Today's soaring commodity prices are contributing handsomely to the bottom line at the various oil and gas industry-related companies.

Regrettably, the extra dollars don't make it any easier to wring more production out of the reservoir. Indeed, those always-complex hydrocarbon-bearing intervals continue to challenge the most talented oil finders - in both the good times and the bad.

It's all about the reservoir model, for the most part.

Despite all the data being thrown into model building, to meaningfully decipher the real world in the subsurface demands much more. In fact, when you get right down to it, the models used as the basis for so many critical decisions throughout the drilling and production process might well be called educated guesstimates.

As drilling becomes increasingly costly and complex, particularly in tough environments such as the North Sea and the deepwater Gulf of Mexico, technology must rise to the occasion. In other words, the reservoir model must be built to depict reality - which won't be cheap or easy.

But there's good news already: A team comprised of seasoned industry players just kicked off an aggressive project to pursue this goal and, simultaneously, to provide a software tool that will behave as a cross-discipline "meeting and control room."

It's called FieldWatch, and it's a joint effort between Roxar, Statoil and the Norwegian Computing Center (NCC). Both Statoil and the Norwegian government are providing financial support via the PETROMAKS research and development program. The project is designed to develop a software tool for right-time integration and application of production data in reservoir modeling and sweep analysis to foster multi-disciplinary collaboration and maximize reservoir performance.

"One target of the project is to close the loop between the subsurface disciplines and the process disciplines," said Roxar's vice-president of technology, Andres Hatløy, who was the company's key driving force behind the FieldWatch program.

"FieldWatch will provide the opportunity to work together in the same environment with all available data at their disposal to provide hints and actual facts on reservoir behavior," Hatløy noted. "This will enable them to make the right assumptions to improve and update the reservoir models to ensure better production forecasts."

Norway historically has been a staunch supporter of new technology development.

"The PETROMAKS projects are funded by the Norwegian government, which asks that you come up with a concept that will effectively improve the Norwegian economy by improving oil recovery," Hatløy said. "At least one industry sponsor, which is Statoil in this case, needs to support the vendor's application for funding because the government wants to ensure what you research and develop is commercially viable, and they want to know you are partnering with someone who will use it in the end."

Maximizing reservoir performance could pay off big-time for Norway. An increase in recovery of only 1% would have a value of approximately US $78.4 billion, according to Sandy Esslemont, chief executive officer at Roxar.

"Given our long relationship with Statoil, we're aware of where the market is going from their point of view and our other customers," noted Daryl Gunn, product manager for production management solutions at Roxar. "We knew the industry's need for tools to integrate production data and reservoir data to increase recovery efficiency and optimize production was especially important in light of forecast increases in operating costs. The digital oil field, or the automation of processes and equipment used to monitor and manage your reservoir, is one way the industry sees to tackle these escalating costs."

"The more real-world data you can get into the model, the more likely you are to get it right faster," Hatløy noted.

Companies spend considerable money acquiring production data, e.g. the meters and gauges used to accomplish this carry a high price tag. But the experts agree there's no easy way to get these data into the reservoir model, which ultimately provides the foundation for all decision making.

"One of the big challenges unsolved in the industry in making 3-D models is how to include production data," said Petter Abrahamsen, research director of Statistical Analysis of Natural Resources (SAND) at the NCC. "This could be historic production data or other kinds of production-related information. Our part in the FieldWatch project is to do numerous enhancements and modifications in the algorithms in Roxar's IRAP RMS software that will enable it to use these data. In fact, the 3-D modeling software modules that model properties in RMS were originally constructed by NCC."

To make a 3-D model of the reservoir so it performs like the real world is very difficult mathematically, Abrahamsen noted. Production data are very ambiguous, so it's impossible to know exactly where the fluids are flowing. Production often is measured only at the top side, i.e., when the hydrocarbons arrive at the platform, and therefore, it represents commingled fluid flows rather than production from each perforated interval, which would provide hard evidence for reservoir quality and flow behavior.

"It's hard to pinpoint the exact 3-D properties of the reservoir from those kinds of vague yet important measurements," Abrahamsen said. "Even though you're able to make a number of different models that perform perfectly according to data already collected, each one of them might behave differently in the future - and what we're aiming for is to be able to predict what will actually happen in the future. By continuously having access to real-time data, the reservoir models can be progressively conditioned to the production data as they arrive, helping to pinpoint what the reservoir really looks like."

Engineers are known to spend years performing history matching, changing the productivity of the wells, the reservoir quality, flow communication patterns, etc., until reaching a better production history match. Yet the results usually look rather artificial.

"Their power for prediction, for forecasting, is usually very poor," Abrahamsen said. "So what we're trying to do is merge geologically sound models that are in accordance with the history data. If we can incorporate production into the model, we will get a model that will be more accurate in predicting the future.

Not surprisingly, there's tremendous enthusiasm emanating from Statoil about the FieldWatch program.

"We support this project because it is aligned with our vision of becoming a leader in utilization of integrated operation concepts, and R&D to develop better methods and tools is an important part of our efforts in this area," said Peter Eilsø Nielsen, senior advisor of Integrated Operations, Statoil. "The project is also a natural extension of the cooperation between Statoil and Roxar on improved reservoir modeling methods the last 15 years."

Indeed, this is the kind of effort that has the potential to give a company a leg up over its peers as it reaps the benefits of a new technology early on.

"The idea behind sponsorship is you get a say and therefore are more likely to get features you particularly want moved up the priority list," Hatløy noted. "Also, there's the advantage that you understand the technology from the beginning and start to gear your working practices to using that technology. You know how it will fit in right away."

There's considerable industry interest in FieldWatch, he added, including:

• Fewer applications needed to establish and maintain an efficient reservoir exploitation and production optimization workflow;

• Less time required to update reservoir models when new data become available;

• Cross-disciplinary work environment encouraged by having a common tool to share reservoir competency and knowledge;

• Improved enhanced oil recovery and injection management via better sweep monitoring and analysis; and

• Faster, improved reservoir communication analysis to speed the search for in-fill targets.

"At the end of the day, the core of the e-field requires you have a reservoir model that uses all the available data and reflects the real-world situation," Gunn noted. "Every subsequent step in the workflow is using that information to make decisions.

"If the core is wrong, it doesn't matter what you add."