SCIENTIFIC BRIEF
Executive Summary of ENSO
Submitted to the 110th United States Senate
Authors: Luis Poza and Josh Gellers
Columbia University
December 3, 2006
- Background
ENSO refers to the climatic phenomenon known as El Niño and the Southern Oscillation. El Niño has been observed since at least 1567 [“What is El Niño?” 2002] and was coined by Peruvian fisherman who noticed a warming in the temperatures of their coastal fishing waters during Christmas [Cane 2005]. Paleoclimate records indicate that ENSO has impacted the Earth’s climate for more than 130,000 years [Cane 2005]. Sir Gilbert Walker is credited with statistically identifying atmospheric teleconnections observed in the Pacific during the 1920s and paved the way for further analysis of the ENSO mechanism. Today, scientists have come to recognize El Niño as a global pattern of responses to a fairly localized coupled phenomenon occurring in the tropical Pacific. ENSO can also be perceived as a system of coupled atmospheric instability(i.e. the system is subject to continuously reinforcing changes under certain conditions, disabling a return to a state of equilibrium) that occurs approximately every 2 to 10 years [Kump et al 2004].
This ENSO cycle is marked by two distinct phases that vary in strength- El Niño and La Niña. These warm and cold ENSO events, respectively, feature strong negative correlations with regards to physical characteristics in various parts of the world (though not necessarily experienced as exactly opposite climatic effects). ENSO itself is indexed by the Southern Oscillation Index, a “see-sawing of atmospheric mass” as measured by sea-level pressure difference between Darwin, Australia and Tahiti [Cane 2005]. The value assigned to the SOI determines whether we are experiencing an El Niño (negative SOI) or La Niña (positive SOI). Another common index of ENSO involves the sea-surface temperatures (SSTs) over an area of the equatorial Pacific known as NINO3.4 (5S-5N; 170W-120W). Even a slight warming (0.5 °C) of NINO3.4 over a sustained period of time can be indicative of a warm ENSO event [“NOAA Announces…” 2004].
II. ENSO Dynamics
The dynamics of the ENSO mechanism can best be explained through the groundbreaking research conducted by Jacob Bjerknes in the 1960s. Bjerknes explained how the oscillation that occurs in the equatorial Pacific is due to a chain of events that comprise a coupled positive feedback. Looking to ENSO’s most salient region of impact, the chain of events that occur in the atmosphere over the tropical Pacific are governed by the Walker Cell circulation. In the Walker Cell, high a low pressure center over warm waters in the west causes the surrounding air to warm and rise to the cooler troposphere where it diverges and finally sinks over an area of low pressure in the east.according to the Ideal Gas Law.[MC1] Easterlies (winds originating from the east) travel westward across the Pacific due to the Earth’s rotation[MC2], pushing the warm surface layer of the ocean across the equator and also driving a poleward component of the surface flow (the Ekman transport) allowing cooler water to rise in its place. However, the strength of the winds is largely determined by the temperature gradient. This temperature gradient is what causes pressure to be high in the east and low in the west. Bjerknes noted how, under normal conditions, the sun radiates the tropical Pacific equally but the east is peculiarly cooler than the west (Fig. 1). This is because equatorial upwelling reveals colder waters infused by the cool water coming from the South Pacific via the Humboldt Current. The tropical Pacific features a thermocline, or site where ocean “temperature drops rapidly with increasing depth” [Kump et al 2004], that slopes downward from east to west. This means that the colder waters in the east are more easily accessible as opposed to the placement of the pile of warm water normally prevalent in the west. As Cane [2005] summarizes, “colder temperatures in the east drive stronger easterlies which in turn drive greater upwelling, pull the thermocline up more strongly, and transport cold waters faster, making the temperatures colder still.”
