El Niño 2026: How Strong It Is, and What It Actually Means

Quick Answer: What's Happening Right Now

The tropical Pacific is running a strengthening El Niño, and on 8 October 2026 NOAA rewrote the headline sentence of its discussion for the first time since August: a strong-to-very-strong event is now called likely through January–March 2027, and the chance of a historic peak, stronger than any El Niño since records began in 1950, rose again to 83% for October–December.

Measure	Where it stands
Official status	El Niño Advisory (in effect since 11 June 2026)
Latest official index	RONI +1.7 °C for July–August–September 2026
Chance it's "very strong" in Oct–Dec	100% (rounded)
Chance it exceeds every event since 1950	83% (Oct–Dec, RONI +2.5 °C or more)
Chance El Niño conditions last into Feb–Apr 2027	98%
Expected peak	Late autumn to early winter 2026

NOAA's Climate Prediction Center put it in one sentence on 8 October 2026:

"El Niño continues to strengthen, with a strong-to-very-strong El Niño likely through January-March 2027 (remaining greater than an 83% chance)."

That is not the sentence CPC ran in August and September, which was word for word identical in both months: "El Niño is strengthening, with a greater than 90% chance of a very strong event during the Northern Hemisphere fall and winter 2026-27." The change is not a retreat. The new sentence asks a different question, how long the event stays at strong or better, and the numbers under it all went up. October–December is now at 100% for very strong on CPC's table, up from 98%, and the chance of a historic event rose from 75% to 83%. CPC also now prices that record for three overlapping seasons instead of one: 54% for September–November, 83% for October–December and 70% for November–January. In July the synopsis had read "El Niño continues and will strengthen through the end of the year, with a 97% chance it will persist through early spring 2027," which is a forecast about duration rather than size. The emphasis moved to size in August and held there through September. With the autumn seasons already at 100%, October's sentence goes back to how long the event lasts.

A note on freshness: CPC re-issues its ENSO discussion on the second Thursday of each month. The most recent landed on 8 October 2026, and the next is due 12 November 2026. Every number in this post is stamped with the date it came from. If you're reading this well after publication, check the current discussion before quoting anything.

The Number You've Seen Is Probably One of Five

Here's the thing almost every article about this event gets wrong, and it's worth getting right before anything else.

If you've seen a number attached to this El Niño, it was probably +3.4, +2.5, +2.2, +2.1 or +1.7. All of them are correct, and all of them describe the same patch of ocean, the Niño 3.4 box. There are five values in circulation, and this month they land on five different printed figures. That is still most of why coverage of this event garbles them: three of the five sit within 0.4 °C of each other, so it is easy to quote one while meaning another. They differ on two axes: what window of time they average over, and whether background tropical warming has been removed.

Enlarge

Forecasters watch four boxes strung along the equator. Niño 3.4 (the middle one, straddling the Date Line side of the basin) is the one that defines the event, because that's where the ocean most reliably talks back to the atmosphere.

First axis: the time window. Two are a weekly snapshot, one is a single calendar month, and two are three-month averages.

Reading	What it is	Period
+3.4	Weekly, absolute	week of 21 Oct 2026
+2.5	Weekly, relative	week of 21 Oct 2026
+2.2	ONI (legacy)	July–September 2026
+2.1	Monthly, relative	September 2026
+1.7	RONI (official)	July–September 2026

Weekly values are noisy and jump around; the three-month averages are what actually classify an event. So the weekly +2.5 is not NOAA's classification number even though it's also "relative." RONI is, and it currently reads +1.7.

This month the table does not collide at all. The weekly absolute (+3.4), the weekly relative (+2.5), the legacy ONI for July–September (+2.2), the single-month September value on the relative index that CPC quoted in its 8 October discussion (+2.1) and RONI (+1.7) each sit at a figure of their own. That is an accident of where the numbers happen to sit this month, and it will not necessarily last. While it does, the thing to watch is the cluster in the middle: the weekly relative, the legacy ONI and the monthly relative all fall within 0.4 °C of each other and describe three different things. RONI, the one that classifies the event, is the odd one out, about half a degree below the legacy index that people still reach for.

Second axis: absolute vs relative.

• Absolute anomaly: how far above the 1991–2020 average the water is.
• Relative anomaly: the same thing, minus the average warming across the whole tropics.

That second one matters more than it sounds. The entire tropical ocean has warmed. If you only measure a departure from a fixed average, every recent El Niño looks bigger than the one before it, partly for reasons that have nothing to do with El Niño. Subtracting the tropics-wide background strips that out and leaves the part that's actually the Pacific see-saw.

