LA-ICP-MS Trace Element Analysis of Planktonic Foraminifera and Application to Marine Isotope Stage 31 in the Southwest Pacific Ocean

Abstract

Trace element/Ca ratios were measured by LA-ICP-MS in Gs. ruber and N. incompta from a wide range of core top and plankton tow samples in the Southwest Pacific Ocean, with particular focus on linking measured Mg/Ca ratios to observed (near-) surface ocean temperatures. Mean Mg/Ca ratios measured in the ultimate chamber F of Gs. ruber are significantly lower compared to chambers F-2 and F-1, which show no statistical difference. These observations led to the development of three new Mg/Ca – ocean temperature calibrations that can be used to reconstruct SST from LA-ICP-MS trace element analysis of Gs. ruber over a temperature range of 14.7-28.8°C. In contrast to the LA-ICP-MS study of Gs. ruber, the planktonic species N. incompta shows no systematic difference in Mg/Ca between the final four visible chambers at one core top site, nor between the F-3 and F chambers at any sites. In addition, there was no correlation between Mg/Ca ratios and ocean temperature in either reticulate or crystalline forms. This may reflect migration of this species within the water column that is not uni-directional, that this species does not dwell at the same depths at each core top site, or point towards further unidentified controls on Mg incorporation into N. incompta that require further study. The chamber specific calibrations developed for Gs. ruber in this study, and a calibration developed using the same techniques for G. bulloides [Marr et al., 2011] were applied down core to Mg/Ca ratios measured from ODP Site 1123. The SST derived from Mg/Ca ratios reveal that during the MIS-31 interglacial, SSTs were approximately 4-5°C warmer than today and 8-9°C warmer than those from MIS-29 and 30. A comparison of SSTs measured from Gs. ruber and G. bulloides, suggests that they are responding to local insolation changes. G. bulloides records colder temperatures than Gs. ruber, which reflects differences in their relative depth in the water column. Paired Mg/Ca and δ 18O data reveal significant changes in ice volume over the sampling period. Increases in SSTs recorded by the planktonic foraminifera lead the seawater stable isotope record by 10 kyr suggesting a significant influence from changes in Northern Hemisphere ice sheet volume during MIS-31. In some intervals, the SST leads the benthic stable isotope record by 8 kyr and shows deviations in benthic δ18O from synchronous planktic samples. In G. bulloides and Gs. ruber, Mn and Mg were the only trace elements to show systematic glacial-interglacial changes from MIS-34 to MIS-29. This correlation could imply that Mn/Ca ratios in the foraminifera are recording changes in ocean chemistry related to changing water mass circulation at ODP Site 1123 as past ocean temperatures changed. Size-normalised weights (SNW) of G. bulloides tests show systematic variations from MIS- 34 to MIS-29. For much of the record, SNW is anti-correlated with SST in a manner similar to the modern relationship between SNW and SST in the Southwest Pacific Ocean. However, immediately prior to the MIS-31 Southern Hemisphere insolation maxima, SNW increase with SST suggesting a fundamental change in surface ocean carbonate chemistry occurred that is unique to this time.