Studies of Straying and Gene Flow, and Development of Selective Fishery Regimes Using Mass-marked Chinook and Chum Salmon from Tulalip Hatchery

This is a proposal submitted to the National Ocean and Atmospheric Administration (NOAA) and funded as pproject no. ___________. The funding period for the current award is November 1, 1996 through October 31, 1997.

Principal Investigator:
Kit Rawson
Tulalip Fisheries
7615 Totem Beach Rd.
Marysville, WA 98271 USA
(360) 651-4478
krawson@tulalip.nsn.us

Introduction

The role of hatchery production in the management of fish, and Pacific salmon in particular, has undergone review and scrutiny in recent years (see, for example, papers in Schramm and Piper 1995). Hatchery production of Pacific salmon is meant to mitigate for losses in natural production or, in some cases, to augment natural production levels. However, over the past decade, a number of researchers have documented that some hatchery production may not be resulting in the expected benefits due to a number of factors. Several of these factors address potential detrimental effects of hatchery production upon wild stocks. The research proposed here will address two of these: deterioration of wild stock gene pools due to incorporation of genes from hatchery strays into wild populations (Campton 1995) and overharvest of wild fish by fisheries harvesting at the higher rates appropriate to the hatchery fish mixed with them (Lichatowich and MacIntyre 1987).

The Tulalip Tribes, a recognized treaty tribe with reservation located in Marysville, Washington, approximately 50 miles north of Seattle, is one of the principal tribal comanagers of the wild salmon and steelhead stocks produced in the Stillaguamish and Snohomish river system which enter Puget Sound near the Tulalip Reservation. Due to recent years' declines in stock sizes, the allowable harvest rates on these natural stocks have declined to the point where the tribal fisheries department keeps most of the tribe's usual and accustomed fishing area closed to salmon fishing for most of the year. As partial compensation for this loss of fishing opportunity, the tribe operates the Tulalip Hatchery, located on the reservation. The challenge of managing the production and harvest of hatchery fish, without undue impacts to local wild stock production, appears to have been met to date. With the exception of chinook salmon, the local wild stocks regularly meet established escapement goals, while hatchery fish provide an opportunity for the tribe to exercise a meaningful treaty rights fishery.

The Tulalip Hatchery, in operation since 1982, releases approximately 1.5 million chinook salmon fingerlings, 1.0 million coho salmon smolts, and 4.0 million chum salmon fry each year. The adults returning from these releases contribute to numerous Canadian and US fisheries. Of these, the principal ones of interest to the Tulalip Tribes are the fishery in Area 8A, which is managed for a mixture of wild stocks from the Stillaguamish and Snohomish river systems and hatchery fish from Tulalip, and Area 8D, which is a terminal area fishery managed for fish produced by the Tulalip Hatchery (Figure 1).

State/tribal comangement plans (for example, Washington Department of Fish and Wildlife, et al. 1995) call for management in the Stillaguamish/ Snohomish region to give priority to achieving escapement goals for natural stocks returning to the two river systems. Despite declining run sizes, the management objectives have largely been achieved in recent years for all species except chinook salmon. This has been due to unprecedented restrictions on wild stock harvest, meaning that the ability of the Tulalip Tribes to exercise their treaty-reserved fishing rights is becoming increasingly reliant on Tulalip hatchery production.

When little or no harvest is allowed on local wild stocks, Tulalip tribal harvests must be restricted to limited periods of time in Area 8D in order to focus the fishery on the hatchery production. Since 1994 the non-Indian recreational fishery has operated under the same restriction. One goal of the research proposed here is to increase both the permissible area and time of salmon harvest without increasing the harvest rates on wild stocks above allowable levels. The other goal of this research is to document the extent to which straying of Tulalip Hatchery fish into natural stock spawning areas of the Stillaguamish and Snohomish Rivers may be occurring so that the primacy of natural stock management in this region may be maintained.

Both of these goals address needs of the National Marine Fisheries Service (NMFS) in development of recovery plans for species listed under the Endangered Species Act. This work is part of ongoing work by NMFS to investigate these issues.

