El Niño

Bernard N. Meisner, Ph.D., CCM
Techniques Development Meteorologist
Scientific Services Division
Southern Region Headquarters
National Weather Service
Fort Worth, Texas

What is El Niño?
Interactions between the atmosphere and the ocean in the tropical Pacific, as well as the physical shape and thermal structure of the tropical Pacific Ocean, result in alternations between two climate extremes. Each extreme typically lasts from one to two years, with two to seven years between events. The two extremes are frequently called El Niño and La Nina, while Southern Oscillation is the name given to the associated atmospheric fluctuations.

The terminology concerning El Niño and the Southern Oscillation has not been formally defined and may vary from author to author (Aceituno, 1992). For example, El Niño (The Male Child) originally referred to a warm current that develops almost every year along the coast of Ecuador and Peru at Christmas time. El Niño is now commonly used to refer to persistent, large-scale warm events in the central Pacific. Corresponding cold events have been called anti-El Niño, counter-El Niño, El Viejo (the grandfather) and La Nina (the female child).

The term Southern Oscillation was originally defined as an oscillation in the atmospheric pressure field between the Pacific and Indian Oceans, with related changes in temperature and precipitation in those regions. The term is now sometimes used to refer to the self-sustained oscillation of the combined ocean-atmosphere system in the central Pacific. The acronym ENSO (El Niño/Southern Oscillation) was first used to differentiate the large-scale warm events from those confined to coastal South America. ENSO is now frequently used to identify the coupled ocean-atmosphere system in the tropical Pacific. In this paper El Niño will refer to the persistent, large-scale warm event and La Nina will refer to the corresponding cold event. Southern Oscillation will refer to the large-scale variations in atmospheric pressure between the eastern and western Pacific, and the resulting changes in atmospheric circulation and precipitation.

ENSO has been called the single most important influence on extreme climate events in many regions of the global tropics. The four most important parameters that contribute to the El Niño/Southern Oscillation phenomenon are the sea surface temperature, the thermal structure of the ocean, the atmospheric winds, and the tropical rainfall.

What are the normal conditions in the tropical Pacific?
The tropical Pacific Ocean can be considered as comprised of three layers. A shallow, warm, well-mixed layer overlies a deep cold, stratified layer. The transition zone between these layers is called the thermocline. Within the thermocline the water temperature decreases very rapidly with increasing depth. The warm layer is about 200 meters (656 ft) deep in the western Pacific, but it is typically only 50 meters (164 ft) deep, or less, in the eastern Pacific (Figure 1).

Surface air pressure over the eastern South Pacific is normally greater than that over the western South Pacific. As a result, air near the equator flows from east to the west. This air flow is the trade winds. The frictional drag of the westward flowing air on the ocean's surface results in the water also being driven westward across the Pacific, where it accumulates. The slope of the ocean surface increases towards the west until it balances the frictional drag of the winds. You may note that whenever you blow on a cup of coffee the liquid flows towards, and accumulates at, the far side of the cup until a balance is achieved. The same is true, on a much larger scale, in the tropical Pacific Ocean.

Under the influence of the trade winds, sea level in the western Pacific is normally about 50 cm (20 in) higher than in the eastern Pacific. As warm water accumulates in the western Pacific it depresses the thermocline. Since warm water is less dense than cold water, a deeper layer of warm water is required to produce the same hydrostatic pressure as a layer of cold water. In the eastern Pacific the westward transport of the warm surface waters by the tradewinds causes the thermocline to rise towards the surface.

The surface waters of the western Pacific are very warm, with temperatures greater than 28oC (82.5oF) common (Figure 2). The surface waters of the eastern Pacific, in contrast, are among the coldest found near the equator, and can be as cool as 20oC (68oF). This is due to the upwelling of cold sub-surface water near the South American coast. This water is brought to the surface by the combination of the persistent southerly winds along the coast and the Coriolis Effect that results from the earth's rotation. As the wind pushes the water towards the equator, the Coriolis Effect results in the water having a component of its flow to the west. As the surface water moves away from the coast, sub-surface water rises to replace it. This cold, upwelled water is very rich in nutrients and can have a high oxygen content. Marine life is very abundant in such cold water. As the surface water flows westward from the South American coast, additional upwelling of sub-surface waters along the equator maintains a tongue of cold water across the equatorial Pacific Ocean (Figure 2).

