Blame-storming North Pacific Climate Extremes Since 2013

Nate Mantua

Southwest Fisheries Science Center

Santa Cruz, CA

Collaborators: Jim Johnstone, Nick Bond, Megan Cronin, Howard Freeland,

Manu DiLorenzo, Paul Fiedler, Mike Alexander, Mike Jacox, and Steven Bograd

SWFSC - FED

March 22, 2017

2017 Salmon Ocean Ecology Meeting, Seattle

El Niño

Who or what is to blame?

National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 2

• 2013-14: The warm blob in the Gulf of

Alaska

• 2014-15: Record warming of the

California Current System

• 2015: record warm winter in Western

North America and a snow drought

• 2016: record high SSTs in the eastern

Bering Sea

Johnstone and Mantua, PNAS, 2014, DOI: 10.1073/pnas.1318371111

Global warming, but Eastern

Pacific cooling:

(1998-2013) – (1977-1997)

NASA Global Land-Ocean Temperature Index

Leading EOFs/PCs of NEP SST and SLP (Johnstone and Mantua, 2014: PNAS)

SST1: 30% variance

SLP1: 22% variance

SST PC1

SLP PC1

4

Relate variations in SLP and SST patterns with a Stochastic Climate Model

SSTt = α SSTt-1 + ßSLPt + εt

Coefficients first from lag-1 autoregression of SST Slight adjustments guided by sensitivity experiments

α = 0.81 (persistence term) ß = 0.27 (SLP perturbation)

SSTt = 0.81 SSTt-1 + 0.27 SLPt + εt

5

(Johnstone and Mantua, 2014: PNAS)

SST1 modeled from SLP1 forcing + persistence

The simple stochastic

climate model does about

equally well reproducing

observed monthly and

annual mean variations in

SST1

Century long warming

trend in the NEP Arc

related to trend to lower

SLP between Hawaii and

the West Coast

6 (Johnstone and Mantua, 2014: PNAS)

Marine heat waves Hobday et al. 2016, Prog. In Oceanogr:

a prolonged warm-water event described

by its duration, intensity, rate of evolution

and spatial extent

• recent events in Western Australia

(2011), the Mediterranean (2003), and

NW Atlantic (2012)

Scannell et al. 2016, Geophys. Res.

Letts: for N. Atlantic and N. Pacific,

small area anomalies more frequent

than large area anomalies, fitting a

power law relationship

• Marine heat waves are influenced

by intrinsic climate variability

Western N. Pacific Eastern N. Pacific N. Atlantic

Sea Surface Temperature Anomaly (relative to 1981-2010)

Extraordinary conditions • Northeast Pacific warmed before

the Eastern Tropical Pacific

– Gulf of Alaska has been

extremely warm since Fall

2013

– Baja and So. California near-

shore was extremely warm

from June 2014 – August

2016

• Wide-ranging impacts on Pacific

marine life

8

June 2014

False Killer Whales, 24 Feb 2015; Photo: M. Robbins

California Current Sea Surface Temperature Anomaly (relative to 1981-2010)

Extraordinary conditions

9

False Killer Whales, 24 Feb 2015; Photo: M. Robbins Fiedler and Mantua, in prep

2014-15: record

warm in

California

• Surface air temperature

record for July 2014-June

2015 was almost off the

charts, ~ 1 °C warmer

than the previous record

• This “hot drought” was

amplified ~30% by high

temperatures

• 2016 just a bit cooler than

2014

Record high Eastern Bering Sea

SSTs in Summer 2016

July 15, 2016

Ridiculously Resilient Ridge(s) are to blame!

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 12

Image from ww2.kqed.org

• A few “ridge patterns” kept storms away and temperatures high in CA

• What caused the atmosphere to get stuck like this?

Unusual warm-west cool-east temperatures in the tropical Pacific? The warm Arctic

and reduced sea ice cover? The warmth of the northeast Pacific Ocean? A

combination of these factors?

