Science for Hire: Why Industry's Deep Pockets May Be Depleting the Last of Our Fisheries
Photo Credit: AFP
This story first appeared on Public Trust Project.
It’s sunset in Woods Hole, Massachusetts, a time when the little town’s handful of shops board up for the night, and the lights click off at no fewer than six marine research institutes.
But at the far end of the town, one block from the churning Atlantic, 10 weary scientists sit around a big square table arguing about cod. They’ve been at it since 8 a.m.. Each has blocked two weeks in his or her calendar for this single purpose.
The group is meeting at the Woods Hole Aquarium, the oldest saltwater aquarium in the country, which also houses office space for 45 federal scientists working for the National Oceanic and Atmospheric Administration (NOAA). There are no cod fish to be seen in the aquarium’s tanks. There aren’t many cod in the ocean either, compared to historic levels, and it is on this point that the scientists are squabbling.
It’s not that they all disagree. In fact, there is consensus on most points among nine NOAA scientists who are attending the meeting. Their job it is to analyze fish populations in New England waters for the federal government – among them cod, pollock, and flounder.
But one prominent dissenter withholds his consensus vote: Doug Butterworth, the lone representative of the fishing industry at the session.
Butterworth is a renowned scientist based at the University of Cape Town in South Africa. He’s been lauded for his extensive contributions to the science of fishery management in many different countries. But he also has a controversial side job: he’s employed by commercial fishermen in the Gulf of Maine, who fly him into town every now and again to wage scientific war on the NOAA biologists.
It’s a role that has gained him some notoriety. “Doug seems like he’s got a retainer up here,” said Dr. Liz Brooks, a NOAA fisheries biologist who has worked on measuring pollock and cod populations.
Cod is an iconic species. Fishermen began harvesting them in the 17th century, when the fish were so abundant in New England it was said that you could walk across the sea on their backs. Over the centuries, cod’s abundance made it what Paul Greenberg called a “workaday fish,” in his lovely book Four Fish.Cod became a common, plentiful source of protein for the working class.
By the 20th century, innovations in fishing technology led to the rise of giant, industrial factory ships, which began to displace small-scale cod fleets, and a new era of overfishing was born. In the mid 1990s, scientists feared the cod population would collapse altogether due to extreme fishing pressure.
Finally, the government made the tough call to close Georges Bank to commercial cod fishing in 1994. The waters – an elevated shelf of sea floor in the Gulf of Maine – were once among the most productive fishing grounds in the nation. Since then, portions of Georges Bank and the Gulf of Maine have reopened to cod fishermen, but fishing pressure is just one-third of what it was 15 years ago. Even so, fishermen have been engaged in an elaborate dance with NOAA to prove that there are plenty of fish in the sea for the taking. Doug Butterworth has been a key part of their strategy.
Butterworth’s tactic is always the same: to challenge the notion that fish populations are as small as the government says they are, helping to ward off regulations that reduce the amount of fish that industry is permitted to take from the sea. To understand how he operates, one must dig into the complex details of fisheries science – a science that holds profound ramifications for the future of fish in our oceans.
Just over a year ago, in October 2011, this same group of NOAA researchers made a startling discovery. A newly completed computer model (called a “stock assessment”) indicated that the cod population in the Gulf of Maine had fallen so dramatically that even a complete moratorium on cod fishing would not allow the population to rebuild by 2014 -- the cutoff date mandated by federal law.
That year, commercial cod fishermen had hauled in approximately 8 million pounds of cod from the muddy bottom of the Gulf of Maine. Those working fishermen feared that the finding that cod stocks were perilously low would all but put them out of business by triggering regulations severely curtailing the amount of fishing allowed.
The scientists were surprised by what they found. A previous stock assessment, completed in 2008, had identified a cod population on the upswing – in fact, cod had been held up as an example of a fishery well on the way to recovery after near collapse. But between 2008 and 2011, 48 million pounds of fish had seemingly vanished.
