Atmospheric methane removal may reduce climate risks

Atmospheric methane removal—breaking down the 1.9 parts per million (ppm) of methane in the atmosphere faster than natural systems already do—may be a critical tool to address the methane emissions gap and lower climate risks [1].

Limiting peak global warming to 1.5 °C or 2 °C will require both reaching net-zero emissions of long-lived greenhouse gases (particularly carbon dioxide) to stabilize temperatures and cutting short-lived climate pollutants (particularly methane) to lower near-term warming. The rate of near-term warming will dictate the prevalence of climate extremes and the speed at which humanity and nature must adapt [2]. The higher the peak temperature, the greater the risk of passing various Earth system tipping points, initiating irreversible changes and feedback loops that may destabilize climate and potentially societies [3].

Even if all currently available methane emissions reductions approaches were implemented simultaneously, we show below that there would still be a methane emissions gap between the resulting level of emissions and what is needed to limit global warming to 2 °C. This emissions gap will likely consist of two parts: residual hard-to-abate anthropogenic emissions and anthropogenically amplified natural emissions.

To be consistent with a scenario that keeps global warming below 2 °C (e.g. SSP1-2.6 [4]), anthropogenic methane emissions must be cut by nearly 50% from today's ∼380 million metric tons (Mt) per year down to ∼200 Mt yr⁻¹ or less by 2050 [4, 5]. Lowering methane emissions to 200 Mt yr⁻¹ would require the global implementation of all currently technologically feasible emissions reductions simultaneously [6].

Meanwhile, anthropogenically amplified natural emissions are already rising as temperatures rise and precipitation patterns change [7]. Wetland methane emissions are predicted to increase by 20–100 Mt yr⁻¹ above their baseline by 2050 depending on the global emissions trajectory and climate sensitivity [7, 8]. As global warming passes poorly understood thresholds, abrupt permafrost thaw could release a further 5–20 Mt yr⁻¹ of methane by 2050 [9]. Together, these systems could release 25–120 Mt yr⁻¹ of methane by 2050, contributing ∼0.05 °C–0.2 °C of warming [10] that is not accounted for in IPCC scenarios [4].

By 2050, the methane emissions gap for a 2 °C-consistent scenario will therefore be the sum of the residual anthropogenic emissions that remain in excess of 200 Mt yr⁻¹ and the predicted 25–120 Mt yr⁻¹ of unaccounted-for anthropogenically amplified natural emissions.

The primary tool to lower the methane emissions gap is anthropogenic methane emissions reductions. Lowering global warming overall would also lessen anthropogenically amplified natural emissions and therefore shrink the methane emissions gap. If proven viable, atmospheric methane removal—enhancing its natural sinks—could be a third way to lower the methane emissions gap.

Furthermore, methane removal may become even more valuable for lowering long-term climate risks if global warming exceeds 2 °C. Under full implementation of all current climate policies, global warming above preindustrial levels is predicted to reach 2.6 °C by 2100 (and continue rising) with a wide uncertainty range of 1.7 °C–3 °C [11]. Climate pledges, if implemented, would lower temperature projections substantially [11] but there are also risks to temperatures rising even faster: emissions could remain higher than modeled, the climate sensitivity could be on the higher end of its expected range, tipping points such as Antarctic ice sheet collapse could be crossed, and wetland and permafrost feedbacks could emit hundreds of Mt yr⁻¹ of methane by 2200 [8]. The marginal economic impact per unit of atmospheric methane increases in high-warming scenarios, so the value of removing methane will likely be higher in these scenarios [12].

Given the possibilities of high emissions, climate sensitivity, or climate feedbacks, a broad portfolio of climate actions would minimize risk. There are potential synergistic (or antagonistic) effects that must be considered, particularly for methane because its atmospheric lifetime depends on the anthropogenic emissions of other gases that influence methane sinks. For example, nitrogen oxides generate hydroxyl radicals (the primary sink for methane) while hydrogen and carbon monoxide compete with methane for hydroxyl radicals. Therefore, decreasing emissions of nitrogen oxides (for air quality improvements) or increasing emissions of hydrogen (from leaks if hydrogen is used as an energy source) or carbon monoxide (from wildfires) would lengthen methane's atmospheric lifetime [13, 14]. Methane removal must be additional to—not a replacement for—other climate actions, including greenhouse gas emissions reductions and carbon dioxide removal.

