Fig. 1: Emissions of SF₆ in China.

Emissions of SF₆ in China derived in this study (black line) are compared to several previous bottom–up (plot a)^{7,8,10,16,17,18,19,20,21,22,23} and top–down (plot b)^{10,25,33,68,69} emission estimates. The gray shading represents the 68% uncertainty intervals of the top–down emissions in this study. Error bars for the cited emissions denote the 68% uncertainty intervals (or the 1-sigma uncertainties), with exceptions for bottom–up emissions in Fang et al.⁷ where the 95% uncertainty intervals were quoted; uncertainties in the bottom–up emissions in Simmonds et al.¹⁰ were estimated based on the range of activity data; uncertainties in the top–down emissions in Fang et al.²⁵ were determined through sensitivity inversion tests; and Vollmer et al.³³ defined uncertainties based on different a priori emissions. Please refer to the respective references for detailed definitions. The numbers in the parentheses after each of the legends represent the years covered by that study. All known SF₆ emissions in China since 2005 reported by previous studies are displayed in the plot for a complete comparison, while emissions in early years that do not overlap with the time period covered by this study are not discussed in the main text. Previous top–down estimates from Simmonds et al.¹⁰ (yellow line in plot b) have focused on eastern China. They used population density as a proxy to extrapolate to a national total. The “ISC” in the legends of top–down estimates indicates the use of an interspecies correlation method in that particular study, and all other top–down studies without “ISC” in the legend used an inversion method.