SPECIAL REPORT: POINT OF VIEW: New SEG chief describes seismic methods lifting Forties field oil flow

Oct. 4, 2010
Modern seismic methods support a "drilling machine" raising oil production in a UK North Sea field that, measured by discovery date, is just a decade shy of qualifying as an antique.

Modern seismic methods support a "drilling machine" raising oil production in a UK North Sea field that, measured by discovery date, is just a decade shy of qualifying as an antique. Not all new seismic tools apply to Forties field off Scotland, says incoming Society of Exploration Geophysicists Pres. Klaas Koster. But new methods that do apply enhance each of the main steps of seismic work—acquisition, processing, and interpretation—and are crucial to maintaining production from the old giant.

As exploitation manager for Apache North Sea Ltd., Koster managed the recent acquisition and processing of about 300 sq km of time-lapse 3D seismic data over the field. His team soon will begin interpretation to assess reservoir changes that have occurred since the last 3D survey over the same area, shot in 2005.

Quantitative interpretation, especially of time-lapse data, is Koster's specialty. At the start of his career, he says, "time-lapse was not at all routinely applied." Now the method, also called 4D, is an important tool of reservoir management.

"I have seen the recognition of the value of time-lapse, the whole development of the criteria that are necessary to make it successful," he says. "That would be the one big, big change in quantitative interpretation that has occurred in my 20 years in the industry."

Another important change is development of the computational capacity needed to apply new methods.

"A lot of the fundamentals of how to do quantitative interpretation were long established before I entered the industry," Koster says. "But the ability to apply them to large 3D volumes with many wells has only been possible with increasing compute power."

Advances like those have been critical at Forties field, of which Apache North Sea became operator when it purchased BP's 97.14% interest for $630 million in 2003. Field production at that time had fallen to 40,000 boe/d of oil and gas. With Apache as operator, monthly average production has been as high as 71,000 boe/d. At its peak in 1980, the field produced more than 500,000 b/d of oil alone.

Assembly line

The operator is keeping three drillstrings busy continuously, an activity level that in turn keeps the geophysicists busy.

"The pace is close to having a target drilled every other week," Koster says. "We have to come up with a new target every other week." Because Forties is a mature field—discovered in 1970, on production since 1975, and now producing water at nearly six times the rate of oil flow—"the moment you stop drilling it immediately starts going into decline. That translates directly into your bottom line. It's an assembly line."

Little time passes between the identification of a drilling target and when that well reaches total depth. A reservoir model continually updated on the basis of drilling results generates information for drilling decisions. With decades of production history from 300 wells, "the amount of data that we have there, quite apart from the geophysical data, is staggering," Koster says.

Career highlights


Klaas Koster's first job was with Amoco in Tulsa, where he initially worked on multicomponent acquisition. He subsequently joined a team of Amoco and Pemex geophysicists to find solutions for some of Mexico's seismic data quality challenges. His next assignment was at Amoco's Denver office, where he worked on 3D acquisition and processing. He also worked with Schlumberger colleagues on early applications of dipole-sonic methods. Koster started working on quantitative interpretation and seismic inversion in 1994 for Shell, covering fields in countries that included Greenland, Nigeria, Peru, and New Zealand. In 1997, he moved to Norway, where he was part of the multidisciplinary team responsible for time-lapse monitoring of Draugen field. In 2000, Koster joined Woodside, where he became head of quantitative interpretation and worked frequently with geophysicists at Curtin University and Commonwealth Scientific and Industrial Research Organization of Australia. He became head of subsurface technology implementation with Shell in New Orleans in 2004, leading a team that also covered geomechanics and geochemistry. He joined Apache recently in Aberdeen as senior technical advisor responsible for subsurface activities in Forties field.


Koster holds a PhD in geophysics from Delft in The Netherlands.


Koster joined SEG in 1986. He was elected president of Australian SEG in 2003 and vice-president of SEG in 2004. He was awarded life membership of SEG in 2008 and is the recipient of a best paper in The Leading Edge award. He has served as chair of the Governance Review, Constitution and Bylaws, International Showcase, and Meetings Review and Planning committees.

The pressure to supply targets to what he describes as "that drilling machine" means interpretation of the time-lapse data newly recorded and processed must occur quickly.

"The uptake of the results is going to be almost instant," Koster says. And technical improvements since the 2005 survey should mean unprecedented quality of data and their interpretation.

Acquisition advances

An acquisition method used in the recent survey that wasn't available in 2005 involves steerable sources and receivers. In the earlier survey, "streamers were still relatively dumb pieces of kit that you just towed behind the boat," Koster says. "Sources pretty much ended up where they ended up."

