In-depth: Experts assess the feasibility of ‘negative emissions’

To limit climate change to “well below 2C”, as nationsagreed to do in Paris last December, modelling shows it is likely that removing carbon dioxide emissions from the atmosphere later on this century will be necessary.
Scientists have imagined a range of “negative emissions” technologies, or NETs, that could do just that, asexplained by Carbon Brief yesterday. But are any of them realistic in practice?

Carbon Brief reached out to a number of scientists, policy experts and campaigners who have studied both the necessity and feasibility of negative emissions.

The Paris Agreement calls for “holding the increase in the global average temperature to well below 2C above pre-industrial levels and to pursue efforts to limit the temperature increase to 1.5C above pre-industrial levels”. However, as the IPCC AR5 report showed, the majority of modelling to date assumes a significant global-scale deployment of negative emissions technologies in the second half of this century, if such temperature limits are to be achieved.

These are the responses we received, first as sample quotes, then, below, in full:

Ottmar Edenhofer, Intergovernmental Panel on Climate Change – “Whether we can achieve the deployment needed for the aims declared in Paris last year…depends very much on which negative emissions options will end up to be at our disposal ultimately.”
Detlef van Vuuren, PBL Netherlands Environmental Assessment Agency – “At the moment, BECCS and large-scale reforestation seem to be most promising based on a combination of technology-readiness, costs, and potential. At the same time, both options are not without challenges.”
David Keith, Harvard University – “I am skeptical that BECCS should be used beyond narrow niches…Instead, I would focus on accelerated weathering and air capture which could, in principle, be scaled to many gigatonnes per year with low land footprint.”
David MacKay, former chief DECC scientist – “Do I think it is a realistic view of what the world will do? No, not at the moment, because I think the Paris discussions completely ducked this issue, which is one of the most important issues out there.”
Pete Smith, University of Aberdeen – “For BECCS, there are significant issues with competition for land if it is implemented at the median rate projected by integrated assessment models, and water use is also significant.”
Oliver Geden, German Institute for International and Security Affairs – “A strategic debate about how to use carbon dioxide removal within a broader portfolio of climate policy measures is clearly lacking.”
Hannah Mowat, Fern – “If countries reduce emissions fast enough, then the level of CO2 that must be removed is entirely feasible, and, if done through [forest] restoration, can be positive.”
Rob Bailey, Chatham House – “It is clearly less risky not to emit a tonne of CO2 in the first place, than to emit one in expectation of being able to sequester it for an unknown period of time, at unknown cost, with unknown consequences, at an unknown date and place in the future.”
Joeri Rogelj, International Institute for Applied Systems Analysis – “Without CO2 emissions being penalised or strongly discouraged in some way, a large-scale deployment does never seem realistic.”

Stephan Singer, WWF – “A debate on the Paris objectives must not start with ‘negative emissions’, since this might be used to delay actions towards later decades.”
Sabine Fuss, Mercator Research Institute – “There is no champion to be singled out here and we will have to look for the portfolio of negative emissions technologies that minimises unwanted effects on non-climate policy goals.”
John Lanchbery, RSPB – “A large proportion, if not all of this, could probably be achieved by the conservation and enhancement of natural forests, peatlands and other natural sinks and reservoirs – without recourse to negative emissions technologies.”
Glen Peters, CICERO – “The best carbon dioxide removal strategy would be to use a mix of technologies, with each technology located to avoid its limitations.”
Noah Deich, Centre for Carbon Removal – “We must engage the diverse set of stakeholders that will develop and deploy carbon removal systems today to ensure we can meet tomorrow’s climate, economic, and social goals.”
Mike Childs, Friends of the Earth – “Critically, use of land always brings issues of land rights and justice. It would be a disaster to see poor communities thrown of their land for negative emissions — land grabs are well documented for biofuels.”
Florian Kraxner, International Institute for Applied Systems Analysis – “Biomass co-firing with existing retrofitted or newly built coal power plants seems to be an easy entry for the BECCS technology since it is an expensive technology and economy of scale is key.”
Sami Yassa, Natural Resources Defense Council –“There is no scientific basis for assuming that BECCS can deliver “negative emissions” after accounting for direct and indirect life-cycle emissions.”
Tim Lenton, University of Exeter – “We can start on this now, at low cost, and as part of already internationally signed-up-to targets under the Bonn Challenge and subsequent New York Declaration.”

