Lecture 5: The El Nino Southern Oscillation (ENSO)

1
2. Natural Climate Variability    
2.1 Introduction 2.2 Interannual Variability 2.3
Decadal Variability 2.4 Climate Prediction 2.5
Variability of High Impact Weather
2
2.1 Introduction What is Interannual and
Decadal Variability?
Time series of SSTs in the East Pacific (region
Nino3.4)
3
2.1 Introduction What is Interannual and
Decadal Variability?
Time series of SSTs in the North Atlantic
highlighting the Atlantic Multidecadal
Oscillation (AMO) source Knight et al (2005).
4
2.1 Introduction What is Interannual and
Decadal Variability?
The NAO index is defined as the anomalous
difference between the polar low and the
subtropical high during the winter season
(December through March). The positive phase is
associated with more storminess in the Atlantic
storm track and the negative phase with reduced
storminess.
5
2.1 Introduction Why is Climate Variability
important?

• Interannual-to-Decadal Variability impacts
  Societies
• Food resources
• Water security
• Health
• Demographics
• Energy

Time series (1941-2001) of average normalized
April-October rainfall departure for 20 stations
in the West African Soudano-Sahel zone (11-18N
and West of 10E) following methodology of Lamb
and Peppler, 1992).
6
2.1 Introduction Can we predict SI Variability?
7
Section 2.2 Interannual Variability  2.2.1 El
Nino Southern Oscillation (ENSO)   (i)
Observations (ii) Theory for ENSO (iii)
Impacts 2.2.2 Interannual variability in
Atlantic SSTs 2.2.3 The North Atlantic
Oscillation (NAO)  
8
2.2.1 ENSO Observations

• Philander, S.G.H, 1990 El Nino, La Nina and the
  Southern Oscillation
• Useful El Nino pages
• http//www.pmel.noaa.gov/tao/elnino/nino-home.htm
  l
• http//www.gfdl.noaa.gov/atw/enso

9
Mean State of the tropical Pacific coupled
Ocean-Atmosphere System

• Warmest SSTs in west, cold tongue in east
• Precipitation associated with warmest SSTs
• Easterly trade winds advect equatorial surface
  waters westward

10
The tropical Pacific thermocline
11
The Walker Circulation

• Mean ascent, and low surface pressure, over
  warmest SST associated with deep convection
• Subsidence, and high surface pressure, in
  non-convecting regions
• Equatorial trades blow from high to low pressure
  (balanced by friction since Coriolis force gt0)

Low slp
High slp
12
El Nino

• During El Nino trade winds slacken
• E-W tilt of thermocline upwelling of cold water
  are reduced.
• SST rises in central/eastern equatorial Pacific
• Changes Walker Circulation

13
El Nino
El Niño 82/83
Mean climate
14
SST anomalies during El Nino
Dec 1982 Sept 1987
15
The 1997/98 El Nino
Jan 1997
16
The 1997/98 El Nino
Nov 1997
Jun 1997
17
The 1997/98 El Nino
Nov 1997
18
The 1997/98 El Nino
Mar 1998
19
The 1997/98 El Nino
Mar 1998
Jan 1997
20
The 1997/98 El Nino
Jun 1997
21
The 1997/98 El Nino
Nov 1997
22
The 1997/98 El Nino
Mar 1998
23
What about La Nina?
In La Nina conditions SST in the central and
eastern equatorial Pacific is unusually cold
easterly trade winds are unusually strong
Dec 1982
Nov 1988
La Nina conditions sometimes occur in the year
following an El Nino event (e.g. 1988 followed
1987 El Nino)
24
Walker Circulation

• British mathematician, director general of
  observations for India (formed after monsoon
  failure of 1877- worst famine in Indian history)
• Arrived in 1904, shortly after huge famine caused
  by drought
• Goal to predict Indian Monsoon
• Found that many global climate variations,
  including Monsoon rains in India, were correlated
  with the Southern Oscillation

25
The Southern Oscillation Index
Darwin
Tahiti

• Mean pressure is lower at Darwin than Tahiti
• The term Southern Oscillation was also coined
  by Gilbert Walker
• The SOI measures the strength of the Pacific
  Walker circulation

26
(No Transcript)
27
ENSO

• Bjerknes recognised that the El Nino warming of
  the ocean was related to variations in the SOI.
• During El Nino
• SOI is low
• Trades are weak
• Precipitation is enhanced over central
  equatorial Pacific (indicated by low OLR)

