Abstract: The surface meteorological data and mass concentration data of fine particulate matter (PM_2.5) and ozone (O₃) from 2017 to 2020 in Xinyang are collected. The Kolmogorov-Zurbenko (KZ) filtering method is employed to decompose the original concentration series into short-term, seasonal, and long-term components. Then the stepwise regressionis used to establish the linear regression model between pollutants baseline and short-term components and the corresponding scale meteorological elements. The residual differences between the baseline and short-term components are filtered and series reestablished. Finally, the long-term variations of PM_2.5 and O₃ without the influence of meteorology are obtained. These long-term variations are only related to the emission of pollutants. The results show that the fluctuations of PM_2.5 and O₃ concentrations are mainly caused by pollution emissions, and meteorological variations in short-term and seasonal change. Meteorological conditions have a great influence on the seasonal component of PM_2.5 and the long-term component of O₃. The emission of PM_2.5 weakens, and the emission of O₃ pollution increases first and then decreases from October, 2018, in Xinyang. The study also reveals that, from 2017 to 2020, due to emissions, the long-term components of PM_2.5 and O₃ decreases by 3.5 and 1.5 μg/(m³·a), respectively.