ecember 29, 2003

VIA FACSIMILE (301) 713-1917

Dr. Christopher Rogers

Chief

Highly Migratory Species Management Division

National Marine Fisheries Service

1315 East-West Highway

Silver Spring, MD 20910

Dear Chris:

On November 28, 2003, the National Marine Fisheries Service ("NMFS") announced its intent to prepare a Supplemental Environmental Impact Statement ("SEIS") pursuant to the National Environmental Policy Act ("NEPA") to assess the potential effects of actions under a yet to be promulgated proposed rule relating to turtle bycatch in the Atlantic pelagic longline fishery. 68 Fed. Reg. 66783 (November 28, 2003). Blue Water Fishermen's Association ("BWFA") appreciates this opportunity to submit comments in response to NMFS' notice of intent to prepare the SEIS.

Following three years of successful pelagic longline sea turtle bycatch reduction cooperative research, our fishermen, together with our NMFS and academia research partners, have clearly demonstrated bycatch reductions in excess of the mandate contained within the Endangered Species Act ("ESA") June, 2001 Biological Opinion. It is now critically important for NMFS to implement an appropriate transition from this successful research program to the application of practical measures that meet the legal mandates while providing a reasonable opportunity for U.S. fishermen to fully utilize the U.S. quota for highly migratory species ("HMS") allocated pursuant to the International Convention for the Conservation of Atlantic Tunas ("ICCAT"). The socio-economic benefits of this fishery to the Nation require that the Distant Water Fleet secure access to the Northeast Distant Area ("NED") fishing grounds for the next Grand Banks season beginning in June of 2004. Indeed, the experimental fishery cannot be considered a success unless we achieve a reopening of the NED using practical measures that allow the fleet to operate in a viable manner.

BWFA represents a majority of the active U.S commercial fishermen, vessel owners, fish dealers, and service and supplier companies with an interest in the sustainable pelagic longline harvest of Atlantic swordfish, Atlantic tunas, Atlantic pelagic sharks and mahi-mahi. BWFA's membership has dedicated over a decade spearheading the effective international recovery of the fish stocks that our future depends upon. The severely diminished Distant Water sector of this domestic fishery has worked diligently to address incidental catch issues and to develop practical solutions that satisfy domestic mandates — mandates which undeniably impose higher conservation standards on this fishery than on any other domestic or international Atlantic HMS hook and line fishery.

BWFA and its members also recognize that their industry depends on the continued viability of Atlantic HMS stocks. Thus, BWFA has aggressively pursued international conservation and management of Atlantic HMS stocks throughout their range. Members of BWFA serve on NMFS' Atlantic HMS and Billfish Advisory Panels on the United States ICCAT Advisory Committee ("IAC"), including serving as technical advisors to species working groups which provide advice and support to the IAC, and as members of the U.S. ICCAT Delegation. BWFA members also generally participate in the proceedings of ICCAT and in the development of the fishery management program for Atlantic HMS.

Before offering specific comments on the proposed SEIS, BWFA will place the issue into its historic context, a context which is critical to understanding the results of the NED experimental fishery.

Sea Turtle Bycatch

During the 1960s, 1970s, and the early 1980s, fishermen operating in the Atlantic HMS fishery encountered very few sea turtles. Even sighting sea turtles on the pelagic longline fishing grounds was uncommon. In those years, interactions were nearly non-existent. However, as sea turtle beach protection and shrimp trawl bycatch regulations worked to increase sea turtle populations, loggerhead and leatherback sea turtles have dramatically increased in the areas where we fish. Since the mid-1980s, the increasing number of turtles on the fishing grounds has increased the possibility of turtle interactions.

Our fishermen's first experiences with ESA sea turtle concerns occurred in a joint NMFS, academia and industry section 7 consultation process in 1994. At that time, our representatives were extremely candid concerning all aspects of pelagic longline sea turtle interactions. Many of the fishing techniques and equipment that have been successfully demonstrated through the NED research were first raised by our fishermen and discussed during the 1994 consultation. The Safe Handling and Release Guidelines initially developed by BWFA were subsequently implemented in the Pacific pelagic longline fishery in 1995. At that time, Atlantic pelagic longline sea turtle interactions, although recognized as increasing with the improvement in sea turtle populations, were not deemed to warrant a "jeopardy" finding under the ESA. Unfortunately, from all perspectives, our ideas were shelved for several years.

In 2000, NMFS estimated from two observed NED fishing trips that the "taking" of loggerhead sea turtles in 1999 exceeded the existing Incidental Take Statement ("ITS") issued under the ESA. At that time, BWFA advised NMFS that such "clustered" turtle interactions are rare events that can be substantially avoided by communication among the vessel captains. The extrapolation NMFS engaged in to determine total turtle interactions was not sound. Indeed, the 100% observer program implemented by NMFS has proven BWFA to be correct. Nevertheless, in 2000, a section 7 consultation was initiated under the ESA which resulted in the 2000 Biological Opinion which contained a "jeopardy" finding.

Following the issuance of that Biological Opinion, NMFS promulgated an Emergency Rule in October 2000 which closed an "L-shaped" area within the NED and required the entire U.S. Atlantic pelagic longline fleet to carry approved line cutters and dip nets. The Emergency Rule was finalized by an Interim Final Rule issued in April, 2001. The 2000 Biological Opinion also required gear modifications for leader lengths to be at least 110% of the float line length and the use of corrodible hooks (90 day delay). These requirements were implemented by regulation on July 9, 2002. 67 Fed. Reg. 45393.

The 2001 Biological Opinion

On June 8, 2001, NMFS issued a Biological Opinion on the Re-initiation of Consultation on the Atlantic Highly Migratory Species Fishery Management Plan and its Associated Fisheries ("BiOp") pursuant to a consultation initiated under Section 7 of the ESA. The BiOp concluded that the Atlantic pelagic longline fishery was likely to jeopardize the continued existence of loggerhead and leatherback sea turtles, species protected under the ESA. The BiOp required the closure of the Atlantic Ocean's 2.6 million square nautical mile NED statistical area to U.S. longline fishermen. It also required certain gear modifications and research. BiOp at 111-114. NMFS implemented the BiOp's findings by issuing rules under the Magnuson-Stevens Fishery Conservation and Management Act.

