UCAR/NCAR/UOP > News Center > Fact Sheets > El Niño Search

El Niño  

Thumbnail sketch of ocean - atmosphere coupling during El Niño. The details of this interaction are described in this article, its accompanying figures, and animations portraying the evolution of sea surface temperatures during the El Niños of 1982-83 and 1997-98. Links to other El Niño related sites are also provided.

Children of the Tropics: El Niño and La Niña


Bob Henson & Kevin E. Trenberth

February 1998 • updated October 2001

For many years, coastal residents of Peru had noticed a strange feature of the eastern Pacific Ocean waters that border their home. This region of tropical yet relatively cool water is host to one of the world's most productive fisheries and a large bird population. In the first months of each year, a warm southward current usually modified the cool waters. But every few years, this warming started early (in December), was far stronger, and lasted as long as a year or two. Torrential rains fell on the arid land; as one early observer put it, "the desert becomes a garden." Warm waters flowing south brought water snakes, bananas, and coconuts from equatorial rain forests. However, the same current shut off the deeper, cooler waters that are crucial to sustaining the region's marine life.

This is El Niño, "the Christ child," so named because of its frequent late December appearance. Once thought to affect only a narrow strip of water off Peru, it is now recognized as a large-scale oceanic warming that affects most of the tropical Pacific. The meteorological effects related to El Niño and its counterpart, La Niña (a cooling of the eastern tropical Pacific), extend throughout the Pacific Rim to eastern Africa and beyond.

Among the many global phenomena connected to the El Niño-Southern Oscillation (ENSO) are Asian monsoons. The migration of the monsoon each year has been linked to the tropical Pacific rainfall patterns that are part of ENSO. The image highlights late monsoon season thunderstorms near Jamalpur, northern Bangladesh, 90.5°E, 25.0°N, 31 August 1985. From bottom right to top left one is looking southwest across Bangladesh to the Bay of Bengal and the northeastern coast of India. Evident in much of the photo is the silt laden Brahmaputra River and its branches in floodstage, flowing towards the Bay of Bengal. Jamalpur lies southwest of the Khasi Hills which form much of the state of Meghalaya in northeastern India. Geographical atlases typically show precipitation in the Khasi Hills to fall in the range of 3000-5000 mm (120-195 inches) annually, depending on elevation. However, in many localities of the Khasi Hills, annual rainfall amounts far exceed the upper limit of this range, noteworthy being Cherrapunji [elevation 1312 m (4306 ft)] on the Shillong plateau where the monsoon and strong orographic effects conspire to produce 11,420 mm (450 inches) of rain annually. Up to 9700 mm (382 inches) of this precipitation falls during the monsoon from late May to mid September, with monthly amounts in June and July often exceeding 2500 mm (98 inches). NASA image STS51F-31-069.
El Niño is normally accompanied by a change in atmospheric circulation called the Southern Oscillation. Together, the ENSO (El Niño-Southern Oscillation) phenomenon is one of the main sources of interannual variability in weather and climate around the world. Since recognizing some 25 years ago that the oceanic and atmospheric parts of ENSO are strongly linked, scientists have moved steadily toward a deeper understanding of ENSO. Climate forecasters have taken the first steps toward predicting the onset of El Niño and La Niña events months in advance. Still, much remains to be learned about these children of the tropics.

The Basics of ENSO

It was the atmospheric part of ENSO-the Southern Oscillation, or SO-that first attracted the attention of scientists. Sir Gilbert Walker documented and named the SO in the 1930s. Other persistent patterns of high and low pressure had been previously noted in the North Pacific and North Atlantic; thus, the "southern" in SO.

The clearest sign of the SO is the inverse relationship between surface air pressure at two sites: Darwin, Australia, and the South Pacific island of Tahiti. As seen in Figure 1, high pressure at one site is almost always concurrent with low pressure at the other, and vice versa. The pattern reverses every few years. It represents a standing wave or "see-saw", a mass of air oscillating back and forth across the International Date Line in the tropics and subtropics.

