Guest Post by Willis Eschenbach
I’ve said before that I consider myself a climate heretic rather than a climate skeptic. A skeptic doubts parts of things. A heretic questions the basic assumptions underlying the whole field. My heresy is that I do not think that temperature is a linear function of greenhouse gas forcing. I think that the temperature is regulated by a variety of “emergent” phenomena. In my last post, Watts Available, I discussed my view of how thunderstorms act to cap the tropical Pacific temperature.
In addition to my underlying heresy, I consider the various climate phenomena in what some people think is a backwards or improper manner. I’m not so much interested in how these phenomena work. Instead, I’m much more interested in what they do when they do work. I was told that this is called a “functional” analysis of a situation, which makes sense—I’m looking to see what the function of a phenomenon is.
Doing a functional style of analysis leads me to notice things like how thunderstorms operate on the exact same principle as your household refrigerator. And it also leads to my heretical view of another emergent temperature regulatory phenomenon, what’s commonly called the “El Nino”.
Let’s take a leisurely perambulation through the work of the El Nino/La Nina interchange, and consider what are known as “emergent” phenomena.
Let me start by discussing emergence and the class of phenomena known as “emergent”. Here are the defining characteristics of emergent phenomena.
- Emergent phenomena appear, often very quickly, out of what might be termed a “featureless background”. For example, a day in the tropical oceans typically dawns clear. The clear air usually persists until late morning, when suddenly and without warning, out of the featureless blue sky background, puffy white cumulus clouds rapidly form and cover half the sky. These cumulus clouds are an emergent phenomenon.
- Generally, emergent phenomena are not what might be termed naively or obviously predictable before emergence. For example, suppose you’d lived your entire life in tropical clear blue morning skies without ever seeing or knowing anything about clouds. There’s no way you’d look up and say “Y’know what? I think a whole bunch of giant white bulbous masses could suddenly appear way up in the sky!”. People would call you crazy.
- Next, emergent phenomena are generally not permanent. For example, the tropical cumulus clouds above typically dissipate before dawn. Emergent phenomena usually have an emergence time, a lifespan, and a dissipation time.
- Emergent phenomena are often, but far from always, associated with a phase change. For example, the clouds mentioned above are associated with condensation, which is a phase change of water from a vapor in the air to tiny liquid droplets in the clouds.
- Emergent phenomena are often mobile and wander around the landscape. An excellent example of this kind of emergent phenomenon are the familiar “dust devils” that are frequently seen moving across dry hot landscapes.
- Emergent phenomena involve flow systems far from equilibrium.
- Emergent phenomena, as the name states, emerge spontaneously when the conditions are right.
- The conditions for that emergence are often threshold-based. Once the threshold is crossed, many individual examples of the phenomenon may appear quickly. This is true, for example, of the tropical cumulus clouds discussed above. Once the morning gets warm enough, and some local temperature threshold is passed, a skyful of cumulus clouds rapidly materializes out of nowhere.
- Emergent phenomena are generally not cyclical. They don’t repeat or move about in any predictable fashion. It’s why the predictions of the emergent phenomenon known as tropical cyclones have “cones” rather than a single line.
From the smallest to the largest, the emergent phenomena that I think work together to regulate the global temperature include:
• Dust devils
• Rayleigh-Benard circulation of both the atmosphere and the ocean
• Daily cumulus cloud fields
• Tropical (convective) thunderstorms
• Squall lines and other thunderstorm aggregations
• Tropical cyclones
• The El Nino/La Nina alteration, discussed in this post
• Ocean-wide circulation shifts such as the Pacific Decadal Oscillation (PDO), Atlantic Multidecal Oscillation (AMO) and the like.
All of these are thermoregulatory emergent phenomena. When the local temperature exceeds a certain level, they emerge and cool the surface down in a wide variety of ways.
With that discussion of emergent phenomena as prologue, let’s look at what’s happening in the Pacific. Here’s a movie of the month-by-month sea surface temperatures (SSTs). Take particular notice of the tongue of cooler water that extends out a variable distance offshore from South America along the Equator.
Figure 1. Month by month temperature variations, Reynolds Optimally Interpolated sea surface temperature dataset. The blue box shows the “NINO34” area, from 5°N to 5°S, and from 170°W to 120°W
So where in all of that endless motion is the El Nino and the La Nina? Here’s a drawing from NOAA showing the normal Pacific conditions.
Figure 2. ORIGINAL CAPTION The map (top surface) shows the Pacific Ocean from the Americas (brown area, right edge) to Australia (brown area, left edge) Graphic shows the sea surface temperature (colors from blue to red for cold to hot), the atmospheric circulation (black arrows), the ocean current (white arrows), and the “thermocline” (blue subsurface sheet). The thermocline is the bottom of the mixed layer—above the thermocline, the ocean is mixed on a regular basis, and below it there is little mixing. As a result, the water above the thermocline is warmer, often much warmer, than the water below the thermocline.
