Frequently (well, at least once)-asked-questions about El Niño

El Niño is an intermittent disruption of the climate system centered in the equatorial Pacific that has effects on short-term climate around the Pacific basin.

To understand El Niño, it is necessary to understand the normal trade-wind system in the tropical Pacific. (I have a schematic diagram of this system). The sun heats the equatorial regions more strongly than the rest of the globe, so air tends to rise from the surface there, replaced by inflow from the subtropics. The Coriolis effect turns these inflows to the right in the northern hemisphere and to the left in the southern, resulting in the great trade-wind belts that blow equatorward and westward over the width of the the tropical Pacific. This sets up the coupled ocean-atmosphere interaction in the tropical Pacific in which the winds determine the water temperature but the water temperature also determines the winds, in a chicken and egg situation. In this system, we can start a description at any point in the cycle. For example, we observe that there is cool water in the east and warm in the west (see the figure "Mean SST and winds in the tropical Pacific"). The winds blow towards the warm water, since that heats the atmosphere and makes the air rise, then other air flows in to fill the gap. (These are the trade winds, that the Spanish used to sail from their colonies in South America to the Philippines). Because of the force of the trades, sea level at Indonesia is about 1/2 meter higher than at Peru. At the same time the trade winds act on the ocean as well. The westward winds along the equator push the warm water (heated by the sun) off to the west, drawing up the thermocline and exposing the cooler water underneath in the east. This upwelling cools the eastern surface water, and we have returned to the starting place of the description.

So if you ask me "Why are there trade winds?", I will say "That's simple, there are trade winds because the water is cold in the east and warm in the west". But if you ask me "Why is the water cold in the east and warm in the west?", I'll say "That's simple too, it's because there are trade winds". The ocean and the atmosphere are inextricably coupled together.

The thermocline is the boundary between the cold deep water (that extends to the bottom of the ocean and around the world) and the warmer upper layer. In the tropics it forms a relatively sharp boundary. The trade winds, in piling up warm water in the west, make a thick (150 meter) warm layer in the west that pushes the thermocline down there, while it rises in the east. The shallow (30 meter deep) eastern thermocline allows the winds to pull up water from below, water that is generally much richer in nutrients than the surface layer. (This is so because life lives mostly in the sunlit zone. Available nutrients there are quickly taken up by some form of life (plankton). Debris from the living creatures rains down into the lower layer, where nutrients dissolve, making the deep water rich, so that wherever there is upwelling, there is abundant life.)

Note that the rising air over the western Pacific is associated with rainfall (see the figure of rainfall and winds just above). When air rises it cools, and can hold less evaporated water. The water comes out as rain. But in returning to a liquid state, it releases the heat that was used to evaporate it from the ocean surface (heat that came from the sun), and this middle atmosphere heating amplifies the rising motion. This is a principal mechanism for heat from the sun to warm the atmosphere (the atmosphere by itself is relatively transparent to solar radiation).

Because the warm pool pumps great amounts of heat and moisture into the upper atmosphere, this system is one of the major driving forces of world climate. The huge source of heat helps set the path of the jet streams (storm tracks) that control temperate-zone weather, much as a large rock in a stream determines the pattern of water flow, including wavey motions that extend well downstream of the rock. Therefore, when the warm pool changes shape or position, the effects ripple outward to affect much of the world's weather.

During El Niño events, this entire system relaxes. The trade winds weaken, particularly west of the Dateline, and the piled-up water in the west sloshes back east, carrying the warm pool with it. The region of rising air moves east with the warm pool, and so does the pumping of heat and moisture into the upper atmosphere, distorting the usual paths of the jet streams, which eventually causes the changes in weather around the world. With weakened trade winds, the upwelling in the east correspondingly weakens; as the warm pool moves east the upwelled water is also not as cool as during normal periods. When eastern SST becomes warm the east-to-west temperature contrast is small, and so the trade winds weaken even further, leading to a complete collapse with essentially flat conditions across the entire equatorial Pacific.

We don't know what initiates El Niño. But we do see that it apparently does not take much to destabilize the strong-trade-wind/large-temperature-contrast non-El Niño state discusses in the section above. This is because of ocean-atmosphere coupling. (See a schematic diagram of the ocean-atmosphere system during the onset stage).

Suppose a relatively brief opposing wind occurs over the west Pacific warm pool. It may last for as little as one month. This starts an eastward current that pushes the warm pool a little bit east of it's usual position. If the ocean and atmosphere were not coupled, then this motion would soon stop when it ran into the trade winds. But we have shown that the trade winds exist because of the temperature contrast between east and west. If the central Pacific is warmed by flow from the west, even a small amount, then the region of rising air will tend to move east with the warm water. That means the trade winds will also shrink back east a little, since those winds are caused by the rising air. But then the pressure of the trades holding up the sea surface slope to the west is weakened, and even more west Pacific water tries to slosh eastward. That warms the central Pacific a little bit more, and the rising air moves further east, and the trades shrink more. This collapse continues until the water is warm across the Pacific, the trades are weak, and the thermocline and sea surface slope flatten out. That is El Niño.

While we can describe the collapse pretty well, we don't seem to know what makes the system ready to collapse, or what exactly causes the opposing winds in the first place.

There are two main theories at present. The first is that the event is initiated by the reflection from the western boundary of the Pacific of an oceanic Rossby wave (type of low-frequency planetary wave that moves only west). The reflected wave is supposed to lower the thermocline in the west-central Pacific and thereby warm the SST by reducing the efficiency of upwelling to cool the surface. Then that makes winds blow towards the (slightly) warmer water and really start the event. The nice part about this theory is that the Rossby waves can be observed for months before the reflection, which implies that El Niño is predictable.

The other idea is that the trigger is essentially random. The tropical convection (organized largescale thunderstorm activity) in the rising air tends to occur in bursts that last for about a month, and these bursts propagate out of the Indian Ocean (known as the Madden-Julian Oscillation). Since the storms are geostrophic (rotating according to the turning of the earth, which means they rotate clockwise in the southern hemisphere and counter-clockwise in the north), storm winds on the equator always blow towards the east. If the storms are strong enough, or last long enough, then those eastward winds may be enought to start the sloshing. But specific Madden-Julian Oscillation events are not predictable much in advance (just as specific weather events are not predictable in advance), and so to the extent that this is the main element, then El Niño will not be predictable.

In my opinion both these two processes can be important in different El Niños. Some models that did not have the MJO storms were successful in predicting the events of 1986-87 and 1991-92. That suggests that the Rossby wave part was a main influence at that time. But those same models have failed to predict the events since then, and the westerlies have appeared to come from nowhere. It is also quite possible that these two general sets of ideas are incomplete, and that there are other causes entirely. The fact that we have very intermittent skill at predicting the major turns of the ENSO cycle (as opposed to the very good forecasts that can be made once an event has begun) suggests that there remain important elements that are await explanation.

The most severe effects of El Niño are found close to the equator. The usual pattern of deserts in Peru and heavy rainfall over Indonesia and the west Pacific reverses. Forest and range fires can occur in Indonesia (as has been happening in recent months, exacerbated by deliberate burning) and Australia, while Peru suffers flooding, with accompanying epidemics of cholera and other sewage-borne diseases. The food chain in the rich upwelling region is disrupted, so fish die off, with consequent hardship for the birds, mammals and people that survive on that stock. The warmer water near Central America spawns more and stronger hurricanes, which can go as far west as Hawaii. The entire sequence of the event lasts about one year, and events are usually separated by 2-7 years, in an irregular and not-well-understood pattern (see a time series of occurrences of El Niño since 1950).

