- Climate
Agriculture emits greenhouse gases, but less than using fossil fuels, despite recent paper’s claim
Scientific evidence shows that using fossil fuels is the main source of greenhouse gas emissions – but a controversial paper recently claimed otherwise.
On 11 March 2025, a new paper published in Environmental Research Letters (ERL) titled “Increased transparency in accounting conventions could benefit climate policy” suggested that ‘it is agriculture, not fossil fuels, that are the main emission source’. This idea was then shared in several social media posts and articles online, portraying the paper as a big breakthrough in our understanding of climate change. But is that really the case?
To answer this question, we spoke to several climate scientists with relevant expertise, including a lead contributor to the Global Carbon Project – a global project with an international network of scientists who are developing a more complete picture of Earth’s carbon cycle, including where carbon dioxide (CO2) is emitted and stored.
Main Takeaways:
- Greenhouse gas accounting is the process of determining the amounts of greenhouse gases being emitted to the atmosphere and stored in our land and oceans
- The balance between emissions and storage dictates how much greenhouse gas remains in our atmosphere and warms the planet
- Fossil fuels are the leading source of greenhouse gas emissions based on widely-accepted greenhouse gas accounting methodology used by the most reputable scientific organizations such as the Intergovernmental Panel on Climate Change
- Land use/agriculture emits CO2 through deforestation, but also ‘sinks’ or stores CO2 through reforestation – well-accepted methodology accounts for both
- A recent paper used unusual accounting methods by looking only at gross emissions and omitting CO2 sinks (storage) through reforestation; we interviewed several climate scientists who explained that this method is misleading
- Looking at greenhouse gas emissions but not sinks is like evaluating the status of your bank account based only on your income without looking at your expenses
Using fossil fuels emits more greenhouse gases than any other source
Before diving into the details of why the use of fossil fuels is widely accepted as the main source of greenhouse gas emissions[1], it’s important to note that agriculture and other land-use changes (like deforestation) also play a major role. It should not be seen as an ‘either or’ scenario of which influences climate change – they both do, just to different degrees.
Science Feedback contacted Dr. Piers Forster*, Professor of Physical Climate Change and founding Director of the Priestley Centre for Climate Futures at the University of Leeds, who explained:
Piers Forster
Professor, University of Leeds
“It is known that historically both agriculture and fossil fuel emissions have driven climate change. Both are important. For CO2 emissions agricultural contributions are much smaller than fossil fuel emissions. For methane emissions they are roughly equal. This [ERL] paper doesn’t really add to what is already known.”
* Note: Forster, is a trustee of the United Bank of Carbon – an environmental charity affiliated with the University of Leeds that conducts research and protects, plants, and restores trees and forests.
So what exactly do scientists know? As we’ve covered in past articles, greenhouse gases – such as CO2 and methane – warm our planet by trapping excess heat that would otherwise escape to space.
Because recent global warming has been driven by rises in CO2, scientists carefully track where it is being emitted and in what quantities to understand climate change drivers. This tracking is known as ‘greenhouse gas accounting’ (or ‘carbon accounting’ when looking at CO2 specifically).
Greenhouse gas accounting can be thought of like the accounting of money. Just as you have income and expenses – or money coming in and out – Earth’s atmosphere gains and loses greenhouse gases through ‘emissions’ and ‘sinks’.
Emissions sources – like the burning of fossil fuels – add greenhouse gases to our atmosphere, while sinks remove them (by absorbing them into our oceans, storing them in plants and trees through photosynthesis, etc.). The balance between emissions and sinks determines how much greenhouse gas actually builds up in the atmosphere over time.
To determine this balance, climate scientists carefully account for the different sources and sizes of emissions and sinks on Earth. Every year (since 2006) the Global Carbon Project (mentioned earlier) publishes a report called the Global Carbon Budget (GCB) – which tracks the emissions and sinks of carbon on our planet to account for the ongoing rise in atmospheric CO2 (Figure 1).
In Figure 1, note that fossil fuel emissions (large upward-facing gray arrow) are roughly 9 times greater than emissions from agriculture and other land use* (upward-facing orange arrow).
* Note: Forests can grow or shrink without human intervention, but in our article, deforestation and reforestation are referring to direct human intervention (like cutting down or regrowing forests).
