
Report 1 on El Niño presented a few features of the coastal upwelling along the west coast of the Americas, a gentle introduction to a complex global shift in atmospheric and oceanic circulations and to the local effect called El Niño. In this report we will continue that discussion by describing the large circulation shift called the Southern Oscillation and how that relates to El Niño so that the entire ocean-wide phenomenon is called “El Niño/Southern Oscillation” or ENSO.
Imagine a bathtub

Before we start this discussion let us try a thought exercise to help vision the features of ENSO. Imagine the entire Pacific Ocean is in your bathtub (assuming you are one of the lucky people in the world who has a bathtub). At one end of the tub we place a powerful fan, like the one used to simulate wind in the movie “The Perfect Storm.” That had to be a big fan, right? Now run the fan at full force so that, at the far end of the tub, water is piled up, and at the upwind, fan end the water level is down. The friction of the wind on the surface pushes the water down wind so there is a slope. As long as the fan roars away the water level has the upward slope and the system is in equilibrium, another word for steady.
Do you have that vision of the bathtub with water set up at the far end? Now, shut off the fan. What happens? The surface water runs down hill and the surface becomes level. This, to the first order of simplicity, describes El Niño-Southern Oscillation, or for short, ENSO. Of course reality is much more complex than water in a bathtub.
Now, think of the bathtub as the equatorial Pacific Ocean, including Peru, Central America, New Guinea, and Northern Australia. The wind machine becomes the trade winds which force the surface waters toward the equator and, more importantly, pushes them toward the west. The result is that, under normal conditions, the western side of the Pacific is around 4 meters higher than the coast of South America.
As we have said many times, El Niño is not an isolated weather event, local to Peru and South America. It is not even localized to the North and South American continents. Rather, it is a re-adjustment to the global ocean and atmospheric circulations whereby weather patterns are disrupted, often in extreme ways. In this report I want to describe a few of the major global patterns in an attempt to give a clue, just a clue, to the magical complexity of this regular event.
Reader, stay with me now. A linked chain defines the most important features in oceanic circulation: The figure above shows the Pacific Ocean circulation features for normal conditions. (a) Solar heating: Solar radiation at the equator is intense and heats the sea surface. (b) Convection: The sea surface heats the atmosphere. Hot air rises and therefore, along the equator, the air rises high into the atmosphere. (c) Equatorial convergence: Low level air flows toward the equator to replace the rising air. (d) Trade winds: The equatorward flow is turned westward by the rotation of the Earth. (e) Mid-ocean high cell: Moist air rises from the equator, rains out all of its moisture, and sinks as dry air at around 30 degrees north and south of the equator. Large oceanic cells of high pressure with circular air flow are formed. (Clockwise flow in the northern hemisphere and counter-clockwise in the southern hemisphere.) (f) Upwelling regions: The northern hemisphere high cell drives northerly coastal winds along the west coast of North America. Ocean currents move to the right of the wind (see the previous report) which creates coastal upwelling. Along Peru, the trade winds cause southerly coastal winds (from the south) which drive offshore currents and create a trade wind driven upwelling. (g) Warm Pool: Intense solar heating, the trade winds, and resulting ocean circulations form a massive region of very warm sea surface temperature (SST) in the tropical western Pacific Ocean. This is one of the most important features in global atmospheric dynamics; we will discuss it further in the next section.
Figure 1. A NASA derived map of the global sea surface temperature (SST) for July of 2001. This was a normal, e.g. non-El Niño/Niña year. Several features, which are discussed in this report, are shown. The trade winds deliver a broad westward force across the equatorial Pacific. The trades and associated ocean circulations contribute to the massive pool of warm water in the western Pacific, the “warm pool,” and the coastal winds along North and South America lead to the upwelling regions, notably on the California and Peruvian coasts. (Click on the map to bring up a high resolution version.) Image credit: NASA
The Warm Pool, the firebox for the atmospheric engine
Figure 1 shows the features we discussed above on a real satellite-derived map of global sea surface temperature. The yellow color denotes very warm surface water and the tropical western Pacific warm pool region stands out as a huge region. The official definition of the warm pool is water that is above 28 degrees Centigrade (about 82 deg F). Typically the size of the warm pool is greater than the United States.
Inside the warm pool convective clouds, cumulo-nimbus towers, extend to the stratosphere where their tops are sheared off by the strong westerly (from the west) winds above. The rising air, having come across thousands of miles of ocean, is full of water which falls from the clouds. The rainfall in the warm pool is typically 3-5 meters (9-15 feet) per year making this one of the rainiest places on the Earth.
Exactly how and why the warm pool forms as intensely as it does is not well understood. The physics of the warm pool is still a subject of research and debate. But scientists all agree that if one thinks of the atmosphere as a steam heat engine (hot, moist air rising and cooler, dry air falling) then the warm pool would be the “hot box” for the entire world’s atmosphere.
Now, turn off the trade winds
Think back to the bath tub analogy. When the fan stops blowing, the water level in the tub relaxes back toward level. This is what happens in the Pacific in an event called the Southern Oscillation. The Southern Oscillation has a period of from 3-6 years. It is called an oscillation because the winds cycle in a back-and-forth way. On one part of the oscillation, the trade winds drop considerably. Sometimes the wind direction changes from easterly (from the east) to westerly in what is called a westerly wind burst because it tends to come on suddenly.
