In 1991, longline fishermen in the waters off Hawai’i set approximately 12.3 million hooks. They caught 9.9 million pounds of swordfish, 5.8 million pounds of various large tuna – and, according to the best estimates of the National Marine Fisheries Service, more than 750 endangered or threatened sea turtles, of which about 150 are believed to have died by the time they were retrieved.
The turtle haul is unintentional. Still, since the estimated taking is higher than that allowed by NMFS, it constitutes a violation of the federal Endangered Species Act. But practically every year that the longline fishery has operated since the late 1980’s, the Hawai’i based longline fleet has hooked more turtles than NMFS has allowed.
Renewed Consultations
The Endangered Species Act does not absolutely ban any injury to or killing of the animals that appear on the federal lists of endangered and threatened species. Section 7 of the act does, however, require that if any such “takings” are unavoidable in connection with a federally sponsored project that is otherwise desirable, those takings must be limited to a number whose loss is not expected to impair the species’ chance for recovery. In the case of sea turtles, the agency that establishes the takings limit, through so-called Section 7 consultation, is the National Marine Fisheries Service. The federal involvement that triggers Section 7 is the Pelagic Fishery Management Plan authorized by the Western Pacific Regional Fishery Management Council, one of eight fishery management councils nationally that are established under the Magnuson Fishery Conservation and Management Act.
The incidental take limit is based on a biological opinion that NMFS prepares to determine whether a given project places a protected species in jeopardy. The opinion takes into account the level of activity proposed as well as other factors (the species population, habits, and the like).
In 1985, NMFS issued its first biological opinion for the pelagic fishery management plan developed by the Western Pacific Fishery Management Council. At that time, the longline fleet consisted of fewer than 40 boats. NMFS looked at the expected level of longline fishing activity in all of the areas subject to the Fishery Management Council’s jurisdiction (an area that includes waters of Hawai’i as well as those of Guam, American Samoa, the Commonwealth of the Northern Marianas, and other U.S. flag islands in the Pacific).
Turtle Takings
That first biological opinion considered the fishery’s impact on four species of turtles: green turtles (Chelonia mydas), which are federally listed as threatened; olive ridley turtles (Lepidochelys olivacea), listed as threatened; leatherback turtles (Dermochelys coriacea), listed as endangered; and hawksbill turtles (Eretmochelys imbricata), also listed as endangered. An acceptable annual level of take,” NMFS determined, “is 50 individuals of each species. Of these 50 takes, mortality shall not exceed 25 animals for each species.” As a condition of these takes, NMFS required that it be notified when each taking occurred and if it involved a death; also, “every effort must be made to return the turtles to the sea alive, with as little harm as possible.”
Starting in the late 19808, the Hawai’i based longline fleet experienced a four-fold increase in vessel numbers. The heightened activity caused NMFS to undertake once more a formal Endangered Species Act consultation with the Fishery Management Council. That second biological opinion, issued on May 15, 1991, authorized an annual taking of 25 sea turtles, with no more than one death each of leatherback, olive ridley, and green turtles.
Although the longline fleet had grown, the basis for the NMFS biological opinion in 1991 did not reflect most of that growth. NMFS had assumed an annual fishing effort of 1.4 million hooks set by the longline fleet, when in fact, the actual number of hooks was almost nine times this amount (12.3 million hooks).
A Third Try
When NMFS reviewed the logbooks of longline vessels and the 1991 annual report of the Fishery Management Council, NMFS found that the incidental take of sea turtles far exceeded the authorized annual take. That prompted NMFS to reinitiate Section 7 consultation in early 1993.
This third consultation considered the impact of longline fishing in Hawai’i on the three turtle species considered in 1991, plus the hawksbill (originally considered in 1985, but dropped in 1991), and the loggerhead turtle (Caretta caretta), which is listed as threatened.
On the basis of this consultation, NMFS issued its “Statement Regarding Incidental Taking” of June 1993. NMFS would allow a total take of 752 sea turtles. No more than 299 of that number were to be injured or retrieved dead, with NMFS defining injury as “turtles that are released still entangled with fishing gear and 25 percent of those turtles released alive where the swallowed hook was not retrieved and the fishing line was cut. And in no case are more than 150 leatherback turtles to be taken “in a manner that is observed to result in mortality or serious injury.” The numbers were based on NMFS’ estimate of the actual numbers of turtles taken by the longline fishing fleet the previous year.
The biological opinion concluded that the operation of longliners in Hawai’i over the next 12 months, at present levels, was “not likely to jeopardize the continued existence of any threatened or endangered species.” However, NMFS added, “this fishery adversely affects: leatherback, hawksbill, green, olive ridley and loggerhead sea turtles and … the authorized level of take established by this biological opinion may not likely be sustained by these species on a continuing basis without the risk of jeopardizing their continued existence.”
