Australia’s management of its fisheries and marine ecosystems is widely considered among the best in the world. But the first independent peer-reviewed study of targeted fish stocks in Australian waters has found alarming population declines.

“Regardless of its reputation for sustainable fishery management, overfishing has apparently contributed to … declining biomass of large fishes on Australian reefs,” wrote Graham J. Edgar and Rick D. Stuart-Smith of the Institute for Marine and Antarctic Studies at the University of Tasmania, and Trevor J. Ward of the University of Technology Sydney, in a paper published in the journal Aquatic Conservation.

The study found a 36 percent decline over the past 10 years in the total biomass of fish 20cm (8in) or more in length in zones where fishing is unrestricted. Where fishing is regulated, the decline was 18 percent. No-fishing areas, in contrast, saw a 4 percent increase in biomass. To test whether the fall was a function of environmental pressures other than fishing, the scientists looked at species that are not fished and found their drop in biomass was “non-significant.”

“Populations of exploited fishes generally rose within marine reserves and declined outside the reserves, whereas unexploited species showed little difference in population trends within or outside reserves,” the researchers wrote.

They reached those conclusions by analyzing a decade’s worth of fish counts conducted by the Reef Life Survey, an independent visual census that does not rely on fisheries data or involvement. Australian fisheries authorities, on the other hand, evaluate fishery stocks by looking at “catch per unit effort.”

That approach assumes through complex mathematical models that how much is caught per unit effort gives an idea of how many fish are left in the sea. So without actually going underwater and counting fish, the declines in catch found over the same period could be attributed to new fishing restrictions put in place after 2005, the year the researchers chose as the baseline against which to assess fish populations over the following decade. But the 31 percent overall drop in biomass found in the visual census corresponded closely with the 32 percent decline in catch reported by fisheries, indicating, the authors wrote, “that falling catches reflect declining fish populations, at least in part.”

The study generated an extraordinary rebuke from the Seafood Industry Australia (SIA), an industry group representing the nation’s fishers. SIA claimed “a number of flaws in the paper’s research and methodology” and cited the Australian Bureau of Agriculture and Resource Economics and Sciences (ABARES), which found in 2017 that “for the fourth consecutive year there were no stocks classified as subject to overfishing in any fisheries managed solely by the Australian government.”

Said Jane Lovell, the SIA’s chief executive: “Reports like this impact community perception of the fishing industry, and we’ve been beaten up for a long time by environmental NGOs and reports like this.” Comparing the new study with ABARES’s finding, she said, “the two stories don’t line up.”

Two scientists not involved in the study backed the paper and its methodology. “There are no perfect data, but these are as good as we’ve ever got, and anyone who’d ignore these warnings would do it at their own peril, I think,” said David Booth, a professor of marine ecology at University of Technology Sydney.

Callum Roberts, a marine conservation biologist at the University of York in the United Kingdom, said the paper, “shows … for reef environments there’s been no improvement in fisheries management, and populations of fish on those reefs have gone down in [both] limited fishing and open access areas, so whatever managers are doing in those environments isn’t working.”

The study’s authors seemed to support that assessment. “To maintain some naturally functioning food webs … and associated ecosystem services … a greatly expanded network of fully protected marine protected areas is needed,” they wrote in the paper.

The findings come as Australia’s conservative government proposes to open 44 marine protected areas to recreational and commercial fishing. “Independent scientific review – substantially ignored by this government – recommended an increase in no-take protections, but the government has taken the polar opposite approach to that,” said Michelle Grady, oceans director for Australia at the Pew Charitable Trusts.

“A highly protected marine park area provides conservation outcomes that no level of fisheries management can match,” said Tony Burke, a Labor Party member of the Australian House of Representatives and shadow minister for environment and water. “There are principles that have been used to establish national parks on land for more than 100 years, and every time it’s worked out. In the ocean, we see that the exact same principles apply.”

Australia’s minister for environment and energy, Josh Frydenberg, defended opening the reserves to fishing. “At more than 3.3 million square km [1.3 million square miles], Australia has one of the largest areas of marine parks in the world, second only to the United States,” he said in a statement issued by his office. “The new plans improve on plans introduced but never implemented by the Labor Party in 2012 by delivering a more balanced and scientific evidence-based approach to ocean protection, enabling tourism and well-managed fishing activity supporting local communities, local jobs and regional communities.”

Lovell argued that the paper’s use of 2005 as the baseline is “invalid and grossly misleading” because it was “the year before Australia adopted its current, more responsive fisheries management approach.” She said that because catch limits were introduced after 2005, it would make sense that the researchers found fewer caught fish. But Roberts and Booth said that if the new management approach had been effective, the data from counts of fish in the water should have shown more fish after 2005, not less.

“I think that’s a spurious criticism, really,” said Roberts.

“I didn’t understand that at all,” said Booth. “Every measure they have in the paper is on the way down, so you might have slight issues with each, but it’s consistent findings.”

Lovell also took issue with the method of counting fish, suggesting that the citizen scientists who swam underwater transects were unqualified for the work. But the Reef Life Survey data the volunteers contributed to, which alongside catch data informed the paper’s conclusions, “has been validated through numerous scientific publications, including seven papers published or in press with Nature,” one of the most widely cited scientific journals, said co-author Edgar.

Lovell raised concerns that the reef data was insufficient to draw conclusions about Australia’s fisheries as a whole because so much seafood is extracted farther offshore. But Booth pointed out that counting fish in deep water is impossible because divers can descend only to about 30m (100ft). Scientists can use underwater cameras to find fish, but the necessity of baiting them for efficiency biases their data, so they look at catch numbers. “We make the assumption that catch is in some way proportional to the abundance of fish,” Edgar said. “That’s what fisheries [do] as well.” The paper, he says, indicates “maybe we need to push for [more] fisheries-independent data sets offshore.”

The ABARES findings, he said, are “hypothetical constructs … reliant on a conceptual mathematical model that have a huge number of assumptions built in. Ours was an empirical field assessment of what are things actually doing in the water.”

“Fisheries management works by using a particular framework,” added Edgar, who noted his institute receives funding from the fishing industry.” We tested those using a fisheries-independent approach to see whether the final conclusion is correct that fisheries are sustainable. The hypothesis that fisheries are sustainable is not supported by the field data and that raises flags for the process [by which fisheries are] coming to their conclusions.”

Lovell trumpeted that 36 percent of Australia’s marine habitat is in highly protected marine park. But the government proposal now before Parliament would open 17 percent of that protected area to commercial fishing. Sixteen percent of reserves would be open to recreational fishing.

“We provide 1 billion meals to Australian families every year, and 70 percent of seafood eaten in this country is imported,” she said. “So if we’re at the pinnacle of fisheries management and only 30 percent of seafood consumed is produced under that management structure, you could argue that if we stop using our own resources we’re actually making the situation worse globally and in other countries.”