The world needs to reach net-zero emissions — a balancing point where the amount of greenhouse gases emitted into the atmosphere is equal to the amount removed — by around 2050 to limit warming to 1.5 degrees C (2.7 degrees F) and avoid rapidly worsening climate impacts. More than 100 countries, covering over 80% of global greenhouse gas emissions, have already set have a net-zero target.
Actions to cut emissions, like scaling up clean energy or switching to electric vehicles, can get us most of the way to net zero. But they won’t get us all the way. Some emissions will very likely be leftover by midcentury, whether because the technology to eliminate them doesn’t exist, isn’t widely available or can’t be deployed quickly enough.
These remaining or “residual” emissions will need to be balanced out with an equal amount of carbon dioxide removal — the “net” part of net zero.
While the concept of residual emissions seems simple, it raises important questions: Where would these emissions likely come from? Will carbon removal — still a largely nascent set of technologies and approaches — be available at a sufficient scale to address them? And how should countries be considering residual emissions and carbon removal in their climate plans?
1) What Are ‘Residual’ Emissions?
Residual emissions are generally defined as those that are still being released into the atmosphere when net zero is reached. They are what remain after efforts to reduce and avoid emissions have been taken. To achieve net zero, these residual emissions will have to be counterbalanced by carbon dioxide removal (CDR).
It’s likely that most residual emissions will come from sectors or sources that are considered “hard to abate,” meaning their emissions are difficult to reduce for economic or technological reasons. While there are different definitions, hard-to-abate emissions often include those from heavy industry sectors like cement and steel, where low-carbon solutions are not readily available, costly or still in the early stages of development. They can also refer to a broader set of emission sources, including agriculture, aviation, shipping and some industrial processes.
Despite this overlap, “residual” and “hard-to-abate” emissions are not the same thing. For instance, what is residual at net zero may include emissions that could have been avoided or reduced.
Ultimately, what is considered hard to abate can be subjective and political — since industries or countries could claim certain emissions are hard to abate as an excuse to avoid addressing them — while residual emissions are objectively whatever is left over at net zero. (Though residuals will also be the result of political and economic decisions about both emissions reductions and carbon removal deployment.)
2) How Do Residual Emissions Relate to Carbon Removal?
To achieve net zero, the world will need enough carbon dioxide removal to counterbalance any residual emissions, such that the sum of residual emissions and removals is zero. This carbon removal can come from both nature-based approaches (like planting trees) and novel, technological approaches (like direct air capture).
Technically, net zero could be achieved with any balance of emissions reductions and removals (as seen in the illustration above). However, there is a big difference between a future where emissions remain high and we rely heavily on carbon removals, and one where emissions and removals are both low. Pathways that maximize emission reductions and minimize the future need for CDR are the most likely to minimize climate hazards. This is because reductions and removals are not equivalent in their environmental, economic or social impacts.
Steep, near-term emissions reductions can limit peak warming and help minimize the potential for crossing irreversible climate tipping points. But pathways that allow for higher levels of removal in place of steep emissions reductions risk allowing higher peak temperatures, which could lead to even more dangerous heat waves, fires, storms and floods.
3) How Much Residual Emissions Are Expected, and Where Would They Likely Come From?
Even the most ambitious climate scenarios (which limit warming to 1.5 degrees C with limited or no overshoot) project that about 5-10 gigatonnes of CO2 (GtCO2) per year of emissions will remain around midcentury and will need to be counterbalanced by CDR. That’s about equal to 1-2 times the United States’ current net GHG emissions.
How Are Residual Emissions Estimated?
Many global estimates of residual emissions at net zero come from integrated assessment model (IAM)-based scenarios, which show how various climate policy interventions may interact with societal structures and Earth systems processes to influence temperatures. While IAMs provide global top-down estimates, countries must also assess CDR needs in their own climate plans. Bottom-up, sectoral assessments may provide more precises estimates, as they assess each sector individually — using estimates of abatement technology development — to project which emissions will be un-abatable by midcentury.
However, these estimates involve a lot of uncertainty. Many factors — including future technological and economic developments, cost reductions, product substitutions, demand shifts and behavior changes — will influence how much we can reduce emissions over the coming decades and therefore how much will be leftover, or residual, by midcentury.
