The bulk of the eastern Pacific halibut fishery occurs in relatively shallow waters of the coastal shelf from the US Pacific northwest through the Bering Sea and Aleutian Islands.  This reflects the summer distribution of the species, since no targeted winter fishery is allowed.  However, we are examining the possibility of a limited winter fishery, primarily in response to market concerns raised by the industry.  While we believe that winter fishing can be conducted with little negative biological impact to the stock overall, we are concerned that biomass distribution is considerably different in winter than summer.  Spawning occurs during winter along the shelf break, from at least the Queen Charlotte Islands to the Pribilof Canyon.  Rather than simply moving offshore from the locations at which they feed, conventional tagging data suggest that many halibut may move considerable distances along shore during seasonal migration.  Halibut appear to aggregate in the northern GOA during winter, and may emigrate from British Columbia and areas south of the Alaska Peninsula.  As a result, fishers in areas from which halibut emigrate in winter would likely experience reduced catch per unit effort (CPUE) relative to fishers in areas where halibut tend to aggregate.  Disparity in catch rates between regions may be nontrivial.  Thus, winter fishing may require that some areas be managed at lower fishing effort than others, or that fishing only be allowed in certain regions.  In order to best structure a winter fishery, we must understand seasonal migration and winter distribution.

The IPHC has a history of conventional tagging studies to address a variety of issues, but the vast majority of returns occur during the summer when the majority of fishing takes place.  In this study we have begun using Pop-up Satellite-transmitting Archival Tags (PSATs) to study seasonal migration.  PSATs represent a novel technology that has proven effective for use on halibut in Alaska, and allow us to study movements and behavior without the need to recapture fish.  PSATs are an archival tag, meaning that they collect and store temperature, depth and light data at regular intervals (every four minutes in the case of our tags) while attached to the fish.  PSATs are relatively large, measuring about 1” in diameter and 6” long, with a 1½” x 2” float and 5” antenna on the upper end.
 

IPHC sea-sampler Ayala Knott applys a PSAT to a large halibut during the 2002 summer setline survey.  (Photo by Bonnie Gauthier.)

The tags are attached to fish via a dart and flexible leader anchored in the fish just below the dorsal fin, about half way along the body.  On a pre-programmed date, the tag releases from the leader, floats to the surface (i.e., pops up), and stored data are transmitted to us via satellite.  More importantly, the satellite that receives the signal determines the tag’s final location, in much the same way that an EPIRB is used to determine the location of a vessel.  Thus, we can determine where the fish was on the pop-up date.  If a fish is captured with a PSAT attached, we can offload highly detailed environmental data, retrieving temperature and depth readings taken every four minutes throughout the fish’s time at liberty.  After the body of the tag releases, the fish will still carry the leader, which can then serve as a conventional "spaghetti" tag, giving us information about whether fish return to their tagging location in the future.  Thus, the IPHC offers a $500 reward for the return of any PSAT tag-body that is recovered, and a tagging-program hat for the return of tag information for fish that bear only the leader.

Twelve halibut were tagged during IPHC's 2002 summer setline survey in the Gulf of Alaska.  Tagged fish were released along two lines: one line in the western Gulf of Alaska extending from Sanak Island in the south to near Seward in the north, and a second in the eastern Gulf from the north end of Vancouver Island to Icy Strait.  Tagged fish ranged from 106-150 cm forklength.  PSATs were programmed to pop up on January 15, 2003.  In addition, collaborators at the US Geological Survey in Anchorage (Dr. Jennifer Nielsen) and the University of Alaska, Fairbanks (Andy Seitz: aseitz@ims.uaf.edu) expanded the tagging into the Bering Sea.  Using vessel support from the local commercial fishery, Andy deployed 12 tags at St. Paul Island in the Pribilofs.  These fish were tagged during 3 fishing trips, with 4 fish tagged during each trip.  Nine were tagged with PSATs programmed to pop up in early January 2003, and 3 programmed to pop up in early May 2003.
 

Of the 12 tags deployed in the Gulf, 10 successfully popped up and transmitted their locations and behavioral data.  The two lost tags were both deployed in southeast Alaska's Area 2C, and the reason for their loss is unknown.  They may still be attached to the fish, may have fallen off prematurely, or the fish may have died prior to the pop-up date.  In any case, the 10 tags that transmitted have provided us with some interesting information that suggests at least two general movement patterns.  Four of the tagged fish appeared to move little during their time at liberty, being located in January very near to where they were tagged during the summer.  One of these fish was tagged in Dixon Entrance near Cape St. James, B.C., one was tagged just offshore of Baranof Island, Alaska, and the other two were tagged along the southern Alaska Peninsula.
 

