Depletion of a brine layer at the base of ridge-crest hydrothermal systems

Abstract

The variable salinity of fluid venting from mid-ocean ridges is indicative of mixing between hydrothermal seawater and fluids that have undergone supercritical phase separation. In order to study the stability of a brine-saturated layer that may form in the lowermost part of the hydrothermal system, we have performed numerical simulations of a system that has returned into the subcritical regime. For typical geological parameters, it is shown that the interface between the brine layer and the overlying fluids is not very stable, but vanishes by one of two dynamical mechanisms: convective breakdown or vertical migration. This contradicts the conventional picture of a steady, layered convective system in which the brine is depleted only by dispersion and diffusion across the interface. The depletion mechanism depends on the fluid-dynamical stability of the brine layer. Convection within the brine layer results either in the convective breakdown (for low excess salinity of the brine, as compared to seawater) or the upward migration of the interface (for higher excess salinities). Consequently, the depletion times are much shorter than for models with pure dispersion/ diffusion across the interface. If the brine layer is static, high-chlorinity liquid is entrained slowly by the convecting overlying fluids, leading to downward migration of the interface. This gradual depletion of the brine layer results in almost constant vent salinities, in agreement with measured salinities of chronic high-chlorinity vents. ß

Laboratory and field-based investigations of subsurface geochemical processes in seafloor hydrothermal systems

“…5.3). As stated previously, Mn abundances in Vienna Woods fluids only suggest temperatures near 350°C and so it is possible that the high Cl endmembers do not reflect active phase separation at depth in 2006, but rather entrainment/depletion of residual brines formed during previous phase separation events (Shanks and Seyfried, 1987;Von Damm and Bischoff, 1987;Von Damm, 1988;Butterfield et al, 1997;Schoofs and Hansen, 2000;Von Damm et al, 2005). Brine entrainment would be consistent with the chronic venting of Cl-rich fluids at Vienna Woods since 1990 (Lisitsyn et al, 1993;Auzende et al, 1996;Gamo et al, 1997;Douville et al, 1999;Fourre et al, 2006) and the apparent absence of any Cl-depleted conjugate fluids.…”

A new Fe/Mn geothermometer for hydrothermal systems: Implications for high-salinity fluids at 13°N on the East Pacific Rise

“…This suggests that Grand Bonum is venting a vapor phase fluid, despite being chloride-enriched relative to seawater. Numerical modeling efforts have grown quite sophisticated, yet, for pressures and temperatures applicable to most ridge crest hydrothermal systems, they have not been able to reproduce venting of chloride-enriched fluids unless the magmatic heat source is diminished to the point where active phase separation is no longer taking place (Schoofs and Hansen, 2000;Coumou et al, 2009;Lewis and Lowell, 2009). This does not appear to be the case for EPR 13°N.…”

Fluid Phase Separation Processes in Submarine Hydrothermal Systems

“…Not surprisingly, dissolved volatile species also exhibit distinctly different partitioning in the vapor fraction relative to transition metals (e.g., Fe, Cu), indicative of the important role of phase separation in inducing great oxidation events in the subseafl oor reaction zone, while delivering volatile-enriched but metal-depleted vapor fl uids to the seafl oor. ocean ridges (Seyfried et al 2003), but also to better understand formation and consequences of thermal convection of a putative brine (liquid) layer during phase separation/segregation processes (Bischoff and Rosenbauer 1989;Lowell and Germanovich 1997;Schoofs and Hansen 2000;Kawada et al 2004;Geiger et al 2005;Fontaine and Wilcock 2006). Brine storage is an important component of the subseafl oor reaction zone, with brine layers appearing to be stably stratifi ed but unable to convect (Kawada et al 2004;Fontaine and Wilcock 2006).…”