Synthesis Workshop Summary
The Climate-Ocean Regime Shift Hypothesis
of the Steller Sea Lion Decline
December 4-5, 2003
Newport Beach, CA
11 Proposals funded by NOAA OAR/NOS CIFAR
Object: Synthesis Review Paper of the Research Results
Relate temporal variability in the physical system
to the ecosystem changes
Workshop Participants
Funded Investigators:
M. Alexander, A. Capotondi (CDC)
S. Bograd (PFEL)
K. Coyle (UAF)
B. Finney (UAF)
G. Hunt, J. Jahncke (UCI)
N. Kachel (PMEL)
N. Mantua, C. Marzban (UW)
H. Maschner (Idaho State)
W. Maslowski (NPS)
A. Miller, D. Neilson, E. Di Lorenzo, H.-J. Kim (Scripps)
S. Okkonen (UAF)
T. Royer, C. Grosch (ODU)
A. Trites, E. Gregr (UBC)
J. Wang (ARL)
Program Manager:
J. Calder (NOAA)
Invited Contributors:
L. Fritz (NMFS)
J. Overland (PMEL)
Synthesis Paper Highlights
Title: The Climate-Ocean Regime Shift Hypothesis
of the Steller Sea Lion Decline
Forum: Fisheries Oceanography
Authors: Trites, Maschner and all contributors
Thesis: Spatial and temporal variations in the ocean climate system
are creating adaptive opportunities for high trophic levels
which is the underlying mechanism for the decline of the
Steller sea lion populations in the western Gulf of Alaska.
Important results:
Temporal issue - 1970s to 1990s changes
Spatial issue - East vs west asymmetry in Gulf of Alaska
Biogeographic transition point at 170W
Basin-scale climate changes have regionally sensitive impacts
Upscaling from local complexities to broadscale regularities
Eddy variance changes in western Gulf
Outline
The Climate-Ocean Regime Shift Hypothesis
of the Steller Sea Lion Decline
Physical Ocean-Atmosphere Data Analysis
Ocean Model Results: Coarse and Eddy-Permitting
Bio-Geography: Samalga Pass Transition Point
Paleo-Perspective: Isotopes and Archeology
Outline
The Climate-Ocean Regime Shift Hypothesis
of the Steller Sea Lion Decline
Physical Ocean-Atmosphere Data Analysis
Ocean Model Results: Coarse and Eddy-Permitting
Bio-Geography: Samalga Pass Transition Point
Paleo-Perspective: Isotopes and Archeology
Sea Surface Temperature Trends in the Gulf of Alaska
• Analysis of COADS SST, 1950-97
• SST clusters into 5 dynamic regions
• regional differences in long-term
SST behavior in eastern Subarctic
Pacific
Bograd et al., 2004
NL PC1: 32% of variance
Nonlinear PCA
results from
Marzan,
Mantua, and
Hare:
based on 22
abiotic indices
from the Pacific
for 1965-2001
NL PC2: 23% of variance
North Pacific Regime Shifts: State Changes in Forcing
SSTA
SLPA,
WindA
“cool”
“warm”
“cool?”
Peterson and Schwing, 2003
Modeled Gulf of Alaska coastal discharge
Royer and Grosch
Changes in Ekman pumping
Pre-shift mean
conditions
1960-75
Dec-May
Change
after shift
(1977-97)
-(1960-75)
Capotondi et al.
Outline
The Climate-Ocean Regime Shift Hypothesis
of the Steller Sea Lion Decline
Physical Ocean-Atmosphere Data Analysis
Ocean Model Results: Coarse and Eddy-Permitting
Bio-Geography: Samalga Pass Transition Point
Paleo-Perspective: Isotopes and Archeology
Pycnocline (26.4) depth changes
Period2 (77-97) – Period1 (64-75)
P = OWS P
G = GAK1
Model vs. Obs PAPA
1970
1980
Model vs. Obs GAK
1990
2000
1970
1980
Capotondi et al.
Eddy-Permitting Model
Eddy Surface Currents
Before 76-77 Shift
After 76-77 Shift
Difference
Kodiak Is.
