Microseismic Monitoring and Geomechanical Modelling of CO2 - bris

1.3. THESIS OVERVIEW
as microseismicity must surely be viewed as a manifestation of wider geomechanical deformation in
and around a reservoir.
Furthermore, because the waves from events in the reservoir recorded on downhole geophones have
travelled through only reservoir and caprock materials, wave propagation effects such as anisotropy
can be directly attributed to these materials. Teanby et al. (2004a) show how analysis of shear wave
splitting (SWS) can be performed on recorded microseismic events. By considering the magnitude of
splitting and the orientation of the faster S-wave it is possible to identify the orientation and number
density of fracture sets that act as flow paths within the reservoir, as well as image stress-induced
anisotropy. Both stress changes and the presence of faults and fractures can significantly influence the
security of CO 2 storage, and so the possibility of detecting them using microseismic monitoring will be
of great use to a reservoir engineer. Furthermore, microseismic recording arrays, once installed, cost
little to maintain and operate. As such, they will provide a more cost effective method of monitoring
storage security over the long term, especially after injection has ceased and the field has been shut
in.
1.3.3 Thesis outline
I will begin this thesis by introducing the Weyburn reservoir, currently the largest CO 2 storage site
in the world. It is also the first CCS site to deploy microseismic monitoring, and in Chapter 2 I
will discuss the results of this monitoring program, showing microseismic event locations and how
they correlate with injection and production activities. In Chapter 3 I develop a novel approach to
invert shear-wave splitting measurements for fracture properties. I use synthetic models to show the
sensitivity of SWS analysis to the range of ray coverage available, before using the technique to image
the fractures at Weyburn.
One of the key observations made at Weyburn is a very low rate of microseismic activity. This
has lead to the suggestion that CO 2 may have an inherently lower seismic deformation efficiency than
other fluids such as oil or water. If CO 2 injection and/or migration does not generate microseismicity
then this has obvious implications for the feasibility of microseismic monitoring for CCS. To evaluate
this issue, in Chapter 4 I discuss a second microseismic dataset where both CO 2 and water have been
injected into the same reservoir (a different North American oil-field). This allows me to make a
direct comparison of the microseismic response to injection of the two fluids, and to discuss whether
the abundant experience of water injection found in the oil industry will be applicable to CO 2 injection.
In Chapter 5 I outline the geomechanical modelling tools that were developed as part of the Integrated
Petroleum Engineering, Geomechanics and Geophysics (IPEGG) research consortium. This
consortium has developed a method to couple industry-standard fluid-flow simulations (such as Eclipse,
MORE, VIP or MoReS) with a finite element geomechanical solver (ELFEN). I develop some simple
numerical models to demonstrate the sensitivity of injection-induced stress changes to reservoir geometry
and material properties. By examining the stress evolution, and in particular the development
of differential stresses, I can determine which geometries and material properties are most prone to
fracturing as a result of injection, and where fracturing is likely to occur.
7