Science Response 2016/021
Pre-season run size forecasts for Fraser River Sockeye (Oncorhynchus nerka) Salmon in 2016
Context
The Fraser Sockeye return forecast is not a single value. Instead, the forecast is a probability distribution: at the 25% probability level, for example, there is a one in four chance returns will fall at or below the forecast value. The forecast value at the middle of the distribution, the 50% probability level, indicates where there is an equal chance returns will fall above or below this value. A stock’s forecast distribution reflects its past survival. If a stock’s return falls at the lower end of its forecast distribution, it is because the stock has experienced lower survival compared to previous years with similar numbers of spawners. Conversely, a return that falls at the higher end of a stock’s forecast distribution will indicate higher than average survival. If survival has been exceptional, then returns can fall outside the lowest or highest forecast value presented.
Most Fraser Sockeye mature as four year olds. Four year olds returning in 2016 are the offspring of adults that spawned in 2012. This consistent age-at-maturity is one of the factors that leads to persistent four year patterns in return abundances. The 2016 cycle has the lowest average return (3.9 M) of any of the four cycles and most of the forecast range for 2016 falls below this average. The 2016 forecast distribution indicates there is a one in 10 chance that the return will be less than 800,000, and a nine in 10 chance that the return will be less than 8.2 M. The median forecast (equal chances of higher or lower returns) is 2.3 million. A major factor driving the lower forecast in 2016 is the low escapement in 2012. For over half of the Fraser Sockeye stocks, spawner abundances in 2012 were below average and for six of these stocks they were the lowest on record. Bowron, Late Shuswap and Portage, for example, had particularly low effective female abundances in 2012 of only 10 fish each. Chilko, the stock expected to contribute the most to the 2016 returns, had a small brood year escapement in 2012 that was almost one third of its cycle average. If Fraser Sockeye survivals are below average, they could compound the impact of the low 2012 escapements, and result in return abundances that are less than the median forecast of 2.3 million.
With the exception of the large returns in 2010, total Fraser Sockeye returns have been less than their median forecasts (50% probability level) for much of the past decade. The returns for a number of stocks that contributed the most to last year’s 2015 forecast fell at the lower end of their forecast distributions. This pattern is consistent with lower than average survival during this period. For example, the marine survival for Chilko Sockeye, that contribute the most to the total 2016 forecast, has not exceeded 8% in each of the last 18 years. Yet, the median 2016 forecast for Chilko implies a marine survival of 9%. Thus, the potential impact of lower than average survival on the 2016 returns warrants consideration.
The 2016 return is concentrated among a few key stocks. Summer Run stocks contribute the most (70%) to the median (50% probability level) total forecast, especially Chilko (44%), and to a lesser extent Late Stuart and Stellako (20% combined), and Harrison (8%). Of the Early Summer Run stocks, two lower Fraser populations (Chilliwack: 6%; Pitt: 4%) are major contributors to the overall forecast. The forecast return to the Late Run aggregate is particularly poor. Even at the highest probability level (90%), the forecast (366,000) is close to half the cycle average return for this group (689,000). In the Late Run Timing group, Cultus Sockeye have been assessed as ‘endangered’ by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC). The forecast distribution for the Cultus return indicates there is a nine in 10 chance of a return less than 17,000 fish. Almost all of these fish will be of hatchery origin. This forecast range is below its cycle average of 22,000, and also well below this stock’s lower Wild Salmon Policy benchmark of 12,000 wild effective total spawners.
Unusually warm ocean temperatures developed in the Northeast Pacific Ocean in the second half of 2013, and these conditions have persisted to the time of this publication. These warmer temperatures, referred to as the ‘warm blob’, cover a broad spatial area in the Northeast Pacific Ocean, are as much as 3°C above average, and extend down to depths of 100 m. Fraser Sockeye stocks that returned in 2015 spent their final two years of life in these warmer waters, and a number of these stocks, including Chilko Sockeye, experienced low survival. However, it is unclear whether these high ocean temperatures influenced the poor Fraser Sockeye returns in 2015 because of the inconsistent responses across Fraser Sockeye stocks (Early Stuart, Raft, and Nadina had average to above average survivals), the information that suggests that freshwater strongly influenced the survival of some stocks (e.g. Weaver, Birkenhead and Shuswap), and because of the average to above average survival of other Sockeye stocks outside the Fraser (Barkley Sound, Nass, Columbia Sockeye). The inconsistent response of sockeye stocks to the ‘warm blob’ in the Northeast Pacific Ocean confounds attempts to draw an inference about the potential marine survival of four year old Fraser Sockeye returning in 2016.
We can, however, use estimates of returns for the 2015 four year olds to help generate estimates of 2016 five year olds. These siblings were together as eggs in their spawning gravel, and as juveniles in their rearing lakes and the North Pacific Ocean. The only difference between these siblings is that five year olds spend an additional year in the ocean. Sibling models were used to quantify the relationships between four year old and five year old returns for three Fraser River sockeye populations (Quesnel, Weaver and Birkenhead) with significant fractions of five year old fish expected to return in 2016. Due to the poor return of 2015 four year olds for these stocks, the application of sibling models reduced the total median forecast for these stocks by up to 75%. A similar approach was used for Harrison Sockeye where estimates of three year old returns in 2015 were used to forecast their four year old siblings in 2016, which reduced this stock’s forecast by 50%.
For Scotch, Seymour and Late Shuswap, Larkin models were applied since the 2015 median forecasts from these models were much closer to the actual returns compared to the Ricker model.
Due to the low total return of four year olds in 2015, a sensitivity analysis was conducted that compared forecasts resulting from adjusted five year old returns to those resulting from standard forecast models. This analysis adjusted the forecasts of five year old returns for all stocks, not just those that had a higher proportion of five year olds expected in 2016, as was the case for the official forecast. The total adjusted median forecast return (2.7 M at the mid-point of the forecast distribution) was slightly higher than the official median forecast (2.3 M) as presented in this document. This was attributed to the higher survival for a few stocks in 2015 for the Early Summer and Summer Run Timing groups. The Early Stuart forecasts were identical and the Late Run forecast was slightly smaller for the adjusted forecast.
This Science Response Report results from the Science Response Process of December 14, 2015 on the Pre-season abundance forecast for Fraser River Sockeye Salmon returns in 2016. The 2016 forecast relies on methods of past CSAS processes and publications (Cass et al. 2006; DFO 2006, 2009, 2011, 2012, 2013, 2014a, 2014b, 2015a, 2015b; Grant et al. 2010; Grant and MacDonald 2012; MacDonald and Grant 2012).
To support the 2016 Fraser Sockeye forecast, an additional Science Response process occurred on January 21-22 to summarize data and information on fish condition and/or survival from the 2013 spawners and their offspring. This Science Response will be posted on the Fisheries and Oceans Canada (DFO) Science Advisory Schedule as it becomes available.
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.
- Date modified: