…and We Can Determine its Effects through Linear Regression Analysis, Then Remove Those Effects, Leaving the Anthropogenic Global Warming Signal
This is the 5th post in the series about myths and failed arguments about the El Niño-Southern Oscillation—myths that were created by proponents of the hypothesis of human-induced global warming. The intents of these myths are obvious: they downplay and obscure the very obvious roles Mother Nature has played in global warming. Links to past posts about other myths and a listing of future posts are provided at the end of this one. This series of posts are reprints from my book Who Turned on the Heat?
For four years, I’ve searched for signs of an anthropogenic warming signal in the global satellite-era sea surface temperature record. It’s difficult if not impossible to find (1) if you divide the global oceans into logical subsets and (2) if you understand the role that strong El Niño events played in that warming. In this post, we’ll illustrate the impacts of those strong El Niños on GISS Land-Ocean Temperature Index (LOTI) data, which is the merger of land surface air temperature and sea surface temperature anomaly data.
The title of this myth is long, but self-explanatory: “ENSO Only Adds Noise to the Instrument Temperature Record and We Can Determine its Effects through Linear Regression Analysis, Then Remove Those Effects, Leaving the Anthropogenic Global Warming Signal.” The flawed assumption that perpetuates this myth is that the year-to-year wiggles associated with El Niños and La Niñas ride on top of a warming signal created by manmade greenhouse gases. The realities are, (1) strong El Niño events are responsible for the vast majority of the warming, and (2) ocean heat content data indicates El Niños are fueled naturally. If this subject is new to you, refer to my illustrated essay “The Manmade Global Warming Challenge” [42MB] for an overview.
This myth has been discussed in a number of posts at my blog Climate Observations, many of which were cross posted at WattsUpWithThat. Most recently, an overview of the flaws of this myth was presented in the video included in the post The Blatant Errors in the SkepticalScience Video “Global Warming over the Last 16 Years”, which was cross posted at WUWT here. My YouTube video included in that post is On the SkepticalScience Video “Global Warming over the Last 16 Years”. This post is a further discussion of the flaws in the SkepticalScience video—a more detailed look at those errors in the SkepticalScience argument is planned for next week.
The following is Chapter 7.5 from my book Who Turned on the Heat? For those with copies, it starts on page 459. I also discovered a typo, which I struck and corrected in this post.
7.5 ENSO Only Adds Noise to the Instrument Temperature Record and We Can Determine its Effects through Linear Regression Analysis, Then Remove Those Effects, Leaving the Anthropogenic Global Warming Signal
The relationship between ENSO events and global surface temperatures is known. During an El Niño event, some places around the globe warm in response and other places cool. The warming exceeds the cooling, so global temperatures, as a whole, warm in response to an El Niño event. On the other hand, during a La Niña event, the cooling around the globe exceeds the warming and, as a whole, global temperatures cool in response to a La Niña event. Early work on the relationship between ENSO and weather (temperature and precipitation) around the globe, as noted earlier in the book, include Berlage (1976) Southern Oscillation and World Weather, Newell and Weare (1976) Factors Governing Tropospheric Mean Temperature, Angell (1981) Comparison of Variations in Atmospheric Quantities with Sea Surface Temperature Variations in the Equatorial Eastern Pacific, Pan and Oort (1983) Global Climate Variations Connected with Sea Surface Temperature Anomalies in the Eastern Equatorial Pacific Ocean for the 1958–73 Period.
Somewhere along the line, possibly Jones in the 1989 book The influence of ENSO on global temperatures [not available online], a study used a statistical tool such as correlation or regression analysis to determine the linear relationship between an ENSO index and global temperature. With that factor and an appropriate time lag between the ENSO index and global surface temperatures, they then crossed a hurdle. They subtracted the scaled ENSO index from the global temperature data and claimed the difference was caused by manmade greenhouse gases.
That is, with a statistical tool, they used an ENSO index and global surface temperatures to determine how much global surface temperatures warmed and cooled in response to the ENSO signal represented by the ENSO index. They also determine the time lag between the change in the ENSO index and the response in global surface temperatures. For example, let’s say a study determined that global surface temperatures varied 0.18 deg C for every 1.0 deg change in the ENSO index and that global surface temperatures lagged the ENSO index by 3 months. Further, let’s say the study also included a dataset called Aerosol Optical Depth (AOD) to account for the sun-shading effects of volcanic aerosols spewed into the stratosphere by explosive volcanic eruptions. ENSO and volcanic aerosols are the two primary causes of the year-to-year wiggles in the global surface temperature record. Like before, also for the example, we’ll exclude polar data, so we’ll only examine the GISS Land-Ocean Temperature Index data from 65S-65N.
