Chengdu Plain Urban Agglomeration (CPUA) is one of the most serious areas suffering from ozone pollution in China. A comprehensive field observation focused on the ozone production rate and its sensitivity was conducted at CPUA in the summer of 2019. Six sampling sites were set and two ozone pollution episodes were recognized. The daily maximum 8-h average (MDA8) O3 concentration reached 137.9 ppbv in the urban sites during the ozone pollution episode. Peak concentration of O3 was closely related to intense solar radiation, high temperatures, and precursor emissions. The OH-HO2-RO2 radical chemistry and ozone production rate (P(O3)) were calculated using an observation-based model (OBM). The daily peak OH concentration varied in the range of 3-13 × 106 molecules cm−3, and peak HO2 and RO2 were in the range of 2–14 × 108 molecules cm−3 during ozone pollution episodes. During the ozone pollution episode, the average maximum of P(O3) in suburban sites (about 30 ppbv h-1.) was compared with urban sites, and the maximum of P(O3) was 18 ppbv h-1 in rural sites. The relative incremental reactivity (RIR) results demonstrate that it was a VOCs-limited regime in the central urban area of Chengdu, with NOx suppression effect in some regions. In the southern neighboring suburb of Chengdu, it was VOCs-limited as well. However, the northern suburban area was a transition region. In the remote rural areas of the southern CPUA, it was highly NOx-limited. Local ozone production driven by the photochemical process is crucial to the ozone pollution formation in CPUA. The geographically differentiated recognition of the ozone regime found by this study can help to tailor control strategies for local conditions and avoid the negative effects of a one-size-fits-all approach.

Abstract A comprehensive understanding of meteorological, emission and chemical influences on severe haze is essential for air pollution mitigation. However, the nonlinearity of the atmospheric system greatly hinders this understanding. In this study, we developed the quantitative decoupling analysis (QDA) method by applying the Factor Separation (FS) method into the model processes to quantify the effects of emissions (E), meteorology (M), chemical reactions (C), and their nonlinear interactions and impact on fine particulate matter (PM2.5) pollution. Taking a heavy-haze episode in Beijing as an example, we show that different from the integrated process rate (IPR) and the scenario analysis approach (SAA) in previous studies, the QDA method explicitly demonstrate the nonlinear effects by decomposing the variation of PM2.5 concentration into individual contributions of E, M and C terms as well as the contributions from interactions among these processes. Results showed that M dominated the hourly fluctuation of the PM2.5 concentration. The C terms increase with increasing the level of haze, reaching maximum (0.37 μg · \$\mathit\cdot \$ m−3 · \$\mathit\cdot \$ h−1) at the maintenance stage. Moreover, our method reveals that there are non-negligible non-linear effects of meteorological, emission, and chemical processes during pollution stage, with the mean accounting for 50% of the increase in PM2.5 concentrations, which is often ignored in the current air pollution control strategies. This study highlights that the QDA approach can be used to gain insight into the formation of heavy pollution, and to identify uncertainty in numerical models.

In this study, the chemical mechanisms of O3 and nitrate formation as well as the control strategy were investigated based on extensive observations in Tai'an city in the NCP and an observation-constrained box model. The results showed that O3 pollution was severe with the maximum hourly O3 concentration reaching 150 ppb. Higher O3 concentration was typically accompanied by higher PM2.5 concentrations, which could be ascribed to the common precursors of VOCs and NOx. The modeled averaged peak concentrations of OH, HO2, and RO2 were relatively higher compared to previous observations, indicating strong atmospheric oxidation capacity in the study area. The ROx production rate increased from 2.8 ppb h−1 to 5 ppb h−1 from the clean case to the heavily polluted case and was dominated by HONO photolysis, followed by HCHO photolysis. The contribution of radical-self combination to radical termination gradually exceeded NO2 + OH from clean to polluted cases, indicating that O3 formation shifted to a more NOx-limited regime. The O3 production rate increased from 14 ppb h−1 to 22 ppb h−1 from clean to heavily polluted cases. The relative incremental reactivity (RIR) results showed that VOCs and NOx had comparable RIR values during most days, which suggested that decreasing VOCs or NOx was both effective in alleviating O3 pollution. In addition, HCHO, with the largest RIR value, made important contribution to O3 production. The Empirical Kinetic Modeling Approach (EKMA) revealed that synergistic control of O3 and nitrate can be achieved by decreasing both NOx and VOCs emissions (e.g., alkenes) with the ratio of 3:1. This study emphasized the importance of NOx abatement for the synergistic control of O3 and nitrate pollution in the Tai'an area as the sustained emissions control has shifted the O3 and nitrate formation to a more NOx-limited regime.

