History suggests: don't bet on La Nina this year

Typical (Average) El Nino, Traditional El Nino, and El Nino Modoki Events

Guest Post by Bob Tisdale

Recently, there have been a number of posts around the blogosphere about the current El Nino or about Sea Surface Temperatures (SST). Accompanying them are predictions by the authors of those posts or by commenters of a pending La Nina event. But the “typical” El Nino event is not followed by a La Nina event. Also, the current 2009/10 El Nino event is an El Nino Modoki; that is, simply, the area with elevated SST anomalies is located more towards the center of the tropical Pacific than a traditional El Nino event; and few La Nina events follow El Nino Modoki.

A number of months ago I noticed some of my visitors arrived from Google searches of “typical El Nino” or “average El Nino”. I prepared this post for them back then but got sidetracked and never posted it.

This post looks at the development and decay of the average El Nino, of the average traditional El Nino, and of the average El Nino Modoki. I’ve also segmented the data into two periods, before and after 1979 to illustrate the change in development and strength of El Nino events. Last, as references, are spaghetti plots of the development and decay of all El Nino events since 1950 (excluding the current El Nino, since it’s not complete). The post could also be used by those bloggers who like to make predictions or by those wanting to see whether prognostications have any basis in history.

THE AVERAGE EL NINO

Figure 1 illustrates the development and decay of the average El Nino event for the period of 1950 through 2007. It starts in January of the development year and ends in December of the following (decay) year. To create the graph, I averaged the SST anomaly (ONI) values for the 24 months associated with each official El Nino event identified on the CPC’s Oceanic Nino Index (ONI) webpage:

http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ensoyears.shtml

The average El Nino reaches the +0.5 deg C threshold of an El Nino in late May, peaks in December, then quickly decays until it drops below the +0.5 deg C El Nino threshold in mid March. The SST anomalies of the average El Nino do drop below zero, but during the following ENSO season they do not cross the -0.5 deg C threshold for a La Nina event.

http://i39.tinypic.com/k3vuvo.png

Figure 1

BEFORE AND AFTER 1979

The frequency and magnitude of ENSO events changed about 1976. Between the mid-1940s and the mid-1970s, La Nina events dominated (with a period of El Nino dominance in the 1960s), and after, El Nino events were dominant. This can be illustrated with a long-term graph of NINO3.4 SST anomalies smoothed with a 121-month filter, Figure 2.

http://i43.tinypic.com/33agh3c.jpg

Figure 2

But studies such as Trenberth et al (2002) divide the data into periods before and after 1979, based on the development of El Nino events, so I’ve divided the data in this post at 1979. (The 1976/77 event was a weak traditional El Nino, and the 1977/78 El Nino was a weak El Nino Modoki.) Link to Trenberth et al (2002) “Evolution of El Nino–Southern Oscillation and global atmospheric surface temperatures”:

http://www.cgd.ucar.edu/cas/papers/2000JD000298.pdf

The average SST anomalies of the El Nino events before and after 1979 are shown in Figure 3. It comes as no surprise that El Nino events after 1979 are stronger and last longer than those before the cutoff year. Still, even in more recent decades, the average El Nino is not followed by a La Nina.

http://i40.tinypic.com/2wpto8w.png

Figure 3

TRADITIONAL EL NINO VERSUS EL NINO MODOKI

Central Pacific versus Eastern Pacific El Nino events are discussed in a number of recent papers. Ashok et al (2007) “El Nino Modoki and its Possible Teleconnection”… https://www.jamstec.go.jp/frcgc/research/d1/iod/publications/modoki-ashok.pdf

…provides an equation that can be used to identify El Nino Modoki:

“EMI= [SSTA]A-0.5*[SSTA]B-0.5*[SSTA]C …(1)

“The square bracket in Equation (1) represents the area-averaged SSTA over each of theregions A (165E-140W, 10S-10N), B (110W-70W, 15S-5N), and C (125E-145E, 10S-20N), respectively.”

Ashok et al further describe the basis for their selection of El Nino Modoki events: “Based on the time series of the EMI shown in Figure 4a, we have identified seven typical El Niño Modoki events that lasted from boreal summer through boreal winter, peaking in one of these seasons (seasonal standard deviations for boreal summer and winter are 0.5ºC and 0.54ºC respectively). These typical El Niño Modoki events occurred in 1986, 1990, 1991, 1992, 1994, 2002, and 2004. Additionally, we identified a typical El Niño Modoki during the boreal winter of 1979-80 that lasted through the summer of 1980, though its amplitude fell below the threshold of 0.7 σ by then.” And they clarify with the footnote, “We call an El Niño Modoki event ‘typical’ when its amplitude of the index is equal to or greater than 0.7 σ, where σ is the seasonal standard deviation.”

