Read our COVID-19 research and news.

Go fish. Better management of small, unmonitored fisheries, such as in Kupang, Indonesia (shown), could lead to as much as a 40% increase to the global sustainable fish harvest.

Image courtesy of Dawn Dougherty

Netting Better Data on Global Fish Stocks

Overfishing is a major global concern—but that concern tends to focus on just a few well-studied fish, such as salmon and herring. Current assessments cover only 20% of the world's fish stocks, so the true state of most of the world's fish populations is still murky. Now, a new method based on catch data and fish characteristics suggests that those unstudied stocks are declining—but also that better management of global fisheries could boost the status of many of those stocks and could also increase the global sustainable fish harvest by as much as 40%.

When fisheries are managed well, they can be a source of large amounts of food and economic value without irreversibly disrupting ecosystems and depleting fish numbers. But industry lobbying, corruption, fears that reducing fishing capacity will lead to unemployment, and other economic and political obstacles have made sustainable fisheries management difficult in practice.

Meanwhile, even as they discuss measures for sustainable management, conservation scientists and fisheries scientists are locked in a disagreement about whether global fish stocks are in crisis. In 2006, marine ecologist Boris Worm of Dalhousie University in Halifax, Canada, and colleagues published a paper in Science that projected that if current practices remain unchanged, all fish stocks would collapse by 2048. This projection received heavy criticism from fisheries scientists, who said that the number of recovering stocks actually shows an overall improvement.

Fish like salmon and herring are, in general, strictly managed. Their condition is assessed through sampling to determine stock size, as well as through catch data and population models that examine factors such as the fish's growth, how often they reach maturity, and how quickly they reproduce.

But no such assessment takes place for 80% of current fish stocks, which include more than 7000 populations. This neglect applies to many uncommon fish species, as well as to some stocks of well-known species such as cod and tuna. So scientists try to get a grip on these stocks in several ways. They extrapolate from assessed stocks to provide population estimates for unassessed fish of similar size and habitat. They also use catch data provided by the United Nations' Food and Agriculture Organization (FAO) and tap into the knowledge of local fishermen about the size and health of caught fish. These figures contain several flaws, however. Local knowledge can be biased and limited, and the FAO data incorporate mislabeled and misreported catches and don't account for illegal fishing.

Now, environmental economist Christopher Costello of the University of California, Santa Barbara, and colleagues propose a novel method. To estimate the size of the unassessed stocks, the team first collected as much information as possible about assessed stocks, including the assessment data, FAO catch data, and data from the FishBase database. FishBase, run by the Leibniz Institute of Marine Sciences in Kiel, Germany, contains information on traits such as taxonomy and geographic distribution. Since all of these data, except the assessment data, are also available for the unassessed stocks, they built a model using all these data to calculate the stock size for these unevaluated stocks. Costello's team then compared these biomass estimates to the sustainable population sizes at which catches would be optimal without shrinking the stocks, as calculated by regional fishing authorities (such as the International Council for the Exploration of the Sea for most European stocks).

They found that 64% of the unassessed stocks are below their sustainable numbers—comparable to the below-sustainability estimate of 63% for assessed stocks reported in Science in 2009. The findings indicate that unassessed stocks, such as cod and miscellaneous coast fish, are in decline—and are in particularly poor condition in developed areas like the Mediterranean Sea. (The team declined to publish data on the health of individual stocks, but reported only aggregate results, emphasizing that the study isn't intended as a replacement for actual assessments.) The analysis also showed that whereas many assessed stocks have been recovering for about 10 years, most are still just below the level for maximum sustainable yield, which means they are only slightly overfished.

So although the oceans will not be depleted in 2048, Costello says, "the small-scale, unassessed stocks need recovery efforts."

That conclusion validates the longstanding concerns of many fisheries researchers. "The paper rightly concludes that the fish stocks around the world are in worse shape than predicted by FAO," says Rainer Froese, a marine ecologist at the Helmholtz Centre for Ocean Research Kiel who was not involved in the study. And Daniel Pauly, a fisheries scientist at the University of British Columbia, Vancouver, in Canada who was also not involved in the study, notes that the conclusions of the paper stand in contrast to previous statements by one of its co-authors, marine biologist Ray Hilborn of the University of Washington, Seattle. "This paper implicitly refutes the position taken very publicly by … Ray Hilborn that the world's fisheries are recovering," a position Pauly says Hilborn had inferred based only on assessed stocks in developed countries. "With his associates," Pauly adds, "[Hilborn] heaped abuse on anybody who dare suggest that his rose-tinted views may not reflect the reality of developing country fisheries."

Hilborn, however, says that the study doesn't actually suggest that the state of most unassessed fish stocks is worrisome, although he admits they are in decline. But the study found that the most economically valuable species, which make up 99% of the catch, are currently not overfished, he writes in an e-mail. "They are roughly in the same condition as the assessed stocks."

Despite these differences among co-authors, the paper does present a common conclusion. Better assessments are needed to reverse the declining trend for currently unassessed stocks.