To make sure the computer model wasn’t erring, Michael Palmer, the head NOAA scientist for the 2011 stock assessment, spent six months running the numbers. “It was by far the best stock assessment I’ve ever seen,” said Jud Crawford, science and policy manager at the Pew Environment Group. “Mike Palmer did every conceivable analysis.”
Palmer realized that a few lucky trawls by government boats surveying the Gulf of Maine in the early 2000s had given NOAA a misleading impression of the health of the cod population. The trawl nets had mysteriously filled with juvenile fish who were expected to build into a healthy, bountiful “year class” of adults. But they never did.
The New England groundfishing industry – the group of companies that harvest fish inhabiting the deep, chilly Gulf of Maine waters that stretch from Cape Cod to Nova Scotia, were not pleased with the findings.
“I’m telling you, it’s out there,” cod fishermen Russell Sherman told the New York Times. “We’ve had no problems locating codfish.”
“We don’t trust your data,” New Hampshire charter boat fishermen Bill Wagner explained to regulators at a meeting that was reported on by the Associated Press.
Facing the possibility of potential restrictions to their harvest, industry representatives took action. Butterworth appeared at the cod meetings bearing an alternate computer model -- one that projected nearly 40 percent more cod in Gulf of Maine waters than the model run by the NOAA scientists. His participation was commissioned by the Northeast Seafood Coalition and the Associated Fisheries of Maine.
Butterworth used a statistical strategy called “dome shaped selectivity” -- fisheries jargon for a model that assumes that the largest, most fecund animals escape capture by commercial fishermen, and therefore are not factored into the models run by government scientists. This in turn raises estimates of the number of fish in the ocean – a very good thing if you are a commercial fisherman trying to avoid regulations.
“With dome shaped selectivity, you are assuming that there are older fish out there, so the model creates them. But if you’re wrong, you are saying the stock is in much better shape than in fact it is,” said Dick Brame of the Coastal Conservation Association. “Dome shaped selectivity is the du jour way to influence stock assessments.”
Because Butterworth’s 2011 cod model found more cod fish than NOAA’s model, the scientists had to come to a consensus. Industry participation in the scientific process is permitted under the Magnuson Stevens Act, which enables cooperative research among scientists, fishery managers, educational institutions, and stakeholders.
Building a consensus among participants is not mandated by the law, but it adds to the credibility of the scientists’ report. Dr. Paul Rago, a NOAA fishery biologist who supervises the New England groundfish assessments, says that it’s “something that you desire to have as a way of providing scientific advice to the [fishery] managers.”
Consensus wasn’t easy to come by.
NOAA scientists spent a week forming a fragile scientific agreement with Butterworth. Collectively, they decided to use the government’s model, but they would employ some of Butterworth’s dome-shaped selectivity – effectively agreeing to generate more cod. Initially, Butterworth agreed to the compromise, and a report was drafted with preliminary recommendations to the New England Fishery Management Council, the body that regulates cod and other species in the Gulf of Maine.
Shortly after, Butterworth reneged on the consensus agreement. He distributed a paper to New England fishery managers that made a case for greater dome selectivity, among other modifications to the model.
Normally, when a stock assessment is completed, it goes through a rigorous peer review process. A selective group of independent scientists from around the world spend a week in Woods Hole at NOAA’s Northeast Fishery Science Center, learning about the computer model and evaluating whether it accurately models the number of fish in the sea. Only when the peer reviewers sign off on the assessment do fishery managers set regulations, such as limiting the amount of fishing. It’s not taken lightly since those kinds of decrees can affect the livelihoods of fishermen.
This time, the rules of engagement were ignored. Fishing industry allies pounced before the peer reviewers had reached a verdict on the cod assessment. Sen. John Kerry of Massachusetts issued a press release calling on NOAA to “immediately conduct a new assessment” with “complete information that has been developed with the consent of all stakeholders.”
