Schlumberger’s Patrick Smith reviews technological developments that are increasing the accuracy of 4D seismic reservoir monitoring.
 |
|
Schematic representation of the reconstruction of a multi-measurement shot record. The red lines represent the streamer locations of a previous baseline survey. The pale blue lines represent multi-measurement streamers at 75m spacing. The dotted lines represent the dense 6.25m x 6.25m grid of traces generated by the wavefield reconstruction process. This dense grid of traces can be accurately interpolated to the receiver positions of the baseline survey streamers.Image from Schlumberger. |
Continuous improvement in the quality, reliability and consistency of surface seismic imaging technology has led to an increased use of the time-lapse (4D) method to monitor changes in producing hydrocarbon reservoirs. This approach involves acquisition of 3D seismic surveys at intervals from before and during production of a reservoir. Differences between the repeated seismic images can indicate subsurface changes in pressure or fluid content, leading to improved reservoir management decisions, identification of flow barriers, and location of untapped compartments suitable for infill drilling.
However, the effect of these changes can be very subtle and may be masked by inconsistencies between the acquired surveys. Minimization of variability in the acquisition and analysis processes is essential for reliable delivery of repeated seismic images in which the only differences are due to subsurface changes.
Marine time-lapse projects require a stable seismic source with an accurately known signature and stable receivers with known characteristics. Source and receiver locations should be accurately repeated from one survey to the next. Changes in environmental conditions such as tides, water velocity, and wave heights should be accounted for. In addition, data processing of newly-acquired datasets must match that of previous vintages and be completed with rapid turnaround; otherwise the value of the data may not be realized in time to support drilling schedules.
The Q-Marine point-receiver marine streamer seismic system, introduced in 2000, was designed to maximize time-lapse survey repetition accuracy. IsoMetrix marine isometric seismic technology, launched in 2012, builds on the technology introduced by Q-Marine and, through its multi-measurement streamer reconstruction, has the capability to take repeatability to an unprecedented level of accuracy. Meanwhile, robust and efficient workflows have been developed to deliver reliable and timely results from data processing.
Repeatable in-sea equipment
The Q-Marine system provides several features designed to maximize the repeatability of in-sea equipment performance and positioning between vintages of surveys. The calibrated marine source (CMS) solution removes shot-to-shot variations in the source signature due to factors such as pressure, array geometry, and dropouts. In addition to compensating for relative differences due to such source variations, CMS helps to remove absolute variations between surveys. In addition, it also benefits imaging and inversion processes, such as amplitude-versus-offset inversion, and rock property characterization by accurately collapsing low-frequency source bubble energy.
 |
| Fine isometrically-sampled seismic data provides the possibility to separate and remove interference such as noise from nearby seismic activity, drilling, or oilfield vessels, reducing non-productive acquisition time in producing fields that may have several such sources of noise. The new system provides broadband data with enhanced signal-to-noise ratios including in the low frequency range (e.g. between 2 and 8Hz), which benefits both 3D and 4D inversion processes that can help to indicate rock properties. Fine-scale isometric subsurface characterization means that interpretation attributes can be generated independent of the orientation of viewing. This translates into more detailed representations of subsurface structures and stratigraphic variations, and enables a new level of insight into the geology from seabed to reservoir. Image from Schlumberger. |
Point-receiver acquisition enables the application of advanced digital technology that effectively attenuates many types of noise that typically affect sensors in towed streamers. The Q-Fin marine seismic streamer steering system is used to adjust the position and depth of streamers during a survey, and the capability to attenuate high levels of noise allows deployment of more powerful steering devices that exert stronger lateral forces. The dynamic spread control (DSC) system uses current information and data from a dense in-sea real-time acoustic positioning network to automatically and independently steer vessels, sources and streamers to achieve an accurate repeat of a previous survey. During a repeat survey in the North Sea, despite less predictable ocean currents and higher natural feather relative to the baseline survey, DSC enabled 95% of source positions to be within 2.5m of planned positions. Streamer feather angle was repeated within a margin of less than 2.5° for 95% of the time. A further benefit of point-receiver recording is that filters can be applied to emulate the spatial frequency response of previous surveys acquired using conventional hydrophone arrays.
Corrections for variations in sea level due to tides can be applied either using tide table predictions, in which case correction for atmospheric pressure is advisable, or from differential global positioning system (DGPS) tide height measurements, which typically provide 10-15cm accuracy. Wave height correction can be calculated and applied during data processing to address perturbations in the seismic measurements induced by the roughness of the sea surface, which become significant once other survey repetition challenges have been addressed. Variations in seismic velocity through the water layer resulting from factors such as temperature or salinity can be accounted for using measurements from a moving vessel profiler on the seismic vessel to derive a unique space- and depth-variant water column velocity for each sail line. Accurate survey repetition minimizes the need for statistical matching processes, and deterministic time-lapse processing workflows have been developed that take advantage of this to ensure that the 4D seismic measurements are accurate from surface to below the reservoir.
Multi-measurement streamer reconstruction
With modern steering technology, source locations can be repeated from one survey to the next to within around 2m; however, even with streamer steering, it can be difficult to consistently repeat receiver locations. It is common to minimize receiver positioning errors by deploying streamers at half the spacing of the baseline survey, but this increases cost. The introduction of multi-measurement streamer technology enables data acquired with economically feasible streamer separations to be accurately reconstructed at user-defined locations and datums. Two or more multi-measurement surveys may be reconstructed at common locations, and these surveys will benefit from the broader spatial and temporal bandwidth provided by the new system. A multi-measurement monitor survey may also be reconstructed at the receiver locations, and with the same essential characteristics, of a previous conventional marine streamer dataset.
The multi-measurement system—IsoMetrix technology—is based on a unique point-receiver streamer system that combines hydrophones with calibrated accelerometers that measure particle acceleration in the seismic wavefield. For each seismic shot record, the measured pressure (P), vertical (Z), and crossline horizontal (Y) components of the pressure gradients are combined to create an estimate of the full 3D broadband seismic wavefield sampled on a 6.25m-by-6.25m point-receiver surface grid. Figure 1 shows how this densely sampled grid enables accurate computation of a set of “virtual” streamers that match the exact positions of those from a previous survey.
The new system allows streamers to be towed further apart and deeper than conventional arrangements, delivering data of equal or better bandwidth. A 2013 Barents Sea survey saw multimeasurement streamer data acquired at 23m tow depth to minimise ambient noise. Streamer steering was used to closely match the locations of an earlier Q-Marine, hydrophone-only, sail line acquired at 8m depth with a similar configuration.
The IsoMetrix dataset was processed to simulate the Q-Marine data and, as shown in Figure 2, the results show a close match between the two datasets, with no coherent energy in the difference section that might adversely influence 4D analysis.
Patrick Smith is a senior area geophysicist with Schlumberger and has more than 34 years of experience in seismic processing. He has worked with time-lapse data since 1991 when the first commercial monitor surveys were acquired in the North Sea. Smith has worked extensively with both Q-Marine and IsoMetrix time-lapse projects. Based in Norway, he provides support to time-lapse seismic processing teams worldwide. Smith holds a degree in geophysics from the University of Reading.
Subscribe