Figure 1. This diagram depicts the tropical Pacific under normal (i.e. neutral) conditions. The system governing atmospheric activity is the Walker Cell. Also of note are the easterlies dragging the surface layer of the ocean across the Pacific from east to west and the dramatic slope of the thermocline. (Image obtained from NOAA)
During an El Niño, the Walker Cell is disrupted,effectively weakening or even reversing the easterlies. Without winds dragging across the surface of the equatorial Pacific, the thermocline deepens in the east and shoals in the west lessening the upwelling effect (Fig. 2). As more warm waters pile in the center of the tropical Pacific due to the weakened easterlies, the region becomes a hotbed for convective activity. The pressure over this area becomes higher lower[MC3] and the air is uplifted. Once the warm air reaches the cooler troposphere, it diverges to both the east and west. The cooler rising air condenses and precipitation occurs. at sites on either side of the Pacific where the pressure is low.[MC4] This has the effect of increasing rainfall in Peru due to increased convection. The air once again converges in the center of the tropical Pacific and the cycle continues. It should also be noted that the atmospheric conditions in this region have implications for global atmospheric circulation and various regional impacts.
Figure 2. This diagram depicts the change in the ocean-atmosphere system that occurs during an El Niño. The thermocline has sunk in the east and risen in the west, the easterlies have weakened, giving rise to a reversal of surface flow now leading toward the east, and the Walker Cell has been thoroughly disrupted, resulting in a hotbed of convective activity over the central Pacific and thus more precipitation in parts of western South America. (Image obtained from NOAA)
III. Regional Impacts of ENSO
The impacts of ENSO are far-reaching. As can be expected, dramatic differences in SST, SLP, and precipitation can be observed at the western and eastern ends of the equatorial Pacific, where the anomalous instability itself occurs. ENSO notably influences the probability of precipitation anomalies across the globe, with effects both positive and negative. Beyond the possible impacts, the very probabilistic nature of said impacts in itself presents headaches as farmers, fishermen, climatologists, and policymakers all must cope with uncertainty and variability.
While effects vary by region, it should first be noted that they are widespread. Figure 3 shows the areas of impact during ENSO positive (El Niño) and ENSO negative (La Niña) stages.
Figure 3.
Above portion (a) shows areas with increased probability of positive rainfall anomaly in El Niño years in blue. Areas in orange have higher probability of negative rainfall anomaly. Bottom section (b) shows probable positive and negative rainfall anomaly during La Niña years.
[Holmgren et al 2001]
Understanding this ample range of ENSO impacts elucidates its particular effects in distinct regions.
Well known among ENSO’s consequences is the collapse of fisheries in the eastern Pacific during the El Niño phase. Warmer waters create inhospitable conditions for plankton. The subsequent population decrease in turn reduces stocks of fish like anchoveta, which feeds on plankton and which accounts for much of Peru’s fishing industry [Holmgren 2001, Jaksic 2001]. Besides this economic hardship, the regions of Peru, Ecuador, and northern Chile also face higher chances of notably greater volumes of precipitation. This anomaly has been linked to destructive flooding and landslides [Holmgren et al 2001] as well as to eruption of rodent populations given an increase in desert blooming and perennial herb growth [Jaksic 2001].
Also affected in South America is northeastern Brazil, primarily the semiarid state of Ceará. Statistically significant connections exist between rainfall in Ceará and Pacific SST, noting that in ENSO positive years below normal rainfall is more likely than usual [Battitsi and Sarachik 1995]. The damages in crop yield and population migration resulting from drought is such that the state of Ceará felt compelled to create an agency that provides monthly bulletins with climate conditions, ENSO forecasts, and local precipitation data as well as land management advice for dealing with these impacts [Battisti and Sarachik 1995].
Equally to be expected from ENSO is higher probability of anomalous circumstances in the western Pacific. In northeastern Australia, chances increase for below normal rainfall conditions to arise midway through an El Niño year and persist for as much as 10 months thereafter [Chiew 1998]. Correspondingly, stream flow levels are likely relatively lower during El Niño years while drought conditions (as determined by the Palmer Drought Severity Index) more likely to prevail [Chiew 1998]. In fact, looking back on 79 years of rainfall data, Chiew [1998] finds five of the six lowest rainfall indexes occur during ENSO positive years.
Figure 4.
Chiew (1998) finds statistically significant correlation between rainfall and SOI
Remaining in the scope of the western equatorial Pacific, eastern Asia similarly is more likely to experience anomalous climate as a result of ENSO. Probability increases that the winter monsoon along the east Asian coast is weaker than normal in ENSO positive years [Wang 2000]. Moreover, climate in southeastern China and Korea tends to be warmer and wetter during ENSO winter and following spring seasons [Wang 2000].