RONI does one more thing after that subtraction: it rescales the result so its variance matches the traditional index. Without that step the numbers would all shrink and the familiar weak/moderate/strong thresholds would stop meaning anything. It's the reason RONI values are still comparable to the ONI scale you may remember.

NOAA made the switch official on 1 February 2026. RONI is now the index NOAA classifies El Niño and La Niña on, replacing the traditional ONI. The legacy files keep publishing for continuity, which is exactly why several numbers are floating around at once.

The 2023–24 event shows why the change was made. On the old index it peaked at +1.99 °C in November–January, a hair from the "very strong" cutoff. On RONI it peaked at +1.42 °C in October–December, which lands in the moderate category. Same event, two indices, and a two-category difference in how it gets filed.

(Those two peaks fall in different three-month seasons, so the gap between them isn't a single clean quantity you can subtract and label. The point is the classification, not the arithmetic.)

So when you see a headline number, the useful question is which index. For this post: RONI is +1.7 °C for July–September 2026, and that's the number NOAA's official probabilities are built on.

What El Niño Actually Is

Strip away the jargon and El Niño is a story about wind, and what happens when the wind gives up.

Normally, trade winds blow steadily east to west along the equator, dragging surface water toward Indonesia. Warm water piles up in the west, enough that sea level there sits roughly a foot and a half higher than off South America. With the warm surface layer hauled westward, cold water wells up along the South American coast to replace it.

That warm western pool is where the thunderstorms live. Warm water evaporates, air rises, rain falls, and the rising air flows back east high above before sinking over the cold eastern Pacific. That loop is the Walker circulation.

Enlarge

Now weaken those trade winds. The piled-up warm water sloshes back east. The boundary between warm surface water and cold deep water, the thermocline, flattens out. Upwelling off South America is suppressed. It doesn't stop so much as stop mattering, because with the thermocline pushed deeper it draws up warm water instead of cold. The eastern Pacific warms.

And here's the part that makes El Niño self-sustaining: warmer water in the east means a smaller temperature contrast across the basin, and it's that contrast that drives the trade winds in the first place. Weaker contrast, weaker winds. Weaker winds, warmer east. The ocean and atmosphere egg each other on, a loop meteorologists call the Bjerknes feedback.

The consequence that reaches you has nothing to do with water temperature directly. It's that the planet's largest concentration of thunderstorms picks up and moves several thousand miles east. That relocated heat engine punches into the upper atmosphere, bends the jet streams, and rearranges weather patterns continents away.

How This One Spun Up in Four Months

What makes the 2026 event worth writing about isn't just its size. It's how fast it assembled. CPC's own month-by-month verdict tells the story better than any paraphrase:

Enlarge

Discussion	CPC's read on the coupled ocean–atmosphere system
12 March 2026	"remained consistent with La Niña"
9 April 2026	"reflected ENSO-neutral conditions"
14 May 2026	"reflected ENSO-neutral conditions"
11 June 2026	"reflected the onset of El Niño conditions"
9 July 2026	"reflected a strengthening El Niño"
13 August 2026	"reflected a strengthening El Niño"
10 September 2026	"reflected a strengthening El Niño"
8 October 2026	"reflected a strengthening El Niño"

Seven months, cool-leaning to strengthening El Niño. The last four rows are word for word identical, and that is the point: since July, CPC's diagnosis of the coupled system has not moved at all. The odds attached to it have, and in October so did the headline sentence.

The ocean moved first. The volume of warm water stored in the equatorial Pacific, the fuel tank for any El Niño, went from below average in July 2025 to the highest single month in the entire record, which begins in 1980, by August 2026, the latest month posted. July 2026 is now second, and both edge past April 1997, the previous top month. That is a genuinely enormous recharge, and it happened in about thirteen months.

Enlarge

Then the atmosphere committed. Through spring, forecasters kept hedging, because a warm ocean alone doesn't make an El Niño. The atmosphere has to respond. CPC said as much on 14 May: "The strongest El Niño events in the historical record are characterized by significant ocean-atmosphere coupling through the summer, and it remains to be seen whether this occurs in 2026."

It occurred. Three independent atmospheric fingerprints flipped sign between March and July 2026:

• Trade winds collapsed. Low-level winds over the western Pacific didn't just weaken. They reversed to westerly, running about 2 standard deviations from normal. Over the central Pacific the anomaly hit −2.4σ.
• The storm belt moved. Outgoing longwave radiation near the Date Line, a proxy for deep convection, swung from +5.6 W/m² in March to −19.4 W/m² in July. Negative means more cloud, more storms. The thunderstorms relocated, right on cue.
• The pressure see-saw tipped hard. The Southern Oscillation Index went from +1.2 in March to −2.4 in July, the most negative reading of the sequence.