Three “conservation recommendations” accompanied the biological opinion. NMFS should “undertake research to determine the fate of turtles released alive” so that it can “more accurately estimate the impact of this fishery on listed turtles.” NMFS also proposed to the Fishery Management Council that it amend its Pelagic Fishery Management Plan to “preclude increases in fishing effort until NMFS has determined that incidental sea turtle mortality has been managed at a level that will not preclude recovery.” Finally, NMFS recommended that it work with the U.S. Department of State in working out an exchange of information with other nations that fish in the Pacific so it can better determine the impact of longline fishing on an international scale.
Reasonable, Prudent
Over and above the conservation recommendations, NMFS notified the Western Pacific Regional Fishery Management Council that if longline fishing was to continue under the Council’s authority, the following “reasonable and prudent measures” would have to be carried out.
By July 1993, a voluntary observer program, in which observers were placed aboard vessels agreeing to take them, was to be in place, with a mandatory observer program to be instituted “as soon as practicable.” (From July to December, only two observer trips were made under the voluntary program, owing largely to the lack of federal observers.) The mandatory observer program began in March, after 15 observers had been hired and trained. Under this program, all vessels in the Hawaiian longline fishery must notify NMFS before the start of their trips and accept NMFS observers if requested. At most, the observer program is intended to cover only 10 percent of all longline trips.
On the basis of information provided by the observers and other information, NMFS may “impose appropriate conservation measures, either under regulations implementing the Pelagic [Fishery Management Plan,] or under the authority of the Endangered Species Act.” Such actions could include area or seasonal closures, gear restrictions, quotas, and the like.
NMFS is to undertake research to find out the fate of turtles released alive. To this end, it prepared a research plan to assess “marine turtle hooking mortality” in December 1993.
Fishermen that catch turtles are required to handle them “with due care,” to observe the turtles’ activity, to “resuscitate” them if necessary and to return them to the water.
Finally Section 7 consultation is to resume within 12 months – that is, by June 1994.
A Chilly Reception
The response of the Western Pacific Regional Fishery Management Council to NMFS’ findings and recommendations was cool. A subcommittee of the Council’s Scientific and Statistical Committee met in Honolulu in late August 1993 to review the biological opinion and its recommendations. In its report, that subcommittee warned, “Extrapolations from data collected elsewhere … must be viewed with caution…. although use of data from other sources is warranted until locally generated data provide management guidance.” It recommended that NMFS undertake a research program even broader than that proposed by NMFS “to get as much information as possible on those incidentally caught.”
The panel did not like NMFS recommendation that there was no increase in fishing effort until the present level of turtle hooking was determined to pose no harm to the species’ chances of recovery. In the first place, the panel stated, “proving this negative is an impossible scientific endeavor.” In addition, this recommendation “is contrary to the Council’s desire to allow vessel owners… to upgrade their vessels.” It acknowledged that the pending Amendment 7 to the Pelagic Fishery Management Plan (discussed elsewhere in this issue) “might lead to some increase in the number of vessels actively fishing, and may also lead to an increase in harvesting effort by some vessels.” In any event, “a 12-month limit on harvesting effort is unlikely to have any significant impact on cumulative turtle captures or mortality,” the panel determined.
At the meeting of the whole Council in September 1993, the Scientific and Statistical Committee’s concerns were repeated by Council Chairman Rufo Lujan. According to minutes of the meeting, Lujan said he “was bothered by the fact that the biological opinion relied on information from areas outside the region. The Mediterranean fleet and information from Japan was used to arrive at the conclusions in the biological opinion… Furthermore, Lujan did not find a sound basis for the conclusions arrived at in the biological opinion.”
Even after the meeting, the Council pursued its case against the NMFS biological opinion. In October 1993, Council Chairman Lujan wrote Rolland A. Schmitten, assistant administrator for Fisheries of the National Oceanic and Atmospheric Administration, complaining that NMFS failed to include the Council and the longline industry in its third Section 7 consultation. NMFS had released no draft opinion “and there was no comment period,” he wrote, adding: “We requested preliminary drafts of the opinion several times with no success.
“We were explicitly excluded from the consultation, but now that it has been completed maybe you can help. We would appreciate your providing the existing administrative record of this consultation, including a chronology and summaries of any meetings held, of personal contacts between biologists and industry, of written correspondence and phone conversations on the topic, and any other material used in this consultation.” According to Council staff, Lujan’s letter has not been answered yet.
Meager Data
In response to Lujan’s concerns at the September Council meeting, NMFS biologist Gene Nitta acknowledged that the data were “meager,” according to the minutes, but suggested that this could be owing, at least in part, to poor reporting of turtle takes by the longline fleet itself. This point is discussed further in the biological opinion. In 1991, “of the 140 vessels that submitted logbooks [to NMFS], only 24 vessels reported interactions with sea turtles,” the biological opinion states. “There were 61 total incidental takes including three mortalities and one injury…” reported in the longline logbooks for 1991. On the basis of the logbook data, NMFS calculated an “incidental take rate of .005 turtles per 1,000 hooks, and an observed mortality rate of .0002 turtles per 1,000 hooks.”
