Berkeley Earth

Scientists thought they understood global warming. Then the past three years happened. The last 30 years are the fastest warming period since 1880, according to a Washington Post analysis of NASA data. By John Muyskens and Shannon Osaka

Excellent @washingtonpost.com piece on the signs that global warming is accelerating by @shannonosaka.bsky.social and @johnmuyskens.bsky.social, featuring @hausfath.bsky.social @rarohde.bsky.social @cjsmith.eu

www.washingtonpost.com/climate-envi...

Warming stripes for different layers of the atmosphere and ocean from 1960-2025. The ocean and troposphere are warming, the stratosphere is cooling.

Climate 'fingerprints' mark human activity from the top of the atmosphere to the bottom of the ocean

theconversation.com/climate-fing...

With cold outbreaks sweeping parts of the US, some have argued that climate change is to blame. But the proposed mechanism remains quite controversial in the scientific community, and the number of extreme cold events have been decreasing almost everywhere: www.theclimatebrink....

The State of the Climate 2026 | Ep242: Zeke Hausfather YouTube video by Cleaning Up Podcast

In this week's episode of @cleaninguppod.bsky.social, Bryony sat down with US-based climate scientist @hausfath.bsky.social for an update on all things climate science. Zeke is extremely grounded, no exaggerations and fake scenarios. But oh, the comments already!
www.youtube.com/watch?v=EzyS...

If updates to the legacy 1° product would still be useful, feel free to reach out to data@berkeleyearth.org
and we’ll do our best to accommodate! (3/3)

In the meantime, we’ve made drop-in 1° and 5° resolution datasets derived from the high-resolution product available without registration on our Data Overview page (berkeleyearth.org/data/) under Global Temperature Data → High-Resolution TAVG/TMIN/TMAX Beta. (2/3)

Hi @zacklabe.com - regarding the 1° product: we’re in the process of transitioning our core product to the new high-resolution (0.25°) dataset, and will be moving away from issuing regular updates to the legacy 1° product going forward. We hope to finalize this transition in 1-3 months. (1/3)

Berkeley Earth High-Resolution Data Set Access Request Please complete the form below. You will receive an email with access instructions within 24 hours. For technical questions about the high-resolution data products, please contact data@berkeleyeart...

Following the resolution of a bug that delayed the export of the most recent monthly data update, the public files are now updated through December 2025 and are available on our website (or via request below). Thx to all who flagged this for us!

Access is available by request: tinyurl.com/ty6wbyjm

Hi @edhawkins.org Thanks for flagging this...the bug in our export pipeline that prevented the Dec update from publishing correctly has been resolved. The global TAVG files are now available thru Dec 2025. Appreciate you checking in (and your patience)!

12-month moving average of global mean temperature from 1965 to 2025, with the linear trend 1970-2019 highlighted and the upward divergence in recent years.

For 50 years, global warming had a very consistent trend (+0.19 °C/decade) with a boring, predictable range of natural variations around it.

During the last three years, we've broken out above that range, suggesting the pace of change has quickened.

Time series of annual average global mean temperature 1850-2025.

While the world sometimes seems metaphorically on fire right now, global warming is still progressing as well.

In 2025, global warming delivered the 3rd warmest year since measurements began. A modest step down from 2024's records, but still well above 20th century norms.

🧪🧵

Bar chart showing global annual temperature rise since 1940 relative to pre-industrial levels. 2025 ranks as the third-warmest year on record, just slightly cooler than 2023, with 2024 the warmest. The 2023–2025 average exceeds 1.5°C for the first time.

Line graph of global temperature anomalies from 1850 to 2025 relative to the 1850–1900 average. Temperatures rise sharply after 1970, reaching about 1.3°C above pre-industrial levels by 2025. Error bars show uncertainty, and a shaded range highlights likely values for 2026.

Heatmap of monthly global temperature anomalies from 1980 to 2025. Nearly all months since 2023 are warmer than average, with multiple record-breaking months highlighted, including September 2023 and January 2025.

World maps showing surface air temperature anomalies for 2023, 2024, and 2025. Most regions are warmer than average, with widespread red shading indicating much warmer-than-average conditions across land and ocean, especially in 2024 and 2025.

Looking for additional confirmation?

@copernicusecmwf.bsky.social & @berkeleyearth.org independently agree.

Copernicus: 2025 was 0.13°C below 2024, the warmest year on record, and only 0.01°C less than 2023, underscoring how close these years are at the top.

Lest we forget the greatest global challenge we all are facing...

There are numerous superb write ups of 2025 already. See for example, Berkeley Earth

bsky.app/profile/berk...

A chart back to 1970 showing the global average temperate rise, which matches the warming effects of CO₂, methane and other warming gases, until an anomalous heat spike after 2023.

Heat and greenhouse gas pollution rise in lock-step — until 2023, when the thermostat goes bonkers.

