Zum Hauptinhalt springen

Focus on reducing methane pollution from all sources, not distractions over metrics

May 20, 2024 Category: Policy Work Area: Methane
cows

According to the Intergovernmental Panel on Climate Change (IPCC), human-caused greenhouse gas emissions have warmed the planet by about 1.1 degree Celsius since 1900.1 This may seem to be a small temperature increase, but it is one that is causing huge impacts around the world including rising sea levels, more powerful storms, longer droughts, and worse wildfires. Methane is responsible for a large portion of this total warming – 0.5°C – and methane levels are rapidly rising in the atmosphere (see Fig. 1), making swift reductions in methane pollution a critical step to bending the curve on climate change. 

Methane pollution comes from the fossil fuel, waste, and agriculture sectors,2 and emissions from all sources are expected to grow. We need to find ways to reduce emissions from all sources, including agriculture, which emits about 40% of global methane pollution. Reducing agricultural methane poses a unique challenge, as global demand for animal protein is increasing due to population and income growth.3 However, economic analysis finds that to limit warming to 1.5°C, agricultural methane should be reduced by ~25% while methane from other sectors is cut more aggressively.

Methane and carbon dioxide warm the climate over different timescales, so it is best to set methane and CO2 reduction targets separately.4,5 Still, we still need ways to compare the effects of methane and CO2, and climate scientists have developed several new analytical approaches to compare methane and CO2. In this piece, we discuss the shortcomings of a fairly new metric, known as the “GWP*” (pronounced “GWP Star,” where GWP stands for Global Warming Potential), that is being used to argue that methane from some sources is far less harmful than conventional analysis suggests.  

Wichtigste Erkenntnisse: 

  • It is inappropriate to use GWP* to compare emissions between countries. This isn’t what GWP* was created for, and it leads to inequitable outcomes. 
  • Using GWP* to calculate CO2e emissions for methane “builds in” an expectation of continued high levels of anthropogenic methane emissions. But we urgently need to cut methane emissions.  The correct “baseline” for metrics, therefore, should be a baseline of not emitting.  
  • GWP* should not be used to distract any industries or sectors from the critical goal of achieving meaningful reductions in methane emissions. 

Why compare the harm CO2 and methane cause, and what are appropriate ways to do it? 

Carbon dioxide (CO2) is the most important climate pollutant: it causes the most warming, and unlike methane it stays in the atmosphere and continues to warm the climate almost indefinitely.  Because of CO2’s importance, the harm from methane and other climate pollutants is commonly compared to CO2. This is typically done by using global warming potential (GWP) to convert an amount of methane emissions to a carbon dioxide equivalent (CO2e) by multiplying the GWP by the mass of methane emitted. Typically, a 100-year GWP is recommended by the IPCC. However, this is a simplistic approach because methane and CO2 warm the climate over different timescales. When not done carefully, this exercise can lead to misleading and unhelpful results and comparisons. 

Nevertheless, metrics such as GWP and CO2e are needed – for example, to consider the implications of policies that lead to changes in emissions of both CO2 and methane. To address the limitations of GWP and CO2e, multiple alternative metrics for equivalence between pollutants such as methane and CO2 have been developed (e.g., natural science-oriented models such as the Global Temperature Potential or GTP and the Temperature Proxy Index or TEMP, economically-oriented metrics such as the Global Cost Potential or GCP and the Global Damage Potential or GDP, and others).

These metrics differ in many ways, such as the aspect of climate change they are considering (e.g., radiative forcing, temperature change, economic harm), the way they treat climate damage over time, geographic scope, the types of model they use, their baseline of comparison (e.g., emissions versus the absence of that emission or emissions versus a defined reference emission level), and their uncertainty level. All these metrics have limitations, and scientists have pointed out their advantages and disadvantages depending on the context in which they are used. The topic of climate metrics is a complex, ongoing debate within the scientific community – with the newest subject of debate being GWP*.  

What is GWP* and how does it work? 

