Three weeks in April that closed the question.
On April 9, 2026, NOAA's Climate Prediction Center issued a formal El Niño Watch. Two weeks later, on April 24, the World Meteorological Organization elevated the framing — high confidence in the onset of El Niño, followed by further intensification. A question lives underneath both announcements. The public still tends to ask it wrongly.
The conventional reading of ENSO — the El Niño / La Niña oscillation in the Pacific — runs like this: warm years spike upward, cool years pull the global mean back to a long-run average, and the system oscillates around a slowly rising baseline. The data since 2015 reads differently. Each La Niña year now lands above the previous El Niño year from a decade before. The system is not oscillating around a baseline. It is climbing a staircase.
The mechanism that closes the trap is Earth's energy imbalance. More than 91% of the excess heat from greenhouse gas accumulation is absorbed by the ocean — a number the WMO published, for the first time, as a key climate indicator in March 2026. Each El Niño transfers some of that ocean-stored heat to the atmosphere and lifts surface temperatures. The subsequent La Niña suppresses Pacific sea-surface temperatures again, but cannot return the global mean to its previous floor — because the reservoir is now permanently warmer. The 2026 El Niño the WMO just confirmed will set a new peak. The La Niña that follows will not drain it.
The timeline that built the floor. The numbers that confirm it. The forecast that is locking in the next step.
NOAA and the WMO both confirmed a new El Niño in April 2026. The big picture: the system is no longer oscillating. Each La Niña cool phase now lands warmer than the El Niño hot years of a decade before. The ocean is absorbing over 91% of excess heat — and that reservoir only goes in one direction.
Eleven years. Two peaks. No reset.
The decade 2015–2025 contains the eleven hottest years in 176 years of instrumental records, in order. That is not noise around a trend. Each ENSO cycle ratchets the floor higher than the one before it.
Five years of floor.
Global mean temperature anomaly above the 1850–1900 pre-industrial baseline. Five consecutive years. Two ENSO phases. The cool years are now warmer than the hot years of a decade earlier.
Half a century of no return.
Annual global mean temperature anomaly above the 1850–1900 pre-industrial baseline. 1970 to 2026. The staircase is not a metaphor. It is the shape of the record.
What ENSO moves. What it can't.
Earth has an energy imbalance — more heat coming in than going out. In March 2026, the WMO put numbers on where that excess heat actually goes, publishing the breakdown for the first time as a key climate indicator. The numbers explain why La Niña no longer resets the floor.
~1% in the atmosphere.
Roughly 1% of Earth's excess heat sits in the atmosphere — the surface temperature reading the public watches. An additional 5% is in the land, 3% in ice. ENSO shifts heat between Pacific surface waters and the lower atmosphere, which is why El Niño years feel hotter and La Niña years feel cooler in the global mean. This is the heat the system can move on a one-to-three-year timescale.
91% in the ocean.
More than 91% of the excess heat from greenhouse gas accumulation is absorbed by the ocean — a reservoir that, on human timescales, never releases its heat back to space. Ocean heat content set a new record in 2025. The annual increase was equivalent to roughly 39 times total human energy production. Ocean warming from 2005–2025 has more than doubled the rate of 1960–2005. ENSO can scoop a thin layer off the top during El Niño. La Niña cannot push it back.
What the next eighteen months hold.
Models from 13 modeling groups and 637 individual runs converge on a clear picture. 2026 builds the El Niño. 2027 cashes the lag.
High confidence.
On April 24, 2026, the WMO's Chief of Climate Prediction issued the agency's strongest formulation yet on the 2026 El Niño. In February, the probability had stood at 30–40%. By late April: high confidence in onset.
“Climate models are now strongly aligned, and there is high confidence in the onset of El Niño, followed by further intensification.”
— Wilfran Moufouma Okia · Chief of Climate Prediction, WMO · 24 April 2026
When the WMO released its annual climate report in March 2026, Secretary-General Celeste Saulo placed the structural framing on record: "Human activities are increasingly disrupting the natural equilibrium and we live with consequences for hundreds and thousands of years." The energy imbalance, she added, has reached a new high. The next ENSO cycle will play out on top of it.
The lag cashes in 2027.
Pacific sea-surface temperatures peak first. The global mean temperature peak follows roughly three months later. The 1998, 2016, and 2024 record years all sat after their El Niño's Pacific peak.
If the 2026 El Niño peaks in late 2026 — which the model consensus now suggests — the temperature record it sets will most likely show up in 2027, not 2026 itself. Carbon Brief's central projection for 2026 is 1.47°C above pre-industrial, with a 19% probability of breaking the all-time record this year. The probability of a new record in 2027 runs higher still. Models project 2027 as a likely candidate for the all-time warmest year, once the ENSO lag cashes in.
What the staircase predicts next is the part that doesn't make headlines. After the 2026–27 El Niño peak, ENSO will swing back toward neutral or La Niña. That cooling — like the 2022 La Niña before it, like the 2025 La Niña before that — will fail to return the global mean to its pre-El-Niño floor. The ocean reservoir will be warmer still. The next floor sits above 1.44°C; the one after that, higher again.
The framing the WMO chose for its 2025 report is the framing that has to travel: not 'global temperatures', but Earth's energy imbalance. Surface temperature is the symptom — the tenths of a degree that ENSO toggles up and down. The reservoir is the disease. It has only one direction.
El Niño peaks in the Pacific first. The global temperature record it breaks shows up about three months later. If the 2026 El Niño peaks late in the year, the all-time temperature record will likely land in 2027. The La Niña after that will cool the Pacific surface but won't lower the ocean reservoir — the floor will be higher than before.