An analysis and opinion essay predicting the decline of the Bakken Shale play by a recent Forbes contributor with a master’s degree in petroleum geology has pushed some buttons among North Dakota energy officials, who have questioned the soundness of the author’s reasoning.

“The decline in Bakken oil production that started [two years ago] is probably not reversible,” wrote Art Berman, who claims to have 37 years of oil/gas experience and numerous interviews with major U.S. news media over the years. Berman has it all wrong, according to Department of Mineral Resources (DMR) Director Lynn Helms, North Dakota’s top oil/gas official; and Justin Kringstad, director of the state Pipeline Authority for the nation’s second-largest oil producing state.

Kringstad will be releasing his own study on well economics later this year. In the meantime the North Dakota legislature is expected to begin releasing projections for future state revenues based on oil/gas production forecasts. The lawmakers rely a lot on Helms’ and Kringstad’s work when they make those calculations every two years.

In responding to Berman’s analysis for Forbes, state officials found inaccuracies in his assessment of well performance, economics, gas/oil ratios (GOR) and water/oil ratios (WOR), a DMR spokesperson said. “We want to put what we feel is the most accurate information out there for decision makers,” she said.

The headline on the negative analysis is “The Beginning of the End for the Bakken Shale Play,” something Helms in the past has characterized as being targeted for about 20 to 30 years from now. In 2009 Berman published a pessimistic outlook for the Barnett Shale, estimating that only around 10 Tcf would be produced from the basin, and of that, close to 7 Tcf — 70% — would be noncommercial even at $7/Mcf gas prices.

Through 2016 the Barnett has produced nearly 20 Tcf and was continuing to produce over the last year at a rate of 3.8 Bcf/d, according to Railroad Commission of Texas statistics. Barnett gas was selling at about $2.63/MMBtu Thursday.

A combination of steadily improving well performance and continuing reductions in breakeven costs has pointed to the Bakken continuing to evolve as a more robust U.S. shale play as opposed to it facing the “beginning of the end,” Helms said, adding that Kringstad’s approach to measuring well performance year over year using “raw data” is superior to the approach used by Berman.

Helms cited Berman’s use of WOR as a source of concern for the Bakken as being off base. “The water in oil drilling is largely coming from the Three Forks formation drilling in a deeper zone under the Bakken, and Three Forks is somewhat connected to even deeper zones containing more water,” he said.

“The Three Forks has 150% of the WOR of the Bakken, so what we are seeing in the trends is not well depletion of the middle-Bakken, but more and more drilling of wells into the Three Forks and the production of more water from there.”

Kringstad’s forecasts show the Bakken oil output growing steadily through the mid-2030s between 1 million and 1.8 million b/d, charting conservative and robust growth cases for the shale play.

“In response to Berman and as part of this week’s release of data in North Dakota, we wanted to make sure that the most current information was part of the mix,” Helms said.

Kringstad’s well performance data shows that since 2011, performance has improved steadily on a per-well basis through 2016. “This well performance data is critical for the midstream operators’ to develop plans for new pipeline and processing infrastructure,” he said. Midstream operators keep close tabs on both oil and natural gas performance.

“Natural gas has been even more pronounced year over year in terms of improvement in how wells are performing,” Kringstad said. “The big question is can the industry continue this march forward?” His data so far show that it can.

With activity in the past two years being concentrated on the Bakken’s core including McKenzie County, Kringstad examined seven fields in the northern end of that county, and found well performance was about 25% higher last year than it was a few years earlier. “When you are talking multiple wells, high-performing wells in that area, the 25% increase means a substantial additional amount of gas needs to be moved out of the region,” he said.

Since 2012, Kringstad said that the gas-to-oil ratio has continued to increase each year as there are more mature wells in the Bakken mix. “We expect to continue to see that as we see wells continue to mature,” he said. “As reservoir well pressures decrease and with increasing production, the gas-oil ratio is going to continue to increase. This is well known in the industry and something we expect will continue to develop over time.”

He thinks this also has significant implications for additional gas pipeline and processing infrastructure in North Dakota for years to come.