This two-dimensional picture was extended vertically by renowned meteorologist Jacob Bjerknes in 1969. He noted that trade winds across the tropical Pacific flow from east to west. To complete the loop, he theorized, air must rise above the western Pacific, flow back east at high altitudes, then descend over the eastern Pacific. Bjerknes called this the Walker circulation (in honor of Sir Gilbert); he also was the first to recognize that it was intimately connected to the oceanic changes of El Niño and La Niña.

The persistent easterly trade winds are a key ingredient in the ENSO process. They have two major effects:

As warm surface water collects in the western Pacific, it tends to push down the thermocline, the boundary separating well-mixed surface waters from deeper, colder waters. The thermocline is usually about 40 meters (130 feet) deep in the eastern Pacific but varies from 100 to 200 meters (330-660 feet) deep in the west. Figure 2 depicts a sequence of longitude-depth cross-sections of mean temperature in the equatorial Pacific Ocean for the months of December 1996, and April, August, and December of 1997. This period corresponds to the onset and intensification of the 1997-98 El Niño event. Clearly evident in Figure 2 is the normal pooling of very warm water and the depression of the thermocline in the western Pacific in December of 1996, and the eastward march of warmer than normal temperatures (positive temperature anomalies) from this region during the 1997-98 El Niño event. Notice the extensive cap of exceptionally warm surface waters in the eastern Pacific in December of 1997, the tell-tale signal of the arrival of El Niño in the Pacific coastal waters of equatorial South America. One can well imagine that the presence of this abnormally warm water profoundly influences and interferes with the normal upwelling of cold deep water and the delivery of life-sustaining nutrients to the base of the marine food chain.

An animation of sea surface temperature (SST) anomalies from March of 1992 through December of 1997 also illustrates the early stages of the 1997-98 El Niño in the Pacific Ocean, and completes the three-dimensional picture of the eastward migration of the equatorial warm pool during this event. (1997-98 El Niño SST animation, 1.7MBytes, MPEG format, courtesy of Tim Scheitlin, NCAR Scientific Computing Division Visualization Lab.)

The persistent oceanic heat surrounding Indonesia and other western-Pacific islands leads to frequent thunderstorms and some of the heaviest rainfall on Earth. The rainfall is abetted by the upward motion produced by the Walker circulation. The distribution of SSTs drives the enhanced rainfall, Walker circulation, and associated trade winds, which in turn are responsible for the ocean currents and the distribution of SSTs. The atmosphere drives the ocean and the ocean drives the atmosphere in a truly coupled mode of behavior. An example of this coupling can be found in Figure 3, which shows higher than normal rainfall centered on the equator and International Date Line during the strong El Niño of December-January-February of 1982-83. The enhanced rainfall is attributable to the presence of anomalously warm SSTs during El Niño, and in this example accounts for upwards of 4-6 mm per day (about 0.2 inch per day) more precipitation than what is typical. [A 1982-83 El Niño SST animation (1.2MBytes, MPEG format, courtesy of Tim Scheitlin, NCAR Scientific Computing Division Visualization Lab) is provided which portrays the state of Pacific Ocean SST anomalies from October of 1981 to August of 1985.] Figure 4 summarizes the salient features of the atmosphere-ocean interaction and coupling during normal and El Niño conditions.

Anatomy of El Niño

The sea-level pressure at Darwin can be used as an index of the SO and, by extension, as a guide to the major ENSO events of the past. Figure 5 shows the Darwin pressure anomalies (deviations above or below normal) of the past century, smoothed to eliminate short-term effects. Positive anomalies of Dawrin sea-level pressure correspond to El Niño events, negative anomalies to La Niña. Note that

Figure 5 and other evidence like it reveals that ENSO is a quasi-periodic yet highly variable phenomenon. Sometimes the warm waters generated by an El Niño flow all the way across the Pacific. The 1997-98 event increased surface water temperatures near Peru by 5°C (9°F). In the much weaker event of 1986-87, the warm water extended eastward only as far the mid-Pacific (near 170°W) and raised the temperatures there a modest 1°C (1.8°F) or so. In still other cases, warm anomalies first appear offshore of Peru and then progress westward to meet the preexisting warm pool.