Effects of El Niño and La Niña on North America are shown in a schematic diagram with text explanation from NOAA's Climate Prediction Center.
Back to top

The name El Niño (referring to the Christ child) was originally given by Peruvian fisherman to a warm current that appeared each year around Christmas. What we now call El Niño seemed to them like a stronger event of the same type, and the usage of the term changed to refer only to the irregular strong events. It wasn't until the 1960s that it was widely realized that this was not just a local Peruvian occurrence, but was associated with changes over the entire tropical Pacific and beyond.

The following quote is given in the introduction to an excellent (scholarly) book by George Philander of Princeton University ("El Niño, La Niña and the Southern Oscillation", Academic Press, 1990). These are remarks quoted from Senor Federico Alfonso Pezet's address to the Sixth International Geographical Congress in Lima, Peru in 1895.

In the year 1891, Senor Dr Luis Carranza, President of the Lima Geographical Society, contributed a small article to the Bulletin of that Society, calling attention to the fact that a countercurrent flowing from north to south had been observed between the ports of Paita and Pacasmayo.

The Paita sailors, who frequently navigate along the coast in small craft, either to the north or the south of that port, name this countercurrent the current of "El Niño" (the child Jesus) because it has been observed to appear immediately after Christmas.

As this countercurrent has been noticed on different occasions, and its appearance along the Peruvian coast has been concurrent with rains in latitudes where it seldom if ever rains to any great extent, I wish, on the present occasion, to call the attention of the distinguished geographers here assembled to this phenomenon, which exercises, undoubtedly, a very great influence on the climatic conditions of that part of the world.

The Southern Oscillation was named by Sir Gilbert Walker in 1923, noting that "when pressure is high in the Pacific Ocean it tends to be low in the Indian Ocean from Africa to Australia". This was the first recognition that changes across the tropical Pacific and beyond were not isolated phenomena but were connected as part of a larger oscillation. Walker was Director of Observatories in India and was mostly concerned with variations in the Indian monsoon.

The first real description of El Niño/Southern Oscillation in terms of physical mechanisms was by Prof Jacob Bjerknes of the Univ of California, Los Angeles in 1969.
Back to top

To answer this question we need to consider how much heat change is involved in climate oscillations such as El Niño.

The increase in SST during January through July 1997, averaged over the equatorial Pacific (10S-10N, 150E-75W) was about 1.62°C.

But the quantity we need is heat content, which proportional to the mass of water times its temperature. This can be estimated from the TAO buoy array, by multiplying the density of water (1030 kg/m3) times the heat capacity of seawater (4000 J/kg/K) times the area of the TAO array times the vertical integral of buoy temperature with depth. (see the note on units and symbols below) (The area was estimated by dividing the array into regions each assumed to be represented by one buoy. Each day, the above calculation was made summing over the area of only those buoys with data for that day). Performing this calculation with TAO temperatures, the time series of heat content in the upper 500m shows that this quantity increased by about 3.5e22 J during July 1996 through March 1997, then decreased slightly. (See also other related calculations).

Now we can compare that heating to human engineering power. One of the most powerful devices humans have created is the H-bomb. The heating during El Niño can be related to the power of the H-bomb as follows. One kiloton of TNT is about 4.18e12 J (Reference). Therefore. 3.5e22 J is about equivalent to 8e9 kilotons, or about 400,000 20-Mt bombs. That assumes the power of those bombs could be directed entirely to heating (i.e. not to moving) the water.

Another way to look at this is to estimate the heating that would be done by a bomb if all the energy went into heating. Consider a 1-Mt H-bomb, releasing 4.18e15 J of energy. That energy would heat a volume of about 1 cubic kilometer of seawater 1°C. That certainly seems like a lot of energy, but the volume of just the upper water of the equatorial Pacific (10°S-10°N, upper 200m) is about 6 million cubic kilometers. Compared to H-bombs, climate changes are a different order of beast entirely.

Another way to look at it is in terms of power plants. Increasing the heat content by 3.5e22 J over the period July 1996 to March 1997 is a heating rate of about 1.5e15 W (1.5 PW). A large power plant can generate about 1000 MW, so the El Niño-related heating is equivalent to the total output of roughly 1,500,000 power plants working continuously for 8 months.

Of course, this argument is misleading in that El Niño probably does not actually heat that much water. What it does is redistribute the heat already in the ocean, first moving it in to the TAO array region, then out. However, it probably is the case that the work required to move that much water is similar in magnitude to the work required to do the heating as above. This calculation is harder to do because we can't measure the currents nearly as well as the temperatures.

Even if we could make the kind of deliberate changes of ocean temperature suggested above, it would have many unpredictable effects, and we do not understand the complex processes well enough to know what would happen if it was done. It is important to remember that El Niño is not the only climate fluctuation that is occurring. El Niños occur on a changing background state, which is probably the reason that the various El Niños are somewhat different. So it would not be simple to figure out how to modify the ocean temperatures to get the changes we wanted, even if we had the technological means to do that.

In any case, even if we had a way to channel these effects into direct heating of ocean water, it is clear from the magnitudes involved that people are nowhere near the ability to modify the climate on this scale.

There is one exception. By burning fossil fuels at the rate we do, we increase the CO2 concentration in the atmosphere in a way that could well cause climate changes (global warming), as most scientists think is now happening. The difference is that this change represents the "efforts" of all the industrial world acting over a century to just get the change under way.

A note on units used here:
kg = kilogram.
m = meter.
K = degrees Kelvin (same units as °C, but starting from absolute zero or -273.15°C)
J = Joule. This is a unit to measure work or energy.
W = Watt. 1 W = 1 J/second. This is the power (rate of doing work).
Abbreviations for large numbers: M = meta (10e6 or 1 million). P = peta (10e15 or 1 million billion).

Back to top

Here's an idea. This experiment is similar to what actual scientists are doing right now.

The project is to construct some forecast models of El Niño's development over the next few months. We don't know what it will do. Will it get more intense?, weaken?, remain strong?, and if so for how long? These are the subject of much debate in the scientific community right now, and many efforts are under way to predict and understand it.

The models would be forecasts made using several assumptions, and the main result would be graphs showing how the forecasts compared with actual evolving conditions.

One model would be called "persistence". That is, whatever conditions are occurring now, they will continue. Surprisingly, persistence is often a hard-to-beat forecast, and weather forecasters score themselves on how much better than persistence thay can do. A second model is continuation of the trend. That is, if the sea surface temperature (SST) is warming up it will continue to warm at the same rate. Obviously that can't go on forever but in many ways a trend is a good indicator of future trends. A third model is random changes. Get a random number generator (or pick numbers out of a hat). Each day or week, use the random numbers to predict what the change of SST will be (scale the numbers to keep it reasonable). Those are three simple models that can be used to project forward from current conditions. Essentially that's what weather forecast models do, just more sophisticatedly (see question 14). Maybe you can think of some other ways to make forecasts (if you get something that works, send it in!)

Choose a few buoys from our network in different regions of the tropical Pacific (for example, on the equator, off the equator, in the east, and the west). Get the data from our web page (click for detailed instructions to get this data). Make and graph predictions for each buoy chosen for a month or two ahead, then collect observations as they come in (the data files are updated daily). Graph the observations against the three predictions. My guess is that each model would be successful in some regions for some periods of time. Other extensions would be to compare forecasts beginning at different times. Perhaps a forecast begun with September comditions is good for 3 months, but one begun in December is only good for one month. Etc.