Given that the GCB is one of the most robust resources on carbon emissions and sinks, we interviewed the Global Carbon Budget Office director and the lead author of GCB reports, Dr. Pierre Friedlingstein, to ask about the findings in the recent ERL paper.
Overall, Friedlingstein told Science Feedback that when proper accounting methods are used, scientists find that fossil fuels are the main source of greenhouse gas emissions. This is explained in GCB reports and also a conclusion of the most recent report from the Intergovernmental Panel on Climate Change (IPCC)[1], the world’s leading authority on climate change information. So why does the recent ERL paper deviate from this conclusion?
Friedlingstein pointed out that there is a major discrepancy between the widely-accepted methodology for carbon accounting and that which is used in the recent ERL paper. The ERL paper advocates for using gross emissions instead of net emissions – or looking only at what is emitted, not what is stored.
Going back to the money analogy, Friedlingstein explains that this approach is problematic because it is like evaluating the status of your bank account based only on your income without looking at your expenses. Looking only at gross emissions of each sector ignores a key difference between sectors: some sectors almost entirely emit greenhouse gases, while others emit and store them.
Land-use change involves both deforestation (which reduces its carbon sink) and reforestation (which increases it) – looking only at gross emissions ignores the CO2 that’s removed. But fossil fuels are quite straightforward: they are responsible for emissions, but almost no sinks.
As Friedlingstein explains:
Pierre Friedlingstein
Professor, University of Exeter
“He [Gerard Wedderburn-Bisshop] makes a claim in this paper and in a previous paper as well, that we – as the Global Carbon Budget, IPCC, essentially everyone – are making a major mistake because when we talk about the fossil fuel emissions, we just account for the gross fluxes of what is going out [being emitted]. But when we account for land-use change/agriculture, we account for net flux, which is the difference between what is going out [being emitted] and what is going in [being stored]. And he is saying this is not correct – that instead, we should treat things as ‘like to like’ and look at gross fluxes only. This might sound sensible, but it is complete nonsense because fossil fuels are essentially a one way flux [burning them emits greenhouse gases but doesn’t store them]. When we burn fossil fuels, coal, or gas, we release CO2 to the atmosphere.”
Although Wedderburn-Bisshop claims that fossil fuels are ‘unfairly’ looked at as a gross emission source, this is not the case. Interestingly, cement emissions are often included with fossil fuel emissions – due to the fossil fuels used for heating to produce it – and Friedlingstein notes that solidified cement can absorb some CO2 through something called cement carbonation.
So in a sense, cement carbonation is a ‘sink’ for the fossil fuel sector, and it is indeed accounted for as such by climate scientists. But Friedlingstein explained to Science Feedback that “this amount is very small relative to global emissions from fossil fuels and is basically negligible” – especially compared to the ratio of emissions and sinks of agriculture and other land use.
For example, in 2023, cement only took in roughly 0.21 billion tonnes of carbon*, while the fossil fuel sector emitted 10.31 billion tonnes of carbon[2]. This is only 2% of the total emissions. In contrast, for land use, Friedlingstein explains:
Pierre Friedlingstein
Professor, University of Exeter
“the net is about half of the gross, because we deforest about two billion tons of carbon and we reforest about 1 billion tons of carbon. Therefore, for land use, the net is 1 billion tons of carbon which is emitted.”
* Note: 1 billion tonnes of carbon = 3.664 billion tonnes of CO2
When scientists calculate net emissions, it’s based on scientific reasoning, not ‘fairness’
In an earlier paper, Wedderburn-Bisshop (author of the recent ERL paper) laid the groundwork of his arguments for gross accounting across all sectors. Wedderburn-Bisshop’s main argument is that reforestation allows for uptake of CO2 emitted from both fossil fuels and deforestation.
In other words, he argues that newly grown forests do not choose which CO2 to uptake based on its original emission source, and thus this uptake should not be ‘subtracted’ only from land use/agriculture emissions.
However, Friedlingstein pointed out that this reasoning is flawed: given that both deforestation and reforestation are directly tied to human interventions, it makes sense to account for them together. Humans cause deforestation – emitting CO2 and reducing its uptake – but then also reforest around the world, increasing CO2 uptake. Friedlingstein explains reforestation must be accounted for in the total balance otherwise something would be missing.