This project will address several research and data needs identified by the Pacific Fishery Management Council including1 research to improve the ability to identify stocks in ocean fisheries and escapement, research in mass-marking of hatchery stocks, development of freshwater sampling programs designed to allow the estimation of the contribution of hatchery fish in river systems where both hatchery and wild chinook salmon are present, and characterization of gene pools.

Despite documented reductions in chinook harvest rates (Puget Sound Salmon Stock Review Group 1992), the escapement levels for this species have not improved in the Stillaguamish/Snohomish Region. The Snohomish chinook stock is regarded as not rebuilding by the Chinook technical Committee of the Pacific Salmon Commission and has triggered the overfishing definition under the Pacific Fishery Management Council's Salmon Management Plan. The research proposed here will provide significant new information important for improved management of this important stock.

Work to Date

The proposed research on chum salmon will build on mass-marking and initial mark-recovery work undertaken by the Tulalip Tribes since 1990. During brood years 1990 through 1993 Tulalip hatchery chum salmon were genetically marked by selection of spawners to change the allele frequencies at two of the 21 loci used for electrophoretic stock separation analysis (Rawson 1995). This genetic mark is easily detectable in mixtures of this stock with other Puget Sound and Canadian chum stocks through standard genetic stock identification (GSI) methods using electrophoresis (Pella and Milner 1987). One advantage of the particular mark chosen is that, Tulalip hatchery fish can be distinguished from non-Tulalip fish by laboratory analysis of only one tissue, instead of the usual three tissues, at approximately one-third the cost of the usual analysis required for GSI (Rawson 1995).

During the fall of 1994, when the first 4-year-old marked chum salmon returned to Tulalip hatchery, initial fishery and in-river sampling for the marked fish was undertaken. Preliminary results from fishery sampling documented the wild versus hatchery timing difference hypothesized by Rawson (1995). Initial in-river results detected no Tulalip hatchery contribution to the natural spawning population sampled in the Snohomish system nor to the Stillaguamish tribal in-river fishery nor to the return to the Stillaguamish hatchery at Harvey Creek. Tulalip hatchery fish were apparently contributing to natural spawners in Quilceda Creek, located on the Tulalip Reservation near Tulalip Bay. Progeny fry from Quilceda Creek were collected in the spring of 1995 to see if the hatchery contribution to the 1995 outmigrants was the same as the apparent hatchery contribution to the spawning adult population, but results from the genetic analysis of these fish is not available at this time. Fishery sampling work was repeated in 1995, but results are not yet available. This proposal includes funding for more intensive in-river and terminal area fishery sampling in the 1996 return year.

Mass-marking of hatchery chinook salmon at Tulalip and Skykomish hatcheries began with the 1993 brood year (Rawson, Bengston, and Volk 1995, 1996). The Skykomish hatchery, operated by the State of Washington, is located on the Wallace River, a tributary to the Skykomish in the Snohomish system. All chinook production2 from both hatcheries is marked by thermal inducement of unique banding patterns on otoliths following the method developed by Volk, Schroeder, and Fresh (1990). Fish originating in one of these hatcheries can be distinguished from naturally-produced fish by laboratory analysis of otoliths extracted from samples of adults caught in the fishery or found as carcasses on the spawning grounds. Limited sampling of carcasses in the Snohomish system is planned for the 1996 spawning season using other funding. This proposal includes funding for spawning ground and terminal area fishery sampling for the 1997 return year as well as laboratory analysis of otoliths collected in both 1996 and 1997..

Research Plan

Task 1. Chum Salmon Studies.

In-River Samples

In-river sampling of adults will take place in the Stillaguamish and Snohomish River systems as well as Quilceda Creek. Sampling will take place during the 1996 spawning season, and the design will be an expansion of the preliminary work undertaken in 1994.

Within the Stillaguamish system sampling work will be subcontracted to the Stillaguamish Tribe. They will collect 200 samples (50/week for 4 weeks) from the tribal fishery which takes place in the lower mainstem Stillaguamish River and 200 samples (4 samples of 50 fish each spread evenly through the period of chum return to the hatchery) from the tribal hatchery at Harvey Creek. The Stillaguamish Tribe will also take samples from 200 chum salmon on the spawning grounds at Squire Creek, tributary to the North Fork Stillaguamish, and 200 chum salmon on the spawning grounds in Jim Creek tributary to the South Fork Stillaguamish. All Stillaguamish system samples will involve the collection of muscle tissue only and will be analyzed on for allele frequencies at the Tulalip hatchery marker loci.