Ashok et al appeared to use two definitions of an El Nino Modoki: first, the average of boreal summer through boreal winter for most events, and, second, the average of the boreal winter for the 1979 event. Using the average boreal summer through winter (June through February) El Nino Modoki Index and the boreal winter El Nino Modoki Index, Figure 4, as references, I’ve identified the typical El Nino Modoki events before 1979 (based primarily on the boreal winter data when they conflict). These along with traditional El Nino events are shown in Table 1, as are the breakdown of El Nino events after 1979.

http://i40.tinypic.com/16kc3kg.png

Figure 4

###############

http://i39.tinypic.com/24e7v2t.png

Table 1

Note 1: El Nino Modoki events identified by Ashok et al that do not qualify as official El Nino events on the ONI Index have been excluded.

Note 2: As illustrated in Table 1, there were more El Nino Modoki before 1979 than after, yet in press releases we’re advised that El Nino Modoki events are new, and that this NEW TYPE is resulting in a greater number of hurricanes with greater frequency and more potential to make landfall.” Refer to the press release…http://media-newswire.com/release_1094000.html…for the Hye-Mi Kim, et al (2009) paper “Impact of Shifting Patterns of Pacific Ocean Warming on North AtlanticTropical Cyclones”:

http://www.sciencemag.org/cgi/content/abstract/325/5936/77

The press release describes El Nino Modoki as “new” more than once. The “newness” of El Nino Modoki was also contradicted by data in my July 6, 2009 post There Is Nothing New About The El Nino Modoki.

Figure 5 compares the average El Nino Modoki and Traditional El Nino event since 1950. The typical Traditional El Nino is stronger than the El Nino Modoki and it results in a La Nina event, where the typical El Nino Modoki decays to a neutral SST anomaly of ~0.

http://i44.tinypic.com/5pkhn8.png

Figure 5

MORE COMPARISONS

Figures 6 through 9 provide further comparisons of El Nino Modoki and Traditional El Nino events before and after 1979. I won’t discuss these individually, other than to call your attention to the comparison of El Nino Modoki and Traditional El Nino events prior to 1979, Figure 8. Note that El Nino Modoki events were stronger and their durations were longer than Traditional El Nino events.

http://i41.tinypic.com/qz222u.png

Figure 6

###############

http://i41.tinypic.com/bk4ux.png

Figure 7

###############

http://i42.tinypic.com/dqzon.png

Figure 8

###############

http://i39.tinypic.com/elcfa0.png

Figure 9

###############

COMPARISONS OF INDIVIDUAL EL NINO EVENTS

Figure 10 compares the ONI SST anomalies for the 8 Traditional El Nino events from 1950 to 2007. Dashes are used to identify the El Nino events before 1979. Of the 8 Traditional events, only two El Nino events did not transition into La Nina events. The 1976/77 El Nino was followed by the 1977/78 El Nino Modoki.

http://i44.tinypic.com/jtxvg9.png

Figure 10

And the 1951/52 El Nino was not followed by a La Nina. The 1951/52 El Nino is also anomalous in that it peaks before the typical El Nino peak months of November, December, and January. However, looking at maps of ICOADS SST anomaly data (the basis for the Hadley Centre and NCDC’s SST data) for the tropical Pacific for October through December 1951 and for January 1952, Figure 11, we can see that there were few to no SST readings during those months in the NINO3.4 region (and most of the tropical Pacific for that matter), so the 1951/52 El Nino data could be considered suspect. (Always keep in mind that much of SST data before the eras of buoys and satellites are infilled.)

http://i42.tinypic.com/qod3bq.png

Figure 11

And Figure 12 is a comparison of the 10 El Nino Modoki events. I’ve also identified the earlier events with dashes. Of the 10 El Nino Modoki, only 2 events transitioned into La Nina events, the 1963/64 and 1994/95 El Nino events. The SST anomalies during the ENSO season following the 2004/05 El Nino dipped below the La Nina threshold, but did not remain there long enough to be considered an official La Nina.

http://i44.tinypic.com/72deeq.png

Figure 12

CLOSING COMMENT

Will a La Nina follow the 2009/10 El Nino? Considering that only 2 of 10 El Nino Modoki events since 1950 were followed by La Nina events, the odds are against it. But nature does provide surprises.

SOURCES

The ONI data is available through the NOAA CPC webpage:

http://www.cpc.noaa.gov/products/analysis_monitoring/ensostuff/ensoyears.shtml

The HADISST SST anomaly data used for the El Nino Modoki graph, and the ICOADS data used for the tropical Pacific SST maps are available through the KNMI Climate Explorer:

http://climexp.knmi.nl/selectfield_obs.cgi?someone@somewhere

I also used the KNMI Climate Explorer to create the maps.