Given these uncertainties, it’s challenging to pinpoint exactly where residual emissions will come from. Scenarios analyzed by the Intergovernmental Panel on Climate Change (IPCC) offer some estimates of residual emissions sources at midcentury, which are outlined below.
In pathways that limit warming to 1.5 degrees C, non-CO2 emissions such as methane (mainly from agriculture) make up the majority of residual emissions. Other residual emissions come from transport and industry, though the relative amounts vary significantly depending on the scenario.
| Sector | Activities that may generate residual emissions |
|---|---|
|
Agriculture |
Livestock management: Agricultural emissions from ruminant animal digestion (such as cows, sheep and goats) are expected to increase along with the global population and demand for animal products. While there are several ways to reduce livestock-related emissions, it will be technologically difficult to fully eliminate them. Fertilizer use: Nitrogen-based fertilizer creates nitrous oxide. Research shows that emissions from both production and use of fertilizer could be reduced by 80% by 2050, but the remaining emissions may need to be counterbalanced by carbon removal. Rice cultivation: Rice production can emit methane, as well as nitrous oxide with fertilizer application. More sustainable farming practices can help reduce emissions from rice but may not be enough to fully eliminate them. And shifting to sustainable rice farming methods can be challenging to incentivize. |
|
Transportation |
Aviation: There are no commercially competitive substitutes to liquid fossil fuels for aviation today. Reducing demand for air transport, improving energy efficiency and developing sustainable fuel alternatives can all help. However, more research is needed to determine what technologies are sustainable and cost-effective at scale. Shipping: There are currently no commercially competitive substitutes for the fossil fuels used in maritime shipping, either. And increasing international trade is leading to rising energy demand for shipping. |
|
Industry |
Cement and steel: Producing these materials requires high heat that is difficult to achieve without fossil fuel combustion. It also generates “process emissions” from the chemical reactions that happen during production. Cement and steel already face slim economic margins and are trade exposed, which makes it difficult to incentivize investments in novel and often costly low-carbon technologies. |
Sources: IPCC 2023, Buck et al 2023
Industrial and transport infrastructure — such as ships, planes or industrial production plants — typically lasts between 30 and 70 years. This means policies and incentives to advance innovation and deployment of lower carbon solutions are critical to avoid locking in high emissions processes and systems over the long term.
Other types of emissions may not be hard to abate in terms of technology development, availability or cost, but may be residual due to slow turnover rates. For example, electric vehicles are becoming more common. But there will very likely still be gas-powered cars on the road by 2050 due to the often decade-or-longer lifetime of an average vehicle. The emissions from these cars would therefore be residual.
4) How Are Countries Planning for Residual Emissions and Carbon Removal?
As of June 2025, more than 100 countries had set economy-wide net-zero emissions targets. Fifty-eight of them had begun laying out plans to achieve these targets in their long-term strategies (LTSs), which guide long-term low-emissions development. But just having a net-zero target doesn’t mean residual emissions are fully addressed. Not all countries with net-zero targets in their long-term strategies quantify expected residual emissions, and among those that do, they don’t all indicate the sources.
Across the countries that estimate residual emissions in their long-term strategies, agriculture accounts for the largest proportion of expected residual emissions: 35% on average. Following agriculture, the energy sector makes up around 20% of residual emissions, mostly from the use of fossil fuels in homes for heating and cooking. Other sources mentioned are industry, as well as small amounts of residual emissions from transport and waste.
Many countries also articulate plans to use conventional CDR approaches, like tree restoration, in their long-term strategies. And 26 mention plans to use or consider the use of novel, technological carbon removal approaches like direct air capture.
Countries also take different approaches in considering how to estimate and address their residual emissions. Canada, for example, presents a scenario with lower emissions reductions and high residual emissions from continued fossil fuel production and consumption, which it would counterbalance with a high level of novel CDR. Sweden, on the other hand, explicitly aims to minimize residual emissions as much as possible to avoid over-reliance on CDR.
5) Why Is Overreliance on Carbon Removal Risky?
