Seismic Imaging Technology Part IV

PART IV: THE ULTIMATE MARINE SEISMIC ACQUISITION TECHNIQUE?

Unattainable criteria?

Illustration of OBS acquisition. The principle of reciprocity states that a receiver (yellow star) can be considered the source, and all sources on the surface to be receivers. A partial image of the subsurface is obtained by migration of this shot record.In reservoirs within geologically complex settings, the seismic images are still not perfect.  Advanced 3D imaging requires huge amounts of data, densely recorded over a large area, to properly reconstruct the geology and stratigraphy at a high resolution. The optimum seismic data collection technique is yet to be found. The ultimate survey method may be unattainable, but it should try to satisfy three criteria: full illumination, full noise suppression, and low to moderate cost.

Two routes of future development are envisaged. One that has the potential to revolutionize seismic acquisition is the development and deployment of autonomous underwater vehicles (AUVs). These would float down to the seabed in thousands, in order to record seismic data over several periods of weeks, while a shooting vessel traverses the surface. The other is to significantly expand the number of streamers towed behind the vessel, so that the receiver coverage gets much larger, preferably with the smallest receiver intervals possible, both crossline and inline.

To follow our line of reasoning, it is necessary to know a bit more about Ocean Bottom Surveying (OBS) and imaging. In GEO ExPro No 3/2008 we hopefully convinced everyone that OBS produces superior seismic imaging. What is needed is to make OBS cheaper, so that we can afford to sample on a much denser grid.

Lord Rayleigh helped the geophysicist

In OBS acquisition, cables or nodes with geophones and hydrophones are deployed on the ocean bottom. Data are recorded while a shooting vessel traverses the surface on a 50m by 50m grid. When geophysicists process the data, they can take advantage of the principle of acoustic reciprocity, credited to the famous physicist Lord Rayleigh. He showed that sources and receivers can be interchanged without affecting the recorded signal. Thus, in OBS the receiver location can be considered the source location, and the source location considered the receiver location.  Imagine, therefore, that one receiver at the seabed is the source, and all shots distributed over the sea surface are receivers distributed over the same surface. Then, from this set of data, a partial image of the geology can be obtained by the process of shot profile depth migration. The sum of all partial images, one from each receiver in the OBS experiment, gives the full seismic image.

Autonomous underwater vehicles

The challenge is to make OBS acquisition much cheaper, either by using cables or nodes. In this article we do not address cables, which may face limitations in deep water, or in areas with rugged seafloor or extensive seafloor infrastructure. The major node advantage is that it gives the ability to deploy large receiver grids and minimize the shot overlap burden inherent in many cable-based methodologies. Using today’s OBS nodes, typically four nodes can be deployed per hour by using remotely operated vehicles (ROVs) to obtain good precision and coupling of the sensors to the seabed. In one day, therefore, 96 nodes can be deployed. If the survey requires 5,000 node locations, the operation time for deployment is 50 days. The time for recovery is somewhat less, but with today’s vessel rates, the cost would be very high. Imagine, therefore, if say 500 nodes can be deployed each day. Is this possible? We believe the oil and gas industry is ready for more cost-effective solutions. Seven-eight years ago it was proposed that AUVs could be dropped from the back of a vessel one-by-one with an interval of only minutes. Driven by gravity and buoyancy, the AUVs rapidly glide into a pre-planned grid on the sea bottom. When the survey is complete, they receive a return command, blow their water ballast, and ascend to the surface where they are picked up by the vessel.)

Unfortunately, this project did not get sufficient financial support from the industry at the time. But the idea is great, and deserves attention from anyone who wants cheap high quality seismic.

There are several major hurdles though which AUV technology must pass. The single most important one will be to increase the life of the battery, while still keeping it small enough to fit into the AUV.  A typical 4-component case solution, recently used in a survey in Angola, shows that the battery is several times bigger than the node itself, in order to ensure the capacity for long term recording and to provide robustness against unexpected survey downtime. But, by relying on progress taking place outside the petroleum industry, like the defense and space industries which have long experience in producing and employing robots with months of runtime, we think that battery and data recording units in the future will become sufficiently small to be carried by AUVs.

40 streamers and beyond?

Conventional 3D marine seismic surveys typically use a plurality of uniformly spaced streamers in parallel lines. The streamers range between 3 and 8km long, with receiver group intervals of 12.5m or 25m. The lateral or crossline spacing between streamers is typically 50-100 m. The relatively large spacing reduces streamer fouling due to crosscurrents and vessel course changes.

Ten years back in time, the maximum number of streamers that could be towed was around five. Today, PGS is capable of towing up to 21 streamers using their newest Ramform vessel.

This progress in streamer technology is expected to continue apace. In the future we could hope to tow 41 streamers with a 50m crossline separation. If this is combined with the new two-component hydrophone/geophone streamer (see GEO Expro No 4/2007) it might be that the future single vessel marine streamer concept can provide two-component, dense sampled receiver technology similar to today’s OBS systems used in the North Sea. There, the receiver area for one shot location (recall the reciprocity principle) going into the depth migration process is 8km inline by 2km crossline. Thus, OBS geometry, which we know produces superior imaging in the North Sea, can in fact be simulated by one streamer vessel towing 41 streamers with 50m separation. To simulate the split-spread geometry of OBS, the streamer vessel would need to traverse the sail line in both directions with streamers that are 4km long. The towing of 40 streamers and beyond is a major engineering challenge.

Subsalt imaging in ultradeep water will require an even larger crossline receiver spread than 2km, and thus a two-vessel operation will be necessary.

Marine seismic web?

Around 10 years ago Western Atlas suggested that a vessel could tow a lattice like a web instead of individual streamers to carry hydrophones . We believe this is an excellent idea that needs to be investigated. Every geophysicist knows that the acoustic wavefield is more accurately represented when the receivers are spaced closely together. However, geophysicists tend to close their eyes to this knowledge, and very seldom challenge the way we sample the acoustic wavefield.

Undersampling at higher frequencies has known disruptive effects on reflections from steeply dipping geologic formations such as the flanks of salt bodies. For this reason, the need exists for an improved marine seismic data acquisition system. The marine seismic web would be one such unique system for supporting seismic sensors towed in the sea behind a seismic vessel.

Another high-end use of the marine seismic web would be for seismic reservoir monitoring. By repeating the surveys with the web, one would not need to worry much about repeating the receiver positions between two surveys, simply because the acoustic wavefield is then well sampled.

Conclusions

Lasse Amundsen is Chief Scientist Exploration Technology at Statoil. He is adjunct professor at the Norwegian University of Science and Technology (NTNU) and at the University of Houston, Texas.Martin Landrø is a professor in Applied Geophysics at the Norwegian University of Science and Technology (NTNU), Department of Petroleum Engineering and Applied Geophysics, Trondheim, Norway.