Fisheries managers may need to rethink how they set catch limits because new research is questioning a fundamental part of the equation.

Sustainable harvests typically target the growth of a fish stock each year, which is called productivity. How productive a stock is has long thought to be a direct consequence of the abundance of the fish population.

Essentially, the long-held assumption was, the more fish in the sea, the more productive they are up to a certain point, and thus the higher the catch limit.

Research in this week's Proceedings of the National Academy of Sciences challenges that assumption.

A team from the University of Washington, Rutgers University and Argentina's Centro Nacional Patagónico analyzed 230 fish stocks from around the world and found that abundance drove productivity in 18 percent of the stocks.

For most of the stocks, shifting to a higher or lower productive state had little or nothing to do with how many fish were there. Rather, other environmental factors were the primary force behind productivity shifts in 69 percent of the stocks.

"That was a very surprising finding," said Olaf Jensen, a study co-author and fisheries biologist at Rutgers. "Most management is based on the fundamental idea that there is a repeatable relationship between productivity and abundance."

While it is unclear what exactly sparked the shifts, the researchers suspect things like changes in predator, prey or competitor populations or changes in habitat, such as temperature.

The team used records in a global fish stock database that was unveiled last year. They calculated what productivity would have been without fishing, which includes physical growth, addition of new offspring and natural survival. Then they tested four models -- abundance, environmental conditions, a combination of the two or random -- to see which one best matched the productivity pattern.

Sustainable harvests are typically targeted around annual productivity to preserve the population at a steady level. Jensen compared it to withdrawing only interest on a savings account, while keeping the principal in the bank.

Mangers used abundance (the principal) to determine how much growth (interest) to anticipate and harvest. But the study shows productivity is much more like a volatile stock market, fluctuating between high and low states.

"We have to be a lot more modest about our ability to manage populations for a specific population size," Jensen said. "There are all kinds of environmental variables beyond our control."

The study was party funded by the National Oceanic and Atmospheric Administration, which manages hundreds of fish stocks.

Richard Methot, the agency's stock assessment science adviser, welcomed the study.

"What this paper points out is the frequency that irregular changes in productivity are a bit more common than we had comfortably assumed previously," he said.

Fisheries managers will have to be more prepared to adjust harvest rates based on environmental conditions, rather than "driving through the middle of all the possible shifts," Methot said.

The science adviser said NOAA officials would try to make catch-limit forecasts more responsive to these possible changes, now that they know they play such a key role in productivity.

"We have forecasts in place today that, in principle, they can take into account that additional information, but very few cases actually do that," Methot said.

Managers have been working on implementing adjustments based on environmental factors, he said. For example, after a 1995 paper found California sardines fluctuated with ocean temperature, managers modulated harvests as temperatures went up and down. However, that relationship has failed to hold up in recent years, indicating other factors are at play, Methot said.

He agreed that it is imperative to have regular stock assessments to detect shifts in aggressively fished stocks. But taking it to the next level to find the early indicators to predict when shifts will occur will be a significant challenge, he added.