Portions of eastern Africa also experience climate variability in connection with ENSO. While El Niño years customarily bring above normal rainfall to coastal areas of Kenya and Tanzania, the northern areas of Kenya and Uganda, the basin of Lake Victoria, and the northern areas of Tanzania experience increased likelihood of below normal precipitation during the “long rains” occurring in the months from March to May, which are crucial to agricultural production [Indeje et al 2000]. Aside from, or perhaps compounding, lower precipitation volumes, these affected areas in eastern Africa also are more likely to encounter a late onset to the long rains and an early cessation [Indeje et al 2000]. In fact, about 50% of East African seasonal rainfall variance can be attributed to ENSO (albeit indirectly, as this climatological anomaly is also linked to Indian Ocean SST, itself impacted by ENSO [Goddard and Graham 1999]) [Indeje et al 2000], when one takes into account the fact that in La Niña years the “short rains” of Northern Hemisphere winter are more probably abbreviated [Indeje et al 2000].
Finally, climate variability in North America also connects to ENSO. Warming and increased precipitation along the west coast of the United States, which are more likely in El Niño years, leads to flooding, reduced fishery stocks, and significant alterations to terrestrial ecosystems. In the arid islands of the Gulf of California along Mexico’s northwest, the increase in precipitation can be such that plant cover increases from 0-4% during ENSO neutral years to 54-89% during ENSO positive years [Holmgren et al 2001]. Further, hurricane activity in the Atlantic, a usual burden for the eastern United States, is likely to be suppressed in El Niño years given the change in wind shear [Cane 2006].
In brief, the impacts of ENSO extend far beyond the equatorial Pacific basins. Fortunately, the ability to accurately forecast El Niño and La Niña events well in advance has helped to temper impacts on crop loss and property damage as those most affected receive aid in understanding probabilistic forecasts and adapt to likely anomalies [Battisti and Sarachik 1995]. It bears noting, however, that there exists some contention as to the actual significance of ENSO in these climatic events. Goddard and Dilley [2005], in a study tracking climate related disasters found only weak ENSO effects on precipitation perturbations over land[MC5], and found no increased frequency of climate-related disasters during El Niño or La Niña years. Similarly, while positive ENSO conditions correlated strongly to weakening of India’s summer monsoon (less rainfall and shorter duration), that pairing has had broken down in recent decades [Kumar et al 1999], only to return strongly in 2002. A final complication occurs in that, for some regions, impacts are seasonal rather than annual (such as in eastern Africa where the “short rains” of October through December are more likely to increase in ENSO years, despite earlier drought conditions [Indeje et al 2000, Goddard 2006] ). For more detailed information, maps, and relevant graphs of probable impacts, readers can visit the ENSO webpage hosted by the International Research Institute for Climate and Society at
IV. ENSO in the Future
The challenge in modeling changes in the characteristics of ENSO in future conditions with warmer average global temperature and higher greenhouse gas concentrations arises in that few models agree in replicating ENSO conditions of the past [Cane 2005]. While some models find the amplitude and strength of El Niño events in particular to be increasing, others find no such trend and in fact suggest the strongest single event was in 1877 [MC6][Cane 2005]. Models that are “flux corrected”, that is, the models’ time-average fluxes are replaced by observed average fluxes added to the models’ anomalous fluxes [Goddard 2006], show a stronger ENSO at higher frequency under increased greenhouse gas concentrations, but the use of such models is largely avoided because of uncertainties to the models’ reflection of variability and sensitivity to greenhouse gases [Cane 2005][MC7]. More recent studies undertaken by Doherty and Hulme [2002, in Cane 2005] found some support (6 out of 12 CGCM’s, while four suggested no change and two suggested the opposite) for stronger La Niña-like conditions on average. While this finding corresponds to the Bjerknes feedback that drives ENSO, there is also the chance that increased warming will supercede the feedback and actually maintain warmer temperatures in the eastern equatorial Pacific simply because of increased global temperature raising SST [Cane 2005].[MC8]
- Key Points
1)El Niño and the Southern Oscillation (ENSO) has been an instrumental factor in shaping the Earth’s climate for at least the past 130,000 years. The El Niño event has been recorded as far back as 1567 and the ENSO teleconnections were first identified in the 1920s by Sir Gilbert Walker.
