Mg/Ca thermometry in planktic foraminifera: Improving paleotemperature estimations for G. bulloides and N. pachyderma left

• Published in: Geochemistry, Geophysics, Geosystems
• Main Authors: Riveiros, Natalia Vazquez, Govin, Aline, Waelbroeck, Claire, Mackensen, Andreas, Michel, Elisabeth, Moreira, Santiago, Bouinot, Thomas, Caillon, Nicolas, Orgun, Ayche, Brandon, Margaux
• Format: Text
• Language: English
• Published: Amer Geophysical Union
• Subjects: envir; geo; Southern Ocean; Neogloboquadrina pachyderma; North Atlantic; Sea ice
• Online Access: https://doi.org/10.1002/2015GC006234; https://archimer.ifremer.fr/doc/00324/43480/42916.pdf; https://archimer.ifremer.fr/doc/00324/43480/42917.docx

Planktic foraminiferal Mg/Ca ratios have become a fundamental seawater temperature proxy in past climate reconstructions, due to the temperature dependence of Mg uptake into foraminiferal calcite. However, empirical calibrations for single species from methodologically consistent data are still lacking. Here we present species-specific calibrations of Mg/Ca versus calcification temperature for two commonly used species of planktic foraminifera: Globigerina bulloides and Neogloboquadrina pachyderma left, based on a series of Southern Ocean and North Atlantic core tops. Combining these new data with previously published data, we derive an integrated G. bulloides Mg/Ca-temperature calibration for mid and high latitudes of both hemispheres between 2 and 18°C, where Mg/Ca = 1.006 ± 0.032 * e0.065 ± 0.003*Tiso (R2 = 0.82). G. bulloides is found to calcify deeper in the Southern Ocean (∼ 200 m) than in the North Atlantic (top 50 m). We also propose a Mg/Ca temperature calibration to describe the temperature response in N. pachyderma left that calcified away from the influence of sea ice in the Southern Ocean, valid between ∼ −1 and 9°C, of the form Mg/Ca = 0.580 ± 0.016 * e0.084 ± 0.006*Tiso (R2 = 0.70). These calibrations account for uncertainties on Mg/Ca measurements and calcification temperature that were carefully estimated and propagated using Monte Carlo iterations. The 1σ propagated error in Mg/Ca-derived temperatures is 1.1°C for G. bulloides and 0.9°C for N. pachyderma left for the presented data sets. Geographical extension of genotypes must be assessed when choosing to develop regional or global calibrations.