As a reminder, the Arctic temperature data is also impacted by a phenomenon called polar amplification, which skews the data. As also noted before, GISS, for all intents and purposes, deletes sea surface temperature data in areas of seasonal sea ice in both polar oceans. They then replace it with land surface temperature data, which will exaggerate the long-term warming there because land surface temperatures vary much more than sea surface temperatures. There’s no need to include those biases in our discussion.
The three datasets are shown in Figure 7-14. Note that I haven’t scaled the volcanic aerosol data, and that it shows a dip of about 0.15 deg C in 1991 in response to the eruption of Mount Pinatubo. Climate studies have estimated the impact of Mount Pinatubo to have been in the range of 0.2 to 0.5 deg C. I’ll show you what they missed—later.
The climate studies that attempt to remove the impacts of ENSO and the volcanic eruptions, very simply, then subtract the scaled and lagged ENSO index and Aerosol Optical Depth data from the global surface temperature data and assume the warming shown is caused by anthropogenic greenhouse gases. See Figure 7-15. It’s nonsense, but there are a good number of studies that attempt to show this. Why is it nonsense?
There are two very obvious erroneous assumptions made by those studies. First, they assume the ENSO index represents the all of the effects of the entire ENSO process. We know that’s not correct. The ENSO index only represents the effects of ENSO on the variable or variables being measured. For NINO3.4 sea surface temperature anomalies, the ENSO index only represents the impacts of El Niño and La Niña on the sea surface temperatures of a small region along the equatorial Pacific. The Southern Oscillation Index only represents the effects of ENSO on the sea level pressures of Tahiti and Darwin, Australia. Those ENSO indices do not account for the effects of the warm surface waters in the tropical Pacific that are left over after an El Niño and they do not account for the warm subsurface waters that are returned to the west Pacific and East Indian Ocean by Rossby waves. The Multivariate ENSO Index, which includes a number of other different variables, also fails to capture these important aspects of ENSO.
The second erroneous assumption—maybe it’s actually the second and third erroneous assumptions—is they assume the effects of ENSO on global surface temperature are proportional or linear to the ENSO index and that those effects are all the same sign. We know the effects of ENSO on global sea surface temperatures are NOT proportional to the ENSO index. The sea surface temperatures of the East Indian and West Pacific Ocean warm in response to an El Niño but fail to cool during the trailing La Niña event. Also, we’ve already shown and discussed a number of times that there are places around the globe that warm in response to an El Niño and some that cool in response to an El Niño, and vice versa for a La Niña. As a reminder, the map in Figure 7-16 presents the correlation of GISS Land-Ocean Temperature Index data with NINO3.4 sea surface temperature anomalies, with a 3-month lag.
Let’s smooth the ENSO- and volcano-adjusted GISS Land-Ocean Temperature Index data with a 13-month running-average filter, Figure 7-17, and see what we can see. You may have noticed the upward shifts during the 1986/87/88 and 1997/98 El Niño events in the “raw” (unsmoothed) data (Figure 7-15), but they stand out remarkably well in the smoothed data. The periods before, between, and after the 1986/87/88 and 1997/98 El Niño events (highlighted in blue) still show signs of ENSO-related variations, but they’re inversely related to the NINO3.4 data. Stop and think about that for a moment. Considering the different time lags and when and where global surface temperatures are impacted by ENSO at those assorted time lags, it’s very logical that there are noticeable ENSO related signals visible that are the opposite sign after you remove the major variations of the same sign. In fact, you should expect to see them.