- More eddies north of Kodiak
- Fewer eddies southwards
Miller et al.
Naval Postgraduate School numerical model
23 year (1979-2001) simulation with real winds
NPS Model SST along shelf break
Deep Salinity along shelf break
Okkonen et al.
Physical-Biological
Ocean Hindcast
(Chai et al. model)
Decrease in large
zooplankton
(and large phyto,
small zoo) after the
1976-77 shift…
…But occurs only
during spring bloom
Alexander et al.
Outline
The Climate-Ocean Regime Shift Hypothesis
of the Steller Sea Lion Decline
Physical Ocean-Atmosphere Data Analysis
Ocean Model Results: Coarse and Eddy-Permitting
Bio-Geography: Samalga Pass Transition Point
Paleo-Perspective: Isotopes and Archeology
Locations of Passes sampled May-June 2002
Ak
u
Pa
ss
as
s
alg
s
as
kP
m
Sa
as
aP
ss
Pa
kta
Pacific Ocean
s
s
Pas
ss
Pa
nga
a
Tan
am
gu
Se
u
Am
na
Um
SamalgaBering
Pass forms
Sea a boundary
between shelf water to the east
and open ocean waters to the west
tan
Un
im
ak
P
>
Coyle et al.
Surface
Salinity
along
Aleutian
Islands
East of
Samalga Pass,
ACC has
strong influence
West of
Samalga Pass,
Alaska Stream
has strong
influence
Ladd et al.
(2004)
Abundance vs. Transect May-June 2002
Unweighted Means vs Transect and Species
600
Calanus marshallae
N. plumchrus & N. flemingeri
Acartia spp.
400
300
200
100
0
Ak1
Un1
Tng
Sgm
Amk
Smg
Unk
Ak2
Un2
Transect
Unweighted Means by Transect and Species
0.6
Mean Abundance (No. m )
0.5
Thysanoessa inermis
Euphausia pacifica
-3
Mean Abundance (No. m-3)
500
0.4
0.3
0.2
0.1
0.0
Un1
Ak1
Tng
Sgm
Transect
Smg
Un2
Ak2
Coyle et al.
Seabirds Reflect Biogeographical Boundaries
Short-tailed shearwaters
54
52
-178
-176
West of
Northern fulmars
Samalga Pass: 54
Fulmars and
52
Auklets
(open ocean
-178
-176
food web)
Tufted puffins
dominant
-174
-172
-170
-168
-166
-164
-174
-172
-170
-168
-166
-164
-174
-172
-170
-168
-166
-164
-174
-172
-170
-168
-166
-164
54
52
-178
-176
Small alcids
Jahncke
et al.
54
52
-178
-176
East of
Samalga Pass:
Shearwaters
and Puffins
(coastal
food web)
dominant
Outline
The Climate-Ocean Regime Shift Hypothesis
of the Steller Sea Lion Decline
Physical Ocean-Atmosphere Data Analysis
Ocean Model Results: Coarse and Eddy-Permitting
Bio-Geography: Samalga Pass Transition Point
Paleo-Perspective: Isotopes and Archeology
Bering Sea: Skan Bay Core
30
Paleoproductivity
> 1976 Regime-shift
-20.0
20
-21.0
d13C - decrease
Opal - variable
Gulf of Alaska:
d13C - increase
Opal - increase
10
-22.0
-23
Opal (%)
Paleoproductivity Bering Sea:
d13C
Gulf of Alaska: GAK 4 Core
d13C
-24
10
d13C
Opal (%)
8
-25
1700 1750 1800 1850 1900 1950 2000
Finney et al.
Year AD
Opal (%)
12
GOA GAK 4
Skan Bay
Relative Abundance of SSL
based on Archaeological Data
Climate
400
1200
1600
2000
2400
cool, mesic
2800
Years Before Present
800
cool, wet
3200
warm, mesic
3600
50
40
30
20
10
Below Average Percent SSL
0
10
20
30
40
50
Above Average Percent SSL
Maschner et al.