Over the eastern Pacific, the northeast and southeast trade winds flow around and out of the subtropical anticyclones at low levels and into a convergence zone just north of the equator (Figure 3). A second convergence zone extends across the South Pacific from northwest to southeast. Within each zone the converging surface air rises, producing clouds and abundant rainfall, and then flows back toward the poles at high altitudes. As the trade winds are the major circulation feature over the eastern Pacific, a great monsoon circulation dominates the western Pacific. In the monsoon, the low level air flows across the equator from the winter to the summer hemisphere. There it rises, produces clouds and abundant rainfalls, and then flows back into the winter hemisphere at high levels.

Why is it called El Niño?
Every year around Christmas time as the winds subside, the temperature of the water along the coast of Ecuador and Peru increases. This warming is part of the natural annual variation in the oceanic circulation, although the degree and extent of the warming varies from year to year. The name El Niño, Spanish for The Male Child, was given to this warming by the Peruvian fishermen as a reference to the Christ Child. As the water temperature increased the fish departed for colder waters and the fishermen used this time to repair their boats and nets.

How is this El Niño related to the major weather disruptions that are often reported in the media?
About once every three to seven years, the water along the western coast of South America remains warmer than normal for a year or longer. The warmer water extends along the equator from the coast of South America to the international date line. Significant changes in the circulation of the atmosphere are associated with these changes in sea surface temperature. Unusual weather is often experienced throughout the Tropics, such as droughts in Indonesia and Hawaii, floods in Ecuador, and hurricanes in Tahiti. Large weather anomalies may occur beyond the Tropics as well, especially in the winter hemisphere. These extra-tropical weather anomalies were first noted in statistical studies of surface temperature and precipitation and given the name "teleconnections."

These major warm oceanic events were first called catastrophic El Niño events. Over time, the adjective catastrophic was dropped, and these warm events were just called El Niños (or more properly, Los Niños). Scientists now often use the term warm event, or warm phase of the Southern Oscillation, for these catastrophic events. Table 1 gives the years of recent El Niño events.

What is the Southern Oscillation?
Southern Oscillation refers to a periodic see-saw in the air pressures over the eastern and western South Pacific Ocean. Air pressure is a measure of the mass of air over a given area. The greater the mass of air over the area, the higher the air pressure. In the early part of this century, Sir Gilbert Walker was appointed Director-General of Observatories in India. He investigated ways to predict the timing and intensity of the Indian Monsoon. During this work, he found that the air pressures over much of the Southern Hemisphere and the tropical Northern Hemisphere were highly correlated over long periods (Figure 4).

Whenever pressures are higher than normal over the western South Pacific, they are lower than normal over the eastern South Pacific, and vice versa. Walker described the Southern Oscillation as follows: "When pressure is high in the Pacific Ocean it tends to be low in the Indian Ocean from Africa to Australia; these conditions are associated with low temperatures in both these areas, and rainfall varies in the opposite direction to pressure. Conditions are related differently in winter and summer, and it is therefore necessary to examine separately the seasons of December to February and June to August." (Walker and Bliss, 1932). Walker also discovered two Northern Oscillations--one in the Atlantic and another in the Pacific--but neither are as large nor as globally significant as the Southern Oscillation.

Scientists often monitor the Southern Oscillation by comparing the air pressures at Darwin, Australia with those at Papeete, Tahiti. The Southern Oscillation Index is based on the difference between the pressures at these two locations (Figure 5). The index is positive when the pressure is above normal in the eastern Pacific and below normal in the western Pacific. Persistent high values of the index are associated with La Nina events. The index becomes negative and remains negative during El Niño events.

