ScienceIndependently confirmed3 publishers3 min readPublished Updated
Every one of the eight hottest months ever measured has arrived since July 2023
Copernicus put August 2026 at 16.96C, a hundredth of a degree above July 2023 and inside its own uncertainty, so it calls the month a tie. The clustering of the record months is the firmer finding.
The Scientist · Science desk
What happened
- The month ran 1.65C above pre-industrial levels, above the 1.5C long-term figure written into the 2015 Paris agreement.
- NOAA said summer in the Lower 48 was hotter than the 1936 Dust Bowl peak for the first time, averaging 74.37F, nearly four tenths of a degree above the mark set 90 years earlier.
- Ocean temperatures in August tied the March 2024 record for the warmest month measured, and 22 August was the hottest day yet recorded in the world's seas.
Why it matters
- constraint Adaptation specifications sized to August's numbers carry a stated expiry, because Copernicus expects further records before December, so the design object has to be a trend with a widening tail rather than a high-water mark.
- decision UNEP's overshoot position moves the planning question from avoiding 1.5C to reversing warming after the fact, which pulls carbon dioxide removal into portfolios that were assembled on the assumption of avoidance.
- exposure A global mean will not size anyone's risk: Europe's excess deaths and the Nepal and Tibet glacier collapse came from regional extremes that a 16.96C index cannot locate.
- contradiction Dessler says it is time to stop treating each new record as surprising, and Barnes says this is not a stable new normal; together they mean planners can assume neither surprise nor stability, which is a harder brief than either alone.
The weakest number in the Copernicus release is the ranking itself. A gap of one hundredth of a degree, measured with an uncertainty of one hundredth of a degree, cannot order two months, which is why Samantha Burgess used the word tie [1]. As far as the instrument is concerned, July 2023 at 16.95C and August 2026 at 16.96C are the same value [1][2].
The clustering is much harder to argue with. All eight of the warmest months in the Copernicus series sit between July 2023 and August 2026 [12]. The series begins in 1940 [13], which gives 1,040 monthly values through last month, and that 38-month window is about 3.7 per cent of them [23]. A record whose entire top eight lands in its most recent 3.7 per cent is what a shifted distribution looks like when you sample it monthly. (Burgess also says humans have not experienced temperatures warmer than this summer in 100,000 years [19]; that statement comes from outside the instrumental datasets, which start in 1940 and, for NOAA's global series, 1880 [13].)
Burgess puts El Nino's own contribution at one or two tenths of a degree C [7]. Against August's 1.65C anomaly, that is 6 to 12 per cent of the total [20], which is the arithmetic under Friederike Otto's point that the driver is fossil fuel burning and that warming makes every El Nino hotter [8]. The operationally useful form of that figure is what expires when the El Nino does: about two tenths of a degree, at most.
What 16.96C does not measure is exposure. New Scientist reports at least 35,000 more deaths in Europe during the summer's heatwaves than would be statistically expected, and lower than usual harvests of wheat and other crops [15]; it links the August heat to a glacier collapse in Nepal and Tibet with at least 1,300 dead and more than 5,500 still missing [26]. Four European heatwaves and severe drought fed wildfires near Bordeaux and Madrid [16]. A global mean is an index, not a hazard map, and none of that shows up in it.
The Paris comparison deserves the same care. August ran 0.15C above the agreement's 1.5C figure [21], but that figure is defined on a long-term average rather than a single month [4]. The institutions have moved regardless: the UN Environment Programme now says the world will overshoot 1.5C and should focus on bringing temperatures back down afterwards, including through carbon dioxide removal [17], and the UN has described consistent exceedance as unavoidable [18].
Andrew Dessler of Texas A&M says it is time to stop acting surprised that a warming planet keeps getting warmer, and that the open question is not whether records keep falling but whether anything is done about it [9]. Claire Barnes of Imperial College London answers the obvious follow-up directly: asked whether this is the new normal, she says no, because records will keep tumbling until fossil fuel burning stops and the extremes will worsen faster than adaptation can keep pace [10]. Zeke Hausfather of Berkeley Earth reads August, not the traditionally warmer July, taking the all-time mark as a sign the warming is accelerating [14]. Burgess expects more records as the El Nino intensifies through December, with some months perhaps 2C above average [24], which would be 0.35C beyond August [25]. Scientific American reports that scientists expect 2027 to set the annual record whether or not 2026 does [11]. A specification written against August 2026 has a stated expiry, and Copernicus expects the first challenge to it before the year is out.
What to watch
- Whether Copernicus's forecast of monthly anomalies up to 2C above average verifies through December, or the El Nino peaks earlier and monthly figures fall back.
Clarity's read
What the record supports and how the coverage leans. The claims behind it follow.
Reality
- Evidence78
- Adoption
- Insufficient
- Hype gap+15
- Incentives
- Insufficient
- Confidence74
Perspective Coverage
3 publishers- Builder
- Builder 18%
- Operator
- Operator 54%
- Investor
- Investor 28%
Claim ledger
Ranked by verification strength, evidence, and original report placement.
- [1]
August 2026 averaged 16.96 degrees Celsius (62.53F) globally, technically one-hundredth of a degree Celsius above the all-time global mark set in July 2023, but because that is also the uncertainty level it is considered a tie, said Samantha Burgess, strategic climate lead for Copernicus.
ReportedSupportedSource: Copernicus Climate Change Service, via phys.org4 sources— create a free account to open themView cited source - [2]
The previous record, set in July 2023, was 16.95 degrees Celsius (62.51F).