2)The ENSO cycle relates a chain reaction of climatic events that take place within the tropical Pacific ocean-atmosphere system, reinforcing itself both during cold events (La Niña) and degenerating during warm events (El Niño). This mechanism operates as a coupled positive feedback (posited by Bjerknes) often measured by divergences in sea-level pressure between Darwin, Australia and Tahiti in the form of the Southern Oscillation Index (SOI) or NINO 3.4 SST Index.
3)ENSO impacts vary greatly across regions, depending on the phase of the cycle and time of year. Some of the most dramatic effects can be seen during the warm phase; anomalously high precipitation in Peru and below-normal rainfall in northeastern Brazil, drought in northeastern Australia, abnormally high rainfall in coastal Kenya and Tanzania, increased precipitation in the western U.S. and reduced hurricane activity in the Atlantic. These initial effects may incite secondary impacts such as flooding, fires, crop failure, and the spread of disease.
4)The future of ENSO is opaque. Current models do not concur with a high degree of certainty that global warming has had a strengthening effect on ENSO, yet some models do purport that ENSO has grown more powerful in recent years.
Recommendation: It behooves the newly minted 110th United States Senate here assembled to support measures that will enable researchers to better understand ENSO and its future impacts, as its consequences are severe and more refined forecasting techniques can assist in implementing preventative tactics. [MC9]
References:
Battisti, D.S., and E.S. Sarachik, 1995; “Understanding and Predicting ENSO”; For the US National Report to IUGG: Contributions in Oceanography (1991-1994), Revs. Geophys. (Supp.), 1367-1376
Cane, M., 2005. “The evolution of ENSO, past and future.” Earth & Planet. Sci. Lett., 230, 227-240.
Cane, M; October 5, 2006; “Mean State of the Tropics Air-Sea Interaction”; lecture for EESC W4400: “Dynamics of Climate Variability and Climate Change”; Columbia University Graduate School of Arts and Sciences
Chiew, F.H.S., Piechota, T.C, Dracup, J.A, and McMahon, T.A; 1998; “El Niño/Southern Oscillation and Australian rainfall, streamflow and drought: Links and potential for forecasting”; Journal of Hydrology 204; p. 138-149; Elsevier Ltd; UK
Goddard, L; 2006; comments via personal emails to authors
Goddard, L. and Dilley, M; 2005. “El Niño: Catastrophe or opportunity”. Journal of Climate; 18(5); 651-665; American Meteorological Society, Boston, MA.
Goddard, L. and Graham, N; 1999; “Importance of the Indian Ocean for simulating rainfall differences over eastern and southern Africa”; Journal of Geophysical Research 104(D16); p. 19,099 – 19,116
Holmgren, M, Scheffer, M, Ezcurra, E, Gutiérrez, J, and Mohren, G; February 2001; “El Niño Effects on the Dynamics of Terrestrial Ecosystems”; Trends in Ecology & Evolution 16(2); p. 89-94; Elsevier Ltd; UK
Indeje, M, Semazzi, F, and Ogallo, F; Jan 2000; “ENSO Signals in East African Rainfall Seasons”; International Journal of Climatology 20(1); p. 19-46; Royal Meteorological Society, UK
Jaksic, F; 2001; “Ecological Effects of El Niño in Terrestrial Ecosystems of South America”; Ecography 24(3); p. 241-250; Blackwell Synergy; Copenhagen
Kumar, K, Rajagopalan, B, and Cane, M; June 1999; “On the Weakening Relationship between the Indian Monsoon and ENSO”; Science 284(2); p. 2156-2159; American Association for the Advancement of Science; Washington DC
Kump, Lee R., James F. Kastin, and Robert G. Crane. The Earth System. 2nd ed. Upper Saddle River: Pearson Education, Inc., 2004. 94, 308.
“NOAA Announces the Return of El Niño.” NOAA. 10 Sept. 2004. U.S. Department of Commerce. 4 Dec. 2006 <
Wang, B, Wu, R, Fu, X; May 2000; “Pacific East Asian Teleconnection: How Does ENSO Affect East Asian Climate?”; Journal of Climate 13(9); p. 1517-1536;
American Meteorological Society, Boston MA
“What is El Niño?”Environment Canada. 21 Aug. 2002. Government of Canada. 4 Dec. 2006 <
I, Luis Poza, assert that I am solely responsible for the content within and research behind sections III and IV of this report and the accompanying presentation, and that all sources consulted have been properly cited.