Chances are, many more reservoir problems could be solved if time and money were no object.

But oil companies don't have that luxury. They need reliable answers quickly and inexpensively. And over the years that has often meant short-cuts, assumptions and guesstimates that have called into doubt the quality of the finished product.

Two major technical advancements are helping to change that - the integration of seismic information with the other geoscience disciplines (geology and petrophysics) and the visualization technologies that allow for the quality control (QC) of this work. With computationally intense procedures taking a mere fraction of the time they took 5 years ago, the process of reservoir characterization can utilize much more of the data that's being acquired and find meaningful relationships within the data that lead to better reservoir descriptions and decisions. And with visualization centers bringing a variety of ideas together in one room, it's much easier to integrate data and expertise to develop innovative workflows that result in a more realistic conclusion.

At companies like Veritas Geophysical Corp., this has meant that 3-D seismic has the chance to be viewed as more than an exploration tool that's acquired in the field, processed and handed off to the client. But officials at Veritas realized the company didn't have all of the internal expertise needed to move into advanced reservoir characterization. In 1998 it formed Veritas Exploration Services (VES) to act as a consulting service to oil company clients who had complex reservoir problems to solve. And it hired a range of disciplines, from geologists to petrophysicists to reservoir engineers, many of them experienced oil finders prepared to provide the range of expertise needed to solve these problems.

The methodology

Using its visualization centers to interactively work with the data, VES consultants have developed solutions to make a meaningful correlation between seismic data and geological, petrophysical and engineering information.

"Traditionally, reservoir characterization has been done by interpolating between available well control based on geologic trends," Robert Hobbs, president of VES, said. "Because we have the power of 3-D seismic, we can relate the seismic traces in between the wells to arrive at an interpretation of the geology that has the higher spatial resolution of the seismic."

For instance, in a field in the Gulf of Mexico, the client was trying to get a better understanding of what was turning out to be a very complex reservoir. VES consultants were able to relate seismic attribute information to the presence of sand and fluid types. By combining well information and the seismic attributes, the extraction of reservoir parameters was achieved at a much higher spatial resolution, preserving the true heterogeneity of the hydrocarbon accumulation.

"When you drill a well, you need to be able to predict what's away from that well," Hobbs said. "We geostatistically make that connection between the geology and the seismic to fill in the gaps."
Brad Bankhead, vice president of reservoir technologies, added, "If you make a discovery in the deepwater Gulf of Mexico, key to the economic success of this discovery is determining the appropriate facility design that will be required to extract the hydrocarbons. That makes a big difference in the cost of the total development. So being able to more accurately interpolate the reservoir properties at a higher resolution between the wells is a huge advantage."

VES utilizes seismic data tied to wells to extract rock properties using geostatistical methods. Bankhead said the seismic response is modeled against the well logs to look for correlations between attributes and rock properties. This can be done multiple times with multiple attributes and reservoir properties very rapidly. In one example, the team modeled porosity against multiple seismic attributes and found acoustic impedance to yield the best relationship. The regular seismic reflectivity data didn't highlight the productive stratigraphy, but by integrating the acoustic impedance, an important channel system was made evident in the data. "It's actually a very complicated depositional system," Bankhead said. "With this type of complex stratigraphy, how are you going to interpolate rock properties in between wells without integrating the seismic attributes?"

One of the major benefits to this type of workflow is that it enables interpreters to combine the spatial resolution of seismic data with the vertical resolution of well log data, called a stochastic inversion. Once the correlation is discovered between the seismic attribute and the reservoir property, that correlation is interpolated throughout the rock volume. Then a log-scale property can be generated at every seismic trace. Seismic wavelets are then combined to create a synthetic volume.

"We don't do that one time," Bankhead said. "At every trace we'll generate several hundred synthetic traces. The synthetic trace that has the highest correlation is the one utilized in the final interpretation."

The power of processing

Bankhead said that one of the advantages of housing a reservoir characterization group within a larger geophysical contracting company is the access to the seismic processing group. "One of the things we do a lot of is reprocessing," he said. "If seismic data is not processed appropriately, it's hard to extract rock properties from it. So we spend a lot of time with our processors actually calibrating well data during the reprocessing phase."

Hobbs added, "VES could be an independent consulting company. But because we have strong onsite seismic processing expertise, we can optimize the processing results, geared specifically to reservoir parameter extraction from the seismic data. If we understand the rock properties, understand the seismic response to those rock properties and understand the key reservoir problems, then it's easy to iterate because we have this seamless organization within Veritas, including acquisition. Now we're looking at reservoir problems and designing acquisition programs to solve those specific problems."

This results in significant timesavings. A recent 350 sq mile (900 sq km) exploration 3-D job for El Paso in the Santos Basin of Brazil was shot, processed and interpreted, with well locations chosen, in 7 months. "We had the deepwater people from El Paso in our visualization center, and they completed their management reviews alongside our seismic interpreters," Hobbs said. "The initial exploration test of the prospect resulted in a significant gas discovery on a new trap style that before then had been untested in the Santos Basin. They've now drilled a delineation well off of our seismic data and interpretation work, and I think they've upgraded that find to almost 500 Bcf.

"The fact that we were able to merge acquisition, processing and interpretation in a seamless fashion enabled that client to have a completed project in a very short period of time."

A changing mindset

"It's a bit of a paradigm shift for our company in that we are positioning ourselves to be problem-solvers to the E&P industry, and there are challenges," Hobbs said. "One of the challenges is that there are things we do that oil companies aren't quite ready to let go of yet on the basic interpretation side. We're working through those. But some oil companies have gotten past that and recognize that we have expertise here that they may not have on their own staff."

Moving seismic from an exploration tool to a drilling and development tool also has its challenges. Bankhead said the cost to do a full-scale reservoir characterization is minimal compared to the money being spent on offshore developments. But, reservoir engineers have often not recognized the value seismic offers to the reservoir modeling workflow.

"It seems logical to VES personnel to utilize the vertical resolution of well logs integrated with seismic data to aid in the interpolation of reservoir properties in the inner well space," he said. "It's taken awhile, but many of our clients now recognize the value of seismic data integration in the reservoir modeling process." The increased spatial resolution the seismic data brings to solving the reservoir heterogeneity problem is of enormous value to the modeling workflow, he added.

But to put all the pieces together requires input from a variety of disciplines. "Reservoir characterization is truly an integration of all of the disciplines," Hobbs said. "You need engineering, geophysical, geological and petrophysical expertise. Utilizing and combining the data and interpretation from each of the disciplines in a logical manner to solve complex reservoir problems is the workflow we utilize to conduct reservoir characterization projects."

Comments