Abstract Geological carbon storage aims at long-term storage of carbon dioxide (CO 2 ) in deep geological formations to reduce anthropogenic CO 2 emissions into the atmosphere. The viability of CO 2 storage hinges on how much of the CO 2 can be injected into the storage formation. However, the rate of CO 2 injection can be limited by low injectivity of the storage formation and limits placed on the injection pressure to maintain seal integrity. Therefore, pressure evolution during CO 2 injection is an important consideration for CO 2 storage operation and is investigated in this study. We use geological characteristics of a typical oil field near a salt diapir in the Gulf Coast basin in the Southern United States. In this case CO 2 is injected into a complex sedimentary rocks deposited in a fluviatile environment, juxtaposing volumes of high- and low-permeability rocks. The rock volume in which CO 2 injection is taking place can be modelled, for the purpose of this study, as a simple reservoir surrounded by mud rocks which exhibit small, but finite, permeability and high compressibility. The mud rock is expected to behave as a capillary seal preventing upward CO 2 migration, but may allow significant pressure dissipation. This pressure attenuation can be important from an operational standpoint as a fault in the vicinity of the injection well requires an accurate estimate of pressure evolution to avoid leakage along the fault. Most numerical simulations of geological CO 2 storage have focused on the behavior of fluids within a storage formation and assumed that overlying and underlying rocks are impermeable and incompressible. Such a reservoir model has immediate pressure buildup and drawdown corresponding to start and end of injection. Numerical simulations using reservoir parameters and a simple geomechanical model based on rock and fluid compressibilities were performed with the commercial simulator GEM from CMG. They show that overlying and underlying mud rocks attenuate pressure build-up within a reservoir during injection and extend the period of pressure recovery after the end of injection. The attenuation of pressure propagation within a target formation can reduce the probability of creation and/or reactivation of geological discontinuities. In addition, strong pressure gradients detected just outside the storage formation require high numerical resolution at the boundaries between the storage formation and mud rocks, otherwise extensive numerical diffusion will lead to unphysical pressure dissipation. It is therefore not just important to include the surrounding rocks but also to discretize them appropriately, and this is currently not standard practice.
[1]
P. Domenico,et al.
Physical and chemical hydrogeology
,
1990
.
[2]
G. Chierici,et al.
Water Drive From Interbedded Shale Compaction In Superpressured Gas Reservoirs A Model Study
,
1978
.
[3]
The Effect of Faults on Dynamics of CO2 Plumes
,
2009
.
[4]
Serge A. Shapiro,et al.
Characterization of fluid transport properties of reservoirs using induced microseismicity
,
2002
.
[5]
C. Tsang,et al.
A study of caprock hydromechanical changes associated with CO2-injection into a brine formation
,
2002
.
[6]
André Revil,et al.
Permeability of shaly sands
,
1999
.