The BiOp's jeopardy finding and corresponding NED closure were based on the BiOp's conclusion that turtle mortality incidental to the NED pelagic longline fishery must be reduced by 55%. BiOp at 111. The 55% reduction number was based on models regarding loggerhead turtles, the accuracy of which BWFA contests. The BiOp did not contain a similar analysis for leatherbacks. Instead, the BiOp assumed that the loggerhead model was sufficient to justify a 55% mortality reduction requirement for leatherbacks. BiOp at 111.

The BiOp found that the NED closure would reduce the number of loggerheads and leatherbacks "captured in the fishery by 51% and 49%, respectively, each year . . . ." BiOp at 111. The BiOp concluded that this level of reduction, coupled with the reduction in takes resulting from the gear modifications required in the BiOp, would achieve the mortality reductions necessary to avoid jeopardy. BiOp at 112.

The BiOp's focus on the NED stems from the BiOp's finding that 75% of the loggerhead incidental takes and 40% of the leatherback incidental takes occurring in the Atlantic pelagic longline fishery occur in the NED. BiOp at 82. The BiOp also determined that 19% of the loggerheads captured in the NED were from the northern nesting group of loggerheads. BiOp at 32-33.

The BiOp identified five nesting groups of loggerheads in the western Atlantic. The BiOp reported that the southern nesting group, typically referred to as the south Florida nesting group, had increased 5.3%-5.4% annually between 1978-1990 and has been increasing approximately 4% since 1990. BiOp at 34. The BiOp also reported that the northern nesting group of loggerheads, occurring from northeast Florida to North Carolina, has been increasing approximately 2.8%-2.9% annually since 1990. BiOp at 34 and 73. According to the BiOp, the south Florida nesting group of loggerheads, with about 83,400 nests each year, contains approximately 90.7% of the loggerhead nests in the western Atlantic. In contrast, the 7,500 nests in the northern nesting group comprise about 8.5% of the western Atlantic loggerhead nests. Only 0.8% of the western Atlantic nests were from Florida Panhandle sites which has about 1,200 nests annually. BiOp at 31-34. The BiOp found that the Dry Tortugas Islands nesting group has about 200 nests per year and the Yucatan Peninsula nesting group in Mexico hasd about 1,000 nests annually. BiOp at 31.

With respect to leatherbacks, the BiOp reported that the two U.S. nesting groups of leatherbacks have been increasing by 10.3% and 7.5% annually since the early 1980's. BiOp at 40.

To analyze overall loggerhead and leatherback population levels, the BiOp found that adult nesting females account for less than one percent of the total population. BiOp at 34. The BiOp concluded that the Atlantic population of nesting leatherback females totals about 15,000 in the western Atlantic, 4,000 in the Caribbean and 4,700 in western Africa, for a total of 23,700. BiOp at 40. Thus, the total Atlantic leatherback population would approximate 2,370,000. Significantly, DNA analyses cannot distinguish between leatherbacks nesting in the Atlantic region based on DNA analyses done on leatherbacks nesting in Florida, French Guiana and Africa. BiOp at 40.

Regarding the overall loggerhead population, the BiOp noted that the two largest nesting aggregations of loggerheads occur on the Masirah and Kuria Maria Islands in Oman and along the southeast U.S. coast. BiOp at 31. The BiOp concluded that the adult female nesting population along the U.S. Atlantic and Gulf coasts was approximately 44,970. BiOp at 33. As to the Oman nesting groups, the BiOp noted that the Masirah Island nesting female population was 30,000. BiOp at 31. The BiOp had no estimate for the Kuria Maria Island nesting group. Thus, the number of U.S. origin Atlantic loggerhead nesting females from the U.S. nesting population is 44,970, which means that the Atlantic population approximates 4,497,000.

To determine the impact of the fishery on these turtle populations, NMFS assumed that turtles which were lightly hooked incurred a 27% post-release mortality and that turtles ingesting the hook incurred a 42% post-release mortality. BiOp at 87. However, according to NMFS observers, no leatherbacks ingested hooks in 1999 or 2000. BiOp at 86. In 1999 and 2000, 84.2% and 81.3% respectively of all leatherback turtles were observed hooked in their flippers. BiOp at 86. In those two years, 72.7% and 82.1% respectively of loggerhead turtles were observed hooked in their beak or mouth. BiOp at 85. The BiOp also determined that using the recently required gear removal equipment, the trailing line when all fishing gear could not be removed had a mean length of 3.25 feet for loggerheads and 3.8 feet for leatherbacks. BiOp at 90.

The Petition To List Certain Loggerhead Nesting Groups As Distinct Populations

Subsequent to issuance of the BiOp, NMFS received a petition requesting that the northern nesting group of loggerhead turtles and the Florida Panhandle nesting group of loggerhead turtles each be reclassified as a distinct population segment ("DPS") under the ESA. Classification as a DPS means that each group would be considered as a separate and distinct population under the ESA.

Currently, NMFS has listed the entire loggerhead sea turtle population as a single unit and has classified it as threatened throughout its range. However, NMFS has determined that it will "treat [loggerhead] sea turtle populations in the Atlantic Ocean separately from those in the Pacific Ocean for the purposes of section 7 consultations under the ESA" since the Atlantic and Pacific populations are separated by large oceanic expanses and are managed separately. 68 Fed. Reg. 53947, 53948 (September 15, 2003). Thus, the effect of the petition, if it had been granted, would be to have the northern nesting group and the Florida Panhandle nesting group considered as separately listed populations under the ESA and separated from the rest of the Atlantic loggerhead population for ESA consultation purposes.