Another simple project is to determine how significant an effect El Niño has on your local region. Do this by gathering an assortment of local weather time series from your region (monthly rainfall, temperature, etc) (available at the web pages of the National Weather service). Then get an index of El Niño like the Southern Oscillation Index (see Question 16 for a description and graphic, and download the values at NOAA's Climate Prediction Center. The data links are: values for 1951-today and 1882-1950. Note that the SOI monthly values are very jumpy and must be smoothed by a 5-month running mean). Compare the turns of the El Niño/La Niña cycle with changes in your local weather; this could either be through a listing of El Niño/La Niña years and good/bad local weather, or by correlation of the two time series (send me e-mail for how to do correlation). You will probably find out that some aspects of your local weather are related to the El Niño/La Niña cycle and some are not. Also that some strong El Niño or La Niña years make a difference but some do not. This reflects the fact that, far from the center of action in the tropical Pacific, El Niño is only one of many influences on weather.

If your are pretty good at math and computer programming (at least 8th-grade math), then I have a more advanced project that you can find here.

Other web sites for classroom lessons on El Niño:
At the University of Illinois (Champaign-Urbana)
At the Oak Ridge National Laboratory
Back to top

It is important to remember that El Niño is not the only effect on weather. It is but one fluctuation among many, and the weather we experience is the sum of all of these overlaid and interacting. Most of these interactions are poorly understood, particularly the longer-term ones, and as we get longer and longer records we become aware of more and more complexity. The different El Niños occur on different background states, at different times of year, and therefore have different total effects. Therefore, one cannot simply speak of the isolated effects of El Niño on weather in San Diego. There is only the ever-changing combination of influences. That is the main reason why we cannot produce reliable long-term forecasts.

Scientists study El Niño partly for its own effects, but also partly as an example of how this kind of climate oscillation interacts with the rest of the climate system. We know there are many such oscillations, and we would like to be able to fit the whole picture together. We hope that what we learn about the climate system from studying El Niño will help us understand other, less obvious, variability.
Back to top

El Niño is not "responsible" for any particular storm. But El Niño is a vital element all the same. For example, winter is not responsible for any particular snowstorm; but winter is certainly a crucial part of why the northern hemisphere has snow in January.

There are two ways in which El Niño adds punch to the storms over North America. First, it draws the jet streams further south in the eastern Pacific. That means more storms in general across the southern tier of the U.S. Second, El Niño adds punch to the storms because the stronger and eastward-displaced warm pool (see question 1) pumps additional moisture into the upper atmosphere, where it is blown towards North America and caught up in the more-southerly jet streams.

So while it is impossible to prove that El Niño caused the particular tornadoes that struck central Florida in late February, just as it is impossible to say that winter caused a particular snowstorm, it is the likely suspect. Perhaps there would have been strong tornadoes on this day in Florida without El Niño, but it appears not.

We cannot run experiments to see what a parallel earth without an El Niño this year would do. But a group at NOAA's Climate Diagnostic Center in Boulder, CO, is trying something similar using numerical forecast models. First, they run the model with the actual conditions (that is, including this year's El Niño-induced SST anomalies in the Pacific) and produce weather forecasts, just like the regular ones. Then they make another run, in which everything is the same as the first one except that they change the Pacific SST to be like a "normal" year. The difference between these forecasts gives an indication of the effect of El Niño conditions in the Pacific on the specific weather events being forecast. The results for forecasts of Florida precipitation during the tornadoes of Feb 23 suggest that El Niño did indeed add punch to these storms. Visit the web site of this forecast group.
Back to top

The reason that you don't see much publicity about the causes of El Niño is that we don't understand the origins of the event. We do, however, have a pretty good understanding of how it evolves once it has begun, and that gives a useful ability to make forecasts 6-9 months ahead for some regions. That is the information you see because that is the present state of reasonably-secure knowledge. Of course, there are a variety of theories, and many scientists are working on various aspects of the genesis, which would presumably extend the predictive skill out another few months or even years.

The fact is, at several points over the past two decades we thought we had working theories of what causes El Niño. Unfortunately (or perhaps fortunately for those who like scientific challenges), nature has shown that those theories were at best incomplete. For example, during the mid-1980s a group at Columbia University developed a fairly simple theory and wrote a computer model to produce predictions based on it. This was successful in predicting the 1986-87 and 91-92 events almost a year in advance, and they were (figuratively?) breaking out the champagne. Then along came the event of 1993, then another in 94-95, and most prominently the present event, none of which developed according to the ideas in their theory.

The main reason this is so difficult is that the processes that cause El Niños involve the full complexity of ocean-atmosphere interaction on a global scale. We have developed a reasonably good understanding of how the atmosphere works (at least in theory), once the sea surface temperature (SST) that drives the atmospheric circulation is known. (We are somewhat further behind when it comes to the ocean, which is much harder to observe). That works pretty well to run the atmospheric models to make short-term weather forecasts, because the ocean changes rather slowly. But, when you consider longer-term phenomena like El Niño, it is not enough to specify the SST; one must consider how the ocean will evolve under the winds, and then how the altered ocean will modify the winds, and so on, in many tricky and sensitive feedback loops. We are just beginning to be able to see how these fundamentally coupled disturbances work, and generally only in very idealized cases. Remember that for a long time meteorologists only talked to meteorologists, and oceanographers only to oceanographers. Now we are really at the initial stages of being able to think about these coupled problems.

Nevertheless, it seems to me that we are able to a lot of good for society even at our present stage of ignorance, since even without knowing what drives El Niño, we can recognize it, and then know (largely from statistics of past events) what the effects will be on regions far removed from the tropical Pacific. That's what the publicity is about, and that is appropriate, although perhaps not so satisfying to those such as yourself who ask "why?".
Back to top

We observe that large-amplitude climate signals occur well outside the tropics during El Niño events. The question is, are those precursors of El Niño, or are they far-field effects essentially incidental to the equatorial Pacific "main event"? Both points of view have their advocates.

On the one hand, there is no doubt that the entire ocean-atmosphere system is interconnected, and each El Niño event occurs on the background of existing conditions, which includes the positions of the mid-latitude jet streams as an important component. In addition, newly-developing theories of the slow (decadal) evolution of the equatorial circulation point to a key role for water masses that originated in the subtropics or higher latitudes. These water masses are part of a global-scale meridional overturning circulation, in which water sinks at high latitudes (due to evaporation or winter cooling making it more dense), and travel a circuitous subsurface route to the equator, where they upwell. Then the water flows poleward at the surface back to the subtropical gyres, warming under the sun to complete the cycle, which probably takes several decades or so. Now researchers have shown that the sinking water masses can have different temperature and salinity properties, and that as these reach the equator, they modify the conditions prevailing there, and it is thought that these slow changes may be the reason why some periods have many El Niños (like the 1990s) while others don't.

On the other hand, numerical models of the equatorial ocean-atmosphere system (including both fairly simple to quite complex formulations) commonly develop interannual climate oscillations similar to El Niño. This indicates that at least the basic phenomena is intrinsic to the general situation of a wide ocean spanning the equator, and theory bears this out. Therefore I think the prevailing sentiment among climate researchers is that the mid-latitude influences referred to above can modify individual El Niños, but that fundamentally it is an instability of the equatorial ocean-atmosphere system that would occur in recognizable form regardless of what happens poleward of say 15-20 degrees latitude.

