Historical and future changes and present-day uncertainties of ozone in China from CMIP6 models

Abstract. Ozone (O₃) contributes to global climate change and poses a direct threat to human health. This study analyzes historical and future changes, as well as current uncertainties, in surface O₃ concentrations in China, based on CMIP6 and the Tracking Air Pollution in China (TAP) dataset. The results are as follows: (1) The Multi-Model Ensemble Mean (MME) of CMIP6 simulated O₃ concentrations is higher during June–August (JJA), averaging 105 μg·m^-3, and lowest during December–February (DJF) at 55 μg·m^-3. (2) CMIP6 models generally underestimate O₃ concentrations in most regions of China, with the most significant underestimation occurring in East China. (3) The MME-simulated O₃ concentrations exhibit lower Bias, MAE, and RMSE over natural land surfaces compared to those over anthropogenic land surfaces. The Bias reaches its minimum under cloudy conditions and peaks under partly cloudy conditions. Furthermore, the Bias generally increases with rising PM_2.5 concentrations, however, once PM_2.5 exceeds a specific threshold, the Bias begins to decline. (4) Over the entire historical period, the MME simulates an increase of 39.3 μg·m^-3 in the annual mean surface O₃ concentration in China. (5) Under future SSP scenarios, MME projects generally increasing O₃ under weak mitigation (SSP3-7.0), with East China rising by 26.9 %. Strong mitigation (SSP1-2.6) leads to widespread decreases, especially in Southwest and South China (>30 μg·m^-3). (6) Differences in climate treatment, circulation, chemistry, and precursor emissions create substantial uncertainties, emphasizing the need to understand how emissions (including precursors and PM_2.5), climate, and model processes jointly affect future O₃ projections.