That's the difference between a warm patch of ocean and an actual El Niño. It's also what turned a hedged spring forecast into four months of escalation: a 63% chance of a very strong event in the 11 June discussion (for November–January), 81% on 9 July (October–December), then 95% on 13 August, 98% on 10 September and 100% on 8 October for that same season, with a record priced in on top.

How Strong Is It, Really?

Here's where honesty matters, because there are two defensible answers depending on which index you pick, and picking the flattering one is how "biggest El Niño ever" headlines get made.

Compare like with like: every event measured at the same point in the calendar, July–August–September.

Event	ONI	RONI	Peak RONI
1982–83	+0.99	+1.14	+2.40
1997–98	+1.79	+1.84	+2.28
2015–16	+1.73	+1.54	+2.25
2023–24	+1.25	+0.79	+1.42
2026–27	+2.16	+1.69	—

(The first two columns are July–September values for each event. ONI is the legacy index; RONI is the one NOAA now classifies on. The last column is what each event eventually peaked at.)

On the old index, 2026 is running ahead of every historical analogue at this point in the year, 1997 included; no other July–September in the ONI record, which begins in 1950, is as warm. On RONI, the index that actually classifies the event, it's ahead of 2015 and still behind 1997. That split is the whole argument for using the relative index: the legacy number says fastest start on record, the official one says second to 1997.

Both statements are true. The second one is the fair one.

Two more pieces of context worth holding onto:

• On the ocean-heat side, this event resembles 1997 more than 2015. The stored warm water in June 2026 was almost exactly the June 1997 value, and more than double anything the 2015–16 super El Niño ever managed. That's the fuel, not the fire, but it's a lot of fuel.
• The warmth is stacked toward South America. The far-eastern Niño 1+2 box read +5.4 °C in the week centered 28 October, the highest single week anywhere in the record that begins in 1981. It edges past the +5.3 °C readings of the weeks centered 30 September and 7 October, and the weeks centered 14 and 21 October each read +5.2 °C. The next three highest values are all from this year (+4.9, +4.7 and +4.6 °C, in the weeks centered 23 September, 16 September and 9 September), and 1983's peak of +4.5 °C is the only earlier week that comes close. Every other week above +4.0 °C belongs to 1983 or this year. (CPC revises these weekly values retroactively, so treat the exact ordering as provisional. The company it keeps is the durable part.) Meanwhile Niño 4, way out west, sits at +1.4 °C absolute and +0.2 °C relative, barely above the tropics-wide average even as the eastern boxes surge. This is an east-Pacific-flavored event, and that shape matters for where the atmosphere responds.

Enlarge

For reference, the strongest events on record peaked at RONI +2.40 (1982–83), +2.28 (1997–98) and +2.25 (2015–16). The "very strong" threshold NOAA's probabilities are built around is +2.0. Today's value is +1.7. CPC has drawn a second line above all three of those historical peaks, at +2.5, and as of 8 October 2026 gives October–December an 83% chance of clearing it, up from 75% a month earlier and 69% in August.

What Forecasters Actually Expect

NOAA publishes two probability tables. The first is about whether El Niño exists at all, and it's about as emphatic as seasonal forecasting gets: 100% for every three-month season from September 2026 through March 2027, easing to 98% by February–April, 85% by March–May and 45% by April–June 2027. La Niña sits at 0% in every row through March–May, then registers 1% for April–June and 13% for May–July 2027, the last row, where neutral (70%) has overtaken El Niño (17%).

The second table is about how strong, and it's the more interesting one:

Season	Strong	Very strong	Was, 10 Sep
Sep–Nov	0%	100%	97%
Oct–Dec	0%	100%	98%
Nov–Jan	4%	96%	93%
Dec–Feb	16%	81%	75%
Jan–Mar	37%	46%	39%
Feb–Apr	32%	6%	5%
Mar–May	5%	0%	0%
Apr–Jun	0%	0%	0%
May–Jul	0%	0%	not issued

(Issued 8 October 2026, verified against RONI, with the 10 September issuance in the last column for comparison. Seasons run from September 2026 through July 2027. "Very strong" means at least +2.0 °C, and that is the highest published category: CPC's +2.5 °C record threshold sits inside it rather than beside it. Rows don't total 100%; the remainder sits in the moderate, weak and neutral categories.)