28
The Nino3 SST index
Note that El Nino events do not occur regularly
29
Why is El Nino important?

• Major climate and economic impacts on countries
  around the tropical Pacific, and further afield.
• Droughts in some regions, floods in others
• Collapse of coastal fishery in
  Peru (largest average annual catch of
  marine fish in the world)

30

• Impacts on global climate, ecosystems and society
  

31
(No Transcript)
32
(No Transcript)
33
(No Transcript)
34
Observing El Nino
35
(No Transcript)
36
(No Transcript)
37
(No Transcript)
38
(No Transcript)
39
TAO Data
TAO on Youtube https//www.youtube.com/watch?v
nzBAWirHMvA Live TAO data http//www.pmel.noaa
.gov/tao/jsdisplay/
40
Satellite Observations of Sea Surface Height
41
(No Transcript)
42

• Chronology of Events in the History of
  Understanding El Niño and La Niña
• late 1800s Fishermen coin the name El Niño to
  refer to the periodic warm waters that appear off
  the coasts of Peru and Ecuador around Christmas.
• 1928 Gilbert Walker describes the Southern
  Oscillation, the seesaw pattern of atmospheric
  pressure readings on the eastern and western
  sides of the Pacific Ocean.

43

• 1957 Large El Niño occurs and is tracked by
  scientists participating in the International
  Geophysical Year. Results reveal that El Niño
  affects not just the coasts of Peru and Ecuador
  but the entire Pacific Ocean.
• 1969 Jacob Bjerknes, of the University of
  California, Los Angeles, publishes a seminal
  paper that links the Southern Oscillation to El
  Niño.

44

• 1975 Klaus Wyrtki, of the University of Hawaii,
  tracks sea levels across the Pacific and
  establishes that an eastward flow of warm surface
  waters from the western Pacific causes sea
  surface temperatures to rise in the eastern
  Pacific.
• 1976 Researchers use an idealized computer model
  of the ocean to demonstrate that winds over the
  far western equatorial Pacific can cause sea
  surface temperature changes off Peru.

45

• 1982 A severe El Niño develops in an unexpected
  manner, but its evolution is recorded in detail
  with newly developed ocean buoys.
• 1985 Several nations launch the Tropical
  Ocean-Global Atmosphere (TOGA) program, a 10-year
  study of tropical oceans and the global
  atmosphere.
• 1986 Researchers design the first coupled model
  of ocean and atmosphere that accurately predicts
  an El Niño event in 1986.

46

• 1988 Researchers explain how the "memory" of the
  ocean--the lag between a change in the winds and
  the response of the ocean--influences
  terminations of El Niño and the onset of La Niña.
• 1996-1997 The array of instruments monitoring
  the Pacific, plus coupled ocean-atmosphere
  models, enable scientists to warn the public of
  an impending El Niño event.

47
Section 2.3 Interannual Variability  2.3.1 El
Nino Southern Oscillation (ENSO)   (i)
Observations (ii) Theory for ENSO (iii)
Impacts 2.3.2 Interannual variability in
Atlantic SSTs 2.3.3 The North Atlantic
Oscillation (NAO)  
48
Section 2.2.1 (ii) Theory of ENSO The
Bjerknes Feedback A mechanism for growth of El
Nino or La Nina  
49
Fig.1 Schematic of Ekman Spiral in the Ocean
50
Equatorial Upwelling
51
Coastal Upwelling

• Motion of surface waters away from coast requires
  upwelling of water from below to satisfy
  continuity of mass.

Andes Mts.
S. Pacific Ocean
52
Observed SST Distribution
Winds have amajor influenceon tropical
SSTpattern.
Equatorial Upwelling
Coastal Upwelling
53
La Niña conditionsStrong cold tongue
El Niño conditionsCold tongue absent
54
Fig.2 SSTs in Pacific in Dec
55
Upwelling linked to hurricanes
56
(No Transcript)
57
Fig.3 From S.-P. Xie (U Hawaii)
58
ENSO mechanism the Bjerknes feedback
Fig.4
59
Bjerknes feedback - equatorial sections
atmos
ocean
longitude
longitude
Fig.5
60
Bjerknes feedback - equatorial sections
El Niño
La Niña
atmos
ocean
longitude
longitude
Fig.5
61
Section 2.2.1 (ii) Theory of ENSO The
Delayed Oscillator  
62

• How does the phase of ENSO reverse?
• What triggers an El Nino event?

