It works for 4 year olds. :P
What the paper is saying is simply that the salty water is displaced, making room for the CO
2http://onlinelibrary.wiley.com/doi/10.1002/ghg3.1/full#sec1-6…
In the southern San Joaquin Basin, the deep Vedder Sand has been considered an important target formation for GCS in California (Fig. 3). The formation pinches out toward the south, north, and west. To the east, the Vedder Sand (and its equivalent sandstones) outcrops along the edge of the Sierra Nevada mountain range. Its salinity is relatively modest, ranging from 29 000 mg liter−1 in the deeper portions of the formation to less than 100 mg liter−1 in the outcrop region. As a result, we use the term ‘water’ in this subsection to refer to the resident fluid of variable salinity. The primary seal is formed by the Temblor-Freeman shale, except in the northern area where the Vedder Sand connects with the overlying Olcese Sand, another possible storage formation. Numerous oilfields exist in the basin, with their oil/gas pools in different formations, including the Vedder Sand. The oilfields act like closed, partially closed, or open subsystems, evidenced by strong variations in pressure behavior observed during petroleum extraction. For example, the pressure decrease (induced by production of petroleum and produced water) observed at wells and the subsidence imaged using InSAR data indicate that the Kern River oilfield is a closed subsystem bounded by faults and a formation outcrop.22,23 In summary, the Vedder Sand in the southern San Joaquin Basin forms a partially closed storage system with three closed boundaries and one open boundary, and comprises some localized, fault-bounded closed and partially closed subsystems. Several major faults may act as partial groundwater barriers to regional groundwater flow.
A large-scale numerical model of 84 km by 112 km domain size was developed to understand the scale and magnitude of pressure build-up in the partially closed system of the southern San Joaquin Basin. The model represents most of the major geologic and stratigraphic features discussed above. The storage scenario assumes an injection rate of 5 Mt CO2/year at one well (located between the Greeley and Pond faults) for a period of 50 years. The model accounts for pressure attenuation by diffuse water leakage through seals, by focused water leakage through the seal-pinchout area, and by water discharge into the outcrop area of the storage formation, and also represents the effect of fault zones on pressure-build-up propagation. In addition to the base case (with caprock permeability of 10−18 m2 and baserock permeability of 7 × 10−17 m2), we reduced the cap- and baserock permeability to 10−21 m2 for sensitivity analysis. As shown in Fig. 3b (the base case), the pressure perturbation in the Vedder Sand is confined by the southern, western, and northern boundaries of the storage formation at 50 years of injection. The pressure build-up is above 1.10 MPa near the injection center and more than 0.50 MPa in the central area of the basin bounded by the Greeley and Pond faults. In the southwestern region of the storage formation, the pressure build-up is higher than 0.30 MPa, showing the effect of the formation boundaries. The open eastern boundary allows local resident water to flow into shallower formations, without noticeable pressure build-up. Pressure build-up is also less significant in the northern region of the storage formation, because the local absence of the seal there allows water to migrate into overlying aquifers.
The volumetric balance at the end of injection is as follows. The total volume of water displacement includes 333.5 × 106 m3 displaced by free-phase CO2 (with an average density of 656 kg/m3 of the 218.8 Mt free-phase CO2) and 25.6 × 106 m3 by dissolved CO2. This volume is accommodated by 98.9 × 106 m3 of pore volume made available by both pore and water compressibilities in response to pressure build-up in the storage formation, 147.6 × 106 m3 of water migrating from the storage formation into overlying and underlying formations, and 112.6 × 106 m3 of the water migrating through the Vedder outcrop boundary and through the northern and western open boundaries for all formations except the Vedder Sand. This shows that pressure attenuation by water migration from the storage formation accounts for 72% of the additional pore volumes needed to store the injected CO2 volume.
In comparison to the base case, pressure build-up is higher within the entire storage formation if the seal permeability is too small to allow for pressure relief (Fig. 3c). At the end of injection, the pressure increase compared to initial hydrostatic conditions is above 1.45 MPa near the injection center, over 0.8 MPa in the region between the Greeley and Pond faults, and more than 0.7 in the southwestern region. The total volume of water displaced by free-phase CO2 (333.0 × 106 m3) and by dissolved CO2 (24.9 × 106 m3) is 357.9 × 106 m3, very close to that in the base case, indicating that the seal permeability has much less impact on CO2 plume evolution (as long as there is no CO2 leakage through the caprock) than on pressure build-up. This total volume of displaced water is accommodated by 160.4 × 106 m3 pore volume made available by compressibilities in the storage formation, 7.9 × 106 m3 cumulative water volume leaked through the northern area (where the caprock is absent) and stored in the overlying formations, and 189.6 × 106 m3 cumulative water volume migrating through the Vedder outcrop boundary and the seal-pinchout area out of the system. The simulation results in both cases indicate that the water outflow from the system is an important mechanism for pressure attenuation, accounting for 31% and 53% of the total displaced water volumes, respectively. Note that the salinity of the outflowing water through the outcrop boundary is very low, and no environmental impact on shallow groundwater resources is expected.
At the end of injection (the base case), the injected CO2 mass (250 Mt in total) is safely stored in the storage formation, either as dissolved CO2 (31.2 Mt) or as free-phase CO2 (218.8 Mt). The CO2 plume is located between the Greeley and Pond faults. With time, the plume of free-phase CO2 continues to migrate updip while more and more CO2 becomes trapped. Simulation results show that at 1000 years, the total injected CO2 mass is safely contained in the storage formation, either by residual trapping (189.6 Mt) or by dissolution trapping (60.4 Mt), leaving no mobile free-phase CO2 in the model domain.
…