How is El Niño related to the Southern Oscillation?
Strictly speaking, El Niño is an oceanic phenomenon, while the Southern Oscillation is an atmospheric phenomenon. During the warm phase of the ENSO cycle, air pressure falls over the eastern Pacific and rises over the western Pacific. As the pressure difference from east to west decreases, the easterly trade winds that blow across the equatorial Pacific become weaker. In some warm years the flow is reversed and the winds blow from the west. As the winds weaken, the warm water that had accumulated in the western Pacific flows back to the east. At the same time subsurface perturbations, called Kelvin waves, move eastward in a narrow belt centered on the equator to the coast of South America where they depress the thermocline. These waves take about three months to cross the Pacific from west to east. The result is an increase in the water temperatures in the central and eastern Pacific, and a warm event (catastrophic El Niño) occurs.

During the cold phase of the ENSO cycle, air pressure rises over the eastern Pacific and falls over the western Pacific. The easterly trade winds become stronger than normal. During some cold years the easterlies may extend all the way across the Pacific to Asia. The warm water confined to the western Pacific and cold, subsurface water is upwelled along the equator and the South American coast The result is a decrease in the water temperatures in the central and eastern Pacific, and a cold event (La Nina) occurs.

What are the effects of El Niño and La Nina?
El Niño, and its opposite, La Nina, produce major disruptions in the world's weather. The extreme El Niño of 1982-83 resulted in over 1,000 deaths, and nearly nine billion dollars in damage. Most of the effects are felt within the Tropics, but some higher latitude regions are often regularly affected as well.

Sir Gilbert Walker in India, H. P. Berlage in the Netherlands, and Jacob Bjerknes of UCLA were among the first to note that El Niño events perturbed the atmosphere at points far removed from the equatorial Pacific. Chester Ropelewski and Michael Halpert of the National Weather Service's Climate Analysis Center near Washington, DC have documented the consistent relationships among El Niño, La Nina and regional temperature and precipitation anomalies around the globe. Figures 6 and 7 summarize their results.

In the South Pacific, tropical cyclones form far to the east of their normal spawning grounds during El Niño events. Six tropical storms struck French Polynesia during the summer of 1982-83, an island chain where such storms are rarely seen in normal years. In some cases, twin storms form over the warm water in the central Pacific, one on either side of the equator.

El Niño also seems to inhibit the formation of hurricanes over the Atlantic Ocean. The condensational heating of the atmosphere that occurs in the rainstorms over the eastern equatorial Pacific during El Niño events increases the strength of the upper tropospheric westerly winds over the tropical Atlantic. The increased wind shear between the upper tropospheric westerlies and the lower tropospheric easterly trade winds inhibits the formation of tropical cyclones (Figure 8). William Gray of Colorado State University, includes El Niño as a parameter in his equations for predicting the number and intensity of Atlantic hurricanes. Years with strong El Niño events in progress average four storms fewer than normal; years with moderate El Niño events tend to have two fewer storms (Figure 9). Neville Nicholls of the Bureau of Meteorology Research Centre in Melbourne, Australia has found there are also fewer tropical cyclones near Australia during El Niño years.

El Niño events also have significant biological consequences. Figure 10 shows the record of the bird population and anchovy catches off the Peruvian coast. During El Niño the water along the coast of South America is warmer, contains fewer nutrients and less dissolved oxygen. Reproduction rates of some species of marine life drop to almost zero, and many adult fish and birds die as the food supply is diminished. Note the significant decrease in the number of birds during both the 1957 and 1965 El Niño events. As the Peruvian fishing industry grew during the 1960's the bird population never returned to what it had once been. Instead it remained nearly constant. During the 1982 El Niño, nearly all the adult birds on Christmas Island in the equatorial eastern Pacific abandoned their nestlings as the schools of small fish and squid--that are their principal food source--disappeared.

