Research Document - 2001/107

Forecast for Southern British Columbia Coho Salmon in 2001

By Simpson, K., Dobson, D., Irvine, J. Holtby, B., Tanasichuk, R.W.

Abstract

This Research Document documents forecasts of marine survival, abundance and distribution for the coho salmon of southern British Columbia (Strait of Georgia, west Vancouver Island, lower Fraser and interior Fraser including the Thompson River) for return year 2001.

Marine survival

Recommendations for the marine survival forecast for the five hatchery indicators and two wild coho indicators are given in the following table:

For populations around the Strait of Georgia, survivals are forecast to change very little. Survival will remain poor throughout southern BC and survival is forecast to improve only marginally at Black Creek. None of the sibling models performed better than the statistical models. However, all the regressions predict better survivals, particularly in the lower Fraser system, and marine data also suggest that the forecasts are more likely to be conservative than not.

Euphausiid and sibling survival forecasts are presented for Robertson Creek coho on the west coast of Vancouver Island (wVI). Bearing in mind that both badly under-forecast the 2000 survival of 7.6%, they predict a survival in 2001 that is similar to the mean 1990's survival and about half of last year's unusually large survival. Using an alternate estimate of Robertson escapements (see below) resulted in larger survivals in the time series, including a larger forecast. However, the 2001 forecast remains in the same proportion to last year and the 10 year average of the new time series. Carnation Creek has been added to the forecast this year. The forecast survival of 10.2% is a substantial increase over the 4.8% last year and is better than the 1990's average of 8.4%.

Abundance

Forecasting abundance is highly problematic, particularly in the present regime of low exploitation. The abundance in 2000 was estimated to be 560,000 for populations in the Georgia Basin, which is more than the upper 95% confidence limit for the forecasted abundance of 250,000. This year the best model is the LLY model, meaning that, like survivals, we are forecasting no change in abundance. This 2001abundance forecast of 560,000 is 40% of the long term mean abundance of 1.4 million.

The estimated abundance of the wVI aggregate in 2000 (920,000) was extreme compared to the forecast (270,000; upper 99% confidence limit: 900,000). This was attributable to a very large escapement to the hatchery. Although there was also a large relative increase in the escapement to Carnation Creek, all our other escapement data indicated only a slight improvement in abundance. This stimulated us to examine the effect of using Stamp Falls counts as an alternate estimate of Robertson escapement, which we think were underestimated before 2000 to an unknown degree. The abundance estimate for 2000 was slightly higher using this data (1.15 million). However, 1999 and many preceding years were higher yet so the difference from 1999 to 2000, although still large, was about half the difference using hatchery escapements. Escapement in 2000 to the Somass system, including Robertson, was obviously large compared to most other wVI areas. This occurrence highlights the vulnerability of wVI forecasts based on only one exploitation and survival rate indicator. The forecast of wVI abundance in 2001, based on the 3YRA forecast model of abundances, is for 460,000 (50% CI: 310,000 - 670,000). The 2000 estimate and 2001 forecast represent 175% and 84% of the long term mean abundances, respectively.

The forecast total abundance of Thompson River watershed coho in 2001 is ~17,500, similar to the estimated abundance in the brood year and only about 20% of the long term mean abundance. The overall abundance of Thompson coho has not increased significantly in recent years but spawner numbers are increasing. Escapements in 2000 and 1999 were both larger than brood year escapements. Greater proportions of fish surviving to maturity are returning to spawn because of significant reductions in fishing pressure. Thus, assuming marine survivals and fishing pressures remain low, the outlook for Thompson and other interior Fraser coho is for slow but gradual improvement.

Distribution

In the hypothetical circumstance of historical patterns of fishing, the predicted proportion of catch inside the Strait of Georgia (pinside) is 0.70 (50%CI: 0.60-0.78), which is a very strong inside distribution. There is less than a 1% chance of a moderate outside year occurring if that is defined as pinside < 0.3. Some Georgia Basin stocks, including Thompson and Cowichan, have a greater propensity to rear outside even in inside years. Therefore, Thompson coho will probably occur in both areas in 2001, as they have in previous inside years.

Complete document (PDF, 452 KB)

Accessibility Notice

This document is available in PDF format. If the document is not accessible to you, please contact the Secretariat to obtain another appropriate format, such as regular print, large print, Braille or audio version.