In the IPCC’s 5th Assessment Report (AR5), negative emissions have mostly been achieved through a combination of bioenergy with carbon capture and storage (BECCS), where different technologies have been used by different models, ie not only biofuels. However, a handful of scenarios also included afforestation, ie an expansion of terrestrial carbon sinks, and since AR5, the community is working to integrate other technologies, such as direct air capture. While energy-intensive and still expensive, it does not compete for land with food production and could thus be part of a sustainable solution for low stabilisation targets (Smith et al. 2016).

Whether we can achieve the deployment needed for the aims declared in Paris last year thus depends very much on which negative emissions options will end up to be at our disposal ultimately. It is clear that the infrastructure needed for BECCS, in particular, is massive in many of the current low-stabilization pathways and that we are late in ramping this up. On average, these pathways require investments into BECCS of $138bn and $123bn per year for electricity and biofuel respectively in 2050 (Smith et al. 2016). However, this is not only related to technology, but also to unintended side effects and social acceptance. Based on the uncertainties, future costs and uncertain social acceptance of these technologies, it would be irresponsible to postpone immediate action on climate policy, and we would miss the short-term entry points into an effective climate policy. In order to provide the incentives for short-term action and for a massive decarbonization it will be necessary to price carbon – something that has also been acknowledged in the Paris Agreement. The carbon price would need to be increasing in order to achieve a dynamically efficient reduction of emissions, and should be combined with a transfer mechanism, which could also raise political feasibility (Edenhofer et al. 2015).

At the moment, BECCS and large-scale reforestation seem to be most promising based on a combination of technology-readiness, costs, and potential. At the same time, both options are not without challenges (eg Van Vuuren et al., 2013; Smith et al., 2015). These concern the large amount of land that might be needed, the unknown effectiveness of large scale storage, but also the societal concerns.

Models indicate that for meeting the 2C target, negative emission options are attractive, as they could smooth the short-term transition somewhat — although even pathways with negative emissions are below current “intended nationally determined contributions” (INDCs). The problem is, however, that large-scale deployment is only really needed in the longer term (several decades – partly because stabilising and possibly declining population would ease the land-use concerns a bit). At the same time, our current decisions depend on our assessment whether these technologies will be there. Counting on negative emissions, we could create a lock-in overshooting the budget first, which can only be solved by negative emissions. Emissions targets for 2050, for instance, are 40-60% with negative emissions globally, but 60-70% without(Van Vuuren et al., 2015).

I do find it hard to assess exactly how much potential there is for negative emissions in the long-term. I would assume that there will likely be some potential, but it would be attractive if we can make sure to rely on this as little as possible. In that context, one could try to design reduction strategies that are as much as possible still robust against not having large amounts of these technologies, and at the same time start an international research demonstration programme. Information from that programme could be used to re-adjust the strategy. In any case, in the short-term decisions on this will be needed.

The climate problem is mostly caused by humans moving carbon from the lithosphere to the biosphere at a rate that is in the order of 100 times faster than the natural rate at which carbon degases from the lithosphere. Once there, carbon can be repartitioned between land, ocean, and atmosphere on much faster timescales. Sharp thinking about carbon dioxide removal (CDR) begins with dividing it into two broad classes. Some CDR technologies, such as afforestation or the manipulation of soil carbon, are inherently short-term. They repartition carbon between atmosphere and land, building up carbon stocks that could be rapidly released in a warming world. Other CDR technologies, such as BECCS, air capture with CCS, or the addition of alkalinity to the oceans, which may be thought of as accelerated weathering, are inherently long-term. They (roughly) reverse the geological flux that is causing the problem.

Both may be useful, but it’s nonsense to compare them one-to-one. All else equal, a tonne of carbon removed by injecting it into a deep geological reservoir, or by adding alkalinity to the ocean, buys us more environmental protection than a tonne of carbon captured in a forest or in biochar mixed into soils. Both both deserve more attention and research, but it’s dumb policy to treat them equally.

