompta, Orbulina universa, Globigerinoides ruber (white) and G. sacculifer, and globorotaliids Globorotalia inflata, G. hirsuta and G. truncatulinoides), collected live from the North Atlantic, the Southeast Atlantic, the Northeast Pacific and the Norwegian Sea. Mg/Ca ratios for N. incompta, O. universa, G. ruber, G. sacculifer and G. truncatulinoides are similar to available North Atlantic core-top studies and consistent with previous calibration equations. In contrast, some G. bulloides, G. inflata and G. hirsuta Mg/Ca ratios are higher than predicted based on ?18O values, and exhibit considerable scatter. This elevation may be in part related to the impact of potential isotopic disequilibrium effects on ?18O-derived temperatures, which the Mg/Ca ratios are compared to. Another factor that may affect Mg/Ca ratios in some plankton samples is the lack of low-Mg test components (e.g., final chambers or gametogenic calcite), because of the incompleteness of the life cycle at the time of collection. N. incompta Mg/Ca ratios are correlated with salinity, with Mg/Ca changing about 16% per salinity unit, suggesting that salinity may have an important influence on Mg/Ca of some species even in non-extreme salinity environments. This is the first extensive multispecific plankton tow Mg/Ca data set from different oceanographic regions, which has been used to test the Mg/Ca temperature proxy in the context of published calibration data, highlighting the complex physiological/ecological controls on the acquisition of the proxy signal