A global evaluation of temperature and carbonate ion control on Mg/Ca ratios of ostracoda genus Krithe.

• Published in: Geochemistry, Geophysics, Geosystems
• Main Authors: Elmore, A.C., Sosdian, S., Rosenthal, Y., Wright, J.D.
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: American Geophysical Union 2012
• Subjects: New Zealand; Norwegian Sea; North Atlantic
• Online Access: http://dro.dur.ac.uk/19985/; http://dro.dur.ac.uk/19985/1/19985.pdf; https://doi.org/10.1029/2012GC004073

[1] Improving estimates of past ocean temperatures is paramount to our understanding of ocean circulation and its role in climate change. Magnesium/calcium (Mg/Ca) ratios of carapaces of the benthic ostracod genus Krithe were determined from new, globally distributed core top samples from the Norwegian Sea, Cape Hatteras shelf, Gulf of Mexico, Sulawesi Margin (Indonesia), New Zealand shelf, Ceara Rise, and the North Atlantic. A linear regression of the Krithe Mg/Ca ratios and bottom water temperature (BWT) reveals a significant correlation for locations where temperature during carapace calcification was above ∼3°C, which can be described by the equation Mg/Ca = (0.972 ± 0.152) * BWT + (7.948 ± 1.103) consistent with previous North Atlantic calibrations. Deviations from the global calibration line below ∼3°C follow the same pattern observed for benthic foraminifera, suggesting that the incorporation of magnesium into ostracodal calcite may be secondarily controlled by changes in carbonate ion concentration. Therefore, we propose a linear regression that describes the relationship between magnesium incorporation, temperature, and carbonate saturation for low temperatures (<3°C); Mg/Ca = (0.972 ± 0.152) * BWT + (0.100 ± 0.030) * Δ[CO32−]) + (4.440 ± 1.103) (1 SE = ± 0.3°C). While the standard error of the calibration is small, it requires an accurate knowledge of past Δ[CO32−] concentration, which necessitates additional proxy data. Applying the calibration to glacial samples from the deep North Atlantic Ocean we show that estimates of bottom water temperatures generated from the new Δ[CO32−]- corrected equations are more consistent with results from oxygen isotopes and pore water studies.