Pollution characteristics and influencing factors of atmospheric PAN in industrial area of Shenzhen

In this study, peroxyacetylnitrate (PAN) and precursor VOCs were measured in a typical industrial area in Shenzhen in the autumn (September 25 to October 31) of 2018, 2020 and 2021 to analyze the atmospheric photochemical pollution, and the generalized additive model (GAM) was used to study the effects of meteorological factors and precursors on the formation of PAN. During the sampling, the average mixing ratio of PAN showed the highest in 2018 (1.01×10-9), followed by 2021 (0.90×10-9) and 2020 (0.63×10-9). The diurnal variation showed that the nighttime background concentration of PAN was similar among the three years, and the interannual concentration difference depended on the formation of photochemical reaction during the daytime. The correlation between PAN and O3 was significant, and the R2 was 0.64~0.75. GAM model simulation results showed that multi-factor analysis could better characterize the nonlinear relationship between PAN and meteorological factors and key precursors, and determination coefficient R2 was 0.85~0.95, which were significantly better than single-factor analysis. PAN concentration was non-linear related to Ox, the photolysis rate of NO2 (JNO2), temperature, relative humidity (RH), acetaldehyde and NO. Among them, Ox, JNO2 and acetaldehyde showed significant positively non-linear correlation with PAN, and temperature and RH showed significant negatively non-linear correlation with PAN when temperature and RH were higher than 25°C and 50%. © 2023 Chinese Society for Environmental Sciences. All rights reserved.