Jeff Masters Blog

The El Niño event of 2015-16 is now history, according to NOAA’s Climate Prediction Center (CPC), which certified the demise in an advisory on Thursday. It wasn’t exactly a tough call: the warm equatorial waters over the eastern tropical Pacific that have signaled El Niño’s presence for more than a year are pretty much gone. El Niño events are defined mainly by sea surface temperatures (SSTs) that are least 0.5°C above average across the benchmark Niño3.4 region, which straddles the central and eastern equatorial Pacific. CPC also considers whether the atmosphere and ocean are acting in sync to support El Niño conditions (for example, trade winds over the tropical Pacific typically weaken as El Niño sets in). In the months since El Niño reached its peak intensity in November 2015, sea surface temperatures (SSTs) have steadily dropped across the Niño3.4 region, and other measures of El Niño strength have waned as well. Last week’s Niño3.4 SST anomaly (departure from average) dipped to -0.2°C, its lowest value since mid-March 2014. Below the surface of the tropical equatorial Pacific, cool anomalies now predominate, in part because of a strong upwelling oceanic Kelvin wave--an eastward-moving impulse that’s brought cooler water toward the surface along its path. Just as a downwelling Kelvin wave can help stimulate El Niño’s formation, upwelling Kelvin waves often feed into the development of La Niña, as one seems to be doing right now.


Figure 1. Anomalies (departure from average) in ocean temperature (degrees C) below the equatorial Pacific. The widespread blue values are consistent with the demise of El Niño and the expected development of La Niña. Image credit: NOAA/NWS/CPC.


Figure 2. Weekly anomalies (departure from average) in sea surface temperature (SST, in degrees C) across the “Niño regions” of the eastern and central tropical Pacific (see map at top of image). SSTs in the Niño3.4 region (third panel up from bottom) are at least 0.5°C above average during El Niño and at least 0.5°C below average during La Niña. NOAA employs a separate SST analysis across overlapping three-month intervals when assessing the total duration of an El Niño event. Image credit: NOAA/NWS/CPC.


An El Niño event becomes official for the record books once El Niño conditions have been in place for at least five overlapping three-month periods. By this definition, the most recent El Niño began in February-April-May 2015 and will most likely end in April-May-June 2016. Many El Niño events last just one year. This one would have been a rare two-year episode, extending back to late autumn 2014, were it not for a single three-month period (Jan-Mar 2015) that fell just below the El Niño threshold in the ocean dataset that NOAA uses (ERSSTv4).

A global footprint
Like the previous “super” El Niño events of 1982-83 and 1997-98‚ which are the only comparably strong events in the last 66 years of NOAA records, El Niño 2015-16 was a high-impact phenomenon. Among the repercussions linked to it:

--Devastating drought, fire, and air pollution in Indonesia and neighboring countries in fall 2015, which caused many thousands of people to fall ill and cost Indonesia more than $16 billion

--The wettest month on record across Texas and Oklahoma (May 2015), with subsequent rounds of torrential rain and major flooding over parts of the south-central U.S. in late 2015 and early 2016

--South Africa’s most expensive natural disaster on record, a severe drought culminating in the summer of 2015-16 that hammered crops and water supplies

--An intense heat wave in spring 2016 across Southeast Asia, with more than half of the major weather stations in Thailand setting all-time record highs

--Record-warm upper-ocean temperatures that produced the third global coral bleaching event on record and caused massive damage to the Great Barrier Reef, where more than 20% of the entire reef’s coral may have been killed.


Figure 3. Left: Coral researcher Kim Cobb (@coralsncaves, Georgia Institute of Technology) placing a conductivity-temperature-depth sensor on a healthy reef at Kirimati Island in the Southwest Pacific several years ago. Right: The same reef as of late April 2016, showing a large number of dead and dying coral colonies as a result of prolonged above-average water temperatures. Learn more about Cobb’s research and the devastation to coral reefs this year at this ENSO Blog entry (May 27, 2016). Image credits: Pamela Grothe (left), Kim Cobb (right).


One place where El Niño failed to behave as expected was Southern California, where forecasts and historical analogs suggested that heavier-than-usual rains were quite likely after four years of punishing drought. Instead, the West Coast rains were shunted hundreds of miles north of where they usually materialize during strong El Niño events. Seattle ended up with its wettest winter on record--a stupendous 45.51” of rain from October 1 through June 7--while Los Angeles saw only 6.88”, just 47% of its long-term average for that same period. Although the absence of El Niño rains was a big blow to Southern California, the unusual outcome together with ample data collected by a rapid-response NOAA field project in the Pacific may help spur some fascinating and important research.