Read the shape, not just the peak. It holds at 96% or higher for three straight seasons, September through January, then decays through late winter: 81% for December–February, 46% for January–March, 6% for February–April. Every official product stops in mid-2027, so anything you read about La Niña in late 2027 is somebody extrapolating.

Note what moved. Every season from September–November through February–April ticked up, and the biggest moves came in late winter (December–February from 75% to 81%, January–March from 39% to 46%). That is the change behind CPC's new synopsis: with autumn already at 100%, the open question has shifted to how long the event stays strong. The shape is still the one August drew, and this is still a forecast being confirmed rather than revised.

Independent guidance points the same way on timing. The international model plume run by Columbia's IRI has 23 of 26 models clearing +2.0 °C in October–December. The World Meteorological Organization's multi-model ensemble has it "peaking in November." ECMWF's seasonal model peaks it in December. Those aren't three disagreements. They're three ways of describing the same late-autumn peak.

One caveat on that IRI figure: it's measured on the traditional Niño 3.4 index, not RONI, so its +2.0 °C line is not the same line NOAA's 100% call is drawn at. Treat it as corroborating the strengthening and the peak timing, not as a second vote on the "very strong" call.

CPC's own consolidated outlook now puts the median October–December relative Niño-3.4 value at +2.74 °C, inside a 5-to-95% range of +2.25 to +3.23 °C. In September the median was +2.67 °C with a range of +2.12 to +3.23 °C, and in August +2.66 °C with +1.95 to +3.37 °C. So the median rose this month, after barely moving in September, and the low end rose again while the top held at +3.23 °C. Two months ago the bottom of that 90% span sat just below the +2.0 °C "very strong" line; now it sits a quarter of a degree above it. That is what a 100% call looks like from the distribution side, and the distribution has been squeezing from below rather than stretching upward. That distribution, not the binned strength table, is where the 83% record figure comes from: the published strength categories stop at +2.0 °C and have no +2.5 bin at all.

One caution about model output: some individual model members are showing peaks near +4 °C. Do not take those at face value. ECMWF says plainly that a single model "is typically overconfident, and the eventual outcome can sometimes lie outside the predicted range." The credible signal is the cluster, not the tail.

Hurricane Season: the First Real Consequence

This is where El Niño is already showing up in daily weather, and NOAA updated its outlook on 6 August 2026. That remains the current outlook; NOAA issues an initial Atlantic outlook in May and a single update in early August, so nothing has superseded it. One caveat on timing: it predates the ENSO escalation by a week and states its own input plainly, a 48% chance of a very strong El Niño during August–October. CPC has since raised that season twice, to 75% on 13 August and 77% on 10 September, and its 8 October table, which no longer includes August–October, has September–November at 100%. If anything the outlook's suppression assumption has firmed up rather than weakened.

El Niño suppresses Atlantic hurricanes through a straightforward mechanism: all those relocated Pacific thunderstorms strengthen upper-level winds over the Atlantic, which increases vertical wind shear, the change in wind with height. Shear tears developing storms apart before they can organize. NOAA's phrasing: "El Niño increases wind shear, which tends to prevent tropical storms from forming or from getting stronger once formed."

The statistical record behind that is unusually clean:

"All hurricane seasons coincident with strong (greater than or equal to 1.5 °C Niño 3.4) El Niño events since 1950 have featured below-normal seasonal activity."

NOAA's updated 2026 Atlantic outlook:

Measure	2026 outlook	1991–2020 average
Chance of a below-normal season	75%	—
Named storms	7–13	14
Hurricanes	2–6	7
Major hurricanes	0–2	3

As of NHC's advisories on the evening of 8 October the Atlantic has produced nine named storms and one hurricane. Arthur in mid-June and Bertha in late July, both in the Gulf, then Cristobal near the Azores on 12 August, Dolly in the deep tropics on 27 August and Edouard in the Gulf on 31 August. Dolly lasted about a day and a half before falling apart into an open tropical wave. September added Fay, a depression southwest of the Azores on 19 September that was post-tropical by the 29th, Gonzalo south-southeast of the Cabo Verde Islands on 25 September and Hanna in the central subtropical Atlantic on 28 September. Then came Isaias. It formed as a depression over the Gulf on 6 October, was named on 7 October and became the season's first hurricane on 8 October. At NHC's 7 PM CDT advisory that day it had 100 mph winds, with a hurricane warning in effect for the northern Gulf Coast from Ocean Springs, Mississippi to the Bay/Gulf County line in Florida. If you are anywhere near it, NHC's advisories, not this page, are what to follow. That is textbook for an El Niño year in one respect: much of the activity that does happen is displaced toward the Gulf and subtropics, closer to home, and four of the nine storms have formed in the Gulf. Put the count against NHC's own climatology and the gap is still stark. In an average season the first hurricane forms by 11 August, the second by 26 August, the third by 7 September, the fourth by 16 September and the fifth by 28 September. This season went past all five benchmark dates with none, and its first arrived on 8 October, nearly two months after the average first date, in the stretch that is normally the busiest of the year. That is the suppression showing up in the observed record rather than just the forecast. One hurricane does not overturn it: nine named storms already sits inside the outlook's 7–13 range for the whole season, but one hurricane is still below the floor of its 2–6 range.

