I’ve moved to WordPress: http://bobtisdale.wordpress.com/

Monday, July 27, 2009

Regression Analyses Do Not Capture The Multiyear Aftereffects Of Significant El Nino Events


This post illustrates why regression analyses do not capture the multiyear aftereffects of significant El Nino events. To emphasize this, I’ve provided a detailed explanation of the processes that take place before, during, and after those significant El Nino events, using graphics and videos from earlier posts.


Regression analyses are used by climatologists to determine and illustrate the impact on global temperature of one or more variables, such as ENSO, Solar Irradiance, and Volcanic Aerosols. Figure 1 shows the results of one such study. It is a multi-cell illustration of “Surface Temperature Variability Components” from Lean and Rind (2008) “How Natural and Anthropogenic Influences Alter Global and Regional Surface Temperatures: 1889 to 2006” [GEOPHYSICAL RESEARCH LETTERS, VOL. 35, L18701, doi:10.1029/2008GL034864, 2008].
Link to Paper:
Figure 1

My Figure 1 is Figure 2 from Lean and Rind (2008). Under the heading of “Datasets”, Lean and Rind write, “Monthly fluctuations in ENSO, volcanic aerosols, solar irradiance and anthropogenic influences are shown in Figure 2. The multivariate ENSO index, a weighted average of the main ENSO features contained in sea-level pressure, surface wind, surface sea and air temperature, and cloudiness [Wolter and Timlin, 1998], extends from 1950 to 2006. It is augmented with an index derived from Japan Meteorologial Agency sea surface temperatures from 1868 [Meyers et al., 1999]. Volcanic aerosols in the stratosphere are compiled by [Sato et al., 1993] since 1850, updated from giss.nasa.gov to 1999 and extended to the present with zero values. The adopted solar forcing, consistent with IPCC [2007], is less than half that reported in prior IPCC assessments. Monthly irradiances since 1882 are estimate d from competing effects of sunspots and faculae in observations made by space-based radiometers, extended into the past using solar flux transport simulations [Wang et al., 2005]. The anthropogenic forcing is the net effect of eight different components, including greenhouse gases, landuse and snow albedo changes, and (admittedly uncertain) tropospheric aerosols [Hansen et al., 2007] (inset, Figure 2d).”

Lean and Rind then go on to detail the analyses they performed. Under the heading of “Amplitudes and Patterns of Natural and Anthropogenic Influences,” they state, “Natural changes cannot account for the significant long-term warming in the historical global surface temperature anomalies. Linear trends in temperature attributed to ENSO, volcanic aerosols and solar irradiance over the past 118 years (depicted by the lines in Figure 2) are, respectively, 0.002, -0.001 and 0.007 K per decade. Only by associating the surface warming with anthropogenic forcing is it possible to reconstruct the observed temperature anomalies.”

Basically, using a short-term comparison of NINO3.4 SST anomalies and Global RSS MSU TLT anomalies, my Figure 2, regression analyses like those used by Lean and Rind argue that natural variables cannot explain the upward divergence of global temperature from NINO3.4 SST anomalies. And if natural variables cannot explain the additional rise in global temperature, then the anthropogenic global warming hypothesis dictates that anthropogenic forcings must cause the rest. BUT…
Figure 2


Regression analyses regard El Nino events as a climate forcing of varying frequency and magnitude, the same way they consider other natural forcings such as volcanic aerosols and solar irradiance. They do not consider the multiyear processes that can occur after those El Nino events. Before presenting these, I’ll first provide a detailed description of the processes that take place before, during, and after significant El Nino events.


For those new to the process of El Nino events, Bill Kessler and David B. Enfield, both of NOAA, provide excellent descriptions of ENSO in their ENSO Q&A web pages. Link to Bill Kessler’s:
Link to David B. Enfield’s:

I’ll expand on their descriptions.

During non-El Nino years (La Nina and ENSO-neutral years), warm water accumulates in an area of the western tropical Pacific known as the Pacific Warm Pool (PWP); also known as the Indo-Pacific Warm Pool (IPWP). Refer to Figure 3.
Figure 3 (Source CRCES. Link to follow.)