It may sound ‘fair’ that, because forests take in CO2 from all sources, these sinks should be excluded from total land use emissions. However, this neglects the fact that climate scientists are specifically accounting for direct human intervention with these forests (through deforestation and reforestation) when measuring land-use emissions.
Natural carbon sinks from forests – those that occur in forests we aren’t directly altering – are not lumped in with land use emission accounting, but instead accounted for separately. In Figure 2, you can see how these emissions have changed over time; note that net land-use change emissions (orange area) have remained relatively stable while fossil fuel emissions (gray area) steadily rose until 1950, then saw a sharp rise.
Science Feedback also contacted Dr. Drew Shindell, Distinguished Professor of Earth Science at Duke University, who explained:
Drew Shindell
Distinguished Professor of Earth Science, Duke University
“I had a quick look at the paper and in my opinion it is patently unsound. With fossil sources of carbon dioxide, there is only output and no input on societal timescales (geologic inputs take millions of years) and so using net or gross emissions is the same. Using gross emissions for agriculture is not sensible as there are large fluxes in both directions […] the gross flux from agriculture is largely carbon that recently came out of the atmosphere and so is just cycling back and forth into and out of plants, just like deciduous trees do every year when they grow and then shed leaves. It has been understood by researchers for decades that only a change in the net flux could drive climate change.”
Among the accounting methods used in the recent ERL paper, the biggest shift came from the switching from the well-accepted net accounting to the controversial method of gross accounting. In the paper, this increased agriculture’s contribution by 10%, while lowering that of fossil fuels by 11%, which – based on the paper’s methods – ‘tips the scale’ toward agriculture as the main contributor.
However, the ERL paper also used two other methods that further increased agriculture’s share of total global emissions – something called effective radiative forcing (ERF) and adding in fossil fuel’s ‘cooling’ emissions – as we will explain below.
The climate impacts of emissions depends on metrics and timescale – it’s important to specify both
The climate effects of greenhouse gases mainly depend on two things: how effective a greenhouse gas is at warming our planet and how long the gas lasts in our atmosphere[1]. Metrics play a key role in capturing this information for greenhouse gas accounting.
There are different metrics used to explain the warming impacts of different greenhouse gases in our atmosphere. The standard metric for greenhouse gas accounting is GWP-100, which measures the warming impact of different greenhouse gases relative to CO2 over a 100 year time span. The ERL paper, however, advocates for the use of effective radiative forcing (ERF), a different metric which tells a different story. As Forster explains:
Piers Forster
Professor, University of Leeds
“[the ERL paper] uses effective radiative forcing which is a snapshot in time and mixes drivers of climate change operating on different timescales. If you are comparing the climate impact of different emissions, you need to specify the metric you care about and a time period. This study uses ERF as the metric which is fine but fails on the time horizon point. The case in point is that if coal fired power stations cleaned up their SO2 [aerosol] pollution tomorrow (which is happening), the attribution of ERF to fossil fuel would jump to a much larger number, even with their method choice.”
The ERL paper points out that burning fossil fuels not only emits planet-warming greenhouse gases but also aerosols which cool the planet. While this is true, what Forster points out above is that aerosols are extremely short-lived in our atmosphere compared to CO2. So ignoring the ‘time’ component does not capture how the warming would play out as emissions change – it only represents a snapshot in time.
The IPCC explains that atmospheric aerosols usually last one day to two weeks in the lowest layer of our atmosphere (the troposphere), and about one year in the second lowest layer (the stratosphere)[3]. However, CO2 can last for centuries or longer[4].
So, although it is true that burning fossil fuels emits cooling aerosols that partially offset greenhouse gas warming – this is already well-known and reported by the IPCC[1] – it’s important to keep in mind that these cooling effects rapidly diminish when their emissions cease. In fact, this trend has already started.
As explained in a 2024 Nature Communications Earth & Environment paper, recent reductions in aerosols have increased Earth’s energy imbalance, which drives further warming[5]. The authors explain that accelerated global warming is expected in the next decade as a result of further reductions in aerosol emissions due to air quality legislation[5].
As explained to Science Feedback by Dr. Jyoti Singh, Postdoctoral Scientist at the Center for Climate Systems Research (CCSR) at Columbia University:
Jyoti Singh
Postdoctoral Scientist at the Center for Climate Systems Research (CCSR), Columbia University
“Aerosols like sulfates may have provided a temporary cooling effect, but they are also harmful to human health and are now declining due to stronger air quality regulations […] That makes the cooling effect a short-term, unstable factor, which could create ambiguity if used to offset fossil fuel emissions in long-term accounting.”