The Pacific gets the opposite treatment. NOAA's East Pacific outlook calls for 15–22 named storms, above average, and NHC's archive already lists nineteen in that basin: the eighteen names from Amanda through Simon, plus Nolo, a depression numbered in the East Pacific that was named by the Central Pacific Hurricane Center. Nine of them reached hurricane strength (Fausto, Genevieve, Karina, Lowell, Marie, Nolo, Odalys, Polo and Rachel), against a 1991–2020 average of 15 named storms and 8 hurricanes for the entire season. Lowell rapidly intensified into a Category 5 on 2 September, and Nolo reached Category 4 with 145 mph winds before crossing the International Date Line. The two basins are running the same El Niño in opposite directions.

A suppressed season is not a safe season. NOAA repeats this every year and it's worth repeating here: this is not a landfall forecast, and it only takes one storm. NWS Director Ken Graham made the point about Arthur, which never became a hurricane and still delivered record rainfall to parts of Louisiana. Edouard made it again in September: also never a hurricane, and its remnants were still dropping flooding rain west of Tyler, Texas three days after it formed. Isaias is making it a third time, and this one is a hurricane. If you want the details on how the forecast cone works and what it doesn't show you, we've covered the cone of uncertainty and why storms intensify quickly separately.

What It Means for North American Winter

CPC's October winter outlook is now out. The seasonal set issued 15 October 2026 replaces September's, and the next update is due 19 November 2026. It is separate from the 8 October ENSO discussion above: that one forecasts the Pacific event itself; this one translates the climate signals into U.S. temperature and precipitation odds. The winter picture is revised monthly, so treat it as the current read rather than a settled forecast.

The mechanism is well established. During El Niño the Pacific jet stream strengthens and extends across the southern United States, while the polar jet retreats north into Canada. That favors a wetter southern tier and a milder northern tier.

Here is CPC's actual December–February 2026–27 outlook, from the set issued 15 October 2026, rather than a generic composite. The temperature map and precipitation map are saved copies of that issuance, with the valid season printed on each. CPC's discussion opens with November–January; the table below is for December–February.

Region	Temp	Precip
Pacific Northwest coast	Above normal favored	Above normal favored on the Oregon coast; equal chances farther north
Interior Northwest & N. Rockies	Above normal favored	Below normal favored in the northern Rockies; equal chances farther west
Upper Midwest & Great Lakes	Above normal favored	Below normal favored in parts of the Great Lakes; equal chances elsewhere
Northeast	Above normal favored	Above normal favored toward the coast; equal chances inland
Ohio & Tennessee Valleys	Above normal favored	Mostly equal chances
Southeast & Florida	Near normal favored toward the Gulf, southern Atlantic coast and Florida; above normal favored farther north	Above normal favored, strongest near the coast
Gulf Coast	Near normal favored east of Texas; equal chances in Texas	Above normal favored, except parts of far southern Texas
California & Southwest	Above normal favored along much of California's coast; equal chances across much of the Southwest	Above normal favored across much of the region
Central & Southern Plains	Above normal favored north and east; equal chances farther south and west	Above normal favored across much of the region
Alaska	Above normal favored east, strongest in the southeast; equal chances west	Below normal favored across much of the Mainland; equal chances in the southeast

These are broad regions, and the boundaries cut across them. Cells name the favored category, not a confidence band or a guaranteed outcome. Probability strength varies within each region; consult the dated maps for the bands. Equal chances means no category is favored, not a near-normal forecast. The maps give category probabilities, not the number of degrees or inches to expect.

