Seismic is only as good as the answers it provides. A new way of thinking might squeeze a little more useful information out of those squiggly lines.
If you're a geophysicist, bear with me - I'm going to have to start spouting blasphemies here in a moment to discuss a different philosophy and how it might impact the use of seismic to generate a geologic model.
Ready? Here goes. Suppose we threw out all of our suppositions about the linearity of the subsurface and embraced the fact that it's a highly nonlinear environment? Well, sure, that brings up all sorts of problems, like we have to throw out most of the assumptions that have gotten us this far in our seismic processing and interpretation. But might that nonlinearity also give us some new insights?
Tawassul Khan, chief executive officer of Nonlinear Seismic Imaging Inc., is exploring that very possibility. With a career spanning more than four decades, Khan has a breadth of experience, in operations and in research, and in operator and service company capacities. Nonlinearity, he said, is a concept of a single input having multiple possible outputs, meaning it's impossible to predict the reaction. "If we look at it from one point of view, it's a nuisance," he said. "The mathematical models that we have cannot address the nonlinearity.
"On the other hand, a nonlinear system has certain properties that can be utilized to better understand the rock properties. The sensitivity of the nonlinear response of the reservoir rocks due to porosity, permeability and microfractures is larger than that of the standard linear measurements of the seismic velocity, attenuation and modulus."
Khan maintains that understanding rock properties might be the missing link that will make enhanced oil recovery (EOR) economically justifiable. "Industry needs something," he said. "The report put together by (Vice President) Cheney and (President) Bush pointed out that anywhere from 30% to 70% of oil and 10% to 20% of natural gas is not recovered in field development, and it recommends that the secretaries of energy and the interior promote oil and gas recovery from existing wells through new technology."
He agreed with Dr. W.L. Fisher at the University of Texas at Austin that EOR efforts have proven costly and mostly unsuccessful. "The limitation is not due to extraction techniques but due to the inability of accurately mapping the geologic architecture of the complex reservoirs," he said. "We do not have the technology that can map the reservoir geology with the accuracy that we need. There's a lot of ambiguity in the work we do today. There's no unique result from it. We're hoping that this technology will reduce that ambiguity."
As it turns out, the types of rocks that oil companies are most interested in - those that have high porosity, permeability and fractures - are more nonlinear than their shale or clay neighbors. "The very fact that a rock is porous or permeable means that it will be nonlinear," he said. "There are millions of little springs connected together, and each is going to be different because the composition of the rock is not orderly."
Khan has developed several patents to use his theory to determine porosity and permeability and better define fractures.
While the methodology is somewhat complex, the overriding idea is simple; in fact, he compared it to a patient drinking a barium cocktail before an X-ray so that only the organs of interest will show up in the image. By using a method that seeks out only the nonlinear formations, the method effectively filters out areas of little interest and highlights areas of greater interest.
It's all done through frequencies. In one model, that of fracture detection, a downhole source emits a frequency of 1,000 Hz while a surface vibrator emits at a much lower frequency, in the 10 Hz range. "We have two signals," Khan said. "When we have no fractures, these signals do not intermix or affect each other. When we have a fracture, it gets squeezed by the lower 'modulation' frequency, which results in amplitude changes of the higher 'carrier' frequency signal. The changes in the higher frequency amplitude and its harmonics directly relate to the compression and the rarefaction cycles of the lower frequency."
In another example, related to the reservoir porosity, "It's the interaction of the two frequencies, and that interaction is going to be prominent only in the rocks that are porous, permeable and fractured."
Khan's patents include fracture mapping and determining rock properties and permeability. He also hopes to apply for a patent on using the same methodology to determine the viscosity of the fluid in the rocks, enabling engineers to determine whether it's gas or oil in the formation. This will have particular benefits in 4-D seismic studies, he said.
Khan's plan is not to invest in seismic acquisition equipment and start shooting nonlinear surveys but to offer the intellectual capital he has developed to a service company that can put it to use. If his methods truly offer a unique interpretation of the subsurface, it shouldn't be a hard sell - the process uses the same equipment as a standard land seismic survey, and the data processing doesn't take significantly longer.
"We're not replacing anything; what we're doing is adding on," he said.
His hope is that this new information will answer some of the questions being asked about the reservoir by petroleum engineers. "They're interested in the flow mechanisms," he said. "If there's a fault, they want to know if it's a conductive fault or a seal. If there are fractures, they want to know their exact location and orientation. Those are things geophysicists cannot answer. And if we answer, sometimes our answers are not the right ones.
"We have lost some of our credibility, and we have to work to get it back."

Comments