Wenyue Xu

Effect of seafloor temperature and pressure variations on methane flux from a gas hydrate layer: Comparison between current and late Paleocene climate conditions

We investigate the response of a methane hydrate layer in marine sediments to cyclic seafloor perturbations of temperature and pressure in order to determine the change in seafloor methane flux resulting from gas hydrate dissociation or accumulation. By using a one-dimensional model describing mass, energy, and methane transport through porous sediments we show that seafloor pressure changes have negligible effect on methane transport to the seafloor. The effect of seafloor temperature perturbations is more pronounced than that of pressure. With an initial seafloor temperature of 3°C, which corresponds to current conditions on Earth, a +4°C seafloor temperature perturbation occurring over 104 years does not significantly effect methane transport. Thus such a perturbation is not likely to have a significant impact on the current global climate or to give rise to an event such as the δ13C excursion during the late Paleocene thermal maximum (LPTM). If the initial seafloor temperature is assumed to be 11°C, which corresponds to the conditions of the late Paleocene, a +4°C temperature perturbation over a period of 104 years could result in complete dissociation of methane hydrate layers situated at water depths around 1200 m. In this case, the calculations show that the change in methane flux might be able to explain the δ13C excursion of marine carbonate fossils during the LPTM. This result is weakened because the simplifications in the model tend to yield overestimates of the change of methane flux. The principal point is that strong coupling between methane transport and seafloor temperature occurs because significant hydrate accumulation and dissociation take place near the seafloor only when the seafloor temperature is relatively high. This was the case during the late Paleocene, but it is not the case at present.

Sub-critical two-phase seawater convection near a dike

Abstract We address the circulation of hydrothermal seawater near an igneous dike emplaced in the oceanic crust. By using the two-phase finite difference hydrothermal code GTHM, we are able to treat the sub-critical two-phase flow that occurs just after emplacement as well as the later single-phase circulation, which occurs as the dike cools. We investigate the effects of bulk rock permeability and dike width. The simulations show that for a 2 m wide dike emplaced in country rock with uniform permeability of 10−9 m2, two-phase flow may occur briefly adjacent to the dike margin in a region that is less than 0.1 m across. The width of the two-phase region and the duration of two-phase flow vary inversely with permeability, but they increase as the dike width increases. During two-phase flow, the advective heat flux at the seafloor fluctuates about its mean and the temperature near the seafloor remains nearly constant. The mean heat flux increases with permeability, but is independent of dike width. The model is used in conjunction with chlorinity data from ‘A’ vent near 9°N on the East Pacific Rise to indicate that the permeability there is ∼10−12 m2. The model suggests, however, that an additional heat source is required to account for the high-temperature vent fluids at 9°N that have persisted for more than 3 years. The calculations also show that heat transport near a high permeability dike (i.e., ≥10−9 m2) is consistent with the heat transport measured for the CoAxial event plumes. The calculations further suggest, however, that for a dike to generate an event plume, the zone of high permeability should be concentrated near the dike margin.

Oscillatory instability of one-dimensional two-phase hydrothermal flow in heterogeneous porous media

We derive a general analytical solution for one-dimensional steady state two-phase hydrothermal flow in porous media. The solution indicates that saturation jumps associated with discontinuities in material properties of porous media, such as permeability and thermal conductivity, and phase change boundaries are common in these systems. Using linear stability analysis, we then show that the saturation jump associated with discontinuities in permeability or thermal conductivity of porous media can be unstable under infinitesimal perturbation. We also provide a simple numerical example to show that saturation waves attributed to the instability can propagate away from the discontinuity as relatively undamped oscillations. Because rapid spatial changes in material properties are likely in nature, we suggest that two-phase hydrothermal systems may often exhibit oscillatory behavior. When observational data on oscillations become available, such data may yield information on permeability and/or thermal conductivity structures of two-phase hydrothermal systems.

An alternative model of the Lassen hydrothermal system, Lassen Volcanic National Park, California

We describe an alternative two-dimensional numerical model for the Lassen hydrothermal system. As in earlier models, the new model consists of a vertical upflow zone with a low permeability caprock near its top and a permeable horizontal channel that connects to the upflow zone at depth. The new model, however, also includes a recirculation region beneath the horizontal channel and a recharge zone. Magmatic heating is explicitly represented by specifying heat flows near the bottom of the upflow zone. Simulations show that a periodically appearing vapor-dominated zone, which sometimes consists of pure-steam, develops beneath the caprock and feeds steam-type discharges at high elevations, while the horizontal channel feeds the hot-spring discharge region at lower elevations. This model is able to reproduce the inferred ∼1:1 discharge ratio between water and steam and to match estimates on current heat output from the system. The development of the vapor-dominated zone does not require dramatic changes in geohydrologic conditions. Our model also shows that both the vapor-dominated zone and the whole system evolve quasiperiodically with a period of ∼103 years. These oscillations are related to the unstable two-phase hydrothermal fluid flow within the vapor-dominated zone. The observed high-frequency oscillations near the vapor-dominated zone superimposed on low frequency ones probably result from the oscillatory instability introduced by rapid changes in permeability of the low-permeability caprock. The results indicate that the system is probably still in a transient state.