The 1972, and later the 1982, El Niño events were catastrophic to the Peruvian fishing industry. Overfishing may have also contributed to the major decrease in the 1972 catch. The anchovies caught in the ocean off Peru represented about 10% of the world's protein supply. In an apparently unrelated event, the Russian wheat crop also failed severely in 1972. The result was significant increases in the price of bread, corn, chickens, etc. as demand exceeded supply.

What causes El Niño?
El Niño is a natural occurrence resulting from the size, physical shape and thermal structure of the tropical Pacific, and the interactions between the ocean and the atmosphere there. The internal coupling between the ocean and the atmosphere has been shown to produce recurring El Niño events, without any external forcing required. The implication is that El Niño events are a natural mode of variability of the ocean-atmosphere system of the tropical Pacific Ocean. Interactions among the surface winds, the tropical rainfalls, the sea surface temperature and the ocean currents are sufficient to produce recurring warm and cold events.

Jacob Bjerknes, a meteorologist at the University of California at Los Angeles, suggested in 1969 that the normal distribution of sea surface temperature in the equatorial Pacific--cold in the east and warm in the west--results in a huge, thermally-driven circulation in the overlying atmosphere. Air sinks and produces clear skies over the cold water of the eastern Pacific, becomes warm and moist as it flows westward in the trade winds, produces clouds and rain as it rises over the warm water of the western Pacific near Indonesia and northern Australia, then flows eastward at high altitude toward the South American coast. He called this the Walker Circulation, in honor of Sir Gilbert Walker, who first studied the Southern Oscillation at the turn of the century. Bjerknes recognized that the Walker Circulation was the link between the Southern Oscillation and El Niño.

During El Niño events, the east-west sea surface temperature gradient is decreased. As a result, the Walker Circulation is reduced, or even reversed. Air is removed from over the eastern Pacific and accumulated over the western Pacific. The Southern Oscillation Index drops as surface pressures rise in the west and fall in the east. A positive feedback exists between the ocean and the atmosphere. The sea surface temperatures are modified by the resulting changes in the atmospheric circulation. The water becomes warmer than normal in the central and eastern Pacific, and cooler than normal in the western Pacific. The atmospheric circulation, in turn, responds to the changes in the sea surface temperature distribution. Air rises over the central and eastern Pacific, producing clouds and rains, and sinks over the western Pacific, resulting in clear skies and drought.

Can volcanic eruptions cause El Niño events?
In April, 1982 the volcano El Chichon erupted in Yucatan, Mexico. The gas and dust cloud from this volcano entered the stratosphere and circled the globe in a few weeks. This warm cloud interfered with the satellite measurements of sea surface temperature. The water appeared colder than it was because the satellites were sensing radiation emitted by the dust in the cold stratosphere instead of that from the warm water. By this time, El Niño of 1982-83 was already developing.

Again in 1992, Mount Pinatubo in the Philippines erupted violently. As before, El Niño was already underway. These coincidences have led some, including Paul Handler the University of Illinois at Urbana-Champaign, to suggest that large volcanic eruptions can cause El Niño events. Neville Nicholls of the Bureau of Meteorology Research Centre in Melbourne, Australia has shown there is no relationship between low-latitude volcanic eruptions and El Niño. Harry van Loon and his colleagues at the National Center for Atmospheric Research in Boulder, CO have noted that the eruptions of Mt. Agung in 1963 and El Chichon in 1982 actually suppressed the stratosphere's response to the El Niño events of those years.

Computer models of the ocean-atmosphere system are able to predict accurately the occurrence of El Niño without external forcing such as volcanoes. In fact, a statistical study of data over the last century revealed that eruptions were more likely after El Niño than before! Applying Occam's Razor, one would conclude there is no relation between volcanoes and El Niño.

Are all Los Niños the same?
Although there are many features that are common from one El Niño to the next, there are also significant differences. These include such factors as the intensity, timing, evolution and duration of each event. The Los Niños of 1972 and 1982 were very strong; those of 1976 and 1986 were moderate in intensity; while those of 1963 and 1969 were weak. Intensity here refers to the magnitude of the sea surface temperatures, the amounts of the anomalous tropical rainfalls and the strength of the wind anomalies.

