Slide #1.

El Niño – Southern Oscillation (ENSO) ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #2.

Outline • Definitions • History • Determining ENSO phase – Southern Oscillation Index (SOI) – Multivariate ENSO Index (MEI) • Physics of ENSO – Bjerknes theory – Delayed Oscillator Theory – Stochastic Theory • ENSO Teleconnections – Tropical Cyclone Activity – U.S. Climate • ENSO Predictability – Zebiak-Cane Model – Aspects of ENSO Forecasts ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #3.

Definitions • El Niño – An anomalous warming in eastern Pacific ocean temperatures • La Niña – An anomalous cooling in eastern and central Pacific ocean temperatures • Southern Oscillation – Pressure fluctuations in the tropics with centers of action in the western Pacific/eastern Indian Oceans and the southeastern Pacific ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #4.

http://scijinks.jpl.nasa.gov/en/educators/gallery/oceans/NinoNina_L.jpg ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #5.

Characteristics of El Niño • • • • Anomalous low (high) pressure in the eastern (western) Pacific Weak or even reversed trade winds across Pacific Dry (Wet) conditions in the west (east) Pacific Deep thermocline in the east – upwelling capped ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #6.

Characteristics of La Niña • • • • Anomalous low (high) pressure in the western (eastern) Pacific Stronger than normal trade winds across Pacific Dry (Wet) conditions in the east (west) Pacific Deep thermocline in the west – shallow in the east ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #7.

Southern Oscillation Note high correlations and dipole pattern from west to east Correlation between Southern Oscillation Index and Sea Level Pressure ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #8.

Brief History of ENSO • 1877 – Drought and famine in south Asia – Monsoons fail • 1891 – Term “El Niño” given to warm current off Peru that appears after Christmas. Luis Carranza writes about the current • 1897 – Noted that pressure oscillations between Sydney Australia and Buenos Aires Argentina out of phase • 1902 – Norman Lockyer confirms oscillation and determines period of 3.8 years ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #9.

Brief History of ENSO • 1923 – 1937 – Walker wants to predict Indian monsoon variability. Publishes series of papers that shows the monsoon is correlated with pressure oscillations in the Pacific – Coined phrase “Southern Oscillation” ATMS 373 Sir Gilbert Walker C.C. Hennon, UNC Asheville
More slides like this


Slide #10.

Brief History of ENSO • 1940’s – 1950’s – Interest in phenomenon decreases • Late 1950’s – Good SST data reveals correlation between Pacific SST changes and Southern Oscillation – Bjerknes (UCLA) says this didn’t happen by accident ATMS 373 Jacob Bjerknes C.C. Hennon, UNC Asheville
More slides like this


Slide #11.

Brief History of ENSO • 1969 – Bjerknes proposes that SST changes force displacement of the Walker Circulation ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #12.

Determining Phase of the ENSO • Southern Oscillation Index (SOI) – Tracks see-saw in pressure between eastern Pacific/Indian Ocean and central Pacific – Uses pressure observations from Tahiti and Darwin, Australia • Seasonal trends removed – Usually analyzed as a 3-6 month running mean ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #13.

SOI • One way of calculating SOI: Avg. Tahiti MSLP for month – Avg. Darwin MSLP for month  P SOI 10 * diff  Pdiffav  Long term avg. of Pdiff for that month (climatology) SD( Pdiff ) Long-term standard deviation of Pdiff for the month ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #14.

La Niña El Niño ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #15.

Determining Phase of the ENSO • SST-based methods use measurements of sea surface temperature in various averaging regions of the Pacific – Niño 3.4 – Trans-Niño Index (TNI) – Determined from anomalies between the Niño 1+2 and Niño 4 regions ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #16.

Averaging regions for SST ENSO Indices ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #17.

Determining Phase of the ENSO • Multivariate ENSO Indexs (MEI) – Hybrid indicator of ENSO. Accounts for both atmospheric and oceanic variables – Sea-level pressure – Zonal component of surface wind – Meridional component of surface wind – SST – Surface air temperature – Cloud fraction ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #18.

MEI See http://www.cdc.noaa.gov/people/klaus.wolter/MEI/mei.html for more information ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #19.

Physical Explanation of ENSO • Bjerknes – Circular relationship between ocean and atmosphere changes – Could not determine which came first • 1960’s – Oceanic Kelvin and Rossby waves identified as having key roles – Kelvin waves move eastward at 2-3 m/s – Rossby waves more westward at 0.6 – 0.8 m/s – Both carry energy and momentum gained from surface wind stresses ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #20.

Kelvin Waves • Recall that Kelvin waves are equatorially trapped by reversal of sign of Coriolis Force across the Equator • Created by wind anomalies in the tropical Pacific – Westerly wind bursts are intense westerly wind anomalies, sometimes associated with back side of MJO • Move across basin in a few months – Travel poleward at eastern boundary ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #21.

Kelvin Waves ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #22.

Bjerknes Hypothesis • Positive feedback loop • Explains the onset of a warm or cold ENSO event but does not explain the transition between the two ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #23.

