A novel approach to identify the spatial characteristics of ozone-precursor sensitivity based on interpretable...

Area of Science:

• Environmental Science
• Atmospheric Chemistry
• Data Science

Background:

• Tropospheric ozone (O3) pollution is a significant environmental issue in China.
• Current methods for identifying ozone precursor sensitivity (OPS) have limitations in speed and accuracy.
• Effective O3 pollution control requires timely and precise OPS identification.

Purpose of the Study:

• To develop a novel, rapid, and accurate method for identifying OPS.
• To provide a tool for formulating effective O3 pollution control strategies.
• To analyze the spatiotemporal heterogeneity of OPS in the Guangdong-Hong Kong-Macao Greater Bay Area (GBA).

Main Methods:

• Developed a novel OPS identification approach using eXtreme Gradient Boosting and Shapley additive explanation (SHAP) algorithm.
• Incorporated volatile organic compound (VOC) photochemical decay adjustment with meteorological and pollutant monitoring data.
• Classified OPS into NOX-limited, VOCs-limited, and transition regimes based on SHAP value differences between scenarios.

Main Results:

• Demonstrated large spatiotemporal variations in OPS across the GBA during an autumn 2022 ozone pollution episode.
• Observed a shift in OPS from NOX-limited to VOCs-limited from September to October.
• Identified VOCs-limited conditions in central GBA areas and NOX-limited conditions in peripheral areas.

Conclusions:

• The developed method enables accurate OPS identification on a second timescale.
• This innovative approach provides a state-of-the-art tool for guiding spatial-specific O3 control strategies.
• The findings highlight the complex and dynamic nature of O3 pollution in the GBA, necessitating tailored control measures.