Quantifying European SF6 emissions (2005-2021) using a large ensemble of atmospheric inversions

Sulfur hexafluoride (SF₆) is an extremely potent (GWP₁₀₀ = 24,300) and long-lived greenhouse gas whose atmospheric concentrations continue to rise due to anthropogenic emissions. Europe represents a particularly relevant test case for investigating SF₆ emissions, as successive EU F-gas regulations over the past two decades have aimed to substantially reduce emissions. A key question is whether these regulatory measures are reflected in observed emission trends and whether reported national inventories are consistent with observation-based estimates.

 In this study, we quantify European SF₆ emissions for the period 2005–2021 using a large ensemble of atmospheric inversions with a strong focus on uncertainty characterization. Uncertainties are assessed using an extensive set of sensitivity tests in which key inversion parameters are systematically varied, while final uncertainties are quantified via a Monte Carlo ensemble that randomly samples combinations of these parameters. This allows us to identify the main sources of uncertainty and to evaluate the robustness of inferred emission trends.

Our analysis focuses on countries with relatively dense observational coverage - the United Kingdom, Germany, France, and Italy - while also examining aggregated emissions for the EU-27.  The inversion results reveal declining SF₆ emissions in all studied regions except Italy, broadly consistent with the timing of EU F-gas regulations (842/2006, 517/2014). In several countries, inferred emissions exceed reported national inventories, although the agreement generally improves in more recent years. At the EU-27 scale, emissions exhibit a pronounced decline between 2017 and 2018, coinciding with a marked reduction in emissions from southwestern Germany, suggesting regional actions were taken as the 2014 regulation took effect.

Our sensitivity tests highlight the crucial role of dense and sustained atmospheric monitoring networks for robust inversion-based emission estimates. In particular, expansions of the UK observing system in 2012 and 2014 lead to significant reductions in emission uncertainties, demonstrating the importance of comprehensive observational networks in refining emission estimates.