GWP* is a newer metric developed to address a particular problem involving modeling global decarbonization budgets corresponding to specific temperature targets, as described in Box 1. GWP and GWP* are quite different: 

  • GWP is based on the total annual emission value, meaning that all emissions within a year are treated as contributing to environmental pollution. So, for GWP, the “baseline” for emissions is zero. 
  • GWP* focuses on the change in emissions over time. In this frame, the environmental impact of emissions today is evaluated relative to a baseline of emissions in the past. This means that even if today’s emissions are high, GWP* considers their impact on climate to be low if past emissions were similarly high. 

Calculating GWP* is not straightforward compared to calculating GWP, but the scientists who developed GWP* have reported that it can be approximated by “multiply[ing] the current methane emissions rate by 128 and subtract[ing] the methane emission rate of 20 years ago multiplied by 120.”6,7 

Therefore, the results of calculating CO2e using GWP or GWP* can be very different. Under GWP*, when methane emissions are constant over time, the calculation suggests that little or no additional warming occurs because GWP* is evaluating the warming impacts of methane pollution by comparing it to a baseline in which humans continue polluting methane at levels similar to today. However, human emissions have made current concentrations of methane in the atmosphere far too high and the appropriate baseline for evaluating the effects of methane pollution should be a baseline of not polluting. Thus, using GWP* to calculate CO2e emissions for methane builds an expectation of continued high levels of anthropogenic methane emissions into its calculations, which is not appropriate given the urgency of cutting methane.8 

Box 1 – Why was GWP* created?

One focus of climate policy is reducing peak temperature rise as much as possible – that’s the direct point of the 1.5°C and 2.0°C temperature goals. As researchers pointed out in the papers proposing GWP*, temperature at any point in time is essentially determined by: 

  1. Cumulative emissions (total tons) of CO2 and other long-lived climate pollutants, such as N2O; and  
  2. The recent rate (tons per year) of emissions of methane and other short-lived climate pollutants.   

One potentially useful way of thinking about the climate challenge is the climate budget.  How much climate pollution can global society still emit, without exceeding a temperature threshold?  Typically, this is calculated with climate models with a budget expressed in terms of the remaining tons of CO2e that can be emitted without exceeding the 1.5° or 2° C targets.  In principle, under a simple CO2e budget, there’s a direct trade-off between the amount of CO2 emissions allowed and the amount of methane emissions allowed. By this logic, the more methane pollution the world manages to cut, the more CO2 could be emitted before peak warming is reached.  

GWP* was proposed because in this context, emissions profiles that are within the CO2e budget (as calculated with GWP) but have very aggressive methane abatement strategies might still lead to exceeding the temperature target. That is, aggressive reductions in methane emissions could lead to numerically sufficient CO2e reductions, i.e., CO2e values would be below the budget for that temperature target, but temperature-wise we would be above the target. This happens because methane has a relatively short time in the atmosphere, and methane emitted in the near term would have been almost entirely naturally removed from the atmosphere by the time we reach peak temperature (which may be around 2060). When used properly in this context, GWP* allows a modeler to correctly balance CO2 and methane abatement and ensure that the temperature target is met.  

What impacts could GWP* have on policy decisions? 

Metrics should be selected to fit the goal of an analysis or a specific policy and require several considerations: 

  • Choosing a metric requires multidisciplinary perspectives (e.g., biogeochemistry, environmental economics, political science), in addition to climate physics – and value judgment.
  • Metrics must be accurate from a climate physics point of view. However, their application to policy should not be based solely on physical accuracy, but also on scientific, economic, and political appropriateness.
  • When metrics are used for policy making, they must also be fair and equitable.  

A number of analyses use GWP* to examine emissions from particular sectors or nations.9,10,11,12 In this context, GWP* can lead to inappropriate conclusions, such as the notion that two countries with similar methane emissions have very different climate impacts due to differing histories.13,14 

Consider two countries, A and B.  

Country A has had a stable population of cattle for several decades. In 2000, A’s cattle emitted a million tons of methane. In 2020, the country’s cattle still emitted a million tons of methane. The GWP* approach would estimate country A’s 2020 emissions to be equivalent to 8 million tons of CO2.  

Country B has rapidly developed since 2000, with population growth and decreased poverty, and increased consumption of meat and dairy. In 2000, B’s smaller and less productive cattle herd only emitted half a million tons of methane. In 2020, with greater demand, B had a similar population of cattle to A, so its cattle also emitted one million tons of methane in that year.  The GWP* approach would estimate the climate impact from B’s 2020 methane emissions to be equivalent to 68 million tons of CO2!   