NMFS issued its findings on the requested action on September 15, 2003. 68 Fed. Reg. 53947 (September 15, 2003). In describing the legal test applicable to designating a DPS, NMFS noted that to find that a population is a DPS, NMFS must first determine whether the population is discrete relative to the remainder of the taxon. Under NMFS policy, a population is discrete if it is markedly separated from other populations because of "physical, physiological, ecological, or behavioral factors (quantitative measures of genetic or morphological discontinuity may provide evidence of this separation) . . . ." Id. at 53949.

If a population is discrete, NMFS next considers if the population is biologically or ecologically significant. However, NMFS considers the significance test only if NMFS first finds the population is discrete. If the discreteness and significance factors are both satisfied, a population may be listed as a DPS. Id. at 53949-50.

In applying the test for discreteness discussed above to the northern and Florida Panhandle nesting groups, NMFS found that measures of morphological characteristics such as carapace length and width, and body depth did not show distinctions among the five different loggerhead nesting groups found in the western Atlantic. Id. at 53953. NMFS then cited recent genetic studies indicating that, although some subdivisions may exist, some widely separated populations "have no statistically significant differences" in their DNA analysis. Id. NMFS also noted that genetic analyses for immature loggerheads demonstrate that loggerheads from the south Florida, northern, and Florida Panhandle nesting groups intermix in foraging areas. This is also the case with respect to adult females. Id. NMFS then analyzed the annual variation in nesting activity and determined that it did not indicate a marked separation of the subpopulations. Id.

Furthermore, NMFS found that DNA information indicates "males likely interbreed with females across subpopulations, and thus the subpopulations are not separable on this basis." Id. at 53952. In this regard, NMFS determined that the south Florida nesting group produces 20% male hatchlings, the Yucatan nesting group produces 31% males and the northern nesting group produces 65% males. The sex ratios of hatchlings from the Dry Tortugas and Florida Panhandle nesting assemblages were not assessed. Id. at 53951. Nevertheless, NMFS estimated that the percentage of males produced from the Florida Panhandle nesting group is higher than that of the south Florida nesting group. Id. at 53952. NMFS went on to state that although the south Florida nesting assemblage apparently produces only about 20% males, the total number of males produced is likely to be greater than that produced by the northern and Florida Panhandle assemblages due to the larger size of the south Florida group. Id. at 53953.

After considering all of the data, NMFS found that the various subpopulations "are not markedly separated . . . ." Id. at 53952. NMFS then concluded that "on the basis of the best available scientific and commercial information, we find that the Northern and Florida Panhandle subpopulations of the loggerhead sea turtle are not discrete, and therefore they are not distinct population segments and do not qualify for reclassification" under the ESA. Id. at 53953.

The Experimental Fishery

In 2001, the entire NED (2.6 million nautical square miles) was closed to U.S. longline fishing because of leatherback and loggerhead sea turtle interactions. Amazingly, even while the closure process proceeded, both the National Oceanic and Atmospheric Administration ("NOAA") and the affected fishermen realized that this situation presented an opportunity to initiate a cooperative research program to resolve a global bycatch roblem. While the U.S. pelagic longline fishery is only a minor contributor to this problem, the shared vision was to attempt to develop turtle avoidance and mitigation methods readily exportable to foreign longline fleets in order to benefit global sea turtle populations.

Under the direction of BWFA and the Fisheries Research Institute ("FRI"), a willing segment of the Distant Water fleet was contracted by NOAA to conduct bycatch reduction research in the NED which includes the highly productive swordfish fishing grounds of the Grand Banks of Newfoundland. The objective of the three year research program was to develop gear modifications and fishing techniques to reduce the incidental take and mortality of leatherback and loggerhead sea turtles. Throughout the project, each participating vessel had one or more NMFS observers on board to ensure that all research protocols were met and to document the results.

Research To Avoid Turtle Interactions

In the project's early stages, eight boats fishing 186 research sets with 164,600 hooks tested ideas from the Pacific pelagic longline fishery consisting of blue-dyed squid bait and controlled spacing of hooks. These treatments showed no statistically significant reductions although potentially interesting discoveries included elevated loggerhead interactions occurring with increased "daylight soak time." Surprisingly, satellite tracking showed some turtles swimming southwest against the Gulf Stream currents apparently to return to the U.S. mainland. Vessel Captains participating in the research found that daily communications to avoid migrating sea turtle clusters were helpful but did not achieve the desired level of reduction in turtle interaction.

In the second year of the experimental fishery, thirteen boats made 489 sets fishing 427,385 hooks under an experimental design that included our fishermen's suggestions presented in 1994 of large circle style hooks and mackerel bait, which not only demonstrated reductions up to 92% in sea turtle interactions but also increased the target catch and size of swordfish. Tank tests were also conducted to identify the size and style of hook which would be unacceptable to loggerhead turtles which pursue the baited hook while simultaneously avoiding leatherbacks, which do not go after the bait but are snagged by the hook and/or entangled in the gear.

According to NMFS (www.mslabs.noaa.gov/mslabs/harvest/sea_turtle_mitigation.htm ), the combination of circle hooks and mackerel bait achieved a 92% reduction in interactions with loggerhead turtles. Circle hooks with squid bait reduced incidental loggerhead bycatch by 85%. The use of mackerel bait alone, regardless of hook type, reduced leatherback bycatch by 66% with J hooks and by 67% with circle hooks. Circle hooks with squid bait achieved a 50% reduction in leatherback interactions. These percentages are comparisons measured between the control hook/bait and the treatment hooks and baits used in the experiment. The control hook and bait were essentially the standard 9/0 big game fish hook with a 25-30 degree offset and baited with whole squid. This was the gear that was generally used in the NED fishery prior to the BiOp.

In 2003, eleven boats made 539 longline sets and fished 577,423 hooks to continue to explore practical reductions through fine tuning the treatments that were successful in 2002. The research team also utilized Hooking Time Recorders ("HTRs") and Time Depth Recorders ("TDRs") to measure any temporal and/or spatial differences between target and bycatch species. Using an inflatable Zodiac, one vessel successfully captured "control" loggerhead turtles for satellite tagging and tested a leatherback lifting cage and hoist. Continued satellite tagging will be important to determine post release stress and will add to our knowledge of sea turtle migrations.

