Transcript 130

ICES/NAFO Decadal Symposium Santander, Spain May 12th 2011
The serial recruitment failure to North Sea fish
stocks during the 2000s, is climate to blame?
Geir Ottersen
E. M. Olsen, T. Falkenhaug, P. Licandro, M. Llope
and others in the RECNOR team
Centre for Ecological and
Evolutionary Synthesis
Serial recruitment failure
Herring
Sandeel
Cod
Norway pout
Increasing sea temperatures
M. Llope
Increasing ambient temperatures IBTS Q1
Cod
Herring
G. Dingsør, G. Ottersen et al. In prep
Switch from C finmarchicus
-> C helgolandicus
M. Edwards (2008)
G. Beaugrand
Dynamics of C. finmarchicus (prefered food of larval cod)
and C. helgolandicus co-occuring in the Skagerrak
Monitoring of plankton at station in the Skagerrak:
Sampling of zooplankton: 2 times per month since 1994
WP2 vertical net tows (180µm), 50 – 0 m
Calanus finmarchicus
female
Samples recently reanalysed for identification of C. fin and C. hel
Aims:
- To describe the seasonal and interannual variation in relative
proportions of the two species.
- Reveal possible causes for the observed variations.
Calanus helgolandicus
female
T. Falkenhaug, E. Bagøien, C. Broms work in prep.
autumn
spring
2005
2006
2007
2008
Numbers/m2
20000
18000
16000
14000
12000
10000
8000
6000
4000
2000
0
Chel
Cfin
27/01/2005
08/03/2005
19/04/2005
01/06/2005
13/07/2005
13/09/2005
26/10/2005
14/12/2005
12/02/2006
31/03/2006
15/05/2006
01/07/2006
31/08/2006
09/10/2006
07/12/2006
06/02/2007
19/03/2007
03/05/2007
18/06/2007
07/08/2007
18/09/2007
03/11/2007
08/01/2008
04/03/2008
07/04/2008
14/05/2008
27/07/2008
13/09/2008
04/11/2008
16/12/2008
Calanus CV-VI
•Seasonal variation: >80% C. finmarchicus in spring (Jan-June);
>80% C. helgolandicus in autumn (July-November)
•Interannual variations: The relative proportion of the two species differs
between years: ”finmarchicus years” and ”helgolandicus years”
Long term changes and interannual variations in ratio of
C. finmarchicus/ C. helgolandicus
1= 100% C. finmarchicus (blue)
0=100% C. helgolandicus (red)
Month
CVI females
The period of C. helgolandicus
dominance (ratio>0,5) has appeared
earlier in the season in recent years
(2004-2008).
Year
Conclusions calanus fin vs cal helg
• C. finmarchicus occur in high abundance in spring, while C.
helgolandicus peaks later in the season at lower abundances.
• The annual temperature regime in this region (2-20 ºC) allows
both species to co-occur, but are seasonally separated
through their different temperature optima (niche separation).
• The seasonal increase in temperature triggers a shift from a
system dominated by C. finmarchicus to a system dominated
by C. helgolandcus. This shift occurs in June, at ~13 ºC.
• Higher temperatures, earlier in the season will trigger earlier
shifts from C.fin to C.hel.
• This is bad news for early life stages of cod, which have
Cal.fin. as preferred food.
North Sea Herring
• Year class strength determined from pelagic larval
to juvenile stage (1th winter)
• Low survival through this stage recent years
• YCS of 0-ringers and 1-ringers negative correlated
with bottom temperatures
Nash & Dickey-Collas 2005
The reduced herring larval survival does not
appear to be due to the fishery, maybe it is related
to changes in the plankton food of herring larvae?
Payne et al. 2009
Are recent planton changes of significance to herring larvae?
Since 2001
- Decrease of biomass of small (< 2mm) plankton size
fraction, i.e., the prey of the herring larvae
- Increase of biomass of mesozooplankton > 2mm), i.e.,
potential competitors and predators of herring larvae
A combined effect of predation (top-down) and
competition for food (bottom-up) could be a possible
cause of the low survival rate of herring larvae
Licandro et al. In prep.
Enhancing stock-recruitment models for North Sea cod
by including climate and zooplankton
Modelling the Spawning Stock-Recruitment
relationship for North Sea cod by a linear relation?
?
?
Modelling the Spawning Stock-Recruitment
relationship for North Sea cod by a Ricker type relation??
Modelling the Spawning Stock-Recruitment
relationship for North Sea cod by a Beverton-Holt type relation??
Enhancing the S-R relation by including environmental
effects in a combined Beverton-Holt and Ricker model
A-priori set of stock (S) and recruitment (R) models
Model
1
2
3
4
Structure
1
log(R/S) = a + log(exp(-b•S))  log(R)-log(S)=a-bS
2
log(R/S) = a – log(1 + exp(c)•S/maxS)
3
log(R/S = a + log(exp(-b•S)•(1-Z) + 1/(1 + exp(c)•S/maxS)•Z)
4
log(R/S) = a – (a1•T) + log(exp(-b·S)•(1-Z) + 1/(1 + exp(c)•S/maxS)•Z)
Traditional Ricker model (overcompensation)
Traditional Beverton-Holt model
Combined Ricker-Beverton-Holt model including a Z effect only
Combined Ricker-Beverton-Holt model including Z and T effects
In combined models Ricker term dominate at low food levels, B-H at higher
Model 4 best model as selected by the Akaike Information Criteria (AIC)
800
800
100 150 200 250 300
600
400
0
800
50
50
100 150 200 250 300
0
400
600
Very warm
200
400
600
Warm
Average
200
400
200
400
200
800
0
200
Recruitment
Cold
600
Very cold
600
800
North Sea cod: Effects of spawning stock biomass, food availability
(zooplankton index), and sea surface temperature on recruitment at age 1
0
50
100 150 200 250 300
0
50
100 150 200 250 300
Spawning stock biomass
50
100 150 200 250 300
Conclusions
stock-recruitment models for North Sea cod
Our results suggest that the stock-recruitment relationship of North Sea
cod is not stationary, but that its shape depends on environmental
conditions, i.e food (zooplankton) availability and sea temperature
A full recovery of North Sea cod is not to be expected until the
environment – both food availability and temperature - becomes more
favourable
The future: Effects of climate change on the
survival of larval cod (estimated by models)
Trond Kristiansen (IMR) and others
Lofoten
North Sea
Projected temperature development (value today=0.0)
North Sea
Lofoten
Predicted survival rate in Lofoten
(distinct increase)
Predicted survival rate in the North Sea
(weak decrease)
MAIN CONCLUSIONS (that partly answer our initial question)
Temperatures increased, also ambient winter temp of herring and cod
Changes in zooplankton community
both in the NS proper (CPR) and the Skagerrak (WP2 net tows )
In particular decrease in Cal fin and increase in Cal hel abundance
shift to Cal hel dominance earlier in year with higher temp (Skagerrak)
Decrease in biomass of small plankton, increase of mesozooplankton:
Shift from prey of herring larvae towards more competitors and predators
unfavorable for herring recruitment
Shape of stock-recruitment relationship of North Sea cod depends on
zooplankton availability and temperature
Present situation unfavorable for cod recruitment
IBM predicts higher future temperatures and lower survival for larval cod
Thanks, that’s all!