Across the energy system I favour choices which have a low land footprint. Humans biggest impact on the natural world is (arguably) through land disturbance. We must be cautious of technologies that aim to remediate the carbon problem while greatly expanding our impact on the land. I am skeptical, therefore, that it makes environmental sense to use BECCS beyond narrow niches based around waste biomass. Instead, I would focus on accelerated weathering and air capture which could, in principle, be scaled to many gigatonnes per year with low land footprint. (Reader beware: I founded a company developing air capture.) Both these technologies are in their infancy and require much more research and development before we can meaningfully assess their cost and environmental impacts.

My views on this list [of which negative emissions technologies offer the most promise] haven’t changed much since I wrote the final chapter of Sustainable Energy Without the Hot Air:

I think all four of these deserve R&D. In terms of “promise”, I’d put (a) and (b) top of the list in that order; then (c) and (d) as equal third most promising.

On bioenergy with CCS, one new thing I have learned is this, from Dan Schrag at Harvard. He reckons that a promising way of proceeding with biomass is to use aFischer-Tropsch process to turn biomass into two things: liquid fuels, and pure CO2. Roughly half the carbon atoms turn into alkanes and half into CO2.

The other important caveat is the “genuinely sustainable” thing. As the MacKay-Stephenson BEAC report [DECC’s biomass emissions report] shows (and to the surprise of many including me), there are many bioenergy intensification options that actually create a very, very long lasting carbon debt; so, in many locations, the best results in carbon terms might be achieved by intensifying reforestation rather than by increasing biomass.

In all the IPCC working group 3 modelling, I think it is correct to say that they assumed that only (b) on this list was possible. The others were not included in the modelling. This could be grounds for optimism, since I think (a) may well be do-able at a reasonable carbon price.

A concern about the IPCC-WG3 modelling of BECCS, incidentally, is that I expect it assumes perfectly rational and well-informed behaviour. So, in the model, no-one would deforest an area to make a quick buck, because they would be aware of the loss of carbon stocks. Whereas, in reality, it is very difficult to measure carbon stocks in the landscape and, if there are subsidies for biomass without correct carbon stock measurement, it is quite possible that the subsidies would lead to biomass activities that have bad carbon effects in the landscape.

Well, I would say that [the scale of negative emissions technologies to meet the aims of the Paris Agreement] is technically deliverable, just about, but the way I always put it is this… The required scale of burial of CO2 by 2100 (measured as a mass buried per year) is, according to both back-of-envelope calculations and the IPCC WG3, about five times as great as today’s oil industry (measured in the same units as a mass extracted per year).

Is this technically deliverable? Yes, in principle, but only if many governments make clear that this is their intention, and agree a mechanism, for example, an agreement on a global carbon price, to get it delivered. Do I think it is a realistic view of what the world will do? No, not at the moment, because I think the Paris discussions completely ducked this issue, which is one of the most important issues out there.

None of the negative emission technologies (NETs) considered by Smith (2016) andSmith et al. (2016) can be implemented at scale without implications. For bioenergy with carbon capture and storage (BECCS), there are significant issues with competition for land if BECCS is implemented at the median rate projected by integrated assessment models, and water use is also significant. One advantage of BECCS relative to other NETs is that it produces rather than requires energy. Similar land and water constraints face afforestation/reforestation. For enhanced weathering of rocks that naturally absorb CO2, whilst the land areas required are vast, crushed rock could be spread on land without changing the land use, perhaps also providing benefits in terms of soil fertility (by raising the pH of acidic soils). The process is, however, currently costly and the mining and grinding of the rock is energy intensive. Direct air capture using chemicals is currently extremely costly and requires extremely high energy inputs, but it has a low land and water footprint. Soil carbon sequestration can be applied on land without changing land use, and provides a range of co-benefits. It is inexpensive, but the sinks created are finite in duration and reversible. Biochar can produce some energy, but the more biochar that is produced, the less energy is generated. The land and water footprint for spreading biochar are negligible, but the land and water footprint of the biomass used as a feedstock for biochar can be large, as for BECCS.

Further research is urgently required into NETs if we are to meet the challenging targets under the Paris Agreement. Among these research and development challenges is the need for more detailed research into the optimal use of land, including multi-functional land uses that could allow different (eg food and biomass) to be derived from the same land. Research also needs to examine how adverse environmental impacts (eg on water, biodiversity) associated with land-based NETs could be minimized, or even reversed. More R&D into alternative end fuels is also required, including the extraction of high value co-products from biomass before use as a bioenergy feedstock for BECCS. Implications of transporting feedstocks for BECCS or biochar over large distances also need to be better understood. For any technology involving CCS, more large-scale demonstration projects are required to demonstrate efficacy of carbon storage and to learn by doing – to allow costs to be reduced and efficiencies improved ahead of larger scale roll-out.