During the El Niño of 1972 warm water first appeared off the coast of Peru late in the year and spread westward, while in 1982 the warm water first appeared in the western Pacific in May and spread eastward. While most Los Niños persist for no more than 18 months, the Los Niños of 1941 and 1991 both lasted more than two years.

The evolution of the thermocline movements is rather consistent from one El Niño to the next. Conditions in the central equatorial Pacific are more similar from one event to the next, compared to those in the eastern and western Pacific.

How can El Niño be predicted?
Stephen Zebiak and Mark Cane of the Lamont-Doherty Geological Institute of Columbia University have developed a simple coupled ocean-atmosphere model that appears to do a very good job of predicting El Niño events. Their ocean model includes both a well-mixed surface layer, and the stratified deep-ocean layer, with a thermocline of varying depth separating the two. This model ocean responds to variations in the near-surface winds. The circulation of their model atmosphere, in turn, responds to sea surface temperature anomalies, particularly in or near the two principal convergence zones in the tropical Pacific.

During a simulation, the model atmosphere first acts on the ocean, perturbing the sea surface temperatures and oceanic structure through wind stress. The model atmosphere then is allowed to respond to the updated oceanic conditions. The newly computed atmospheric circulation then acts on the ocean, and the coupled process continues.

Zebiak and Cane have noted that the heat content of the near-equatorial Pacific Ocean is critical in producing warm events. The heat content increases before a warm event, and dissipates rapidly during the event. The advection of warm water into the region, the degree of vertical mixing of the water, and the amount of solar radiation all contribute to the oceanic heat content.

The modeled warm events in some simulations are irregular in both amplitude and spacing in a manner very similar to those that are observed. The annual variations in the ocean and the atmosphere modulate the amount of interaction between the two. The annual cycle thus controls the time of the year during which El Niño events may develop, as well as dissipate. Zebiak and Cane have used their model to predict the occurrence of El Niño events as much as a year in advance. The Zebiak-Cane model also successfully predicted that 1990 would not be an El Niño year.

There are other models that have been developed to predict El Niño events. Scientists often use more than one numerical model to simulate phenomena; the different models, with varying degrees of complexity, allow them to evaluate the importance of various processes (such as upwelling, advection, thermodynamics, the role of boundaries, etc.) in the phenomenon they are studying. Tim Barnett and his colleagues at the Scripps Institute for Oceanography in California have developed a statistical prediction model for El Niño. This model attempts to identify the important features of the global atmospheric circulation that are precursors of El Niño events. James O'Brien and his colleagues at Florida State University have developed a single layer ocean model. The ocean in their model evolves while being driven by an unchanging, observed wind field. Both the Scripps and Florida State models successfully forecast the El Niño events of 1986-87 and 1991-92.

The fully-coupled ocean-atmosphere computer models are the best suited for use in routine operational climate forecasts. The Climate Analysis Center of the National Weather Service has proposed using a coupled model in issuing seasonal forecasts of temperature and precipitation as much as a year in advance.

Can a forecast of El Niño assist U. S. climate forecasters?
The precursors of an El Niño can sometimes be detected months in advance of its onset. Because El Niño lasts for such a long time the knowledge that an El Niño is occurring, or will soon occur, can be used by those who make long-range (monthly and seasonal) outlooks of the weather across the U. S. The areas most often affected are shown in Figures 6 and 7. Note that these figures have been produced by combining weather data for several El Niño events.

Figure 11 shows the temperature anomalies observed across the contiguous U.S. during recent El Niño winters. It is the winter weather that is most affected by El Niño. During El Niño winters, on average, it is warmer than normal in New England and the Pacific Northwest, and colder than normal along the Gulf Coast. However, there are large variations among the years when an El Niño is in progress. That makes it difficult to predict accurately the winter weather for much of the nation even when forecasters know an El Niño is occurring. Weather and climate anomalies always occur, and for many different reasons. El Niño is only one factor that may contribute to the weather of any particular season. Because El Niño has a significant influence on the weather over a large portion of the globe, it merits continued efforts at analysis and prediction.