We can show the inverse relationship between ENSO and the ENSO- and volcano-adjusted GISS surface temperature data, by scaling and inverting the NINO3.4 sea surface temperature anomalies. In Figures 7-18, 7-19 and 7-20, the NINO3.4 data was scaled by a factor of 0.12, which is less than the scaling factor we used when we originally adjusted the data; then we used a scaling factor of 0.18. The NINO3.4 data in the three upcoming figures has not been lagged. This helps to show the timing of the upward shifts, which occur during the transitions from El Niño to La Niña. The upward shifts trail the NINO3.4 data by a couple of months. The lagged relationship between the inverted NINO3.4 data and the adjusted GISS Land-Ocean temperature stands out best in Figure 7-20, but it can also be seen in Figure 7-19, for the period between the 1986/87/88 and 1997/98 El Niño. You’re probably asking yourself why I’m using three graphs of the same two datasets. In Figure 7-18, I’ve offset the NINO3.4 data by -0.15 deg C, to show how well the adjusted GISS data and the inverted and scaled NINO3.4 data agree before the 1986/87/88 El Niño. The NINO3.4 data is offset -0.05 deg C in Figure 7-19 to show how well the two datasets agree between the 1986/87/88 and 1997/98 El Niño events. Then, in Figure 7-20, the inverted and scaled NINO3.4 sea surface temperature anomalies were offset +0.15 deg C to show how well that inverted NINO3.4 data now agrees with the ENSO- and volcano-adjusted GISS Land-Ocean Temperature Index data after the 1997/98 El Niño. The changes in the offsets provide a very rough approximation of the shifts in surface temperatures caused by the 1986/87/88 and 1997/98 El Niño events.
It all makes sense, and there was no need to rely on greenhouse gases to explain the warming of land plus sea surface temperatures from November 1981 to March 2012, for the latitudes of 65S-65N. If you’re not aware, the latitudes of 65S-65N cover about 90% of the planet, and the GISS data north of 65N and south of 65S has been modified by their deletion of sea surface temperature data and replacing it with land surface temperature data.
Recall that in Figure
5-14 7-14 we assumed the eruption of Mount Pinatubo only cooled global surface temperatures by about 0.15 deg C. We also noted that climate studies determined that global surface temperatures cooled 0.2 to 0.5 deg C in response to the volcano. Some of those studies removed the linear effects of ENSO to determine the response of global temperatures to Mount Pinatubo. They assumed incorrectly that the response of global surface temperatures to ENSO were all the same sign. They didn’t bother to consider the fact that there are areas around the globe that cool in response to El Niño and warm in response to La Niña AND that these inverse effects may be strongest at a different time lag. Now, if you would, please scroll back up to Figure 7-19. Note how well the scaled and inverted NINO3.4 data agrees with the ENSO- and volcano-adjusted GISS Land-Ocean Temperature Index data at the time of the Mount Pinatubo eruption. The adjusted GISS land plus sea surface temperature data is cooling in response to the string of El Niño events from 1991 to 1995. The studies appear to have overestimated the response of global surface temperatures to the eruption of Mount Pinatubo, because they failed to notice that the ENSO-adjusted surface temperatures (which were adjusted for surface temperature response to ENSO with the same sign) were cooling in response to the El Niño events.
Let’s look at the climate model hindcasts for the period of 1999 to 2000 that were and will be used by the IPCC in the 4th and 5th Assessment Reports, Figure 7-21. The data has been detrended to approximately “zero” before and after the responses to the 1991 eruption of Mount Pinatubo. I’ve also presented the Aerosol Optical Depth data required to account for the actual dip and rebound caused by it. The multi-model ensemble mean of the climate models prepared for the IPCC’s upcoming 5th Assessment Report more than doubles the response needed to explain the dip and rebound.
In summary, ENSO is a coupled ocean-atmosphere process and its effects on Global Surface Temperatures are not represented by an ENSO index. ENSO indices cannot account for the impacts of the warm water released by an El Niño event, returned to the West Pacific and redistributed from there. Therefore, any scientific papers that attempt to determine manmade global warming by removing the linear effects of ENSO with a scaled and lagged ENSO index and subtracting it from global surface temperatures are fatally flawed. Papers that make this error-filled, misleading effort include:
Foster and Rahmstorf (2011) Global Temperature Evolution 1979–2010
Lean and Rind (2009) How Will Earth’s Surface Temperature Change in Future Decades?
Fawcett (2008) Has the world cooled since 1998?
Trenberth et al (2002) Evolution of El Niño–Southern Oscillation and global atmospheric surface temperatures
Wigley, T. M. L. (2000) ENSO, volcanoes, and record-breaking temperatures
A COUPLE OF FINAL NOTES FOR THIS CHAPTER
Trenberth et al (2002) is a widely cited paper about ENSO. I’ve used it as a reference in this book. Only a very small portion of that paper makes the error discussed in this chapter. Trenberth et al also provide a disclaimer in the second paragraph of their Conclusions, (their paragraph 52, my boldface):
The main tool used in this study is correlation and regression analysis that, through least squares fitting, tends to emphasize the larger events. This seems appropriate as it is in those events that the signal is clearly larger than the noise. Moreover, the method properly weights each event (unlike many composite analyses). Although it is possible to use regression to eliminate the linear portion of the global mean temperature signal associated with ENSO, the processes that contribute regionally to the global mean differ considerably, and the linear approach likely leaves an ENSO residual.
