Persistent contamination in benthic-foraminifera-based Mg ∕ Ca thermometry using standard cleaning methods

Abstract. Mg/Ca ratios measured in benthic foraminifera have been explored as a potential palaeothermometry proxy for bottom water temperatures (BWT). Mg/Ca-BWT calibrations from the Indian Ocean are rare and comprise conflicting results. Inconsistencies between studies suggest that calibrations may need to be region specific. The aim of this study was to develop Mg/Ca-BWT calibrations based on species-specific benthic foraminifera (Uvigerina peregrina, Cibicidoides wuellerstorfi, and Cibicidoides mundulus) in the tropical western Indian Ocean and to optimize the chemical cleaning procedure by Barker et al. (2003) applied to samples analysed in this study. The majority of samples of C. mundulus and U. peregrina, however, remained contaminated, rendering those data unusable for Mg/Ca core-top calibrations. Only Mg/Ca ratios in C. wuellerstorfi allowed a tentative Mg/Ca-BWT calibration with the relationship being: Mg/Ca=0.19±0.02⋅BWT+1.07±0.03, r2=0.87 and n=4). While this result differs to some degree from previous studies, it principally suggests that existing core-top calibrations from the wider Indian Ocean can be applied to core-tops in the western Indian Ocean. The agreement of Mg/Ca ratios at lower temperatures in C. wuellerstorfi, C. mundulus, and U. peregrina with Mg/Ca ratios reported for these species at low temperatures in other studies supports this conclusion. The clear difference in contamination between Cibicidoides spp. and U. peregrina, despite using the same cleaning procedure, supports the findings of previous studies that suggest different rigour might be required for different species. Many other uncertainties surrounding the Mg/Ca proxy exist and more calibration studies are required to improve this method.