HR: 0800h
AN: T31B-0499    [Abstracts]
TI: Numerical Modeling of Convection in a Porous Layer with Temperature-Dependent Permeability: Application to Seafloor Hydrothermal Systems
AU: * Yang, Y
EM: yang.yang@eas.gatech.edu
AF: Georgia Institute of Technology, School of Earth and Atmospheric Sciences, Atlanta, GA 30341 United States
AU: Lowell, R P
EM: bob.lowell@eas.gatech.edu
AF: Georgia Institute of Technology, School of Earth and Atmospheric Sciences, Atlanta, GA 30341 United States
AB: Tectonic and magmatic events in the oceanic crust can alter the permeability distribution and generate regions of new crack-related permeability. Hydrothermal fluids moving into newly created cracks encounter rocks that are initially at different temperature than the fluid. The thermoelastic stresses that result generate a temperature dependent permeability that exerts feedback on the circulation pathways. Using a 1-D model, Germanovich et al. (2001)¹ showed that thermoelastic stresses could result in bifurcated flow systems in which steady state discharge temperatures could vary dramatically in response to small parameter changes. Here we perform a more detailed analysis of the role of thermoelastic stresses on convection in a porous medium using the finite difference code GTH. The code is modified to incorporate temperature dependent permeability. We considered a 2-D rectangular system one kilometer deep and five kilometers wide, and considered both heterogeneous and homogeneous scenarios. In the heterogeneous simulation, we constructed a 1 km-wide recharge zone with initial permeability k₀ = 10^-13m² and 100 m-wide cross-flow and discharge zones with k₀ = 10^-12m². In the interior part of the model, we assigned k₀ = 10^-16m². We also simulated the effect of a layer of high porosity and permeability pillows. The side boundaries were impermeable and insulated; the lower boundary was also impermeable and either at constant temperature or constant heat flux, with values appropriate for black smoker flows. The top boundary was open with seawater entering at a constant temperature and exiting at the upwelling plume temperature. In the homogenous simulation, we assigned the initial permeability k₀ = 10^-12m² throughout the whole domain, and the boundary conditions are the same as the heterogeneous matter. To describe the thermoelastically controlled, temperature dependent permeability we used the form k(T) = k₀ [1-γ(T -T_i)]³H[1-γ(T - T_i)] + k_res., where H is the Heavyside step function. The first term in the expression expresses the main temperature dependent permeability; the latter term represents a residual permeability that remains after the primary permeability is closed. For the basic heterogeneous model, the results tended to validate the initial results of Germanovich et al. (2001). For the homogenous models, the thermoelastic effects tended to affect plume spacing and shape, the magnitude of the effect depending upon whether or not the primary permeability was closed. The affect of high porosity pillows in the heterogeneous system was promote mixing between the deep upwelling fluid with shallow circulation. Such a model could explain Galapagos-type hydrothermal systems in which the seafloor venting occurs at relatively low temperature despite the geochemical evidence of a high-temperature end-member fluid. ¹Germanovich et al., J. Geophys. Res., 106, 473-496, 2001.
DE: 3017 Hydrothermal systems (0450, 1034, 3616, 4832, 8135, 8424)
DE: 3035 Midocean ridge processes
DE: 8135 Hydrothermal systems (0450, 1034, 3017, 3616, 4832, 8424)
SC: Tectonophysics [T]
MN: Fall Meeting 2005