Search

Driven by precursor emissions, meteorological conditions, and other factors, atmospheric ozone (O 3 ) has become the main pollutant affecting urban air quality in summer. The current deductive models driven by physical and chemical mechanisms require a large number of parameters for the analysis of O 3 pollution, and the calculation timeliness is poor. The data-driven inductive models are efficient but have problems such as poor explanation. In this study, an explainable model of data-driven Correlation-ML-SHAP was established to reveal the strongly correlated influencing factors of O 3 concentration. Additionally, the machine learning ML module coupled with the explainable SHAP module was used to calculate the contributions of driving factors to O 3 concentration, so as to realize the quantitative analysis of driving factors. The O 3 pollution process in the summer of 2021 in Jincheng City was used as an example to carry out the application research. The results showed that the Correlation-ML-SHAP model could reveal and use strong driving factors to simulate O 3 concentration and quantify influence contribution, and the ML module used the XGBoost model to achieve the best simulation accuracy. Air temperature, solar radiation, relative humidity, and precursor emission level were the strong driving factors of O 3 pollution in Jincheng City in summer 2021, and the contribution weights were 32.1%, 21.3%, 16.5%, and 15.6%. The contribution weights of air temperature, solar radiation, and precursor emission level increased by 3.4%, 1.2%, and 1.2% on polluted days, respectively, and the contribution weights of precursor emission level rose to third place on polluted days. Each driving factor had a nonlinear interaction effect on O 3 concentration. When the air temperature exceeded 24℃, or the relative humidity was lower than 70%, there was a 94.9% and 94.1% probability of positive contribution to O 3 pollution, respectively. Under such meteorological conditions, ρ (NO 2 ) exceeded 9 μg·m -3 , or ρ (CO) exceeded 0.7 mg·m -3 , and there was a 94.9% and 99.3% probability of positive contribution to O 3 pollution, respectively. The southeast wind speed was lower than 5.8 m·s -1 , or the south wind speed was lower than 5.3 m·s -1 , both of which contributed positively to O 3 pollution. The model quantitatively analyzed the influence contribution of various driving factors on urban O 3 concentration, which could provide a basis for the prevention and control of urban atmospheric O 3 pollution in summer.

Based on the daily average concentration of PM 2.5 , social influencing factor data, and meteorological data of 11 cities in Shanxi Province from 2015 to 2019, the concentration period of PM 2.5 was determined using wavelet transform. The correlation between PM 2.5 and social influencing factors and meteorological factors was explored respectively through Spearman correlation and the wavelet coherence spectrum, and the main influencing factors of long-term and short-term management and control of PM 2.5 were determined. The results showed that the concentration of PM 2.5 in Shanxi Province showed an upward trend from 2015 to 2017, with an average annual increase rate of 4.3% and a downward trend from 2018 to 2019, with an average annual decrease rate of 4.2%. The average concentration of PM 2.5 showed a "U" distribution, with the highest value in January (95 μg·m -3 ) and the lowest in August (34 μg·m -3 ); the average value in winter was approximately twice that in summer. The ρ (PM 2.5 ) in southern cities such as Linfen was 62 μg·m -3 , and the average value in Datong and other northern cities was 45 μg·m -3 , which was high in the south and low in the north. There were significant periodic changes in PM 2.5 concentration in the 11 cities, including a long period of approximately 293 d and a short period of approximately 27 d. Among them, the energy consumption level and industrial structure were the strong driving factors affecting the PM 2.5 concentration in the long period of Shanxi Province. In the short period, it was greatly affected by the change in atmospheric circulation, and different cities were affected by typical meteorological factors. Linfen, Yuncheng, Datong, Shuozhou, and Xinzhou were vulnerable to wind speed; Jinzhong and Luliang were vulnerable to temperature; and Taiyuan, Jincheng, Yangquan, and Changzhi were uniquely and significantly affected by relative humidity. Therefore, industrial structure adjustment and energy structure adjustment are key to the long-term control of atmospheric PM 2.5 and the long-term improvement of air quality in Shanxi Province. The differential impact of different urban meteorological factors on PM 2.5 should be considered when carrying out short-term regional joint prevention and control.