In winter 2013-14 extreme warm temperature anomalies developed under the

influence of an extremely persistent high pressure anomaly in the NEP

SSTa in February 2014 Sea Level Pressure anomalies:

Oct 2013-Jan 2014

Mechanisms “warming” the Blob

National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 14

Bond et al; Geophysical Research Letters

Volume 42, Issue 9, pages 3414-3420, 5 MAY 2015 DOI: 10.1002/2015GL063306

http://onlinelibrary.wiley.com/doi/10.1002/2015GL063306/full#grl52851-fig-0004

-9

-8

-7

-6

-5

-4

-3

-2

-1

01980 1985 1990 1995 2000 2005 2010 Advec on

NetSurfaceFluxes

Entrainment

MixedLayerTempChange

• Blob region September to

February mixed-layer

temperature changes from the

NCEP GODAS

• Weak wind stress and wind

stress curl reduced

entrainment, reduced net

surface heat fluxes, and

reduced advective cooling all

combined to create extreme

warm anomalies in fall/winter

2013-14

Advection

Net Sfc

heat flux

entrainment

ML Temp

change

What Caused the warm Blob in 2013/14?

National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 15

• The proximate cause: persistent ridge of high pressure – and the lack of

stormy weather and strong winds (Bond et al. 2015, GRL)

• Forcings: warm SSTA in the far western tropical Pacific and cold SSTA

in the eastern tropical Pacific (Wang et al. 2014, Hartmann 2015; Lee et

al 2015; Seager et al 2014); reduced Arctic/subarctic sea ice (Lee et al.

2015, Kug et al. 2015), and the warm blob itself (Lee et al. 2015)

Images from: https://www.climate.gov/news-features/blogs/enso/tropics-prime-suspect-behind-warm-cold-split-over-north-america-during

“North Pacific mode” SSTA pattern 500mb ht. anomalies associated with NPM

H

L

L

H L

L L

observed simulated

CCS SSTa Index CCS SSTa Index

Winter 2014 Winter 2015

SLPa SLPa SLPa SLPa

SSTa SSTa SSTa SSTa

C C

mbar mbar

ARC SSTA GOA SSTA

• Winter (JFM) 2014

warm blob evolved

into an ARC-like

warming pattern fall

14/winter 15

• The associated

persistent SLP

anomalies in the

Northeast Pacific also

changed

SLPa GOA Index

SSTa GOA Index

SLPa AR1 model

R=0.66

SLPa ARC Index

SSTa ARC Index

SLPa AR1 model

Gulf of Alaska NE Pacific ARC 2014/15

2000 2010 1990 1980 2000 2010 1990 1980

2014/15

R=0.79

0

2

-2

0

2

-2

0

2

-2

0

2

-2

0

2

-2

0

2

-2

AR1 AR1

Figures from Di Lorenzo and Mantua (2016):

Nature Climate Change

GOA warming

followed by “ARC”

pattern warming has

happened before:

1957-58, 1962-63,

and 1991-92

Never as extreme as

2014-15 though

SSTa ARC Index

3/27/2017 19

50-member ensemble using

observed tropical SSTA as

forcing, climatological SSTs

and a 50m “slab-ocean”

elsewhere

• Ensemble average

response has a 1.5C

anomaly for JFM2015 in

the ARC region

• 9 ensemble members

matched or exceeded the

observed 3 C anomaly

Atmospheric

circulation model

experiments

Persistence and evolution of SST/SLP anomalies involved well-

known extratropical-tropical-extratropical teleconnections

• However, only about half the magnitude of the Arc SST anomaly

could be related to tropical-origin forcing

+ “noise” (internal atmospheric

variability)

30-member ensemble from NCAR’s CESM under historical forcing (1800-

2005)+RCP8.5 (2006-2100)

A role for anthropogenic forcing? Earth System model

experiments

• PDO and NPGO patterns in the model very similar to observed patterns

• Model ensemble shows increasing variance in the PDO (+7%) and NPGO (+16%)

patterns over the 1920-2100 period

• variance trends in the observations are not so simple

2013/14 2015/16

SLPA Oct-Mar 2013-14

SSTA Oct-Mar 2013-14 SSTA Oct-Mar 2014-15 SSTA Oct-Mar 2015-16

SLPA Oct-Mar 2014-15 SLPA Oct-Mar 2015-16

H

L

L

H L

SLP-forced AR1 model for

ARC SST pattern

generally does a good job

matching observations,

but not in 2016

3/27/2017 24

May-Sept 2016

A period of extraordinary

atmospheric conditions

followed the end of the El

Niño winter … record high

sea level pressure

anomalies over the N.

Pacific from May-

September 2016

What happens after 5

months of fair weather

across the N. Pacific and

Bering Sea???

