The monitor survey was calibrated to match the phase, amplitude and statics to those of the baseline, effectively enhancing production- induced changes while suppressing differences created by other factors. Differences in survey acquisition, processing, and near surface velocities during data acquisition can reduce the repeatability of a monitor survey, altering the phase, amplitude and static solution between surveys. However, observable time-lapse differences are a result of multiple factors, in which reservoir changes are one of many constituents. The strong contrast between the ideal baseline survey and a production influenced monitor survey has allowed for a detailed interpretation of reservoir changes due to production to be made. The monitor data contains steam chambers and elevated temperatures within the reservoir zone, albeit a lower S:N ratio due to noise contamination from production and surface activities.
The baseline survey was recorded prior to production and possesses a high S:N ratio, while the monitor survey was recorded after nine years of steam injection and production.
The 4D dataset analyzed in this study is an ideal candidate for time-lapse interpretation. Difference volumes form the foundation for time-lapse seismic analysis, ideally integrated with reservoir characterization, geological modeling and reservoir production (Johnston, 1997). To aid in the time-lapse interpretation, a difference volume was produced through the subtraction of the baseline data from the monitor data, producing a third dataset comprised of traces that are different between surveys. The analysis of monitor surveys may allow for the detection of both large and subtle changes within the reservoir (Johnston, 1997). The objective of timelapse seismic monitoring is to image production induced changes within the reservoir and to identify areas of bypassed reserves, or regions in which current steam injection is not optimally stimulating reserves. Time-lapse monitoring is comprised of a baseline survey, ideally recorded before the onset of production, and a monitor survey recorded after a period of oil, gas or water production (Clifford, et al., 2003 Kalantzis, 1996). Outside of the amplitude anomalies, the channel sands were bound by muddy IHS bedding, creating baffles to steam flow. The channel sands were observed to intersect amplitude anomalies with the monitor volume. McMurray Formation channels sands were observed within the seismic data through the analysis of the semblance attribute, as well as within the geological data as low gamma ray values on well log cross sections. Geological well log information was integrated with the geophysical observations. The attenuation of high frequencies beneath steam chambers was observed within the monitor survey, characterized by lowfrequency shadows, observable on the Devonian reflection underlying the amplitude anomalies. An analysis of the amplitude anomalies yielded a spatial display of reservoir steam chamber distributions. The decrease of velocity was interpreted to be due to the increase in reservoir temperature and decrease in differential pressure created from the injection of high temperature steam into the McMurray Formation reservoir.
High amplitude anomalies were observed on the monitor data in conjunction with apparent time-thickening of the reservoir interval due to a decrease in the P-wave velocity. Two 3-D seismic datasets and their difference volume were interpreted and analyzed for the presence of amplitude anomalies and time delays related to the injection of steam into a shallow, heavy oil reservoir.