Download PDF Download Document

Cedar River and Lake Washington Sockeye Salmon Biological Reference Point Estimates

Category: Fish/Shellfish Research and Management - Fish/Shellfish Research

Date Published: June 15, 2009

Number of Pages: 65

Author(s): Scott McPherson, MSc., Alaska Department of Fish and Game and James C. Woodey, Ph.D., Fisheries Consultant


  1. The objective of the current report is to provide Washington Department of Fisheries and Wildlife with estimates of current Cedar River and Lake Washington natural origin recruit (NOR) sockeye biological reference points (BRPs) to assist in management of the stocks. Analysis of Cedar River and total Lake Washington NOR sockeye abundances in relation to parental spawner population sizes indicated that productivity in the most recent period, 1982-2002, has declined compared to the 1967-1981 period.
  2. When data from 1982-2002 were analyzed, our analysis of optimum yield via bias-corrected Ricker stock-recruit (S-R) models produced a relatively low maximum sustained yield (MSY) escapement estimate for Cedar River NOR sockeye of approximately 82,000 spawners. The combined Lake Washington NOR sockeye populations MSY escapement was estimated to be 102,000 natural spawners.
  3. Further investigation revealed that even-year brood sockeye recruitment in the 1982-2002 period has been substantially higher than recorded for odd-year broods. Using the Lake Washington spawning escapements to Lock-based recruit dataset, the MSY escapement for even-year broods was relatively stable over the full time period predicting that, on average, 150,000 spawners would generate recruitment of approximately 313,000 fish, yielding 163,000 in surplus production at an optimum harvest rate (HR) of 52%.
  4. For the odd-year broods, the 1983-2001 data indicated a MSY escapement of 69,500 spawners compared to 184,000 spawners for the 1967-1981period. The cause of the decline in odd-year brood productivity in the recent time period could not be determined through S-R analysis.
  5. Comparative S-R parameter estimates for eight reference stocks from Washington and southern British Columbia clearly show that current Cedar River productivity is far lower than any of these stocks. The Ricker “a” (alpha) value for the 1982-2002 Cedar River NOR was 1.9 vs. a mean of 8.7 for reference stocks. At MSY, yield from the Cedar River NORs was estimated to be 35,000 from a mean recruitment of 117,000 fish giving an optimum harvest rate of 30% while the mean HR at MSY for the reference stocks was 73%.
  6. Analysis of Cedar River NOR sockeye fry recruitment relationships and estimates of fry to presmolt and to adult recruit survival indicates that a bottleneck to production occurs in Lake Washington. Estimated fry to adult survival appears to be quite low and was negatively related to fry abundance. Odd-year brood fry to adult survival was substantially lower than for even-year broods at a given fry abundance, particularly at higher abundance levels. This pattern of survival was found in hatchery origin (HOR) Cedar River sockeye fry, as well. We estimated through optimization analyses that current yield is highest at approximately 12 million NOR fry for odd-year broods and 21 million fry for even-year broods. The escapements required for these levels of fry recruitment (51,000 Cedar River spawners for odd-year broods and 90,000 for even-year broods) follow the pattern estimated via S-R analysis. Recent levels of NOR fry production have been up to 39 million fry.
  7. We hypothesize that foraging by large even-year brood abundances of longfin smelt and other species in late winter exacerbated by large numbers of sockeye fry entering the south end of the lake from the Cedar River prior to the spring bloom period depletes preferred food resources and leads to an extended period of low growth and high mortality that constrains Cedar River NOR sockeye productivity. Interspecific competition between large even-year age 0 smelt populations and juvenile sockeye may be responsible for the lower survival of odd-year brood sockeye fry from the Cedar River. Predation of sockeye fry by large populations of maturing even-year brood longfin smelt or by piscivorous fish after mature even-year broods of smelt spawn and die may account for the differential survival of even- and odd-year brood sockeye.
  8. The productivity of the Cedar River sockeye population is very low and would generally limit harvest opportunities at any level of escapement. Yields on future odd-year returns will likely be low under current conditions (MSY yields = 28,000 and 80,500, depending on the assumption for recruitment). Even-year brood recruitment estimates at MSY would likely provide for more regular surplus production and harvest opportunities.