- Jiangsu Open University, Nanjing, China (lihm@jsou.edu.cn)
Ozone (O3) is a secondary pollutant in the atmosphere formed by photochemical reactions that endangers human health and ecosystems. Since the mid-1990s, Asian regions have experienced the fastest O3 increase rate of 2–8 ppb per decade at remote surface sites and in the lower free troposphere across the world. Therefore, a deeper understanding of the long-term changes and causes of tropospheric O3 concentrations is of significance in both the environment and climate policy making.
In this study, to quantify the impacts of future climate change on O3 pollution, near-surface O3 concentrations over Asia in 2020–2100 are projected using a machine learning (ML) method along with multi-source data. The ML model is trained with assimilated O3 data from a global atmospheric chemical transport model and real-time observations. The ML model is then used to predict future O3 with meteorological fields from CMIP6 multi-model simulations under various climate scenarios. The climate penalty on future O3 is robust over most regions of Asia. The near-surface O3 levels are projected to increase by 5 %–20 % over South China, Southeast Asia, and South India under the high-forcing scenarios in the last decade of 21st century, compared to the first decade of 2020–2100. We also find that the summertime O3 pollution over eastern China will expand from North China to South China and extend into the cold season in a warmer future.
Unlike the traditional “black box” ML models, we predict near‐surface O3 concentrations in China in 2030 and 2060 based on a process‐based interpretable ML method, integrated with physical and chemical processes of O3, natural emissions of O3 precursors, and other multi‐source data. The direct (via changing physical and chemical processes of O3) and indirect (via changing natural emissions of O3 precursors) impacts of future climate change on O3 concentrations are quantitatively analyzed. The results suggest that the climate‐driven O3 levels are projected to decrease by more than 0.4 ppb in 2060 over eastern China under a carbon neutral scenario relative to a high emission scenario. The physical and chemical processes under climate change play a more important role in regulating O3 concentrations than natural emissions in the future under the carbon neutral scenario.
How to cite: Li, H.: Projecting future climate change impacts on ozone pollution with machine learning, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5011, https://doi.org/10.5194/egusphere-egu26-5011, 2026.