Time Lapse Borehole Seismic Monitoring of CO2 Injection in an Oil Reservoir
Abstract
A series of seismic borehole experiments were conducted in a diatomite reservoir to monitor the injection of CO2 into a hydrofracture zone. This work has application to subsurface sequestration of CO2 and enhanced oil recovery. A comprehensive pre-injection experiment was conducted acquiring seismic cross-well and single-well data. The seismic waves were excited by two different source types, a high frequency piezoelectric source with a spectrum between 0.8 and 3.5 KHz and an orbital vibrator with a spectrum between 70 and 350 Hz. The receiver well was located 30m from the source well and contained 3-component clamping geophones as well as hydrophones. The injection well was located about 15m from the source well. During the pre-injection phase water was injected into the hydrofrac zone. The set of seismic experiments was repeated after a time interval of 7 months during which CO2 was injected into the hydrofractured zone. The initial experiment was intended to estimate a baseline of reservoir parameters within the hydrofractured zone, while the post injection experiment provides the opportunity to determine and monitor changes in the formation due to CO2 injection. The questions to be answered ranged from the detectability of CO2 by seismic waves to the variability of reservoir pressure due to gas injection. In the case of the high frequency source, we find that the P-wave velocities, during the pre-injection experiment, exhibit relatively low values between 1700-1900m/s, and that these estimates decrease to 1600-1800m/s during the post-injection phase. However, some zones reveal an increase in P-wave velocity, possibly indicating an increase in pore pressure. Differencing the pre- and post-injection data reveals the possible locations of CO2, which is expected to be in gas form at the injection depth range from 480 to 620m. The orbital vibrator data support the findings listed above. In addition to the P-wave data, the orbital vibrator source generates horizontally polarized S-waves. Initial results of the pre-injection data reveal slow S-wave velocities between 600-800m/s, which translates into Poisson ratios between 0.3-0.4. We expect to present post-injection S-wave results as well as differenced S-wave estimates. The single-well data show velocities slower than the cross-well velocities possibly reflecting the disruption of the medium by the borehole and its effect on the propagation of elastic waves around the source well. Furthermore, single well reflection data reveal an arrival that may be reflected off the hydrofracture between the source and receiver wells.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2001
- Bibcode:
- 2001AGUFM.S31A0581G
- Keywords:
-
- 0915 Downhole methods;
- 0935 Seismic methods (3025);
- 7203 Body wave propagation;
- 8180 Tomography