Sampling of natural spawners in the Snohomish system will take place at Skykomish slough. There will be samples collected from 200 fish, and these will involve the collection of three tissues and analysis of allele frequencies at all 21 loci used for GSI analysis. This fuller analysis will add to the natural stock baseline database for the Snohomish system as well as providing information on the possible contribution of Tulalip hatchery fish to this spawning area. Sampling of 200 adult chum will also take place in Quilceda Creek at several locations to expand upon the preliminary work of 1994. These samples will be analyzed only for the allele frequencies at the Tulalip hatchery marker loci.

In-river sampling of juvenile chum will be conducted in the spring of 1997, to investigate intergenerational gene flow, in those areas where the presence of Tulalip hatchery fish was detected from analysis of 1996 adult in-river samples. This will require that analysis of the adult tissues be completed within two months of collection. We have been assured that the WDFW genetics laboratory can accommodate this need given sufficient advance notice. We expect that we will again find evidence of Tulalip hatchery straying into the Quilceda Creek system, and have programmed for the collection of up to 400 juvenile chum (100/site at four sites) by electroshocking in that area. The allocation of samples among areas will be adjusted if evidence of hatchery straying is found in any of the other natural spawning areas and those areas are sampled for juveniles as well.

We have also planned for collecting GSI samples from 200 fish at the Tulalip hatchery rack to provide a new baseline for that stock.

Fishery Sampling

We propose to conduct weekly GSI sampling of the chum fishery in Areas 8A and 8D during the 1996 chum management period (late October through mid-December). Standard GSI fishery sampling protocol calls for 200 samples/week, with analysis of allele frequencies at 21 loci, for analysis of mixed-stock fisheries such as the one in Area 8A. For extreme terminal fisheries, such as the one in Area 8D, 100 samples/week, with analysis of allele frequencies at only the two Tulalip genetically marked loci are sufficient to estimate the weekly contribution of hatchery-origin fish which is expected to be high. A total of 1000 samples (5 fishing weeks) will be taken from Area 8A and a total of 500 samples (5 fishing weeks) will be taken from Area 8D. We will also contract with the Stillaguamish Tribe to take 200 samples (50/week for 4 weeks) from the Stillaguamish Tribe's fishery in the Stillaguamish River.

Task 2. Chinook Salmon Studies

In-River Samples

The basic sampling design for sampling chinook carcasses in the Snohomish system has been worked out by the Tulalip Tribes and Washington Department of Fish and Wildlife and will be tested during the 1996 return year (Tulalip Tribes participation in 1996 sampling will be supported by other funding). We have identified 6 areas within the Snohomish system where we are likely to be able to sample chinook carcasses for otoliths. These areas also encompass the range of chinook stocks identified within this system by the comanagers (see Washington Department of Fish and Wildlife and Western Washington Treaty Indian tribes 1994).

Subdivision of the system also constitutes a stratification of the spawning population according to a priori expected hatchery contribution rates. Once the otoliths from each stratum's sample are analyzed the ratio of hatchery otoliths to the total number sampled will be an estimate of the contribution of hatchery fish to the natural spawning population. Since an individual total spawner escapement estimate will be available for each stratum, this ratio can be expanded for each stratum and summed to provide an overall, system-wide, estimate of hatchery contribution to spawning escapement. The sampling goal is 100 fish per stratum. However, realistic expectations projected stratum by stratum based on the previous experience of field biologists, leads us to expect a total of 400 otolith samples from the 6 strata selected. Tulalip tribal staff, funded by this grant, will be wholly or partly responsible for sampling three of the in-river strata identified. Sampling in the other strata will be done by the Washington Department of Fish and Wildlife and Snohomish Public Utility District as part of work already programmed.

An additional 200 otoliths (spread out through the spawning period) will be taken from chinook salmon spawned at Skykomish hatchery to monitor for the possible straying of naturally-produced fish into the hatchery population.

The above sampling plan will be attempted in 1996. It will be repeated in 1997 unless modified due to the 1996 experience. We are requesting funding for laboratory analysis of both the 1996 and 1997 otolith samples and for the Tulalip Tribes participation in the 1997 sampling.