Some of the warm water in the Pacific Warm Pool is water that returns there after El Nino events (the Equatorial Countercurrent in the Pacific relaxes after an El Nino and the North and South Equatorial Currents move the warm water back from the eastern to the western equatorial Pacific). More on that later. Some of the warm water in the Pacific Warm Pool results from solar radiation that warms the tropical Pacific and from the trade winds that push those warm surface waters from east to west in the Pacific during La Nina events and during ENSO-neutral periods. And some of the buildup of warm water in the Pacific Warm Pool occurs during the El Nino event itself, when cloud amounts over the Pacific Warm Pool drop significantly, causing a major rise in downwelling shortwave radiation (visible light). During the 1997/98 El Nino, it has been estimated that downwelling shortwave radiation rose as much as 25 watts/sq meter over the PWP. Refer to Figure 4. (This change in downwelling shortwave radiation was discussed in my post Recharging The Pacific Warm Pool Part 2.)

Figure 4

Figure 4 is from the Pavlakis et al (2008) paper “ENSO Surface Shortwave Radiation Forcing over the Tropical Pacific”:

The accumulation of warm water in the Pacific Warm Pool over months and years from trade winds pushing surface waters west, the periodic transport of the warm water out of the PWP by El Nino events, the blast of downwelling shortwave radiation during El Nino events, and the replenishment of the warm water during the subsequent La Nina all cause the size and temperature of the Pacific Warm Pool to vary.

Figure 5 illustrates the variations in area and temperature of the Pacific Warm Pool. The illustration is from the CRCES webpage “Natural decadal-multidecadal variability of the Indo-Pacific Warm Pool and its impacts on global climate” by Mehta and Mehta:
Figure 5

CRCES also provides a Quicktime movie (2.7MB) of the annual variations in Indo-Pacific Warm Pool area and SST anomalies here:

The variability of the Pacific Warm Pool can also be seen in the Western Equatorial Pacific Warm Water Volume, Figure 6, which is from my post Equatorial Pacific Warm Water Volume.
Figure 6

Note how, during the 1997/98 El Nino, the Western Equatorial Pacific Warm Water Volume (light blue curve) drops as NINO3.4 SST anomalies (black curve) rise. This is one indication that the warm water is being carried away from the Pacific Warm Pool during the El Nino event. Also note how quickly the Western Equatorial Pacific Warm Water Volume replenishes itself. It has “recharged” by the second phase of the 1998/99/00 La Nina.

The direction shifts in the Pacific Equatorial Currents that are part of an El Nino show how the warm water volume of the Pacific Warm Pool is lowered during those events. The Equatorial Countercurrent increases in size and carries the warm water from the Pacific Warm Pool to the east. When the El Nino ends, the Equatorial Countercurrent ebbs, and the North and South Equatorial Currents carry the warm water back to the west, to the Pacific Warm Pool. These shifts can be seen in Video 1 “Equatorial Currents Before, During, and After The 1997/98 El Nino” from my post of the same name:

Video 1

And there are subsurface changes that take place during an El Nino event. The warm water that was in the Pacific Warm Pool, most of it below the surface, shifts east during the El Nino, where it rises to the surface. These changes in the subsurface waters of the Pacific can be seen in my Video 2 “Cross-Sectional Views of Three Significant El Nino Events – Part 1”. Link to post:

Video 2

Though not discussed in Video 2, the rise of the thermocline at the end of the 1997/98 El Nino is visible. “Rewind” to minute 3:00 and start the video. After the commentary, the thermocline rises, further illustrating that warm water that was once below the surface of the Pacific Ocean has been brought to the surface by the El Nino.

Some BUT NOT ALL of the warm water that had sloshed east during the El Nino returns to the Pacific Warm Pool during the subsequent La Nina. And the warm water that doesn’t return to the Pacific Warm Pool is carried westward by the Equatorial Currents of the Pacific, Figure 7, to the surface of the Western Pacific and the Eastern Indian Oceans.
Figure 7

There, the warm water raises the surface temperature of the Western Pacific and the Eastern Indian Oceans, Figure 8.
Figure 8

The transport of this warm water and its aftereffects can be seen in Video 3 “Recharging The Pacific Warm Pool”. Link to post:

Video 3

In other words, warm water that was below the surface of the Pacific Warm Pool (and not included in the calculation of global temperature anomaly) is redistributed around the surface of the nearby oceans by the El Nino, (and it is now included in the calculation of global temperature). Phrased yet another way, before that El Nino, the warm water was not included in surface temperature record but afterward the warm water was included in surface temperature record. This raises global temperature anomalies without any heat input. Keep in mind that the rearranging of waters during an El Nino does not in and of itself create heat; it only shifts warm water from below the surface of the Pacific Ocean to the surface where it impacts temperature measurements.


