Fisheries Pêches and Oceans et Océans DFO Science Pacific Region Stock Status Report D6-01 (1998)

D r if tw o o d R i WATERSHED v e r Sockeye Salmon Distribution

M i dd le R .

Trembleur Lake

0 25 50 m ile s

Stuart Lake 0 40 80 km

S tua rt R iver Francois Lake Fras er Lake R iver Nadin a R. o k a ch Fras er e PRINCE N GEORGE W B River il o lo w Chila ko w r R . o R n (upper) i

v R

e iv r e

r er iv R West R oad Bowron Lake C o ttonwood R. Q u QUESNEL e s Mitchell River n e l River Quesnel Lake

The 1997 Fraser Sockeye Cycle River r

e v

WILLIAMS i

C R

LAKE l C e hi Horsefly R. a lc r o n tin w o a r s e t iv e p R r R r m o iv R ilk er e o

r h h . v r C i e T e R v v i t i f R R a

s o R

k m

e a .

s h d

R t a A r T

r o N. Barriere . r

R u e N h Lake mic o

s Mo y m

Chilko e a Taseko r S

Lake r Adams Background F e Lakes R iv Lake Shuswap . e r t C Lake r h r

a e tc iv

p o R a e c agl

Br n S E id o

ge B

R

.

R iver Kamloops L. Mara Lake

B r e ir Anderson i v k LILLOOET e Lake KAMLOOPS R n n S n ou o M abe l h o th Thom s e s p La ke Cyclic fluctuations in abundance are characteristic L a Seton Cr. p i d L. Shuswap R. J l m o lo R N hn o s o . ic ton e Gates Cr. h o e t T l Str a ait R M. Shuswap R. V iv A e of many fish populations with a single predominant N r C . O R

U r R. NORTH te S R tlatch V t aha a r U iv N BEND w reproductive age. Fraser sockeye mainly mature E a e d it . r Harrison l R P S . o o t f it a Lake Hell's R C G t v C I R h la S e e Gate l o . e L R h a r h A g . a i i l u a i and spawn at age-four. Of approximately 20 sockeye N s q Pitt R . o D . C Lake R n Stav e o ris HOPE Lake a r populations in the Fraser River watershed that are H CHILLIWACK . Chilliw ac k R. CANADA M ISSION R er dd counted routinely, eight exhibit persistent four-year Ve U.S.A. cycles with a predictable dominant-year cycle line every four years. The 1997 cycle or cycle line refers to the sequence of years 1997, 1993, 1989...1901, Hudson Bay Company records prior to 1873 etc. The term ‘dominant-year cycle line’ refers to indicate that all the major Fraser River sockeye the sequence of years wherein the run size is runs exhibited four-year cycles, with the persistently larger than the other cycle lines. Run dominant cycle occurring in the same year. size is persistently lowest on off-year cycle lines and Later documents of the intermediate on the subdominant cycle line. Department of Fisheries also confirm that spawning abundance to the upper Fraser River The processes that maintain population cycles in populations cycled in synchrony between 1901 Fraser sockeye are poorly understood. There is and 1913. strong evidence that cycles are not merely echoes of past random catastrophic perturbations that persist Estimates of the number of spawners on the due to the simple age structure of these populations. 1901 line up to 1913 (1897-1913) range from Hypotheses to explain cycles centre on identifying five million in 1909 to 13 million in 1901. The agents that impose higher mortality when abundance International Pacific Salmon Fishery Com- is low than when abundance is high (i.e. depensatory mission (1973) reported catches of 35 million in mortality). Various ecological hypotheses have been 1913. Dominant runs prior to 1913 may have proposed that theorise predation during early been about 100 million sockeye. sockeye freshwater development is depensatory. Other explanations for population cycles centre on Owing to obstructions in the Fraser canyon at evidence for depensatory fishing mortality. Hell’s Gate, most sockeye returning to the Fraser River failed to reach the spawning Considerable evidence has been compiled to suggest grounds in 1913 and consequently the 1901 that the 1901 dominant cycle line, destroyed by rock (1997) dominant cycle was lost. So too ended dumping during railroad construction at Hell’s Gate the synchrony in the temporal pattern of cycles in 1913-14, had persisted since the time of the first among cyclic populations. The build-up of runs Hudson’s Bay Company (1820) records and perhaps in the years that followed eventually resulted in since the first reference to Fraser River sockeye less pronounced differences in sockeye (1793). abundance among the four cycle lines.

July 1998 Pacific Region The 1997 Fraser River Sockeye Cycle

Individual runs rebuilt beginning in 1926, with large increases to the lower Adams River on the 1998 cycle line. Spawning stock size for other runs began Run Size For The 1997 Cycle Line to increase in the 1930s. Sustained increases in all runs, however, were only achieved following QUESNEL LATE STUART construction of fishways in the Fraser Canyon along EARLY STUART with a five-year fishing closure on early and mid- OTHER season run timing groups during 1946-50.

