By LOUISE S. DURHAM
EXPLORER Correspondent

30-Year-Old Algorithm Back in Vogue

Computing Muscle Revives WEM

Related Stories: Wave Equation Migration (WEM) 3D Visualization AAPG Digital Pubs Update aapg.org Online Maps (May05)

Graphics courtesy of TGS

Click to enlarge examples

It was a long time coming, but the disconnect between soaring commodity prices and languishing oil company stocks appears to be history -- at least for now.

Still, it's a pressure cooker out there, and companies are producing all out and scurrying to find the hydrocarbons to replace what's being produced -- a more daunting mission than ever before with the bulk of the "easy" finds already claimed.

The biggest challenges for many oil finders lie deep below the waters of the Gulf of Mexico, where numerous massive salt bodies wreak havoc on the travel path of seismic waves as they propagate through the earth during the seismic data collection stage. Indeed, these salt features taunt the most knowledgeable geoscientists as they seek to acquire the ultimate image of the reservoir in the complex subsalt environment.

"There's increasing interest in the subsalt among the companies, and without decent imaging techniques, it's basically impossible to see anything there," said Colin Murdoch, senior vice president of data processing at Veritas DGC.

"The deepwater Gulf of Mexico is the driving force behind some of the imaging techniques currently used," Murdoch said. "Also, the drastic decrease in the cost of computing -- not just clusters, but all the surrounding technology -- is letting us do things people have wanted to do for a long time but couldn't."

The tendency of oil companies to farmout most of the research work to the contractors plays a key role in advancing seismic imaging, according to Jerry Young, vice president of imaging technology at Veritas.

"There is a lot more revenue from the oil company to the vendor that's driving the whole market forward," Young said, "especially in the deepwater Gulf of Mexico subsalt imaging."

Pass the Salt

In the forefront of the imaging technologies in vogue today is wave equation migration (WEM), which has been around for close to 30 years but without the computing power necessary to use it until the past five years or so, Young noted.

The WEM algorithm does a better job of imaging beneath the salt as opposed to Kirchhoff migration, which is quite good above salt, according to Pete Bennion, vice president at TGS Imaging.

"The ability to do wave equation 3-D depth migration where we can output pre-stack gathers as well has been a tremendous step forward for the industry," Bennion said. "It's where the market is headed.

"Most of our big data sets will go through wave equation 3-D depth migration," he continued. "In fact, the industry is heading toward doing time and depth processing together on new data sets."

One of the big issues in subsalt imaging is how well velocity models can be derived.

"As an industry, our ability to image far exceeds our ability to derive velocity models from the data," said Nick Bernitsas, senior vice president at GX Technology. "Going forward, the emphasis should be on improving our tools for velocity model building so they're more accurate and have higher resolution."

"Depth migration requires high resolution velocities," Murdoch noted. "It requires you to know the answer before you run the migration."

Other key technical issues specific to subsalt noted by Young and others:

• How well multiples can be removed -- Veritas is pushing the use of 3-D SRME (surface related multiple elimination).
• There is a need for various flavors of WEM having differing strengths and weaknesses.
• The impact of the acquisition scheme, particularly regarding illumination issues that might exist, i.e., the seismic energy didn't hit the target or, if it did, it wasn't able to get back to the surface to be recorded by the array of receivers deployed at the time.

Some folks who think this might be a problem are considering the potential for wide azimuth surveys to provide the answer, Murdoch noted.

Challenges Remain

Despite the widespread use of WEM technology, it's not without flaws. In fact, there's much work yet to be done.

One of the problems with wave equation migrations is they are dip-limited in general -- a big downside in salt territory, where events such as near-vertical salt flanks are commonplace. Advances in technology are needed to handle steep dips, particularly those that exceed 90 degrees.

Among the companies working to address this issue is Applied Geophysical Services, which has developed a proprietary 3-D prestack depth migration algorithm akin to the beam method.

"It will handle steep dips beautifully where wave equation doesn't," said Kevin Sherwood, vice president processing. "It's different from all others -- we have the best parts of Kirchhoff and wave equation in the algorithm to image steep dip and overturned dip and also get very good subsalt response."

Still, the consensus among the experts is there's no panacea at the moment.

"Sometimes you just can't define the shape of the salt body in the Gulf of Mexico very well," Bernitsas said. "Salt distorts the signal, and sometimes you can get it unraveled very well and sometimes not at all.

"I don't think anyone thinks the problem will be solved 100 percent of the time, because there are limitations imposed by Mother Nature," he noted. "But there are things on the horizon that suggest we'll do a better job going forward."

Efforts to overcome the complicated subsalt imaging problems in the Gulf must encompass much more than migration algorithms.

"It's getting the right mix of a number of different technologies working all at once, from angle gathers to visualization systems," Murdoch noted. "There's a whole series of bits and pieces to have in place and working effectively.

"Some people give the impression wave equation appears, and all problems are solved," he said, "but it's just a stage in the technology."

There may be a whole big bag of tools, but companies better have the people to use them, Murdoch cautioned.

"It's a huge issue, having the interpreters who can do a decent job on doing salt," he said. "If the model you build is no good, the imaging algorithm won't put the seismic data together to give a clear picture of what's going on.

"The biggest limiting factor for imaging is having enough highly skilled qualified people to perform the difficult, complex work to get a good result."

"The people, the technology, the know-how, the work-flows are all important," Bernitsas noted. "The demands on the data and the processing have to go hand in hand; you can't have sloppy processing and extract amplitude attributes that make sense."

There's a whole series of processing techniques that go into the data prior to the migration and algorithm stage. If the data are not prepared correctly or sampled correctly the migration techniques won't provide a good result.

"This often gets missed by people these days because it's nicer to talk about the more glamorous components," Murdoch said. "To work those other algorithms that come before the migration stuff is critical to providing a good image."