Seismic monitoring of an onshore carbonate reservoir in a desert environment has been achieved for the first time. Optimizing data repeatability was key to detecting the weak 4D (time-lapse) signal resulting from a fluid-injection program, which was achieved through a combination of specialized survey design, careful acquisition, and dedicated 4D processing. The hybrid acquisition system utilized buried geophones, which significantly reduced 4D noise caused by variability in the near-surface environment. Despite the extensive acquisition efforts, time-lapse processing is an essential component of achieving highly repeatable data. A fit-for-purpose workflow was developed to reduce the remaining 4D noise using a combination of parallel and simultaneous processing. Processing steps leading to the largest improvement in reflection signal-to-noise ratio, such as noise attenuation, amplitude balancing, and supergrouping, produced the largest reduction in 4D noise. Outstanding final migrated data repeatability has been achieved, comparable to levels reported for the more favorable permanent marine systems. However, the need to use surface sources results in a seasonal imprint on data repeatability, which hinders the interpretation of surveys acquired during different seasons. In the absence of a fully buried acquisition system, advanced processing techniques such as surface-consistent matching filters may be required to resolve these variations.
Distributed Acoustic Sensing Using Dark Fiber for Near-Surface Characterization and Broadband Seismic Event Detection