Figure 4. Percent of average precipitation for the period from October 1, 2015, through June 7, 2016, based on provisional data. Most of the rainfall in a West Coast water year (October 1 - September 30) has fallen by June. Image credit: Western Regional Climate Center.

La Niña into 2018? It’s quite possible
There is high confidence (though not ironclad certainty) that El Niño will be quickly replaced later this year by La Niña. Even though the interplay between ocean and atmosphere that leads to El Niño is not fully understood, the mechanics of its downfall are more clear-cut, as explained nicely in a NOAA blog post last January. In a nutshell, El Niño’s inclination to push both wind and water eastward across the tropical Pacific can trigger oceanic Rossby waves. These huge, shallow, slow-moving impulses on either side of the equator can move west, bounce off the west end of the Pacific basin, and return to stimulate a cooling of the central and eastern tropical Pacific, often putting an end to El Niño and sometimes ushering in La Niña. (This process is in addition to the oceanic Kelvin waves noted above, with various types of interrelationships adding to the complexity.) On top of conceptual understanding, we know from past experience that most strong El Niño events tend to be followed rather promptly by at least a year of La Niña conditions. Moreover, seasonal climate models that include oceanic conditions agree strongly on the development of La Niña later this year. With all this in mind, the June outlook from forecasters at NOAA and the International Research Center for Climate and Society (IRI), released on Thursday, gives La Niña a 70-75% chance of being in place by late summer and continuing into at least early 2017.


Figure 5. The early-June outlook issued by forecasters at the NOAA/NWS Climate Prediction Center and the International Research Center for Climate and Society. Image credit: IRI/CPC.


Because many La Niña events last two or three years, there is already a slightly enhanced probability of La Niña in 2017-18, if history is any guide. A 2014 modeling study by Pedro DiNezio (@txgaucho, now at the University of Texas) and Clara Deser (National Center for Atmospheric Research) used a highly sophisticated global climate model (CCSM4) to analyze 252 La Niña events that appeared in 1300 years of simulated climate. The study found that about a third of the first-year La Niña events returned or persisted for a second year, but with marked variations from century to century. The historical record shows that second-year La Niñas have been even more frequent in the last 150 years than in the 1300-year simulation. WU member Eric Webb (@webberweather, North Carolina State University) has examined El Niño and La Niña events going back to 1865 using a blend of multiple datasets from several agencies. From the 23 “first-year” La Niña events in Webb’s database (focusing on the core winter period of December through February, or DJF), 11 of those events featured at least one more subsequent year of La Niña. Similarly, in the official NOAA database going back to 1950, exactly half of the 12 first-year La Niña winters were followed by a second year of La Niña. (NOAA’s Emily Becker digs further into the post-1950 statistics in a recent blog post.)

Considering this climatology, if we assume that La Niña materializes in 2016-17 as expected, the historical odds are about even that 2017-18 would also be a La Niña winter (potentially with a short neutral break in between). I wouldn’t bet the farm on this outcome, though--especially since the 75% odds of getting a La Niña this coming autumn, while quite high, are not a guarantee! “Some folks are pointing out that the subsurface cooling is not very strong relative to some past events,” says Michelle L’Heureux (NOAA/CPC), “so we still have to see if it is enough (and that the atmosphere couples to it, because right now there is no such coupling).” The next step toward making an actual La Niña forecast for 2017-18 would be to get past the early-2017 “spring predictability barrier”, when ongoing El Niño and La Niña events tend to weaken and climate model guidance become less robust.

What shoaling can tell us
Here’s one promising research angle that may help: The DiNezio-Deser study cited above found that the depth of the thermocline (the oceanic boundary that separates warmer surface waters from cooler subsurface waters) six months before the onset of La Niña was correlated with SSTs a year and a half later. When the thermocline shoals more strongly (becomes shallower) after El Niño, then there is a greater chance of a subsequent multi-year La Niña. According to DiNezio and Deser, this process could provide an 18-month lead time for predicting whether La Niña conditions might return a second year. In general, the thermocline shoals much more strongly after a strong El Niño event, which raises the odds that the subsequent La Niña will last more than a year. After the “super” El Niño of 1997-98, the thermocline rose by more than 40 meters (130 feet), and the subsequent La Niña lasted for three years. The El Niño of 2015-16 was just as strong as in 1997-98, but as of late May 2016, the thermocline had shoaled by no more than 20 meters (65 feet). “This makes the prediction of the return of La Nina for an additional year more challenging,” said DiNezio in an email. “The duration of the upcoming La Nina is not set in stone and will be difficult to predict. I like this event because it makes the prediction so much more challenging!”