More data fiddling – this time in NOAA's ENSO data

Plot below showing ONI -vs- Aqua Channel 5 Temperature from lukewarmplanet (not Tisdale) to illustrate what he is talking about in his upcoming book. – Anthony

Comments on NOAA’s Recent Changes to the Oceanic NINO Index (ONI)

Guest post by Bob Tisdale

As many of you are aware, I’m writing another book. The working title is The Ignored Driver of Global Climate: El Niño- Southern Oscillation. I’m about two-thirds of the way done, I believe. There’s lots of new illustrations (so far about 70% are new), and they should help those who are having trouble understanding the processes of El Niño-Southern Oscillation (ENSO).

I’ve just finished a discussion of NOAA’s recent changes to their Oceanic NINO Index, also known as ONI, in the chapter about ENSO Indices. I believe you’ll find it interesting. (I’ve left the illustration numbers as they presently exist in the book draft.)

NOAA’s Climate Prediction Center (CPC) recently modified their Oceanic NINO Index (ONI). Refer to their Description of Changes to Oceanic NINO Index webpage for a complete write-up. There they note in the opening paragraph:

Due to a significant warming trend in the Niño-3.4 region since 1950, El Niño and La Niña episodes that are defined by a single fixed 30-year base period (e.g. 1971-2000) are increasingly incorporating longer-term trends that do not reflect interannual ENSO variability. In order to remove this warming trend, CPC is adopting a new strategy to update the base period.

NOAA does not attribute the “significant warming trend” to anthropogenic greenhouse gases, but anytime that phrase is used it implies manmade warming to many persons. Unfortunately, what NOAA has actually done with their changes is minimize the impact of the 1976 Pacific Climate Shift on NINO3.4 sea surface temperature anomalies.

I cannot fathom why they would do that when the 1976 Climate Shift is the subject of numerous scientific studies. Google scholar has 176 returns for “1976 climate shift”, in quotes. It is an accepted, well-documented phenomenon.

The Oceanic NINO Index is based on the NOAA ERSST.v3b sea surface temperature dataset. Yup, that’s the dataset that NOAA introduced in 2008 with bias-corrected satellite data and then quickly modified, removing the satellite-based data, when “users” at NOAA discovered that the satellite data made global sea surface temperatures in 1998 warmer than 2003 by a couple hundredths of a deg C. Refer to the discussion of Figure 2-23 in Chapter 2.

Figure 4-21 illustrates the ERSST.v3b-based NINO3.4 sea surface temperature anomalies, on which the Oceanic NINO Index data is based. The data does in fact have a positive linear trend of slightly less than 0.06 Deg C per decade. Note that I’ve highlighted 1976 to point out the Climate Shift.

So let’s look at the data before and after the climate shift.

The ERSST.v3b-based period average sea surface temperature anomalies for the NINO3.4 region from January 1950 to December 1975 and from January 1977 to May 2012 are shown in Figure 4-22. The average sea surface temperature anomalies after the 1976 Pacific Climate Shift are about 0.3 deg C higher than they were before it. By the way, that shift impacted the entire Eastern Pacific Ocean, not just the eastern equatorial Pacific.

And of course the linear trends before and after the climate shift are negative, Figure 4-23, and that implies the climate shift is responsible for a good portion of the overall positive linear trend from 1950 to present.

To “remove” the “significant warming trend” caused in part by the 1976 climate shift in the sea surface temperatures of the NINO3.4 region, NOAA no longer uses a single set of base years (1971-2000) for the anomalies in their Oceanic NINO index. They now use a series of shifting base years. They explain:

ONI values during 1950-1955 will be based on the 1936-1965 base period, ONI values during 1956-1960 will be based on the 1941-1970 base period, and so on and so forth.

The result: NOAA has eliminated the positive trend in what used to be sea surface temperature anomalies of the NINO3.4 region. See Figure 4-24. One can’t even call them sea surface temperature anomalies anymore with the sliding base years. But for now, we’ll treat them as anomalies.

With the changes, NOAA has minimized the difference in the period-average NINO3.4 sea surface temperature “anomalies” before and after the 1976 shift. Based on the “raw” ERSST.v3b data, the climate shift caused NINO3.4 sea surface temperature anomalies to shift up 0.3 deg C, but the “corrections” dropped the shift to about 0.04 deg C, as shown in Figure 2-25.

And the changes to the way NOAA calculates sea surface temperature “anomalies” for the Oceanic NINO Index has resulted in more severe negative trends before and after the climate shift. See Figure 4-26.

I won’t speculate about why NOAA would want to minimize the effect of the 1976 Pacific Climate Shift. The reason given for the changes seems odd at best. Consider this though: There are scientific studies where the authors remove the linear effects of ENSO on global surface temperatures by simply scaling and subtracting an ENSO index from global surface temperatures. The authors then mistakenly claim the remaining trend in global surface temperatures is the result of anthropogenic global warming. This faulty method of determining the effects of ENSO on global surface temperature is discussed further in Section 6.  Now, because NOAA has flattened the Oceanic NINO index trend, if someone were to use it in one of those misleading scientific studies, the remaining trend in global surface temperature residuals would be a little bit higher than if they had used a sea surface temperature anomaly-based ENSO index.

In summary, because the Oceanic NINO index no longer represents the sea surface temperature anomalies of the NINO3.4 region using a single base period, and because NOAA has minimized the impact of the 1976 Pacific Climate Shift in it, and because that climate shift exists in all sea surface temperature datasets, I, personally, would not use Oceanic NINO index as an ENSO index. Then again, I don’t believe NOAA cares one way or the other if I use their Oceanic NINO index.

INTERESTED IN LEARNING MORE ABOUT THE EL NIÑO-SOUTHERN OSCILLATION AND DON’T WANT TO WAIT FOR MY NEW BOOK?

About one-quarter of my book If the IPCC was Selling Manmade Global Warming as a Product, Would the FTC Stop their deceptive Ads?, Section 6, is about the processes that are part of El Niño and La Niña  events. Many of the discussions are rewordings (expansions and simplifications) of my posts here at Climate Observations, so you could save a few bucks and read dozens of posts.   But the book provides a single resource and reference for you and includes a very basic, well-illustrated introduction to El Niño, La Niña, and ENSO-neutral conditions written in simple terms.  Included in that section are discussions of how La Niña events are not the opposite of El Niño events and how and why certain parts of the global oceans warm in response to certain El Niño AND to the La Niña events that follow them.  The El Niño-Southern Oscillation is a marvelous process Mother Nature has devised to enhance or slow the distribution of heat from the tropics to the poles.  It is process that naturally varies in intensity, and due to those variations, it is capable of warming or cooling global temperatures over multiyear and multidecadal periods.   The individual chapter titles of Section 6 will give you an idea of the topics discussed.  See pages 9 and 10 of the introduction, table of contents, and closing of my book in pdf form here.