Stress‐dependent permeability and the formation of seafloor event plumes

Abstract

Abstract. We address the question of formation of event plumes following dike emplacement in a hydrothermal upflow zone at a mid-ocean ridge. We assume a preexisting low-to moderate-temperature single-pass hydrothermal system and suggest that dike emplacement provides a damaged zone of high permeability along its margins as well as the heat required to drive the event plume. We also consider the role of thermoelastic stresses in limiting the heat output of the event plume. Our calculations show that event plumes can result from dike emplacement into a preexisting, moderate-temperature hydrothermal system, provided the local permeability generated by the dike is emission, the circulation decays rapidly to its original temperature; however, the heat output from the chronic plume is greater because of the increased permeability resulting from dike emplacement. The decay of heat output to preevent plume levels requires that the newly created permeability be sealed, perhaps as a result of chemical precipitation in the cracks.

The response of two-phase hydrothermal systems to changing magmatic heat input at mid-ocean ridges

“…The fixed temperature condition assumes that magma serves as an infinite reservoir of heat and heat transport is controlled by the vigor of the hydrothermal system, which is characterized by the Rayleigh number [e.g., Bejan, 1995;Lowell and Germanovich, 2004]. On the other hand, the thermal heat output of a constant heat flux condition is controlled by the rate at which the heat is conducted from magma body [e.g., Lowell andGermanovich, 1994, 2004;Germanovich et al, 2000Germanovich et al, , 2001. Neither of these boundary conditions is realistic because heat transfer from a convecting magma body near its liquidus will cause magma to cool and crystallize [e.g., Cann and Srens, 1982;Lowell and Rona, 1985].…”

Episodic thermal perturbations associated with groundwater flow: An example from Kilauea Volcano, Hawaii

“…Because of the large horizontal permeability of unaltered Hawaiian basalt flows [Souza and Voss, 1987], part of the ascending fluid plume is channeled laterally at shallow depths where rocks are as yet unaltered and relatively permeable. This lateral flow causes thermoelastic expansion [Germanovich et al, 2000], precipitation of secondary minerals [Morrow et al, 1981;Moore et al, 1983;Vaughan et al, 1986;Martin and Lowell, 2000], and alteration of the host basalt, decreasing permeability and causing a relatively rapid transition from advection-dominated to conduction-dominated heat transport. Such episodic hydrothermal activity is consistent with recent observations from mid-ocean-ridge hydrothermal systems [Baker et al, 1987;Johnson et al, 2000], where temperature transients were observed over timescales of weeks.…”

“…Further, temperature data obtained from fluid inclusions in secondary minerals often indicate that past fluid temperatures in the rock were higher [Keith and Bargar, 1988;Bargar and Fournier, 1988;Bargar et al, 1995;Gonzalez-Partida et al, 1997], whereas the standard interpretations imply temperatures increasing toward a maximum at steady state. Laboratory experiments [Summers et al, 1978;Morrow et al, 1981;Moore et al, 1983;Vaughan et al, 1986;Moore et al, 1994] and thermodynamic, kinetic, and thermoelastic considerations imply that in many cases hydrothermal fluid flow under a horizontal temperature gradient results in rapid mineral precipitation and rock contraction, decreasing permeability with time [Fournier, 1987;Germanovich and Lowell, 1992;Nur and Walder, 1992;Bolton et al, 1996;Martin and Lowell, 1997;Germanovich et al, 2000;Martin and Lowell, 2000;Huertas et al, 2000]. For example, during an experiment in which heated water was forced through a cylindrical sample of granite down a temperature gradient (300°-92°C), the measured permeability dropped by a factor of $25 in just 2 weeks [Moore et al, 1983].…”

A preliminary 1‐D model investigation of tidal variations of temperature and chlorinity at the Grotto mound, Endeavour Segment, Juan de Fuca Ridge

“…In practice, we estimate dynamic viscosity as a function of fluid temperature as l5C 1 =ðC 2 1T f Þ, where T f 5370 C, C 1 50:032 Pa s= C, and C 2 515:4 C [Germanovich et al, 2000]. We calculate R and c using the formulas given in Jupp and Schultz [2004b] and the typical values of layer 2A/2B crustal properties given in Crone and Wilcock [2005].…”