Annual Trend In Run Size For Stocks That Are

Dominant On The 1997 Cycle Line MILLIONS OF FISH

EARLY STUART 0 5 10 15 20 25 53 57 61 65 69 73 77 81 85 89 93 97 YEAR 0.0 1.0 53 57 61 65 69 73 77 81 85 89 93 97 Annual returns of Quesnel Lake sockeye have LATE STUART increased dramatically since the 1950s from <1 million fish in 1961 to 13 million in 1993. Annual 0246 returns of early Stuart sockeye on the 1997 cycle

MILLIONS OF FISH 53 57 61 65 69 73 77 81 85 89 93 97 QUESNEL have been highly variable with little trend since the 1950s and have ranged from 256,000 sockeye in 1965 to 1.5 million in 1997. The late Stuart run has 048 increased since 1981 with record returns of 3.4 53 57 61 65 69 73 77 81 85 89 93 97 YEAR million in 1989 and >5 million in 1993.

Estimates of the number of spawners on the 1997 cycle line remained relatively constant during 1953- The Fishery 81 and averaged 1.1 million fish/year. Large increases in run sizes on the 1997 cycle have Following construction of the first cannery occurred since 1981. The number of spawners, catch on the Fraser River in 1866, the commercial and total returns (catch+spawners) increased gillnet fishery developed rapidly. The early steadily from 7.5 million in 1981 to 24 million in commercial fishery was intense and harvest 1993. The returns in 1993 were the highest on rates on the non-dominant cycles were record since 1913. higher than on the dominant 1901 cycle. Populations of sockeye that rear in Quesnel Lake Thus it seems likely that cyclic trends in and the largest runs to the Stuart-Takla-Trembleur Fraser River sockeye production prior to lakes region (early Stuart and late runs) 1913 were exaggerated and at least partially are now dominant on the 1997 cycle line. Cyclic maintained by different fishing patterns populations that return to Shuswap and Chilko Lake are now lowest on the 1997 cycle. among cycles (e.g. depensatory fishing mortality). Reconstructed estimates of The increase in run size to record levels on the 1997 catches ranged from 35-50 million on the cycle beginning in 1981 resulted from increases in 1901 line before 1913. As a result of habitat spawning abundance and higher than average destruction, the most notable being the survival rates in the 1980s. Increases in survival were highest for Quesnel Lake sockeye. effects of the Hell’s Gate slide in 1913 and the overfishing in the years that followed, catches declined from 35 million in 1913 to an average of 1.9 million fish/year during

2 Pacific Region The 1997 Fraser River Sockeye Cycle

1915-30 for all cycles. During 1930-50 commercial catch, the purse seine fishery catches increased to an average of 2.6 took 49% and the gillnet fishery took 36%. million fish/year. Estimates of native catches were about 5% of the total estimated catch in 1993. Catch And Harvest Rate For The 1997 Cycle Line Canadian commercial catches are mainly taken in the troll fishery off the west coast of , purse seine and gillnet fisheries in Johnstone and Juan de Fuca

HARVEST RATE straits, and the gillnet fishery in the Fraser River. Smaller commercial fisheries also

occur in northern and central B.C. and within the . Sockeye are harvested HARVEST RATE MILLIONS OF FISH in native food fisheries throughout the Fraser River watershed. Recreational catches are 0 5 10 15 0.5 0.6 0.7presently 0.8 small but estimates of sports 53 57 61 65 69 73 77 81 85 89 93 97 YEAR catches have recently increased.

During 1937-84 the International Pacific U.S. fisheries occur mainly in the net Salmon Fishery Commission (IPSFC) was fisheries in Juan de Fuca Strait and southern responsible for management of fisheries in a approaches to the Fraser River located in ‘Convention Area’ that included Canadian U.S. waters. Some Fraser sockeye are also and U.S. fisheries in southern B.C. and U.S. taken in southeastern Alaska. U.S. catches waters of Washington State. During that have remained relatively stable and were period catches from Convention Area about 18% of the total catch in 1993. fisheries were shared equally between Canada and the United States. With Harvest rates on the 1997 cycle have been ratification of the Pacific Salmon Treaty in variable, ranging from 64% in1997 to 85% in 1985, the Fraser River Panel, under auspices 1985, with an averaging of 78%. High of the Pacific Salmon Commission, has Fraser River discharge rates in 1997 resulted regulated management of Fraser sockeye in high in-river mortality of migrating adult fisheries. The Fraser River Panel consists of sockeye. Harvest rates were reduced in Canadian and U.S. representatives. Its 1997 to increase spawning escapement to the purpose is to ensure spawning targets set by Fraser River in an attempt to compensate for Canada, as well as international and domestic the unusually high in-river mortality. catch allocation goals, are met.