The ENSO “residuals” are a significant contributor to the warming of Global sea surface temperatures during the satellite era, as we’ve shown throughout this book.
A more recent paper, Compo and Sardeshmukh (2010) Removing ENSO-Related Variations from the Climate Record, seems to be a very important step in the right direction. They write (my boldface):
An important question in assessing twentieth-century climate is to what extent have ENSO-related variations contributed to the observed trends. Isolating such contributions is challenging for several reasons, including ambiguities arising from how ENSO is defined. In particular, defining ENSO in terms of a single index and ENSO-related variations in terms of regressions on that index, as done in many previous studies, can lead to wrong conclusions. This paper argues that ENSO is best viewed not as a number but as an evolving dynamical process for this purpose.
Note: While Compo and Sardeshmukh made a step in the right direction, they missed a very important aspect of ENSO. They overlooked the significance of the huge volume of warm water that is left over after certain El Niño events, and they failed to account for its contribution to the warming of global Sea Surface Temperature anomalies since about 1975/76.
[END OF REPRINT]
INTERESTED IN LEARNING MORE?
Why should you be interested? As noted in the post, the sea surface temperature records indicate El Niño and La Niña events are responsible for the warming of global sea surface temperature anomalies over the past 30 years, not manmade greenhouse gases. I’ve searched sea surface temperature records for more than 4 years, and I’ve searched ocean heat content records for more than 3 years, and I can find no evidence of an anthropogenic greenhouse gas signal. That is, the data indicates the warming of the global oceans has been caused by Mother Nature, not anthropogenic greenhouse gases.
I’ve recently published my e-book (pdf) about the phenomena called El Niño and La Niña. It’s titled Who Turned on the Heat? with the subtitle The Unsuspected Global Warming Culprit, El Niño Southern Oscillation. It is intended for persons (with or without technical backgrounds) interested in learning about El Niño and La Niña events and in understanding the natural causes of the warming of our global oceans for the past 30 years. Because land surface air temperatures simply exaggerate the natural warming of the global oceans over annual and multidecadal time periods, the vast majority of the warming taking place on land is natural as well. The book is the product of years of research of the satellite-era sea surface temperature data that’s available to the public via the internet. It presents how the data accounts for its warming—and there are no indications the warming was caused by manmade greenhouse gases. None at all.
Who Turned on the Heat? was introduced in the blog post Everything You Ever Wanted to Know about El Niño and La Niña… …Well Just about Everything. The Free Preview includes the Table of Contents; the Introduction; the beginning of Section 1, with the cartoon-like illustrations; the discussion About the Cover; and the Closing.
Please buy a copy. (Credit/Debit Card through PayPal. You do NOT need to open a PayPal account.) Simply scroll down to the “Don’t Have a PayPal Account” purchase option. It’s only US$8.00.
THE ENSO MYTH SERIES
1. El Niño-Southern Oscillation Myth 1: El Niño and La Niña Events are Cyclical. Refer also to the cross post at WattsUpWithThat for comments.
3. Myth – ENSO Has No Trend and Cannot Contribute to Long-Term Warming. The WattsUpWithThat cross post is here.
4. ENSO Myth Number 4 – The Variations in the East Pacific and the East Indian-West Pacific Sea Surface Temperatures Counteract One Another. See the WUWT cross post here for more comments.
Myth – The Effects of La Niña Events on Global Surface Temperatures Oppose those of El Niño Events
Failed Argument – El Niño Events Don’t Create Heat
Myth – El Niño Events Dominated the Recent Warming Period Because of Greenhouse Gases
Myth – The Warm Water Available for El Niño Events Can Only be Explained by Anthropogenic Greenhouse Gas Forcing
Myth – The Frequency and Strength of El Niño and La Niña Events are Dictated by the Pacific Decadal Oscillation
The Sea Surface Temperature anomaly data are available through the NOAA NOMADS website:
GISS Land-Ocean Temperature Index data are available through the KNMI Climate Explorer.
And the GISS aerosol optical depth data can be found here.