Understanding how NETs will impact the Earth system is also a key research challenge, for example, the extent to which reduced atmospheric CO2 concentrations could reduce the strength of the land and ocean sinks in the future.

Despite the increasing prominence of carbon dioxide removal (CDR) in emissions pathways compatible with 2 or 1.5C nobody can say today if we will ever see significant deployment – and if so, when exactly. Regarding economic and technological feasibility, afforestation and bioenergy with carbon capture and storage (BECCS) look promising today. In terms of social and political acceptability, non-terrestrial technologies like direct air capture (DAC) seem to offer the most promise, but they will need significant technological and cost improvements, depending on massive research and development efforts. Similar to mitigation technologies today, we will definitely see country specific priorities. Sweden and Finland, with their large pulp & paper industries, might prioritise BECCS, while countries like Saudi Arabia might opt for DAC.

When accounting for all dimensions of feasibility, including social and political, it’s hard to imagine that carbon removal on the order of 600-800GtCO2 – equaling 15-20 years of current annual emissions – can be realised during the 21st century. Based on terrestrial CDR only (like in today’s integrated assessment models) one would need approximately 500+ million hectares of additional land, that’s 1.5 times the size of India. That’s obviously a political no-go, and the main reason why negative emissions haven’t been part of high-level climate negotiations so far, despite the fact that carbon removal has been seriously discussed in the IPCC since 2007 and is an integral part ofRCP2.6, the IPCC scenario consistent with 2C. Until now, the introduction of CDR has mainly had the effect of covering political inaction. A strategic debate about how to use CDR within a broader portfolio of climate policy measures is clearly lacking. Most policymakers don’t even know the difference between net and gross negative emissions. For 2C, the world should cross the line into net zero around 2070, but the phase-in of carbon removal technologies will have to happen way before 2050.

To hold global temperature increase to well below 2C above pre-industrial levels, we need to keep forests standing and fossil fuels in the ground. Radical emissions reductions are the only way to limit the volume of total carbon removals needed to the levels that could be environmentally and socially sustainable. The only promising approach to achieving negative emissions is the restoration of terrestrial ecosystems, including accelerating the recovery of degraded forests. Such restoration has thepotential to achieve a maximum estimated amount of 330GtCO2 of removals by the end of the century.

Restoration of degraded natural ecosystems is not only possible today, but is an urgent intervention to meet multiple other environmental objectives, such as maintaining and enhancing biodiversity and halting desertification. These actions are also likely to be socially acceptable and effective if done with full consent and by rural communities and forest peoples. Evidence suggests that local people are the best guardians of forests and other ecosystems.

There are currently no technologies to remove CO2 from the atmosphere that can be employed at scale. It is very doubtful any will be available at scale within the timescale required. Furthermore, many of the proposed technologies are likely to have a dire social and environmental impact on food security, community land rights and biodiversity.

It is important also to note that land-based removals are not equivalent to ongoing emissions from fossil fuels. We cannot rely on land-based carbon removals and continue to emit fossil fuel greenhouse gases while remaining within our carbon budget, as these removals are reversible stores of carbon, prone to fire, disease or clearing. They should, therefore, be considered as a safety net, helping to sustain the basic needs of human life, build resilience and allow us to adapt to a changing climate. They should not be used to artificially increase the carbon budget.

The scale of deployment required of negative emissions depends entirely on how fast countries reduce emissions from fossil fuel use, and from forest and land clearing. The amount of CO2 that needs to be removed from the atmosphere to limit warming to below 2C is linked to the amount of CO2 that is put in the atmosphere. If countries reduce emissions fast enough, then the level of CO2 that must be removed is entirely feasible, and if done through restoration can be positive.

However, the level of ambition shown by countries in their Nationally Determined Contributions (NDCs) puts us on a pathway upwards of 3.6C. So, at the moment, much greater levels of ambition are needed from countries to put us on a path of emissions reductions that are steep enough to minimise any reliance on negative emissions (possibly to zero for 2C), to give us the greatest possible chance of staying below the 2 and 1.5C limits. Nothing should distract us from the need to shift to a fossil free world in the next decades.