The broad pattern is familiar, but October's reasoning matters:

• The negative PDO is still there, but CPC now gives it less weight. September's discussion raised the possibility that it would dampen the usual El Niño impacts. October's says its winter influence may be minimal given the very strong El Niño. That is a change in how CPC weighs the competing signals, not evidence that the PDO has turned positive.
• The West Coast wet signal reaches Oregon. CPC favors above-normal precipitation across much of the West Coast and Southwest, with the wet area extending as far north as the Oregon coast. It puts less confidence in this than in the Southeast because the forecast tools show a wider range of outcomes. September's discussion left open an increase in California's odds of above-normal precipitation, pending reassessment of the model forecasts. In October's December–February outlook, the highest displayed band is still 50–60% near the coast. For how individual storms deliver that rain, see atmospheric rivers and what El Niño does to them.
• The Southeast remains the strongest wet signal. CPC gives parts of the extreme Southeast and northern Florida more than 70% odds of above-normal precipitation through January–March. The wet tilt also runs up the Eastern Seaboard. That is a seasonal precipitation call, not a snowfall forecast.
• The dry areas remain in the northern Rockies and parts of the Great Lakes. CPC expects them to shift gradually north through winter and spring. They do not cover the whole northern tier.
• Alaska has two different stories. Eastern Alaska is favored warm, with the strongest odds in southeastern Mainland and Southeast Alaska. Much of the Mainland is favored dry as the storm track shifts south; Southeast Alaska is equal chances for DJF precipitation.

El Niño is the main driver, but CPC also uses model ensembles, statistical guidance, long-term temperature trends and other climate signals. It considers the winter's expected westerly Quasi-Biennial Oscillation (QBO) supportive of a steadier stratospheric polar vortex and warmer northern and eastern U.S. conditions. That shifts the odds; it does not rule out a polar vortex disruption or a cold spell.

Two things that surprise people:

The classic "cool southern winter" may not show up. CPC favors near-normal temperatures in parts of the Southeast and Florida, while much of the Southwest and parts of the southern Plains have equal chances. Those are different calls. Some models favor a colder Southeast later in winter, especially January–March and February–April, but CPC judges that they may overstate rain and cloud cover's cooling effect. Statistical guidance, which includes long-term warming and warmer nights, pulls the other way. The later below-normal temperature tilt CPC describes for parts of the Southwest is for February–April and March–May, not December–February.

These maps do not forecast your snow total. They cover seasonal average temperature and total precipitation. Snow depends on storm tracks and temperatures at the scale of individual systems, which is why even three-day snow forecasts vary so much. A wetter winter does not automatically mean a snowier one.

Related reading: for what the seven strong El Niño winters since 1950 actually did in each U.S. region, see What an El Niño Winter Usually Brings, Region by Region.

Around the World

The rainfall risk is spread across both hemispheres. WMO's seasonal update leans dry for southern and eastern Australia, the Indian subcontinent, Central America and the Caribbean, and northwestern South America, and wet for the central and eastern equatorial Pacific, the Greater Horn of Africa, southeastern South America, and western North America. Every one of those is a shift in the odds, not a forecast.

The global temperature story lands in 2027, not 2026. El Niño's peak effect on global average temperature arrives roughly three months after the Pacific peaks, and it has to work its way through the atmosphere after that. With this event peaking around November–December 2026, the imprint mostly falls in the following calendar year. WMO puts it plainly: El Niño events "exert their strongest influence on global temperatures during the year following their onset." Carbon Brief's July 2026 analysis gives 2026 about a 35% chance of being the warmest year on record, and 2027 about a 92% chance.

Coral reefs are the quiet casualty. The fourth global bleaching event began in 2023 and, in NOAA's assessment, likely ended in 2025. It hit roughly 84% of the world's reef area, the largest on record, well past the 68% of the 2014–17 event. The first, third and fourth global bleaching events all coincided with strong El Niño periods. A very strong event peaking in November–December 2026 places the southern-hemisphere bleaching season of early 2027 squarely in the risk window.

Enlarge

What This Forecast Does Not Tell You

Five things worth being clear about, because the gap between "strong El Niño" and "I know what my winter looks like" is wider than most coverage admits.

1. "Very strong" and "record" are two different lines, and CPC prices both. The ≥ +2.0 °C category is open-ended, so the 100% (rounded) attached to October–December is the probability of clearing that line, not of setting a record. The record question has its own number: as of 8 October, an 83% chance that the October–December RONI comes in at +2.5 °C or more, which would clear the +2.40 peak of 1982–83, the highest value in the RONI record. Three things to hold onto. It is a probability, not a prediction, and the rest of CPC's distribution still includes outcomes where the 1982–83 mark survives. It applies to a three-month season, not to the event as a whole: CPC now prices it for three overlapping seasons (54% for September–November, 83% for October–December, 70% for November–January), and the same forecast gives December–February an 81% chance of merely staying very strong. And it is drawn on RONI, the relative index. On the legacy ONI, +2.5 °C is not a record at all, because 2015–16 peaked at +2.59.

