The State of the Climate report, the most comprehensive annual global climate report, is published each year as a supplement to the Bulletin of the American Meteorological Society. The newly released State of the Climate in 2025 features contributions from hundreds of authors worldwide, and details record-breaking global greenhouse gas concentrations, fossil fuel emissions, sea level, and ocean heat, as well as other key trends, events, and phenomena.
In this post, we speak with Global Climate chapter lead editor Robert Dunn of the UK Met Office about overall global indicators and anomalies in 2025. Read the Global Climate chapter here [PDF].
What was your role in the report?
I am the lead editor for Chapter 2 – Global Climate – ably supported by Nadine Gobron (JRC-ISPRA), Gary Morris (NOAA GML), and Josh Blannin (UK Met Office, now University of Exeter). We share the management of the different sections across the editorial team, with Nadine looking after the Cryosphere, Terrestrial, and some of the Hydrological sections; Gary the Atmospheric Composition; Josh the Temperature sections; and then I look after the rest, alongside producing the figures and overview, and the overall management of the chapter. I have been leading the chapter since 2016, and have been an editor and author since before that time.
Producing the report is one of the highlights of my year. It is a substantial effort and incredibly worthwhile. This valuable collection of assessments of the conditions of the previous calendar year is a vital component of efforts to understand our current climate. Each year the editors and authors strive to improve what we can say has occurred in our climate, how much we can say, and how clearly we can say it. As a previous State of the Climate editor once said, this report is not just for the now, it is also for future historians who will be able to look and read about what the community knew shortly after the events described—whether that’s from the perspective of a few years’ or a few decades’ time.
What was unexpected or differed from the norm last year?
To answer this question, it really depends on what we mean by “the norm.” In one perspective, when we compare to the average conditions measured over a multidecadal baseline (usually a 30-year period as recommended by the World Meteorological Organization, e.g., 1961–1990, 1991–2020) then 2025 was an exceptional year. Temperature metrics were well above normal (the average conditions between 1991 and 2020), and this difference from the 1991–2020 average was reflected across most of the other sections in the report. 2025 only comes across as unexceptional when compared with 2024 or 2023.
The other perspective is to look at the long-term change for the metrics we monitor. Many of these are directly or indirectly influenced by the increasing atmospheric concentrations of greenhouse gases from the burning of fossil fuels. The concentrations of the three main greenhouse gases again reached record levels in 2025. And so, for many metrics, although 2025 did not set record levels, it was a year where values were consistent with what we expect to see as a consequence of our continued emission of greenhouse gases.
What are some noteworthy trends or events you’d want to highlight from the global climate in 2025?
As I said before, the fact the temperature metrics didn’t set new global records, but were still far above long-term normals, is important to reiterate. Similarly, that greenhouse gas concentrations did set new records is also key to remember. However, I’d like to highlight some other metrics in our chapter as well.
- Globally, the amount of carbon emitted from fires in 2025 was the lowest in the record, which started in 2003. Relative to 2024, which was one of the highest annual totals since 2015, it decreased by 26%. This is a rare glimmer of a change for the better, as it is driven by a long-term reduction of fires in many savanna and grassland regions, which dominate the global values and are influenced by agricultural expansion. Africa (which contributed ~40% of the global emissions) and South America also had their lowest emissions on record. Despite this positive global picture, North America (especially Canada) experienced its third consecutive year of extreme fire activity, similar to levels in 2024 and second only to 2023. And the EU also had the highest annual total, driven by unprecedented activity on the Iberian Peninsula. And these are examples of an increasing trend in forested regions where climate change-induced droughts increase flammability.
- Global soil moisture conditions were drier on average in 2025 than in any other year over the 1991–2025 period. This was driven by exceptionally low levels in the Northern Hemisphere throughout the year. In the Southern Hemisphere, conditions became less dry as the year progressed.
Dry soils can exacerbate high air temperatures during summer, as there is less water to absorb some of the available energy through evaporation. Low soil moisture also can lead to vegetation drying out, with impacts on crops. As we see in 2026, with substantial fire activity in France, and again in Canada and the Iberian Peninsula, as well as elsewhere around the world, monitoring these two metrics is vital for understanding trends and changes in recent years. - Finally, and for the 38th year in a row, glaciers have had negative mass balance. This means they have lost more ice in the summer than they gained in the winter, and so have become thinner and shorter. Since 1970, around 9,000 gigatons of water have been lost from glaciers—and 41% of this was over the last decade.
What else is new in 2025’s report?
For the report on 2025 we have resisted the urge to add in new metrics or methods to keep the report manageable and accessible. However, our authors continue to use and select the combination of datasets which in their expert assessment give the best understanding of conditions during 2025. Over recent years, datasets which use machine learning tools are becoming part of the available data landscape. In this edition of the State of the Climate, machine learning has been used to extend the carbon monoxide reanalysis after the discontinuation of the MOPITT instrument, so enabling the ongoing monitoring of this atmospheric constituent.
What do you hope to see/what do you worry about in the year ahead?
Although 2025 didn’t set many records across the metrics that we cover in Chapter 2, which focuses on the global picture, for a large number it was only second or third behind 2024 and/or 2023. This was despite ENSO conditions being neutral or weakly La Niña type at that time, with no strong El Niño to boost global temperatures and other characteristics. In fact, 2025 was the warmest year without an El Niño.
At the time of writing, halfway through 2026, there are indications of an extremely strong El Niño forming. The climate in 2025, being only slightly “below” 2023 and 2024, has potentially set some concerning precursor conditions for 2026. However, until the assessments to place 2026 (and maybe 2027) in context are completed, the extent of these is not fully clear.
Anything else you’d like to mention?
This report is something which can only be done through collaboration, dedication, and teamwork. As editors, we rely on the goodwill and efforts of all our authors (over 250 in the Global Climate chapter), who themselves use data products created, curated, and maintained by many more. This report builds on the hard work and expertise of all these people from all over the world, and we are incredibly grateful to them all.