And those upward step changes after the 1986/87/88 and 1997/98 El Nino events can be seen in the sea surface temperatures of the East Indian and West Pacific Ocean, the black curve in Figure 9. Also illustrated in Figure 9 are scaled NINO3.4 SST anomalies (purple curve) and Sato Index data (green curve), which I’ve added to illustrate the timing of explosive volcanic eruptions that impact sea surface temperature (and global temperature).
Figure 9

The area represented by the East Indian and West Pacific Ocean SST anomalies (the black curve in Figure 9) is shown in Figure 10.
Figure 10

Refer to my posts for further information:Can El Nino Events Explain All of the Global Warming Since 1976? – Part 1
Can El Nino Events Explain All of the Global Warming Since 1976? – Part 2


During the El Nino events, heat from the surplus of warm surface waters along the equatorial Pacific is pumped into the atmosphere where it is carried around the globe. This raises land surface temperatures, (not illustrated). And the higher atmospheric temperature also raises the surface temperature of the oceans outside of the tropical Pacific. These increases in SST can be seen in Video 4 “Global SST Anomaly Animation 1996 to 2009”. Video 4 is from my post “Animations of Weekly SST Anomaly Maps from January 3, 1996 to July 1, 2009.” There is no narrative with Video 4. The description is included in the post.

Video 4

The exchange of heat from atmosphere to ocean in the East Indian and West Pacific Oceans adds to the elevated surface temperatures that are caused by the warm water that had been carried there by ocean currents, discussed earlier. The El Nino also warms the East Pacific, South Atlantic, and West Indian Oceans through the atmosphere. Those portions of ocean basins are in turn cooled by the La Nina event that follows. But there is another portion of an ocean basin where the heat from the El Nino lingers; that is, the SSTs of that ocean basin are not impacted proportionately by the La Nina. And that ocean basin is the North Atlantic.


The title of the linked post “There Are Also El Nino-Induced Step Changes In The North Atlantic” explains the content. And these SST anomaly step changes in the North Atlantic correlate well with the step changes in the East Indian and West Pacific Oceans, though they result from different aftereffects of the significant El Nino events. Refer to Figure 11. Keep in mind that the North Atlantic is also impacted by the Atlantic Multidecadal Oscillation.

Figure 11

Assuming the North Atlantic represents approximately 15% of the global ocean surface area, then the East Indian and West Pacific plus the North Atlantic account for approximately 40% of the global ocean surface area. In the years that follow significant El Nino events, ocean currents and atmosphere-ocean processes “mix” the lingering elevated SST anomalies of the East Indian, West Pacific and North Atlantic Oceans with the remaining 60% of the global oceans. This causes the rise in global SST anomalies that presents itself as the divergence of Global SST anomalies from NINO3.4 SST anomalies, similar to that shown in Figure 2. That natural increase in SST anomalies is mistaken for warming due to anthropogenic causes.


The RSS MSU Time-Latitude Plots of Global TLT illustrate the transport of heat from the tropics toward the poles that result from significant El Nino events. This is illustrated and discussed in detail in my post “RSS MSU TLT Time-Latitude Plots...Show Climate Responses That Cannot Be Easily Illustrated With Time-Series Graphs Alone”. In that post, I combined Time-Series Graphs with the Time-Latitude Plots to show the effects of the significant El Nino events. But even without the time-series graphs, the 1997/98 El Nino is easy to find in Figure 12. It appears as an area of elevated tropical TLT anomalies that begins in 1998 and ends about a year later. Note that most of the heat that had been in the tropics is transported to the mid-to-high latitudes of the Northern Hemisphere, where it lingers through the 1998/99/00 La Nina. Regression analyses cannot capture that lingering aftereffect of an El Nino.
Figure 12