Some Outstanding Questions
The Climate-Ocean Regime Shift Hypothesis
of the Steller Sea Lion Decline
What mechanisms control the restructuring of the ecosystem by
climate, especially concerning fish?
How do human activities and ecosystem interactions work with
variable climate forcing to alter Steller sea lion populations?
Do Steller sea lions feed in eddies?
What can be said about changes in sub-surface conditions?
How does vertical mixing vary in space and time?
What other regime shifts may have occurred?
Synthesis Paper Highlights
Title: The Climate-Ocean Regime Shift Hypothesis of
the Steller Sea Lion Decline
Forum: Fisheries Oceanography
Authors: Trites, Maschner and all contributors
Thesis: Spatial and temporal variations in the ocean climate system
are creating adaptive opportunities for high trophic levels
which is the underlying mechanism for the decline of the
Steller sea lion populations in the western Gulf of Alaska.
Important results:
Temporal issue - 1970s to 1990s changes
Spatial issue - East vs west asymmetry in Gulf of Alaska
Biogeographic transition point at 170W
Basin-scale climate changes have regionally sensitive impacts
Upscaling from local complexities to broadscale regularities
Eddy variance changes in western Gulf
Extras
Sea Surface Temperature Trends in the Gulf of Alaska
(a) trend 1
• loadings increase west to east
• warming from early 70’s
• accelerated warming after 76
(b) trend 2
• accentuates trend 1
• pronounced drop after 58
• rapid rise after 72
• positive loadings = strong
warming after 72
(c) trend 3
• strong warming in eastern
GOA during 57-58 event
• broad warming after 76
(d) trend 4
• strong interannual
variability
• impacts from El Niño events
• strong loadings in coastal
BC and WWD
Bograd et al., 2004
Nonlinear PCA
results from
Marzan,
Mantua, and
Hare:
based on 45
biotic indices
from the Bering
Sea and Gulf of
Alaska for 19652001
NL PC1: 39% of variance
NL PC2: 18% of variance
Kendall’s  trends in Alaska adult SSL data
(Marzban, Mantua and Hare)
99% ci --
99% ci -For SSL data from 1970-2001
Seasonal Amplitude and Phase Changes
in Coastal Freshwater Discharge
Royer and Grosch
Wavelet Analysis of Coastal Freshwater Discharge
Royer and Grosch
Naval Postgraduate School numerical model
23 year (1979-2001) simulation with real winds
Red line indicates shelf break
(for Hovmuller diagram)
Okkonen et al.
T and S vs. Transect May – June 2002
Temperature vs Transect; Unweighted Means
Wilks lambda=.02144, F(64, 381.4 )=5.6421, p=0.0000
Vertical bars denote 0.95 confidence intervals
6.5
5.5
5.0
4.5
Unimak Pass 2
Samalga Pass
Amukta Pass
Seguam Pass
Tananga Pass
Unimak Pass 1
Akutan Pass 1
3.5
Akutan Pass 2
UpperMixedTemp
LowerMixedTemp
MeanTemp
4.0
Umnak Pass
Temperature ( oC)
6.0
Transect
Salinity vs. Transect; Unweighted Means
Wilks lambda=.02144, F(64, 381.4 )=5.6421, p=0.0000
Vertical bars denote 0.95 confidence intervals
33.8
33.6
33.4
UpperMixedSal
LowerMixedSal
MeanSal
33.0
32.8
32.6
32.4
32.2
32.0
Transect
Coyle et al.
Unimak Pass 2
Akutan Pass 2
Umnak Pass
Samalga Pass
Amukta Pass
Seguam Pass
Tananga Pass
31.6
Unimak Pass 1
31.8
Akutan Pass 1
Salinity (PSU)
33.2
Schematic of Traditional Aleut Knowledge of
Recent North Pacific Cycles
2000
1990
1980
1970
1960
1950
High
Average
Gadids
Low
unknown
1930
unknown
1920
unknown
1910
unknown
1900
unknown
1890
unknown
1880
unknown
1870
Average
Year
1940
High
King Crab
Maschner et al.