Acknowledgements: Eugene Rasmusson of the University of Maryland, Chester Ropelewski of the Climate Analysis Center of the National Weather Service, Valerie Voss and Jeff Wilhelm of the Cable News Network Weather Center, Robert Weinbeck of the Project Atmosphere Office, and Carolyn Hawthorne and Faye McCollum, Atmospheric Education Resource Agents for Project Atmosphere, made many valuable suggestions that improved this paper.

Further Reading:
Barber, Richard and Francisco P. Chavez, 1983: Biological consequences of El Niño. Science (16 December) 222:1203-1210.
Barnett, Timothy, Nicholas E. Graham, Mark A. Cane, Stephen E. Zebiak, Sean C. Dolan, James O'Brien and D. Legler, 1988: On the prediction of the El Niño of 1986-1987. Science (8 July) 241:192-196.
Canby, Thomas Y., 1984: El Niño's ill wind. National Geographic (February 1984) 165:144-183.
Cane, Mark A., 1983: Oceanographic events during El Niño. Science (16 December) 222:1189-1195.
Cane, Mark A., Stephen E. Zebiak and Sean C. Dolan, 1986: Experimental forecasts of El Niño. Nature (26 June) 321:827-832.
Cromie, William J., 1980: When comes an El Niño? Science80 (March/April) 36-43.
Gill, A. E. and Eugene M. Rasmusson, 1983: The 1982-83 climate anomaly in the equatorial Pacific. Nature (17 November) 306:229-234.
Graham, Nicholas E. and Warren B. White, 1988: The El Niño cycle: A natural oscillator of the Pacific Ocean-atmosphere system. Science (3 June) 240:1293-1302.
Halpern, David, Stanley P. Hayes, Ants Leetmaa, Donald V. Hansen and George S. Philander, 1983: Oceanographic observations of the 1982 warming of the tropical eastern Pacific. Science (16 September) 221:1173-1175.
Jeffreys, William H. and James O. Berger, 1992: Ockham's razor and Bayesian analysis. American Scientist (January-February) 80:64-72.
Kerr, Richard A., 1990: Who will win the El Niño sweepstakes this time? Science (27 April) 248:445-445.
Kerr, Richard A., 1991: El Niño winners and losers declared. Science (8 March) 251:1182-1182.
Kerr, Richard A., 1992: A successful forecast of an El Niño winter. Science (24 January) 255:402-402.
Linden, Eugene, 1988: Big chill for the greenhouse. Time (31 October) p 90.
McPhaden, M. J. and J. Picaut, 1990: El Niño-Southern Oscillation displacements of the western equatorial Pacific warm pool. Science (7 December) 250:1385-1388.
Philander, S. George, 1989: El Niño and La Nina. American Scientist (Sep-Oct) 77:451-459.
Philander, S. George, 1990: El Niño, La Nina, and the Southern Oscillation. Academic Press, 289pp.
Ramage, Colin S., 1986: El Niño. Scientific American (June) 254, 76-83.
Rasmusson, Eugene M. and John M. Wallace, 1983: Meteorological aspects of the El Niño/Southern Oscillation. Science (16 December 1983) 222:1195-1202.
Rasmusson, Eugene M. and J. Michael Hall, 1983: El Niño--the great equatorial Pacific Ocean warming event 1982-83. Weatherwise (August) 166-175.
Rasmusson, Eugene M., 1985: El Niño and Variations in Climate. American Scientist (March-April) 73:168-177.
Ropelewski, Chester F., 1992: Predicting El Niño events. Nature (9 April) 356:476-477.
Rosen, Richard D., David A. Salstein, T. Marshall Eubanks, Jean O. Dickey and J. Alan Steppe, 1984: An El Niño signaling atmospheric angular momentum and earth rotation. Science (27 July) 225:411-414.
Schreiber, Ralph W. and Elizabeth Anne Schreiber, 1984: Central Pacific seabirds and the El Niño Southern Oscillation: 1982 and 1983 perspectives. Science (17 August) 225:713-716.
Smith, Robert L., 1983: Peru coastal currents during El Niño: 1976 and 1982. Science (30 September) 221:1397-1398.
Trenberth, Kevin E., Grant W. Branstator and Phillip A. Arkin, 1988: Origins of the 1988 North American drought. Science (23 December) 242:1640-1645.
Walker, Gilbert T. and E. W. Bliss, 1932: World Weather V. Memoirs Royal Meteorological Society, 4:53-84.