The ERL paper passed peer review, but climate scientists are skeptical about it
Overall, the recent ERL paper did not use well-accepted methodologies and two of the climate scientists we spoke to about this paper were even surprised that it passed the peer-review process. (This emphasizes the importance of not relying on a single paper to support broad scientific conclusions.)
Furthermore, while it is not always problematic to publish as a single author, both papers published by Gerard Wedderburn-Bisshop on this subject did not include collaborators. As explained to Science Feedback by Dr. Anna Sörensson, Senior Researcher at the University of Buenos Aires:
Anna Sörensson
Senior researcher , University of Buenos Aires
“Overall, the author seems to be almost only publishing as a single author, and for these kinds of complex issues it would be much better to collaborate with a diverse and international research community. Acting solo is, to me, diminishing the importance and credibility of this paper.”
Additionally, as Wedderburn-Bisshop explains in his paper: “The author is biased against industrial scale deforestation, witnessed while monitoring deforestation for the Queensland government”. Wedderburn-Bisshop is also the executive director of the World Preservation Foundation, an organization with an objective to spread information that ‘animal agriculture is the leading cause of climate change’ – another potential bias toward the decisions made for his recent paper.
That being said, the author is bringing attention to the emissions coming from land use and agriculture, which are an important part of the conversation on greenhouse gases, and hold a high potential for emission reductions[6].
Scientists’ Feedback
Questions from Science Feedback:
1. Is the methodology used in the paper a well-accepted way of determining the global warming contributions of different greenhouse gas emission sources? If not, what issues/uncertainties do you find in that methodology?
2. What does leading scientific evidence show is the primary emission source that is driving recent warming? Is there generally a scientific consensus on that, and if so, does the recent Environmental Research Letters paper challenge it in a meaningful way?
3. People online are using this paper to claim that animal agriculture is a bigger contributor to climate change than fossil fuels. Do you think that conclusion can be supported by this recent paper alone, or is more evidence needed?
Piers Forster
Professor, University of Leeds
1. “No, it uses effective radiative forcing which is a snapshot in time and mixes drivers of climate change operating on different timescales. If you are comparing the climate impact of different emissions, you need to specify the metric you care about and a time period. This study uses ERF as the metric which is fine but fails on the time horizon point.
The case in point is that if coal fired power stations cleaned up their SO2 pollution tomorrow, (which is happening), the attribution of ERF to fossil fuel would jump to a much larger number, even with their method choice.
I have issues with the numbers anyway – our 2024 indicators paper shows that around 70% of CO2 emissions are fossil fuel related. This assumes much smaller numbers
2. Yes, sustained high levels of greenhouse gas emissions from fossil fuel and cement production combined with cleaning up of air pollution. See section 7.2 of our indicator paper[7] from last year.
3. It is known that historically both agriculture and fossil fuel emissions have driven climate change. Both are important. For CO2 emissions agricultural contributions are much smaller than fossil fuel emissions. For methane emissions they are roughly equal. This paper doesn’t really add to what is already known.”
Pierre Friedlingstein
Professor, University of Exeter
Essentially he’s [Gerard Wedderburn-Bisshop] doing two major changes to the normal way that we determine the contribution of different sectors to climate change. And he comes to the conclusion that land use/agriculture is the dominant driver of climate change and that fossil fuel is negligible, essentially, which is indeed highly controversial.
And the way he [Gerard Wedderburn-Bisshop] does it includes one element which is totally wrong, and another element which is right. So I can start with what is wrong: he makes a claim in this paper in a previous paper as well, that we – as the Global Carbon Budget, IPCC, essentially everyone – are making a major mistake because when we talk about the fossil fuel emissions, we just account for the gross fluxes of what is going out [being emitted]. But when we account for land-use change, we account for net flux, which is the difference between what is going out [being emitted] and what is going in [being stored]. And he is saying this is not correct – that instead, we should treat things as ‘like to like’ and look at gross fluxes only. This might sound sensible, but it is complete nonsense because fossil fuels are essentially a one way flux [they emit greenhouse gases but don’t store them]. When we burn fossil fuels, coal, or gas, we release CO2 to the atmosphere. When we make cement we also emit CO2. So there’s essentially a one way flux that goes into the atmosphere. There’s only a tiny flux that goes in, which if you look at the global carbon budget papers, we describe this flux, it’s called cement carbonation, but it’s only at a small scale.