Summer 2016 30-65N, 160E-140W

25

The blob makes a comeback!

SSTA - September 2016

1980 2015

http://www.cpc.ncep.noaa.gov/products/GODAS/

Rapid growth of near-

surface warm anomalies

beginning in June, rapid

decline beginning in

October

X

JAN

2013

JAN

2014

JAN

2015

JAN

2016

JAN

2017

3/27/2017 26

Sept 30 2016 SSTA Oct 31 2016 SSTA

L

October takes a bite

out of the Blob

Change in Weekly SST Departures over the Last Four

Weeks

During the last four weeks, small regions of equatorial SST anomalies increased or

decreased across the central and eastern Pacific.

October 2016 SLPA

Water year

precipitation

“% normal”

Oct 1-Mar 21

102

103

88

106

121

112

98

13098

83

132

119

126

78

113

102

116

43

105

127

118

97

95

112

109

109

128

106

127

170

90

117

120

176

132

91

130

110

104

27

97

110

111

95

106

101

120

133

110

101

159

93

105

-997

197

103

-997

135

150

90

106

86

152

97

95

115

29

145

106

197

115

130

1

111

144123

21

127

178105

104

112

166

154

188

40

124

79

23

107

135

0-997

108

94

57

99

99

93

110

106

61

174

109

107

113

112

129

-997

135

129

34

109

145

177

61

0-997

111

157

140

177

104

163

-997

109

126

204

140

150

Mar 21, 2017Current Snow WaterEquivalent (SWE)Basin-wide Percent of 1981-2010 Median

unavailable *

<50%

50 - 69%

70 - 89%

90 - 109%

110 - 129%

130 - 149%

>= 150%

Prepared by:USDA/NRCS National Water and Climate Center Portland, Oregonhttp://www.wcc.nrcs.usda.gov

Provisional data subject to revision

Westwide SNOTEL Current Snow Water Equivalent (SWE) % of Normal

0 150 30075

Miles

* Data unavailable at time of posting

or measurement is not representative

at this time of year

The snow water equivalent percent of normal represents the current snow water equivalent found at selected SNOTEL sites in or near the basin compared to the average value for those sites on this day. Data based on the first reading of the day (typically 00:00).

A very different weather

pattern from Nov – Feb 2017

The Northeast Pacific Ocean is starting to

look more normal … even a little cool in the

“original Blob” region

March 20, 2017

We We

36.5

37.0

37.5

38.0

38.5

39.0

−124 −123 −122

x

y

Nutrients on shelf and shelf break

Primary

productivity on

shelf

Spring transition

to more

consistent, deeper

upwelling Condition and

production of krill

on shelf

Physics Primary

production

Forage

Poleward transport events from GoF Retention of krill in GoF

Spring

increase of

primary

production

Salmon emigrate to shelf

Cohort survival

Salmon condition

Pt Reyes

GoF

38N

37N

123W

Chinook salmon

MB

Abundance of

rockfish on shelf Introduction of nutrients

California

Growth and

production of krill

on shelf during

winter

Strength of NPH

Rockfish feed on

krill and krill eggs

Coastal

upwelling

in winter

Winter

Spring

Canada

Mexico

Pacific

Ocean

Pt Arena

Predation

Freshwater

conditions

Wells et al. 2016; Mar. Ecol. Prog. Ser.

