The 2026 El Nino Could Be the Strongest in 75 Years as Models Show 100% Probability Through February

The Strongest El Niño on Record Is Forming, and Tech Infrastructure Is in Its Path

El Niño conditions in the tropical Pacific are intensifying faster than any event in the modern observational record, with forecast models from 14 of the world's leading climate centers converging on a peak that would exceed all previous records. The median of 667 model runs analyzed by Carbon Brief puts sea surface temperatures in the Niño 3.4 region at 3.59 degrees Celsius above normal at peak intensity, surpassing the previous record set during 2015-16 by roughly 0.8 degrees Celsius. 1 The gap between the strongest and fifth-strongest El Niño over the past 150 years is only about 0.5 degrees, which puts the projected margin in perspective. 2

The event crossed the 0.5-degree El Niño threshold in April, reached 1.6 degrees by June, and shot above 2 degrees in the first weeks of July. 1 That is one of the fastest onsets in the observational record. It is intensifying more rapidly than the benchmark 1997-98 super El Niño, despite 2026 beginning in La Niña, the cool phase of the Pacific cycle that suppresses ocean temperatures. 2

Of the 667 model runs Carbon Brief examined across 14 modeling groups, more than 91% predict the strongest El Niño since reliable records began. 1

The convergence is unprecedented

Every one of the 14 modeling groups, using systems including CFS, NMME, C3S, CanSIPS, and SINTEX-F, produces a median peak exceeding the 2-degree threshold for a "very strong" El Niño, with most models peaking in November or December. 1 Models are fed historical data, and an event of this magnitude has no modern analogue. Zeke Hausfather, a climate scientist at Berkeley Earth, acknowledges the uncertainty but points to the model convergence itself as grounds for confidence. 2

Even using NOAA's Relative Oceanic Niño Index, a more conservative measure that adjusts for background tropical ocean warming to isolate the El Niño signal, roughly 77% of model runs still predict a new record. Under this metric the prior record shifts from 2015-16 to 1982-83, but the 2026 forecast remains dominant. 3

The downstream shock

El Niño does not simply raise temperatures. It rearranges global atmospheric circulation, shifting rainfall eastward across the Pacific and drying out Indonesia, parts of South America, and parts of Africa. 3 Fisheries are already feeling the effects: the Peruvian government banned anchovy catches this spring as warm Pacific waters reduced catches. 2

For the technology economy, these disruptions compound the strain on power grids already absorbing the AI data center buildout. Extreme heat events from a record-strength El Niño will drive up cooling demand and peak electricity loads while utilities are simultaneously managing surging compute power consumption. For agricultural technology and biotech companies that depend on stable commodity supply chains, El Niño-driven droughts in key growing regions threaten inputs that are already under pressure.

The economic stakes are large enough to model now. A study by Justin Mankin, a climate scientist at Dartmouth College, found that the 1997-98 El Niño reduced global economic growth by approximately $5.7 trillion in the years that followed. If current projections hold, the 2026 event could produce cumulative economic losses of around $10 trillion by 2032. 2

November's peak, 2027's record

The El Niño peak arrives in November or December. 1 The odds of 2026 becoming the hottest year on record have nearly doubled since April, rising from 19% to 35%. 1 Carbon Brief projects 2027 at approximately 1.7 degrees Celsius above preindustrial levels, which would set a new annual temperature record. 1

The economic damage from the 1997-98 event compounded for five years after it peaked. 2 This event is projected to be larger, and it arrives on a grid carrying AI compute loads that did not exist a decade ago. When the Niño 3.4 region peaks above 3 degrees Celsius in November or December, the extreme weather that follows will collide with peak data center demand on the same transmission circuits. Every power purchase agreement, every interconnection study, every capacity plan built on historical weather data is about to encounter a season the baseline has never seen. The models have shown this since July. November is four months away.