The study, by researchers from Duke University's Nicholas School of the Environment and Earth Sciences, was published in the March 2004 online and print editions of the research journal Ecology Letters. First author was Rebecca Lewison, a research associate at the Duke University Marine Laboratory. Co-authors were Sloan Freeman, another Duke research associate, and Larry Crowder, who is the Stephen Toth Professor of Marine Biology at Duke. Their research was funded by the Pew Charitable Trusts.
Longlines are lengths of monofilament fishing lines that can stretch for 40 miles and dangle thousands of individually baited hooks. They are set at optimal depths and times to catch tuna and swordfish.
Because the environmentally protected loggerheads and leatherbacks frequent the same zones where these longlines are strung, many sea turtles are either hooked attempting to swallow the bait or are entangled in the fishing gear, the study noted. Such unintentional captures are classified as "bycatch."
"There have been few attempts to quantify the magnitude and extent of protected species bycatch even for fisheries in which bycatch is perceived as a pressing concern," Crowder and his colleagues wrote in their Ecology Letters research paper. "This is, in part, a consequence of limited data."
In the face of those shortcomings, the Duke team mined available turtle bycatch data from the 13 nations that collect such information. And they extrapolated estimates for areas like the Indian Ocean where bycatch data was unavailable.
They also collected the most current fishing information from three primary sources: The International Commission for the Conservation of Atlantic Tunas, the Indian Ocean Tuna Commission and the Secretariat for the Pacific Community Oceanic Fisheries Programme.
To obtain a global picture of where and how frequently turtles were being caught, the researchers then superimposed fishing and bycatch data on a grid map of all Earth's oceans. They also added available demographic data for loggerheads and leatherbacks.
In an interview, Crowder said "the Ecology Letters paper is the first to do a global assessment of sea turtle bycatch in longline fisheries. That's what sets it apart. There had been some earlier regional analyses of sea turtle bycatch, but nothing global prior to this."
The problem was particularly acute in the Pacific Ocean, found the researchers. In fact, more turtles "are killed than nest in the Pacific," Crowder told a February symposium on marine animal conservation at the American Association for the Advancement of Science 2004 annual meeting in Seattle.
The published study located four "primary hotspots" for longline fishing: in the central and southern Pacific Ocean, the southern Atlantic and the Mediterranean Sea. Crowder said those longlining hot spots mark sectors where currents converge in a way that boosts the productivity of marine life, which in turn attracts both hungry turtles and commercial fish.
The authors estimated that longline fleets from 40 different countries set about 1.4 billion hooks in the studied year of 2000 - the equivalent of about 3.8 million hooks each day. And their results suggest that longline fishing worldwide was "likely to have caught at least 200,000 loggerheads and 50,000 leatherback turtles in 2000," they wrote.
In his interview, Crowder added that loggerheads tend to become bycatch much more frequently than leatherbacks because loggerheads are much more interested in nibbling longline bait. "Leatherbacks very often are not caught by being hooked in the mouth but they're sometimes hooked in a flipper or have a line wrapped around their flippers," he said.
Using National Marine Fisheries Service bycatch mortality figures, the study estimated that "tens of thousands" of the total hooked or entangled turtles ultimately died from those encounters.
The authors especially warned of "serious consequences" for the future of loggerhead and leatherback in the Pacific, where they wrote that "precipitous declines" in the numbers of nesting females are already being recorded.
Because the air breathing turtles can often still surface after they are snared, most deaths are not caused by drowning but by "injury related to hooking or entangling," Crowder said.
Previous research, the study also noted, revealed that longlines set to catch swordfish snare turtles at a 10 times greater rate than tuna longlines. Crowder said such a difference arises because tuna longlines tend to be set deeper in the water than the depths where turtles tend to frequent, and during daylight hours.
"Swordfish gear tends to be fished at night, and in shallow water," he said. "So if you just look at where the gear is relative to where the turtles are, it's more likely that they're going to bump into gear set for swordfish," he said.
The study noted that "the United States has implemented both temporary and permanent fishery closures to reduce turtle bycatch and protect turtle populations." But it adds that only about 2 percent of worldwide longline fish landings are taken by U.S. fishing vessels.
Crowder added in his interview that a number of ideas are already under consideration to address the turtle bycatch problem. Two current favorites are substituting circle-shaped longline hooks for the present "J" shape, and altering the kind of bait used.
"The National Marine Fisheries Service is proposing that circle hooks and changes in bait will largely solve the problem," Crowder said. "I tend to think it's a recommendation that is very promising, but not yet ready to implement in the fishery."
Other suggestions include altering fishing practices, like changing the depths at which the hooks are set. "Perhaps places could be located in the ocean where there are still swordfish to catch but fewer turtles, redirecting the fishery to where the bycatch rate might not be as high," Crowder said.
Technology can aid efforts to restrict or close off certain areas from fishing for a time, he said. By international agreement, longline fishing boats might be required to install sensors so that satellites can be used to monitor them for compliance.
Such "vessel monitoring systems" are already used off regions like New England, where both cod and scallop fishing are banned in large areas, he said.
"The additional costs are minor, considering the costs of other electronics that state-of-the-art fishing vessels now have on board. That might seem invasive to individual fishermen. On the other hand, if one fisherman is playing by the rules he would want to be reassured that all the other fishermen are too.
"I guess our paper raises the stakes a bit to say this is something that we need to deal with fairly quickly," Crowder said. "It puts the pressure on to research those potential solutions as quickly as possible."