So if you have concrete in buildings, they absorb CO2 but it’s at a small scale. We emit 10 billion tons of carbon from fossil fuel burning, but cement carbonation – which is the uptake of CO2 in concrete buildings – is only 0.2 billion tons of carbon. And we account for this in the carbon budgets because although it’s small, it’s not zero. So in a sense it is a net flux for fossil fuel emissions as well but the difference between the emissions and uptake is essentially negligible.
For land use change, it’s much more important because it goes both ways. I mean, in many places, like in the tropics, we deforest, and deforestation leads to CO2 emissions to the atmosphere. In many other places, we reforest. We reforest in Europe, China, and we’ve reforested in the U.S. for decades. And if you look at the net flux, this is a difference between the place where you have deforestation and the place where you have reforestation. And you have to take both into account. Otherwise, you’re just wrong. So in a sense, it is the net flux that matters in more ways. It’s like your bank accounts, what matters is net between income and your expenses.
If you just calculate the expenses in your bank accounts, it doesn’t give you any information on what’s happening in your bank account. It’s only half the story. It’s exactly the same for the atmosphere. It’s always the net that matters. It’s the net for fossil fuels, but the net is almost equal to the gross. It’s the net for land use too, but the net is about half of the gross, because we deforest about two billion tons of carbon and we reforest about 1 billion tons of carbon. Therefore for land use, the net is 1 billion tons of carbons, which is emitted. So, if you do the proper comparison of net emissions, for CO2 it is 10 billion tons of carbon, and for land use is one billion tons of carbon. And this is correct; this is what we report in the GCB (global carbon budget) and what is reported by the IPCC.
The well-accepted methodology for accounting is that presented in the Global Carbon Budget. In the last paper [Global Carbon Budget 2024] we had more than 100 contributors and it’s something that’s reviewed by the entire community. It’s an open review and then it goes to IPCC, and the IPCC then re-assesses the understanding of the carbon budget using the Global Carbon Budget estimate, but also using all other estimates that will have been published over the last seven years. And this is reviewed by the community three times and then it’s presented to the government and is reviewed again and then it’s finally accepted. So, yes, the Global Carbon Budget methodology of using net flux is the standard view.
There is one argument in the [ERL] paper which is right – sulfate aerosols are associated with fossil fuel burning, because when you burn coal (other gases, too, but mainly coal) you release sulfate. They are cooling the planet because they are tiny particles [which reflect sunlight]. So they partly counterbalance warming – this part of his paper is correct. But the cooling effect of aerosols is extremely temporary. If we stopped emitting aerosols, the cooling effect would be gone in a week. And the cooling from aerosols is like half a degree. So you are cooling the planet by half a degree, but you are warming the planet by 1.5 because of CO2 and methane and therefore the net is about 1 degree C of warming.
Anna Sörensson
Senior researcher , University of Buenos Aires
I agree that the abstract of the article can be interpreted as “don’t blame fossil, blame agriculture” which is also the messages I found on X (very few publications fortunately).
My personal opinion is that it is good to debate and question existing methodologies. The problem is the policy implications that Gerard Wedderburn-Bisshop refers to. Although he does have a section on the damage of aerosols, this gets lost in the abstract and conclusions which is what most people read. Overall, the author seems to be almost only publishing as a single author, and for these kinds of complex issues it would be much better to collaborate with a diverse and international research community. Acting solo is, to me, diminishing the importance and credibility of this paper.