Methane is removed naturally from the atmosphere through oxidation (i.e. the chemical or biological conversion to carbon dioxide and water). Because methane is 43 times more potent than carbon dioxide per molecule, this conversion substantially reduces warming [15]. Roughly 95% of the methane sink is due to highly reactive atmospheric radicals (mainly hydroxyl, with a minor contribution from chlorine), while ∼5% is due to soil methanotrophs (methane-consuming microbes) [5]. Methane is far more dilute in the atmosphere than carbon dioxide (1.9 compared to 420 ppm), which makes capturing or reacting with it more difficult. However, in contrast to carbon dioxide, methane can be oxidized in situ by enhancing its natural sinks, avoiding the need for permanent capture and storage.

Proposed atmospheric methane removal approaches can be grouped into four categories. Three of these categories involve open-system approaches that modify the atmosphere or ecosystems: (i) atmospheric oxidation enhancement (increasing the generation of methane-destroying radicals in the open atmosphere); (ii) methane oxidizing coatings (covering surfaces with photocatalysts); and (iii) terrestrial methanotrophy enhancement (increasing the quantity or efficiency of methane-consuming microbes on the Earth's surface, including soils). The final category involves closed-system localized approaches: (iv) methane oxidizing reactors (using catalytic processes or methanotrophs to oxidize methane within an engineered system).

(i) Atmospheric oxidation enhancement approaches have two key advantages: sunlight provides the energy needed to start the chemical reactions and atmospheric mixing moves reactive gas-phase radicals through the air to react with methane molecules. The most studied approach in this category is iron salt aerosols, in which iron-based particles could be lofted into the atmosphere to catalytically generate chlorine radicals; evidence of a natural analog to this approach was recently observed in Saharan dust over the Atlantic [16]. Plume modeling and global Earth system modeling are needed to help evaluate the efficiency and full impacts of this approach and for other nascent atmospheric oxidation enhancement approaches such as releasing photocatalytic aerosols or hydrogen peroxide aerosols. Once modeling and smog chamber studies are completed, small-scale field tests would likely be required to investigate the efficiency of these aerosols in real-world conditions (if such field research obtains social license to operate).

However, enhancing the oxidative capacity of the atmosphere would do much more than just oxidize methane. Weighing the possible benefits and harms requires a more comprehensive understanding of the influence of aerosol size, altitude of deployment, and atmospheric conditions on potential unintended side effects, such as the formation of ground-level ozone or chlorinated compounds and the depletion of stratospheric ozone.

(ii) Methane oxidizing coatings also rely on sunlight for energy and do not require engineered airflow in the open atmosphere. However, the efficiency and reaction rates of methane oxidizing coatings are more likely to be limited by mass transfer (i.e. getting the methane to the reactive photocatalyst surface) than they are for atmospheric oxidation enhancement approaches [17]. Photocatalysts may be added to coatings already applied for other purposes, such as white paints used to cool buildings through increased albedo. The benefits of any methane destroyed would have to be weighed against other potential changes in radiative forcing, including potentially lowered albedo of the coating, and the possible byproducts such as volatile organic compounds.

(iii) Terrestrial methanotrophy enhancement attempts to increase methane uptake using natural biological sinks. The potential climate impacts, costs, and scalability of these approaches are poorly understood. Proposed approaches in this category include soil amendments and selectively breeding or engineering methanotrophs to increase their affinity for methane. The factors limiting natural biological methane sinks—such as the presence of nitrogen and copper, water saturation, and methane concentration [18]—must be better understood in order to evaluate the potential for scaling up enhancement of these sinks. As for all approaches, any increase in methane oxidation must be weighed against potential negative side effects, including nitrous oxide production and water quality impacts from runoff.

(iv) Methane oxidizing reactors—powered by heat, light, or methanotrophs—are a promising approach at emission sources, where methane concentrations are elevated. To be cost-effective at atmospheric concentrations, however, substantial breakthroughs are needed. For heat-based reactors, the operating temperature likely must be reduced from the current low of 300 °C down to below 30 °C [15]. For light-based reactors, including those that use photocatalysts and photolysis-based radical generation, the quantum efficiency (the ratio of oxidized methane molecules to incident photons) will need to be increased—from the current high of 0.8% potentially to at least 9% [15, 19]. For bioreactors, substantial technological breakthroughs are required to lower the lowest concentration at which they can operate from ∼500 ppm down to 1.9 ppm [20]. The requirements that these technologies will have to meet to achieve cost-effectiveness and scalability become even more stringent when considering the technoeconomics and energetics of moving climate-relevant quantities of air.

