Understanding the spatial and seasonal variation of the ground-level ozone in Southeast China with an interpretable machine learning and multi-source remote sensing.

Ground-level ozone (O) pollution poses significant threats to both human health and air quality. This study uses ground observations and satellite retrievals to explore the spatiotemporal characteristics of ground-level O in Zhejiang Province, China. We created data-driven machine learning models that include meteorological, geographical and atmospheric parameters from multi-source remote sensing products, achieving good performance (Pearson's r of 0.81) in explaining regional O dynamics. Analyses revealed the crucial roles of temperature, relative humidity, total column O, and the distributions and interactions of precursor (volatile organic compounds and nitrogen oxides) in driving the varied O patterns observed in Zhejiang. Furthermore, the interpretable modeling quantified multifactor interactions that sustain high O levels in spring and autumn, suppress O levels in summer, and inhibit O formation in winter. This work demonstrates the value of a combined approach using satellite and machine learning as an effective novel tool for regional air quality assessment and control.