Abstract: Solar radiation, relative humidity, and ambient temperature are three key meteorological controlling factors for the photochemical reaction of ground-level ozone (O₃). Their distribution forms play an extremely important role in the nonlinear evolution of O₃. The hourly O₃ concentration data and corresponding meteorological data in the same period from 2016 to 2019 during summer in Chengdu were collected and Maximum Likelihood Estimation method was used to estimate parameters for probability distribution functions. At the same time, Kolmogorov-Smirnov test statistics (K-S test), root-mean-square-error (RMSE), Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC) were further combined to analyze the goodness of fit of multiple probability distribution functions of the three meteorological controlling factors for O₃ photochemical reaction,respectively. The results indicate that:(1) Johnson SB distribution function had the best fitting effect for both solar radiation and relative humidity. Additionally, under different O₃ concentration levels, the probability distribution of solar radiation and relative humidity kept obey Johnson SB distribution at significant level of α=0.05 in K-S test. (2) However, Johnson SU distribution function had the best fitting effect for ambient temperature. Moreover, the probability distributions of ambient temperature obey Johnson SU distribution under different O₃ concentration levels at significant level of α=0.05 in K-S test. (3) The probability distribution characteristics of the three meteorological controlling factors vary significantly and regularly with the changing of O₃ concentration level. These research results mentioned above could provide technical supports for the construction of O₃ pollution potential model during summer in Chengdu.

Key words: ozone, photochemical reaction, controlling meteorological factor, probability distribution, Chengdu