The accomplishments of the NED research are remarkable. Compared to the turtle catch per unit of effort ("CPUE") data compiled in 1999, there was an 83.8 % reduction in overall loggerhead interactions and a 70 % reduction in overall leatherback interactions during 2002. The preliminary figures for 2003 are a 88.3 % reduction for loggerheads and 86.2 % for leatherbacks.

Research to Reduce Harm

In the 2002 experimental fishery, the use of circle hooks also resulted in a change in hooking location which is expected to cause an increase in turtle post-release survival. Out of 95 loggerhead turtles captured in 2002, only 11.6% were on the experimental hooks and, of those, only 3 circle style hooks were ingested. In the same year, leatherbacks were most frequently entangled or hooked externally. Only rarely was the hook found in or near the mouth. Indeed, of the 121 leatherback turtles which were hooked, 113, or 93.4%, were hooked externally. Of the hooks which were snagged in or near the mouth, none were swallowed.

In addition, the NED research team's efforts to develop mitigation tools and protocols to carefully release those turtles that could not be avoided progressed continuously. Dip nets for lifting small turtles onboard, dehookers, and line cutters for disentangling all turtles were among the tools developed. The team was extremely fortunate to have NMFS and industry gear specialists who converted ideas into exceptionally useful tools and who worked together to have those tools rapidly deployed for onboard evaluation.

The 2002 experiment clearly demonstrated improvements in the removal of gear from hooked turtles through the use of this equipment. During the experiment, specific gear removal tools were provided to the fishermen. Line cutters and dip nets were the required equipment on all pelagic vessels at the time of the BiOp. 65 Fed. Reg. 60899 (October 12, 2000), 66 Fed. Reg. 17270 (March 30, 2001). During the 2002 experiment, fishermen were also issued long handled dehookers, short handled dehookers, mono cutters, bolt cutters, needle-nose pliers, mouth openers, and mouth gags.

Using this additional equipment, NMFS found that of the 36 loggerheads hooked externally, the hook was removed in 35 cases. Where the hook had been swallowed, it was removed in 7 of 59 instances. For leatherbacks, where most of the hooking was external, the hooks were removed from external locations 46% of the time. The hook was removed 33% of the time when it was in or near the mouth. No leatherback swallowed a hook. If complete gear removal was unsuccessful, the trailing line left in place was generally less than 3 centimeters, or 1.2 inches, in the case of loggerheads. For leatherbacks, where hook and line remained, trailing gear was usually less than 0.6 meters, or 23.6 inches.

In 2003, the NED research continued to develop new technologies and methods for hook and gear removal and the 2003 data will show significant increases in removal percentages. For turtles too large to boat, a turtle holding tether was introduced late in the season, but showed immediate improvements for leatherback turtles. . Also, for loggerhead turtles ingested hook removals increased because the dehooking protocols were revised for the 2003 research program to allow the removal of deeper hooks providing the eye of the hook is visible. The 2003 NED Research program results will be released in January 2004.

Exportability

The combined impact of all U.S. hook and line fisheries on endangered and threatened sea turtles is insignificant relative to the effect of foreign hook and line fleets. Indeed, the U.S. fleet accounts for only 5-8% of the total hooks fished in the Atlantic Ocean. ( BiOp at 37) . For this reason, the experimental fishery research focused on gear modifications and mitigation tools/protocols that are exportable to other fleets.

The Careful Handling/release tools (technologies) pioneered by the NED team and tested during the experimental fishery are already being exported at a much higher rate than anyone involved in the program had previously imagined. Over a dozen different foreign fleets have already ordered some equipment from the gear manufacturers that helped design and fine tune this equipment during the research program. At this point, over thirty-five countries have made inquiries. Fishermen do not wish to leave hooks and gear on a turtle and tools that can more efficiently disengage turtles are a benefit to the fishermen and the turtle.

Slightly offset (an offset is critical for baiting practicality) circle style hooks will be the most beneficial and easily exportable item. Circle hooks will be adopted more quickly in tuna directed fisheries not only because they increase the directed catch but also because there is a significant tendency of the hook to lodge in the corner of the jaw resulting in a catch that remains livelier for longer on the gear providing an opportunity for better processing and a higher quality product. Improving the quantity and quality of the target catch provides a compelling economic incentive for foreign fleets to use this type of equipment.

The results of the experimental fishery regarding the effect of different bait types will be used in varying ways by foreign fleets. For the most part, requiring specific bait types and or sizes is not exportable because specific bait types and sizes are not available in all areas at all times. However, what is readily exportable to foreign fleets is the knowledge that directed swordfish fishing can be enhanced by using a large mackerel (and perhaps other types of whole finfish) baits on an 18/0 or greater circle hook. What is also clear is that this type of information developed through the experimental fishery will be adopted by foreign pelagic longline fisheries with little to no focused effort by NMFS because of the economic advantages of this gear.

Application of Other Experimental Data

In applying data from the NED experimental fishery, BWFA cautions NMFS against immediately assuming that all longline fisheries are the same and that data from other sea turtle bycatch research programs in the Azores, Canada, Hawaii and elsewhere can automatically be applied to Atlantic pelagic longline fisheries. For example, while research in Canada can be very helpful regarding hook and bait type because it is a relatively similar swordfish directed fishery, there are many significant differences between the Canadian and U.S. fisheries that preclude complete use of the Canadian data. Canadian gear is generally fished much shallower in the water column because of the availability of colder water edges, especially in the shallower Continental Shelf areas where bottom contours create upwelling which enhances these frontal edges. The availability of these prime relatively colder water edges provides the Canadian fleet with higher CPUEs for swordfish, and may result in lower CPUEs for loggerhead turtles but potentially higher CPUEs for leatherback turtles.

The Azores study has very limited fishing effort that is primarily directed on blue sharks. The rigging of segments of wire to retain sharks renders the gear non-comparable to swordfish directed fisheries. Other than research results relating to hook placement in the turtles (relatively similar size loggerhead turtles), most of this research cannot be applied in the NED or compared with the NED or Canadian data.