Country B’s calculated 2020 impact is eight and a half times higher than country A’s 2020 emissions because B only polluted half as much as A in 2000.

Setting policy using these calculations would be unfair, unequal, and unethical.  It would penalize developing countries with increasing livestock, and reward developed countries with historically high levels of methane emissions.

This problem has been noted by others in the scientific community.15,16 

As described in Box 2, using GWP* in a policy context can produce even stranger results, since GWP* calculations for a country with dropping emissions can produce negative CO2e results. These results must be used with caution. It is easy to interpret these emissions as “climate cooling”17,18,19,20 – analogous to physically removing CO2 from the atmosphere, which leads to a cooler future than if the CO2 were not removed.  But ongoing methane emissions are directly responsible for current and continued warming, and therefore it is more appropriate to describe a reduction in emissions as “less warming,” not “cooling.” Even the authors who developed GWP* warn that “while some ongoing CH4 emissions may be able to give no further temperature increases from those emissions, maintaining these emissions into the future means they will continue to contribute to our elevated temperatures, and the resulting climate damages we will experience”.21 

Box 2 – A tale of two countries

Another challenging situation if GWP* is used to support policy arises when we consider a country with declining methane emissions, due, for example, to declining cattle herd sizes. Imagine these two countries: 

Country CCountry D
Emissions (tons)
KategorieYear 2000Year 2020Year 2000Year 2020
Methane from Cattle1,000,000800,000600,000800,000
Industrial CO2140,000,000110,000,00075,000,000110,000,000
2020 CO2e using GWP100132,400,000132,400,000
2020 CO2e using GWP*92,400,000140,400,000

This example is simplistic but illustrates the way that GWP* leads to peculiar outcomes which obscure valid comparisons between countries. Both countries C and D have identical emissions in 2020, and over the next decades, their 2020 emissions will both warm the climate by the same amount. But country C, with higher historic emissions, will be “credited” for these emissions in GWP* accounting, resulting in total calculated CO2e that is lower than actual CO2 emissions. Meanwhile, country D, which has contributed less to the warming experienced in 2020 (because it put less pollution in the air historically), is effectively penalized because those emissions were lower. Policies supporting the outcomes described would be unethical. 

 

We need to reduce methane from all sources – not just some 

The authors of GWP* have argued that there is no unfairness in the metric per se, and that the question of fairness and historical emissions accountability should be left to policy makers to decide.22 This is correct because it is the use of a metric that may yield unfair policies (and indeed, it is easy to point to other ways that metrics can lead to unfair comparisons), but it is also important to note that the incorrect use of GWP* could support allegations that sectors such as the livestock industries in some countries are not a major driver of the problem. This is based on analysis that leads to inequitable conclusions, and it is critical that these arguments not affect policymaking or industry’s commitment to finding real ways to reduce emissions. 

GWP* should not be used to argue for a “free pass” for methane.

Methane emitted today always creates a warmer world over the next decades, compared to a world where those emissions don’t occur.

Policies using GWP* may inaccurately “credit” emitters who merely hold steady or slightly reduce methane emissions, suggesting low or even “negative” emissions. This does not reflect the behavior of methane in the atmosphere, which always warms the climate.  Arguments based on GWP* that minimize the impact of agricultural methane will understandably lead to allegations of greenwashing,23 and it is encouraging to see the industry and extension services acknowledging the limitations of GWP*.24 

Metrics should support fair and equitable policies, not distract us from reducing emissions 

When modeling or planning decarbonization for a jurisdiction, no one metric can work adequately.25,26 Moreover, there is a broad consensus amongst climate scientists that this type of planning should separate methane (and other short-lived climate pollutants) from CO2, N2O, and other long-lived greenhouse gases, a so-called multi-basket approach.27,28 This approach would bring more clarity to policy making that sets emission reduction targets and help ensure that the various sectors reduce their emissions rather than obscuring them.  

Physically, a ton of methane warms the climate almost precisely the same whether it comes from sheep, a landfill, or an oil well,29 and it doesn’t matter whether that source is new or has been emitting for decades.

