2006 Upper Cook Inlet Sockeye Salmon
Fishery Forecast

Issued Thursday, December 29, 2005

FORECAST AREA: Upper Cook Inlet

SPECIES: Sockeye Salmon

FORECAST OF 2006 RUN:

FORECAST METHODS:

The major sockeye salmon systems in Upper Cook Inlet (UCI) are the Kenai, Kasilof, Susitna and Crescent Rivers, and Fish Creek. Spawner, sibling, fry, and smolt data, if available, were examined for each system. Four models were used to forecast the return of sockeye salmon to UCI in 2006: (1) the relationship between adult returns and spawners, (2) the relationship between adult returns and fry, (3) the relationship between adult returns and smolts, and (4) the relationship between adult returns and siblings. In most cases, sibling relationships were used. The return of age 1.3 sockeye salmon to the Kenai River in 2006 was forecast using the fry model. The fry-model prediction was based on the abundance of sockeye salmon fry rearing in Skilak and Kenai lakes in the fall of 2002. The abundance of smolts emigrating from Tustumena Lake (estimated by the Cook Inlet Aquaculture Association) was used to forecast returns of sockeye salmon to the Kasilof River in 2006. This is the fifth time this model has been used. The aggregate escapement goal is the sum of the upper end of the escapement goal ranges for each of the major sockeye salmon producing systems in UCI. An approximate eighty percent confidence interval for the total forecasted run was calculated using the squared deviations between past forecasts and actual runs as the forecast variance (mean square error).

FORECAST DISCUSSION

In 2005, the commercial harvest of sockeye salmon in UCI was 5.1 million, while the preseason forecast was 4.1 million. The higher than expected commercial harvest in 2005 was largely due to a stronger than expected return of 5-year old sockeye salmon to the Kenai and Kasilof rivers. In 2005, the total return of sockeye salmon was 5.5 million to the Kenai River, 1.2 million to the Kasilof River, 143 thousand to the Susitna River, 173 thousand to the Crescent River, and 22 thousand to Fish Creek. The forecast return of sockeye salmon in 2005 was 3.3 million to the Kenai River, 911 thousand to the Kasilof River, 432 thousand to the Susitna River, 160 thousand to the Crescent River, and 27 thousand to Fish Creek.

A run of 3.6 million sockeye salmon is forecasted to return to Upper Cook Inlet in 2006 with a harvest by all user groups of 2.1 million sockeye salmon. The forecasted harvest in 2006 is about 2.5 million fish below the 20-year average harvest. A fry model was used to forecast the return of age 1.3 sockeye salmon to the Kenai River. The fry model predicted a return of 1.1 million age 1.3 sockeye salmon to the Kenai River, which is less than one half of the 20-year average return for this age class. The fry model has provided more accurate forecasts of age 1.3 sockeye salmon runs to the Kenai River than the sibling model in 5 of the past 9 years, but this year the sibling model forecast of 0.9 million fish was similar to the fry model forecast. The forecast return to the Kasilof River is slightly below the 20-year average return of 957 thousand. Smolt models were used to forecast the returns of sockeye salmon to Kasilof River in 2006. Smolt models for Kasilof River salmon have provided more accurate forecasts than other models over the past 10 years. Age 1.2 and 1.3 sockeye salmon typically comprise about 69% of the run to the Kasilof River. These fish emigrated from Tustumena Lake as smolts in 2003 and 2004. The smolt population estimate in 2003 was the third highest since 1981, while the smolt population estimate in 2004 was near the long-term average. The forecast return to Susitna River is much lower than the 20-year average return of 468 thousand. Age 1.3 and 2.3 sockeye salmon typically comprise 68% of the run to this system. The below average forecast is due to the poor runs of age 1.2 and 2.2 sockeye salmon in 2005, which were used to forecast the runs of age 1.3 and 2.3 sockeye salmon in 2006 using sibling models. The forecast return to Fish Creek is only 27% of the 20-year average return of 164 thousand. Age 1.2 sockeye salmon typically comprise 58% of the run to this system. Only 32.3 thousand age 1.2 sockeye salmon are forecast to return to Fish Creek in 2006. This forecast is based upon a count of only 256 thousand sockeye salmon smolts emigrating in 2004, which is 40% of the 20-year average smolt abundance in this system.

Forecast runs to individual freshwater systems are as follows:

Mark Willette
Research Project Leader
Upper Cook Inlet