Methane and CO2 emissions from the natural gas supply chain have been shown to vary widely but
there is little understanding about the distribution of emissions across supply chain routes,
processes, regions and operational practises. This study defines the distribution of total methane
and CO2 emissions from the natural gas supply chain, identifying the contribution from each stage
and quantifying the effect of key parameters on emissions. The study uses recent high-resolution
emissions measurements with estimates of parameter distributions to build a probabilistic emissions
model for a variety of technological supply chain scenarios. The distribution of emissions resembles
a log-log-logistic distribution for most supply chain scenarios, indicating an extremely heavy tailed
skew: median estimates which represent typical facilities are modest at 18 – 24 g CO2 eq./ MJ HHV,
but mean estimates which account for the heavy tail are 22 – 107 g CO2 eq./ MJ HHV. To place these
values into context, emissions associated with natural gas combustion (e.g. for heat) are
approximately 55 g CO2/ MJ HHV. Thus, some supply chain scenarios are major contributors to total
greenhouse gas emissions from natural gas. For methane-only emissions, median estimates are 0.8 –
2.2% of total methane production, with mean emissions of 1.6 - 5.5%. The heavy tail distribution is
the signature of the disproportionately large emitting equipment known as super-emitters, which
appear at all stages of the supply chain. The study analyses the impact of different technological
options and identifies a set of best technological option (BTO) scenarios. This suggests that
emissions-minimising technology can reduce supply chain emissions significantly, with this study
estimating median emissions of 0.9% of production. However, even with the emissions-minimising
technologies, evidence suggests that the influence of the super-emitters remains. Therefore,
emissions-minimising technology is only part of the solution: reducing the impact of super emitters
requires more effective detection and rectification, as well as pre-emptive maintenance processes.