However, at present we do not know with any certainty how El Niño events are initiated. Therefore the tropical/extra-tropical debate cannot be said to be resolved. Another regional influence that is cited by some is the Asian monsoon. As with the mid-latitude situation, the fact that El Niño develops in models without an Indian Ocean again indicates that the Pacific ocean/atmosphere system is probably sufficient, but we really don't know for sure. These topics are the subject of much current research.
Back to top

The idea that volcanoes cause El Niño events originally gained prominence because of the eruption of Mt Chichon in Mexico in Feb 1982 (preceding the El Niño of 1982-83), and the eruption of Mt Pinatubo in the Philippines) in June 1991 (preceding the El Niño of 1991-92). However, when the time series of El Niños is compared to the time series of volcanic eruptions it becomes clear that the relationship is coincidental. There are numerous large volcanic eruptions around the world and almost as many El Niños. In that situation there is almost always an eruption at some time preceding any El Niño. Scientists are now convinced that this relation is coincidental.

Certain experiments bear this out. For example, several computer models predicted the onset of the 1991-92 event as early as January 1991, based on the state of the ocean-atmosphere system at that time, well before Pinatubo. That indicates that the ocean-atmosphere system was already generating the El Niño, and Pinatubo occurred coincidentally. Computer models integrating the equations of fluid motion and the flow of heat routinely produce El Niño-like variability completely on their own. Of course, computer models are not reality, but these experiments suggest that El Niño is a natural mode of variability of the ocean-atmosphere system, as much as, for example, a thunderstorm. While we do not have a complete picture of how the El Niño cycle operates, these models (and a developing theoretical understanding) suggest that the fluid envelope of the earth is prone to developing various kinds of instabilities, ranging from storm systems lasting a few hours or days, to El Niño, to longer-term fluctuations that we are just beginning to explore. There is no reason to think that external processes such as volcanoes are a necessary element.

None of this is to say that volcanoes don't affect the climate. They most certainly do, and since El Niños occur against the background existing climate, there is little doubt that volcanic eruptions that eject large amounts of dust into the stratosphere must modify the frequency, character and strength of El Niño events, possibly in important ways. The distinction I make is between "slowly modify the background" and "cause" El Niño.

As far as deep-ocean vents modifying the ocean temperatures, researchers now think that this source of heat does contribute to the long-term evolution of the ocean state. The chemical signatures of undersea vents are of great interest as tracers of the slow deep circulation of the ocean, and therefore these signatures are studied carefully. (The deep circulation is so slow that its currents cannot be measured directly, so we look at tongues of chemical tracers to estimate the speed, direction and transport of the flows). Numerous scientific papers discuss these questions, studying a variety of chemical constituents. What is consistently found is that the traces spread extremely slowly through the water column and are vastly diluted. There is little doubt that over very long periods the effects of undersea venting on the ocean are large, both for their heat and for their contribution to the chemical makeup of the ocean. However, these effects occur on timescales of thousands of years, and certainly do not produce the kind of rapid signals that characterize El Niño. To trigger an El Niño event, one would look for a signal that produced surface variability on a month to month or year to year timescale, and undersea venting has never been observed to do that.

It is indeed tempting to look for nice clean causes of complex oscillations like the El Niño cycle. Unfortunately (or perhaps fortunately for those of us who like scientific challenges), it seems that the ocean-atmosphere system is well-capable of generating these oscillations on its own, and the task now is to understand how this happens. Volcanoes and sea-floor venting are part of the slowly-changing background state on which phenomena like El Niño occur, and add to the complexity of the task.
Back to top

El Niño is part of the natural rhythm of the ocean-atmosphere system, as much as winter cold or summer thunderstorms or any other weather phenomenon.

In a complicated system like this, each feature fills a role in the grand scheme of things. We cannot say exactly what the role of El Niño is, but we do observe that these events drain the west Pacific of heat that is built up over several years by the trade winds. In any case El Niño does not exist in isolation, and any changes in it would reverberate around the whole system in unpredictable ways. Further, as part of the natural environment of the Pacific basin that the animals, fish, birds and plants have adapted to over the millenia (we know that El Niños have been occurring as far back as we can see), it is not clear that stopping El Niño would even be desirable. Even if it were possible to make El Niño disappear, we have no idea what the result would be. Very possibly we wouldn't like it!
Back to top

As you probably know, Drake Passage has some of the worst weather on the planet. Anyone going there should be prepared for severe conditions at any time of year, even the height of summer. However, one goes to Antarctica to experience the majesty of wild nature, and I'm sure it will be rewarding and beautiful nonetheless.

As far as I know there are no reported effects of El Niño on Drake Passage. That doesn't mean there aren't any, but they haven't made it into the mainstream scientific literature, at least the part that I've read. Despite all the present media hype, it is important to remember that El Niño is not the only game in town. El Niño is only one fluctuation among many, and all these things are overlaid and interacting. That's one of the main reasons why we can't produce reliable forecasts for all regions. In some places, such as the U.S. Gulf Coast, the correlation of cool, rainy winters with El Niño is well-established, and we can give a forecast with some confidence. In others, such as the U.S. northeast, the correlations are weak and may be coincidental. Perhaps some of the year-to-year variations we see in such regions are due to El Niño, but perhaps not. There have only been 10 or so El Niños since the 1950s, so we really have a weak basis on which to compute statistics, unless the signal is very clear.

The tides are controlled by the sun-moon-earth geometry. El Niño has little effect on this and so has little or no effect on the tides anywhere. (Of course, El Niño does affect sea level, by perhaps 20-30 cm, as seen in satellite altimetry data, but these changes happen over a period of months and are completely different than the tides). Perhaps surprisingly, during the largest El Niño in 1982-83, it was noted that the anomalous eastward motion of the atmosphere (due to weaker trade winds, see question 1) actually did cause the earth's rotation to slow measurably, increasing the length of a day by almost 1 millisecond. This occurred because the rotation (more precisely the angular momentum) of the entire mass of the earth/ocean/atmosphere system remains constant, so when the atmosphere "went faster" in the direction of rotation (towards the east), the solid earth slowed down! This presumably had a tiny effect on the tides.

I have seen nothing to indicate that El Niño might pose an extra danger to travellers in Drake Passage, who presumably are going on well-maintained ships with well-trained crews prepared for weather. Your Mom should enjoy her trip. I always wanted to go there myself and would not let El Niño change my plans.
Back to top

Teleconnection modes can be visualized as the preferred ways the atmosphere oscillates (in a manner roughly akin to a drum head) on time scales of weeks and longer. El Niño disturbs the atmosphere by moving the west Pacific warm pool (region of sea surface temperature greater than 28C, usually found from Indonesia to the dateline) eastward. The warm pool is one of the main sources of heat driving the atmosphere, comparable to a continent during summer. The effects of the warm pool extend upward through the atmosphere because the warm water tends to produce and support extensive organized thunderstorm activity and tropical cyclones. These pump large amounts of heat and moisture to high levels. When the warm pool is displaced east during El Niño, the the enhanced thunderstorm activity in the central equatorial Pacific perturbs the flow in the upper atmosphere, with a ripple effect downstream consisting of a series of alternating high and low pressure regions along a great circle route into North America. One might make an analogy to placing a large rock in a mountain stream. The disturbance spreads downstream in a series of waves, whose length is determined by the size of the rock and the speed of the flow. On a planetary scale, the waves take particular forms and directions due to the sphericity and rotation of the earth, and the general westerly flow of the temperate-latitude jet streams.

