A new study argues that the loss of biodiversity threatens the ability of the world’s oceans and coastal areas to maintain healthy fisheries, a team of scientists said in a recent article published in the journal Science.

The cumulative loss of species—from inconspicuous worms to large fish—sharply reduces the ability of sea life to resist diseases, filter pollutants and rebound from stresses such as overfishing and climate change, the scientists said.

If current trends continue, the scientists predict a global collapse of fish stocks in the middle of the century—an argument disputed by other fishery scientists.

“Whether we looked at tide pools or studies over the entire world’s ocean, we saw the same picture emerging,” said Boris Worm of Dalhousie University, the lead author of the paper. “In losing species, we lose the productivity and stability of entire ecosystems. I was shocked and disturbed by how consistent those trends are—beyond anything we suspected.”

The study did not involve new research, but relied on existing data collected in studies around the world.

The scientists examined results from dozens of controlled small-scale studies in coastal areas that manipulated aquatic populations in local areas. The studies showed that on average, the greater the species diversity in an area, the greater its production and stability—a pattern that was reversed as species were removed.

They saw the same picture emerge when looking at regional coastal areas. The healthiest systems had higher rates of biodiversity. But areas that lost species—or where populations of critical species such as filter-feeding animals or seagrasses had been dramatically reduced—were more susceptible to harmful algae blooms, eutrophication, fish kills, beach closures and other problems.

Turning their attention to large ocean areas, the scientists drew on long-term fisheries records, and concluded that large marine ecosystems with greater diversity of fish species were less prone to stock collapses than areas with fewer species.

“We were all kind of blown away by the consistency of the patterns that we see from the smallest to the largest scales,” said Emmett Duffy, a fisheries scientist at the Virginia Institute of Marine Science, and one of the paper’s 14 co-authors. “The global fisheries data mirror the kinds of patterns that we see in small controlled experiments, and in fact are predicted by simple general theory.”

Conservationists and many scientists have long argued that maintaining species diversity—including small, overlooked species which may unknowingly play important ecological roles—is important for healthy, functioning ecosystems. The Science paper sought to apply that concept to broader marine systems.

“The basic analogy is that an ecosystem is sort of like a machine, and that you need a lot of the parts interacting with one another to support what it does,” Duffy said. “What it does in our case is produce fish and crab harvests, oysters and clean waters. You need a lot of those other inconspicuous species that we don’t think about much to provide sort of a natural infrastructure that supports the products that we take out. When you start losing the parts of that machine, it breaks down.”

In the Bay region, for instance, a die-off of seagrasses in the 1930s closed the bay scallop fishery as the beds, which were critical habitat for the scallops, quickly declined. Likewise, Duffy said, the demise of oysters in the Bay led to a loss of habitat for a variety of reef-dependent species, and the loss of the oysters’ filtering ability contributed to poorer water quality in the Chesapeake.

Conversely, the scientists said, systems that maintain healthy and diverse populations are more productive, and bounce back from disturbances more readily. “The main thing about the loss of species is there is this web of complex interactions that starts getting fouled up when species are lost,” Duffy said. “And that ultimately ripples out to affect the things that we care about like fish production and clean water.”

A more controversial conclusion of the study was a projection which, using collapsed fish stocks as a proxy for loss of diversity, indicates that wild fish stocks being used for seafood would face global collapse by 2048.

“There are some good arguments about biodiversity,” Doug Lipton, an economist and fisheries scientist with the University of Maryland, said of the paper. But he said the fisheries prediction was based on a “big if”—that fisheries would continue to collapse at the same rate as in the past.

In fact, he said, other factors, such as the economics of different fisheries, affect overfishing. And the biology of some species—such as those with low reproductive rates—make them more susceptible to overfishing than others. “The species that are most susceptible to collapse have already collapsed,” he said.

Ed Houde, a fishery scientist with the University of Maryland Center for Environmental Science who has served on many national and international committees dealing with fishing issues, agreed that “fisheries have serious problems, we all know that.” But, he added “in the year 2048, short of an asteroid hitting the Earth, I think we’ll still have some pretty productive fisheries.”

In particular, Houde—in an point also cited by others—faulted the assumption in the paper that a stock was “collapsed” if harvests were 10 percent of their historic highs. Often, peak harvests for many species are not indicators of abundance, he said, and many peak catches were dramatically above sustainable levels. It’s probable, Houde said, that catches that were only 10 percent of historic levels may be the appropriate harvest rate for some species now being regulated.

Globally, he said, fish management has improved and become more precautionary, and U.N. Food and Agriculture Organization figures show that the number of fish stocks in trouble is slowly being reduced.

Regionally, he said, a more precautionary approach to fisheries management can be seen in the Atlantic States Marine Fisheries Commission’s recent action to cap menhaden catches in the Bay, despite its own stock assessment which indicated the coastwide menhaden stock was healthy.

“That was an ecosystem approach that was taken because we were worried about food for predators and the ecosystem services of filtering by menhaden,” Houde said. “We made a social decision that was a precautionary one, recognized ecosystem services and capped the fishery.”

Other scientists have argued that, especially in coastal areas, pollution is a major—and potentially larger—threat to biodiversity than overfishing. While that may be true for coastal areas including the Bay, Duffy said overfishing is almost certainly the more important factor affecting biodiversity in oceans.

Duffy agreed that many fisheries are well-managed, especially in the United States. But, he said, more enlightened management—such as establishing marine sanctuaries, which the paper touted as an effective tool—was needed to preserve many of the world’s troubled fisheries.

“If 10 or 12 fisheries are remaining at midcentury compared to the hundreds that we have now, that is a big loss,” he said. “If our best management practices are not applied much more broadly than they are now, then we are not going to have much seafood and it will be quite expensive.”