document preview

Abstract
Trapping of CO2 improves containment security of geologically stored CO2. To be able to asses the potential of a storage site using reservoir simulators, it is necessary to include all of the possible trapping mechanisms in the numerical algorithms. Currently, four trapping mechanisms are identified in the CO2 sequestration literature: structural, residual, dissolution, and mineralization. Although capillary pressure hysteresis has been accounted for in the historical development of simulators, it has not been recognized as a process for hindering movement of the injected CO2. Capillary pressure hysteresis reduces the
buoyancy driven plume movement significantly compared to relative permeability hysteresis; but it is the latter that has been emphasized in the published literature. In this paper, we focus on a quantitative measure for the contribution of hysteresis in reducing plume transport. Rocks with large pore body to throat size ratio are the best candidates for this mechanism to be operative.

In the present work, a self-consistent relative-permeability capillary-pressure hysteresis model is incorporated within a simulator. With this model, it is possible to compare and contrast hysteresis induced retardation to other mechanisms of trapping. The self-consistent parametrization of all of the transport properties is used to quantify sensitivity compactly. The sensitivity of the CO2 plume shape and the amount of CO2 trapped, to the strength of the capillary pressure hysteresis, is also described.

Simulated results show that CO2 plume shape with and without capillary pressure hysteresis are significantly different. As expected, capillary pressure hysteresis retards the buoyant transport of the CO2 plume. Although a portion of the CO2 is connected, and therefore not residual, the plume remains immobile for all practical purposes. Also, due to decreased driving potential, gravity tonguing below the caprock is reduced in comparison to the case without capillary pressure hysteresis, thus suggesting enhanced storage efficiency. However, the total dissolution of CO2 in saline water is reduced because of the reduced diffusive transport of CO2 within the brine. Thus, one mechanism of containment is offset by the other. Inclusion of accurate hysteresis models is important for qualifying storage sites constrained by spatial domain limits. It is anticipated that site acceptability criteria would change as a result of this study, thus impacting risk evaluation.

Introduction
A significant increase in atmospheric CO2 concentration has been observed since the onset of the last century. The level of CO2 concentration in the atmosphere is growing each year, reaching the highest levels ever seen in the last 400,000 years (Petit et al. 1999). The average concentration for atmospheric CO2 recorded at Mauna Loa Observatory was 387.35 parts per million (ppm) in 2009, compared to 385.57 ppm in the previous year (see e.g. NOAA data, ftp://ftp.cmdl.noaa.gov/ccg/co2/trends/co2 mm mlo.txt). Several studies have shown that the change in CO2 concentration and other greenhouse gases cause a warming effect, necessitating technologies that mitigate CO2 accumulation (Metz et al. 2005). Global temperature change is also thought to cause weather anomalies, with the temperature reaching within 1 ±C of the Holocene maximum. Geological carbon sequestration is one of the technologies proposed to reduce atmospheric accumulation of CO2 and thereby mitigate global climate change (Koide et al. 1992; Bachu 2000; Gunter et al. 1998; Holloway 2001; Bruant et al. 2002; Pruess and Garcia 2002; White et al. 2003; Metz et al. 2005).