The 2026–27 “Super El Niño” Forecast and What It Actually Measures

Picture the Pacific Ocean as a giant thermostat. During an El Niño, that thermostat flips: the central and eastern equatorial Pacific warms more than usual, and the atmosphere above it starts behaving differently, which can ripple into weather patterns around the globe.

In mid-July 2026, forecasters looked at the newest seasonal model runs and found something that feels rare even to seasoned climate watchers: a very high chance that the 2026–27 event will be among the strongest in the observational record, with forecast peaks around the 3.6°C level in the Niño 3.4 region. (carbonbrief.org)

The scary part isn’t the name (“El Niño”). The scary part is what the forecast implies when you ask: what does “strongest ever” mean, mathematically, and why does a changing climate complicate that answer?

The basic ingredient: Niño 3.4 sea-surface-temperature anomalies

El Niño is an ocean–atmosphere pattern (scientists often bundle it with La Niña under ENSO, the El Niño–Southern Oscillation). The most common “strength” metric is based on sea surface temperature (SST): the temperature of the ocean’s upper surface.

To decide whether SST is “unusually warm,” scientists use an anomaly, meaning “departure from an average.” More specifically, NOAA’s widely used Oceanic Niño Index (ONI) monitors SST anomalies in a specific patch called the Niño 3.4 region (5°N–5°S, 120°W–170°W). (climate.gov)

NOAA considers El Niño conditions present when the ONI (a 3-month running mean of SST anomalies) reaches at least +0.5°C. (oceanexplorer.noaa.gov)

So far, so straightforward. The trouble shows up when you say “strongest ever.” Strongest relative to what baseline?

“Record-setting” depends on the baseline you subtract

If the planet were static, “strongest El Niño ever” would be easy: take SST anomalies in Niño 3.4, find the largest peak over time, and compare.

But real life isn’t static. Greenhouse-gas warming changes the overall warmth level of the oceans too. That means a fixed baseline can make the event look stronger than it would have in a cooler world.

This is why the 2026–27 discussion often includes two related measures:

• ONI / raw Niño 3.4 anomaly: “Niño 3.4 warming relative to a climatological average.”
• RONI (Relative Oceanic Niño Index): a “relative” version meant to isolate the ENSO signal from broader tropical background warming.

Carbon Brief’s analysis of the mid-July modeling picture reported the headline numbers many people have already heard: across 667 model runs drawn from 14 seasonal forecasting systems, about 91% of runs project a peak that exceeds the prior modern-record El Niño. The forecast median peak for the raw Niño 3.4 anomaly lands near 3.59–3.6°C, versus ~2.75°C for the 2015–16 record event. ()

That already sounds extreme. But “extreme” is where the baseline question bites.

How ensemble forecasts make “probability” feel real

Seasonal forecasts don’t produce one answer. They produce an ensemble: a collection of many model simulations (“ensemble members”), each starting from slightly different initial ocean/atmosphere states.

A dynamical model is a physics-based computer simulation that solves equations describing fluid motion in the atmosphere and ocean. The atmosphere is chaotic enough that tiny differences at the start can later create different outcomes.

So instead of saying “the peak will be X,” forecasters ask: how often do simulations produce peaks above some threshold? That’s how you get statements like “about 91% of runs…” ()

This is also why the phrase “strongest El Niño ever” is better read as a statistical claim about many plausible futures, not a single deterministic prophecy.

ONI can inflate strength in a warmed ocean—RONI tries to correct that

The Relative Oceanic Niño Index (RONI) is designed specifically to prevent background warming from masquerading as ENSO strength.

NOAA’s CPC (Climate Prediction Center) transitioned to operationally using RONI on February 1, 2026. (prod-01-alb-www-noaa.woc.noaa.gov)

Conceptually, RONI starts from the Niño 3.4 SST anomaly and subtracts the tropical-mean SST departures over a broader latitude band (20°S–20°N). Then it applies a scaling factor so the relative index has an amplitude comparable to the traditional index. (weather.gov)

If that sounds abstract, here’s a plain-language version:

1. Compute how warm Niño 3.4 is compared to its baseline.
2. Compute how warm the broader tropical ocean is compared to its baseline.
3. Subtract step (2) from step (1). The result focuses on extra warmth in Niño 3.4 relative to the tropics as a whole.

Then the scaling factor helps keep the numbers comparable across methods.

To see what this changes, Carbon Brief reported a revised (relative) picture: in RONI terms, the median forecast peak in the latter half of 2026 is around 3.1°C, compared with a prior record of 2.69°C for 1982–83. Even after this correction, 77% of runs still project a new record. ()

So whether you look at the raw Niño 3.4 anomaly or RONI, the models still cluster toward “record-class.” But RONI makes the claim harder to dismiss as a baseline artifact.

Why velocity (trajectory) matters as much as the peak

Another detail that tends to get lost when headlines only mention the maximum: forecasts also care about the trajectory—how fast the event ramps up.

A rapidly developing event can produce earlier and stronger coupling between ocean heat and the atmosphere’s circulation. Carbon Brief noted that the Niño 3.4 index crossed key thresholds quickly after La Niña years, and that model projections indicate the strengthening would continue toward late 2026. ()

This matters because seasonal weather impacts are not driven by a single month’s anomaly. They’re driven by evolving ocean conditions that steer winds and rainfall patterns over multiple overlapping seasons.

A mini “from anomalies to indices” recipe (with dummy numbers)

Writers sometimes summarize indices with a sentence. It’s worth seeing the math structure once.

Let:
- NINO34 = Niño 3.4 SST anomaly (°C)
- TROP = tropical mean SST anomaly across 20°S–20°N (°C)
- s = scaling factor (dimensionless)

A simplified “relative Niño” concept looks like:

RONI_like = (NINO34 - TROP) * s

ECMWF describes the same ingredients—subtracting the tropical-mean anomaly and then scaling by a factor chosen so the relative index matches the traditional index’s amplitude behavior. (charts.ecmwf.int)

The exact operational details can include reference-period choices and how variance is handled, but the backbone idea is subtraction of tropical background warmth.

That subtraction is the heart of why RONI is meant to be a “cleaner” ENSO strength measure in a changing climate. (prod-01-alb-www-noaa.woc.noaa.gov)

So is this “the strongest El Niño ever” or “an artifact of warming”?

The honest answer is: the baseline question doesn’t remove the signal—it changes how strongly we should interpret it.

By mid-July 2026, multiple forecasting perspectives lined up:
- Raw Niño 3.4 anomaly projections cluster near ~3.6°C peak values, with a large fraction of ensemble runs exceeding the 2015–16 modern record. (carbonbrief.org)
- Relative measures like RONI reduce the peak magnitude by subtracting tropical background warming, but still project record-class strength compared with the earlier 1982–83 event. ()
- Forecast discussions emphasize the coupling between atmosphere and ocean evolving under an already-warmed climate baseline. (nature.com)

In other words, the models aren’t just “naming” a warm ocean. They’re producing consistent, ensemble-based evidence that the Niño 3.4 region’s extra warmth is both large and fast-moving compared to historical behavior.

Closing thought: climate data gets better when we question the subtraction

A forecast like this forces beginners and intermediates alike to learn something important: indices are not passive labels. They are decisions about what to subtract and how to scale.

What feels like a single number—“strongest ever”—is actually the outcome of:
- where we measure (Niño 3.4),
- what we call an anomaly,
- how we treat background warming,
- and how we summarize uncertain futures with ensembles.

When those moving parts agree across raw and relative indices, the claim becomes harder to dismiss. And in July 2026, the agreement looked unusually strong. ()