Many methods have been developed to verify the correlation between meteorological conditions and air pollutants; however, all have limitations that lead to biased or incomplete conclusions. Hence, improved methods are urgently required to describe this correlation comprehensively and accurately. In this study, we demonstrated the ability of the Copula function to apply time-varying correlations between meteorological factors and atmospheric pollutants. A mixed Copula model was constructed using meteorological monitoring data for Beijing and Guangzhou from 2014 to 2019 to dynamically analyse the correlation characteristics and tail dependence between these factors. We then performed a correlation analysis for the data from the average, lower, and upper tails to obtain a more accurate and comprehensive correlation description. Dynamic analysis results demonstrated significant seasonal fluctuations between meteorological conditions and pollutants relationships. Moreover, the correlation coefficient variations differ according to their average and tail values. High humidity is more likely to be accompanied by increased NO2 compared with average summer humidity. Our proposed model represents a novel application of the Copula function for determining the factors influencing air pollution. This model emphasizes the tail dependence between meteorological conditions and air pollutant concentrations and can be used to guide more targeted prevention and control strategies.