Northwest Fisheries Science Center

Biological Spring Transition

We suggested earlier that the spring transition could be defined in several ways, one of which was the date that cold water first appeared in mid–shelf waters.  In Figure PST-02, we saw salmon survival correlated with the date when cold water first appeared at our baseline station, NH 05Figure BST-01 shows a similar relationship, but using the date when a northern (cold–water) copepod community first appeared at station NH 05.  We define this as the date of the biological spring transition

Plots of coho salmon survival versus length of the upwelling season (top) and day of the year when the copepod community transitioned (bottom). Figure BST-01. 
Upper panel:  Regression of coho salmon survival vs. day of the year when copepod community structure transitioned to a summer community. The earlier this transition takes place, the higher the coho salmon survival.
Lower panel:  Regression of coho salmon survival vs. length of the biological upwelling season, measured as the number of days that the summer community structure persisted. Numbers indicate the warm (red) and cold (blue) years.

We believe this date may be a more useful indicator of the transition in ocean conditions because it also indicates the first appearance of the kind of food chain that seems most favorable for coho and Chinook salmon; that is, one dominated by large, lipid–rich copepods, euphausiids, and juvenile forage fish.

Thus we suggest that potential feeding conditions for juvenile salmon are more accurately indexed using both northern copepod biomass and the biological spring transition date (as compared to an upwelling index, which is presumed to serve as an index of feeding conditions).  We say this in light of the following two instances wherein the upwelling index alone failed to correctly indicate feeding conditions.

First, during El Niño years, or years with extended periods of weak El Niño–like conditions, upwelling can still be strong (as in 1998), but can produce a warm, low–salinity, low–nutrient water type (rather than the expected cold, salty, and nutrient–rich water).  Upwelling of this water type results in poor plankton production.

A second example of upwelling as a misleading indicator occurred during 2005, when mean total upwelling levels from May to September were "average."  However, the zooplankton community did not transition to a cold–water community until August (Table BST-01).  Therefore, in spite of early upwelling, conditions for salmon feeding, growth, and survival were unfavorable throughout spring and most of summer 2005.

The end of the upwelling season marks the return of a winter community for zooplankton, the timing by which the fall transition is measured. 


 
Table BST-01.  Historical dates of the biological spring transition, as measured by the timing of change in the zooplankton from a winter to a summer community. 
  Arrival of cold–water copepod community  Length of cold–water
copepod presence
Year   Start date End date  (in days)
 
1970   ~20 Mar 20 Oct   214    
1971   20 Mar 6 Nov   231    
1983   21 Jul 19 Aug   29    
1996   25 Jul 7 Oct   74    
1997   15 May 28 Aug   105    
1998   20 Jul 20 Sep   62    
1999   14 May 4 Nov   174    
2000   6 Apr 23 Oct   200    
2001   11Apr 7 Nov   200    
2002   18 Apr 1 Nov   197    
2003   5 June 3 Oct   120    
2004   11 May 14 Oct   156    
2005   28 Jul 28 Sep   62    
2006   30 May 31 Oct   154    
2007   22 Mar 10 Dec   263    
2008   4 Mar 27 Oct   237    
2009   6 Mar 5 Oct   231    
2010   18 Jun 24 Nov   159    
2011   23 Mar 29 Sep   190    
2012   4 May 25 Oct   174    
2013   10 Mar 26 Sep   200    
2014   5 June 7 Sep   94    
2015   Never Never   0    
2016   Never Never   0    
 


These changes in community type occur because of coastal currents, which reverse in spring to flow from the north with the onset of upwelling.  Another reversal occurs in the fall, when the northward–flowing Davidson Current appears on the shelf due to winter downwelling. 

Arrival of the "northern" species in spring signals that the ecosystem is primed to begin a productive upwelling season. Also listed is length of the upwelling season in days, as reckoned by the zooplankton. Note that over the years of 2007-2009 and again in 2011 and 2013, the transition date came very early, in March, whereas 2015 and 2016, the biological spring transition never occurred.

Both the date of "biological spring transition" and "length of the biological upwelling season" also correlate well with counts of adult spring Chinook salmon (Figure BST-02) and adult fall Chinook salmon (Figure BST-03) at Bonneville Dam 2 years later.

Counts of spring Chinook salmon jacks at Bonneville Dam vs. date of the "biological spring transition" and "length of the biological upwelling season."  Figure BST-02.  Spring Chinook salmon adult counts at Bonneville (lagged by 2 years) vs. date of biological spring transition (upper panel) and length of the biological upwelling season (lower panel). Numbers indicate the warm (red) and cold (blue) years. Numbers indicate the warm (red) and cold (blue) years. Years in black were outliers and were excluded from the regression.

*outliers were excluded using Cook's Distance


Fall Chinook adults counts at Bonneville vs. date of the biological spring transition and length of the biological upwelling season.  Figure BST-03.  Fall Chinook salmon adult counts at Bonneville (lagged by 2 years) vs. date of biological spring transition (upper panel) and length of biological upwelling season (lower panel). Numbers indicate the warm (red) and cold (blue) years. Numbers indicate the warm (red) and cold (blue) years. Years in black were outliers and were excluded from the regression.

*outliers were excluded using Cook's Distance