When the trade winds drop, the ocean surface across the entire ocean begins a slow process of adjustment, whereby the sea level changes (down in the west, up in the east) as vast amounts of hot equatorial water flow eastward, toward Peru. The process is very slow because the surface layer is thin, and the distances are thousands of miles. Gyroscopic forces from the rotating Earth focus the flow onto the equator.
The warm pool is part of this adjustment, and it is carried eastward into the center of the Pacific Ocean. It carries its rainfall with it, and desert islands in the central pacific, such as the Galapagos, experience huge rainfall increases.
Figure 2. A map of the SST anomaly (difference of current SST from a 30-year average) during the peak of hurricane season, August-September-October, for 1987. When discussing a quantity such as rainfall or sea surface temperature as they relate to climate, it is much easier and more forceful to consider the anomaly rather than the actual quantity. We say rainfall is 6 cm above “normal,” or that daily maximum temperature is above the long-term average temperature. In discussions on ENSO the anomaly of a quantity is the difference between quantity and the average over the last thirty years (or some such reasonable long length of time). In the above figure for SST, the white areas are not statistically different than the long term average, e.g. no change. Yellow and red areas are warmer water and blue regions are colder. The scale shows the amount of change from the long term average. The Niño 3.4 region, part of the official definition of El Niño is shown. Image credit: NOAA/ESRL.
A Definition of El Niño
A question one might ask is how do we know that an El Niño is here? What temperature, sea level, or other indicator is needed to say, officially, that the ENSO is happening? Or how do we know that it has gone away? The official definition of El Niño was proposed by Kevin E. Trenberth (one of our leading researchers in climate change) of the National Center for Atmospheric Research as follows:
If a five-month running mean of the surface temperature anomaly in the Niño 3.4 region (5N-5S, 120-170W) exceeds 0.4 degC for 6 months or more.
Kevin E. Trenberth, “The definition of El Niño”, Bulletin of the American Meteorological Society, Vol 78 (12), pages 2771-2777, December 1997
The Niño 3.4 region is shown in figure 2. The anomaly at this time was well above the threshold. At the other end of the Southern Oscillation, when the trades are strong, the situation is called La Niña, or the baby girl.

Figure 3. A recent global map of the SST anomaly for the month of March 2010. By the definition above, El Niño is still in progress though diminishing. The most warming at this time is in the central Pacific. Image credit: NOAA/NCEP.
Hurricanes, El Niño, and Modiki El Niño
Not all El Niño events are created equal when it comes to their impact on Atlantic hurricane activity. Differences in El Niño events are described by Dr. Jeff Masters. The global effect of an El Niño depends on how far the warming travels across the Pacific Ocean. If the strong warming travels all the way across the Pacific to Peru, the pattern is called an Eastern Pacific Warming (EPW) pattern. EPW conditions occurred, most recently, during the El Niño years of 1997, 1987, and 1982.
In contrast, more warming occurred in the Central Pacific during the El Niño years of 2004, 2002, 1994, and 1991. The warm pool migration seems to stall in the central Pacific. This pattern is called Central Pacific Warming (CPW).
A recent paper published in the journal “Science” attempts to explain why some El Niño years see high Atlantic hurricane activity. “Impact of Shifting Patterns of Pacific Ocean Warming on North Atlantic Tropical Cyclones”, by Georgia Tech researchers Hye-Mi Kim, Peter Webster, and Judith Curry, theorizes that Atlantic hurricane activity is sensitive to exactly where in the Pacific Ocean El Niño warming occurs.
Over the past 150 years, hurricane damage has averaged $800 million/year in El Niño years and double that during La Niña years. The abnormal warming of the equatorial Eastern Pacific ocean waters in most El Niño events creates an atmospheric circulation pattern that brings strong upper-level winds over the Atlantic, creating high wind shear conditions unfavorable for hurricanes. Yet some El Niño years, like 2004, don’t fit this pattern. Residents of Florida and the Gulf Coast will not soon forget the four major hurricanes that pounded them in 2004–Ivan, Frances, Jeanne, and Charley. Overall, the 15 named storms, 9 hurricanes, and 6 intense hurricanes of the hyperactive hurricane season of 2004 killed over 3000 people–mostly in Haiti, thanks to Hurricane Jeanne–and cost $40 billion in damages.
During EPW years, when the warming occurs primarily in the Eastern Pacific, near the coast of South America, the resulting atmospheric circulation pattern creates very high wind shear over the tropical Atlantic, resulting in fewer hurricanes.
On the other hand, CPW years had lower wind shear over the Atlantic, and thus featured higher hurricane activity than is typical for an El Niño year. One of the paper’s authors, Professor Peter J. Webster, said the variant Central Pacific Warming (CPW) El Niño pattern was discovered in the 1980s by Japanese and Korean researchers, who dubbed it modiki El Niño. Modiki is the Japanese word for “similar, but different.”
Michael Reynolds, Ph.D., RMR Company










