Effect of Relative Permeability Characteristics and Gas/Water Flow on Gas-Hydrate Saturation Distribution

Abstract

The descent of the base of gas hydrate stability zone (BGHSZ) through gas accumulated in a sediment is analyzed. A mechanistic model enables estimating hydrate saturation from initial distribution of gas phase saturation in sediment with known grain size distribution. The initial gas phase saturation is estimated from the profile of capillary entry-pressure with depth. The latter is estimated from grain size variations. A mechanistic model is proposed to determine the relative rates of methane and water transport into the HSZ during hydrate formation. The gas accumulation is assumed to be isolated so that methane transport occurs only within it. If water transport occurs only by co-current flow of gaseous and aqueous phases up to the hydrate stability zone (HSZ), it is not possible to create large hydrate saturations from large initial gas saturations due to limitations on water flux imposed by typical relative permeability curves. Thus the observed large hydrate saturations, such as that observed in Mt. Elbert, Alaska and Mallik, NW Territories and deep Indian Ocean, above the BGHSZ suggest another form of water flow: water moves down through accumulated hydrate from above. This requires the aqueous phase to remain connected within the hydrate-bearing sediment. The ratio of aqueous phase permeability in the hydrate-bearing sediment to the aqueous phase relative permeability at residual gas saturation determines hydrate saturation profile. The model is validated against field data from a hydrate-bearing sand unit in Mt. Elbert.

Physical mechanisms for multiphase flow associated with hydrate formation

“…In this section we derive the generalized fluid flow model to account for capillarity‐driven transport for which the saturation gradients provide the driving force. Here we recapitulate the main features of the model of Behseresht and Bryant [] but include the relevant terms for capillarity‐driven and gravity flow.…”

“…In section 4, it is shown how this dimensionless group can be used to predict the final hydrate saturation profile in terms of having a sharp or diffuse basal contact. Behseresht and Bryant [2011] showed that owing to the very low mobility ratio of the aqueous to gaseous phase, M ≪ 1, in the process of converting a gas accumulation into hydrate accumulation, the viscous contribution to aqueous phase flow is negligible, i.e., u wDV ≈ 0. In the simulations of section 4, it was shown that the capillary flow, u wDC , derived by saturation gradients, is in charge of the aqueous phase flow toward the GHSZ.…”

“…In Case 3, the required aqueous phase flux can be provided only by pressure‐driven flow; the product of saturation gradient and the slope of the P c ‐S w characteristic curve is no longer large enough to meet the demand. This leads to a viscous‐dominated fractional flow of the phases [ Behseresht and Bryant , ] in which gas flux overwhelms the capacity for water flux ( t = 0.7 year and t = 1 year; see Figure ). The GWC rises almost at the speed of the classical Buckley‐Leverett shock front (Figure , left column); before t = 2 years, the flux of aqueous phase is negligible compared to that of the gaseous phase (Figure , left column).…”

“…Behseresht and Bryant [] also showed that in the case of Mount Elbert, the transported gas fraction must have been over half of the total fluid, i.e., R v = 0.55. Their model, however, did not consider the mechanism by which fluids migrated nor did it explain the ratio of phase volumes; Behseresht and Bryant [] showed that pressure‐driven viscous‐dominated flow is unlikely to have been the only mechanism that provided the necessary fluid. Hence, the motivation to validate the more general model presented here using the Mount Elbert observations.…”

A Novel Relative Permeability Model for Gas and Water Flow in Hydrate-Bearing Sediments With Laboratory and Field-Scale Application

“…Experimentally, pore-filling hydrate has been shown to naturally become load-bearing hydrate when the pore saturation exceeds approximately 40% 3 . Hydrate formation models have been developed based on different hypothesis 42,[46][47][48][49][50][51][52][53][54] describing how hydrates are formed, which suggests lack of conclusive evidence on the mechanism behind formation of hydrates. In this study the conceptual model presented in Fig.…”