Days later, the peer reviewers completed their final report – and NOAA’s assessment stood as the best available science: “The panel unanimously recommends that the results of the Gulf of Maine cod stock assessment be used for management of this stock,” they wrote.
But it didn’t matter – enough doubt had been raised over the validity of the science. Nineteen New England lawmakers, including seven U.S. House members and five Senators addressed the Secretary of Commerce, asking that NOAA prioritize further research that considers “the analysis provided by Drs. Butterworth and [Rebecca] Rademeyer.” (Rademeyer is Butterworth’s associate).
Ultimately, the lawmakers got their wish. NOAA leadership instructed its biologists to redo the stock assessment in 2012 – two years earlier than scheduled, at a considerable price to the American taxpayer. Stock assessments cost upwards of $200,000, according to Dr. Rago.
In the meantime, the New England Fishery Management Council decided to cut the harvest of cod by just 20 percent, from 17 million pounds to 14 million pounds. NOAA had recommended a 70 percent decrease in catch in order to preserve Gulf of Maine cod and its ability to replenish itself.
In May 2012, Michael Palmer began preparing a new stock assessment. At the end of the months-long process, his findings remained virtually unchanged.
Five months later, in October, Butterworth arrived on Cape Cod, where he and the NOAA scientists began their two-week long working group on the state of the New England cod fish.
On day one of the meeting, Jim Weinberg, chairman of the stock assessment workshop process at NOAA’s Woods Hole facility, issued parameters to the scientists aimed at minimizing conflict. “The hope is that you will arrive at a consensus,” he said, glancing briefly at Butterworth. “This is a sequential peer review process. It doesn’t work well if we go to them bringing them two, three, or four different opinions.”
Butterworth wasted no time in responding. “Legitimately, there can be more than one defensible assessment,” he said. “There can be multiple views each indicating a different [regulatory] action. Are we going to get into that?”
There was a titter among NOAA employees in the room. “We generally don’t resort to voting or anything like that,” Weinberg said firmly. “The scientific process is typically one of consensus. That means a number of people think it’s the right thing to do, and some group of people can live with it.”
Then they dove in. Palmer presented his model, took questions from the group, and Butterworth went over his alternate model, which used dome shaped selectivity to show a higher “biomass” of cod in the ocean.
“My concern is the thing that’s driving the dome selectivity is this period when you have almost no data. You have almost no information for these older ages of fish,” Dr. Dvora Hart, a stock assessment scientist with NOAA’s Population Dynamics branch, argued in response.
In an interview later that day, Hart told me that Butterworth has typically been hesitant to divulge the methods he uses to demonstrate dome shaped selectivity, and project more fish. “He doesn’t like to show the data,” she said. “He says you only interpret data through the model. I’m a scientist, I want to first look at the data.”
“It irritates me how much time we have to waste with Doug. He talks a lot and its purposeful, so other people won’t talk as much. People don’t want to cut him off because the industry will scream at them that they don’t get heard,” Hart said.
At the end of two weeks, Butterworth refused consensus.
Without an agreement, NOAA must send its report to peer review with both the government’s model and Butterworth’s model on the table. The peer review is scheduled to take place in December; it is unclear what will happen when the reviewers confront two models. “It’s only happened once before,” said Dr. Rago. “If you don’t reach consensus it gets a little murkier.” The peer reviewers will either have to choose one model themselves, or the regulatory process will be in deadlock without a single model to serve as the “best available science.”
Like an expert witness in litigation, Butterworth has used these tactics before – with great success. To the biologists who work at NOAA’s Woods Hole laboratory, he’s a regular presence at stock assessment meetings for pollock, white hake, yellowtail flounder, and more. He is always pushing dome shaped selectivity, that magical maneuver that produces a greater amount of fish.
“Doug shows up looking for domes,” said Dr. Brooks. “I joke that he has to declare his domes before entering the country.”
When fishing is permitted at reasonably high levels, there is little reason for industry to develop its own science. The moment that government science begins to show a population trending downward, however, that can change – quickly.