The ability to more accurately control shot-point and receiver locations serves a vital factor of success in time-lapse surveys: matching those locations from one survey to the next. Geophysicists call that repeatability.

The new capability provided for a reduction in the source-crossline repeatability error from about 10 m in 2005 to less than 3 m in the 2010 survey. It cut the crossline-receiver location error by half by doubling the amount of streamers, which were towed at half the former spacing. Boats in the earlier survey towed six streamers 100 m apart; in the recent survey they towed 11 streamers at 50 m separation.

"Because those streamers are now accurately controlled," Koster adds, "we were able to sail the boats safely close to the platforms." Forties field has five platforms. The need to give them wide berth created "huge holes" in data from earlier surveys, Koster says.

Processing upgrade

Processing improvements, empowered by modern computers, allowed Apache to upgrade the product of the new survey. In 2005, the company used prestack time migration to mitigate positioning errors in recorded reflection data. The result was a representation of the subsurface in which vertical positioning of reflection events is a function of sound travel time.

In the new survey, Apache used anisotropic prestack depth migration. The digitally intensive method accounts for anisotropy—basically, the variation in velocity of sound propagation through the subsurface that occurs with changes in direction of travel. At Forties, the velocity of energy traveling horizontally differs from that of energy traveling vertically. The difference results in an elliptical wave front. Accounting for anisotropy during seismic processing improves the images of faults and lateral discontinuities.

The more-sophisticated migration technique used in the new survey served the need for speed in part because it delivered migration results in depth rather than time and also because much of the processing occurred on computers aboard boats of the contractor, Western Geco. Koster says about 2 months passed between firing of the last shot in the survey and delivery of processed data to interpreters' workstations.

That's critical because the saturation of oil inside the reservoir is always changing. Interpreters need to know the subsurface situation "today or yesterday, very close to when you make your drilling decision," Koster says. Because interpretation involves the continually updated reservoir model, it's important to have the migration output in depth.

"We cannot afford to stay in the base domain of the seismic data, which is time," Koster says. "It's all interpreted in depth. But further than that, it's all interpreted in the reservoir model domain."

Prompt processing and interpretation in depth will support the rapid generation of drilling targets essential to maintaining and raising Forties production. Into the reservoir model, Koster explains, "we plug in real-time drilling results, accurately update the reservoir model, and have those updates influence where we pick our next targets."

Identifying changes

While the techniques for acquiring, processing, and interpreting time-lapse data improve steadily, the basic aim remains to identify changes in seismic responses between sequential 3D surveys. Those changes indicate variations in acoustic impedance of a reservoir that occur as water replaces oil through production. Acoustic impedance describes a rock layer's ability to accommodate sound propagation. Boundaries between strata of different acoustic impedance are what reflect sonic energy in seismic work.

Observations made in this way about changes in oil saturation allow Apache's interpreters working on Forties field to spot attic oil and undrained compartments.

"Those essentially are the targets we go after," Koster says. Over Forties, 3D surveys before the most recent two were shot in 1988, 1996, and 2000.

Some time-lapse surveys look beyond changes in acoustic impedance, mapping areas of varying pressure response. That hasn't happened at Forties.

"That's more something we'll be revisiting in the coming months," Koster says.

Knowledge transfer

Asked about scientific development in the geophysical industry, Koster describes an acceleration of "knowledge transfer" that's reminiscent of his "assembly line" analogy in the generation of Forties field drilling targets.

"Since the industry is truly global," he says, "the information gained by application in one play is soon enough disseminated throughout the industry." SEG, he adds, is critical to that transfer.

"Like most people in most businesses nowadays, one of our most precious commodities is just time. The ability to get training or knowledge transfer happening where and when you need it as geophysicists is tremendously valuable," he says.

He therefore welcomes the ability to deliver content electronically, which helps SEG transmit knowledge beyond its formal meetings and workshops.

"You now have the ability to say, 'Here's my new project. I need to find out about technology such and such.' You can go online and see what's being offered by SEG in terms of training courses" or other information, Koster says, adding that a geophysicist needing training can take the course at his or her own pace. "You're completely in control as the end user of the information."

The new SEG president also welcomes a growing interest within industry in sharing the risk of "research that is deemed important but not believed to give any one company enough of an advantage to justify the development of propriety technology." He expects the trend to broaden SEG's role.

"Research partnerships between companies, with or without government participation, brokered by SEG, are a new development that the SEG intends to pursue further," he says.

More Oil & Gas Journal Current Issue Articles
More Oil & Gas Journal Archives Issue Articles
View Oil and Gas Articles on PennEnergy.com