Original article by Zeke Hausfather republished from Carbon Brief under a CC license.

The year 2025 was in the top-three warmest years on record, with average surface temperatures reaching around 1.44C above pre-industrial levels across eight independent datasets.

The different temperature records confirm that last year was either the second or third warmest since observations began in the mid-1800s, with razor-thin margins between 2025 and 2023.

Last year also set a new record for ocean heat content, with the oceans absorbing more than 90% of the heat trapped by increasing greenhouse gas concentrations in the atmosphere.

Here, Carbon Brief examines the latest data across the Earth’s oceans, atmosphere, cryosphere and surface temperature. (Use the links below to navigate between sections.) 

Noteworthy findings from this 2025 review include…

• Ocean heat content: It was the warmest year on record for ocean heat content and one of the largest year-over-year increases in ocean heat content. In 2025, the oceans added 39 times more heat than all annual human energy use.
• Global surface temperatures: The year 2025 is effectively tied with 2023 as the second-warmest year on record – coming in at between 1.33C and 1.53C above pre-industrial levels across different temperature datasets and 1.44C in the synthesis of all groups.
• Second warmest over land: Global temperatures over the world’s land regions – where humans live and primarily experience climate impacts – were 2C above pre-industrial levels, just below the record set in 2024.
• Third warmest over oceans: Global sea surface temperatures were 1C above pre-industrial levels, dropping from 2024 record levels due to fading El Niño conditions.
• Regional warming: It was the warmest year on record in areas where, collectively, more than 9% of the global population lives. 
• Unusual warmth: The exceptionally warm, record-setting temperatures over the past three years (2023-25) were driven by continued increases in human emissions of greenhouse gases, reductions in planet-cooling sulphur dioxide aerosols, variability related to a strong El Niño event and a strong peak in the 11-year solar cycle.
• Comparison with climate models: Observations for 2025 were nearly identical to the central estimate of climate model projections in the Intergovernmental Panel on Climate Change (IPCC) sixth assessment report (AR6).
• Heating of the atmosphere: It was the second warmest year in the lower troposphere – the lowest part of the atmosphere.
• Sea level rise: Sea levels reached record highs, continuing a notable acceleration over the past three decades.
• Shrinking glaciers and ice sheets: Cumulative ice loss from the world’s glaciers and from the Greenland ice sheet reached a new record high in 2025, contributing to sea level rise.
• Greenhouse gases: Concentrations reached record levels for carbon dioxide (CO2), methane and nitrous oxide.
• Sea ice extent: Arctic sea ice saw its lowest winter peak on record as well as its 10th-lowest summer minimum extent, while Antarctic sea ice saw its third-lowest minimum extent.
• Looking ahead to 2026: Carbon Brief predicts that global average surface temperatures in 2026 are likely to be between the second and fourth warmest on record, similar to 2023 and 2025, at around 1.4C above pre-industrial levels.

Ocean heat content sets a new record

The year 2025 was the warmest on record for the heat content of the world’s oceans. 

Ocean heat content (OHC) increased by around 500 zettajoules – billion trillion joules – since the 1940s.

The heat increase in 2025 alone compared to 2024 – about 23 zettajoules – is around 39 times as much as the total energy produced by all human activities on Earth in 2023 (the latest year in which global primary energy statistics are available). It was also the largest increase in ocean heat content since 2017 (following the strong El Niño event of 2016).

Human-emitted greenhouse gases trap extra heat in the atmosphere. While some of this warms the Earth’s surface, the vast majority – around of 93% – goes into the oceans. About two-thirds of this accumulates in the top 700 metres, but some also ends up in the deep oceans. 

The figure below shows annual OHC estimates from the Chinese Institute for Atmospheric Physics (IAP) between 1950 and present for the upper 700 metres (light blue shading) and 700-2,000 metres (dark blue) of the ocean.

In a new paper published last week, researchers found that the rate of OHC increase over the past 15 years is unprecedented over the observational record in the IAP dataset. More broadly, there has been a distinct acceleration in OHC after 1991 – and recent OHC growth rates are generally consistent with satellite measurements of Earth energy imbalance (EEI).

(Energy imbalance is a measure of how much surplus heat there is in the Earth’s climate system. It is the difference between how much energy enters Earth’s atmosphere from the sun and how excess heat is radiated back into space as the world warms.)

In many ways, OHC represents a much better measure of climate change than global average surface temperatures, because it is where most of the extra heat ends up and is much less variable on a year-to-year basis than surface temperatures. 

