As a recorder of the hydrological cycle and ocean circulation, salinity is one of the most wanted parameters in paleoceanography. Current paleosalinity reconstructions mostly rely on the interpretation of stable oxygen isotope signals combined with an independent paleotemperature proxy. Due to error propagation, this indirect approach is associated with large uncertainties. Recent culture studies and a Red Sea field study have shown that incorporation of Na in foraminiferal shell calcite depends on salinity, providing a potential direct proxy for salinity. However, application of a Na/Ca-based salinity proxy requires further calibration, which should also consider settling of foraminifera through the water column and burial in the sediment. Here we compare Na/Ca in living specimens from Red Sea surface waters with specimens collected from 0- to 500-m water depth and sedimentary specimens from core-tops. This shows that Na/Ca in Globigerinoides ruber and Trilobatus sacculifer shells decrease with increasing water depth and until the sediment surface. For both species, laser-ablation-ICP-Q-MS measurements combined with electron-probe microanalysis show that Na is enriched in the spines. Loss of spines during settling of foraminifera through the water column hence provides a mechanistic explanation for the observed Na decrease in bulk specimens with water depth. In contrast, average Mg/Ca values increase toward the seafloor in both species, coinciding with deposition of gametogenic calcite, which is enriched in Mg but has Na/Ca values similar to that in lamellar calcite. Both spine shedding and gametogenic calcite addition hence affect the average minor/trace element composition of foraminiferal calcite.
- planktonic foraminifera
- salinity proxy