The early Hawaii studies, especially the satellite tagging studies, are hindered by the inefficiency of initial technologies that required substantial fine-tuning over time (i.e. battery failures). Other, potentially important "stealth" gear experiments (various color mainline, leaders, ball-drops, floats, etc.) have not progressed due to continuous lawsuits by environmental activists.

BWFA does look forward to a reasonable and practical application of the successful use of large circle hooks and mitigation tools in other U.S. and foreign hook and line fisheries in the future. We are confident that each fishery can seek ways to use these technologies to benefit both the populations of sea turtles and other bycatch species and the viability of commercial fisheries.

Recommendations Based on the Experimental Fishery

NED Area

Based on the results of the experimental fishery, BWFA recommends that the NED be reopened under practical requirements to use an 18/0 or greater circle style hook with a 10 degree offset (appropriate hook width measurements should be the determining factor). This was the gear used in the experimental fishery which achieved the greatest reduction in turtle interactions. BWFA does not recommend a bait requirement but notes that fishermen will have an incentive to use mackerel when engaged in directed swordfish fishing and to use squid when engaged in directed tuna fishing. This bait flexibility is critical to the present and future viability of the fishery.

BWFA further recommends that longline fishermen operating in the NED be required to use all of the gear removal equipment successfully employed in the experimental fishery. New Captains (other than NED research program Captains) should be required to review, or be trained in using, the Careful Handling/Release tools and protocols. Finally, BWFA recommends that NMFS maintain an observer program for the reopened NED fishery to document the level of implementation success of these protocols prior to phasing observer coverage to the standard required throughout the Atlantic HMS fishery.

Outside the NED Area

Without recent data on the incidental take of sea turtles in the non-NED pelagic longline fishery, we can only make recommendations based upon our general knowledge of the fishery. The comparative results of the NED experimental fishery demonstrate that the standard 8/0 _ 9/0 J-style hook results in more turtle interactions and a greater likelihood of hook ingestion than does the gear employed in the experimental fishery. The Japanese tuna style hook is even more harmful to turtles. It is very apparent that the fishery must move toward a broader use of the type of gear used in the experimental fishery. However, we strongly recommend that NMFS proceed cautiously and not arbitrarily impose requirements beyond those proven by the current data for the areas studied. Thus, we recommend that NMFS require that the non-NED pelagic longline fishery use a 16/0 or greater slightly offset circle hook. We further recommend that NMFS undertake a research program to determine the economic and biological impacts of this requirement in the non-NED coastal fishery.

BWFA also recommends that the non-NED coastal fishery Captains be trained and be required to use the gear removal equipment employed in the NED experimental fishery. Finally, we recommend that NMFS maintain an appropriate observer program in the coastal fishery but also not ignore the importance of similar monitoring across all other HMS hook and line fisheries.

Recalculation of the Incidental Take Statement

The BiOp contained an Incidental Take Statement ("ITS") which was intended to prevent jeopardy. However, the ITS was based on incidental interactions, not on injury and mortality. Although BWFA recognizes that NMFS assumes a portion of the incidental take results in post-release mortality, the critical element to prevent jeopardy to turtles is to reduce injury and mortality. Thus, the ITS should be recalculated to reflect the reduced post-release mortality expected to be achieved due to the reduced turtle interactions and to improvements in gear removal equipment. NMFS must also consider the return of normal levels of U.S. effort in the NED area and the increasing sea turtle populations.

Furthermore, the ITS in the BiOp for loggerhead turtles was calculated based in large part on NMFS' focus on the northern nesting group and the BiOp's assertion that 19% of the loggerheads taken in the NED came from the northern nesting group. Given that the genetic sampling data developed during the experimental fishery shows that the BiOp was incorrect and that approximately only 5% of the loggerheads incidentally taken in the NED are from the northern nesting group, NMFS must recalculate the loggerhead ITS.

Impact of the Proposed Regulatory Program in the NED

To place the results of the experimental fishery into context, recall that the BiOp found that "[t]o comply with its obligation to remove jeopardy [to loggerhead and leatherback turtles], NMFS must take action to reduce the impacts of the U.S. pelagic longline fishery . . . ." BiOp at 111. The principal action taken was closure of the NED. The BiOp found that based on NMFS observer reports the NED closure would reduce the number of loggerhead and leatherback turtles captured in the NED "by 51% and 49%, respectively, each year . . . ." BiOp at 111. The BiOp stated that this "reduction in takes in the NED, in conjunction with the expected reduction in takes resulting from the [required] gear modifications" would be sufficient to remove any jeopardy to the turtles from the Atlantic pelagic longline fishery. BiOp at 112.

The experimental fishery establishes conclusively that the NED can be reopened and meet the incidental take reductions sought in the BiOp. With respect to loggerheads, the 2002 experimental fishery showed that fishing with a circle hook reduced interactions by 92% with mackerel bait and 85% with squid bait compared to the control. The reduction in loggerhead interactions relative to the 1999 CPUE was 84% in 2002 and 88% in 2003. All of these numbers far exceed the 51% reduction in loggerhead incidental takes which the BiOp asserted would be the effect of closing the NED. With respect to leatherbacks, he 2002 experimental fishery showed that using circle hooks with mackerel bait reduced incidental leatherback interactions by 67% while circle hooks baited with squid reduced leatherback turtle interactions by 50%. The reduction in leatherback interactions relative to the 1999 CPUE was 70% in 2002 and 86% in 2003. Again, all of these numbers exceed the 49% incidental take reduction for the NED called for in the BiOp.

A significant component of the BiOp's jeopardy finding was the extent of the post-release turtle mortality which NMFS believed was occurring. This mortality was purported to be caused by the hooks and by trailing line left on the turtle. The experimental fishery demonstrated that required use of the additional Careful Handling/Release equipment, and the use of larger hooks which reduce the likelihood of hook ingestion, will reduce post-release turtle mortality. For example, when the new gear removal equipment was employed, the amount of trailing line left when fishing gear could not be completely removed from the turtle fell from 3.25 feet in the pre-BiOp fishery to 1.2 inches for loggerheads and from 3.8 feet to 23.6 inches for leatherbacks.