Fishery Samples

The Area 8D fishery will be sampled for 6 weeks at 100 fish per week. This fishery is opened three days per week, with four days per week closed. The hypothesis underlying this "pulse fishery" management is that the harvest rate on any wild stocks that may be present can be minimized by allowing 4 days of no fishing for wild stocks to pass through this small area while hatchery fish returning to Tulalip Bay accumulate. This hypothesis will be tested by splitting each week's sample into 50 taken on the first day of the fishery and 50 taken on the third day of the fishery. If the hypothesis underlying the pulse fishery is correct then we expect the hatchery contribution to the first day's fishery to be significantly greater than the hatchery contribution to the third day's.

Project Personnel

Kit Rawson, Harvest Management Biologist with the Tulalip Tribes will supervise all aspects of the project and be responsible for data analysis and reporting of progress and results.

Richard Young, Harvest Management Technician II with the Tulalip Tribes will be responsible for collection of in-river otolith samples, in-river genetic samples in the Snohomish and Quilceda systems, and otolith and genetic samples from the fishery. Mr. Young is experienced in fishery sampling, collection of genetic samples and collection of otolith samples. He will be assisted by one technician for the chinook work and two technicians for the chum work.

John Drotts, harvest management biologist for the Stillaguamish Tribe, will supervise the chum sample collection from the fishery and spawning areas in the Stillaguamish River.

Otolith sample processing will be completed under contract to the Washington Department of Fish and Wildlife under the direction of Eric Volk. Analysis of genetic samples will be completed under contract to the Washington Department of Fish and Wildlife under the direction of Larry LeClair. Both of these individuals and their laboratories have successfully completed similar work under contract to Tulalip during earlier phases of this project. Data entry and management will be the responsibility of Rawson and Young. Ken Currens, geneticist with the Northwest Indian Fisheries Commission, will advise in all aspects of the project related to genetics. Susan Bishop, biologist with the Northwest Indian Fisheries Commission, will advise in project design.

Budget Justification

[This section was removed from the HTML document due to incompatible formatting.]

Project Time Line

[This section was removed from the HTML document due to incompatible formatting.]

References Cited

Pella, J. J. and G. B. Milner. 1987. Use of genetic marks in stock composition analysis. in N. Ryman and F. Utter, eds., Population Genetics and Fishery Management. University of washington Press (Seattle): 274-276.

Puget Sound Salmon Stock Review Group. 1992. Assessment of the status of five stocks of Puget Sound chinook and coho. Technical Report for the Pacific Fishery Management Council (Portland, OR), 113 p.

Rawson, K. 1995. Management applications of the genetic mass-marking of chum salmon at Tulalip Hatchery. in press for Proceedings of the 1995 Pink and Chum Workshop, Bellingham, Washington, March 1995.

Rawson, K., C. Bengston, and E. Volk. 1995. Snohomish River chinook straying evaluation study. Fiscal Year 1994 Progress Report. Tulalip Natural Resources Division Report No. 95-1.

Rawson, K., C. Bengston, and E. Volk. 1996. Snohomish River chinook straying evaluation study. Fiscal Year 1995 Progress Report. Tulalip Natural Resources Division Report No. 96-1.

Schramm, H. L. and R. G. Piper. 1995. Uses and effects of cultured fishes in aquatic ecosystems. American Fisheries Society Symposium 15. American Fisheries Society (Bethesda).

Volk, E. C., S. L. Schroeder, and K. Fresh. 1990. Inducement of unique otolith banding patterns as a practical means to mass-mark juvenile Pacific salmon. American Fisheries Society Symposium 7:203-215.

Washington Department of Fish and Wildlife and Western Washington Treaty Indian Tribes. 1994. 1992 Washington State Salmon and Steelhead Stock Inventory, Appendix One Puget Sound Stocks, North Puget Sound Volume. Olympia, Washington, June 1994.

Washington Department of Fish and Wildlife, Tulalip Tribes, Stillaguamish Tribe, Northwest Indian Fisheries Commission. 1995. 1995 Puget Sound Salmon Forecasts and Management Recommendations -- Stillaguamish/Snohomish Region. Joint Report, 16 p.