ScienceNot yet confirmed elsewhere1 publisher2 min readPublished
Land carbon sink rebounds in 2025 after almost vanishing in 2023 and 2024
Philippe Ciais's team estimates the land carbon sink took up 2.36 gigatonnes of carbon in 2025, 0.71 above the 2015-2022 average. The figure is preliminary and covers a single year, so it addresses the 2023-24 collapse but leaves a slow decline untested.
The Scientist · Science desk
Three figures. Ciais and colleagues estimate the land sink absorbed 2.36 gigatonnes of carbon in 2025. The implied 2015-2022 average is 1.65 gigatonnes. Researchers estimate the sink is 25 per cent smaller than it would have been without climate change and natural variability.
| Figure | Value | Note | Claim |
|---|---|---|---|
| Carbon absorbed by land sink in 2025, estimated | 2.36 Gt of carbon | 5 | |
| 2015-2022 average land sink | 1.65 Gt of carbon | implied average | 12 |
| Sink smaller than without climate change, variability | 25% smaller | researchers' estimate | 14 |
What happened
- Land uptake of CO2 dropped sharply from 2023 in extreme weather, producing a record jump in atmospheric CO2 concentrations in two successive years.
- The preliminary 2025 research also shows a corresponding decline in the growth rate of atmospheric CO2.
- Ciais attributes the recovery largely to tropical forests, where he says the climate became more normal, especially in Africa and the Amazon.
- New Scientist reports that a super El Nino has already begun in the Pacific and is expected to reach record intensity later this year.
Why it matters
- decision Treating 2025 as a new baseline is hard to defend when New Scientist reports that another sharp slump in carbon uptake is expected.
- cost The sink is already 25 per cent smaller than it would have been without climate change and natural variability, and the estimate does not separate those two.
- contradiction Ciais names El Nino and fires as the cause, while O'Sullivan says repeated shocks fuelled by climate change may be weakening the sink, so weather and trend may not be separable.
Ecosystems take up CO2 as they grow and release carbon when they decompose or burn [11]. The land sink is the balance, so a year with less growth and more fire can shrink it from both sides. Philippe Ciais, who led the analysis, said that in 2023 and 2024, El Nino and the Amazon fires almost cancelled the global land sink [3]. He tied that to the atmospheric record: "This cancellation led to a very high CO2 growth rate, because if CO2 is not taken by land, it stays in the atmosphere," he said [4].
The team puts the 2025 sink at 2.36 gigatonnes of carbon, 0.71 above the 2015-2022 average [5]. Subtracting gives an implied average of 1.65 gigatonnes [12]. On that base, 2025 is about 43% higher [13]. The sink overshot its recent baseline. The result is preliminary [2], and New Scientist's account does not give an uncertainty range or say how the sink was estimated, so a 0.71 gigatonne margin is hard to weigh.
On one narrow question we think 2025 gives a firm answer. A sink that had failed for good would not come back to 43% above its 2015-2022 average the next year [13]. Scientists expect the land to shift from sink to source within several decades unless emissions taper [9], and the 2023-24 collapse had researchers worried that shift was arriving decades early [10]. That fear looks mistaken for now. It holds for a collapse, though. One recovery year is one observation, and it cannot show a slow decline running underneath the swings.
Mike O'Sullivan of the University of Exeter, a member of the team, drew the same line. He described the El Nino as a short-term shock, said the sink is recovering the year after and that there is strong evidence of that happening, and said this does not mean the land will maintain long-term resilience to climate change [15].
The Amazon, one of the two regions Ciais credits for the recovery [6], is among the sinks the El Nino now under way is expected to hit with extreme heat and drought [7]. If uptake falls and the next recovery reaches 2025's level, the weather reading gains a second data point. If the recovery falls short, that counts toward the erosion O'Sullivan describes.
What to watch
- Whether the 2025 land sink estimate gets an uncertainty range and holds up when the analysis is finalised.
- How far land uptake falls during the El Nino now under way, and in the Amazon in particular.
- Whether the recovery after that slump reaches the 2025 level or falls short of it.
Clarity's read
What the record supports and how the coverage leans. The claims behind it follow.
Reality
- Evidence50
- Adoption
- Insufficient
- Hype gap−10
- Incentives
- Insufficient
- Confidence45
Claim ledger
Ranked by verification strength, evidence, and original report placement.
- [1]
Since 2023, the amount of carbon dioxide Earth's landscapes have been able to absorb has dropped dramatically in the face of extreme weather, resulting in a record jump in atmospheric CO2 concentrations in two successive years.
- [2]
Preliminary research reveals a strong rebound in the land carbon sink in 2025 and a corresponding decline in the growth rate of atmospheric CO2.
ReportedSupportedSource: New Scientist, reporting the analysis led by Philippe CiaisView cited source - [3]
Philippe Ciais said: "In 2023 and 2024, because of El Nino and the Amazon fires, we had the almost cancellation of the global land sink."
ReportedSupportedSource: Philippe Ciais, Climate and Environment Sciences Laboratory, France, who led the analysisView cited source - [4]
Philippe Ciais said: "This cancellation led to a very high CO2 growth rate, because if CO2 is not taken by land, it stays in the atmosphere."
ReportedSupportedSource: Philippe Ciais, Climate and Environment Sciences Laboratory, FranceView cited source - [5]
Ciais and his colleagues estimate that the land sink absorbed 2.36 gigatonnes of carbon in 2025, which is 0.71 gigatonnes more than the 2015-2022 average.
- [6]
The recovery was driven largely by a dramatic return of growth in tropical forests; Ciais said the climate became more normal, especially in Africa and the Amazon, so plants recovered their carbon uptake capacity.
- [7]
The super El Nino already under way in the Pacific is expected to become the most intense in history later this year, bringing extreme heat and drought to crucial terrestrial sinks such as the Amazon rainforest.
- [8]
Another sharp slump in carbon uptake is expected next year.
- [9]
Scientists expect the world's land biosphere to shift from being a carbon sink to a carbon source within the next several decades unless human-driven emissions taper off.
- [10]
The sudden disappearance of the land sink in 2023 and 2024 had researchers worried that the sink was collapsing decades earlier than expected.
- [11]
Terrestrial ecosystems such as forests and grasslands take up CO2 from the atmosphere as they grow, and emit carbon when they decompose or burn.
- [12]
The implied 2015-2022 average land sink is 1.65 gigatonnes of carbon.
- [13]
The 2025 land sink is about 43% higher than the implied 2015-2022 average.
- [14]
The land sink is already 25 per cent smaller than it would have been without the effects of climate change and natural variability, researchers estimate.
- [15]
Mike O'Sullivan said: "There is a short-term shock, which is the El Nino, and it is recovering the year after. There is strong evidence of that happening. What it doesn't mean is that the land is going to maintain long-term resilience to climate change."
- [16]
O'Sullivan warns that a cycle of repeated weather shocks, fuelled by climate change, may be weakening the global land sink's long-term ability to store carbon.
Sources
1 independent publisher whose own reporting we read for this story.
- newscientist.comThe land’s ability to store carbon has bounced back after two bad years
1 article · October 13, 2026
Topics and entities
Follow any of these and your For You feed starts watching them — no settings page required.
Entities
- Philippe CiaisFollow
- Mike O'SullivanFollow
- New ScientistFollow