Laboratory Seismic Methods for Remote Monitoring of Thermal EOR

Abstract

Summary Laboratory measurements of compressional (P) and shear (S) wave velocities and first-arrival amplitudes at ultrasonic frequencies (0.65 to 1.70 MHz [0.65 × 106 to 1.70X 106 cycles/sec]) in unconsolidated tar and heavy-oil sands indicate that wave-propagation properties in these materials are sensitive to bitumen content. Studies made at elevated temperatures, overburden pressures, and pore pressures revealed that heat and oil content in Venezuelan, Californian, and Canadian reservoir samples had large effects on the measured P and S velocities and amplitudes. Similar results were obtained on well-consolidated sandstones at lower frequencies (1.60 to 3.60 kHz [1.60x 103 to 3.60 × 10 3 cycles/sec]). The velocity and relative P-wave attenuation results show that in reservoir sands with high brine-to-oil ratios, the presence of steam or gas is easily detected. In sands with low brine-to-oil ratios, with oil occupying at least half of the available pore space, the presence of steam is not easily detected, but velocities and attenuation are highly sensitive to the temperature of heated oil. Indeed, measurable seismic properties may become a powerful tool for mapping temperature distribution, and therefore viscosity distribution, within heated reservoirs. The results suggest that seismic wave transmission and possibly reflection methods should be highly successful in locating thermal EOR fronts and in monitoring the distribution of heated tar and heavy oils within a reservoir. The application for this technique is evident: because of the potential to track heated oil remotely, field operators may be able to potential to track heated oil remotely, field operators may be able to determine optimal placement of future production or injection wells. Introduction An extensive, systematic laboratory study was undertaken to evaluate the application of seismic imaging to the mapping of thermal EOR fronts in unconsolidated sand reservoirs. The study was prompted by known effects of steam or gas on wave propagation in consolidated and unconsolidated rocks, by the successful use of seismic reflection to image steam pods at the Street Ranch and the Saner Ranch pilot projects, and by anticipated but unverified seismically detectable changes in oil viscosity and dynamic bulk compressibility and rigidity upon heating of reservoir sands. The paper summarizes the results of a series of experiments on the effects of elevated overburden pressure, pore pressure, temperature, and oil/brine ratio on pressure, temperature, and oil/brine ratio on ultrasonic-frequency pulse-transmission data. Reservoir samples from three locations were studied:a Venezuelan sand from a heavy-oil field on the eastern coast of Lake Maracaibo,a Californian sand from the Kern River area, anda Canadian sample of Athabasca tar sand. The results were used as experimental controls to assess the applicability of seismic transmission and reflection imaging in a thermal EOR pilot project. Magnitudes of laboratory-determined wave-property changes. the intrinsic resolution of transmission and reflection seismic methods, depth and thickness of the target zone, the scale of heterogeneities within the zone of interest, thermal conductivity of the reservoir sand, and the injection rate of steam are all inputs into the seismic forward model. Modeling permits optimization of observation-well spacings with respect to the injector well, signal strength required of the seismic source, and distance between receiver stations; it also permits assessment of expected signal-to-noise ratios, the frequency band of received signals, and the magnitude of changes in seismic properties expected in a specific field test site. In short, forward modeling is used to design the geometry of the field monitoring system.

Temperature‐dependent propagation of P‐ and S‐waves in Cold Lake oil sands: Comparison of theory and experiment

“…The theoretical model based on the wave propagation equations of de la Cruz and Spanos also predicts an intrinsic attenuation dependence on temperature. The predicted effect on attenuation with increasing temperature was found to be in qualitative agreement with measurements by Tosaya et al (1987).…”

“…Previously, Tosaya et al (1987) reported a significant decrease in measured attenuation with temperature for tar sand samples at ultrasonic frequencies. Thus, the theoretical result presented here is in qualitative agreement with their experimental result.…”

“…To this end, laboratory experiments have been performed to try to establish correlations between seismic attributes and reservoir properties. Experiments performed by Tosaya et al (1987) and Wang (1988) show both a velocity and attenuation dependence on temperature for several heavy oilitar sand samples. These experiments provided quantitative evidence that temperature strongly effects ultrasonic attributes, and thus temperature anomalies in reservoirs could be delineated with seismic methods.…”

Processing and interpretation of time-lapse seismic data from a heavy oil field, Alberta, Canada

“…Decreases in shearwave velocities were also observed, although to a lesser extent than the decrease in Pwave velocity. Lastly, Tosaya et al, (1987) proposed that, for high saturation levels, Pwave velocity can serve as a highly accurate reservoir thermometer for enhanced oil 5 recovery projects. However, Paulsson et al (1994) negated this, cautioning that actual temperatures cannot be derived from seismic velocities.…”

1. Heavy-Oil Reservoirs: Their Characterization and Production

“…This raises the possibility of using seismic methods to detect temperature anomalies in reservoirs. Tosaya et al (1987) and Wang and Nur (1988) performed laboratory experiments to demonstrate the velocity and attenuation dependence on temperature for several heavy-oil/tar-sand samples. Wang et al (1990) measured acoustic velocities at ultrasonic frequency (800 kHz) in oil samples of different API as a function of temperature and pressure.…”