>90%
chance of a very strong El Niño this fall and winter
69%
chance of a historic event — stronger than anything since 1950
25/26
forecast models calling it very strong at the projected peak

NOAA Climate Prediction Center, 13 August 2026 · IRI, 19 August 2026.

1Where the event stands

An El Niño Advisory is in effect. In its diagnostic discussion of 13 August 2026, NOAA's Climate Prediction Center stated that El Niño is strengthening, with a greater than 90 percent chance of a very strong event through the Northern Hemisphere fall and winter of 2026–27.

The latest weekly sea-surface temperature departures published by CPC show something worth pausing on — the warmth is heavily concentrated in the eastern Pacific:

Monitoring regionLatest weekly departureWhat it covers
Niño 1+2+3.5 °CFar eastern Pacific, off South America
Niño 3+2.6 °CEastern Pacific
Niño 3.4+1.8 °CThe standard monitoring region
Niño 4+0.1 °CWestern/central Pacific — essentially neutral

Latest weekly departures as published in CPC's ENSO status update of 31 August 2026.

That spread — very warm in the far east, neutral in the west — is the signature of an eastern-Pacific El Niño. It is the same spatial pattern as 1982–83, 1997–98 and 2023–24, and it is not the pattern of 2015–16. Keep that in mind; it matters a great deal by the end of this article.

The model consensus is unusually tight. The International Research Institute for Climate and Society puts El Niño probability at 100 percent from August–October 2026 through February–April 2027. At the projected October–December peak, 25 of 26 international models forecast very strong conditions, and 15 forecast values at or beyond +3.0 °C — which IRI itself describes as well beyond its highest defined intensity category.

2How fast it built

The speed of the onset is the part most coverage has missed. This time last year the Pacific was heading the other way.

Line chart of the Oceanic Niño Index from winter 2025 through mid-2026, rising from La Niña to +1.8
Figure 1. The Oceanic Niño Index, winter 2025 through summer 2026. Each point is a three-month running mean. The winter of 2025–26 was a La Niña winter (Nov–Jan −0.6). The index crossed the +0.5 El Niño threshold in Mar–May 2026 and reached +1.8 by Jun–Aug 2026 — the most recent published value. The forecast band shows the projected Oct–Dec peak. Source: NOAA Climate Prediction Center.

Seven months separate a La Niña winter from the brink of a very strong El Niño. Rapid transitions like this are not unprecedented, but they compress the time available to prepare — and they tend to arrive faster than public messaging adjusts.

3A necessary word about the index

Almost every number you will read about El Niño strength is an Oceanic Niño Index value: the three-month average sea-surface temperature anomaly in the Niño-3.4 region, measured against a thirty-year baseline. NOAA officially defines El Niño as five consecutive overlapping seasons at or above +0.5 °C.

The familiar tiers — weak, moderate, strong, very strong at +0.5, +1.0, +1.5 and +2.0 — are a widely used convention rather than an official NOAA classification. They come from Golden Gate Weather Services and have become the common vocabulary. Worth knowing, because it means "very strong" is a label with a soft edge, not a defined regulatory threshold.

There is a deeper problem with the raw index, and in February 2026 NOAA acted on it.

The traditional Oceanic Niño Index (ONI) relies on a departure from a 30-year average that struggles to keep pace with anomalous changes in tropical sea surface temperature… RONI solves this problem by comparing the ENSO region to the global tropics, thereby reducing the dependency on the climate base period. NOAA Climate Prediction Center, on adopting the Relative Oceanic Niño Index

In plain terms: the entire tropical ocean has warmed. An anomaly measured against a lagging baseline therefore looks bigger than the actual El Niño signal. The Relative ONI subtracts the average anomaly across the whole tropical belt (20°S–20°N) and leaves what is genuinely El Niño.

The most misreported number in this event

CPC's 69 percent chance of a "historic" event refers to a three-month RONI value of +2.5 °C or more — not the ONI. Anyone quoting it as "a 69 percent chance the ONI passes +2.5" has the wrong index. Because RONI currently runs roughly half a degree below ONI, that threshold corresponds to an ONI somewhere around +3.0 — which would be above anything in the record.

4So — are El Niños getting stronger?

This is where the raw index and the corrected one part company, and it is the most interesting thing in the data.

Bar chart of peak ONI for every El Nino since 1950 with RONI markers on six major events
Figure 2. Peak intensity of every El Niño event since 1950. Bars show peak ONI; orange diamonds show peak RONI where CPC publishes it for the major events. On the raw index, the three strongest events on record are 2015–16, 1997–98 and 1982–83 — all within the last 45 years. Corrected for background tropical warming, the ranking reorders: 1982–83 becomes the strongest on record, 1972–73 rises from strong into the very strong tier, and 2023–24 falls from +2.0 to +1.5. Sources: NOAA CPC ONI table (ERSST v6) and RONI table (ERSST v5); the two are not version-matched.