The principal teleconnection mode between the tropical Pacific and the Northern Hemisphere circulation is called the Pacific-North American pattern or PNA. The downstream wave appears as higher pressures than normal over western Canada and lower pressures than normal over the southeastern United States. This in turn modifies the winds and jet streams which are responsible for the temperature and precipitation anomalies over North America.

The mid-latitude jet streams occur fundamentally because of the temperature gradient between the equator and the poles. This gradient is not evenly spread over that entire distance, but tends to be concentrated in relatively narrow bands. For example, in winter it is fairly uniformly hot between the equator and, say, central Mexico, while it is fairly uniformly cold between the northern U.S. and the north pole. The whole gradient is concentrated in perhaps one quarter or less of the equator-pole distance. The jet stream flows in this region. During non-El Niño periods the eastern equatorial Pacific is cool, and the equator-pole gradient is weaker there than in the west. During El Niño periods, the eastern Pacific warms up as much as 4-5°C and the gradient is consequently larger. This contributes to strengthening the jet approaching North America.

Another element affecting North America is that when the eastern Pacific is warm, increased evaporation occurs, as well as tropical convection, and the moisture pumped into the upper atmosphere can be carried northeast by the subtropical jet stream, contributing to heavier rainfall across the southern tier of the U.S., from California to Florida.

Other important teleconnections that have been noted include droughts that often occur in Africa and northeast Brazil during El Niños. In this case the mechanism has to do with eastward displacement of the circulation in the equatorial plane. In non-El Niño years, this circulation consists of rising motion over the warm pool and sinking over the eastern Pacific, connected by surface westward winds and upper eastward winds, known as the Walker cell. (This pattern of vertical motion is why it is usually wet in Indonesia but dry in Peru). When the cell moves east, the sinking motion is correspondingly displaced, producing drought in the equatorial Atlantic region. There are also probably other effects due to the anomalous Atlantic winds changing the local SST. There may also be far-downstream effects associated with the PNA.

Other notable effects in the Atlantic include a reduction in the frequency of hurricanes that usually strike the Caribbean region and the south- eastern U.S. in late summer and fall. It is thought that this is also due to the eastward displacement of the Walker cell, as stronger winds in the upper atmosphere shear off the tops of developing tropical storms in that region, tending to prevent them from becoming hurricanes.
Back to top

There are two main types of forecast. First are statistical forecasts, based on historical records. Second are dynamical forecasts, based on forward integration of numerical models of the coupled ocean/atmosphere system. Each has its strengths and weaknesses, and the results from these can be quite different.

Statistical forecasts correlate observed weather conditions with indices of the occurrence of El Niño. Typically sea surface temperature (SST) in key regions of the equatorial Pacific are used to define "El Niño periods". Alternatively an index known as the "Southern Oscillation Index" (SOI) (see Question 16) is used, based on the surface pressure difference between Tahiti and Darwin, Australia. The advantage of the SOI is that records at those two locations go back a century, while we have only a few decades of SST observations in mid-ocean. Then the correlation of one of these indices with, say, rainfall in California, is the basis for a forecast of the likelihood of reoccurrence of heavy rains in that region during an El Niño winter. These are probably the most common type of forecast that is reported in the media. In some regions, such as the U.S. Gulf Coast, these correlations are quite robust and the statistical forecast is fairly reliable. In others the correlations are weak or marginal.

The strength of statistical forecasts is that they are based on events that actually did occur, but they can fail because El Niño is not an exact, repeating phenomenon. We observe that different events evolve in different patterns, can occur at different times of year, etc. In addition, there are many climate oscillations on all timescales occurring simultaneously, and the weather we experience in any location is the sum of these oscillations and the interactions among them. Most of these are as yet poorly understood, particularly the longer-term ones, and as we get better and longer records we become aware of more and more complexity. Therefore, it is not straightforward to isolate the specific effects of El Niño by averaging over previous events. All these things result in blurring the statistics and reducing the confidence in such a forecast.

Another problem with statistical forecasts is that we do not have good, long-term records of many of the important quantities of interest. Once you go back further than the mid-1950s, the ocean records are sparse and ambiguous, making it hard to determine which are strong El Niño years and which are weak (or even whether or not there is really an El Niño at all). But if attention is limited to the period of good data, then there are really only a handful of events, and the statistics become quite unreliable. (If you measured 5 children from a classroom of 30, you might not get a good estimate of the mean height of the class). Many of the differences among statistical forecasts reported in the media are due to the choice of different averaging periods.

Dynamical forecasts are based on the hydrodynamical equations numerically integrated forward from present observed conditions. These computer models range from relatively simple representations to complex models such as are used in weather forecasting. The simple models are useful in that specific processes can be analyzed and dissected. For a while during the 1980s it appeared that much of the El Niño cycle could be understood in terms of planetary waves bouncing around the Pacific, and this could be reasonably well simulated in a simple model. However, this theory failed to predict the series of El Niño events during the 1990s, and it appears that we must simulate the full complexity of ocean-atmosphere interaction. This is a task of utmost difficulty, compounding the problems of ordinary weather forecasting by the addition of numerous interactions between the ocean and atmosphere.

One of the major difficulties of dynamical forecasting is that we cannot simulate every molecule of air and water, but must reduce the various phenomena to a relatively crude grid mesh representation of the world. Such grids have spacing of typically tens to hundreds of kilometers, due to limitations of computer speed and storage. Consider, for example, the problem of representing clouds in such a model. The grid is far too coarse to resolve individual clouds, and therefore parameterizations are sought to simulate the processes within clouds as an aggregation. But it is the motions and thermodynamics within the individual clouds that actually produce precipitation. In general, rising air produces precipitation, and then condensation of water vapor releases heat which accelerates the rising. To correctly estimate the amount of water and heat released by a cloud one must know the actual speed and humidity of rising air. Therefore the amount of precipitation produced by a group of individual clouds is not the same as that which would be produced by a cloud that had the average properties of the whole region, which includes both cloudy and cloud-free areas. Making an analogy to economics, the total business activity in a society is the sum of the initiatives of the individuals living there, but the net result is not the same as it would be if the society was composed of many identically-behaving "average" individuals. Much current research is devoted to figuring out how to represent complex interactions like these in a way that computers can work with.

Nevertheless, it is my belief that as computers become faster and as our understanding of the physical processes becomes better, we will rely more and more on the dynamical forecasts. They have the tremendous advantage of working forward from the actual present observed conditions, and so avoid the problem of statistically averaging over a number of events that differ in important details. In addition, for low-frequency events like El Niño, it will take decades or centuries to accumulate sufficient realizations to really improve statistical confidence. I suppose I have a personal bias towards this point of view, in that this field offers the opportunity for scientists to make significant progress by advancing the understanding of physical processes within the coupled system, as we have already seen over the past several years.
Back to top

There is a lot of confusion about what El Niño will do this winter. It is true that the media have given it a lot of play, perhaps more than is justified by the uncertainty of the forecasts. Please remember that weather forecasting is not an exact science, particularly when we are forecasting months ahead. However, some things are certain:

* There is a strong El Niño in the tropical Pacific this year.
* Such events "often" lead to heavy winter rains in southern California.

