The El Niño Briefing #7: The warmest September week on record

This week's number is a record, and it is worth being exact about which one.

The warmest September week on record. NOAA's weekly Niño 3.4 anomaly for the week centred 9 September came in at +2.9 °C, up from +2.7 °C the week before. It is the highest reading of the 2026–27 event so far and the warmest September week in NOAA's weekly series, which begins in 1981. The previous September high was +2.1 °C, in 1997.

Only one week in the whole record has been warmer: +3.0 °C on 18 November 2015, at the peak of that event. The difference is timing. 2015 got there in late November; 2026 is 0.1 °C short of it in early September, with the usual October–December peak season still ahead. Closer to South America the numbers are larger still: +4.5 °C in Niño 1+2 off Peru and Ecuador, +3.7 °C in Niño 3.

One caution, because it is the question readers ask: this is the conventional index. The relative index (RONI), which removes the warming of the whole tropical ocean, reads lower — +1.36 for June–August. Both are on the dashboard side by side, with the reason they differ.

A second agency agrees. Australia's Bureau of Meteorology said on 15 September that El Niño "continues to strengthen". Its own relative Niño 3.4 index stood at +2.51 °C for the week ending 13 September, against its El Niño threshold of +0.8 °C, and the Bureau reports that the majority of models it surveys put the peak above +3.0 °C on that index — above the warmest values observed since reliable records began in 1950. That is a model forecast, not an observation, and it is on the Bureau's relative index, not NOAA's. The same update notes the Indian Ocean Dipole index reading +0.65 °C, back above its positive threshold after four weeks in the neutral range; the Bureau is explicit that one week does not mean a positive IOD is underway.

The longer backdrop, from the corals. Two studies this month read El Niño's history out of coral skeletons.

• Galápagos corals covering the last thousand years (Cole et al., Science, 27 August) show year-to-year swings in eastern-Pacific temperature in recent decades that are larger than anything before about 1850 — more than climate models produce from natural variability alone — and the authors trace that to stronger El Niño events.
• Corals from Arno Atoll in the Marshall Islands, in the western Pacific (Communications Earth & Environment, 4 September), suggest the picture is not one-directional: central-Pacific El Niño activity increased between the first and second millennium, while eastern-Pacific activity stayed broadly stable and may even have been stronger than today.

Neither study says anything about how high this event will peak. Together they say that "strongest on record" is measured against a baseline that itself moves, and that where in the Pacific you look matters.

Next. NOAA's weekly file updates on Monday with the week centred 16 September. The next monthly discussion is due on 8 October.

Live dashboard →

— The El Niño Briefing