Look at the raw bars alone and the conclusion seems obvious: El Niño events have been getting stronger, and the three biggest all happened recently. That is the version that usually makes the news.

Now look at the diamonds. Once the background ocean warming is removed, the apparent trend largely dissolves. The correction lifts the older events and lowers the recent ones. The strongest El Niño on the corrected record is no longer 2015–16 but 1982–83, at +2.5. The 1972–73 event — which the raw index files as merely "strong" at +1.8 — rises to +2.3 and joins the very strong tier. And 2023–24, reported everywhere as a very strong El Niño, drops from +2.0 to +1.5.

The scientific literature is consistent with this, and it draws a distinction that most coverage collapses. The IPCC's Sixth Assessment Report is explicit about El Niño's temperature amplitude:

There is no model consensus for a systematic change in intensity of ENSO sea surface temperature variability over the 21st century in any of the SSP scenarios assessed (medium confidence). IPCC AR6, Working Group I, Chapter 4

But about El Niño's rainfall behaviour, the same chapter is confident in the opposite direction:

It is very likely that ENSO rainfall variability… will increase significantly, regardless of amplitude changes in ENSO SST variability, by the second half of the 21st century. IPCC AR6, Working Group I, Chapter 4
The distinction that resolves the argument

The widely cited work of Cai and colleagues projecting a doubling of extreme El Niño events defines "extreme" by the reorganisation of atmospheric convection and rainfall — not by a sea-surface temperature threshold. So that finding and the IPCC's "no consensus on SST amplitude" are not in conflict. They are measuring different things.

The defensible statement is therefore not "El Niños are getting stronger." It is: the ocean signal is not clearly intensifying, but the atmospheric impacts that run through it very likely are. That is a more useful sentence for anyone planning around one.

5What El Niño actually does to rainfall

For Southern California, the relationship is real, useful, and routinely oversold.

Bar chart of Los Angeles water-year rainfall in twelve El Nino years against normal
Figure 3. Los Angeles water-year rainfall in twelve El Niño years, 1958–2024. The water year runs 1 October to 30 September. Normal is 14.25 inches, the 1991–2020 annual normal for Los Angeles Downtown. Ten of twelve events finished at or above normal; the median was about 140 percent of normal. Source: NOAA NCEI, station USW00093134.

Ten of twelve is a meaningful tilt. It is not a forecast. And the two misses are the instructive part:

Note also that event strength and rainfall do not line up in any orderly way. The very strong events produced both the wettest years in the set (1983 at 233 percent, 1998 at 215 percent) and the driest. Plotted against peak ONI, these totals form no trend at all.

6What it does to the coast

The oceanographic signal is more reliable than the rainfall signal, and it arrives earlier. It is also the part most people overlook, because it does not look like weather.

The interaction most people miss

A wet El Niño erodes beaches hard but flushes river sediment to the coast, and the beaches rebuild. A dry El Niño — 2015–16 — delivers the wave energy without the sand. Beaches are stripped and recovery is slow, which leaves less protective beach width going into the following winter. The dry scenario is not the harmless one for the coast. In some respects it is the worse one.

7What El Niño does not do

Three things are worth stating plainly, because each one has caught planners out before.

It does not guarantee a wet winter

2015–16 is the standing rebuttal. Two peer-reviewed explanations compete — one attributing the failure to record-warm Indian Ocean and western Pacific temperatures that displaced the storm track north, the other arguing the seasonal signal was simply overwritten by subseasonal variability. Either way the lesson is the same: the seasonal signal can be entirely overwritten at the timescale on which storms actually happen.

It does not act alone

Analysis of the extreme February 2024 rainfall found El Niño was necessary but not sufficient. It set a permissive background state; the flooding required an exceptionally long-lived Madden–Julian Oscillation event interacting with the North Pacific jet for more than three weeks. El Niño loads the dice. Something else still has to throw them.

Seasonal rainfall totals do not predict damage

This is the least intuitive finding in the record. California's costliest flood seasons are not its El Niño seasons. In one widely used compilation, the 1996–97 season — which occurred in cool-neutral conditions — caused roughly $2.94 billion in California flood damage. The very strong 1997–98 El Niño that followed produced nearly double normal rainfall and roughly $0.86 billion, the least damaging of the thirteen costliest seasons in that set, because the rain fell across nearly double the normal number of rain days.

The practical consequence

Damage is driven by rainfall intensity and storm clustering, not by seasonal total. An inch of rain in an hour matters more than thirty inches across a season. Anyone tracking season-to-date accumulation as a proxy for risk is watching the wrong number.

8What to watch this winter

9Sources