The Time-Latitude Plots also show the impacts of the 1986/87/88 El Nino and limited TLT response to the 1982/83 El Nino. Refer to Figure 13. The 1982/83 El Nino was counteracted by the explosive eruption of El Chichon.
Figure 13


This post primarily discussed the processes and aftereffects of the significant El Nino events of 1986/87/88 and 1997/98, using the 1997/98 El Nino as reference in many of the discussions and links. There were two other significant El Nino events since 1970, the 1972/73 and 1982/83 El Nino events. The 1982/83 El Nino was counteracted by the eruption of El Chichon, which turned it into a nonentity. As illustrated in Figure 14, there are striking similarities between the multiyear periods that followed the 1972/73, 1986/87/88, and the 1997/98 El Nino. This was discussed in detail in my post “Similarities of the Multiyear Periods Following Significant El Nino Events Since 1970.” Are these lesser El Nino events simply aftereffects of the significant El Ninos?

Figure 14


Regression analyses do not account for the multiyear aftereffects of significant El Nino events and do not account for the resulting El Nino-induced step changes in SST, TLT, and Land Surface Temperatures. Regression analyses falsely attribute the divergence of global temperature anomalies from NINO3.4 SST anomalies to anthropogenic causes when, in fact, the divergence is caused by the lingering aftereffects of significant El Nino events. The additional rise in global temperatures after the significant El Nino events is in reality caused by subsurface waters from the Pacific Warm Pool being transported to the surface and remaining there after the El Nino event has ended.


Sources of the data used in the graphs are provided in the linked posts.


Anonymous said...

So the East Pacific is a huge reservoir of heat. And the big El Niños release it. Then I have 2 questions:

1) Do you think that there will be a huge 2009/2010 El Niño?. I am from Peru, and here the big El Niños are the most important climatic fenomena (and the most disastrous)

2)You are skeptic of the greenhouse warming theory, or you simply state that all that heat trapped by greenhouse gases is stored under pacific waters, and then suddenly and periodically released to the atmosphere?

Bob Tisdale said...

Anonymous: You wrote, "So the East Pacific is a huge reservoir of heat. And the big El Niños release it."

The reservoir (Pacific Warm Pool) is in the Western Pacific.

Projections are for a moderate El Nino.

I'm skeptical of anthropogenic global warming.

Kevin Kilty said...

Bob Tisdale;

Your analysis is quite interesting and leads to the following comment and question. Fifteen years ago I tried to get funded a project to analyze absolute humidity trends, changes in sources, and the resulting effect on global temperature trends. Nobody bit on that one. However, your weblog about after-effects brought this proposal to mind again. I wonder if in addition to redistributing stored heat, the El Nino is not also increasing and perhaps redistributing water vapor for a substantial period of time which I imagine to have a substantial effect on winter-time and night-time temperatures. Any thoughts to share?

Bob Tisdale said...

Kevin Kilty: You wrote, "I wonder if in addition to redistributing stored heat, the El Nino is not also increasing and perhaps redistributing water vapor for a substantial period of time which I imagine to have a substantial effect on winter-time and night-time temperatures. Any thoughts to share?"

El Nino events change atmospheric circulation patterns, precipitation, cloud cover and amount. They shift jet streams which would in turn change the locations of clouds and precipitation.

If water vapor varies with SST, then an upward step in SST anomalies of the North Atlantic, East Indian, and West Pacific Oceans should cause a shift in water vapor. Have I plotted it? Nope.

Anonymous said...

The Nino system is as you portray effectively redisributes heat of the Pacific ocean and cleary an El Nino not only causes heating in the eastern pacific it also prevents the cooling here causing upto a 0.2C temperature shift for the globe if 1998 is anything to go by.

It is a cycle though and during La Nina the cooling influence is increased.

Clealy for the system to be causal for the current rise in temperature it would have had to go into a phase where El-Nino predominated and heat previously stored (from ?when) became disipated into the surface system to attenuate the overall surface temperatures. That is it would have to act as if an additional external forcing to be casual for the late 20th temperature rise.

This is not the case as shown by the paper you quote and all your explanation shows is what is already known i.e that acutely an El Nino can add 0.2C to the average global temperatures, however as pointed to this is balanced by the cycles of the Nino system and this balance would have to have changed since the early 1970's for this to be a significant cause of gloabal warming.

