J.F. Gettrust

Density profile and flow of miscible fluid with dissimilar constituent masses

A computer simulation model is used to study the density profile and flow of a miscible gaseous fluid mixture consisting of differing constituent masses (mA=mB/3) through an open matrix. The density profile is found to decay with the height ∝exp(−mA(B)h), consistent with the barometric height law. The flux density shows a power-law increase ∝(pc−p)μ with μ≃2.3 at the porosity 1−p above the pore percolation threshold 1−pc.

Density profile and flow of driven gas in an open porous medium with a computer simulation

A computer simulation model is used to study the flow of gas from a source through open porous media on a simple cubic lattice with a pressure bias. This model is used to represent marine sediments through which fluid flows. Isotropic and layered porous media are considered and the gas fluid is modeled by an interacting lattice gas. In isotropic porous medium with porosity ps=0.35, the density profile is linear in the absence of bias (B=0) and oscillations develop in the presence of bias. At low porosity (ps=0.32), the density profile becomes nonlinear in the presence of bias. The response of the flow rate to bias is linear in media with high porosity – consistent with the Darcy law. In porous medium with low porosity, the response of flow rate is linear only in low bias (B⩽0.3) which crosses over to a different linear response at moderate values of bias (0.30⩽B⩽0.7) before it becomes negative at extreme bias (B⩽0.8). The presence of fault planes (i.e., linear zones of higher porosity within a low-porosity system) broadens the linear-response regime.

SELF-ORGANIZED PHASE SEGREGATION IN A DRIVEN FLOW OF DISSIMILAR PARTICLES MIXTURES

Self-organized patterns in an immiscible fluid mixture of dissimilar particles driven from a source at the bottom are examined as a function of hydrostatic pressure bias by a Monte Carlo computer simulation. As the upward pressure bias competes with sedimentation due to gravity, a multi-phase system emerges: a dissociating solid phase from the source is separated from a migrating gas phase towards the top by an interface of mixed (bi-continuous) phase. Scaling of solid-to-gas phase with the altitude is nonuniversal and depends on both the range of the height/depth and the magnitude of the pressure bias. Onset of phase separation and layering is pronounced at low bias range.

Concentration gradient, diffusion, and flow through open porous medium near percolation threshold via computer simulations

The interacting lattice gas model is used to simulate fluid flow through an open percolating porous medium with the fluid entering at the source-end and leaving from the opposite end. The shape of the steady-state concentration profile and therefore the gradient field depends on the porosity (p). The root mean square (rms) displacements of fluid and its constituents (tracers) show a drift power-law behavior, R∝t in the asymptotic regime (t→∞). The flux current density (j) is found to scale with the porosity according to, j∝(Δp)β with Δp=p−pc and β≃1.7.

Coordinated, High-Resolution Studies of Hydrated Sediments

Abstract : The Naval Research Laboratory (NRL) has been conducting coordinated investigations of marine gas hydrates based on precise co-location of both direct sediment cores and data from a deep-tow multichannel seismic system. The seismic instrument (known as the Deep Towed Acoustics/Geophysics System (DTAGS)) was developed by NRL to support detailed studies of deep-ocean marine sediments by towing both the seismic source (220Hz - 1kHz, 200 dB //1 square Pa @ 1m ) and 48 channel hydrophone array 300 m above the seafloor in up to 6km deep water. This instrument has proven to ideal for studies of marine gas hydrates. Data from this system have been used to study the impact of hydrate dissociation on sediment properties on and near the Blake Ridge and on the Cascadia Margin. Investigations in the Gulf of Mexico are schedules for Spring 2005. Using bottom mounted acoustic transponders for long baseline (LBL) navigation, we have been able to co-locate the deep-tow seismic data with sediment cores, water samples, etc. to constrain models for processes that create and dissociate marine gas hydrates in much greater detail than previously possible. We show relationships between geologic features resolved with the seismic data and geochemical evidence for variability in methane flux through the seafloor. We also present preliminary results from lattice-gas numerical simulations of gas-fluid flow through complex sediments where parameters are predicated on results obtained with DTAGS and associated geochemical samples.

A Monte Carlo simulation of thermal-driven fluid: flux rate and density profile

Effects of uniform temperature and linear temperature gradient on flow rate and density profile of fluid driven from a source to an open system is studied by a Monte Carlo simulation in three dimensions. The steady-state density profile with uniform temperature differs significantly from that of the temperature gradient at all but high-temperature regimes, where the profile is a linear density gradient; the crossover to linear density gradient of the profile is sensitive to range of temperature variation. The response of the flow rate density to the temperature is nonlinear.

GRADIENT DRIVEN FLOW: LATTICE GAS, DIFFUSION EQUATION AND MEASUREMENT SCALES

Tracer diffusion and fluid transport are studied in a model for a geomarine system in which fluid constituents move from regions of high to low concentration. An interacting lattice gas is used to model the system. Collective diffusion of fluid particles in lattice gas is consistent with the solution of the continuum diffusion equation for the concentration profile. Comparison of these results validates the applicability and provides a calibration for arbitrary (time and length) units of the lattice gas. Unlike diffusive motion in an unsteady-state regime, both fluid and tracer exhibit a drift-like transport in a steady-state regime. The transverse components of fluid and tracer displacements differ significantly. While the average tracer motion becomes nondiffusive in the long time regime, the collective motion exhibits an onset of oscillation.