Catches on the 1997 cycle have increased Resource Status linearly in proportion to run size since the late 1960s, from four million sockeye in For most sockeye populations, there is little 1969 to 18 million sockeye in 1993. evidence of diminishing adult returns at the upper range of spawning stock sizes This was the highest recorded catch from all observed since rebuilding began in the 1920s years since 1913. In 1993 the troll fishery and 1930s. For some stocks, we have seen accounted for 15% of the Canadian lower return rates in a few years of high

3 Pacific Region The 1997 Fraser River Sockeye Cycle spawning stock sizes. For example, sockeye that indicates the current cyclic pattern is returns to Quesnel Lake have been about 10- optimal. Rebuilding of off-cycle runs that 12 million fish/year on the dominant cycle co-migrate with much larger dominant runs line in 1985, 1989 and 1993, despite record on the 1997 cycle require careful increases in spawning stock size of more consideration of alternative management than two-fold (750,000-1.9 million). plans so that short-term economic losses are Preliminary estimates of returns in 1997 were minimised. Ultimately, the processes that seven million fish, following a record perpetuate population cycles and prevailing spawning escapement of 2.6 million sockeye climatic factors affecting survival will in 1993. determine the rate that low runs can be rebuilt to optimal long-term levels. The effect of these high spawner densities on survival is not known. Because of high survival variability, independent of the Relationship Between Number Of Spawners And Resulting Run Size For The 1997 Quesnel Lake number of spawners, repeated high spawning Sockeye Cycle Line (Data Points Are Spawning levels are required to assess optimal stock Years) sizes under average conditions.

Trends in recruitment at any particular 85 89 spawning stock are not only affected by 81 random inter-annual variations in survival but also by long-term climatic variations in 93 productivity. Until we have seen spawning stock sizes at repeated and sufficiently high 77 RUN SIZE (millions of fish) 73 levels, the full potential of Fraser River 6965 5761 4953 sockeye will not be known. 024681012

0.0 0.5 1.0 1.5 2.0 2.5 Economic yield from the 1997 cycle line has SPAWNERS (millions of fish) mainly come from a few highly cyclic populations (Late Stuart and Quesnel). Much larger gains on the 1997 cycle depend For More Information on the extent to which off-cycle runs of Chilko and sockeye can be Contact: Alan Cass rebuilt. Less dramatic production increases Pacific Biological Station are also likely if smaller co-migrating runs Nanaimo, B.C. V9R 5K6 can be rebuilt. Tel: (250) 756-7142 The current lack of understanding of the Fax: (250) 756-7138 processes that maintain cyclic patterns in Fraser sockeye is undoubtedly the single E-Mail: [email protected] greatest obstacle to maximising Fraser sockeye potential. Even under pessimistic scenarios, increases in yield should result from increases in spawning abundance to off- cycle runs. Certainly, there is no evidence

4 Pacific Region The 1997 Fraser River Sockeye Cycle

References This report is available from: Adkison, M.D., R.M. Peterman, M.F. Lapointe, D.M. Gillis and J. Korman. PSARC Secretariat 1996. Alternative models of climate Pacific Biological Station effects on sockeye salmon Nanaimo, B.C. V9R 5K6 (Oncorhynchus nerka) productivity in Bristol Bay, Alaska, and the Fraser Phone: (250) 756-7208 River, British Columbia. Fish. Oceanogr. Fax: (250) 756-7209 5: 137-152. Email: [email protected]

Cass, A.J., and C.C. Wood. 1994. ISSN 1480-4913 (for English series) Evaluation of the depensatory fishing ISSN 1480-4921 (for French series) hypothesis as an explanation for population cycles in Fraser River La version française est disponible à l’adresse sockeye salmon (Oncorhynchus nerka). ci-dessus. Can. J. Fish. Aquat. Sci. 51:1839-1854.

Collie, J.S., R.M. Peterman, and C.J. Walters. 1990. Experimental harvest policies for a mixed stock fishery: Fraser River sockeye salmon, Oncorhynchus nerka. Can. J. Fish. Aquat. Sci. 47: 145-155.

Ricker, W.E. 1950. Cyclic dominance among the Fraser River sockeye. Ecology 31: 6-26.

Ricker, W.E. 1997. Cycles of abundance among Fraser River sockeye (Oncorhynchus nerka). Can. J. Fish. Aquat. Sci. 54: 950-968.

Roos, J.F. 1991. Restoring Fraser River Salmon. Pacific Salmon Commission, Vancouver, Canada. 438 p.

5