This is a question of timescales. Before 2050, speculative technologies such as bioenergy with carbon capture and storage (BECCS), direct air capture and ocean geoengineering offer little promise, due to a variety of economic and technological hurdles. For now, less exotic land-use practices, such as soil carbon management, biochar, forestation and wetlands restoration, offer more promise. These are proven, and negative emissions can be achieved with immediate effect.

Beyond 2050, it is possible that sustained R&D will have rendered some of the speculative technologies viable. Of these, BECCS may hold the most promise – principally by virtue of being the least infeasible. Even then, experience to date with bioenergy has demonstrated how hard it is to achieve meaningful emissions reductions while confidently avoiding undesirable impacts on food security and land-use change, and CCS remains stuck at the demonstration stage.

Speculative negative emissions technologies may be worse than chimeras if they result in the false comfort that continued fossil fuel emissions can simply be offset, thereby diverting financial and policy resources from conventional mitigation. This would be reckless. It is clearly less risky not to emit a tonne of CO2 in the first place, than to emit one in expectation of being able to sequester it for an unknown period of time, at unknown cost, with unknown consequences, at an unknown date and place in the future.

Limiting warming to 2C means BECCS could require up to a quarter of global agricultural land – a problem in the context of rising global demand for food that would probably generate social resistance. Large-scale afforestation could encounter similar difficulties.

Major deployment of land-intensive negative emissions technologies would be more feasible were more land available. Dietary change is a major opportunity – the livestock sector uses around 70% of agricultural land and is itself a major source of emissions. Lower global meat consumption could reduce emissions, improve human health and free-up land for afforestation or BECCS, should it become viable.

Any technology deployed at large scale comes with pros and cons, and negative emissions technologies are no exception. Currently, no negative emissions technology entirely avoids potential detrimental societal side effects in a worst case scenario, but neither is there a single (low-carbon) energy technology that exclusively provides benefits. Nevertheless, our society will continue to produce energy in the future, and emissions have to be reduced to meet the Paris Agreement’s objectives. Technology preferences, thus, have to be considered against this backdrop: policies ensuring that detrimental side effects are limited are essential.

Considering these limitations, the most promising negative emissions technology appears to be the combination of centralised bioenergy power plants with carbon capture and storage (BECCS). In contrast to other negative emissions technologies, this technology provides the additional benefit of producing energy instead of merely consuming it. There surely are issues for its up-scaling. In general, negative emissions technologies’ only benefit is the removal of CO2 from the atmosphere. Without CO2emissions being penalised or strongly discouraged in some way, a large-scale deployment does never seem realistic. Then, there are further issues related to land and water competition for biomass production – this is a more general problem, not just for negative emissions – and related to safe ways to transport and store CO2. There is no silver bullet solution to climate change mitigation. The required scale of deployment of negative emissions technologies thus heavily depends on how ambitious mitigation on other fronts is. Research has shown that our dependence on more uncertain and risky negative emissions technologies can be severely limited if ambitious emissions reductions are implemented over the next decade, and if we pay attention to limiting global energy demand. Only with ambitious mitigation on other fronts do the negative emissions requirements for achieving the Paris Agreement’s objectives seem realistic.

Before we even start to talk about “negative” emissions, strong actions on carbon pollution in all sectors and all countries are due immediately. And there shall be no tweaking of environmental, social integrity and equity that must be the leading preconditions of any debate on meeting the 1.5C objective — an unnegotiable survival target for the most vulnerable people and ecosystems. A debate on the Paris objectivesmust not start with “negative emissions”, since this might be used to delay actions towards later decades. Many governments are experts on that.

A 1.5C objective leaves us with a carbon budget of about 300–900 gigatonnes of CO2(GtCO2), the average of about 12 years of present global emissions. We need to mobilise all sustainably existing renewable energy and energy efficiency technologies in the renewable energy and energy efficiency field, and aggressively implement those now. This includes both massively increased investment for mitigation by all financial sources in richer countries and significantly enhanced funding efforts for poor countries’ actions. It also requires a completely renewable-based transport fleet, including aviation and shipping, and early retirement of existing high-carbon assets, such as all coal and many gas plants, in the next two to three decades. Not least, phasing out of HFCs, halting deforestation and dietary changes towards a much lessmeat-based diet are essential. In other words, a 1.5C compliant world needs an annual decarbonisation rate of 8% and more.