Drew Shindell
Distinguished Professor of Earth Science, Duke University
I had a quick look at the paper and in my opinion it is patently unsound. With fossil sources of carbon dioxide, there is only output and no input on societal timescales (geologic inputs take millions of years) and so using net or gross emissions is the same. Using gross emissions for agriculture is not sensible as there are large fluxes in both directions. Humans aren’t an important source of CO2 when we exhale, for example, since the carbon we’re breathing out comes from our food which took that same carbon out of the atmosphere recently. If you just looked at our ‘gross emissions’ we might seem like a substantial source, but that’d be wrong as the net emissions are near zero. Similarly, the gross flux from agriculture is largely carbon that recently came out of the atmosphere and so is just cycling back and forth into and out of plants, just like deciduous trees do every year when they grow and then shed leaves. It has been understood by researchers for decades that only a change in the net flux could drive climate change. This paper follows a long tradition of climate science deniers misusing selected values within the carbon cycle to try to downplay the role of fossil fuels. A good example is former Exxon CEO Lee Raymond, who in a 1997 speech at the World Petroleum Forum, said “Only four percent of the carbon dioxide entering the atmosphere is due to human activities – 96 percent comes from nature. Leaping to radically cut this tiny sliver of the greenhouse pie on the premise it will affect climate defies common sense and lacks foundation in our current understanding of the climate system.” He’s making a version of the same argument nearly 30 years ago as he compares fossil fuel-related CO2 emissions to the gross flux from all natural systems rather than the net flux (approximately zero) to try to mask the fact that the source of his company’s profits is driving climate change. It’s hard to see how this new paper passed peer review at ERL to be honest, as it’s so deeply flawed.
Jyoti Singh
Postdoctoral Scientist at the Center for Climate Systems Research (CCSR), Columbia University
1. Not exactly. The paper does not use conventional IPCC methods, primarily by using gross rather than net emissions for land-use change and assigning cooling effects from fossil fuel-related aerosols as “credits” to the fossil fuel sector. While this approach isn’t widely adopted, I do think it brings attention to important aspects of climate accounting that are often underrepresented.
That said, I will personally be cautious about assigning cooling “credits” to fossil fuel emissions. Aerosols like sulfates may have provided a temporary cooling effect, but they are also harmful to human health and are now declining due to stronger air quality regulations. As noted in Dr. James Hansen’s recent paper, this decline is now contributing to accelerated warming. That makes the cooling effect a short-term, unstable factor, which could create ambiguity if used to offset fossil fuel emissions in long-term accounting.
So while I appreciate the paper’s effort to provoke new thinking, I think we need to treat the cooling component carefully, especially in light of current atmospheric and health trends.
2. There is a strong scientific consensus that fossil fuel emissions are the main driver of anthropogenic global warming, as repeatedly shown in IPCC reports and attribution studies. However, this paper does bring in a fresh perspective using effective radiative forcing (ERF) and alternative accounting frameworks that shift how we view sectoral contributions.
While it doesn’t directly refute the existing consensus, it does challenge us to think more broadly about how we define and distribute climate responsibility, especially between energy and land-based systems. In that sense, I think it meaningfully contributes to ongoing debates around emissions attribution and opens the door to more detailed interdisciplinary work.
3. This paper provides a compelling argument that agriculture, particularly land-based emissions, may have a larger role in historical climate forcing than often assumed. What’s particularly interesting is that even when cooling credits to fossil fuels are excluded, the paper still finds agriculture to be the largest contributor to radiative forcing. That’s a significant finding and one that, I believe, deserves further investigation.
I wouldn’t support using this paper alone to claim animal agriculture is the primary driver of climate change, I do think it strengthens the case for re-evaluating how we account for agricultural emissions and for investing more research into this space.
In summary, I see this paper as a bold and timely contribution. It’s shaped by both scientific reasoning and an advocacy-informed lens, and that’s okay, as long as we remain transparent about assumptions and open to further dialogue. Different communities (scientists, policymakers, and activists) will interpret these results differently, which is why we need more evidence, diverse perspectives, and consistent methodologies going forward.
References:
- 1 – IPCC (2021) Sixth Assessment Report.
- 2 – Friedlingstein et al. (2024) Global Carbon Budget 2024. Earth System Science Data.
- 3 – IPCC (2021) Fifth Assessment Report.
- 4 – IPCC (2021) Fourth Assessment Report.
- 5 – Hodnebrog et al. (2024) Recent reductions in aerosol emissions have increased Earth’s energy imbalance. Nature Communications Earth & Environment.
- 6 – Frank et al. (2018) Agricultural non-CO2 emission reduction potential in the context of the 1.5 °C target. Nature Climate Change.
- 7 – Forster et al. (2024) Indicators of Global Climate Change 2023: annual update of key indicators of the state of the climate system and human influence. Earth System Science Data.
UPDATES:
28 April 2025: This review was updated to correct the spelling of ‘Global Carbon Budget’.