Surface temperatures tied at second warmest

To assess global surface temperatures in 2025, Carbon Brief uses eight independent datasets: NASA, NOAA, the Met Office Hadley Centre/University of East Anglia’s (UEA) HadCRUT5, Berkeley Earth, Copernicus ERA5, Japan’s JRA-3Q, DCENT and China-MST.

The analysis reveals that global surface temperatures were between the second and third warmest since records began in the mid-1800s. Temperatures effectively tied with 2023 within the margin of uncertainty, below the record set last year in 2024. 

The figure below shows global surface temperature records from the eight datasets. 

Global surface temperature records can be calculated back to 1850, though some groups such as NASA GISTEMP choose to start their records in 1880 when more data was available.

Prior to 1850, records exist for some specific regions, but are not sufficiently widespread to calculate global temperatures with high accuracy (though newly published research has attempted to extend this back to 1781). 

These longer surface temperature records are created by combining ship- and buoy-based measurements of ocean sea surface temperatures with temperature readings of the surface air temperature from weather stations on land. (Copernicus ERA5 and JRA-3Q are an exception, as they use weather model-based reanalysis to combine lots of different data sources over time.) 

Some differences between temperature records are apparent early in the record, particularly prior to 1900 when observations are more sparse and results are more sensitive to how different groups fill in the gaps between observations. However, there is strong agreement between the different temperature records for the period since 1970, as shown in the figure below.

Global temperatures over the past three years clearly stand out as much warmer than anything that has come before, well above the prior record set in 2016. More broadly, the 11 warmest years on record all happened in the past 11 years.

Two of the eight datasets analysed by Carbon Brief – NASA and DCENT – had 2025 as the second-warmest year behind 2024, while six of the datasets had 2025 as the third-warmest year behind both 2023 and 2024. 

However, in nearly all cases the difference between 2023 and 2025 falls within each dataset’s published uncertainty range, making it effectively a tie between the two years.

The table below shows the reported 2025 global temperature anomalies (relative to each group’s 1850-1900 pre-industrial baseline), as well as a 2025 value using a common pre-industrial baseline between the 1850-1900 and 1981-2010 periods across the five groups with data back to 1850 (NOAA, Hadley/UEA, Berkeley Earth, DCENT and China-MST). 

Reported temperature anomalies range from as low as 1.33C (NOAA) to as high as 1.53C (DCENT), primarily reflecting differences in the early part of the record. The 2025 values with a common baseline have a much smaller range, from 1.41C (NOAA) to 1.47C (Copernicus).

Separate land and ocean temperatures are not available yet from all of these groups. However, Berkeley Earth reports that global land temperatures in 2025 were the second warmest on record, at 2.03C above pre-industrial levels, while ocean temperatures were the third warmest at 1.03C.

Global land regions – where the global human population lives – has generally been warming around 70% faster than the oceans and 40% faster than the global average since 1970.

The year started off quite hot, with January 2025 setting a new record as the warmest January. All other months of the year ended up being either the second or third warmest on record after 2024 and 2023.

The figure below shows each month of 2025 in dark red, compared to all prior years since 1850. Each year is coloured based on the decade in which it occurred, with the clear warming over time visible, as well as the margin by which both 2023, 2024 and 2025 exceeded past years.

Extreme regional temperatures

While the globe as a whole was tied as the second warmest on record, many different regions of the planet set new records in 2025. 

The figure below shows how global temperature deviated from the average in 2025 across the world. Areas shaded in red were warmer than the baseline period (1951-80) used by Berkeley Earth, whereas the few blue areas experienced cooler temperatures.

Collectively, approximately 770 million people – 9.3% of Earth’s population – live in places that experienced their warmest year on record in 2025. This was mostly concentrated in Asia, including around 450 million people in China.

The figure below highlights regions of the planet that experienced their top-five warmest (red shading) or coldest (blue) temperatures on record in 2025. Overall, around 9% of the planet set a new record, including 11% of the land and 8% of the ocean. No location on the planet experienced record cold temperatures – or even top-five record cold temperatures – for the year as a whole.

Drivers of recent record warmth

Global temperatures over the past three years have been unusually warm, well above what would be expected given the long-term warming trend of around 0.2C per decade since the 1970s. 

Recent research has found that global warming has accelerated in recent years to around 0.27C per decade, though this acceleration is largely in-line with climate model projections under scenarios where greenhouse gas emissions continue to rise while emissions of planet cooling aerosols are reduced.

