The convection–diffusion process of carbon dioxide (CO2) dissolution in a saline reservoir is investigated to shed light on the effects of the permeability heterogeneity. Using sequential Gaussian simulation method, random permeability fields in two and three-dimension (2D and 3D) structures are generated. Quantitative (average amount of the dissolved CO2 and dissolution flux) and qualitative (pattern of the dissolved CO2 and velocity streamlines) measurements are used to investigate the results. A 3D structure shows a slightly higher dissolution flux than a 2D structure in the homogeneous condition. Results in the random permeability fields in 2D indicates an increase in the standard deviation of the permeability nodes enhances the dissolution efficiency, fluctuations in CO2 dissolution flux, separation between the different realizations from the same input parameters, and tendency toward more jagged convective fingers’ shape. Furthermore, the distance between the permeability nodes increases the convective fingers’ dissolution efficiency and jagged structure. The degree of freedom in 3D structures results in a higher chance of escaping from the low permeability zones and reduces the interactions between convective fingers in 3D systems. With the same variance and correlation length between permeability nodes, connectivity between high permeable zones in 3D cases are less than that of 2D cases; therefore, 2D realizations overestimate the dissolution flux of real heterogeneous 3D structures, which should be considered carefully.
Article Highlights
CO2 sequestration in two and three dimensional heterogeneous saline aquifers are investigated.
3D structures in homogeneous conditions show higher dissolution than 2D structures.
2D realizations overestimates the dissolution flux over real heterogeneous 3D reservoirs.