The hard fact is that we do not know very well what will happen in California. We can be more confident about some areas (like the Gulf Coast), where the response to El Niño is pretty robust, but California can go either way so no forecast can be taken too seriously, particularly more than about 3 months in advance. Sorry to be mealy-mouthed, but it's true. There definitely is a good possibility of flooding over the next few months. That's more than a roll of the dice, but it is not certainty. What action you take depends on how you would be affected. It's similar to earthquake preparedness. It makes sense to have your earthquake supplies up to date and handy, but you can't spend your whole life preparing for disaster. One thing can be said reliably: if it's not this El Niño that produces flooding in S. California, it will be another one not too far off.

One of the best forecasts is what we call "persistence". That is, when a pattern is established it tends to remain. If the winter begins rainy, then probably it will continue as such. If there is lots of rain in November that would be a good indication that this El Niño has set up the jet stream to direct moisture there (as opposed to further east), and it would then be more likely to continue.

One of the hardest things for non-specialists to get good information on is the basis behind the various forecasts. A lot of what you hear is based on statistics of past events. But remember, there haven't been very many El Niños since we started being aware that it was a global phenomenon (not just a bunch of unconnected anomalies). There have been 10 since 1950 and only 6 since 1970. That's not a very good basis for statistics, particularly when we observe that different events evolve in different ways. It would be like measuring 5 kids in a classroom. Would that give you a good estimate of the average height of the kids? Maybe you'd happen to get the 5 shortest. So statistical forecasts (noting that El Niño "usually" brings rain to California) are not on a good foundation. How many of those 10 events means "usually"? Then you find that if you just look at the ones since 1970 (observations were extremely spotty before that), then your statistics are probably going to be different. So when you surf the web sites looking for information, you can easily find statistics telling you different things. It doesn't mean they're wrong, or trying to mislead, it just means that we have a very small sample of a highly variable phenomenon.

Another type of forecast is based on a dynamical model of the ocean-atmosphere system. Such a model uses a computer to solve the equations that govern fluid motion and is initialized with the present conditions and then run forward. It has the advantage that is is based on the actual present conditions, not a statistical average, but it has the problem that those conditions are not known with great accuracy. Also, such models are necessarily crude since we can't simulate every molecule. In particular, we can't simulate every cloud, but have a grid structure in which the average cloud amount is estimated at each gridpoint (say 20-50km apart). That's a tricky business, since some effects of clouds are not well simulated by an average cloud. For example, thunderstorms and tropical convection, in addition to dumping a lot of rain, pump a huge amount of heat and moisture into the upper atmosphere in their rising motion. (That is one of the main ways that El Niño affects N America, since that heat and moisture can be carried great distances on the upper winds). But the dynamics of a thunderstorm cloud are specific, and can't be well- resolved with a coarse gridpoint model. Scientists spend a lot of effort trying to make these things approximately right, but the wide spacing between gridpoints is an inherent flaw. (The way we work on these things is to make fine-grid models of single storms, then try to extrapolate those properties to the global models). That is the weakness of the dynamical forecasts. As computers get faster these will improve.

A final thing to remember is that El Niño is not the entire story. Many other oscillations are going on at the same time, so whatever the effects of El Niño, we see them all jumbled up with many other signals. Since we really have only a relatively few years of decent observations, picking these signals apart is a matter of guesswork.

Sorry to not be able to give a more definite answer. In another decade we will probably be at that point (but I realize that doesn't help you this winter).
Back to top

Computer models of the climate system are a sophisticated way of evaulating ideas that are too complicated for the human mind to think through. Perhaps a simple example (that the mind can think through) would help. We know that ocean surface water is heated by the sun during the day and cools off at night. Say the amount of heating is determined just by the length of the day. That means the water would heat more in the summer, since the days are long and the nights short, and cool off more during the winter. We could write these ideas down in equations, specify the values of heating due to various amounts of sunlight, and use a computer to solve them and give us a plot of what the predicted water temperature would be at any time in the future. That's an (overly-simple) computer model of climate.

In reality, of course, we have much more complicated ideas of how the climate system works. For example, there are clouds, and the clouds not only block the sun during the day, cooling off the water, but also tend to insulate it at night, preventing cooling. Do these balance out? Depends on how the amounts of cooling and insulating are specified. Parameters that describe these relationships have to be estimated from observations, then programmed into the model. In addition, clouds are not independent of the water temperature; for example very warm water tends to produce a lot of evaporation, and also rising air above itself and consequently there are tropical rainstorms (similar to the thunderstorms we find occurring over warm summer land). But if there is wind, the clouds may be blown somewhere other than where they were formed. So the pattern of where clouds occur can quickly become extremely complicated. Since we have already specified in our model that the clouds affect the water temperature, that, in turn, means the pattern of water temperature gets more complicated, which feeds back on the cloud pattern, and so on.

Things continue to get more complicated. When air rises over warm water, other air must flow in from the sides to make up the deficit. Therefore if water temperatures are not uniform, there will be wind. When there is wind, this causes ocean currents, which moves water of various temperatures around. If the winds are such as to move the surface water away from some region, colder water from below may be pulled up. Cold water weighs more than warm water, and the difference in density also causes currents to flow.

So I think you can see that computer models of the climate system can start with some fairly simple ideas, but quickly become extremely complicated in practice. The computer power necessary to work through these equations is as large as that in any field of study. As we get more observations, we learn more about the system, and modelers are constantly struggling to represent these processes more accurately. One of the main difficulties is that while we know pretty well how the system will change over a short time (like a day or so), once we ask for longer predictions we come up against the problem that we don't know the initial state perfectly. Therefore there will be some error in the forecast since it won't be starting from exactly where the real system starts from. For a one day forecast, the error probably won't be too great, perhaps some clouds in the wrong place. If we run the model further into the future, soon those wrong clouds will produce erroneous water temperatures, which will produce an even worse cloud pattern, and pretty soon the whole solution is garbage. So there is an inherent limit to the predictability of a system as complicated as the ocean-atmosphere system.

How accurate weather and climate models are depends on how far in advance you want a prediction, and how much detail you want. If I want to know the weather in Seattle for a specific day, then I can get it no more than 3-4 days out. If I only want a general forecast for a season, averaged over a large region, then sometimes our models have useful skill for as much as a few months, particularly if there is a major influence, such as El Niño, that has reasonably-well understood effects. Other times, when there is a more complicated mix of weaker processes, there may be little long-range skill at all.

In general, I don't take seriously forecasts for more than a few months in advance. For example, right now (February 1998) most of the models are suggesting that El Niño will wind down by early summer 1998, and next winter will be a strong La Niña (the opposite phase, in which it is abnormally cold in the tropical Pacific and many, but not all, of the far-field effects of El Niño are reversed). Frankly, I don't have much confidence in a forecast that far ahead. I agree that El Niño is winding down (the signs are obvious in the observations), but I don't think anyone knows what will happen next winter. However, the scientists who make these forecasts do it publically as a means of "ante-ing up" to the forecast competition. Believe me this is a real competition. Everyone wants to be the first to develop a successful El Niño model. And if you don't publish a forecast in advance, you can't claim later that you had it right. So you see a lot of long-range forecasts, but that doesn't mean that anyone, including the authors, necessarily has much confidence in them. Unfortunately, with the media frenzy about El Niño this past year, many of these experimental forecasts were trumpeted around the newspapers and TV shows as if they were truth.
Back to top

The reason you won't find much information connecting El Niño and global warming is that we (meaning the mainstream scientific community) don't really have too much useful to say about it at this point. While we know that El Niño occurs on the background of the large-scale climate, and assume that as the background changes that some aspects of El Niño might also change, we are nowhere near the ability to say what those changes might be. So rather than speculate about such a politically-charged subject, we usually keep our mouths shut.