ReportedSupportedSource: Copernicus data reported by Scientific American4 sources— create a free account to open themView cited source - [3]
Global monthly ocean temperatures in August tied a March 2024 all-time record for warmest month recorded, while 22 August set a record for the hottest day recorded in the world's seas, according to Copernicus.
- [4]
Globally, August was 1.65 degrees Celsius above pre-industrial levels, exceeding the longer term 1.5 degrees Celsius threshold set by the 2015 Paris climate agreement.
- [5]
NOAA announced that for the first time, summer in the Lower 48 states was hotter than during the peak of the Dust Bowl in 1936: June through August 2026 in the contiguous United States averaged 74.37F (23.54C), nearly four tenths of a degree warmer than the mark set 90 years earlier.
ReportedSupportedSource: US National Oceanic and Atmospheric Administration3 sources— create a free account to open themView cited source - [6]
In the United States, August averaged 75.58F (24.21C), beating heat records by more than half a degree Fahrenheit, and the average daytime high was 88.56F (31.42C), also a record.
- [7]
Burgess said El Nino generally adds only one or two tenths of a degree Celsius to the global temperature, with the overwhelming rest of the heat coming from elsewhere.
- [8]
Friederike Otto of Imperial College London said that while El Nino is ramping up the heat, the underlying driver remains the burning of fossil fuels, and that climate change makes every El Nino hotter.
- [9]
Texas A&M climate scientist Andrew Dessler said it is time to stop acting surprised that the warming planet keeps getting warmer, and that the interesting question is not whether heat records keep being broken but whether anything will be done about it.
- [10]
Claire Barnes at Imperial College London said that asked whether this is the 'new normal', the simple answer is no: until fossil fuel burning stops these records will continue to tumble and the extremes will worsen faster than we are able to adapt.
- [11]
Scientists believe that regardless of whether 2026 breaks the record for hottest year, 2027 almost certainly will because of El Nino conditions.
ReportedSupportedSource: Scientific American3 sources— create a free account to open themView cited source - [12]
The eight hottest months on record have all been from July 2023 on, according to Copernicus data.
- [13]
Copernicus data goes back to 1940; NOAA data for the United States goes back to 1895 and its global data, which generally matches Copernicus but takes longer to calculate, dates back to 1880.
- [14]
Berkeley Earth climate scientist Zeke Hausfather said that records being regularly broken, and that August rather than the traditionally warmer July set an all-time record, are signs human-caused climate change is accelerating.
- [15]
At least 35,000 more people died in Europe during the heatwaves than would be statistically expected, and harvests of crops including wheat are expected to be lower than usual.
ReportedSupportedSource: New Scientist2 sources— create a free account to open themView cited source - [16]
Four major heatwaves in Europe, and severe drought, fuelled wildfires near cities including Bordeaux in France and Madrid in Spain and lowered water levels in major rivers.
- [17]
The United Nations Environment Programme said the world will overshoot the Paris Agreement goal of preventing warming of more than 1.5C and should now focus on reversing this warming at a later date, including with approaches such as carbon dioxide removal.
- [18]
The United Nations acknowledged that consistently surpassing the 1.5C threshold is now 'unavoidable'.
- [19]
Humans have not experienced temperatures warmer than this summer in the past 100,000 years, according to Samantha Burgess at Copernicus.
ReportedSupportedSource: Samantha Burgess, Copernicus2 sources— create a free account to open themView cited source - [20]
El Nino's contribution of one to two tenths of a degree Celsius is 6 to 12 per cent of August's 1.65C anomaly above pre-industrial levels.
- [21]
August's 1.65C anomaly was 0.15C above the Paris Agreement's 1.5C figure.
- [22]
The Energy and Climate Intelligence Unit, a UK-based think tank, called on leaders meeting in November for the COP31 climate summit both to reverse the rise in greenhouse gas emissions and to invest more in adapting to current impacts.
- [23]
The July 2023 to August 2026 window that contains all eight of the warmest months is 38 months, about 3.7 per cent of the roughly 1,040 monthly values in the Copernicus record from January 1940 through August 2026.
- [24]
Burgess predicts records will keep being broken as the El Nino intensifies through December, with some months likely to be as much as 2C above average.
- [25]
A month at 2C above average, as Burgess says is possible before December, would be 0.35C beyond August's 1.65C anomaly.
- [26]
The August heat was also linked to a glacier collapse that destroyed several towns in Nepal and Tibet, with at least 1300 people confirmed dead and more than 5500 still missing.
Sources
3 independent publishers whose own reporting we read for this story.
- newscientist.comAugust was the joint hottest month on record
1 article · September 9, 2026
- phys.orgGlobal heat stuck on high. Scientists say August was Earth's hottest month they've measured.
2 articles · September 10, 2026
- scientificamerican.comAugust 2026 ties for hottest month ever recorded
1 article · September 10, 2026
Topics and entities
Follow any of these and your For You feed starts watching them — no settings page required.
Topics
- Global temperature records and the 1.5C thresholdFollow
- El NinoFollow
- Extreme heatFollow
Entities
- Zeke HausfatherFollow
- European Centre for Medium-Range Weather ForecastsFollow
- Energy and Climate Intelligence UnitFollow
- Samantha BurgessFollow
- Paris AgreementFollow
- NOAAFollow
- Friederike OttoFollow
- Andrew DesslerFollow
- UN Environment ProgrammeFollow
- Simon StiellFollow
- Copernicus Climate Change ServiceFollow