Metrics and policy must acknowledge this physical fact – and uncertainty and developing science cannot be used as a justification for delaying work on mitigation.  Therefore, we must work diligently to reduce methane from all countries and sources, accounting for the principles stated in the Paris Agreement in which international policies “will be implemented to reflect equity and the principle of common but differentiated responsibilities and respective capabilities, in the light of different national circumstances.30 


1 The Intergovernmental Panel on Climate Change (2021). Climate Change 2021: The Physical Science Basis, Working Group I, SPM, Fig 2. https://www.ipcc.ch/report/ar6/wg1/downloads/figures/IPCC_AR6_WGI_SPM_Figure_2.png

2 The United Nations Environmental Program, Climate and Clean Air Coalition (2022). Global Methane Assessment: Benefits and Costs of Mitigating Methane Emissions. https://www.ccacoalition.org/sites/default/files/resources//2021_Global-Methane_Assessment_full_0.pdf

3 OECD-FAO (2023). Agricultural Outlook 2023-2032. https://www.oecd-ilibrary.org/agriculture-and-food/oecd-fao-agricultural-outlook-2023-2032_08801ab7-en

4 Tanaka, K, Peters, G. P., & Fuglestvedt, J. S. (2010) Policy Update: Multicomponent climate policy: why do emission metrics matter? Carbon Management, 1:2, 191-197. https://www.tandfonline.com/doi/full/10.4155/cmt.10.28  

5 Allen, M. R., Peters, G. P, Shine, K. P et al. (2022). Indicate separate contributions of long-lived and short-lived greenhouse gases in emission targets. npj Clim Atmos Sci 5, 5.  https://doi.org/10.1038/s41612-021-00226-2

6 Oxford, Department of Physics. Measuring methane impact for climate policy. https://www.physics.ox.ac.uk/research/our-research-action/impact-stories/measuring-methane-impact-climate-policy

7 Smith, M. A., Cain M., & Allen, M. R. (2021). Further improvement of warming-equivalent emissions calculation. npj Clim Atmos Sci 4, 19. https://www.nature.com/articles/s41612-021-00169-8

8 Changing Markets (2023). Seeing stars: The new metric that could allow the meat and dairy industry to avoid climate action. https://changingmarkets.org/wp-content/uploads/2023/11/Seeing-stars-report.pdf

9 Del Prado A., Lynch J., Liu S. et al. (2023) Animal board invited review: Opportunities and challenges in using GWP* to report the impact of ruminant livestock on global temperature change. Animal 17, 5, 100790. https://doi.org/10.1016/j.animal.2023.100790

10 Ridoutt B. (2021). Short communication: Climate impact of Australian livestock production assessed using the GWP* climate metric. Livestock Science 246, 104459. https://doi.org/10.1016/j.livsci.2021.104459

11 Correddu F., Lunesu M. F., Caratzu M. F. et al. (2023) Recalculating the global warming impact of Italian livestock methane emissions with new metrics. Italian Journal of Animal Science 22, 1, 125-135. https://doi.org/10.1080/1828051X.2023.2167616

12 Plaza E. M. (2023). GWP* of U.S. Beef and Dairy Systems. Thesis submitted in partial fulfillment of the requirements for the degree of Master of Science. Colorado State University. https://api.mountainscholar.org/server/api/core/bitstreams/a94329f2-2804-49ac-98d7-451fe2156031/content

13 Rogelj J. & Schleussner C-F. (2019). Unintentional unfairness when applying new greenhouse gas emissions metrics at country level. Environ. Res. Lett. 14 114039 https://iopscience.iop.org/article/10.1088/1748-9326/ab4928

14 Rogelj J. & Schleussner C-F. (2021). Reply to Comment on ‘Unintentional unfairness when applying new greenhouse gas emissions metrics at country level’. Environ. Res. Lett. 16 068002 https://iopscience.iop.org/article/10.1088/1748-9326/ac02ec

15 Meinshausen M. & Nicholls Z. (2022). GWP* is a model, not a metric. Environ. Res. Lett. 17 041002 https://iopscience.iop.org/article/10.1088/1748-9326/ac5930/meta

16 Dooley K., Holz C., Klinsky S. et al. (2021) Ethical choices behind quantifications of fair contributions under the Paris agreement. Nat. Clim. Chang. 11, 300–305 (2021). https://doi.org/10.1038/s41558-021-01015-8