This is not economical in the “classical” sense and truly inconvenient for some incumbents, but beneficial for the planet as a whole. Socially, developmentally and environmentally, this is superior for the billions of the poor and fragile ecosystems rather than relying on large scale BECCS, for instance, with unknown effects on food security. An effective phasing out of fossil fuels, besides other benefits, would also avoid the premature death of four million people annually from air pollution.

Yet, a certain part of negative emissions plays a key role now. Fostering natural carbon sinks in forests, grasslands and soils, if done properly, contribute tremendously to sustainable agriculture and forestry, as well as enhanced biodiversity.

Once this is all done, we might not need any of the other contentious technologies of negative emissions, such as BECCS and relying on unproven and leaky geological layers for CO2 storage for thousands of years. But actions have to be taken now!

Based on current knowledge, BECCS, which is the negative emission technology (NET) prevalent in the IPCC AR5 scenarios, and afforestation and direct air capture (DAC) are the only technologies that would achieve the large carbon removals needed to stabilise at or even below 2°C (Smith et al. 2016). However, BECCS and afforestation take up large areas of land and could thus hamper food security, while DAC needs lots of energy that needs to be produced in a carbon-neutral way. So there is no champion to be singled out here and we will have to look for the portfolio of NETs that minimizes unwanted effects on non-climate policy goals, while ensuring a high probability of reaching ambitious emissions reduction targets.

The term “feasible” has often been used to argue what is doable and not in the debate, but – in my opinion – the concept has not always been very helpful, as its meaning varies from technically feasible, over economically feasible or sustainably realisable, up to politically feasible. What is clear from a techno-economic perspective is that we are running later and later with deployment in order to achieve the negative emission levels indicated by low-stabilisation pathways in the IPCC’s 5th Assessment Report (IPCC 2014, WG3; Fuss et al. 2014; Smith et al. 2016). This is partially due to political feasibility reasons and also sustainability concerns, pointing to an urgent need for more research on this and further emphasising that we will need to look for a portfolio of options, where technologies with lower negative emissions potentials that meet less resistance could complement BECCS and afforestation.

We agree with a recent paper in Nature Climate Change (Smith et al, 2015) by many prominent modellers which concludes “there is no NET (or combination of NETs) currently available that could be implemented to meet the less than 2C target without significant impact on either land, energy, water, nutrient, albedo or cost and so ‘plan A’ must be to immediately and aggressively reduce GHG emissions”.

We are especially concerned about afforestation and bioenergy with carbon capture and storage (BECCS), widely used in IPCC AR5 scenarios, because of their very large land take of up to 6bn hectares, nearly half of the land surface area of the Earth. Land use change of more than a few hundred million hectares, let alone billions, has potentially severe implications for biodiversity and food security.

We have reservations about the practical feasibility and costs of deploying NETs on a large scale and, so far, none have been. As the IPCC AR5 points out for BECCS: “The potential, costs and risks of BECCS are subject to considerable scientific uncertainty.” Even large scale monoculture plantations (afforestation), which are probably the most practical NET, would require vast amounts of water, hundreds of cubic kilometres per year, and would undermine efforts to increase food security, alleviate poverty and conserve biodiversity.

Yet reaching 1.5C will undoubtedly limit climatic impacts on biodiversity and food security, but will probably require negative emissions in the range of 450-1000 GtCO2until 2100, even with aggressive emission reductions. A large proportion, if not all of this, could probably be achieved by the conservation and enhancement of natural forests, peatlands and other natural sinks and reservoirs – without recourse to NETs.

The most promising technologies are likely to change with time as our understanding improves. As of now, most effort has been on understanding bioenergy with carbon capture and storage (BECCS) and afforestation. Several less researched alternatives exist, such as enhanced weathering, direct air capture, ocean fertilization and biochar. Studies find that all carbon dioxide removal technologies have some sort of economic, biophysical, or ecological limitation. The best carbon dioxide removal strategy would be to use a mix of technologies, with each technology located to avoid its limitations.

