Calibration of Mg/Ca thermometry in planktonic foraminifera from a sediment trap time series

[1] Paired Mg/Ca and delta(18)O measurements on planktonic foraminiferal species ( G. ruber white, G. ruber pink, G. sacculifer, G. conglobatus, G. aequilateralis, O. universa, N. dutertrei, P. obliquiloculata, G. inflata, G. truncatulinoides, G. hirsuta, and G. crassaformis) from a 6-year sediment trap time series in the Sargasso Sea were used to define the sensitivity of foraminiferal Mg/Ca to calcification temperature. Habitat depths and calcification temperatures were estimated from comparison of delta(18)O of foraminifera with equilibrium calcite, based on historical temperature and salinity data. When considered together, Mg/Ca ( mmol/ mol) of all species, except two, show a significant ( r = 0.93) relationship with temperature ( T degreesC) of the form Mg/Ca = 0.38 (+/- 0.02) exp 0.090 ( +/- 0.003) T, equivalent to a 9.0 +/- 0.3% change in Mg/Ca for a 1 degreesC change in temperature. Small differences exist in calibrations between species and between different size fractions of the same species. O. universa and G. aequilateralis have higher Mg/Ca than other species, and in general, data can be best described with the same temperature sensitivity for all species and pre-exponential constants in the sequence O. universa > G. aequilateralis approximate to G. bulloides > G. ruber approximate to G. sacculifer approximate to other species. This approach gives an accuracy of +/- 1.2 degreesC in the estimation of calcification temperature. The similar to9% sensitivity to temperature is similar to published studies from culture and core top calibrations, but differences exist from some literature values of pre-exponential constants. Different cleaning methodologies and artefacts of core top dissolution are probably implicated, and perhaps environmental factors yet understood. Planktonic foraminiferal Mg/Ca temperature estimates can be used for reconstructing surface temperatures and mixed and thermocline temperatures ( using G. ruber pink, G. ruber white, G. sacculifer, N. dutertrei, P. obliquiloculata, etc.). The existence of a single Mg thermometry equation is valuable for extinct species, although use of species-specific equations will, where statistically significant, provide more accurate evaluation of Mg/Ca paleotemperature.