“Is the salt winning?”
Such was the question posed by keynote speaker Paul Singer of Total at a subsalt workshop held at the Colorado School of Mines in 2005. This was an unusual gathering in that presenters from oil companies and contractors had assembled to discuss their inability to successfully image under salt—rather than to applaud their successes, which is what happens at most industry conferences.
By the end of the workshop it had become obvious that salt, indeed, was winning, and that we needed a step-change in technology to improve our ability to image subsalt formations consistently. After all, several oil company geophysicists had made it clear that exploration success in deepwater Gulf of Mexico depended on our ability to image subsalt, and most presentations at the gathering suggested that vast areas of the salt basins remained un-imaged. The reasons for our inability to image subsalt in consistent fashion remained unclear. But the prize for whoever eventually managed to overcome this challenge was evident. “Image under salt and reap huge rewards,” said Don Spillman of Shell, thus providing the motivation to invest in new research and development.
Several suggestions were put forward to improve imaging, such as better velocity models, more accurate migration algorithms, multiple attenuation, and improved illumination with wide-azimuth data acquisition. BP presented modeling studies that indicated a step-change improvement in subsalt imaging could be achieved with wide-azimuth (WAZ) data. Modeling studies have shown that these new acquisition approaches provide an opportunity for improved subsalt illumination, signal-to-noise ratio, and attenuation of multiples, compared to conventional narrow-azimuth surveys.
Since then, several wide-azimuth-type surveys have been acquired, with results indicating that these types of surveys do provide a step-change improvement in subsalt imaging. The success of these surveys led the industry to acquire several large-scale multiclient surveys in the deepwater Gulf of Mexico salt basins over the ensuing years. WesternGeco was at the forefront in this effort, and now has the largest data library of WAZ surveys. The WesternGeco method of acquiring WAZ surveys with two recording vessels and two source vessels, known as the “2 x 4 vessel design,” has now become the industry standard.
Incremental enhancements in image quality have also come about through the development of processing algorithms like one-way wave equation (WEM), two-way wave equation (RTM) and true-azimuth 3D surface multiple attenuation (GSMP). These computation-intensive technologies are now routinely available due to cost effective computation capacity and fast algorithms.
In general the industry has made tremendous improvements in the quality of subsalt images over the last five years, however there are challenging areas for which imaging is still quite poor and needs improvement.
Progress has also been made in building better velocity models with azimuthal tomography, introduction of TTI anisotropy combined with other seismic and non seismic measurements such as VSPs, well logs, MMT, gravity, geomechanical stress fields and basin modeling. However interpreting and incorporating salt geometry into the velocity model is the weakest link in our ability to consistently image subsalt sediments.
Although the current wide azimuth data sets provide better illumination of complex salt geometry and subsalt data, they still fail to illuminate in many challenging areas—especially where steep dips appear.
The Next Chapter The industry now needs another significant improvement in illumination quality. This will be achieved through the acquisition of data at all azimuths and longer offsets instead of just wide azimuth at the current offset configuration of 8 x 4 kms. Current datasets are also lacking in low frequencies where the subsalt signal is strongest.
So the next step is to raise WAZ technology to a new level by implementing an acquisition design that can enable the recording of full-azimuth data with a 14 x 14-km offset range. Such a program, known as “Revolution,” was recently developed and introduced by WesternGeco in the Gulf of Mexico. Featuring a unique dual-coil design, the program uses sources and receivers towed at depths necessary to the optimization of low-frequency content.
Three-dimensional finite-difference (FD) modeling of the SEG advanced modeling (SEAM) model shows the illumination improvements obtained by full-azimuth 14 x 14-km data vs. wide-azimuth 8 x 4-km data. Early processing results from the “Revolution” survey confirm the improvements seen in modeling studies. These results have led our clients to underwrite the “Revolution 2” survey and further enquiry into acquiring a “Revolution 3” survey.
Because we are now using the two-way wave equation (RTM) for migration, the next step is to use the two-way wave equation for velocity model building also. One of the most advanced tools for velocity model building, using the two-way wave equation, is known as full-waveform inversion (FWI).We are currently performing FWI on a wide azimuth data set from the Gulf of Mexico. The project, known as “E-Wave,” covers over 30,000 square kilometers and initial results are very encouraging. It appears to be clear that FWI is able to delineate salt reflectivity starting with a sediment velocity model derived from conventional tomography. Several clients have provided pre-funding for this project based on results obtained over the pilot area.
The combination of wide-azimuth data and two-way wave-equation migration algorithms such as RTM, along with true-azimuth 3D surface multiple attenuation, provides significant improvements in salt flank and subsalt imaging compared to conventional narrow-azimuth datasets.
Additional improvements, especially in delineating salt geometry, can be obtained by building velocity models using FWI. Modeling studies suggest that even further improvements in image quality are possible by acquiring full-azimuth data at long offsets with enhanced low frequencies. Current surveys using the novel dual-coil acquisition design will provide data with which to test that assertion.
New surveys using the dual-coil acquisition design will provide full-azimuth data at long offsets. This, combined with enhanced low frequencies, will facilitate the performance of FWI for velocity model building, GSMP for multiple attenuation, and RTM for imaging.
So is the salt still winning? In fact, the war for clarity continues, but we’re making progress, and the more technological ammunition we can stockpile over the coming years, the better off we’ll be.