What is “net zero” for methane?

By: Bill Collins

Recent research is suggesting that the way methane is accounted for in climate targets overemphasises its contribution to climate change at the end of the century. This might mean that countries or sectors (e.g. agriculture) with large methane emissions might have to impose overly stringent CO2 cuts to compensate (but is that a bad thing?), while countries that are able to reduce methane emissions might get away with insufficient CO2 cuts.

The new way of looking at methane reveals that to stabilise temperatures it is the rate of methane emission that is important rather than the total methane emitted (Collins et al. 2020, Allen et al. 2018, Cain et al. 2019). Therefore, a useful way of comparing methane and CO2 is to compare a change in the rate of methane emissions (tonnes per year) with a change in cumulative CO2 emissions (tonnes). This would have considerable implications as to what “net zero” means for each country.

The big step change in understanding from the IPCC 5th Assessment Report (IPCC 2013) was the concept of an almost linear relationship between the amount of carbon dioxide emitted and resulting temperature. From this it follows that to stabilise temperatures the net amount of carbon dioxide emitted will have to decline to zero. This was recognised in the Paris Agreement in which Article 4 refers to achieving “a balance between anthropogenic emissions by sources and removals by sinks of greenhouse gases in the second half of this century”. Some countries (including the UK) have therefore adopted a target of “net zero” greenhouses gas emissions by 2050.

Net zero is clear for CO2, but what about for methane? Methane has a natural removal sink through chemical oxidation in the atmosphere. Should this process be considered part of the “removals by sinks” specified in the Paris Agreement or should only direct capture of methane be counted? Methane behaves very differently to CO2 such that if we maintain global methane emissions at the current level then the concentrations of methane will stabilise and not contribute further to global warming. Reduction in methane emissions will even make a cooling contribution to climate. This contrasts with CO2 where constant emissions lead to a constant warming rate and reduced emissions still lead to a reduced warming rate until they decline to zero.

So how do we include methane in the net zero target? A common way to compare greenhouse gases is through a metric called the 100-year Global Warming Potential (GWP100) that compares the energy imbalance (radiative forcing) caused by emissions of a tonne of each gas averaged over the following 100 years. The GWP100 suggests 1 tonne of methane emission is equivalent to 34 tonnes of CO2 (Collins et al. 2020). Because methane has a short lifetime in the atmosphere (around 12 years) its climate impacts are much smaller by the end of the century (say 2070 or 2095) – only equivalent to 14 or 8 tonnes of CO2 respectively (Collins et al. 2020). This means that GWP100 overestimates the temperature effects of methane on the sort of time frames when we might reach peak warming.

Achieving “net zero in 2050”: schematic comparison of assumptions assuming a constant decline in methane emissions to 2100. CO2 equivalent emissions for methane (dotted red) are based on a rate/cumulative metric of 3000 years. Total CO2 equivalent (dotted black) are the sum of the solid green and dotted red.

Left: Scaling the methane emissions by 30 (GWP100), non-zero methane in 2050 means negative CO2 emissions are needed to compensate. The CO2 equivalent emissions become net zero earlier than 2050 and stay negative leading to an early peak in temperature and a decline afterwards.

Right: using the rate vs cumulative metric (GWP* or combined-GTP – see text), decreasing methane emissions in 2050 means CO2 emissions can be slightly positive. The CO2 equivalent emissions become net zero in 2050 and remain zero leading to a stabilisation of temperature at 2050.

The solution is to put all this together. Temperatures depend on the rate of methane emissions, but the accumulation of CO2 emissions. Hence a useful metric compares changes in these two quantities rather than GWP100 which compares changes in cumulative emissions of both. These rate vs. cumulative metrics are called GWP* in Allen et al. (2018) and combined-GTP in Collins et al. (2020), and a mix of GWP100 and GWP* has been called “CO2-warming equivalent” in Cain et al. (2019). These metrics have values of around 3000 years – depending exactly on how they are calculated. This importance of these new metrics is that a zero change (i.e. constant) emission rate of methane is equivalent to a net zero change in cumulative CO2, and a reduction in methane emission rate of 1 tonne per year is equivalent to a one-off negative emission of around 3000 tonnes of CO2.

Using the standard GWP100 metric for methane means CO2 emissions have to be more strongly negative to compensate, whereas for the GWP* or combined-GTP metric the CO2 emissions can be slightly positive (or more likely less negative to compensate for other gases). The second method leads to the desired stabilisation of climate, but the first method has lower temperatures. Is it better to overvalue methane if it makes us take more severe action on CO2, or to use a metric more closely tied to the temperature stabilisation that may lead to complacency?


Allen, M. R., J. S.Fuglestvedt, K. P. Shine, A. Reisinger, R. T. Pierrehumbert, and P. M. Forster, 2016: New use of global warming potentials to compare cumulative and short-lived climate pollutants. Nat. Climate Change, 6, 773–776. https://doi.org/10.1038/nclimate2998

Cain, M., J. Lynch, M. Allen, J. S. Fuglestvedt, D. J. Frame, and A. H. Macey, 2019: Improved calculation of warming-equivalent emissions for short-lived climate pollutants, npj Climate Atmos. Sci., 2, https://doi.org/10.1038/s41612-019-0086-4

Collins, W. J., D. J. Frame, J. S. Fuglestvedt, and K. P. Shine, 2020: Stable climate metrics for emissions of short and long-lived species – combining steps and pulses. Environ. Res. Lett., 15, 2, 024018 https://doi.org/10.1088/1748-9326/ab6039

This entry was posted in Climate, Climate change, Climate modelling. Bookmark the permalink.