A near-explicit mechanism, the master chemical mechanism (MCMv3.3.1), coupled with the Community Multiscale Air Quality (CMAQ) model (CMAQ-MCM-SOA), was applied to investigate the characteristics of secondary organic aerosol (SOA) during a pollution event in the Yangtze River Delta (YRD) region in summer 2018. Model performances in predicting explicit volatile organic compounds (VOCs), organic aerosol (OA), secondary organic carbon (SOC), and other related pollutants in Taizhou, as well as ozone (O3) and fine particulate matter (PM2.5) in multiple cities in this region, were evaluated against observations and model predictions by the CMAQ model coupled with a lumped photochemical mechanism (SAPRC07tic, S07). MCM and S07 exhibited similar performances in predicting gaseous species, while MCM better captured the observed PM2.5 and inorganic aerosols. Both models underpredicted OA concentrations. When excluding data during biomass burning events, SOC concentrations were underpredicted by the CMAQ-MCM-SOA model (−28.4 %) and overpredicted by the CMAQ-S07 model (134.4 %), with better agreement with observations in the trend captured by the CMAQ-MCM-SOA model. Dicarbonyl SOA accounted for a significant fraction of total SOA in the YRD, while organic nitrates originating from aromatics were the most abundant species contributing to the SOA formation from gas-particle partitioning. The oxygen-to‑carbon ratio (O/C) for SOA and OA were 0.68–0.75 and 0.20–0.65, respectively, indicating a higher oxidation state in the areas influenced by biogenic emissions. Finally, the phase state of SOA was examined by calculating the glass transition temperature (Tg) based on its molecular composition. It was found that semi-solid state characterized SOA in the YRD, which could potentially impact their chemical transformation and lifetimes along with those of their precursors.

PM2.5 pollution has been greatly alleviated in the Beijing-Tianjin-Hebei (BTH) and surrounding area since the implementation of Action Plans for blue skies. Regional air pollution covering multiple cities has recently become common, but an overview from a regional perspective is scarse. In this work, based on PM2.5 chemical composition data collected in “2 + 26” cities during four consecutive autumn-winter seasons (AWS, from October to next March) from October 2016 to March 2020, we determined the main components driving the increase in PM2.5 mass concentration within different PM2.5 concentration regimes. Regionally, in the low PM2.5 regime (daily concentration ≤ 75 μg/m3), the PM2.5 mass remained organic matter (OM)-driven during the four AWSs. In the high PM2.5 regime (daily concentration > 75 μg/m3), regional PM2.5 increase pattern rapidly transformed from OM-driven during the 2016–2017 AWS to secondary inorganics-driven during the latter three AWSs, with nitrate becoming the dominant component, driving PM2.5 increase in almost 90% cities in the region. These transitions not only reflect the effectiveness of policies to control emissions from coal combustion and biomass burning in recent years, but also highlight the need to further reduce nitrogen oxides emissions from diesel vehicles, non-road mobile machinery and industry. Besides, the control of sulfur dioxide and primary organic matter should not be neglected, given the substantial contribution of PM2.5 sulfate under stagnant and humid meteorological conditions, especially in the geographically central and southern parts of the region, and the need for in-depth air quality improvement if the PM2.5 concentration standards are tightened. Considering the dominance of nitrate in driving regional PM2.5 pollution, more research on nitrate formation and its atmospheric and climatic impacts is warranted.