Open AccessA B ayesian, multivariate calibration for G lobigerinoides ruber M g/ C a
Author(s) -
Khider D.,
Huerta G.,
Jackson C.,
Stott L. D.,
EmileGeay J.
Publication year - 2015
Publication title -
geochemistry, geophysics, geosystems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.928
H-Index - 136
ISSN - 1525-2027
DOI - 10.1002/2015gc005844
Subject(s) - salinity , calcite , dissolution , proxy (statistics) , temperature salinity diagrams , geology , oceanography , mineralogy , chemistry , machine learning , computer science
The use of Mg/Ca in marine carbonates as a paleothermometer has been challenged by observations that implicate salinity as a contributing influence on Mg incorporation into biotic calcite and that dissolution at the sea‐floor alters the original Mg/Ca. Yet, these factors have not yet been incorporated into a single calibration model. We introduce a new Bayesian calibration for Globigerinoides ruber Mg/Ca based on 186 globally distributed core top samples, which explicitly takes into account the effect of temperature, salinity, and dissolution on this proxy. Our reported temperature, salinity, and dissolution (here expressed as deep‐water Δ CO 3 2 −) sensitivities are (±2σ) 8.7±0.9%/°C, 3.9±1.2%/psu, and 3.3±1.3%/μmol.kg −1 below a critical threshold of 21 μmol/kg in good agreement with previous culturing and core‐top studies. We then perform a sensitivity experiment on a published record from the western tropical Pacific to investigate the bias introduced by these secondary influences on the interpretation of past temperature variability. This experiment highlights the potential for misinterpretations of past oceanographic changes when the secondary influences of salinity and dissolution are not accounted for. Multiproxy approaches could potentially help deconvolve the contributing influences but this awaits better characterization of the spatio‐temporal relationship between salinity and δ 18 O sw over millennial and orbital timescales.