Methane is the second most important greenhouse gas after carbon dioxide, and also plays a central role in the chemistry of the atmosphere. The combination of its shorter lifetime and higher effectiveness as a greenhouse gas makes it an attractive option for near-term mitigation of climate change. Methane is also a key tropospheric ozone precursor: ozone is a greenhouse gas, and acts as an air pollutant in the troposphere. Therefore, mitigation of methane has both climate and air quality benefits.

A new configuration of the UK Earth System Model, UKESM1-ems, has been developed with an updated methane treatment. Methane emissions are input directly, rather than prescribing a global surface concentration. This thesis focuses on UKESM1-ems and the new capabilities it provides: a more process-based treatment of methane; simulating feedbacks in the methane cycle, and the ability to directly perturb methane emissions.

When compared to the previous, concentration-driven model, UKESM1-ems simulates the methane distribution with a better correlation compared to observations, including an improved latitudinal distribution, interhemispheric gradient and vertical gradient. The observed trend in methane over time is also reproduced, combining the methane emissions inputs, online wetland emissions and online chemistry and transport to simulate the methane mixing ratio. The modelled absolute methane mixing ratio is lower than observations: this is likely due to an underestimate in methane emissions, within the current large uncertainty range for emissions.

Experiments following different emissions pathways are explored using UKESM1-ems. Firstly, an idealised scenario where all anthropogenic methane emissions are removed instantaneously, to attribute the role of future anthropogenic methane. Methane declines to below pre-industrial levels within 12 years and global surface ozone decreases to levels seen in the 1970s. By 2050, 690,000 premature deaths per year and 1 degree of warming can be attributed to anthropogenic methane.

Secondly, the same low-methane scenario is used, with perturbed nitrogen oxide (NOx) and carbon monoxide (CO) emissions, to investigate their impact on the atmospheric oxidising capacity and test the hydroxyl (OH) relationship to NOx and CO. The effect of methane on NOx is also explored. Decreased methane emissions perturb both the NO/NO₂ ratio and the partitioning between NOx and reservoir species, leading to increased NOx in low-methane scenarios.

Finally, a Global Methane Pledge scenario is simulated. This pledge aims to reduce methane emissions by 30% globally by 2030, compared to 2020 values. The new ability of UKESM1-ems to mask emissions from different countries is used to implement this scenario and study regional impacts. The global mean methane mixing ratio decreases by 13% compared to 2020 levels. The expected temperature benefit (0.2°C) following this scenario is not seen in this experiment - this signal is too small and is within the noise and interannual variability of UKESM1-ems. There are global benefits for air quality, with ozone concentrations and population exposure to ozone decreasing in all countries. Global Methane Pledge member countries, where emissions reductions take place, see greater local air quality benefits than non-member countries.