And it hasn't and isn't and as the paper in GEo.REs.Let. demonstrates less than 0.02C can possible be attributed to changes in the El-Nino system and with the recent La Nina this is even less.

You have to look for long trends in the system not just show how it heats acutely.

Of course the extra heat being pushed into the oceans by the GHG effect (and please don't say that doesn't exist e.g. H2O), is changing the telecommunications system via NAO variations and it is unclear how this effect things.

From recent ananlysis it seems that the heat dissipation system that is the teleconnections system was causing rapid warming between 20's-50's then slow warming from the 50-80, then fast from 80-2003ish and then slow again from 2003-2009, just like a choatic system would be expected to do as the external heat forcing increased.

Its about to change again i suspect due to the rapid melting arctic sea ice and this effects on the telecommunications systems and the expanding tropics and moving jetstreams.

June was the second warmest ever!

2005 hottest = year and 2006feb 2007feb hottest twelve month period ever.

And all monthly tops have been in the last 10 years when we should have been cooling due to natural forcings (sunspots and Nina trends).

Now as you provide to evidence that the Nino system is causing a prolonged heating due to chang ein its nature, only that acute El Nino events do (as is well known) think i'll tend to go with the thorough and peer reviewed findings in the paper you quote as the real additional heatring caused overall by the El-Nino La Nino system, that is there hasn't been any.


Bob Tisdale said...

Anonymous (August 2, 2009 5:31 AM): With the exception of one point, I have addressed all of your arguments in this post and the ones linked to it. I've shown with videos of actual conditions, with graphs of actual data from specific ocean areas why your arguments fail. You apparently did not bother to read and comprehend all that was written and illustrated in all of those posts, because if you had, you would not have written what you've written.

What I have not explained in any post so far is that La Nina events are not the opposite of El Nino events.

During a significant El Nino, trade winds relax and warm waters from the Western Equatorial Pacific and from below the surface of the Pacific Warm Pool slosh to the east. The current that carries the water at this stage is the Equatorial Counter Current. It is small to non-existent during La Nina and ENSO-neutral periods. While the water is being carried east by the Equatorial Counter Current, the North and South Equatorial Currents still exist and they carry the warm water that had traveled east back in the opposite direction, westward, where it enters the other currents of the Northern and Southern Pacific Gyres. These carry the warm water away from the tropics, toward the mid latitudes.

At the end of the El Nino, the trade winds resume, and the Pacific Equatorial Counter Current relaxes. The North and South Equatorial Currents return to the area that had been occupied by the Equatorial Counter Current during the El Nino. This change in direction of equatorial flow takes what remains of warm water that had traveled east during the El Nino and returns it to the west and the Pacific Warm Pool.

During the La Nina, convection, total cloud amount, and precipitation follow the warmer water west and the trade winds increase as the warmer water and convection travels farther west. This raises the thermocline in the eastern Pacific, raising lower temperature waters to the surface, increasing the temperature gradient from west to east, which increases the strength of the trade winds. The La Nina occurs when and because the coupled atmosphere-ocean processes overreact during the quest for a normal state.

In effect, the La Nina is only an exaggeration of the “normal” ENSO-neutral state.

Colin Aldridge said...

Bob .. Is there any explanation for why/ how the sub surface warm water pool gets "recharged/ reheated" after it warms the ocean surface during an El Nino. Obviously? the sub surface water must cool as a result of it rising to the surface .. Colin A

Bob Tisdale said...

Colin Aldridge: I wrote two posts on the subject of Recharging the Pacific Warm Pool:


colin Aldridge said...

From Colin Aldridge

I better go read them then!! Many thanks


Tips are now being accepted.

Comment Policy, SST Posts, and Notes

Comments that are political in nature or that have nothing to do with the post will be deleted.
The Smith and Reynolds SST Posts DOES NOT LIST ALL SST POSTS. I stopped using ERSST.v2 data for SST when NOAA deleted it from NOMADS early in 2009.

Please use the search feature in the upper left-hand corner of the page for posts on specific subjects.
NOTE: I’ve discovered that some of the links to older posts provide blank pages. While it’s possible to access that post by scrolling through the history, that’s time consuming. There’s a quick fix for the problem, so if you run into an absent post, please advise me. Thanks.
If you use the graphs, please cite or link to the address of the blog post or this website.