Bjerknes was unable to determine why the SO reverses or why the ENSO changes from warm to cold conditions (El Niño to La Niña). This is still a subject of intense research. Although the atmosphere and ocean act in harmony after an ENSO event begins, some intriguing questions remain. What sets the system off? Is there really a self-sustained cycle in the atmosphere-ocean system? What is the role of other influences?

Recent work using computer models of ENSO hints that the storage of heat throughout the tropical ocean is a key element. Apparently, as rainfall and cloud cover are reduced during La Niña, the increased solar input heats up the ocean, especially in the deep western-Pacific warm pool. During El Niño, heat is transported from the tropics to higher latitudes by ocean currents, and additional heat goes to the atmosphere, mainly through evaporation. Global temperature averages can reflect this heat input, rising by as much as 0.3°C (0.5°F) in the months after a strong ENSO event. Thus, the tropical Pacific Ocean loses heat during El Niño and gains it during La Niña.

Could it be the length of time needed to "recharge" the ocean with heat that determines when ENSO events start and stop? Some model results point in this direction. However, conditions outside the tropics also seem to be important. Atmospheric changes over the South Pacific Ocean often precede SO changes by one to three seasons. Some studies have related the onset of El Niño to anomalous snowfall over Asia and to the southeast Asian monsoon. Ocean wave disturbances that travel across the tropical and subtropical Pacific may also play a role as they reflect off ocean-basin boundaries. Madden-Julian Oscillations, which occur in the atmosphere with periods of 40-50 days typically, contribute to westerly (from the west) wind bursts in the western tropical Pacific and may also play an important role. Even random weather "noise" may initiate events in the coupled system that start an El Niño or La Niña. Perhaps there are multiple ways that an ENSO event can be triggered. Even so, the system's inclination toward an ENSO event can be predicted with increasing success.

Predicting ENSO and its Effects

Even before we know when or how a particular El Niño or La Niña is going to evolve, we can say something about the regional and global effects it is likely to have. This is due to teleconnections: physical relationships that result from the dynamics of atmospheric and oceanic waves. The impact of teleconnections on weather patterns can be illustrated with statistics. Just as high pressure at Darwin tends to occur with low pressure at Tahiti, the presence of El Niño has been correlated with a number of wide-ranging atmospheric events. Figure 7 shows locations that have a consistent increase or decrease in precipitation during El Niño events, with the most common months of that occurrence indicated. Some of the best-established effects are enhanced rainfall over the central Pacific, Peru, Ecuador, and the southern United States and drought in INdonesia, Australia, southern Africa, and northeastern Brazil.

In other locations, the impact of El Niño can have two or more different "flavors." For instance, California can experience very wet conditions (such as in 1940-41, 1982-83, and 1991-92) or drought (1986-87 and 1987-88), depending on how far east the ENSO-related rainfall extends in the tropical Pacific. Predicting which flavor will dominate for a given event is difficult, because very small changes in SSTs can become magnified to produce large differences in rainfall patterns outside the tropics. Precipitation in California is clearly connected to ENSO, but it may vary greatly from one El Niño to the next.

In another approach to prediction, some researchers are using computer models to attempt to reproduce the physics of the ocean and atmosphere as they evolve during ENSO events. This became possible in the 1980s as computer power became sufficient to include ocean-atmosphere interactions in the large-scale climate models used to study such topics as greenhouse warming. Such models have been able to reproduce many of the oceanic and atmospheric effects of ENSO in the tropical Pacific, especially those that occur at the start of an El Niño event.