Recently, Butterworth was hired to present alternate science for menhaden, a keystone species in the Atlantic Ocean that is the main source of food for dozens of important marine predators. A stock assessment from 2010 found that the menhaden stock is at its lowest point on record – just 10 percent of menhaden remain compared to historic levels.
Butterworth was brought on beginning in 2011 by Omega Protein, a company that nets nearly half a billion pounds of menhaden each year. Recently, he has been meeting with fishery managers serving on the Atlantic States Marine Fisheries Commission (ASMFC), the agency that regulates the menhaden fishery, which is preparing to make a critical decision about whether to limit the menhaden harvest this coming December.
At these meetings, Butterworth has submitted a paper entitled “Is Overfishing of Atlantic Menhaden Occurring?” which argues that there is a “clear and sound statistical justification for the introduction of domed selectivity” – meaning there’s more fish out there than you think. Evidence for using the doming is not just statistical, the paper argues, “it is a reflection of emigration of the older menhaden outside (primarily to the north of) customary fishing grounds.”
“Talk about making decisions with limited data,” said Lynn Fegley, the associate director of Maryland’s Fisheries Service, who sits on the ASMFC and has met with Butterworth. “What this hinges on is the idea that there are fish in the northern waters.”
There are no data showing that menhaden are present in northern waters, save for an aerial study conducted by Dr. James Sulikowski, another scientist hired by Omega Protein, who found a limited number of menhaden schooling in the north. Sulikowski himself told me that his research would have to continue for a number of years before it would be statistically significant.
Butterworth’s interventions have had an impact.
In the case of pollock, dome shaped selectivity has translated into huge increases in the quotas allowed for fishermen. In 2010, NOAA scientists presented a stock assessment model that found some limited dome selectivity in the pollock fishery. “Butterworth came in with really high domes and extremely high abundance,” said a NOAA biologist who worked on pollock. The New England Fishery Management Council chose Butterworth’s version of the dome, which enabled the agency to increase catch limits, or quotas, for pollock by 600 percent.
But data suggest that those huge increases in quotas haven’t translated into more fish harvested by industry.
As of Oct. 31, the mid point of 2012 fishing season, pollock fishermen had caught just 28 percent of their allocated Gulf of Maine pollock quota. Cod fishermen had caught 25 percent of theirs, according to data collected by NOAA.
“The most likely explanation for this is that fishermen simply aren’t finding the fish.” Michael Conathan, director of ocean policy at the Center for American Progress, wrote on his organization’s website.
Butterworth defends the right of commercial fishing interests to have a seat at the table.
“Given that I represent industry I feel I am obligated [to speak up] in discussions if a feature that I believe is of importance and is relevant to industry interest is being overlooked. If I am there as a representative of industry I will make sure that it is on the table,” Butterworth told me in a phone interview.
“There is a line in this game between what’s acceptable and what’s unacceptable and I think I have stayed in the bounds of what’s acceptable,” he said.
He cited his work for the government of South Africa, in which he has had to defend the government’s position against private industry. “I’ve had my own battles with industry consultants,” he admitted.
Does Butterworth’s outsized role in Atlantic stock assessments fundamentally undermine good science? Because biologists involved in the process strive for consensus, Butterworth’s flair for casting doubt on the work of government scientists means that he enjoys near veto power.
Some NOAA employees have questioned whether industry involvement complicates their efforts to conduct research that would ultimately improve the stock assessments in the long term.
“If I had time I would do research, and look at more robust ways to capture uncertainty in the assessment models. That has never happened because there’s always a crisis,” said Dr. Liz Brooks. “Someone is always saying ‘your models don’t work, do the same assessment over with same data.’”
“My job is to make sure that the fishery is sustainable. My job is to make sure that 10 years from now, there is a fishing industry,” Michael Palmer, the lead scientist for the cod assessment, told me. “Right now we’re on a treadmill where we are constantly doing stock assessments. There’s very little time to progress in the quality of science.”