Large-Scale Geological Carbon Sequestration - Introduction to Carbon Capture and Sequestration - page 460

(b)

Usual S grmax (~0.2) Half S grmax (~0.1)

Obs

Obs

Inj Inj

S g S g 0 0.05 0.1 0.15 0.2 0.25 0 0.05 0.1 0.15 0.2

(a)

(c)

Figure 10.2.2 Sensitivity of fi eld-scale model predictions to residual gas saturation
Illustration of the sensitivity of fi eld-scale model predictions to model input parameters.
(a) Schematic description of the Frio I sequestration pilot site (a sandstone formation near
Houston, Texas, where about 1.6 kilotonne of CO 2 was injected at a depth of 1,530 m).
(b) Experimental data on CO 2 saturation ( S g ) during the CO 2 injection test; the left and
right sides of the fi gure show S g values (more precisely, changes in resistivity caused by
the presence of CO 2 ) measured as a function of depth at the injection (left-hand side)
and observation wells (right-hand side); the central part of the fi gure shows a recon-
struction of CO 2 saturation in the region between the wells from cross-well seismic
measurements ( Figure 10.5.5 ).
(c) Model predictions of S g for two different values of the maximum residual gas satura-
tion ( S g,r,max = 0.2 or 0.1 on the left-hand and right-hand sides), a model parameter
discussed in more detail in Section 9.7. The model correctly predicts the overall shape
of the CO 2 plume, but not its fi ne structure; model predictions are highly sensitive to
S g,r,max , a poorly-constrained input parameter ( Table 9.7.1 ).
(a) and (c): Reproduced from Doughty et al. [10.3] . (b): Reproduced from Daley et al.
[10.2] . All fi gures reproduced with kind permission from Springer Science + Business
Media.