Research Document - 2000/174

Review of the 1999 Return of Barkley Sound Sockeye Salmon and Forecasts for 2000

By K. Hyatt, W. Luedke, W., Till, J., Rankin, P., Lewis, D.

Abstract

Recent year returns of Barkley Sound sockeye have been gradually increasing from a low of only 200,000 fish in 1995 to 380,000 in 1996 , 465,000 in 1997 and 660,000 in 1998 (Figure 2). Although moderate, recent increases were anticipated by two independent forecasting procedures first developed in 1987 and applied annually since 1988 to predict return variations of Barkley Sound and West Coast Vancouver Island (WCVI) sockeye. Given the gradual recovery of the stock in recent years, aboriginal, recreational and small commercial fisheries were correctly anticipated in 1997 1998 and 1999. Recent year stock increases represent a continuation of a pattern of predictable variations in ocean climate states that have lead to repeated "crashes" (1978, 1985-86, 1989-90, 1994-95) followed within 1-3 years by recoveries (1979-81, 1987-88, 1991-93, 1996-98) of WCVI sockeye returns. The surplus for commercial catch in recent years is directly attributable to the combined effects of: (i) a shift of conditions in the marine environment that supported near average survival of sockeye smolts migrating seaward between 1994 and 1997 and (ii) management to protect escapement as a basis for increases in future year returns.

Over the past 12 years, four independent techniques have been tested for their utility in generating reliable pre-season forecasts of Barkley Sound sockeye returns for harvest managers. The four techniques are known as the Salinity Survival Method (SSM), the Survival Stanza Method (SStM), the Sibling Age Class Method (SACM), and the Salmonid Enhancement Program Biostandard Method (SEPB). Updates on the performance of three of these techniques in 1999 are as follows:

1. The SStM forecast exhibits the best overall performance with a Mean Absolute Percent Error (MAPE) value of approximately 28 % over the most recent 12 years of forecasting (Table 5.). Further, SStM forecasts account for the majority of variations in returns if the extreme observation associated with the 1991 return year is omitted from the analysis (returns = 1.17 SStM forecasts - 66.35, r2 = 0.78, P<0.01).
2. The SSM forecast exhibits the next best performance with a MAPE value of 35 % over the most recent 12 years of forecasting. SSM forecasts also exhibit a statistically significant association with returns if the 1991 return year is omitted from the analysis ( returns = 0.83 SSM forecasts + 35.28, r2 = 0.77, P<0.01).
3. SEPB forecasts have performed well over some return intervals but not others. During the 1988-1998 testing interval SEPB forecasts exhibited a substantially higher MAPE value (54 %) than that displayed by both SSM and SStM forecasts (28-34 %). Large magnitude deviations between SEPB forecasts and actual returns tend to occur in consecutive years thus seriously eroding the confidence of harvest managers as well as fishers in their utility.
4. Different models provide highly divergent forecast alternatives for returns of Barkley Sound sockeye in the year 2000. Midpoint forecast estimates range from a low of 532,000 to a high of 1,900,000 Barkley Sound sockeye. Comparative performance of the various forecast options and supplementary observations of coho marine survival variations as a leading indicator suggest that sockeye returns in the year 2000 are likely to be closer to the lower than the upper end of this range. Given DFO's recent pursuit of a more risk averse approach to management we recommend initial adoption of the SStM forecast range of 485,000 (75 % probability) to 714,000 (25 % probability) sockeye as the preferred, pre-season forecast for the year 2000 (Table 14.).

Complete document (PDF, 114 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.