The experimental fishery also confirmed that it is a rare event if a leatherback turtle swallows a hook. As in 1999 and 2000, no leatherbacks ingested a hook in 2001, 2002, or 2003. The likelihood of ingestion in the future is also remote because, under the regulation proposed in this comment, the hook will be the same one that yielded no ingestions in 2001, 2002 and 2003. Indeed, the hook size and style in the NED fishery under the proposed regulation will be for a circle style hook of 2 1/8 inches wide as compared to the standard J style 1 ½ inch wide hook. The research data documents the substantial impact of these differences.

Use of the larger hooks in the experimental fishery also reduced the rate of hook ingestion by loggerheads since only 27.3% of loggerheads caught with the 18/0 circle hooks swallowed the hook. Equally important, most loggerheads which were hooked externally and in the mouth during the experimental fishery were hooked in places where the gear could be removed. In 2002, for example, the hooks were removed in 35 of 36 instances. Implementing a requirement to use the gear removal equipment employed in the experimental fishery will result in mortality reductions much greater than those anticipated by the gear modifications sought in the BiOp.

Adoption of the proposed regulation will also have minimal impact on the leatherback and loggerhead sea turtle populations. NMFS has estimated the total Atlantic female leatherback nesting population at 23,700. BiOp at 40. If nesting females comprise less than 1% of the total population, the Atlantic leatherback turtle population is 2,370,000. According to the BiOp, as many as 202 leatherbacks could be killed annually by Atlantic pelagic longlining. BiOp at 105. Although BWFA contests that calculation in the BiOp, the facts are that under the regulation proposed in this comment, interactions and, therefore, resulting mortalities will be reduced by 50-86%. In other words, even without the post-release mortality improvement expected to be achieved from the additional gear removal requirements, mortality would be expected to drop by 50%-86%, a reduction in mortality of 101-174 animals. Thus, the mortality attributable to pelagic longlining would approximate 28-101 leatherbacks. As already noted, this estimate does not include the mortality reduction expected to be achieved by use of improved gear removal equipment and techniques. When the possible mortality rate of 28-101 turtles attributable to fishing in the NED is compared to a population of 2,370,000, the impact of the NED fishery is clearly de minimus. Even if the population were a magnitude smaller, the NED fishery would only account for a small fraction of 1% of the total population. A jeopardy finding based on these numbers could only be considered capricious. Indeed, a finding that this level of mortality causes jeopardy would have serious ramifications for other activities which have similar or greater mortality levels.

With respect to loggerheads, NMFS estimated that 44,970 females nest along the U.S. and Gulf coast. BiOp at 31 and 33. Given NMFS' assertion that nesting females are less than 1% of the total population, that produces an approximate Atlantic Ocean loggerhead population estimate of 4,497,000 animals. In the BiOp, NMFS estimated that the mortality caused by the pelagic longline fishery could reach as high as 362. BiOp at 102. Although BWFA contests this calculation, implementation of the regulation proposed in this comment would reduce incidental taking and, therefore, mortality by 85%-88%, even without the improvement in the post-release mortality anticipated from utilization from the new gear removal equipment. Thus, the impact of reopening the NED under the proposed regulatory regime would be a reduction in incidental mortality of 308-319 animals. In other words, the loggerhead mortality expected to be attributed to pelagic longlining would approximate 43-54 turtles. And this number does not include the decreased mortality expected to result from use of the improved gear release equipment and techniques. Again, comparing that expected mortality with a population of approximately 4.5 million loggerheads can only result in the conclusion that the impact is de minimus. Any jeopardy finding given these facts would be arbitrary. Further, a jeopardy finding based on this level of interactions will have ramifications for other activities with equal or greater interactions.

In this regard, it should be noted that the ESA's consultation requirements apply to listed species. 16 U.S.C. § 1536(a)(2). Thus, NMFS' analysis must be with respect to the listed species and not some subset thereof. This is particularly true given NMFS' determination that, based on the best scientific data available, there is no basis to consider the northern nesting group of loggerhead turtles as a distinct population segment for which consultation would be required separately. However, even if NMFS had reached a different result regarding whether northern loggerheads were a distinct population segment, the conclusions would be the same.

In considering the impact of the NED turtle interactions on the northern nesting group, it is significant that the first 119 genetic samples taken during the experimental fishery showed that only 5% of the loggerhead turtles captured in the NED came from the northern nesting group. In addition, 88% of these NED samples were estimated to be from the South Florida nesting group, which the BiOp found is now increasing at an average annual rate of 4% meaning an average of 81,000+ loggerhead turtles are added to the population each year. A full report of all NED research genetic samples is expected in January, 2004.

The estimated number of nesting females in the northern nesting group is 2,976. Given that adult nesting females comprise less than 1% of the population, BiOp at 34, the northern nesting group is comprised of approximately 297,600 animals. If, based on the more recent genetic sampling conducted as part of the experimental fishery, only 5% of the loggerheads taken in the NED originate from the northern nesting group, then only 5% of the expected mortalities, 2-3 animals, can be expected to come from the northern nesting group. Again, the number is de minimus.

In examining the potential impact of the U.S. pelagic longline fishery in the NED upon leatherback and loggerhead turtles, it is also worth bearing in mind that NMFS itself admits thousands of turtles are being taken annually by a wide array of human activities and natural phenomena completely unrelated to the U.S. pelagic longline fishery. BiOp at 74. In fact, NMFS concedes that these other human activities and natural phenomena pose a much larger and more serious threat to the survival and recovery of sea turtles than does the U.S. Atlantic pelagic longline fleet. BiOp at 74. Indeed, NMFS has noted that the United States and 26 other nations participate in longline fisheries throughout the western and northern Atlantic Ocean and that the U.S. fleet accounts for only 5-8% of the total number of hooks fished in the Atlantic Ocean. BiOp at 64 and 37. Moreover, the 28-101 leatherbacks which, under the proposed regulation, might be expected to experience incidental mortality as a result of the pelagic longline fleet pales in comparison to the several thousand leatherbacks which are considered by NMFS to be killed incidental to other U.S. fisheries and activities, none of which NMFS has determined are causing jeopardy to the leatherback population despite that level of mortality.

