A record-breaking severe ozone (O₃) pollution episode occurred in the Pearl River Delta (PRD) in early autumn 2019 when the PRD was under the influence of a Pacific subtropical high followed by Typhoon Mitag. In this study, we analyzed the effects of meteorological and photochemical processes on the O₃ concentration in the PRD during this episode by carrying out the Weather Research Forecast–Community Multiscale Air Quality (WRF-CMAQ) model simulations. Results showed that low relative humidity, high boundary layer height, weak northerly surface wind, and strong downdrafts were the main meteorological factors contributing to O₃ pollution. Moreover, delayed sea breezes that lasted into the night would transport O₃ from the sea back to the land and resulted in secondary O₃ maxima at night. In addition, O₃ and its precursors stored in the residual layer above the surface layer at night can be mixed down to the surface in the next morning, further enhancing the daytime ground-level O₃ concentration on the following day. Photochemical production of O₃, with a daytime average production rate of about 7.2 ppb h⁻¹ (parts per billion), is found to be the predominate positive contributor to the O₃ budget of the boundary layer (0–1260 m) during the entire O₃ episode, while the horizontal and vertical transport fluxes are the dominant negative contributors. This O₃ episode accounted for 10 out of the yearly total of 51 d when the maximum daily 8 h average (MDA8) O₃ concentration exceeded the national standard of 75 ppb in the PRD in 2019. Based on these results, we propose that the enhanced photochemical production of O₃ during the episode is a major cause of the most severe O₃ pollution year since the official O₃ observation started in the PRD in 2006. Moreover, since this O₃ episode is a synoptic-scale phenomenon covering the entire eastern China, we also suggest that the enhanced photochemical production of O₃ in this O₃ episode is a major cause of the extraordinarily high O₃ concentrations observed in eastern China in 2019.

An observation-based method (OBM) is developed to investigate the sensitivity of ozone formation to precursors during two persistent elevated ozone episodes observed at 77 stations in Guangdong. Average OH concentrations derived at the 77 stations between 08:00 and 13:00 local time stay within a narrow range of 2.5×10⁶ to 5.5×10⁶ cm⁻³ with a weak dependence on the NO_x. These values are in good agreement with OH values observed at a rural station in the Pearl River Delta (PRD). They also agree well with a box model constrained by the ambient conditions observed during the two episodes. The OBM has been used to evaluate the ozone production efficiency, ε(NO_x or volatile organic compound, VOC), defined as the number of O₃ molecules produced per molecule of NO_x (or VOC) oxidized. Average values of ε(NO_x) and ε(VOC) determined by the OBM are 3.0 and 2.1 ppb ppb⁻¹, respectively, and both compared well with values in previous studies. Approximately 67 % of the station days exhibit ozone formation sensitivity to NO_x, and approximately 20 % of the station days are in the transitional regime sensitive to both NO_x and VOC, and only approximately 13 % of the station days are sensitive to VOC. These results are in semi-quantitative agreement with the ozone formation sensitivity calculated by the box model constrained by ambient conditions observed during the two episodes. However, our OBM results differ from those of most previous investigations, which suggested that limiting the emission of VOC rather than NO_x would be more effective in reducing ozone reduction in Guangdong.

Accurate estimation of ozone (O₃) concentrations and quantitative meteorological contribution are crucial for effective control of O₃ pollution. In recent years, there has been a growing interest in leveraging machine learning for O₃ pollution research due to its advantages, such as high accuracy, strong generalization, and ease of use. In this study, we utilized meteorological parameters obtained from european center for medium—range weather forecasts (EMCWF) Reanalysis v5 data as input and employed five distinct machine learning methods to estimate values of maximum daily 8‐hr average (MDA8) O₃ concentrations and analyze meteorological contributions. To improve the accuracy and generalization capabilities of the estimation, we employed Grid SearchCV techniques to select optimal parameters and mitigate the risk of overfitting. Additionally, we incorporated meteorological normalization and the SHAP model to quantify the influence of various parameters. Among the models evaluated, the Extreme Gradient Boost model exhibited exceptional performance from 2015 to 2022, yielding determination coefficients of 0.85 and 0.80 for the training and test data sets, respectively. The outcomes of meteorological normalization revealed that meteorological parameters accounted for 87.7% of the impacts in 2018, while emission‐related factors constituted 80.8% of the impacts in 2021. Over the period spanning 2015–2022, 2 m temperature emerged as the most influential parameter affecting daily MDA8 O₃ concentration, with an average contribution of 9.4 μg m⁻³.