2. "Super El Niño" isn't a thing. WMO says so directly: "The term 'super El Niño' is not part of WMO's operational classification system and is therefore not used in official WMO products." There are four official tiers: weak, moderate, strong, very strong. CPC itself now reaches past the top tier with the word "historic," but it does so with a number attached: +2.5 °C on a three-month RONI. Extraordinary language with a threshold behind it is forecasting. Extraordinary language without one still tells you something about the page.

3. 100% doesn't mean certain. It means the probability rounds to 100. CPC notes that a 0% chance "is approximate and not exact."

4. A strong signal in the Pacific is not a strong impact where you live. CPC, 8 October 2026: "With an event of this magnitude, the chances of experiencing impacts consistent with El Niño are larger, though not guaranteed." Australia's Bureau of Meteorology says the same thing more bluntly: "A strong El Niño signal in the Niño 3.4 region does not necessarily mean strong impacts."

5. A seasonal outlook is a different object than a weather forecast. A composite "typical El Niño winter" map is an average over a small handful of past events, each with its own quirks. It shifts your odds over three months. It cannot tell you whether it will snow on a given Tuesday. That's what forecast models are for, and they only see about a week ahead.

One more honest note: because this event is now more likely than not to end up larger than anything in the modern record, there's genuine uncertainty about whether it behaves like the composite at all. CPC flagged this themselves in its August hurricane outlook: a record-scale event "could result in the circulation having impacts we have never observed."

Frequently Asked Questions

Is the 2026 El Niño the strongest ever?

Not yet, and the final rank won't be settled until winter, but as of 8 October 2026 a record is the single most likely outcome. On the relative index NOAA classifies events with, it stands at +1.7 °C for July–September 2026, against record peaks of +2.40 °C in 1982–83 and +2.28 °C in 1997–98. NOAA now gives October–December a 100% chance (rounded) of reaching "very strong" (at least +2.0 °C) and an 83% chance of reaching +2.5 °C or more, which it describes as "a historic event that would exceed the strength of previous El Niño events dating back to 1950." CPC's consolidated outlook for that season has a median of +2.74 °C. Read the 83% for what it is: a probability covering a three-month season on one index, not a prediction that the 1982–83 record will fall.

When will El Niño peak?

Late autumn through early winter 2026. NOAA's strength probabilities peak in September through December, the WMO multi-model ensemble points to November, and ECMWF's seasonal model peaks it in December. Decay follows through late winter and early spring 2027, with NOAA giving a 98% chance El Niño conditions persist into February–April 2027 and 85% into March–May.

Why do different sources give different El Niño temperatures?

Because they use different time windows and different baselines. For the week centered 28 October 2026, the Niño 3.4 region read +3.4 °C as an absolute weekly departure and +2.5 °C as a relative one. Averaged over July–September 2026 instead, the same region read +2.2 °C on the traditional ONI and +1.7 °C on RONI. NOAA classifies El Niño on RONI, a three-month average that removes tropics-wide background warming and then rescales the result to the familiar ONI range. NOAA made it the official index on 1 February 2026. Weekly values are noisy; the three-month indices are what define an event.

Does El Niño mean fewer hurricanes?

In the Atlantic, usually yes. El Niño increases vertical wind shear over the Atlantic, which disrupts developing storms. Every hurricane season coinciding with a strong El Niño since 1950 has been below normal, and NOAA's 6 August 2026 outlook gives this season a 75% chance of below-normal activity. The Pacific runs the opposite way, with an above-average East Pacific season expected. Even so, the Atlantic's first hurricane of 2026, Isaias, formed in the Gulf on 8 October. A below-normal season is not a safe season; landfall risk is not predicted by seasonal outlooks.

Will El Niño give me a warm winter?

It shifts the odds toward a milder winter across the northern United States and a wetter one across much of the southern tier, but it is not a guarantee for any location. CPC's 15 October 2026 outlook for December–February 2026–27 favors warmth across the northern tier and much of the East, near-normal temperatures in parts of the Southeast and Florida, and equal chances across much of the Southwest and parts of the southern Plains. Equal chances does not mean near normal. CPC balances El Niño with long-term warming and other climate signals; individual cold spells can still interrupt a mild season. The next seasonal update is due 19 November.

