Integration yields new tool, not a new technology

Fayetteville Model Has ‘Predictive Value’

In the fast-paced arena of new unconventional finds, the latest hype almost always is all about the latest discovery.

Look at the Fayetteville shale play for example.

Highly touted among the earlier discovered shale gas plays, its high profile status was soon eclipsed in the midst of the near-frenzy generated by these type plays as they rapidly proliferated across the United States.

But ascribing the Fayetteville a second class status also proved to be a tad premature.

Turns out there was – and is – much more to the story.

The U.S. Geological Survey’s 2010 assessment of undiscovered Natural Gas Resources of the Arkoma Basin Province and Geologically Related Areas estimates the potential undiscovered petroleum resources in two of the three Fayetteville assessment units to be 13.2 Tcf of natural gas.

The Fayetteville shale, which is geologically equivalent to the famed Barnett shale in Texas, occurs in the Arkoma Basin. It extends across northern Arkansas from the state’s western edge throughout the north-central region for a total aerial extent of 5,000 miles.

Shale thickness tallies 50 to 500 feet, and it’s found at depths between 1,500 and 6,500 feet.

Southwestern Energy Co. is credited with initially recognizing the economic viability of the Fayetteville shale in 2002, ultimately becoming the first operator in the play.

Today, the Houston-based company is the largest producer in the play with a leasehold topping out at 915,884 net acres and year-end 2010 production hitting 350.2 Bcf net.

Initiating the Integration

Once development was comfortably under way, a team comprising geologists, geophysicists and engineers decided it was time to knuckle down to better identify the intricacies of this shale, using an array of data combined with the company’s own 3-D seismic.

The ensuing one year-plus joint project, implemented via a Southwestern-Schlumberger partnership, wrapped up in early 2011.

John Jeffers
John Jeffers

“We undertook an integrated reservoir characterization project to model major factors believed to most influence Fayetteville shale reservoir producibility, and to determine whether the model could be sufficiently constrained to predict areas of higher production performance,” said AAPG member John Jeffers, director of geosciences for the Fayetteville shale division at Southwestern.

He noted the pilot project was based on the idea of using seismic data for reservoir characterization and integrating it with other types of geological and engineering data and carrying it through to reservoir simulation.

In fact, the study represents an important synergy between 3-D seismic and engineering data.

A reservoir model was developed via integration of all available well, log, petrophysical, sonic, image, core stimulation, production, microseismic data and processed 3-D surface seismic over a specific area.

The static reservoir model was used to history match the short- and long-term production performance and its variations across the exploration area.

History matching production profiles of multiple wells is a critical step toward understanding the key production drivers in unconventional shale gas formations, according to Jeffers.

Predictive Value

Southwestern holds about 1,400 square miles of 3-D seismic in this play, and the project focused on a single locale broken out into three study areas of 10 to 15 square miles, each in the heart of the company’s development area.

“This was intended as a pilot to try out techniques and see if it had value,” said AAPG member Jim Lemaux, staff geophysicist at Southwestern.

“The thing we were hoping to be able to do was build a reservoir simulation model that had predictive value that would let us predict the performance of wells not yet drilled,” Lemaux said.

“We validated it in this model by using blind wells and then seeing if we could predict the performance of wells we hadn’t used in the calibration,” he said, “and we were reasonably successful in doing that.

“We found that this kind of work – carried out carefully and constrained with the right kind of data – does have predictive value,” he said, “and can be used in helping us to determine how to develop an asset like this.”

The three chosen study areas were selected to assess how much data were needed to make a reasonably constrained model. One area was data rich, one not so rich and the other had only one well.

“We wanted to test the limits of predictive value of a seismically based model with varying levels of geologic calibration,” Jeffers said, “and also how much production time was needed to be incorporated into that model from existing producers to get a good history match.

“There was a reasonably good fit in a reservoir simulation world on all three,” he noted. “We were pleasantly surprised by the predictive value of the model.”

Multi-Purpose Data

Southwestern is known to be among the lowest-cost operators in the Fayetteville shale play, and Jeffers noted the study effort is the kind of work that allows them to be more selective in what they do and to continue developing even when gas prices are low.

The company’s 1,400-square-mile trove of seismic data alone represents a nearly $200 million investment, according to Jeffers.

The original reason to acquire the seismic was to use it for structural interpretation to plan horizontal wells, avoid faults and other challenges.

“Through time we realized the possibility of using that data to help us characterize reservoirs and predict well performance,” Lemaux said. “A large part of that was getting development drilling far enough down the road so we had well data to actually calibrate to because it’s difficult to make good use of seismic data for reservoir characterization in this kind of play without a lot of geologic constraint to it.

“At the same time, you need a lot of geophysical coverage to help you interpolate between existing wells,” he continued. “There’s a lot of white space between existing wells, and your seismic volumes help you fill that in with your best estimate of what those properties are doing.”

It’s not unusual to see geoscientists using seismic data in shales only to attempt to identify sweet spots, where to drill and where to skip over.

“Most of our acreage will be developed,” Jeffers said, “and we’re using the seismic data more to determine how to develop each area as it’s helping us to decide appropriate well spacing, completion technique and things like that.

“The high level message,” Jeffers said, is that this project was “all about integrating seismic data with other kinds of information rather than sort of a standalone geophysical project.

“It was not about investigating a new technology,” he said, “but more about integrating engineering data with geology with geophysics with production and completion data to come up with the best solution for how an area is behaving and how to most effectively develop it.”

The Southwestern team and project partner Schlumberger together brought the necessary know-how to the table.

“Schlumberger had skills and experience we didn’t have in reservoir modeling,” Jeffers said. “We had knowledge and experience in this play and operations, as well as the practical side of geophysical applications and field development practices.

“That partnership, the unique structure, was beneficial to the results.”

Comments (0)

 

Emphasis: Seismic