The above cautionary note does NOT, however, mean that one should discount the possibility. Since we see that El Niños in different years vary greatly in their strength it appears the process may be quite sensitive to changes in the background state.

Much of the uncertainty in the question of whether greenhouse warming is affecting the ENSO cycle revolves around the problem of how one would measure the statistical significance of changes in recent El Niños. Some say that the string of warm events during the 1990s are evidence that a general warming trend is starting to change the weather; others say that these variations are within normal limits. The fact is we have only a few events to talk about, which means there is no statistical rigor to any argument for or against this idea. It is simply shooting the breeze. We won't have good statistics about El Niño for another hundred years or so (perhaps even longer if it is truly chaotic), so I don't bother with such arguments at all. To me the interesting stuff is the dynamics and thermodynamics anyway, and on that front we stand a chance of making progress in my lifetime.

One way that a general global warming could affect El Niño is through the sensitivity of tropical deep convection (organized thunderstorms involving strong upward motion over large regions) to the water temperature beneath. Convection occurs only over warm water, which provides both moisture and the heat to produce strongly rising air. But convection does not increase at a steady rate as water warms; at around 27.5°C (81.5°F) it suddenly becomes much more efficient, and the warmer-than-27.5°C areas are often identifiable in satellite images by the strong development of rainstorms penetrating high into the atmosphere. Huge amounts of heat and moisture are pumped out of the tropical ocean into the upper atmosphere, and that energy source is one of the main engines of the global weather systems. Perhaps the primary way that effects of El Niño spread to extra-tropical regions is through increasing the area and location of tropical convection (with its consequent effects on the jet streams) as the warm water extends further east than usual (see Question 1 for more discussion of tropical convection).

Since there are large regions of water with temperature close to 27.5°C (either slightly warmer or slightly cooler), even a small general temperature change could produce large changes in the area in which convection takes place. The area of the tropical Pacific that has mean SST greater than 27°C is 20% larger than that greater than 27.5°C, for example. A mere one-half degree uniform increase could therefore produce a significant change in the amount of tropical convection, perhaps making the normal state appear more like El Niño. (Of course the changes probably would not be uniform, but the example shows how a nonlinear system can be very sensitive to small changes of the input functions). Tropical convection thus has the potential to amplify the disruption of global climate, perhaps far out of proportion to a seemingly small temperature change. So, while the values of temperature rise proposed in the global warming debate appear small, in the sense that in daily life a 1-2 degree change hardly matters, the danger lies in the reorganization of the large nonlinear systems that regulate the entire weather machine. There may well be feedbacks that would damp out some of these influences on the tropical convection process (or possibly amplify them even more). I don't know and I don't think anyone does. But no one should argue that because the overall values of temperature change seem small, or that a computer model of the effects of greenhouse warming may be wrong by a significant percentage, the effect on human society could not be drastic.

The connection could also go the other way, although this is not often considered. The forest fires due to El Niño occurring these past few months in the Amazon and Indonesia are contributing strongly to the increase of C02 in the atmosphere, and also reducing the forest cover that absorbs C02. Therefore El Niño appears to be part of the problem of greenhouse warming.

OK, so now you're wondering why I didn't take my own advice at the start of this section and keep my mouth shut. Sorry.
Back to top

The most widely used scale is known as the Southern Oscillation Index (SOI), which based on the surface (atmospheric) pressure difference between Tahiti and Darwin, Australia. It was noted as far back as the 1920s that these two stations were anticorrelated, so that when Tahiti pressure is high, Darwin pressure is low. This reflects the very large scale of the phenomena, since one would not usually expect such a close relation between such faraway places. When Tahiti pressure is high, that indicates winds blowing towards the west (normal trade winds), and when it is low winds blow to the east (El Niño).

We use the Southern Oscillation Index for three main reasons, even though it is an indirect measure of El Niño, and these locations are not ideally sited for this purpose. First, the time series at Darwin and Tahiti are more than 100 years long, and there is no other record that would allow us to categorize the El Niño cycle that far back. Second, the measurement of atmospheric pressure is simple (it is just the height of a column of mercury in a barometer) and not subject to calibration problems, as, for example, thermometers are. A column of mercury is just a measure of length, which is accurate and easily convertible among inches or centimeters, whereas thermometers are inherently less accurate since they rely on a carefully-made glass tube than may be subject to expansion or irregularities; also the placement of the thermometer (in the sun or shade or breeze, near buildings, etc) can have a large effect. (For example, temperature measured in cities shows a longterm rise associated with the heat generated by urban activities and the increased absorption of solar heat by pavement compared with forest. This is one of the things that makes it hard to detect the signature of greenhouse warming). Therefore, pressure measurements are highly desirable for interpreting long records. Third, pressure tends to be similar over wide regions, whereas more directly-important quantities (SST and winds) can have many local effects that make it hard to interpret single point measurements as representative of the large-scale situation.

The SOI is given in normalized units of standard deviation. It can be used as an intensity scale. For example SOI values for the 1982-83 El Niño were about 3.5 standard deviations, so by this measure that event was roughly twice as strong as the 1991-92 El Niño which measured only about 1.75 in SOI units. By this standard, the present El Niño is about as strong as 1991-92. However, the sea surface temperature anomaly now (November 1997) is about as large as in 1982-83, and some might say that is a more important measure. This shows that there is no single number that summarizes the intensity of events.
See a plot of the SOI since 1950.
And the full record since 1876.
Back to top

For example, here's a sea surface temperature plot showing the pointy wedge (1 Dec 1997)
And for reference, here's yesterday's sea surface temperatures

You see the westward-pointing wedge in plots of anomalies, not in plots of the actual temperatures (compare the top (actual) and bottom (anomalies) panels of the first plot above).

Anomalies mean the normal temperatures in each location have been subtracted from the observed values at the time of the plot. An anomaly plot therefore shows whether the water is warmer or cooler than its normal state, and the normal state is different in different places. For example, if on some (very unusual!) day the temperature was 20°C (68°F) everywhere in the ocean, that would be normal along the coast of Baja California, but anomalously warm here in Seattle, and anomalously cold in the Philippines. An anomaly map for this hypothetical flat 20°C day would show positive values in mid-latitudes and negative values in the tropics.

In the tropical Pacific, normal sea surface temperatures (SSTs) are much colder in the east than in the central/western Pacific (say 23°C (73.4°F) in the east vs 29°C (84.2°F) in the west). (See a plot of the mean SST in the tropical Pacific). Further, the cold water in the east is concentrated in a band along the equator. That may sound strange, since we usually think of the equator as warm, but upwelling of deeper (hence colder) water occurs on the equator in the east, resulting in the cool water found there.

During El Niño, water of about 28°C is found across a huge region from the central Pacific along the equator to South America (See a plot of SST during the height of El Niño (1 Dec 1997)). In terms of anomalies, this is near-normal in the central Pacific, but gets progressively more anomalous along the equator to the east. That's why there appears to be a wedge.
Back to top

This question does not have a simple or straightforward answer, since this is not a settled issue. Fundamentally we are not exactly sure why the Pacific should have an El Niño/La Niña cycle and the Atlantic not. We observe that this is the case and can think of reasons why this makes sense, but if we had no observations to tell us the answer in advance I don't think we would deduce it from theory. For example, if earth did not have an Atlantic or Pacific, and we were studying a new planet with such features for the first time, I doubt that anyone would predict such a distinction. It's always easier to come up with a theory when you know what the answer has to be! Please take my answer below in this spirit.