That rate is extremely low when compared with rates estimated elsewhere. Scientists studying the Japanese tuna longline fleet worldwide estimated an incidental capture rate of 0.1 turtles per 1,000 hooks, the biological opinion notes – a rate twenty times that extrapolated from Hawai’i longliner logbooks. An observer-based study of the Spanish swordfish longline fleet in the Mediterranean found a turtle hook rate of 4.5 turtles per 1,000 hooks – a rate nearly a thousand times higher than that derived from the Hawai’i logbook data.
Fishy Reports
Apart from the jarring discrepancies between the logbook data and information on turtle hooking elsewhere, NMFS has additional reason to suspect the reliability of the logbooks. “Under-reporting and non-reporting of incidental catches must be considered,” NMFS’ biological opinion states. As an indicator of logbook accuracy, a sample of logs from 1991 were compared to landings data (market and auction receipts) and only 47 percent of the logbooks submitted to NMFS were considered acceptable.” In 1992, the accuracy rose to nearly 80 percent, NMFS states but adds: “while the accuracy of reports of fish landed can be readily verified when vessels return to port, the take of protected species cannot be verified except by on-board observers.” To the extent NMFS has observed longline fishing trips (11 trips from 1990 to 1992, before the start-up of its most recent mandatory observer program), it has found a far higher degree of interactions with turtles. “Incidental takes of sea turtles were observed on three of the 11 trips. Three leatherback turtles were entangled and released and two olive ridley turtles were caught on baited hooks, one of which was released alive and the other retrieved dead.”
It was on the basis of those observations and not on the Japanese and Spanish fleet data to which Council Chairman Lujan referred that NMFS arrived at an estimated incidental take rate of .061 turtles per 1,000 hooks, which in turn was used to arrive at the estimate of an annual incidental take of 752 turtles by Hawaiian longliners, with 148 of them retrieved dead.
Post-Release Mortality
According to NMFS’ biological opinion, “many if not most of the turtles caught [by longline gear] are still alive when brought on deck.” But, NMFS adds, “a large percentage” of those caught and released alive have probably received injuries that will eventually be fatal.
When a turtle has swallowed a hook, standard practice is to cut the line as close to the turtle’s mouth as possible. When the turtle is released, it appears active, NMFS says, but its ultimate fate, “with the imbedded hook somewhere in the upper gastrointestinal tract, is questionable. The turtle upon release will swim away and may live for days, weeks or months before dying.”
Should the hook perforate the turtle’s gastrointestinal tract, according to NMFS, the turtle maybe expected to die of peritonitis and septicemia. But another factor could cause earlier death. When turtles are hooked, they tend to struggle on reaching the limit of the line. According to the biological opinion, “the resulting stress on the GI tract would produce a damaging condition known as intussusception, or invagination (telescoping) of one segment of the GI tract into the other. Even greater stress would be expected to result when the fishing line is reeled in and any hooked turtle is dragged along through the water column and hoisted aboard.”
NMFS estimates that between 20 and 30 percent of the turtles released alive after being hooked will eventually die as a result of wounds received.
Ingestion of hooks is not the only threat posed by the longline fleet to the turtles. When turtles are on the line, they make easy prey for sharks. Also, forced submergence can cause turtles to drown on the line.
When leatherback turtles are caught, frequently their enormous size (up to 2,000 pounds) makes it difficult for them to be brought aboard the vessel. They are usually released by cutting the monofilament line at a distance, allowing the turtle to swim away with an embedded hook and trailing a length of line. The line itself can kill sea turtles, according to researchers studying turtles off the Florida coast. Turtles released with hook and line attached will suffer the greatest mortality rates of all, these researchers say.1
Finally, the turtles can be killed in the process of being hauled out of the water. An observer of longline vessels fishing in Micronesia reported an olive ridley turtle was injured when its shell was broken by the vessel’s power block. The turtle was “discarded,” the report states.
Search for Solutions
At a workshop held in November at the Honolulu Laboratory of the National Marine Fisheries Service, turtle experts from around the world discussed ways in which the harm to turtles from longline fishing might be reduced. While some of the measures proposed deal with treating turtles after they have been hooked, others seek to identify ways to reduce turtle hooking in the first place. This might be done by making bait less attractive to turtles (either by employing alternative bait or by treating bait in a way to discourage turtles) or by developing “turtle repellents” devices or chemicals that keep turtles away from longline gear. One scientist suggested hooks with wire guards (designed to keep hooks from becoming entangled with weeds) might minimize turtle injury while still catching fish. Another expert proposed an end to the use of plastic light sticks as fish attractors as well as the development of collapsible hooks or biodegradable hooks.
1 Alan B. Bolten, Karen A. Bjorndal, and Helen R. Martins, “Life History Model for the Loggerhead Sea Turtle (Caretta caretta) Population in the Atlantic: Potential Impacts of a Longline Fishery,” in George H. Balazs and Samuel G. Pooley, eds., Research Plan to Assess Marine Turtle Hooking Mortality: Results of an Expert Workshop, Honolulu Laboratory, National Marine Fisheries Service (December 1993).
— Patricia Tummons
Volume 4, Number 10 April 1994