Due to a robust emission control policy, significant reductions in major air pollutants, such as PM_2.5, SO₂, NO₂, and CO, were observed in China between 2015 and 2020. On the other hand, during the same period, there was a notable increase in ozone (O₃) concentrations, making it a prominent air pollutant in eastern China. The annual mean concentration of maximum daily 8 h average (MDA8) O₃ exhibited alarming linear increases of 2.4, 1.1, and 2.0 ppb yr⁻¹ (ppb is for parts per billion) in three megacity clusters: Beijing–Tianjin–Hebei (BTH), the Yangtze River Delta (YRD), and the Pearl River Delta (PRD), respectively. Meanwhile, there was a significant 3-fold increase in the number of O₃-exceeding days, defined as MDA8 O₃ > 75 ppb. Our analysis indicated that the upward increases in the annual mean concentration of MDA8 were primarily driven by the rise in consecutive O₃-exceeding days. There were expansions of high O₃ in urban centers to rural areas accompanied by a saturation effect so that MDA8 O₃ concentrations at the high-O₃ stations in 2015 remained nearly constant at 100 ppb. Last, we found a close association between O₃ episodes with 4 or more consecutive O₃-exceeding days and the position and strength of tropical cyclones (TCs) in the northwest Pacific and the West Pacific subtropical high (WPSH). The TC and WPSH contributed to meteorological conditions characterized by clear skies, subsiding air motion, high vertical stability in the lower troposphere, increased solar radiation, and a positive temperature anomaly at the surface. These favorable meteorological conditions greatly facilitated the formation of O₃. Thus, we propose that the worsening O₃ increases observed in the BTH, YRD, and PRD regions from 2015 to 2020 can be mostly attributed to enhanced photochemical O₃ production resulting from an increased occurrence of meteorological conditions with high solar radiation and positive temperature anomalies under the influence of the WPSH and TCs.

Multiple linear regression (MLR) models are used to assess the contributions of meteorology/climate and anthropogenic emission control to linear trends of PM_2.5 concentration during the period 2013–2018 in three regions in eastern China, namely Beijing–Tianjin–Hebei (BTH), the Yangtze River Delta (YRD), and the Pearl River Delta (PRD). We find that quantitative contributions to the linear trend of PM_2.5 derived based on MLR results alone are not credible because a good correlation in the MLR analysis does not imply any causal relationship. As an alternative, we propose that the correlation coefficient should be interpreted as the maximum possible contribution of the independent variable to the dependent variable and the residual should be interpreted as the minimum contribution of all other independent variables. Under the new interpretation, the previous MLR results become self-consistent. We also find that the results of a short-term (2013–2018) analysis are significantly different from those of a long-term (1985–2018) analysis for the period 2013–2018 in which they overlap, indicating that MLR results depend critically on the length of time analyzed. The long-term analysis renders a more precise assessment because of additional constraints provided by the long-term data. We therefore suggest that the best estimates of the contributions of emissions and non-emission processes (including meteorology/climate) to the linear trend in PM_2.5 during 2013–2018 are those from the long-term analyses: i.e., emission <51 % and non-emission >49 % for BTH, emission <44 % and non-emission >56 % for YRD, and emission <88 % and non-emission >12 % for PRD.