63
Power Spectrum

• What determines the 4-year periodicity?

64
Recent History of El Niño and La Niña
65
Temporal Evolutionof El Niño/La Niña
66
Equatorial Waves

• Equatorial waves in the ocean are believed to
  play an important role in the onset and
  variability of ENSO
• Two types
• Kelvin waves (propagate eastward along the
  equator and also along coasts)
• Rossby waves (long waves propagate westward)
• The relevant waves are baroclinic internal to
  the ocean, propagating along the density contrast
  of the thermocline
• Equatorial Kelvin waves travel 3 times faster
  than the fastest equatorial Rossby waves

67
How does the phase of ENSO reverse?
Delayed Oscillator Theory
Battisti and Hirst, 1989 Suarez and Schopf, 1988

• Westerly winds force downwelling on Equator and
  upwelling to North and South
• gt Excites Kelvin and Rossby waves

Thermocline depth
Figures from IRI http//iri.columbia.edu/climate/
ENSO/theory/
68
Delayed Oscillator Theory 2
Upwelling Rossby waves
Downwelling Kelvin wave
25 days
75 days
100 days
50 days
69
Delayed Oscillator Theory 3
125 days
225 days
275 days
175 days
Phase has reversed
70
(No Transcript)
71
Delayed Oscillator Theory 4

• Equatorial ocean waves offer a mechanism to
  reverse the phase of perturbations to the
  thermocline depth
• Without further wind forcing waves eventually
  decay
• Thermocline depth perturbations influence SST in
  the upwelling regions of central / eastern
  equatorial Pacific gt coupling to atmosphere
• Bjerknes feedback equatorial waves can generate
  a self-sustaining oscillation

72
(No Transcript)
73
Recharge/Discharge theory (Jin, 1997)
discharging
Source A. Timmermann

• Prior to El Nino heat content in equatorial
  region builds up
• During El Nino heat is discharged eastward and
  polewards

discharged
74
(No Transcript)
75
Equatorial temperature sections
76
What triggers El Nino?

• Still not fully understood
• High heat content (deep thermocline) in the
  equatorial region necessary but not sufficient
  condition
• Westerly wind bursts (few days duration)
  associated with the Madden Julian Oscillation may
  act as one trigger.

G. Vecchi
77
Onset of the 1997/98 El Nino
Downwelling equatorial Kelvin waves triggered by
westerly wind bursts
78
Summary of El Nino onset

• Need high heat content in equatorial Pacific
• Triggering by wind fluctuations (e.g. WWBs) over
  central/western Pacific
• Growth through Bjerknes positive feedback
  mechanism

79
Theories for ENSO Oscillations

• Delayed Oscillator (e.g. Battisti and Hirst,
  1989 Suarez and Schopf, 1988
• Recharge/discharge theory (Jin, 1997)
• Western Pacific Oscillator (e.g. Weisberg, R. H.,
  and C. Wang, 1997)
• Advective-Reflective Oscillator (e.g. Picaut et
  al, 1997)
• Unified Oscillator (Wang, 2001 J Clim)

80
The 1997/98 ENSO event

81

82
Multivariate ENSO Index (upto October 3rd
2014) http//www.esrl.noaa.gov/psd/people/klaus.w
olter/MEI/

Index is based on 6 parameters relevant to phase
of ENSO sea level pressure zonal and
meridional wind at surface sea surface
temperature air temperature cloudiness
83

84

• A weakening and reversal of trade winds in
  Western and central Pacific led to anomalous warm
  SSTs east of date line in early 1997

85

• A weakening and reversal of trade winds in
  Western and central Pacific led to anomalous warm
  SSTs east of date line in early 1997
• Warm waters (gt 29C migrated eastwards, collapse
  of cold tongue in east, which failed to develop.

86

• A weakening and reversal of trade winds in
  Western and central Pacific led to anomalous warm
  SSTs east of date line in early 1997
• Warm waters (gt 29C migrated eastwards, collapse
  of cold tongue in east, which failed to develop.

87

• El Nino developed so rapidly that each month
  between June-Dec 1997 a new monthly record high
  was set for SSTs in eastern equatorial Pacific.

