Did Industry Consultants Hijack the Stock Assessment Process for Atlantic Menhaden?

cast netting for menhaden

At a meeting this week in North Carolina, government scientists were largely silent on the question of how the Atlantic menhaden population should be assessed in 2012.

Fishing industry consultants, on the other hand, dominated the process — so much so that some in attendance questioned whether efforts by industry representatives to steer the discussion might have violated protocol.

The meeting was billed as an official workshop of the Menhaden Stock Assessment Subcommittee, a group of state and federal fisheries experts tasked with updating estimates of menhaden abundance in the Atlantic Ocean by re-running a computer model that was developed and peer reviewed in 2010.

Menhaden is an important prey fish that its defenders say has been depleted from sustained overfishing during the majority of the last fifty years. In November 2011, the Atlantic States Marine Fisheries Commission (ASMFC) changed the reference points used to judge the health of the stock, which will result in a reduction to the amount of menhaden that can legally be netted by commercial interests.

But the ASMFC has yet to set a timeline for the proposed changes, arguing that the Commission is waiting for the results of the 2012 stock assessment. (“We don’t know what the assessment is going to say,” Jack Travelstead, an ASMFC Commissioner from Virginia said at a recent menhaden regulatory meeting on May 2nd. “It may say that we are no longer overfishing.”)

The outcome of the assessment, scheduled for completion in July 2012, is of great interest to Omega Protein, a company that harvests a quarter billion pounds of menhaden out of the Atlantic each year.

It is no surprise then, that Omega hired two consultants to attend the meeting of the stock assessment subcommittee. Doug Butterworth of the University of Cape Town, and Michael Prager, formerly of NOAA’s Beaufort Laboratory, stood on either side of the table, looming over government scientists, and participating in the technical discussions.

Even so, the meeting wasn’t all good news for Omega Protein. Preliminary data from the stock assessment showed that the menhaden population is in bad shape – there has been an increase in fishing mortality and a decrease in abundance since 2008.

“Even relative to the old reference points, things appear to have gotten worse,” Jason McNamee, committee member and biologist with the Rhode Island Division of Fish and Wildlife, told me.

Perhaps in light of these results, Omega Protein’s hired-gun scientists spent the meeting taking issue with the stock assessment model previously vetted through an independent peer review.

Dr. Butterworth noted that the data collected from the Potomac River Pound net index, which measures adult menhaden, showed more fish in the water than in 2008. However, the rest of the model, based on all the other inputs, showed less fish. Butterworth expressed concern about the seeming inconsistency.

“When you have data as mis-fitting to the model as that, one thing goes, either the data or the model,” Butterworth told the Committee. “I worry about putting data that is contradictory into the model and then taking an average.”

Michael Prager, the other Omega Protein consultant, suggested that he thought there were issues with the juvenile abundance index, another tool used to measure the health of the menhaden population, but offered little in the way of specifics or analysis.

The two got some help from Matt Cieri, a stock assessment scientist with the Maine Department of Natural Resources, who argued that the stock assessment could technically be scrapped in favor of a more thorough “benchmark” assessment, scheduled to take place in 2015. “If the diagnostics look like complete crap, we do have the latitude to get in front of the [ASMFC] and say, we do need to take another benchmark,” he said.

Waiting until 2015 would likely delay regulation, which would benefit the industry.

Delaying for a benchmark assessment might also allow Omega Protein to use results from their aerial survey of menhaden in the northern range, which the company commissioned because executives believe data on northern menhaden will present a more favorable picture of stock health. Right now, the survey represents a snapshot of data, rather than a time series needed to pass peer review. But delaying until a benchmark could allow the company time to gather more data, and even lobby Congress to have it included.

Erik Williams, chair of the committee, and a scientist at NOAA’s Beaufort Laboratory, countered Cieri’s proposal to delay. “The model needs to be really bad for us to say that this is not good for [use in menhaden] management. This thing has gone through peer review,” he reminded the group.

Others on the committee felt that the assessment should continue as planned.

“[Industry consultants] picked at the fact that the model was not fitting well to certain data, like the Potomac River pound net index,” Alexei Sharov, a fisheries scientist with the Maryland Department of Natural Resources, said in an interview. “That’s not the best outcome, but it doesn’t mean that the model is perfectly flawed . . . I didn’t see at all that we’ve lost that confidence in the model results. I didn’t hear that from anybody on the committee except for Matt [Cieri].”

Later in the meeting, when the committee turned its attention to “sensitivity runs,” industry scientists were similarly engaged. Sensitivity runs are tweaks made to input data for the computer model, such as the eliminations of certain indices, to ascertain how much the model results depend on particular kinds of data.

Erik Williams posed the question, “which sensitivity runs should be included?”

Butterworth jumped in immediately and suggested five, and all five were approved.
Alexei Sharov of Maryland said he didn’t disagree with the sensitivity runs chosen by Doug Butterworth; however, he noted that Butterworth’s overt participation broke protocol.

“Usually we have our discussion and other people that are present are given an opportunity to make comments either at the beginning or at the end,” he said. “I think some committee members felt that Doug was intervening too much.”

Jason McNamee of Rhode Island agreed. “I think we would have come up with those sensitivity runs ourselves, but [the industry scientists] were standing at the front of the room, which was unfortunate.”

McNamee added that he didn’t feel that industry consultants were driving the process, with the notable exception of the discussion on skipping the update assessment. “I felt that was performing surgery with a sledgehammer,” he said. “99 percent of the comparisons with the benchmark assessment were spot on. Only one percent of it diverged. They triggered that discussion but we were able to get away from it.”

Mike Waine, the ASMFC fishery management plan coordinator for menhaden was overseeing the meeting, and I asked him what the protocol was for allowing outside participation in technical deliberations. “It’s up to the chair of the stock assessment subcommittee how he wants to handle it,” Waine said.

Erik Williams of NOAA had been calling on Doug Butterworth all day. When I asked about his policy on industry participation, he replied, “I’d rather not comment on that. The process allows that. Ultimately this is a scientific process, so to the degree that they can contribute to the science, that’s great. If they interfere with the science, that’s not great.”

The stock assessment subcommittee is scheduled to meet again on May 29th.