36.5

37.0

37.5

38.0

38.5

39.0

−124 −123 −122

x

y

Nutrients on shelf and shelf break

Primary

productivity on

shelf

Spring transition

to more

consistent, deeper

upwelling Condition and

production of krill

on shelf

Physics Primary

production

Forage

Poleward transport events from GoF Retention of krill in GoF

Spring

increase of

primary

production

Salmon emigrate to shelf

Cohort survival

Salmon condition

Pt Reyes

GoF

38N

37N

123W

Chinook salmon

MB

Abundance of

rockfish on shelf Introduction of nutrients

California

Growth and

production of krill

on shelf during

winter

Strength of NPH

Rockfish feed on

krill and krill eggs

Coastal

upwelling

in winter

Winter

Spring

Canada

Mexico

Pacific

Ocean

Pt Arena

2013

Freshwater

conditions

Freshwater

conditions

Predation

36.5

37.0

37.5

38.0

38.5

39.0

−124 −123 −122

x

y

Nutrients on shelf and shelf break

Primary

productivity on

shelf

Spring transition

to more

consistent, deeper

upwelling Condition and

production of krill

on shelf

Physics Primary

production

Forage

Poleward transport events from GoF Retention of krill in GoF

Spring

increase of

primary

production

Salmon emigrate to shelf

Cohort survival

Salmon condition

Pt Reyes

GoF

38N

37N

123W

Chinook salmon

MB

Abundance of

rockfish on shelf Introduction of nutrients

California

Growth and

production of krill

on shelf during

winter

Strength of NPH

Rockfish feed on

krill and krill eggs

Coastal

upwelling

in winter

Winter

Spring

Mexico

Pacific

Ocean

Pt Arena

2014

Freshwater

conditions

Freshwater

conditions

Predation

36.5

37.0

37.5

38.0

38.5

39.0

−124 −123 −122

x

y

Nutrients on shelf and shelf break

shelf

Spring transition

to more

consistent, deeper

upwelling Condition and

production of krill

on shelf

Physics Primary

production

Forage

Poleward transport events from GoF Retention of krill in GoF

Salmon emigrate to shelf

Cohort survival

Salmon condition

Pt Reyes

GoF

38N

37N

123W

Chinook salmon

MB

Abundance of

rockfish on shelf Introduction of nutrients

California

Growth and

production of krill

on shelf during

winter

Strength of NPH

Rockfish feed on

krill and krill eggs

Coastal

upwelling

in winter

Winter

Spring

Mexico

Pacific

Ocean

Pt Arena

2015

shallow deep

shallow deep

Spring

increase of

primary

production

Freshwater

conditions

Freshwater

conditions

Predation

Primary productivity

on shelf

36.5

37.0

37.5

38.0

38.5

39.0

−124 −123 −122

x

y

Nutrients on shelf and shelf break

Primary

productivity on

shelf

Spring transition

to more

consistent, deeper

upwelling Condition and

production of krill

on shelf

Physics Primary

production

Forage

Poleward transport events from GoF Retention of krill in GoF

Salmon emigrate to shelf

Cohort survival

Salmon condition

Pt Reyes

GoF

38N

37N

123W

Chinook salmon

MB

Abundance of

rockfish on shelf Introduction of nutrients

California

Growth and

production of krill

on shelf during

winter

Strength of NPH

Rockfish feed on

krill and krill eggs

Coastal

upwelling

in winter

Winter

Spring

Mexico

Pacific

Ocean

Pt Arena

2016

Primary

productivity on

shelf shallow deep

shallow deep

Freshwater

conditions

Freshwater

conditions

Predation

Spring

increase of

primary

production

A climate timeline for California’s

salmon

National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 34

2012 2013 2014 2015 2016 2017

Yr 1 CA

drought,

carryover

storage

Yr 2 CA

drought,

carryover

storage

Year 3 CA

drought,

record heat

West Coast

“snow

drought”

record high

temp

near average

precip and

snowpack but

an early melt

A very wet

year with an

abundant

snowpack

Cold

productive

NE Pacific

Cold

productive

NE Pacific

NEP in

transition

from good

to bad

Record warm

SSTs,

ecosystem

stress

A still warm

unproductive

NEP, but not

as extreme

An ocean in

transition to

“normal-ish”?

BY 2012

Chinook

Smolt

migration

Ocean year

2

Ocean year

3, most

return

BY 2013

Chinook

Smolt

migration

Ocean year 2 Ocean year

3, most return

BY2014

Chinook

Smolt

migration

Ocean Year 2 Ocean year 3

Most return

BY 2015

Chinook

Smolt

migration

Ocean Year 2

My view: NEP warming was caused by persistent

atmospheric forcing patterns that weakened winds and

the normal upper-ocean cooling processes

Sea Level Pressure anomalies

July-June 2014-15

Surface wind speed anomalies

July-June 2014-15

More questions … • What caused the incredible persistence of the

“resilient ridge(s)” in 2012-2016?

• Prime suspects: tropical SST anomaly patterns,

Arctic warming, feedbacks from Northeast Pacific

warm SSTs, butterflies (internal atmospheric

feedbacks)

• To what extent did anthropogenic forcing play a

role in the 2013-2016 NEP marine heat wave?

Figure from Mike Jacox, NOAA/SWFSC