Infilling 4C ocean bottom cable surveys

null

Such deviations can leave areas of deficient seismic fold (coverage) in the final coverage map, not only where the cable has deviated, but also where the source vessel has deviated to avoid surface obstructions, such as tanker loading buoys and their associated irregular visits from shuttle tankers. Such final coverage plots are typical of surveys acquired in built-up areas.

The challenge is to reduce the areas with insufficient seismic coverage around an active field. But how can this be achieved?

Improved positioning

Clearly, field infrastructure must be avoided when deploying ocean-bottom cables, but the safest minimum closest point of approach (CPA) must take into account the positioning accuracy of, and the ability to control, the cable's touch-down point. Multiwave's near-seabed deployment system (NSDS) enables the cable to be placed to within 1% of water depth from the pre-plotted coordinates.

The curved deviations displayed by the cable on either side of the platform are well controlled, accurately known, and can be placed as close as possible to the edge of the minimum CPA, as agreed at the survey planning stage. Nevertheless, even after the implementation of the NSDS, deficient seismic coverage can remain.

4C data can also be recorded using nodes instead of cables. 4C nodes are self-contained, autonomous seismic recording devices that are deployed on the seabed and left to record 4C data for as long as the batteries last. When the node is recovered, its data-logger is downloaded to the master disc on the recording vessel. Nodes can then be reset and deployed again.

A seabed seismic cable cannot be laid through the legs of a platform or through one side of an anchor spread and out the other. Nodes, however, can be placed anywhere to the same degree of accuracy as a cable. Conceptually, a "virtual cable" extension can be reconstructed under almost any obstruction by planting nodes at intervals equal to the sensor interval of the cable.

Before this technique could be put into commercial operation, it was important to demonstrate that data from 4C nodes was compatible with data from 4C cables and to validate a merging of the two data sets. In the summer of 2003 a series of tests were performed, where nodes were planted to within 1 m of an existing cable station so that a direct seismic comparison could be made.

Data was acquired by both recording systems independently of each other, but from the same overhead source. At the end of each swath, both cable and node were recovered. Data was downloaded from the node before its next re-deployment. This same test was repeated on four other swaths with the node planted less than 1 m from a cable station.

Both systems use non-gimbaled geophones to optimize geophysical coupling to the seabed because the mechanical transfer function from soil to geophone is the most direct. Gimbals and slip rings rely on mechanical clearances that are comparable to the soil motion being measured.

Non-gimbaled data sets do require special attention in processing. The data must be rotated from the geophones' random, as-laid orientation to the axes of the survey grid. For this test, both the cable and the node required independent rotation analyses to produce like-for-like receiver gathers.

Data management

Once a node is deployed, it records continuously until recovery. For each component, a single, contiguous file is recorded that represents several days of continuous data. But how is this data re-formatted to SEG Y?

The first step involves correcting the clock drift within the node, during its week of recording on the seabed. Each component's file can then be merged with shot number information from the source system. The recorded data is split into a more recognizable, one per shot, per component format.

null

Tilt sensors are in the node's recording head, and this data is encoded into the seismic record. These orientation values must be extracted and quality controlled before the data can be rotated to the survey grid.

The node's geophones are mechanically mounted in a galperin, rather than a horizontal/vertical configuration. A galperin system retains orthogonality among three axes, but is symmetrical in relation to the vertical. This approach has the advantage of ensuring that any real mechanical influence in the principal horizontal/vertical axes cannot dominate one particular geophone, and vector fidelity is optimized. The challenge is to perform a full 3D rotation on the raw data set: from galperin to survey horizontal and vertical.

Within the cable, geophones are mounted in the more conventional, orthogonal orientation. The inline geophone is aligned with the cable's longitudinal axis, while the two other geophones are mounted perpendicular to each other in the cable-crossline axis. Data rotation is necessary to bring the information into survey horizontal and vertical alignment. One way to perform rotation of cable data is using hodogram analysis.

For this technique, a first-break window is selected for each un-rotated station, and each seismic sample's amplitude is plotted on a Y-Z hodogram to derive a line of samples. Interpolation of a straight line though the major axis of these points yields the angle by which data must be rotated in order to maximize the energy in the Y component.

After rotating the multi-component data from two independent receiver systems, the records showed that the data was similar enough to merge without significant deviations to the data processing flows.

By understanding the technical and operational risks associated with such an approach, it is possible to incorporate both 4C cables and nodes into commercial surveys. The safe delivery of contiguous multi-component seismic data in and around complex producing fields now becomes an achievable proposition.