Extended Data Fig. 4: Unaccounted-for CCl₄ emissions (a) and their proportion to top-down emissions (b) in China.

The red and grey lines represent the unexplained emissions considering or not considering ~5.1 Gg yr⁻¹ of emissions due to legacy emissions and by-product emissions related to chlorine production and usage. All values shown are the mean values derived from the MCMC method, with error bars for the 68% uncertainty intervals. The linear regression trend for the unexplained CCl₄ emissions in China (lines in plot (a)) is −0.73 ± 0.23 Gg yr⁻² over the 11 years during 2011-2021. The linear regression trend for the proportion of unexplained emissions during the 11-year period from 2011 to 2021 is −0.023 ± 0.007 yr⁻¹ when considering the legacy and chlorine-related emissions (red line in plot (b)), and −0.022 ± 0.006 yr⁻¹ without these emissions (grey line in plot (b)). The negative linear regression trends for both the unaccounted-for CCl₄ emissions and their proportion to top-down emissions in China are significantly different from zero at the 5% level (p < 0.05). Note that only the uncertainties from the top-down emissions are considered here. Including the uncertainties in the bottom-up emissions, which are currently unknown, would likely reduce the confidence in the calculated trends. However, we anticipate that systematic errors in the bottom-up emissions may dominate over random errors, meaning that the negative trends would persist, but with larger uncertainty in the magnitude of the emissions gap.