Bjerknes Hypothesis • Modest change in trade wind strength or equatorial SST distribution caused ENSO phases • No counter-balances – E.g., winds weaken, warm water sloshes eastward, further weakening of the trades • Questions remain: – What causes transition to events of opposite sign? – Why do events last 12-18 months on average? ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #24.

Counter-balance: Delayed Oscillator Process 1. Westerly wind anomaly 2. Creates eastward moving Kelvin wave (orange) and off-equatorial Rossby waves (blue-green). Orange represents a wave that deepens the warm surface layer of the ocean (downwelling wave), blues make surface layer more shallow 3. Kelvin waves travel more quickly than Rossby waves. Takes Kelvin waves about 70 days to cross Pacific; Rossby waves take about 210 days Source: http://iri.columbia.edu/climate/ENSO/theory/index.html ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #25.

Delayed Oscillator Process Rossby wave reaches western boundary and reflects as an upwelling Kelvin wave Kelvin wave reaches eastern boundary and reflects to west as downwelling Rossby wave Now, warm anomalies transported west and cold anomalies transported east Idealized response from one forcing – real life forcings much more complex! ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #26.

Delayed Oscillator Theory • When combined with positive feedbacks between trade wind, SST, and convection, this theory is able to reproduce ENSO cycles of 3-5 years – Comparable to observations • Theory currently accepted ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #27.

Stochastic Theory • Coupled ocean-atmosphere system is actually stable – Not vulnerable to perturbations • ENSO events triggered by random forcings from the atmosphere • Attractive because it suggests that ENSO cycles should be irregular in both length and frequency – Matches observed behavior of ENSO ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #28.

ENSO Teleconnections • Tropical Cyclone Frequency – ENSO alters general circulation of atmosphere – Favored areas of tropical cyclogenesis shift – Reduced frequency during EN include: • Australia: Convection shifts east, monsoon trough weakens • Northwest Pacific (west of 160°E): Monsoon trough shifts away from area • Atlantic: Upper-level (200 mb) westerlies increase, increased vertical wind shear ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #29.

ENSO Teleconnections • Tropical Cyclone Frequency – Increased frequency during EN events: • NW Pacific (east of 160°E to Dateline): Monsoon trough shifts into this area • NE Pacific (140°W to Dateline, near Hawaii): Increased convection due to warmer SSTs – No detectable change • NE Pacific (east of 140°W) ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #30.

ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #31.

EN events: 1982-83, 1987, early 1990s, 1997, 2002 ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #32.

ENSO Teleconnections • United States weather – Changes from shifts in weather patterns (jet streams, storm tracks) forced by SST changes – Major US precipitation changes during EN events: • Southeast US experiences anomalous precipitation (esp. during winter) • California coast experiences high precipitation • Pacific NW and Midwest generally drier – Main factor is shift in sub-tropical jet that brings storms into southeast and southwest US – La Niña exhibits generally opposite patterns ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #33.

ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #34.

ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #35.

ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #36.

ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #37.

ENSO Teleconnections • United States weather – Major US EN temperature changes: • Warm in north, cool in south (more storm systems, less sun) • Temperature anomalies approximately 2°-4°F ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #38.

ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #39.

ENSO Predictability • Zebiak and Cane (1987) first successful model run that realistically simulated an ENSO cycle – Coupled air-sea model – Atmosphere and ocean can “talk” to each other – Ocean capable of reproducing thermocline depth anomalies • Critical – Oceanic Kelvin and Rossby waves advect thermocline changes across basin – Upper ocean can respond to surface wind anomalies • Creates Kelvin waves ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #40.

Zebiak-Cane Model results Area-averaged SST anomalies for the 90-year model simulation. The solid line is NINO3 (5°N-5°S, 90°-150°W), and the dotted line is NINO4 (5°N-5°S, 150°-160°E). [From Zebiak and Cane (1987), Copyright American Meteorological Society]. ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #41.

Highlights of Z-C Model • • • • Recurrence of EN events EN events irregular in time space and amplitude Favored period of 3-4 years Warm episodes tend to peak in June and last from 14-18 months • Warm SST anomalies peak at 2°-3°C at eastern boundary • ALL of these highlights match observations of ENSO events ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #42.

Observations of SST anomalies Recurrence of EN events EN events irregular in time space and amplitude Favored period of 3-4 years Warm episodes tend to peak in June and last from 14-18 months ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #43.

Conclusions from Zebiak-Cane • Necessary condition for onset of EN is above-normal equatorial heat content • All essential mechanisms for ENSO cycle contained in tropical Pacific – No external forcings (e.g. from mid-latitudes) • Random forcing of unknown origin not required – Suggests that ENSO is predictable, even with 1-2 years lead time ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #44.

Aspects of ENSO Forecasting • ENSO forecasts are more difficult from January through April – Called the “spring barrier” • Easier to forecast events from April through June – ENSO cycle usually already established and usually persists until the following spring ATMS 373 C.C. Hennon, UNC Asheville
More slides like this


Slide #45.

Current ENSO Forecasts http://iri.columbia.edu/climate/ENSO/currentinfo/SST_table.html ATMS 373 C.C. Hennon, UNC Asheville
More slides like this