Guest post by Bob Tisdale
The first part of this post, Can El Nino Events Explain All of the Global Warming Since 1976? – Part 1, should be read prior to this the second part. Part 1 gives an overview of the datasets used in the following, illustrates the processes that take place during an El Nino event, and discusses the primary reasons for the step changes in global SST anomalies that result from significant El Nino events–those El Nino events that are not influenced by volcanic eruptions.
In the following, the periods from January 1981 to December 1995 and from January 1976 to December 1981 are examined.
THE STEP CHANGE FROM 1981 TO 1995
As noted in the introduction (Part 1), the volcanic eruptions of El Chichon in 1982 and Mount Pinatubo in 1991 interrupted the normal heat distribution processes of the El Nino events that occurred at or near the same time. Figure 14 illustrates the East Indian-West Pacific SST anomalies, scaled NINO3.4 SST anomalies, and scaled (inverted) Sato Index data for the period of January 1981 to December 1995. (This is another graph you may wish to open in a separate window to keep you from having to scroll back and forth.) Again, the Sato Index and NINO3.4 SST anomaly data are not scaled to any specific level; they are provided for timing purposes only. The volcanic eruptions show up as the two depressions in the Sato Index data (green curve). The smoothing rounds off the start time of the Sato data, making it appear as though the Mean Optical thickness reacted prior to the eruption, but because the SST data is smoothed as well the impact on the discussion is nil.
The 1982/83 El Nino was the ENSO event with the second highest NINO3.4 SST anomaly of the 20th Century, yet there was little to no response by the East Indian-West Pacific SST anomalies to it. The El Chichon eruption effectively suppressed the heat distribution of that El Nino to the East Indian and West Pacific Oceans. In fact, the East Indian-West Pacific SST anomalies reacted quite sharply to the El Chichon eruption; they dropped quickly. Then as the volcanic aerosols subsided, East Indian-West Pacific SST anomalies rebounded to approximately the same level they had been at before the eruption. Considering the lags in the response of the East Indian-West Pacific SST anomalies to El Nino events, part of that rebound from mid-1982 to mid-1983 may be attributable to the 1982/83 El Nino. Then, from mid-1983 to mid-1986, East Indian-West Pacific SST anomalies modulated slightly until being swept up by the 1986/87/88 El Nino, lagging by approximately 7 months. While the SST anomalies of the 1986/87/88 El Nino did not peak as high as the 1982/83 El Nino, the 1986/87/88 El Nino lasted through the summer of 1987, making it a substantial ENSO event. The response of the East Indian-West Pacific SST anomalies was similar to that of the 1997/98 El Nino inasmuch as East Indian-West Pacific SST anomalies shifted significantly (eyeballing it, more than 0.12 deg C at the lowest level after the 1986/87/88 El Nino). East Indian-West Pacific SST anomalies then rose slightly as NINO3.4 SST anomalies rebounded from the 1988/89 La Nina. Note that, like the response to the 1998/99/2000 La Nina, there was little to no response of the East Indian-West Pacific SST anomalies to the 1988/89 La Nina. Then in 1991 two events, the Mount Pinatubo eruption and the beginning of a multiyear El Nino, occurred at the same time. Due to the magnitude of the Mount Pinatubo eruption, and likely its location in the West Pacific, East Indian-West Pacific SST anomalies dropped almost 0.25 deg C over approximately two years. When East Indian-West Pacific SST anomalies finally did rebound, possibly due to the ongoing multiyear El Nino, they did not return to their pre-1991 elevated levels.
In Figure 15, the SST anomaly data for the East Pacific, Atlantic, and West Indian Oceans (red curve) were added to the comparative graph. East Pacific-Atlantic-West Indian Ocean SST anomalies rise and fall from 1981 to 1991, mimicking the variations in NINO3.4 SST anomalies. There was no visible response by the East Pacific-Atlantic-West Indian Ocean SST anomalies to the El Chichon eruption in 1982.
A step change in East Pacific-Atlantic-West Indian Ocean SST anomalies occurs during that period as well. Following the 1986/87/88 El Nino, East Pacific-Atlantic-West Indian Ocean SST anomalies react to the subsequent 1988/89 La Nina. Then the East Pacific-Atlantic-West Indian Ocean SST anomalies (red curve) rise in response to the rebound in NINO3.4 SST anomalies until they nearly match the East Indian-West Pacific SST anomalies (black curve), and they remain at that elevated level. That is, prior to the 1986/87/88 El Nino, the mean of the East Pacific-Atlantic-West Indian Ocean SST anomalies (peak to trough) was approximately 0.05 deg C, but after it, the East Pacific-Atlantic-West Indian Ocean SST anomalies remained almost 0.1 deg C higher, with some minor fluctuations. A final note, the East Pacific-Atlantic-West Indian Ocean SST anomalies did not drop in response to the Mount Pinatubo eruption, but it appears Mount Pinatubo limited the rise of the East Pacific-Atlantic-West Indian Ocean SST anomalies to the El Nino. The minor rise in East Pacific-Atlantic-West Indian Ocean SST anomalies (red curve) countered the significant decrease in the East Indian-West Pacific SST anomalies (black curve), each responding to different natural events and making it appear that there was little reaction in the global SST anomalies to the Mount Pinatubo eruption or the El Nino at that time.
In summary, referring to Figure 16, which is the same graph as Figure 3 (Part 1), the step change in global SST anomalies between 1981 and 1995 was in response to the 1986/87/88 El Nino. The volcanic eruptions of 1982 and 1991 suppressed the normal step response to El Nino events at those times.
THE STEP CHANGES FROM 1976 TO 1981
Note: I changed the smoothing to a 5-month running-average filter for this period.
