15.01.2015 Views

Microseismic Monitoring and Geomechanical Modelling of CO2 - bris

Microseismic Monitoring and Geomechanical Modelling of CO2 - bris

Microseismic Monitoring and Geomechanical Modelling of CO2 - bris

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

CHAPTER 8.<br />

LINKING GEOMECHANICAL MODELLING AND MICROSEISMIC OBSERVATIONS AT WEYBURN<br />

Unit χ ϕ f<br />

Caprock 5 (18.5) MPa 45 ◦<br />

Reservoir 3.5 MPa 40 ◦<br />

Table 8.4: Yield envelope parameters for the Weyburn model. A lower value than that measured<br />

on core samples is used for the cohesion <strong>of</strong> the caprock, which is given in brackets.<br />

3 months <strong>of</strong> injection (timestep 12) <strong>and</strong> after 1 year <strong>of</strong> injection (timestep 16), both in the reservoir<br />

(Figure 8.7) <strong>and</strong> in the overburden (Figure 8.8). From these figures I note that fracture potential<br />

increases in the reservoir during production, while it is relatively unchanged in the overburden. Once<br />

injection begins, there is a sharp increase in fracture potential in the overburden above the injection<br />

wells, while there is a drop in fracture potential in the reservoir at the injection well. The fracture<br />

potentials at the producing wells are relatively unchanged during injection.<br />

In general, there are some qualitative comparisons that can be made between this model <strong>and</strong><br />

the observations made at Weyburn. For instance, the fact that across most <strong>of</strong> the reservoir fracture<br />

potential is not increased by injection matches with the lack <strong>of</strong> seismicity recorded. Also, this model<br />

suggests that fracture potential should be higher at the production wells than at the injection wells,<br />

which matches the observations that the majority <strong>of</strong> events occur close to the producers. However,<br />

this model can not explain why so many events are located in the overburden above the producing<br />

wells, while the models predict that there should be microseismicity above the injection well, where<br />

none is observed. The suitability <strong>of</strong> this model can also be assessed through a comparison <strong>of</strong> the<br />

seismic anisotropy that it predicts.<br />

2000<br />

17<br />

2000<br />

18<br />

1500<br />

16<br />

1500<br />

17.8<br />

15<br />

17.6<br />

Y<br />

1000<br />

Y<br />

1000<br />

14<br />

17.4<br />

500<br />

13<br />

500<br />

17.2<br />

0<br />

1500 2000 2500 3000<br />

X<br />

12<br />

0<br />

1500 2000 2500 3000<br />

X<br />

17<br />

(a)<br />

(b)<br />

Figure 8.5: Map view <strong>of</strong> vertical effective stress in (a) the reservoir <strong>and</strong> (b) the overburden <strong>of</strong><br />

the Weyburn model at the end <strong>of</strong> 1 year <strong>of</strong> CO 2 injection. The horizontal producers <strong>and</strong> vertical<br />

injectors are marked by lines <strong>and</strong> triangles respectively. Contours are in MPa.<br />

154

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!