In 1986, a milestone was reached when the El Niño beginning late that year was successfully predicted months in advance by a computer model at Lamont-Doherty Earth Observatory of Columbia University. Using a state-of-the-art General Circulation Model (GCM) in 1992, NCAR was the first to demonstrate the evolution of ENSO-like behavior in a hypothetical atmosphere containing twice the carbon dioxide of the present (a state likely to be reached by the year 2060). The model indicates that the rainfall anomalies connected to El Niño and La Niña may become stronger in a global-warming scenario.

Forecast Frontiers

In the past, an inadequate understanding of the relevant physical processes and a lack of observational data covering vast areas of tropical oceans have been amongst the principal obstacles for scientists engaged in ENSO prediction. Significant improvement of the observational data base was brought about by the Tropical Ocean Atmosphere (TAO) array of 70 instrument buoys moored throughout the equatorial Pacific Ocean. Completed in 1994, and renamed the TAO/TRITON array in 2000 in recognition of the Japanese contribution of Triangle Trans-Ocean Buoy Network (TRITON) buoys in the western Pacific, the array gathers surface meteorological and oceanographic data and records ocean temperature to a depth of about 500 meters (1650 feet). Data collected by the TAO array (as displayed in Figure 2) played a central role in the early detection of the onset of the 1997-98 El Nio, marking a significant improvement over the detection of the 1982-83 event.

A vast array of ships, aircraft, and buoys collected oceanographic and atmospheric data throughout the western tropical Pacific as part of the Tropical Ocean and Global Atmosphere Program's Coupled Ocean - Atmosphere Response Experiment (TOGA-COARE) from November 1992 through February 1993. Shown is one of the 70 TAO instrument buoys being deployed as part of TOGA. (NOAA image.)
What are the remaining obstacles to predicting El Niño and La Niña? One is to better understand their cyclic yet variable beginnings and endings. The factors leading to the end of an El Niño event are not yet entirely clear, as evidenced by the failure of computer models to predict the end of the El Niño that began late in 1990. The Lamont-Doherty model called for the event to end in 1992. In late 1993, however, it was still in progress and actually intensifying, making it the longest-running El Niño in half a century. This El Niño finally ended in 1995. More recently, several of the major computer models correctly forecast the onset of the 1997-98 El Niño months in advance, although none of the models anticipated its intensity, and one model persisted in forecasting no El Niño even after the 1997-98 event had developed. The use of subsurface ocean data from the TAO array was a major factor in the successful forecast.

To further complicate matters, the baseline against which El Niño and La Niña events are measured may itself be changing. Indications are that the average temperature of the tropical Pacific has risen slightly in the past decade or so. If so, a new benchmark for measuring the onset and conclusion of ENSO events may have to be set.

These lines of research and data acquisition, and in particular the forecast of the 1997-98 El Niño well in advance, captured the public's eye as never before. In fact publicity surrounding El Niño has become so prevalent in the United States that many events of tenuous relation are nevertheless being attributed to El Niño. While many predictions and impacts of the forecasts are being realized, some (such as failure of the Australian wheat crop) are not. Continued verification of the forecasts and post analyses of this event will bring future benefits. It is clear that accurate information on El Niño's impact, and forecasts of its future evolution, will be of great benefit in planning for drought, flood, and temperature extremes and in mitigating the resultant loss of life and property. If we cannot hope to control the effects of ENSO, there is real hope that we can understand and forecast its life cycle.

Related Sites

The following are links to sites that offer web-based El Niño information. The URLs listed here are subject to change and beyond the control of the authors.

NOAA's El Niño Theme Page

IRI ENSO Monitor

Much of the original information in this report was drawn from "General Characteristics of El Niño-Southern Oscillation," by Kevin Trenberth, in Teleconnections Linking Worldwide Climate Anomalies, edited by Michael Glantz, Richard Katz, and Neville Nicholls (Cambridge University Press, 1991).

The National Center for Atmospheric Research and UCAR Office of Programs are operated by UCAR under the sponsorship of the National Science Foundation and other agencies. Opinions, findings, conclusions, or recommendations expressed in this publication do not necessarily reflect the views of any of UCAR's sponsors.