🎁🔗 The 2025 climate numbers are in.
www.bloomberg.com/news/feature...

Update: An initial version of this report was accidentally published with a mislabeled temperature anomaly map showing 2024 averages instead of 2025 averages. We regret the error.

Below is the corrected map showing the 2025 averages.

The tide seems to be finally turning on global coal, as both China and India are now adding enough clean energy to reduce their coal generation: www.carbonbrief.org/...

Global Temperature Report for 2025 - Berkeley Earth 2025 was the 3rd warmest year on Earth since direct observations began, and recent warming appears to be moving faster than expected.

🌍 Berkeley Earth’s 2025 Global Temperature Report is out.
2025 was the 3rd warmest year on record. Annual global temperature reached 1.44 ± 0.09°C above the pre-industrial baseline; 770 million people experienced locally record-warm annual conditions.

Full report: berkeleyearth.org/global-tempe...

In a newly published paper, we find that in 2025 human emissions of greenhouse gases added around 23 billion trillion joules of heat to the world's oceans – 39 times as much as the annual energy produced by all human activity on Earth. link.springer.com/ar...

📊 Berkeley Earth 2025 Annual Temp Report
Release: Jan 14 4am CET | Jan 13 10pm ET | 7pm PT
Embargoed Materials: Jan 13 (TBD)
Media Sign up: tinyurl.com/mrxuy5a2

Press briefing + Q&A will follow at 8am PST | 5pm CET on Jan 14.
Register: tinyurl.com/2hj6j8t3

Inquiries: 📩 media@berkeleyearth.org

📊 Berkeley Earth 2025 Annual Temp Report
Release: Jan 14 4am CET | Jan 13 10pm ET | 7pm PT
Embargoed Materials: Jan 13 (TBD)
Media Sign up: tinyurl.com/mrxuy5a2

Press briefing + Q&A will follow at 8am PST | 5pm CET on Jan 14.
Register: tinyurl.com/2hj6j8t3

Inquiries: 📩 media@berkeleyearth.org

Published today: Importance of beginning industrial-era climate simulations in the eighteenth century

Defining pre-industrial to be 1750, rather than 1850, produces different (simulated) historical climate changes after 1850. Pre-1850 volcanoes & land use matter.

iopscience.iop.org/article/10.1...

The Climate Brink 2025 wrap-up The biggest stories of 2025 and a few predictions for 2026

Over at The Climate Brink @andrewdessler.com and I have an end of the year wrap up chat: www.theclimatebrink.com/p/the-climat...

30-member weather model ensemble experiment simulation changes in near-surface air temperature as a result of Hunga Tonga aerosol and water vapor emissions.

The big special report on the Hunga Tonga eruption is out:

juser.fz-juelich.de/record/10491...

After a couple years of arguing whether the net effect of aerosols+water vapor was warming or cooling, it includes this banger of a model ensemble that literally does cooling then warming.

In a new article over at Carbon Brief, I explore why the past three years – 2023, 2024, and 2025 – have been exceptionally warm. The main culprits turn out to be a combination of El Nino and internal variability, declining aerosols, and a strong solar cycle: www.carbonbrief.org/...

I have a new update to climate model-observation comparisons over at The Climate Brink, covering CMIP3, CMIP5, and CMIP6. Models perform well globally. The latest generation shows too much long-term warming but better reproduces recent trends: www.theclimatebrink....

12-month moving average time series of land-average temperature anomalies 1750-2025 relative to the 1850-1900 average, with 95% uncertainties shown. Two curves show the Berkeley Earth land average with and without the recent addition of weather station data from the HCLIM dataset.

Recently, @berkeleyearth.org expanded our temperature database by integrating the HCLIM data set.

nature.com/articles/s41...

The additional data rescued by HCLIM helps improve early reconstructions, but doesn't fundamentally change the large uncertainties due to sparse sampling.

1/

This #GivingTuesday, we’re reflecting on a year of change and what comes next.
In 2026, Berkeley Earth will launch Synthesis, our most significant update yet—built to deliver clear, local, decision-ready climate intelligence.

💙 Support our work: donate.berkeleyearth.org?mc_cid=58fa7...

New data from @berkeleyearth.org:

🎃 3rd. That is the ranking October 2025 now sits for the WARMEST October of record GLOBALLY, behind 2023 and 2024.
🌍 4% of Earth saw a record-hot October average.
🌡️94%. That's the current odds that 2025 will go down as the THIRD warmest year on record GLOBALLY

October 2025 Temperature Update - Berkeley Earth October 2025 was the 3rd warmest October on record behind 2023 and 2024. 2025 is likely to become the third warmest year on record. La Niña has returned for now but may be short-lived.

For more information on recent weather conditions in a climate context, please refer to our monthly report:

berkeleyearth.org/october-2025...

12/12