A principal difference between the Atlantic and Pacific is the width of the equatorial region. The Pacific is more than twice as wide. This is important to their capacity to sustain an El Niño/La Niña cycle because of the peculiar dynamics of equatorial waves. Equatorial waves are not the familiar surf or swell seen on the surface, but very largescale motions that carry changes in currents and temperature over thousands of kilometers. The period of these waves is measured in months, and they take typically 3 months to more than a year to cross the Pacific. Surprisingly, these waves do not spread out equally in all directions like waves made by dropping a rock in a lake, but preferentially propagate eastward or westward. When winds blow over a large area of the ocean consistently for a month or more, equatorial waves are usually generated, and these then modify conditions over a very large region, including places far removed from where they were generated. For example winds over the far western Pacific make waves that carry the signal to the coast of South America, even though the winds in the South American region may not change at all. The subsurface changes due to the arriving waves can then cause sea surface temperature changes, entirely due to winds occurring many thousands of kilometers to the west.

With the huge distances across the Pacific, one side of the ocean can be reacting to conditions due to one set of waves, while the other can be doing something completely different. As the waves propagate back and forth, a cycle can be set up that oscillates (El Niño/La Niña). The much smaller Atlantic, on the other hand, is not large enough to sustain much of an oscillation, since the waves cross it so quickly, often in only a month or so. This does not allow a cross-ocean contrast to be created, nor an oscillation to be set up. Some indications suggest that some kind of weak oscillation may in fact occur in the Atlantic, but it never gets to the amplitude of that in the Pacific.

A second reason that the Pacific is more important in this regard is that the fundamental driver of the whole ocean-atmosphere circulation is heat. The large width across the Pacific allows the existence of a huge pool of warm water in the west (see Question 1 for reasons why the warm pool is created). The smaller distances across the Atlantic mean that the Atlantic warm pool is very much smaller. The Pacific warm pool is a gigantic source of heat that is one of the main controls on the atmosphere. When the warm pool shifts east (during El Niño) or shrinks west (during La Niña), the effects reverberate around the world, causing the weather disruptions associated with this cycle. In the Atlantic, there is simply not enough of a warm pool to make that much difference to worldwide weather. So even if there is an analogue to El Niño in the Atlantic, it has much less power to cause weather disturbances that affect more than local conditions.
Back to top

The ocean is salty because of the weathering of rocks in the early stages of the formation of the oceans and continents. The ocean probably is continuing to become saltier very slowly (on geologic timescales) as rivers bring dissolved material to the sea. Even though the concentrations of dissolved salts in "fresh" river water is tiny, evaporation from the ocean surface (which is how water gets back to the rivers) removes only pure water so whatever salt enters from rivers remains in the sea and slowly builds up.

Seawater is about 3.5% salt by weight, which means the total quantity of dissolved salt, if dried, would form a layer 45m thick over the entire earth, or 153m thick over the present land area.

The salt in seawater is composed of a variety of components, primarily sodium chloride (table salt), magnesium chloride, magnesium sulfate and calcium carbonate, along with many trace elements. One of the remarkable aspects of ocean salinity is how constant the relative ratios of these constituents are in seawater, no matter where in the oceans one looks. This suggests that the oceans are "well-mixed" on the timescale of salt input. Total salinity, however, varies from about 3.3% to 3.8% in open-ocean surface waters, depending on the rainfall/evaporation/river runoff in different regions. Salinity in bays or near the mouths of rivers is less, for example the salinity of Puget Sound near Seattle is about 2.9-3.2%. Deep-ocean salinity is much less variable.

It should also be noted that a fascinating controversy is now unfolding, regarding the hypothesis that about 10 million small water-rich comet-like objects, each about the size of a small house and weighing about 20-40 tons, bombard the earth annually, depositing water vapor into the upper atmosphere. The protagonists of this theory suggest that this process may have introduced sufficient water into the earth system to fill the oceans over geological time. The evidence was first presented by Frank and Sigwarth of the Univ. of Iowa from images taken by NASA's Polar spacecraft. It is a highly controversial idea about which I can offer no guidance. But if the hypothesis proves viable, it will fundamentally change the ways we view planetary accretion, the origin and evolution of the ocean, and the interplay between the solid earth and the oceans. You can read more about this on Dr Frank's web page.
Back to top

There have been two major developments: First is the much better satellite coverage. Second is the development (by my laboratory) of a simple, inexpensive design for buoys that can be deployed in mid-ocean and remain active for a year or more. Previously, most buoys were placed only in shallow coastal waters or for short times in deep water. Our lab was the first (in the early 1980s) to develop the technology to place and maintain a large network of buoys in mid-ocean for a full year or more. We maintain the TAO (Tropical Atmosphere Ocean) buoy array across the equatorial Pacific that measures SST, surface winds, air temperature and humidity, and subsurface temperature (sometimes currents, too) in the upper 500m. Check the buoy page link for access to the real-time buoy data.

These buoys transmit data daily (in some cases hourly) to the weather forecast network and for research and prediction. The big advantage of moored buoys is their high temporal resolution. Also, knowledge of subsurface processes is necessary to understand and predict how the SST is likely to change over the next few weeks and months. Finally, there is the need for ground-truthing AVHRR (satellite SST) values. SST products are usually a blend of satellite AVHRR ground-truthed by means of buoy and other data. AVHRR by itself can be biased because of unknown amounts of water vapor in the atmospheric column that affects the radar returns. (The AVHRR must be corrected based on an estimate of the atmospheric water, which affects the SST measurement, and many AVHRR satellites have instruments to make these estimates). But where there are buoys, as in the equatorial Pacific, the AVHRR estimate can be fixed by the buoy SST, and this is a better check than a satellite estimate of water vapor.
Back to top

First, there is no question that El Niño has serious effects on marine life in many regions. You can read about some of these through links on our El Niño impacts page.

There is a lot of stuff on that page, but in particular there are links to "Effects of El Niņo on Corals and fish and birds of the Pacific. Also, about 3/4 of the way down the impacts page are several links under the section "West coast of US, Canada, Alaska, Hawaii" that discuss effects on various kinds of marine life.
Some of these links are:

• Impacts on the Pacific Northwest from University of Washington Snowpack, streamflow, marine ecosystems, etc.
• Biological effects of El Niño - from Environment Canada
• El Niño impacts on marine mammels in the California Channel Islands
• El Niņo and Salmon in the Northeast Pacific Ocean
• El Niño forecasts could help save coho salmon

Since normal variations can have large swings, such as occur during El Niño, the boundaries of where particular forms of life can survive are somewhat smaller than they would be if the climate were more constant. For example, palm trees can survive in Seattle (my home town) during most years, since we have generally mild winters. But every 10 years or so we'll have a killing frost, so natural palms do not occur here. But many people (including myself) grow them in gardens and find that they thrive with only occasional protection.
Back to top

I think the reason for el-nino is the large difference in salinity near the Glaciers. I have a strange feeling that the sea beavers are playing havoc with us. There might be high sand dunes at the ocean bed and these might be affecting the ocean flow and hence the difference in temperatures. As most people think the el-nino is not due to the weakening of the trade winds.

Where can I get some references to study the life style of sea beavers?
Does any organisation measure the ocean salinity near the glaciers?

Back to top