We suggest that each proposed methane removal approach should be evaluated using the following criteria:

• (1)  
  Is it climate beneficial? Any warming caused by implementing the approach (e.g. by greenhouse gases emitted during electricity generation) must be outweighed by any cooling (primarily due to the methane removed but also from other radiative forcing impacts).
• (2)  
  Is it cost-effective? Any costs should be outweighed by the economic benefits of the approach. Removal could be incentivized in various ways, including government-driven deployment, a regulatory mandate, or through a direct market price. One benchmark target cost could be the social cost of methane, estimated with high uncertainty to be $2000 per metric ton in 2030 and $3100 in 2050 [15].
• (3)  
  Is it scalable to 1 Mt yr⁻¹ of methane by 2050? Although every amount of methane removed lowers climate risks, we suggest a benchmark scale of relevance for each approach of 1 Mt yr⁻¹ of methane by 2050. Removing 1 Mt yr⁻¹ for three decades would lessen the methane emissions gap and shave ∼0.002 °C from a hypothetical scenario in which global warming peaks at 2 °C in 2080 [10, 21]. Examples of potential constraints on technological scalability include the availability of renewable energy, land for siting, and critical materials, such as metals for catalysts.
• (4)  
  Can it obtain social license to operate at scale? A number of factors must be sufficiently understood so as to inform relevant stakeholders of the tradeoffs before any decisions are made about deployment. These include social concerns such as whether an approach prolongs the use of fossil fuels and biophysical concerns including the effects on air quality, soil health, water quality, and ecosystem resilience.

If proven to be climate beneficial and cost-effective, open-system approaches, particularly atmospheric oxidation enhancement, likely have the largest potential scale and the fastest time to scale. However, open-system approaches may also be tougher to verify and have higher risks of unintended consequences, and will therefore require more complex governance than closed-system approaches. Obtaining robust social license to operate would require coordinated research, governance, and stakeholder engagement.

To reduce the risks and uncertainties surrounding open-system approaches, research is required to improve models, develop new measurement technologies, and increase observations. Modeling will help better constrain the risks, while measurement technologies will reduce uncertainties about the impacts. Because natural methane sinks are poorly constrained today, more frequent and spatially distributed observations would increase confidence in the baseline against which any human-generated increases in these sinks are attributed [5]. Different methane sinks produce different carbon isotope ratios, so isotopic measurements may help attribute methane concentration changes to specific approaches that enhance specific sinks [22]. Modeling, measurement, and observation research is needed now, both for methane and other relevant atmospheric gases and aerosols, such that any hypothetical future methane removal deployment is verifiable.

Governance and stakeholder engagement should progress in parallel with approach-specific research. Earning robust social license to operate will require informing all stakeholders of the tradeoffs inherent to each approach, balancing the risks of action with those of inaction, so that risk analysis facilitates an informed decision-making process. This balance will be particularly important for open-system approaches that will likely have complex, nonlinear, uncertain, and cross-border impacts.

Despite causing one third of anthropogenic greenhouse gas warming, methane emissions reductions and atmospheric removal only receive roughly 1% of funding from international climate finance [23]. More focus and resources should be directed towards methane—implementing existing emission reduction approaches, developing new approaches for reducing dilute emissions, and researching atmospheric removal approaches to better understand their risks, costs, and benefits.

S A is supported by the Stanford Woods Institute Goldman Graduate Fellowship and, with R B J, Grant # SPO1614153WTAQE. R B J acknowledges support from the Gordon and Betty Moore Foundation (Grant # GBMF5439, 'Advancing Understanding of the Global Methane Cycle'). We thank Erika Reinhardt, Paige Brocidiacono, David Mann, Stepher Kent, Brad Abernethy, Eric Davidson, Max Kessler, and Rob Buechler for their helpful comments on drafts of this manuscript.

No new data were created or analyzed in this study.