anomaly. A departure from normal, or average, conditions.

convergence zone. A region of the atmosphere (often somewhat linear) in which the airflow comes together. The term is often applied to those regions where the convergence occurs in the lower part of the atmosphere. The low level convergence results in ascending motion and the formation of clouds with copious rainfalls.

Coriolis Effect. The apparent deflection of a moving object to the right in the northern hemisphere, or to the left in the southern hemisphere, resulting from the rotation of the earth. The Coriolis Effect is due to the fact that the observer's reference system is rotating. The speed of the object is not affected; only its apparent direction of motion.

El Niño. A term originally used to refer to a warm current that develops almost every year along the coast of Ecuador and Peru at Christmas time. El Niño is now commonly used to refer to persistent, large-scale warm events in the central Pacific.

Kelvin wave. A perturbation in the atmosphere or the ocean characterized by alternating centers of high and low pressure along the equator. In the tropical Pacific Ocean, Kelvin waves propagate slowly from west to east, taking as long a six weeks to move from Indonesia to the South American coast.

Occam's Razor. A heuristic principle, or rule of thumb, frequently used by scientists. It has been stated as: "An explanation of the facts should be no more complicated than necessary," and "Among competing hypotheses, favor the simplest one." Attributed to William of Occam, a 14th century English philosopher.

positive feedback. An interaction between two parameters such that a change in one results in a change in the other that, in turn, reinforces the change in the first.

standardized difference. The difference between two numbers, represented in terms of the standard deviation. The standard deviation is a measure of the degree of spread of a set of numbers about the average value of the set. In the case of the Southern Oscillation Index, standard deviations are used to seasonally adjust the pressure differences to a common scale.

thermally driven circulation. A circulation in a fluid (either liquid or gas) in which the warmer, less dense part of the fluid rises, while the colder, more dense part of the fluid sinks. The flow at the top of the fluid is from warm to cold, while at the base of the fluid the flow is from cold to warm.

thermocline. A layer in the ocean in which the temperature decreases rapidly with increasing depth. The thermocline separates the warm, well-mixed surface layer from the cold, stratified deep layer.

trade winds. The wind system, occupying most of the tropics, which blows from the subtropical highs toward the equatorial trough; a major component of the general circulation of the atmosphere. The trade winds are characterized by a great consistency of direction and, to a lesser degree, speed.

upwelling. The rising of water from subsurface layers toward the surface of a body of water. Upwelling is most prominent where persistent winds blow parallel to the coastline such that the resulting wind-driven current is directed away from the coast. Upwelling also is observed where the southern easterly trade winds cross the equator. Upwelling has a distinct influence on a locale by bringing cooler water to the surface. The upwelled water, besides being cooler, is richer in plant nutrients, so that regions of upwelling are generally also areas of rich fisheries.

Walker circulation. An east to west circulation in the tropical atmosphere in which air rises in one location, flows nearly parallel to the equator at high altitudes to another location where in sinks, then flows at low levels back to the location of rising air.

wind shear. The change in wind speed and/or direction between two points.

wind stress. The frictional interaction between the wind and the underlying ocean that results in ocean currents that flow in the same general direction as the wind.