Swordfish is the primary target species in the pelagic longline fishery. During the 2002 experiment, swordfish catch increased approximately 30% using circle hooks with mackerel bait and 63% when the J-control hooks were employed with mackerel bait. The average size of the individual swordfish caught also increased when mackerel bait was used. Swordfish catch was reduced by 29-33% when circle hooks were used with squid bait. An important secondary target species in the pelagic longline fishery is bigeye tuna. The 2002 experimental fishery determined that circle hooks using mackerel bait reduced the catch rate of bigeye tuna by 81%. It will be critical for our fleet to retain the ability to use squid for tuna directed fishing and mackerel for swordfish directed fishing.

The use of circle hooks will reduce harm across all unintentional catch releases including marine mammals, unwanted sharks, rays, marlins, sailfish, small swordfish, other non-marketable finfish and non-allowed tunas due to size restrictions and/or season/quotas. Furthermore, the benefit of circle hooks, which generally result in hooking in the corner of the jaw, is that these hooks will allow all incidental catches to remain livelier on the gear until released. The use of more advance technologies to disentangle and remove hooks from catches that are to be released will also dramatically reduce harm. When this is compared to the normal J-style hook which is more easily deeply ingested and unremovable, the potential to substantially increase post-release survivability of bycatch species is enormous.

Economic Impact

If the NED remains closed, it will have a significant and negative adverse economic impact on the pelagic longline fishery. If vessels are forced to leave the NED, it is estimated that they will experience a reduction in gross revenue of up to 85%. There are 12-14 vessels which have operated almost exclusively in the NED during the summer season and 15-20 additional vessels with a history of making some Grand Banks trips.

If the NED is closed, the 12-14 vessels which operate primarily in the NED have no real alternatives.

Indeed, the annual economics of the NED fleet is heavily based on the 3-5 trips that these vessels made to the NED. NED vessels are unique in the U.S. pelagic longline fleet in that they are larger and capable of operating 4-6 weeks at sea at distances more than 1,000 miles from shore. This increased capacity means the vessels are more expensive to operate than coastal vessels and cannot operate economically in coastal areas. Furthermore, there are no other fisheries in which these vessels can participate because nearly all U.S. fisheries operate under limited access systems prohibiting new entrants and because the cost of converting a longliner to another fishery is prohibitive. Thus, if the NED is closed, the NED fleet will either go out of business or reflag to other nations. Despite ICCAT's efforts to control capacity in Atlantic HMS fisheries, several developing nations continue to solicit U.S. vessels to reflag.

It is extremely difficult to estimate the economic impact of measures that are not yet proposed by the agency. Furthermore, the scientific and economic data collected during the three years of NED research are very extensive but the in-depth analyses which must be conducted will not be completed by the end of this comment period. However, the following options are logically unacceptable because their economic impacts would be devastating.

· Requiring specific baits would eliminate the flexibility necessary for each vessel to pursue the greatest directed fishing market opportunities within their particular area given that each boat is limited by range and trip length.

· Requiring that fishing occur in a specific range of sea surface temperatures is not feasible because each geographical area of these wide-ranging fisheries is different. In the NED, it may be cautious at times to avoid turtles by fishing in water cooler than 68 degrees Fahrenheit. However, during the entire fishing season there may not be any such cooler water in other areas of the fishery outside of the NED. Moreover, it does not seem appropriate to prohibit fishing if the fish are on the warm side of the current and no turtles are present in that particular area. Requiring that fishing occur in waters with certain temperatures may also have negative impacts on the ability to avoid unwanted bluefin tuna or shark catches which can occur on the colder water edges.

· Requiring reduced daylight soak time has already proven to be unsafe and counterproductive because the crews must begin hauling during the night when swordfish that have just taken the hook ( are too lively) , get pulled off the circle hooks during the hauling, and are lost (up to 60% losses of target catch in the NED in 2002).

· Requiring non-offset hooks would make baiting with some baits including mackerel and other whole finfish baits impractical. To remain profitable, baiting during the setting out must be accomplished in a time efficient manner.

Alternatives

Alternative 1

Option A

The preferred alternative is that contained in the BWFA recommendation discussed above; including, (1) require an 18/0 or greater slightly offset circle style hook for pelagic longline fishing within the NED area; (2) as an interim measure, require a 16/0 or greater slightly offset circle style hook for all U.S. Atlantic pelagic longline fishing outside the NED area, while; (3) research to determine the appropriate size hook for this coastal fishery proceeds; (4) require training and the use of all the Careful Handling/Release tools found successful in dehooking and removing gear from sea turtles during the NED experimental fishery; and (5) implement an appropriate observer program to monitor the implementation of these new technologies with a plan to phase down to an appropriate level of Observer coverage for all HMS hook and line fisheries, as quickly as feasible.

We cannot stress enough the importance of undertaking a research program for the U.S. Atlantic pelagic longline fishery outside of the NED to determine the economic impact of 18/0 versus 16/0 circle hooks on secondary catches of yellowfin tuna, albacore tuna and mahi-mahi (primarily directed tuna fisheries with significant secondary catches of swordfish and mahi-mahi). Simply requiring the use of a circle hook throughout the fishery is a major change for our fishermen and it is critical to fully assess the economic and biological impact of that requirement before any such requirement is made permanent for the coastal fishery.

With respect to the use of the Careful Handling/Release tools found successful in dehooking and removing gear from sea turtles during the NED research program, it will be important for NMFS to require an on-line review and/or mandatory workshop on the use of these tools/protocols throughout the fishery. NMFS should also make an effort to export these gear techniques and equipment to foreign fleets.