The East Indian-West Pacific SST anomalies and scaled NINO3.4 SST anomalies for the period of January 1976 to December 1981 are illustrated in Figure 17. There was no volcanic activity during the period, so I deleted the Sato Index data. East Indian-West Pacific SST anomalies rose first (eyeballing it, approximately 0.1 deg C) in a lagged response to the first half of the 1976/77/78 El Nino, then rose again (approximately another 0.03 to 0.04 deg C), responding to the second half of that El Nino.
Then the East Indian-West Pacific SST anomalies respond in a way that was in no way typical of their reaction to all other El Nino events. It may not be unusual if we take a closer look at the 1979/80 El Nino, which was unusual on its own. Refer to Figure 18, which is the raw and smoothed NINO3.4 SST anomaly data for the period of January 1976 to November 2008. The 1979/80 El Nino was not a significant El Nino; its NINO3.4 SST anomalies barely rose above the threshold of 0.5 deg C for a few months. It is so minor it does not register as an El Nino event on the ONI Index. It peaked at approximately 0.7 deg C. It also appears as a gradual rise and fall of NINO3.4 SST anomalies, not a sudden spike typical of other El Ninos.
In Figure 19, the SST anomaly data for the East Pacific, Atlantic, and West Indian Oceans (red curve) were added to the graph. The East Pacific-Atlantic-West Indian Ocean SST anomalies again mimic NINO3.4 SST anomalies, making a specific point at which they acquire an upward step difficult to determine. Note, however, that there are underlying steps in the East Pacific-Atlantic-West Indian Ocean SST anomalies that show themselves in the values at the minimums of its cycles in 1976, 1978, and 1980. In other respects it appears that the East Pacific-Atlantic-West Indian Ocean SST anomalies (red curve) are simply following a “baseline” established by the East Indian-West Pacific SST anomalies (black curve). This could be accomplished by natural ocean-atmospheric heat transfer processes and ocean currents.
Note: The 1976 Pacific Climate Shift also occurred at the start of that period. I illustrated the changes in various SST subsets in that post and the possible influence of the Southern Ocean on the 1976 Pacific Climate Shift.
Figure 20 is a simple recap of the cause of the step change in SST anomalies from 1976 to 1981. It was due primarily to the shift of the SST anomalies in East Indian and West Pacific Oceans in response to the 1976/77/78 El Nino.
A CONFIRMING PHENOMENON?
Two of my first three posts on this blog (Is There A Cumulative ENSO Climate Forcing? & Is There a Cumulative ENSO Forcing? Part 2) dealt with a phenomenon I had discovered in the long-term NINO3.4 SST anomalies provided as part of a Trenberth and Stepaniak study. The appropriate citations are included in the posts linked above. That NINO3.4 SST anomaly data is in fact HADSST data and uses 1950 to 1979 as base years. The dataset and base years are critical for the following. One question I can’t answer is why Trenberth and Stepaniak chose 1950 to 1979 as base years, but using those base years helped to create a unique response when a running total of that NINO3.4 SST anomaly data is graphed. Note the shape of the curve in Figure 21.
(I’ll update that running total graph as soon as I get a chance.) The curve mimics the curve of global temperature anomaly time-series data. The scale is wrong, but the proper coefficient would account for that.
Do the step changes illustrated in this post provide a mechanism for this phenomenon? And does the running total confirm that El Nino events are the primary driver of global temperature?
Figure 22 is a graph of NINO3.4 SST anomaly data from 1976 to 2008 in which I’ve noted El Nino events that were impacted by volcanic eruptions. The questions that came to mind were: What would have happened if El Chichon eruption had NOT been disturbed the heat distribution process of the 1982/83 El Nino? Would the equatorial Pacific have needed all of the additional El Ninos to distribute heat to higher latitudes? The same questions apply to the Mount Pinatubo eruption since it delayed the distribution of equatorial heat another few years.
To check my earlier graphs and to assure that the step changes illustrated in the preceding were not resident in the ERSST.v2 data alone, I plotted the four major datasets again, but this time using HADSST2 data available through the KNMI website. Refer to Figure 23. The same step changes and responses to volcanic eruptions appear in the HADSST data.
There will be those who will note that I used the word “Global” in numerous graphs in this post when in fact I had used data within the coordinates of 60S to 65N, 180W to 180E, excluding the Arctic and Southern Oceans.
It just seemed more appropriate to me to illustrate datasets within the same longitudes.
And there will be those who believe I was misrepresenting the data or hiding additional warming in the areas I excluded.
Nothing could be more from the truth. But to prove the longitudes had little effect on this discussion, Figure 24 is a comparative graph of the two primary datasets used throughout this post, the East Indian-West Pacific SST anomalies (black curve) and the East Pacific-Atlantic-West Indian Ocean SST anomalies (red curve), compared to GLOBAL [90S to 90N, 180W to 180E] SST anomalies.
In summary, step changes in global SST (and global surface temperature) result from El Nino events because warm water that was once below the surface of the Pacific Warm Pool (and not part of the instrument temperature record) is driven to the surface and eventually returned to the surface of the East Indian and West Pacific Oceans (making it a significant part of the instrument temperature record). The other major point of this post was that the heat distribution associated with El Nino events did not occur for all of El Ninos since 1976. The El Chichon and Mount Pinatubo explosive volcanic eruptions suppressed the heat distribution of the 1982/83, the 1991/92, the 1993, and possibly the 1994/95 ENSO events.
Smith and Reynolds Extended Reconstructed SST Sea Surface Temperature Data (ERSST.v2) is available through the NOAA National Operational Model Archive & Distribution System (NOMADS).
It is also available through the KNMI webpage listed below.
The Sato Index Data is available from GISS at:
The HADSST data is available through the KNMI Climate Explorer website. http://climexp.knmi.nl/selectfield_obs.cgi?someone@somewhere