The anhydrite saturation index of the ponded brines and sediment pore waters of the Red Sea deeps

Abstract

We compiled the available chemical analyses of the ponded brines and of the sediment interstitial waters for the Atlantis II, Discovery, Suakin and Valdivia deeps in the Red Sea central valley, along with the DSDP data for interstitial waters of several cores in the rift axis (Sites 225–228 of Leg 23). Recent advances in the calculation of the thermodynamic properties of concentrated solutions at high temperatures and pressures allow the study of the equilibrium conditions between anhydrite and halite, and the brines. We first tested the influence of various parameters as temperature, pressure and solution composition on the calculated anhydrite saturation index in order to provide a clear criterion for equilibrium, which is assumed when the saturation index lies within 0.9 and 1.1. We found that at 62°C and 200 bar, the pressure effect on the saturation indices is of the same magnitude as that of temperature. Our calculations point to the overall undersaturation of the free brines with respect to halite. The calculated halite saturation state of the pore waters at the four DSDP sites is consistent with the mineralogical description of the sediments in which halite is not reported, except at Site 225 where the calculated undersaturation is in contradiction with the reported presence of halite at the base of the core. From the hypothesis of equilibrium between anhydrite and the pore waters in the DSDP cores, we have estimated values of the geothermal gradients off the rift axis which fall within the highest values obtained by extrapolation of direct temperature measurements. This suggests that, at the latitude of the Atlantis II and Suakin deeps, the high heat fluxes measured along the axial trough extend off the rift axis. Our calculations show that the Upper Convective Layer (UCL) of the Atlantis II hydrothermal system has always been undersaturated over the studied period (1965–1985), which is consistent with the absence of anhydrite in sediment underlying this brine. The calcium and sulfate contents of the Lower Convective Layer (LCL) show a parallel evolution: they decrease when temperature increases. Although this behaviour suggests control of brine chemistry by equilibrium with anhydrite, our calculations show that anhydrite has reached saturation in the lower brine only in 1966 and 1976. The general regime leading to the loss of calcium and sulfate outside these periods is tentatively attributed to diffusion. Calculations show that, throughout the whole stratified brine column (from the sediment to normal Red Sea water), there is no chemical potential gradient for calcium sulfate whereas the chemical potential of sodium chloride regularly decreases. We tentatively advocate the coupling of calcium sulfate diffusion with the large sodium chloride gradient to account for the loss in Ca and SO 4. Finally, the pore water chemistry of the Atlantis II and Discovery Deep sediments record two parallel evolutions from anhydrite undersaturation in 1966 to saturation (or a state close to) in 1976. These data are consistent with a periodical brine overspill from Atlantis II into Discovery.