The preferred alternative appropriately utilizes what we have learned thus far through solid research and from scientific data. It also avoids imposing permanent requirements on the coastal fishery before these requirements have been subject to complete scientific testing. The shift of the entire pelagic longline fishery to circle style hooks (regardless of size), which will result in corner of jaw hooking, will not only benefit sea turtles but it will also increase the post-release survivability for all incidental catches; including marlin, sharks, small fish, etc. Significantly, the use of circle hooks in different sizes for the appropriate application (pending additional research) is also "exportable."

Requiring a specific bait type is unnecessary and must be avoided because different baits are required for tuna versus swordfish directed fishing. While swordfish fishing, Captains will have ample incentive to use a mackerel and/or other finfish bait because of the increased catch and size of swordfish. However, requiring a mackerel or finfish bait exclusively would reduce directed tuna fishing to approximately 15% of its current efficiency. Allowable tuna (BET, YFT, ALB) do not prefer finfish baits to squid baits in many areas of the fishery as was demonstrated by the NED scientific data. Also , a mackerel only bait requirement would virtually eliminate economically important secondary catches of mahi-mahi and other finfish. Requiring specific bait types would likely force many remaining vessels in this fishery out of business. NMFS must also recognize that a specific bait requirement is not "exportable" to other fleets.

Option B

If there is any chance that the final rule reopening the NED may be delayed beyond the May/June start of a normal Grand Banks season, then we recommend: (1) an Experimental Fishery beginning June 1, 2004 with the requirements set forth in Option A above for the NED and remaining in place through November 2004, unless superseded by a final rule reopening the NED; and (2) 100% observer coverage to monitor the use of 18/0 slightly offset circle hooks and all of the Careful Handling and Release tools/protocols. Extension of the ESA Section 10 Permit has already been approved. An Exempted Fishing Permit would be required. Compensation would not be necessary to monitor the full implementation of the gear requirements and mitigation tools/protocols expected to be required by the final rule. Any additional research (treatment hooks/baits, Search/Capture/Sat Tag loggerhead control turtles, and/or Leatherback Lift/Sat Tag) that would prevent a vessel from optimal fishing could be negotiated through contract to FRI. This arrangement would give NMFS researchers and managers the time to fully complete the 2003 analyses and the rulemaking. BWFA initially proposed this alternative in late September and remains willing to discuss it with NMFS.

Alternative 2

Under this alternative, NMFS would: (1) require an 18/0 or greater slightly offset circle style hook baited with large mackerel (or other finfish) bait for all U.S. Atlantic pelagic longline directed swordfish fishing; (2) require a 16/0 or greater circle style hook for all U.S. Atlantic pelagic longline directed tuna fishing; (3) undertake research to determine the economic impact of 18/0 versus 16/0 circle hooks on secondary catches of allowable tunas and mahi-mahi; (4)require all the Careful Handling/Release tools found successful in dehooking and removing gear from sea turtles during the NED experimental fishery; and (5) implement an appropriate observer program that includes an equitable plan to monitor all HMS hook and line fisheries.

This approach is not preferred principally because it would be difficult to enforce. This alternative does have merit in that it would allow the flexibility necessary for the fleet to remain economically viable while research is conducted to determine the appropriate hook size for the coastal fishery. Further, this requirement is not "exportable." Finally, as noted above, unless initial training workshops and/or on-line verifiable reviews are developed, the full potential of the Careful Handling and Release tools/protocols may not be realized.

Alternative 3

Under this alternative, NMFS would: (1) require an 18/0 or greater slightly offset circle style hook to be used exclusively for all U.S. Atlantic pelagic longline fishing; (2)require all the Careful Handling/Release tools found successful in dehooking and removing gear from sea turtles during the NED research program; (3) implement an appropriate observer program; and (4) undertake a research program to determine the economic impact of 18/0 circle hooks on target and secondary catches; (4) develop an fair and balanced Observer coverage program for all HMS fisheries.

This alternative would not be acceptable fleet-wide because we simply do not know the economic impact of moving directly to the huge 18/0 or greater circle style hooks. During recent years, there has been significant testing of 16/0 circle hooks and confidence in these hooks has increased steadily. The Canadian pelagic longline fleet has shifted to a 16/0 hook but their fishermen continue to resist the larger size hooks. Generally, the 18/0 circle style hook has not been tested in most areas of our fishery.

As discussed above with respect to other alternatives, training would be the most appropriate way to obtain the maximum benefit from requiring the use of the gear removal equipment developed in the NED experimental fishery.

Alternative 4

Under this alternative, NMFS would adopt the preferred alternative (Alternative # 1, Option A above) except that NMFS would close any coastal "hot-spot" sea turtle interaction area using time and area closures to prevent circumstances in one area from impacting other areas of the U.S. Atlantic pelagic longline fishery. If this alternative is adopted, NMFS will need to conduct research to determine appropriate alternatives to these time/area closures.

The NED area was closed for the past three years because it was a sea turtle interaction "hot-spot." It has been suggested that a similar situation is now occurring in another area of the fishery. We recommend that NMFS not use any time/area option unless absolutely necessary to prevent one area of the fishery from closing down the entire fishery. BWFA has continuously recommended dividing this fishery into its natural geographical segments to logically deal with protected species and avoid the unnecessary closure of the entire fishery due to problems in a single area.

Conclusion

The BiOp reached certain conclusions about what needed to be done to avoid jeopardy to sea turtles by the Atlantic pelagic longline fleet. To achieve that goal, the BiOp concluded that closure of the NED was required. The experimental fishery demonstrated that implementing the regulation proposed in this comment will achieve a reduction in turtle interactions significantly greater than that contemplated by the BiOp's NED closure. Equally important, the experimental fishery demonstrated that the gear removal equipment proposed in this comment will decrease post-release mortality significantly beyond the levels contemplated in the BiOp. This fishery's move to circle hooks and sophisticated mitigation tools will also have significant unquantified benefits across all released bycatch species. Allowing this fishery to operate as a viable fishery following successful cooperative research is essential to the "exportability" of these technologies to benefit sea turtles populations and the conservation of other internationally-shared bycatch species.

Sincerely yours,

Nelson Beideman

Executive Director,

Blue Water Fishermen's Association