17 University of California, Davis, Clear Center. (2022) Global Dairy Platform releases brief on GWP* modeling for cattle sectors. https://clear.ucdavis.edu/news/global-dairy-platform-releases-brief-gwp-modeling-cattle-sectors

18 Cain M. (2019) New Zealand’s farmers have a chance to be climate leaders. Climate Change News. https://www.climatechangenews.com/2019/05/15/new-zealands-farmers-chance-climate-leaders/ 

19 Cady R. A. (2020) A Literature Review of GWP*: A proposed method for estimating global warming potential (GWP*) of short-lived climate pollutants like methane. Global Dairy Platform https://www.globaldairyplatform.com/wp-content/uploads/2020/11/literature-review-of-gwp-nov_20.pdf

20 Liu, S., Proudman J. & Mitloehner F. M. (2021) Rethinking methane from animal agriculture. CABI Agric Biosci 2, 22 (2021). https://doi.org/10.1186/s43170-021-00041-y

21 Lynch, J., Garnett, T., Persson, M., et al (2020). Methane and the sustainability of ruminant livestock (Foodsource: building blocks). Food Climate Research Network, University of Oxford. https://tabledebates.org/sites/default/files/2021-09/FCRN%20Building%20Block%20-%20Methane%20and%20the%20sustainability%20of%20ruminant%20livestock.pdf 

22 Cain M., Shine K., Frame D. et al. (2021) Comment on ‘Unintentional unfairness when applying new greenhouse gas emissions metrics at country level. Environ. Res. Lett. 16 068001 https://iopscience.iop.org/article/10.1088/1748-9326/ac02eb

23 University of California, Davis, Clear Center. (2020) Methane has been the Achilles’ heel for cattle emissions, but it may be part of a climate solution. https://clear.ucdavis.edu/news/methane-has-been-achilles-heel-cattle-emissions-it-may-be-part-climate-solution

24 Mitloehner F. (2023). Seeing stars from the GWP* debate. University of California Davis, Clear Center Seeing stars from the GWP* debate | CLEAR Center (ucdavis.edu)

25 For example, if one tries to design a simple ‘one-basket’ scheme for decarbonization, where the cheapest measures (in terms of cost in dollars per ton of CO2e reductions the measures achieve) are carried out first, one can construct a simplified but reasonable scenario where increasing the GWP for methane slows down methane abatement in the initial years of emissions reductions, and also slows down CO2 abatement. 

26 McCabe D., Cooley R. & Tellinghuisen S. (2020). Re: Policy Framework for Addressing Methane in Colorado’s Greenhouse Gas Regulations. https://cdn.catf.us/wp-content/uploads/2020/09/21092651/AQCC-GWP-methane-June-16-2020-Final.pdf

27 The adoption of GWP as the default metric for policy has been reported because of its simplicity, which facilitates negotiations, and it adds flexibility to countries on the mix of components in their abatement approach.

28 Rypdal K., Berntsen ZT., Fuglestvedt J. S. et al. (2005) Tropospheric ozone and aerosols in climate agreements: scientific and political challenges. Environmental Science & Policy, 8, 1, 29-43. https://doi.org/10.1016/j.envsci.2004.09.003

29 There is one small difference between fossil fuel sources (methane from oil, gas, and coal) and methane from biogenic sources (such as agriculture and waste): methane from fossil fuels is slightly worse than methane from agriculture or waste. This is because when methane oxidizes in the atmosphere, it forms CO2. If methane is from agriculture or waste, the CO2 that forms when it oxidizes isn’t ‘new’ CO2.  It is just part of the natural carbon cycle, just as CO2 from decomposing vegetation or animals and humans exhaling is a natural part of the carbon cycle, not climate pollution. However, when methane from fossil fuels is oxidized in the atmosphere, it’s new CO2, not part of the natural carbon cycle – just like CO2 from a power plant or tailpipe, so it’s climate pollution. However, this difference is minor, as this effect is a very small portion – a few percent – of the warming caused by fossil methane. Regardless of source, once methane is in the atmosphere, it leads to additional warming.

30 United Nations (2015) Paris Agreement. https://unfccc.int/sites/default/files/english_paris_agreement.pdf