Taking the typical heavy air pollution process in Yangquan from December 26, 2018 to January 20, 2019 as an example, the characteristics and cause analysis of heavy air pollution in a mountainous city in winter were analyzed in this study. The results showed that fine particle mass (PM 2.5 ) was the primary pollutant during the heavy pollution period. The water-soluble ions and carbonaceous components were the main components of PM 2.5 . The secondary ions of SO 4 2- , NO 3 - , and NH 4 + had the lager contribution to water-soluble ions (87.7%), and the secondary organic carbon (SOC) was the main component of the carbonaceous components (71.6%). The concentration of the secondary ions during the heavy pollution period increased by 5.3 times compared to levels before the heavy pollution period, and was an important component resulting in the fast increase of PM 2.5 . An analysis of meteorological conditions showed that PM 2.5 and its main components had a significantly positive relationship with humidity and a significantly negative relationship with wind speed. And that pollution became stronger with an increase in humidity and a decrease in wind speed. The typical meteorological characteristics of mountainous cities are high relative humidity and large temperature variations, which can accelerate the formation of secondary pollutants and are the main reasons for the rapid aggravation of PM 2.5 . In addition, the lower average wind speed caused by the relatively closed terrain in mountainous cities makes the diffusion conditions of air pollutants relatively poor, which is one of the reasons for the accumulation of pollutants. The source apportionment results showed that the secondary sources (46.0%) were the most important source of PM 2.5 , followed by coal combustion (32.6%), vehicle exhaust (19.8%), and fugitive dust (1.6%). Therefore, mountainous cities should pay more attention to controlling secondary components, especially secondary ions.

Volatile organic compounds (VOCs) were collected at three environmental sampling sites in Yangquan and quantified by gas chromatography-mass selective detector/flame ionization detector(GC-MSD/FID). The VOC sources were identified by diagnostic ratios and positive matrix factorization (PMF), and environmental impact of VOCs on O 3 and secondary organic aerosol (SOA) were evaluated. The results showed that the average VOC concentration was (82.1±22.7) μg·m -3 , with alkanes being the most abundant group (51.8%), followed by aromatics (17.8%), alkenes (8.0%), and alkynes (3.8%). The diurnal variation of VOCs exhibited a bimodal trend, with twin peaks appearing at 08:00-10:00 and 18:00-20:00, falling to a valley at 12:00-14:00. The results for benzene/toluene (2.1±1.3) and isopentane/ n -pentane (1.7±0.6) showed that the ambient VOCs may be influenced by coal combustion and vehicular emissions. Six sources were extracted by PMF:coal combustion (34.9%), vehicle emissions (18.2%), gasoline evaporation (15.2%), industrial emissions (13.6%), biogenic emissions (9.2%), and solvent usage (9.0%). The average concentration of ozone formation potential (OFP) was 156.6 μg·m -3 , with the highest contribution from alkenes, while the average concentration of secondary organic aerosol formation potential (SOA p ) was 68.7 μg·m -3 , mainly from aromatics (93.4%). In summary, coal combustion was the most abundant source of VOCs, and accelerating the management of coal gangue and energy structure readjustment are the key points to address. Meanwhile, restricting the VOCs from vehicle emissions, gasoline evaporation, and industrial emissions is also required.

The n-alkanes in PM10 and typical emission sources samples collected during heating and non-heating periods in Taiyuan were determined with GC-MS. Meanwhile, the distribution characteristics and source identification of n-alkanes were investigated with diagnostic parameters and principal component analysis (PCA). Concentrations of n-alkanes ranged from 213.74 to 573.32 ng.m-3 and 22.69 to 150.82 ng.m-3 in the heating and non-heating seasons, respectively. The n-alkanes concentrations in suburban districts including JY, JCP, XD and SL were higher than those in urban sites in the heating quarter, and the relative concentration in JS was 7 times higher than that in SL in the other period. The correlation of the total n-alkanes in PM10 with that derived from fossil fuel was higher than the correlation with those from plant in the heating quarter, while the opposite result was detected in the other period, manifesting higher contribution of fossil fuel in the heating days. CPI and % WNA values showed that the contribution from plant wax in the non-heating period was higher than that in the heating period, and the alkanes production rate was elevated along with the increase in environmental pressures. Information on higher organic matter maturity was obtained during the heating period by Cmax and OEP and the existence of UCM bulge confirmed that vehicles were the significant contributor to n-alkanes concentration during the whole year. PCA analysis indicated the major component was the mixture of vehicle emission and higher plant, accounting for 51.28% of the total variances, followed by coal dust, accounting for 43. 14%. Cooperating control of emissions from coal combustions and vehicles would be the effective way to lower the concentrations of the corresponding n-alkanes.