According to analysis from Berkeley Earth, the odds of global temperatures over 2023-25 occurring as a result of greenhouse gas emissions and natural variations in the Earth’s climate alone “is less than one-in-100” and “likely indicates that recent years have been impacted by additional warming factor(s)”.

The figure below shows how the exceptional warming spike of 2023-25 compares to the longer-term warming trend and historical climate variability.

Carbon Brief recently explored the drivers of recent warmth in more detail, finding that it is likely to have been driven by a combination of:

• A strong El Niño event that developed in the latter part of 2023.
• Rapid declines in sulphur dioxide emissions – particularly from international shipping and China.
• An unusual volcanic eruption in Tonga in 2022 that put a large amount of aerosols and water vapor into the upper atmosphere.
• A stronger-than-expected solar cycle.

This is illustrated in the figure below, which provides an estimate of the impact of each of these different factors on 2023 and 2024 temperatures, along with their respective uncertainties. 

The sum of all the factors is shown in the “combined” bar, while the actual warming compared to expectations is shown in red. The upper chart shows 2023, while the lower one shows 2024.

The first bar includes both El Niño and natural year-to-year climate variability; the height of the bar reflects the best estimate of El Niño’s effects, while the uncertainty range encompasses year-to-year variability in global temperatures that may be – at least in part – unrelated to El Niño. 

While a similar analysis has yet to be undertaken for 2025, the end of El Nino conditions and the development of a modest La Nina would have driven temperatures down, while the warming impact of shipping, Chinese aerosol declines would have slightly increased. The warming effect of the solar cycle would likely have remained flat or slightly declined as solar cycle 25 passed its peak. 

Finally, a World Meteorological Organization (WMO) assessment of the Hunga Tonga-Hunga Ha’apai volcano found that “the record-high global surface temperatures in 2023-24 were not due to the Hunga eruption”. 

The report suggested that the volcano had a small cooling effect (-0.03C) globally in 2023 and 2024. This might switch to a small warming effect (+0.03C) in 2025 and 2026 as the planet-cooling aerosols from the volcano fall back down to the surface but some of the stratospheric water vapour remains, it noted.

However, it added, these effects are “indistinguishable from background variability in the current climate”.

El Niño and La Niña are generally the largest drivers of year-to-year variability in global temperatures. The figure below shows the El Niño (red shading) and La Niña (blue) conditions over the past 40 years (collectively referred to as the El Niño-Southern Oscillation, or “ENSO”). 

Carbon Brief has used the historical relationship between ENSO conditions and temperature to effectively remove the effects of El Niño and La Niña events from global temperatures, as shown in the figure below. 

This analysis indicates that El Niño cooled global temperatures in 2025 around -0.05C, following a boost to global temperatures of around 0.12C in 2024, compared to the estimate of global temperatures with both El Niño and La Niña events removed.

This suggests that the shift from El Nino to La Nina conditions can fully explain the decline in global temperatures between 2024 and 2025 and that  2025 would have likely been the warmest year in the observational record if it had not been for the effects of ENSO.

Scientists provided estimates of where they expected 2025 temperatures to end up at the start of the year. 

The figure below shows estimates by four different groups that provided temperature predictions for the year prior to any data being collected – the Met Office, NASA’s Dr Gavin Schmidt, Berkeley Earth and Carbon Brief’s own estimate — compared to what actually transpired.  

Unlike in 2023 –and, to a lesser extent, 2024 –when start-of-year predictions were notably low, 2025 fell reasonably in-line with what was expected. The Met Office estimate was nearly exactly on target, with Berkeley Earth’s being close as well. Carbon Brief and Schmidt’s estimates were a little on the low side, but actual temperatures were well within the estimated error bars.

Observations in-line with climate model projections

Climate models provide physics-based estimates of future warming given different assumptions about future emissions, greenhouse gas concentrations and other climate-influencing factors.

Here, Carbon Brief examines a collection of climate models – known as CMIP6 – used in the 2021 science report of the IPCC’s sixth assessment. 

In CMIP6, model estimates of temperatures prior to 2015 are a “hindcast” using known past climate influences, while temperatures projected from 2015 onward are a “forecast” based on an estimate of how things might change. 