At times, however, the heat piles up in the eastern Pacific near the Americas. In this case both the atmospheric and oceanic circulation changes, as shown in Figure 3. The thermocline deepens, with warmer water near the coast of the Americas.
Figure 3. El Nino conditions. The surface near the Americas is warmer. The thermocline off the coast of the Americas is deeper.
Moving from theory to measurement, here is the sea surface temperature (Figure 4) as well as the sea surface temperature anomaly (seasonal variations removed, Figure 5) during an actual El Nino.
Figure 4. Actual sea surface temperature during the peak month (November) of the large El Nino of 1997-1998. Note the high temperature of the water in the blue rectangle that outlines the NINO34 area. Temperature in this area is diagnostic of the state of the El Nino/La Nina alteration.
Figure 5. Sea surface temperature anomaly (seasonal variations removed) during the peak month (November) of the large El Nino of 1997-1998. This show the large heat buildup along the equator in the eastern Pacific near the Americas.
After an El Nino condition peaks, a strong trade wind starts blowing towards Asia. This blows the warm surface water towards Asia, to the point where the thermocline comes all the way to the surface off the coast of the Americas. When the warm water hits the coast of Asia, it splits in two. One part goes towards the Arctic, and the other goes towards the Antarctic. Here’s the NOAA graphic showing La Nina conditions.
Figure 6. Schematic diagram of the La Nina Condition.
And as above, shown below is an actual La Nina condition. This is the La Nina peak of the same Nino/Nina cycle in Figure 5, which began 12 months earlier, in November 1997.
Figure 7. Reynolds Optimally Interpolated (OI) sea surface temperature for November 1998.
And here is the temperature anomaly at that time:
Figure 8. SST Anomaly (seasonal variations removed) during a peak La Nina condition.
Note in Figure 8 above how the trade winds have exposed the cooler subsurface waters all across the equatorial Pacific. They’ve been laid bare because the warm water has been pushed westwards. You can see above how when the warm water hits Asia/Australia, it is mostly split in two and moves towards the poles.
Now I started this by saying I do functional analysis. I don’t look at what causes the El Ninos or the La Ninas. I’m not trying to understand the processes. Instead, I look at what they do.
When I do that, I see that talking about the El Nino and the La Nina as separate phenomena is incorrect. They function together as the world’s largest pump. What they do is pump trillions of tonnes of warm equatorial Pacific water polewards. So much water is pumped that the elevation of the equatorial Pacific sea surface drops, and the effect is visible in local tide gauges.
Figure 9. The Nino/Nina differences as shown by the TAU/TRITON moored buoys along the Equator. You are looking westward, across the equator in the Pacific Ocean, from a vantage point somewhere in the Andes Mountains in South America. The colored surfaces show TAO/TRITON ocean temperatures. The top surface is the sea-surface, from 8°N to 8°S and from 137°E to 95°W. The shape of the sea surface is determined by TAO/TRITON Dynamic Height data. The wide vertical surface is at 8°S and extends to 500 meters depth. The narrower vertical surface is at 95°W. All of these data come from the TAO/TRITON Array of moored ocean buoys in the Equatorial Pacific.
So … what happens when warm ocean water gets transported towards the poles? More heat is lost to space. Figure 10 shows how much upwelling surface radiation makes it to space, by latitude.
Figure 10. Top Of Atmosphere (TOA) upwelling longwave radiation as a percentage of the upwelling surface radiation, by 1° latitude band. These are monthly averages over the entire period of record.
In Figure 10 above, the low spot at about 7°N or so is the location of the ITCZ, the intertropical convergence zone. When you start moving towards either pole, there’s an immediate and continuing increase in the percentage of the surface thermal radiation that escapes to space.
Now given the functional nature of my analysis, I make a different identification of “El Nino” and “La Nina” than the one normally given.
There are several indices used to evaluate the El Nino/La Nina conditions. An example of an index is that “El Nino conditions” are times when the sea surface temperature (SST) anomaly in the NINO34 Region (blue box) is more than a certain temperature (often around 1°C) warmer than normal. And “La Nina conditions” are when they are more than a degree cooler than normal in the NINO34 region. (There are other identifications, but they all identify Nino and Nina conditions separately, and they all establish a temperature threshold for Nino and Nina conditions. I do neither.)
I don’t look at them separately, or have any set temperatures. This is because I don’t see them as separate phenomena.