Scenarios that rely heavily on CDR to enable the continued use of fossil fuels, as in Canada’s long-term strategy, underscore a common concern: that CDR will be used to compensate for emissions which could otherwise be abated.
There are several risks associated with this kind of planning. Carbon removal is not yet scaling quickly enough to meet the anticipated need at midcentury. And relying on CDR to counterbalance emissions that could otherwise be reduced means that less of this limited capacity — which faces real sustainability limits — would be available to balance out emissions for which there aren’t immediate low-carbon alternatives.
Most CDR technology is still in development.
Countries that estimate large amounts of residual emissions, including emissions that could otherwise be abated, are relying on the promise of a massive CDR scale-up. However, novel carbon removal technologies, like DAC and carbon mineralization, are still relatively early in development and have not yet been deployed at a large scale. It remains uncertain whether and how fast these technologies will be able to scale globally — so counting on them to be available at high levels can be risky.
The potential for large-scale CDR deployment therefore cannot be viewed as an excuse to delay emissions reductions. Countries must plan to slash emissions as deeply as possible, so that residual emissions at net zero are minimized and CDR is only being used where necessary. If carbon removal is not able to scale to the extent needed, countries risk missing not only their CDR goals, but also their broader long-term climate targets.
There are limits to sustainably scaling CDR.
While estimates vary, a growing body of literature suggests that there are limits to the amount of carbon removal that can be deployed sustainably, both in terms of resource use and social impacts.
With constraints on necessary resources like land, water and energy, these studies find that the quantity of CDR that will be available without compromising other sustainability goals is well below what IPCC scenarios show will be technically feasible by midcentury. There are also concerns around the societal impacts of this natural resource usage; for instance, how it could affect food production, land rights, or the livelihoods of rural communities and Indigenous peoples.
As more CDR is used to address residual emissions (including those that could have been abated) to reach net-zero, less will be available to reduce atmospheric CO2 and achieve net-negative emissions — which will be needed to help bring global temperatures back down to lower levels.
6) What Does This Mean for Countries’ Climate Plans?
We know some level of carbon removal will be needed to address residual emissions and help reach net zero. But with CDR techniques still largely nascent and their scalability not guaranteed, it’s highly uncertain how much removal we can count on.
As such, CDR should be planned for as a limited resource on the path to net zero, used to balance out only the most difficult to abate emissions sources. This is particularly true for high-income countries that have more capacity to invest in emissions-reduction interventions.
Countries should pursue several measures in their net-zero planning to ensure that emissions are reduced deeply and rapidly, and that carbon removal is planned for responsibly:
Transparently communicate how residual emissions are being estimated.
How countries identify and estimate residual emissions must be transparent. Using bottom up sectoral assessments may provide a more precise quantification of residual emissions than top-down estimates. This approach could enable countries to transparently communicate what emissions CDR would counterbalance by sector.
Transparently communicating estimated residual emissions — and how those estimates were derived — can help ensure carbon removal is being planned for responsibly and is not delaying emissions reductions. It can also help countries share best practices and develop a more consistent approach to addressing residual emissions through carbon removal.
Establish and clearly communicate separate targets.
Within their overarching climate targets, countries should set separate targets for emissions reductions and removals. This can provide transparency around the relative levels of each. It is the first step to ensuring that countries avoid over-relying on future carbon removal to reach net zero. Some early adopters, such as Sweden, have already set separate targets.
Countries can also voluntarily include information about their estimated residual emissions and plans for CDR in their long-term strategies and nationally determined contributions (NDCs), due in 2025 and widely expected before the UN climate summit in November. NDCs can be used to set intermediate targets not just for CDR, but also around goals for capacity building and support, like public investment in research and development.
Looking Ahead
Countries need to start planning for how to meet their net-zero goals today. These plans must aim to maximize emissions reductions and thereby minimize the level of residual emissions — and CDR needed — to reach net-zero. They should plan for CDR responsibly and transparently, being clear about the expected sources of residual emissions and ensuring they are only relying on CDR to counterbalance those that are most difficult to abate.
The less CDR we use to enable continued use of fossil fuels and other avoidable emissions, the more will be available to enable net zero and ultimately net negative emissions — keeping a safer future within reach.
