Prediction Performance of Permeability Models in Gas-Hydrate-Bearing Sands

Permeability variation in the presence of gas hydrates (GH) is a major unknown in modeling hydrate dissociation in gas-hydrate-bearing sediment. Reduction of permeability in porous media occurs as a result of decreased porosity because of hydrate formation within pore spaces. In the absence of reliable experimental data, theoretical and empirical models have been proposed to establish the relationship between gas-hydrate saturation and permeability. The effectiveness of a particular permeability model in fitting the measured data has largely been qualitative through graphical analysis. In contrast, this paper introduces a quantitative performance measure to evaluate the effectiveness of an individual model in predicting the measured permeability. Second, a hybrid approach based on the weighted combination of existing permeability models is proposed. Permeability measurements from experimental and field studies were used to assess the prediction performance of various permeability models and the proposed hybrid approach.