The remaining six fish moved considerable distances between tagging and pop-up.  The two long-distance champions were a fish that was tagged off the Shumagin Islands and moved to the mouth of the Yakutat Valley, and another that was tagged off northern Vancouver Island and moved to central southeast Alaska.
 

In summary, the Gulf tag returns largely support conventional tagging studies, even though the number of tags deployed was admittedly quite few.  First, it appears that the northern Gulf is somewhat of a wintertime "attractor", given that more of the tagged fish were located in Area 3A on the pop-up date than had been tagged in that Area.  This also appears to be true of Area 2C (southeast Alaska), although that conclusion is a bit more tenuos since only one of the tags released in 2C successfully transmitted to the satellite.  Regardless, the tag locations also support the contention that at least some proportion of the local populations remain at potential spawning sites at the southern edges of Areas 2 & 3, such as around Cape St. James and the within the Shumagin Gully.  Interestingly, two fish appear to have been located considerably farther inshore in January than their summer tagging locations: a fish tagged south of Seward moved into Cook Inlet, and one tagged near Sitka appears to have moved into Chatham Strait.  These are intriguing results that run slightly counter to the notion that halibut should be found in deep water along the continental shelf edge during mid-winter, where they are known to spawn.
 

Of the tags deployed at St. Paul, 7 of the 9 with winter pop-up dates successfully released and transmitted.  One fish was recaptured less than two weeks after tagging, but that fish carried a tag programmed to pop up during early May.  Thus, two fish remain at liberty with tags that should release in May.  The winter pop-ups demonstrated a few different behaviors.  By mid-February, 3 fish had moved to the shelf-edge south of the Islands and another had moved westward into deeper water, although not as far as the shelf-edge.  This was expected.  Winter spawning is known to occur in the Pribilof Canyon, and likely extends northward for some distance.  The longest movement observed was southward to the Aleutian Chain, just north of Amukta Island.  While this was a fairly considerable distance, it is still not too surprising.  If this fish were to spawn near Amukta, it's larvae would probably be swept back to the east and north by prevailing currents, placing its offspring somewhere on the southeast Bering shelf.  What is more intriguiong is that one fish remained very near St. Paul, and another actually moved to the northeast, farther onto the shelf.  These locations are not known to represent spawning areas.
 

So why did 2 out of 10 fish tagged in the Gulf, and 2 of 7 Pribilof fish, appear to be located in shallow water in mid-winter when other evidence suggests that spawning primarily takes place along the shelf-edge?  There are many possibilities.  Perhaps these fish had already spawned prior to the pop-up date and were located in deep water earlier in the season; or perhaps they spawned later in the season and had not yet moved offshore by the pop-up date.  Alternatively, inshore spawning  may occur that we have not adequately described.  In particular, the channels of southeast Alaska and British Columbia may harbor winter residents that do not move offshore to spawn.  Or perhaps halibut do not spawn every year, and those that don't simply do not undertake the same seasonal migrations as their actively-spawning counterparts.  We may be able to address some of these questions once we get a chance to critically analyze the behavioral data transmitted by the tags (i.e., the temperatures and depths experienced by each fish), and others we hope to investigate with future tagging studies.
 

The tags will also allow us to learn about mysteries such as migration timing.  It is important that we understand migration timing as we discuss extending the fishing season to include all or part of the winter.  Today's quotas are based on summer surveys and the distribution of fish on the summer feeding grounds.  If halibut redistribute during the  spawning season, then the fishery will probably need to be altered during the winter months to reflect those changes in distribution.  While the both conventional tagging data and the PSATs indicate that movement to the north during the winter is common, a majhor question is when do the fish leave their summer grounds, and when do they return?  The graph below shows an excellent example fo how the PSATs can help us to answer this question.
 

The data show the average depth experienced by the halibut from the time it was tagged until the pop-up date.  The fish remained at a very constant shallow depth until late September.  From mid-September until mid-December the fish remained in relatively shallow water but began to visit a number of different depth-strata from day-to-day and week-to-week.  Then, during the second half of December, it moved rapidly into deeper water.  From this depth profile one can confidently say that the fish had moved off of its shallow-water feeding grounds by late December, and that it may have begun to relocate as early as late September.  In the future we hope to deploy more tags, leaving them on the fish throughout the winter and into the following summer, in order to gain more information on the timing of both fall and spring migrations.
 

Stay tuned for more details ...