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

Thursday, November 6, 2008

A Different Way to Look at NINO3.4 Data


Due to the random timing and varying intensity of ENSO events, a graph of NINO3.4 SST anomalies, Figure 1, is a difficult illustration from which to gather visual information. Sure, you can pick out the spikes of the El Ninos and La Ninas. But anything else? Maybe an underlying trend, here and there?
Figure 1

Many La Ninas appear to be an overreaction of SST anomalies as the equatorial Pacific returns to “normal” after El Nino events. In fact, in his discussions on the webpage titled "Frequently-(well, at least once)-asked-questions about El Niño", Oceanographer William S. Kessler of the NOAA Pacific Marine Environmental Laboratory states, “Many scientists are coming to the view that there may not be such a thing as La Niña, or at least that it is not just the opposite of El Niño. Perhaps there is just the normal situation that is disturbed every few years by an El Niño.”
While this might be the case in recent years, La Nina events do dominate many periods of the 20th Century, so I’m not necessarily convinced that there are no independent La Nina events. Regardless, how can we illustrate the single El Nino/La Nina event if they do in fact exist? Maybe it’s simple as...


Figure 2 illustrates long-term NINO3.4 SST anomaly data that’s been smoothed with a 25-month running-average filter. Since a “typical” El Nino/La Nina cycle occurs over a two-year period, by using the 25-month filter, the positive El Nino anomalies and the negative La Nina anomalies that follow are merged, assuming a La Nina does follow, which is not always the case. Also, in this way, El Nino events
- that don’t have abnormally high SST anomalies
- but that last more than one year
- and don’t have sizeable La Ninas between them or following them to dampen them
can appear to have the same magnitude as the major El Ninos of 1982/83 and 1997/98. (The same would hold true for periods with multiple-year La Nina events.) Note how the 1939/40/41/42 El Nino becomes the largest ENSO event of the 20th Century. This helps explain the spike in temperature at that time, especially in the Indian Ocean SST data.

Figure 2

Figure 3 is the short-term version of the NINO3.4 SST anomaly data smoothed with a 25-month filter. There actually appears to be an oscillation in the NINO3.4 data, not just the random noise of the raw data. Note that the number and magnitude of El Nino events clearly dominate La Ninas during the period.
Figure 3


Sea Surface Temperature Data is Smith and Reynolds Extended Reconstructed SST (ERSST.v2) available through the NOAA National Operational Model Archive & Distribution System (NOMADS).


Anonymous said...

Have you done an analysis of the comparison between the ENSO 3.4 data and the SOI. Is there a consistent lead by the SOI?

Given that tropical warming events are usually general which latitudes in the tropics show a sea surface temperature variation that is a best fit to these indices?

Bob Tisdale said...

Anonymous: Thanks for making me look at the SOI and NINO3.4 data again. I don't know that I can say that one precedes the other consistently. Those track remarkably close to one another. I'll have to check some other data sets tomorrow morning.

The warming events aren't general. The Tropical West Pacific SST decrease while the East increases and vice versa. Please clarify your question and whether you're refering only to the Pacific.


Bob Tisdale said...

Anonymous: I’ve done a few posts that dealt primarily with the tropics.

My earliest post on the tropics. Actually, it’s 5S to 5N for each of the oceans.

Another post that deals solely with the tropics:

This is the 5th in a follow-up series. Links to the other four are included at the beginning of it:

They get a little more detailed each time I take a look.

Erl said...

Thanks for pointing me to the other comparisons.

I notice that though the amplitude of the fluctuations vary strongly these variations look pretty well synchronized: http://i25.tinypic.com/118ovbd.jpg

I noticed that sea surface temperatures in all oceans start to rise together about 1910 in this graph. http://i27.tinypic.com/205bolj.jpg

Perhaps we focus a little too closely on the big gyrations in the Eastern Pacific and fail to notice that in the big picture all the oceans are moving together. Ocean temperatures obviously vary strongly in the equatorial zone depending upon the strength of circulation coming back from high latitudes and the Antarctic in particular.

In assessing the climatic variations due to these successive warming events are we looking too closely at a very narrow latitude band? Perhaps, in terms of global climatic effects we could look at the change in energy stored in the waters between say 20°south and the equator and 20°north and the equator? Would there be a lot of difference if we looked, not at energy stored but at surface temperatures? I am interested in differences between the hemispheres and locating the latitude zones and particular areas where the energy is gained.

Very interesting to see that water temperature changes at depth before it changes at the surface. Is this possibly due to a change in light intensity due to change in cirrus density in the upper troposphere?

Sorry to be anonymous but sometimes passwords are hard to find.

Bob Tisdale said...

Erl: I agree that the major global oceans all follow the same overall trends. There are minor variations between oceans due to Thermohaline Circulation/Meridional Overturning Circulation. The big unanswered questions for me are: Why do they show a sharp decrease in SST from the late 1800s to the early 1900s? What caused SST to drop during that period? It's not a volcanic eruption. Is the increase in SST from 1910 to 1940 simply a rebound in a cycle? If so, why can't the rise from the 1940 to 2000 simply be a continuation of that overall cycle, with a minor mid-to-late century variation due to the AMO and North Pacific Residual?

Erl said...

How reliable is the data before 1948? I imagine it is based on a small number of observations.

If indeed we are looking at a long cycle in solar wind activity influencing temperature, and the correlations are good even if the mechanism is unknown, a return to 1910 temperature has to involve a lot of cooling over perhaps several solar cycles.

Establishing what drives tropical warming events is in my view the key to understanding the 100 plus year evolution of sea surface temperature that you document.

I don't buy the general notion that ENSO and its less spectacular manifestations in other oceans is due to a 'coupled interaction between ocean and atmosphere' as if this were wholly generated within the Earth system. If the ocean gains or loses heat there is a little problem in logic with that proposition. For the oceans in general to heat all that is required is a change of cloud albedo.

Bob Tisdale said...

Erl: Prior to 1880, the errors in the ERSST.v2 data make it questionable. From there to 1950, the uncertainties lessen. More discussion on that here:


Tips are now being accepted.

Blog Archive

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.