Abstract

Fluids in the pores of the shallow crust of Mars are commonly two-phase mixtures: a H2O-rich CO2 phase usually occupying the upper (shallower) parts of the crust and a CO2-rich H2O phase occupying the lower parts of the upper crust. By analogy to terrestrial aquifers, these H2O- and CO2-filled formations are referred to as Martian CO2-fers or H2O-fers (or in general fluidfers). Here we present numerical experiments of heat and fluid flow in several large-scale fluidfer systems relevant to the Martian shallow crust. Given the lack of data on the Martian subsurface, the model systems and simulation results are presented in the spirit of conceptual models answering questions about what is possible rather than about precisely what is happening in any given Martian system. The simulations are carried out using the TOUGH2 and TOUGH3 numerical codes, developed at LBNL, with the Aquaveo-GMS graphical user interface to model the groundfluid flow. We use the equation of state module ECO2N, which models phase equilibria between the components CO2-H2O-NaCl from 3 °C to 240 °C and pressures from 0.1 MPa to 60 MPa. These simulators were developed for terrestrial systems and we chose for experimentation here specific Martian fluidfer systems where P–T conditions conform to the limits of the simulators. For simplicity in demonstrating Martian fluid-flow phenomena, we use homogeneous porosity and permeability but we include the effects of topography as a key control on large-scale thermo-fluid flow. We also include the coupled thermo-fluid flow effects produced by an igneous intrusion at depth. Because the flows are two-phase porous media flows, we model the effects of capillary pressure, relative permeability, and the Klinkenberg effect. As boundary conditions we use lithostatic pressure at the base of the Martian cryosphere all along the top boundary of the models, while the remaining boundaries are impermeable (closed), apart from above the intrusion where we set a constant temperature of 240 °C and a pressure 1 bar above ambient. In total, we present 11 different simulations for various broad classes of scenarios. For each scenario, we show the time evolution of pressure, temperature, gas saturation and gas density. We find that, (a) as expected, the time scale for convective cooling is much faster than conductive cooling. Therefore, we speculate that the current upper Martian crust has water-freezing temperatures down to the depth of pore closure (10–15 km), unless magmatic intrusions are occurring in the upper crust. (b) The CO2-fers tend to overlie the H2O-fers apart from the deep upper-crustal layers where the CO2 can reach densities comparable to those of the water. Because of this, CO2 eruptions at the surface are possible as both gaseous or liquid (unstable) phases. H2O may also have erupted at the surface if the fluidfers were water-saturated, particularly at the base of relevant topographic relief (e.g., base of a mesa) and above deep intrusions that would have triggered strong hydrothermal convection. This phenomenon, though, would only last for times on the order of 100 ka. (c) In addition to water, also high-density CO2 could perhaps mobilize and precipitate metals potentially leading to ore deposits and carbonate rocks.