Streamline‐based simulation of carbon dioxide storage in a North Sea aquifer

Abstract

[1] We model carbon dioxide (CO 2 ) storage into a deep North Sea aquifer using streamline-based simulation. We assume incompressible flow of liquid-like CO 2 and aqueous phases. We simulate dissolution of CO 2 and a rate-limited precipitation reaction. Advective transport and reactions are solved along streamlines, while dispersion and flow due to gravity segregation of the phases are solved on the underlying grid. Geological storage is modeled on a one million cell model. The distribution of CO 2 after injection is dominated by advective transport due to multiphase flow, and CO 2 moves preferentially through high-permeability channels. Without reaction the regional groundwater flow causes the CO 2 to continue to migrate until it reaches residual saturation, where it continues, slowly, to dissolve. Precipitation leads to a decrease in porosity and permeability, while CO 2 is stored in the solid phase. The storage efficiency is low, around 2%, because of aquifer heterogeneity.

Estimating permeability from quasi-static deformation: Temporal variations and arrival-time inversion

“…Typically, in modeling CO 2 -brine systems it is assumed that capillary effects are negligible (Pruess 2004, Obi andBlunt 2006) and hence p w = p n = p, where p denotes the common fluid pressure. Adding equations (1) together and assuming that the pressure in the aqueous and non-aqueous phases are equal results in…”

Impact of relative permeability hysteresis on geological CO2 storage

“…Many authors have presented simulations of CO 2 injection and migration [see, e.g., Johnson et al, 2000;Ennis-King and Paterson, 2002;Wellman et al, 2003;Pruess et al, 2003;Xu et al, 2003;Doughty and Pruess, 2004;Flett et al, 2004;Kumar et al, 2005;Obi and Blunt, 2006] using a variety of approaches. Because of the density difference between the CO 2 and the brine, the low viscosity CO 2 tends to migrate to the top of the geologic structure.…”

On the role of caprock and fracture zones in dispersing gas plumes in the subsurface

“…Juanes et al, 2006;Obi and Blunt, 2006]. At the leading edge of such a current, where water is displaced by the gas, capillary forces will lead to some residual water being trapped in the pores.…”

“…Some of the original models were based on numerical simulation of multiphase flow but also took into account some of the particular trapping processes as described above, including dissolution, mineralization, and capillary trapping of the gas [e.g., Johnson et al, 2001;Pruess et al, 2003;Xu et al, 2003;Obi and Blunt, 2006]. These numerical models show that, owing to two-phase flow effects, there is typically a dispersed zone at the boundaries of the gas plume across which the gas saturation varies, wheras the saturation in the central core behind the main front shows less variation.…”

“…[4] In typical water-wet rock, the hysteresis associated with the water-gas drainage and imbibition processes, which occur as the nose of the gas plume displaces water and then as water imbibes into the pore spaces vacated by the tail of the gas plume [Bear, 1972;Obi and Blunt, 2006], is important for determining the fraction of the gas plume that becomes trapped in the formation [cf. Juanes et al, 2006;Obi and Blunt, 2006].…”