Research Document - 2010/042

Pre-season run size forecasts for Fraser River Sockeye salmon (Oncorhynchus nerka) in 2010

By S.C.H. Grant, C.G.J. Michielsens, E.J. Porszt and A. Cass

Abstract

• Pre-season forecasts of salmon abundance are generally highly uncertain due the combination of historical variability in annual survival rates (stochastic uncertainty) and uncertainty regarding future survival rates. Fraser River Sockeye Salmon (Fraser Sockeye) forecasts in 2010 are especially uncertain given the decreasing trends in productivity for most stocks in recent years and, in particular, the unexpectedly poor returns in 2009.
• Uncertainty that is attributed to stochastic (random) variability in annual Fraser Sockeye survival is communicated in the 2010 forecast paper through a series of forecasted values that correspond to standardized cumulative probabilities (10%, 25%, 50%, 75%, 90%) (Figure 10). For example, there would be a one in four chance at the 25% probability level that the actual number of returning Sockeye will be at or below the forecasted value given the assumptions about future survival.
• Uncertainty about future Fraser Sockeye survival is communicated in this paper through the presentation of three alternative cases (forecast tables) to characterize stock productivity for the 2010 returns:
• Case 1.  “Long Term Average Productivity” assumes that average stock productivity (across entire stock-recruitment time series) will persist through to 2010 (Table 2). Methods and model ranks were identical to the 2009 forecast (DFO 2009).
• Case 2.  “Recent Productivity” assumes that recent productivity trends will persist through to 2010 (Table 3). To forecast age-4 recruits, this case includes the addition of three new models that consider recent productivity. Model performance was evaluated in recent years only (brood years 1997-2004). Age-5 recruits were forecast using preliminary productivity for the 2005 brood year (2009 returns).
• Case 3.  “Productivity Equivalent to the 2005 Brood Year (2009 Returns)” assumes this low productivity will re-occur in 2010 (Table 5). Age-4 and age-5 recruits were forecast using preliminary productivity for the 2005 brood year (2009 returns),
• At the time of this paper, the 2005 brood year productivity data used to forecast age-5 recruits in the Case 2 forecast and age-4 and age-5 recruits in the Case 3 forecast were preliminary and also do not include the age-5 recruits that will return in 2010.

• The forecast with the greatest degree of belief (as recommended by the March 9, 2010 Canadian Science Advisory Secretariat (CSAS) Regional Advisory Process (RAP)) was Case 2 (“Recent Productivity”). For this case, the number of returning Fraser Sockeye in 2010 will range from 7.0 million-18.3 million at the 25% to 75% probability levels. The same probability range each of the four run timing groups are as follows: Early Stuart Run: 26,000-66,000; Early Summer Run: 374,000-1.6 million; Summer Run: 1.6 million-4.3 million; and Late Run: 5.0 million-12.3 million (Figure 11 and Table 3).
• For Case 1 (“Long-Term Average Productivity”), the forecast range is from 8.4 million to 23.5 million at the 25% to 75% probability levels (Figure 11 and Table 2).
• For Case 3 (“Productivity Equivalent to the 2005 Brood Year (2009 Return)), the forecast range is from 1.6 million to 7.9 million at the 25% to 75% probability levels (Figure 11 and Table 5).

Complete document (PDF, 1854 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.