Generalized Additive Model (GAM) Applied to the Analysis of Ozone Pollution in a City in Eastern China

Ground-level ozone (O3) pollution remains persistently high in China, despite the implementation of stringent emission controls targeting primary pollutants. However, understanding of the drivers and formation mechanisms of this secondary pollutant remains limited. Herein, comprehensive field observations of O3 and its precursors were conducted in a medium-sized city in eastern China. The average O3 concentration was 93.60 ± 61.98 μg·m−3, with severe pollution accounting for 47.05% (high-temperature, low-humidity conditions). The peak O3 concentration during pollution episodes (207.13 ± 34.93 μg·m−3) exceeded that of non-pollution periods (108.77 ± 43.99 μg·m−3) by more than twofold. A generalized additive model (GAM) was employed to identify the key drivers of O3 pollution, revealing relative humidity (RH) (F = 36.95) and volatile organic compounds (VOCs) (F = 8.03) as dominant drivers. Further interaction analysis using the GAM showed synergistic effects between RH and nitric oxide (NOx) as well as the temperature (T) and NOx on O3 evolution. O3 formation sensitivity analysis demonstrated that O3 production was primarily within a VOC-limited regime (VOCs/NOx 10 μg·m−3). The toluene/benzene ratio indicated that Taizhou’s ambient VOCs were dominated by vehicle exhaust emissions, with minor contributions from solvents, oils, and gases, and LPG volatilization, making vehicle exhaust control the core of VOC reduction. The air mass transport from the Yellow Sea also significantly affected the local O3. This study quantifies the effects of multiple factors of summertime O3 pollution and provides scientific support for targeted O3 control strategies in a medium-sized city in eastern China.