Does El Niño cause record global temperatures?

It contributes to them, with a lag. El Niño releases stored ocean heat to the atmosphere, and the peak effect on global average temperature arrives in the calendar year after the event develops. The 2023–24 El Niño contributed to 2024 becoming the warmest year on record. For the current event, that means 2027 is the year to watch. Carbon Brief's analysis gives 2027 about a 92% chance of setting a record, versus about 35% for 2026.

Track the Pacific With WeatherAI

El Niño works on the scale of seasons. Your actual weather still arrives one day at a time, and that's where WeatherAI lives, on your iPhone:

• Ask in plain English. "Is this storm track normal for El Niño?" gets you a real answer for your actual location, not a composite map.
• Hurricane tracking with live positions and official NHC forecast cones for Atlantic and Pacific storms, useful in a season where the risk shifts toward the Gulf.
• Professional-grade radar for when the seasonal outlook stops being useful and the storm is actually overhead.
• Custom alerts so you hear about the rainfall event that matters to you, not the one making national news.

Seasonal outlooks tell you how to bet. They don't tell you what to do on Thursday. Get both.

Download WeatherAI for iPhone →

---

Sources. Every figure above traces to one of these:

• NOAA CPC ENSO Diagnostic Discussion, 8 October 2026: advisory status, the strong-to-very-strong synopsis, the 54%, 83% and 70% record calls and the +2.5 °C RONI threshold, and the September index values. Dated archive: enso_disc_oct2026 (the live page is overwritten each month)
• CPC ENSO Diagnostic Discussions of 10 September, 13 August, 9 July and 11 June 2026: the 98%, 75%, 95%, 69%, 81% and 63% figures, the earlier synopsis sentence and the July wording quoted above
• CPC ENSO probabilities, strength probabilities and RONI outlook, all issued October 2026: the three tables, the strength-category thresholds and the October–December distribution. The September comparison column comes from CPC's archived strength chart for that issuance
• CPC RONI and ONI series, read 8 October 2026: the July–September values, the event comparison table, all historical peaks, and the 2023–24 comparison. July–September is now the last complete season; August–October posts in early November
• CPC weekly absolute and relative SST indices, read 3 November 2026: the +3.4 / +2.5 weekly pair, the regions diagram and the Niño 1+2 ranking. The latest week in both files was the one centered 28 October. CPC revises these files retroactively, so weekly values here are as published on that date
• NWS Public Information Statement 26-05, 13 January 2026: the ONI-to-RONI switch and RONI's definition, including the variance rescaling
• CPC Long-Lead Seasonal Outlook discussion, 15 October 2026, read 21 October: the winter interpretation and regional table. The discussion opens with November–January and separately describes the evolution through winter and spring; the table here is specifically December–February 2026–27. Dated copies of the official DJF temperature and DJF precipitation maps preserve that issuance. Retrieved from CPC's lead02 temperature and lead02 precipitation URLs, with both printed headers checked: issued 15 October 2026, valid Dec–Jan–Feb 2026–27. Lead03 now shows Jan–Feb–Mar 2027. The live discussion and lead URLs roll forward; this seasonal set is superseded on 19 November 2026
• NOAA 2026 Atlantic Hurricane Season Outlook update, 6 August 2026: the seasonal outlook ranges and the 75% below-normal figure
• NHC 2026 tropical cyclone archive and seasonal climatology, read 8 October 2026: the storm counts, formation dates and intensities in both basins, and the 11 August, 26 August, 7 September, 16 September and 28 September benchmark dates for the first five Atlantic hurricanes of an average season, plus the 15 named storms and 8 hurricanes of an average East Pacific season
• NOAA PMEL Warm Water Volume, read 8 October 2026: ocean heat content, the record-high August 2026 month, July 2026 in second place and the 1997 and 2015 comparisons. The series runs from January 1980 and stops at August 2026 as of this reading
• CPC monthly index files for the atmospheric numbers: SOI, outgoing longwave radiation, and 850 hPa trade winds for the west and central Pacific
• Carbon Brief State of the Climate, 24 July 2026: the 35% and 92% record-year projections, which are Carbon Brief's own estimates rather than an agency forecast
• IRI/Columbia ENSO forecast, July 2026 · WMO Global Seasonal Climate Update, prepared 20 July 2026 · Australian BOM · ECMWF on relative Niño indices · NOAA Coral Reef Watch

Keep reading: what the hurricane cone actually shows, why the polar vortex matters in winter, and how weather models differ.