The figure below shows how observations compare to the full ensemble of 37 CMIP6 models under the middle-of-the-road SSP2-4.5 emissions scenario for future projections. The red line represents the average of all the models and the red areas showing the 5th to 95th percentile range. The average of the eight observational temperature datasets are plotted as dots on top of the climate model data.

The chart illustrates how observations have generally been a bit below the model average over the past two decades, were slightly above model average in 2024 and are more or less dead on in 2025.

However, the ensemble of CMIP6 models differs from the main projection of future warming in the recent IPCC AR6 report. A subset of CMIP6 models have unrealistically high climate sensitivity and they reproduce historical observations poorly. 

To account for this, rather than simply averaging all the models – as had been done in prior assessments – the IPCC employed an approach that effectively weights models by their performance. As a result, the models align better with the range of climate sensitivity derived from multiple different lines of evidence. 

The chart below shows the assessed warming projections from the IPCC AR6 report in red, with historical observations since 1850 as black dots.

The chart reveals that observed global surface temperatures (black dots) are further above the modeled central estimate 2023-25, but generally remain within the IPCC assessed range. 

Climate models broadly expect an acceleration of warming in the current period in a scenario like SSP2-4.5 where emissions of CO2 and other greenhouse gases continue to modestly increase, but emissions of planet-cooling aerosols like sulphur dioxide are rapidly reduced.

Second-warmest atmospheric temperatures

In addition to surface measurements over the world’s land and oceans, satellite microwave sounding units have been providing estimates of temperatures at various layers of the atmosphere since 1979. 

The lowest layer of the atmosphere that satellite microwave units provide temperature estimates for is the lower troposphere. This data reflects temperatures a few kilometres above the Earth’s surface. It reveals a pattern of warming in the lowest troposphere that is similar – though not identical – to surface temperature changes. 

The records produced by Remote Sensing Systems (RSS), the University of Alabama, Huntsville (UAH) and NOAA show 2025 as the second warmest year on record in the lower troposphere, after 2024. The chart below shows the three records for the lower troposphere, using a more recent baseline period (1981-2010) given the absence of satellite data before 1979.

The lower troposphere tends to be influenced more strongly by El Niño and La Niña events than the surface. Therefore, satellite records show correspondingly larger warming or cooling spikes during these events. This explains why there was both a bigger increase between 2023 and 2024 and a bigger decline between 2024 and 2025 in the satellite record than in surface records.

The lower-tropospheric temperature records show large differences after the early 2000s. RSS shows an overall rate of warming quite similar to surface temperature records, while UAH and NOAA show considerably slower warming in recent years than has been observed on the surface. 

Greenhouse gas concentrations reach new highs

Greenhouse gas concentrations reached a new high in 2025, driven by human-caused emissions from fossil fuels, land use and agriculture.

Three greenhouse gases – CO2, methane (CH4) and nitrous oxide (N2O) – are responsible for the bulk of additional heat trapped by human activities. CO2 is by far the largest factor, accounting for roughly 42% of the increase in global surface temperatures since the pre-industrial era (1850-1900).

Methane accounts for 28%, while nitrous oxide accounts for around 5%. The remaining 25% comes from other factors including carbon monoxide, black carbon and halocarbons, such as chlorofluorocarbons (CFCs).

Human emissions of greenhouse gases have increased atmospheric concentrations of CO2, methane and nitrous oxide to their highest levels in at least a few million years – if not longer. 

The figure below shows concentrations of these greenhouse gases – in parts per million (ppm) for CO2 and parts per billion (ppb) for methane and nitrous oxide – from the early 1980s through to October 2025 for CO2 and September 2025 for CH4 and N2O (the most recent data currently available).

Sea level is rising rapidly

Modern-day sea levels have risen to a new high, due to a combination of melting land ice (such as glaciers and ice sheets), the thermal expansion of water as it warms and changes in land water storage. 

In recent years, there have been larger contributions to sea level rise from melting ice sheets and glaciers, as warmer temperatures accelerate ice sheet losses in Greenland and Antarctica.

Since the early 1990s, the increase in global sea level has been estimated using altimeter data from satellites. Earlier global sea levels have been reconstructed from a network of global tide gauge measurements. This allows researchers to estimate how sea level has changed since the late 1800s. 

The chart below shows five different modern sea level rise datasets (blue lines), along with satellite altimeter measurements as assessed by AVISO (in black) after 1993. (As sea level rise data has not yet been released for the whole year, the 2025 value is estimated based on data through to November.)