We currently do not know what scale of carbon dioxide removal is feasible, and to minimise the risks of high temperature pathways, the best strategy is rapid decarbonisation of the global economy. In reality, we know it is unlikely that we can decarbonise at a rate sufficient to keep below 1.5/2C. I am quite sure we will have some small-scale carbon dioxide removal, but it is a challenge to scale up to a level that offsets positive emissions and removes carbon from the atmosphere. Each technology has its limitations, but I will point to two major constraints.

We often point to the faster-than-expected deployment of renewables, but rarely point to the slower-than-expected deployment of carbon capture and storage (CCS). CCS is a key technology in scenarios, both with bioenergy and fossil fuels. CCS is a tougher nut to crack than thought due to technical, political and social constraints. According to most emission scenarios, if we don’t have large-scale CCS, then we can’t keep below 1.5/2C.

Most carbon dioxide removal technologies require land. Reduced deforestation and increased afforestation will reduce the available land. Without rapid, perhaps infeasible, yield improvements, food production may take more land. To make more land available may require unprecedented amounts of fertiliser and water to drive yield improvements and improve unproductive land.

A broad portfolio of carbon removal solutions offers the potential to clean up excess CO2 from the atmosphere. Biological solutions – such as ecosystem restoration and “carbon farming” agricultural practices – can harness the power of photosynthesis to capture carbon from the air and store that carbon in plants and soils. Industrial approaches – such as bioenergy with carbon capture and storage, direct air capture, and enhanced rock weatherisation – can help the energy, manufacturing, and mining sectors generate negative emissions. A study from the National Academies estimates that these solutions could reliably sequester billions of tonnes of CO2 each year – all while offering additional environmental and social co-benefits, and new opportunities for economic growth in a carbon-constrained world.

Like all new technologies, however, each of these solutions faces serious challenges to reaching scale in an economically viable, sustainable and equitable way. For example, more science is needed to understand the amount and permanence of carbon sequestration from various agricultural, ecosystem, and geologic carbon removal approaches. New policies are needed to ensure large-scale bioenergy cultivation and ecosystem restoration does not lead to other adverse environmental or social consequences (such as driving up food prices, increasing pressure for deforestation, etc). And all solutions will need the cost curve to come down and find business models that enable their commercialisation.

I have no doubt it is technically feasible to achieve the Paris Agreement goals. To do so, we will need to pick up the pace on reducing CO2 emissions and we will need to invest in the research, development, and demonstration of a broad (and dynamically updating) portfolio of carbon removal technologies, as well as create markets and policies that support their commercialisation in a sustainable and equitable manner. Technology development takes time and is inherently uncertain – we must engage the diverse set of stakeholders that will develop and deploy carbon removal systems today to ensure we can meet tomorrow’s climate, economic, and social goals.

The use of natural carbon sinks, such as soil and biomass (afforestation, BECCs), hold some promise, but the quantity of carbon pollution that can be captured is limited, so is not a replacement for rapid and wholesale reduction of fossil fuel use. It is also constrained by other factors. For example, already humans are already consuming an astonishing proportion of total global biomass production, leaving little for the other species. If we want to use biomass as an energy source with carbon capture and storage we have to see significant reductions in other uses of biomass, particularly meat and dairy consumption. Tom Powell and Tim Lenton from Exeter University found this is a prerequisite for any significant quantity of negative emissions. Similarly, using theglobal pathway calculator produced by DECC, IEA, WRI and others to model a pathway to 1.5C also requires action on diet, with Friends of the Earth’s pathway requiring a 50% decline in global average meat consumption, against a backdrop of increased global consumption. Reducing meat consumption significantly frees up land, some of which could be used for biomass production and some of which must be used for nature to ensure resilient ecosystem services. But, critically, use of land always brings issues of land rights and justice. It would be a disaster to see poor communities thrown of their land for negative emissions — land grabs are well documented for biofuels.

Friends of the Earth’s 2011 report Negatonnes identified that, theoretically, by far the greatest contribution to negative emissions could come from chemical air capture coupled with carbon capture and storage. But it also pointed out that the cost of this would be extremely high, the scale of the industry would need to be absolutely enormous, and that by far the cheapest and most sensible approach to addressing climate change is rapidly cutting carbon pollution now.

[It is feasible to achieve the scale of deployment required to meet the aims of the Paris Agreement] only if we stop thinking that biomass is a limitless resource, that we need to reduce total use, and that we need to significantly cut meat and dairy consumption and be prepared to spend vast sums on chemical air capture.