88

• Development of El Nino in equatorial Pacific was
  significantly modulated by higher frequency
  variability.
• Weakening and reversal of trades was punctuated
  by a number of westerly wind bursts of increasing
  intensity along the equator (linked to several
  MJOs)
• Ocean currents forced by these WWBs, advected
  warm water eastwards.
• Also, WWBs forced downwelling K-waves
  (thermocline depth decreased by 90m in eastern
  Pacific by end of 97) opposite anomaly in
  western Pacific.

89

• Rossby waves excited by the initial weakening of
  the trades propagated westwards towards the west.
  
• Net result was a flattening of the thermocline
  and disapperance of normal east-west gradient
  across the equator, resulting in a weakening of
  the trades etc etc etc (Bjerknes Feedback).

90

• Note anomalous SSTs in early 1998, SST anomalies
  superimposed on seasonal warming.
• Cooling did not take place until winds returned
  close to normal (easterly) in the middle of 1998.
  Very abrupt change associated with upwelling due
  to normal easterly wind strength. Change to La
  Nina.

91
Observed heat content anomalies (from 0 to 400m)
averaged between 2N and 2S from the TAO array

• Note build up of heat content in Western Pacific
  at least a year before onset of the 97/98 El Nino
  due to stronger than normal trade winds (linked
  to weak La Nina).
• Although these conditions helped to set the
  stage for the development of El Nino the onset
  did not occur until the intensification of the
  MJO in late 1996.

92

• The delayed oscillator theory and the 1997/98 El
  Nino
• Evolution of equatorial Pacific on interannual
  timescales is governed by interplay between
  large-scale equatorial wave processes and
  atmosphere-ocean feedbacks.
• The only downwelling Kelvin waves seen in
  1997/98 were associated with a succession of
  MJOs.
• DO theory predicts demise of El Nino to be
  associated with reflected upwelling Rossby waves
  at the western boundary which was seen to occur
  after the onset of the El Nino.
• In addition there was direct wind-forcing of
  upwelling Kelvin waves associated with appearance
  of easterly wind anomalies in the West Pacific in
  late 1997 early 1998 so the demise was likely
  associated with upwelling Kelvin waves due to two
  different processes.

93
Section 2.2 Interannual Variability  2.2.1 El
Nino Southern Oscillation (ENSO)   (i)
Observations (ii) Theory for ENSO (iii)
Impacts 2.2.2 Interannual variability in
Atlantic SSTs 2.2.3 The North Atlantic
Oscillation (NAO)  
94
2.2.1 El Nino Southern Oscillation
(ENSO)   (iii) Impacts
95
Questions What are the impacts? What are the
causes of the impacts? How does ENSO variability
impact remote regions and why?
96
Different Ways of viewing Impacts

• Schematic Overviews
• Station Data Statistics
• Gridded Data Statistics
• Specific Events
• Important to consider these as a function of
  season, location and variable.
• Most common variables considered are T and PPN

97
Schematic Overviews
98
(No Transcript)
99
(No Transcript)
100
Statistics
IRI impacts website http//iridl.ldeo.columbia.e
du/maproom/ENSO/Impacts.html NOAA impacts
website http//www.pmel.noaa.gov/tao/elnino/impac
ts.html Composite ENSO plots http//www.esrl.n
oaa.gov/psd/enso/enso.composite.html Risk of
extremes http//www.esrl.noaa.gov/psd/enso/clima
terisks/
101
Events
http//www.esrl.noaa.gov/psd/enso/enso.different.h
tml
102
Mechanisms for Local Impacts
103
Mechanisms for Remote Impacts
Teleconnections
104
Mechanisms for remote impacts

• Latent heating associated with tropical
  precipitation anomalies excites waves in the
  atmosphere
• Atmospheric equatorial Kelvin and Rossby waves
  propagate zonally. Associated subsidence warms
  troposphere and suppresses precipitation
• Rossby waves can also propagate into extratropics
  (especially in winter) causing large-scale
  circulation anomalies which impact weather. (Note
  that propagation requires a westerly mean flow
  and the relevant Rossby waves have an eastward
  group velocity)

105
Source D. Neelin
106
Source D. Neelin
107
Current ENSO Advisory http//www.cpc.noaa.gov/p
roducts/analysis_monitoring/enso_advisory/