Unlike the standard definitions, I identify the Nino/Nina phenomenon as working together as a pump. In that pump the El Nino is the peak of the intake stroke, and the La Nina is the peak of the discharge stroke. We can see this activity in a graph of the temperature in the NINO34 region (blue rectangle shown in the graphs above).
Figure 11. Sea surface temperature in the NINO34 area. Blue sections show the times when the pumping action is occurring. Red dots show peak El Nino conditions, and blue dots show peak La Nina conditions. Dotted vertical white lines show November of each year.
I have highlighted in blue the times of the pumping action. What I first noticed about them is what the Peruvians noted about them. This is that they all start within a month or so of November, and thus are often strong near Christmas … hence the name “El Nino” for the boy-child.
I noticed another oddity. In all cases highlighted , the duration of the pumping action from the red dot at the top (peak “El Nino”) to the blue dot at the bottom (peak “La Nina”) is one year plus or minus a month or so. This allows us to distinguish the Nino/Nina pumping action from the normal temperature variations found anywhere in nature.
The regular ~ 12-month length of the discharge cycle also shows that the two (El Nino and La Nina) do not exist as independent entities. Instead, they are intimately tied in a single larger one-year-long phenomenon.
Now, recall that the question in functional analysis is, what does this single larger combined Nino/Nina phenomenon do?
I say that the El Nino/La Nina pump is an emergent phenomenon with a lifespan of 12 months. It emerges when enough heat is built up in the eastern equatorial Pacific Ocean. It cools the equatorial Pacific, and hence the entire planet, by
1) exporting the warm equatorial surface waters polewards where the heat is lost more quickly to space, and by
2) exposing the cooler subsurface ocean layer which cools the atmosphere.
So the function of the El Nino/La Nina alteration is to cool the earth by means of a periodic pumping cycle.
Like many other emergent climate phenomena, it is what I call “self-latching”. By this I mean that once the Nino/Nina pump starts, it creates conditions such that it strengthens itself, and thus tends to persist.
Here’s how that works. The strength of the trade winds in the equatorial Pacific is driven by the east-west temperature difference. Now, when the pumping action is underway, the east is getting cooler, and the warm water is piling up in the west. This increases the east-west temperature difference, which in turn increases the east–>west wind strength, which in turn increases the temperature difference, which …
This makes it self-latching, and this positive feedback is responsible for the long duration of the phenomenon once initiated. Once the Nino/Nina phenomenon begins, it generates its own wind. This allows it to continue to runs until the cold water is exposed all along the equator, as you can see in Figure 8 above.
Predictions and Conclusions
Now, any theory such as mine is only as good as its predictions. So how can I determine if the Nino/Nina is actually an emergent phenomena that cools the Pacific when excess heat builds up?
Well … we could start with the observation that the trigger for the pumping is the buildup of heat in the eastern Pacific. So the form of the phenomenon obviously is temperature-limiting (cooling) and thermal-threshold based (happens more when it’s warmer).
What I’d never figured out until this analysis was how to determine whether the Nino/Nina pumping phenomenon as a whole was more frequent or more powerful or both in warmer times than in cooler times. The problem is that we already know that it is triggered by excess heat … but does it increase when the excess heat increases? And how would you measure that increase?
What I realized is that if the pumping increases in warmer times, such as the current post-1981 Reynolds SST record of a gradual slight overall ocean surface warming, we should see differential heating trends in the Pacific.
And what the pattern of larger and smaller trends should look like is what it looks like after a complete pumping cycle—the areas on the way to the pole should show warmer trends, and the eastern Pacific should be cooler. If there is an increase in the number of Nino/Nina cycles, the transfer of energy will show up in the trend. The trend should be smaller in the area along the Equator where the pump exposes cooler water, and the trend should be larger where the pump moves the warm water, which is westwards and towards the poles.
Here’s how that played out at the end of the large 1997-1998 Nino/Nina cycle. I repeat Figure 8 from above to here for comparison.
This is Figure 8 from above.
And here are the sea surface trends during a 36-year period when, as the figure below shows, there’s been a slight SST warming (0.10°C per decade).
Figure 12. Decadal sea surface temperature trends.
My conclusion from the distinct similarity of those last two graphs is that the prediction from my theory is correct—the Nino/Nina pump is indeed a temperature-regulating emergent phenomenon, which opposes any increase in overall tropical Pacific temperature.
Here, the first rain came yesterday after a long dry “fiery but mostly peaceful” wildfire season here … the forest smells of growth, green life, and decay are particularly strong this evening, reminding me of the endless cycles of creation and destruction.
My best regards to everyone,
Post Scriptum: When you comment please quote the exact words that you are discussing. This helps avoid the endless misconceptions that plague the internet. For more info regarding how you can show that I’m wrong, see my post Agreeing To Disagree.