Stringent Emission Controls Are Needed to Reach Clean Air Targets for Cities in China under a Warming Climate.

Quantifying the threat that climate change poses to fine particle (PM 2.5 ) pollution is hampered by large uncertainties in the relationship between PM 2.5 and meteorology. To constrain the impact of climate change on PM 2.5 , statistical models are often employed in a different manner than physical-chemical models to reduce the requirement of input data. A majority of statistical models predict PM 2.5 concentration (often log-transformed) as a simple function of meteorology, which could be biased due to the conversion of precursor gases to PM 2.5 . We reduced this bias by developing a unique statistic model where the sum of PM 2.5 and the weighted precursor gases, rather than the PM 2.5 alone, was predicted as a function of meteorology and a proxy of primary emissions, where the input data of PM 10 , CO, O 3 , NO x , and SO 2 were obtained from routine measurements. This modification, without losing the simplicity of statistical models, reduced the mean-square error from 27 to 17% and increased the coefficient of determination from 47 to 67% in the model cross-validation using daily PM 2.5 observations during 2013-2018 for 74 cities over China. We found a previously unrecognized mechanism that synoptic climate change in the past half-century might have increased low quantiles of PM 2.5 more strenuously than the upper quantiles in large cities over China. Climate change during 1971-2018 was projected to increase the annual mean concentration of PM 2.5 at a degree that could be comparable with the toughest-ever clean air policy during 2013-2018 had counteracted it, as inferred from the decline in the daily concentration of carbon monoxide as an inert gas. Our estimate of the impact of climate change on PM 2.5 is higher than previous statistical models, suggesting that aerosol chemistry might play a more important role than previously thought in the interaction between climate change and air pollution. Our result indicated that air quality might degrade if the future synoptic climate change could continue interacting with aerosol chemistry as it had occurred in the past half-century.