Sea levels have risen by over 0.2 metres (200mm) since 1900. While sea level rise estimates mostly agree in recent decades, larger divergences are evident before 1980. There is also evidence of accelerating sea level rise over the post-1993 period when high-quality satellite altimetry data is available. 

(To understand more on how climate change is accelerating sea level rise, read Carbon Brief’s explainer.)

Shrinking glaciers and ice sheets

A significant portion of global sea level rise is being driven by melting glaciers on land. 

Scientists measure the mass of glaciers around the world using a variety of remote-sensing techniques, as well as through GRACE measurements of the Earth’s gravitational field. The balance between snow falling on a glacier and ice loss through melting and the breaking off – or “calving” – of icebergs determines if glaciers grow or shrink over time.

The World Glacier Monitoring Service is an international consortium that tracks more than 130 different glaciers in 19 different regions around the world. The figure below shows the change in global average glacier mass from 1950 through to the end of 2024. (2025 values are not yet available.) Note that glacier melt is reported in metres of water equivalent, which is a measure of how much mass has been lost on average.

Greenland ice sheets have become a larger contributor to sea level rise in recent years due to accelerating loss of mass. The year 2025 was the 29th in a row where Greenland lost ice overall, with 105bn tonnes of ice lost over the 12 months from September 2024 to August 2025. Greenland last saw an annual net gain of ice in 1996.

The figure below shows the cumulative mass balance change – that is, the net ice loss – from Greenland between 1970 and 2025. The authors find that Greenland has lost around 6tn tonnes of ice over the past 50 years – more than 700 tonnes lost per person for every person on the planet.

Lowest winter Arctic sea ice extent

Arctic sea ice saw its lowest winter peak on record as well as its 10th-lowest summer minimum extent, while Antarctic sea ice saw its third-lowest minimum extent. 

Both the Arctic and Antarctic were at the low end of the historical (1979-2010) range for most of 2025, with new daily lows recorded for Arctic sea ice extent in January, February, March, June and December.

The figure below shows both Arctic (red line) and Antarctic (blue line) sea ice extent for each day of the year, along with how it compares to the historical range (corresponding shading).

Looking ahead to 2026

There is reason for caution when estimating likely temperatures for 2026. 

In 2023, temperatures were significantly higher than predictions made at the start of the year, while 2024 temperatures were towards the high end of annual predictions. Temperatures in 2025 were more in-line with predictions, albeit still on the higher side for three out of the four predictions included above.

There are currently weak La Niña conditions currently present in the tropical Pacific, which are expected to extend through February. This would somewhat suppress temperatures in the first half of the year. However, the latest forecasts suggest a growing likelihood of El Niño conditions developing by June, which may lead to warmer temperatures in late 2026 – and potentially much warmer temperatures in 2027.

Carbon Brief predicts that global average surface temperatures in 2026 are likely to be between the second and fourth warmest on record, similar to 2023 and 2025, at around 1.4C above pre-industrial levels.

This is the fourth published temperature prediction for 2026, after those already produced by the Met Office, NASA’s Dr Gavin Shmidt and Berkeley Earth.

The figure below shows the four different 2026 predictions compared to the average of eight different temperature records explored in this article. (These have been “normalised” to show 2026 warming relative to the 2023-25 average to allow a clear comparison, given that each of the predictions was originally presented for a different temperature record.)

Carbon Brief’s prediction of likely 2026 temperatures is based on a statistical model using the average temperature of the past year, the latest monthly temperature and projections of ENSO conditions over the first three months of 2026.

The Met Office, Dr Schmidt, Berkeley Earth and Carbon Brief estimates all have 2026 ending up as somewhere between the second- and fourth-warmest year on record, with the best estimate as being more or less tied with 2023 and 2025. 

There is a very small chance that 2026 could end up beating 2024 as the warmest year on record, or end up below 2016 as the fifth or sixth warmest year.

However, with the growing likelihood of El Niño conditions developing in the second half of 2026, it is increasingly likely that 2027 will challenge 2024 for the title of the warmest year on record. The rate of warming has notably accelerated over the past 15 years and the period of exceptionally warm years that started in 2023 shows no signs of abating. 

Original article by Zeke Hausfather republished from Carbon Brief under a CC license.

• State of the climate: 2025 on track to be second or third warmest year on record
• State of the climate: 2025 close behind 2024 as the hottest start to a year
• State of the climate: 2024 sets a new record as the first year above 1.5C
• State of the climate: 2024 will be first year above 1.5C of global warming