Statistical significance of PM2.5 and O3 trends in China under long-term memory effects

Over the past decade, the Chinese government has implemented the “Clean Air Action” measures to enhance the atmospheric environmental quality, primarily focusing on curbing PM2.5 and O3 concentrations. The efficacy of these strategies and the underlying causes (human factors or natural variability) of any observed increases or decreases in PM2.5 and O3 concentrations are of great importance. Examining the hourly PM2.5 and O3 concentration time series from six representative regions in China between 2015 and 2021 revealed an overall downward trend in PM2.5 concentrations. However, the O3 concentration time series indicated upward trends in some regions, except for the Northeast area (NE) and Sichuan Basin (SCB). In the context of conventional significance tests, the assumption is typically that the time series' samples are independent and therefore memoryless. However, in situations where the time series exhibits strong autocorrelation and limited sample size, this assumption can lead to an overestimation of the statistical significance of the linear trend. To account for this, we utilized a long-term memory model that can reproduce the long-term persistence of pollutant records to improve the accuracy of significance tests. By comparing the P-values of real and surrogate data generated by the long-term memory model, we found that only PM2.5 concentrations in the Pearl River Delta (PRD) were slightly insignificant. For the remaining five regions, the P-values of PM2.5 concentrations were smaller than the significant level of 0.05, suggesting that the observed downward trends in PM2.5 concentrations are not due to natural variability, thereby confirming the effectiveness of the government's policies aimed at curbing atmospheric particulate matter in recent years. Our results show that O3 pollution is significantly increasing only in the Beijing-Tianjin-Hebei (BTH) region, beyond natural variability. In contrast, the trends of O3 pollution in many regions of China are markedly impacted by natural and climate variability.