September 2026 Was the Second-Warmest September on Record, Copernicus Reports

The Copernicus Climate Change Service (C3S) reported that September 2026 was the second-warmest September on record, with a global average surface air temperature of 16.27°C, which is 1.63°C above the 1850 to 1900 pre-industrial level. Sea surface temperature set a September record. This article sets out the figures from the bulletin published on October 14, 2026 and explains how to read them.

What did Copernicus report for September 2026?

C3S gives these headline values:

• Global surface air temperature: 16.27°C, which is 1.63°C above the pre-industrial level
• Rank: second-warmest September, behind 2023
• Sea surface temperature (60°S to 60°N): 21.05°C, a September record that beat 20.92°C set in 2023
• Streak above 1.5°C: the second consecutive month above that threshold this year, after a run that ran in October and November 2025

By our own arithmetic, the sea surface record is 0.13°C above the previous one, and the implied pre-industrial September air temperature is about 14.64°C (16.27 minus 1.63). C3S does not print that second figure on the page we used.

What is the 1.5°C threshold?

The Paris Agreement sets a long-term goal of limiting global warming to well below 2°C, with efforts to stay within 1.5°C. A single month above 1.5°C is not the same as breaching the long-term goal, which is judged over decades. C3S Director Carlo Buontempo said that crossing the Paris Agreement’s long-term limit is only a matter of time. We report the quotation as stated. The page we used does not give the 12-month average or the year-to-date figure, so we do not state how close the current year is to the threshold. Our article on global CO2 emissions covers the emissions behind the trend, and our entry on climate change gives the background.

What about the oceans and El Niño?

C3S reports that the monthly sea surface temperature in the Niño 3.4 region reached an all-time high for September, though the page gives no numeric value. Niño 3.4 is a region of the tropical Pacific used to monitor El Niño. Our article on the 2026 El Niño advisory covers the NOAA outlook, and our article on record heat in 2026 covers earlier months. Sea surface temperatures matter because the ocean absorbs most of the extra heat in the climate system, which is a general point not stated in the bulletin.

What happened to sea ice?

The Antarctic maximum of 18.19 million square kilometers on September 10 was the third lowest in the 48-year record. The Arctic minimum of 5.06 million square kilometers on September 13 was the 14th lowest in the satellite record. Our article on the Arctic sea ice minimum compares agency reports in detail.

What did the bulletin say about Europe?

C3S reports widespread dryness across much of western and central Europe in September, with wetter-than-average conditions in parts of northern and northeastern Europe and Türkiye. It also reports record-low September river flows on parts of the Rhine, Danube, Vistula and Garonne-Dordogne. The page gives no precipitation numbers, so we report only the pattern.

How to read the numbers

Three points help readers. First, the headline 16.27°C is an absolute air temperature, while 1.63°C is an anomaly, meaning a difference from a baseline. Second, the choice of baseline changes the anomaly: C3S quotes this one against 1850 to 1900, not against the more recent 1991 to 2020 period, whose anomaly the page does not give. Third, a rank such as second-warmest compares one September with all earlier Septembers in the record, and 2023 remains warmer. These are general reading rules and are not statements from C3S.

How does September compare with the year so far?

The bulletin page frames September within a pattern: it was the second month in a row above 1.5°C this year, and the previous run of such months came in October and November 2025. That wording shows that months above the threshold are now recurring rather than isolated, although the page does not count how many months this year have crossed it. Because we do not have the full list, we avoid giving a tally. A reader who wants the monthly sequence can compare the C3S bulletins for each month, which are published in the same series.

Why do monthly records matter?

Monthly values vary with weather and with natural cycles, so one month cannot establish a trend. Scientists look at many months and years together. Our article on Copernicus climate data for 2025 covers the previous full-year summary. Monthly bulletins like this one help track the pattern as it develops, and C3S publishes a new one each month.

Glossary

• Anomaly: the difference between a measured value and a reference average
• Pre-industrial: the baseline years 1850 to 1900 used here
• Sea surface temperature: the temperature of the ocean’s top layer
• Niño 3.4: a tropical Pacific region used to track El Niño
• Reanalysis: a method that combines observations with a model to build a consistent record

These definitions are general explanations and are not quoted from C3S.

What is not covered here?

The bulletin page we used does not give the anomaly against 1991 to 2020, the 12-month average or the year-to-date value. We have not filled them in. For those numbers readers should open the full Copernicus bulletin.

Key facts at a glance

• September 2026 air temperature: 16.27°C, 1.63°C above pre-industrial
• Rank: second-warmest September, behind 2023
• Sea surface temperature: 21.05°C, a September record
• Antarctic sea ice maximum: 18.19 million square kilometers, third lowest
• Arctic sea ice minimum: 5.06 million square kilometers, 14th lowest

Frequently asked questions

How warm was September 2026?

Second-warmest on record, at 1.63°C above pre-industrial levels.

Did the world pass 1.5°C?

The month was above 1.5°C, but the Paris Agreement goal is judged over the long term.

Was the ocean warmer than ever?

C3S reports a September record for sea surface temperature at 21.05°C.

Who publishes the bulletin?

The Copernicus Climate Change Service, part of the European Union’s Earth observation programme. That description is general and is not taken from the bulletin page.