The COVID-19 lockdowns led to major reductions in air pollutant emissions. Here, we quantitatively evaluate changes in ambient NO₂, O₃, and PM_2.5 concentrations arising from these emission changes in 11 cities globally by applying a deweathering machine learning technique. Sudden decreases in deweathered NO₂ concentrations and increases in O₃ were observed in almost all cities. However, the decline in NO₂ concentrations attributable to the lockdowns was not as large as expected, at reductions of 10 to 50%. Accordingly, O₃ increased by 2 to 30% (except for London), the total gaseous oxidant (O_x = NO₂ + O₃) showed limited change, and PM_2.5 concentrations decreased in most cities studied but increased in London and Paris. Our results demonstrate the need for a sophisticated analysis to quantify air quality impacts of interventions and indicate that true air quality improvements were notably more limited than some earlier reports or observational data suggested.

The Campaign on Atmospheric Aerosol Research network of China (CARE-China) is a long-term project for the study of the spatio-temporal distributions of physical aerosol characteristics as well as the chemical components and optical properties of aerosols over China. This study presents the first long-term data sets from this project, including 3 years of observations of online PM_2.5 mass concentrations (2012–2014) and 1 year of observations of PM_2.5 compositions (2012–2013) from the CARE-China network. The average PM_2.5 concentration at 20 urban sites is 73.2 µg m⁻³ (16.8–126.9 µg m⁻³), which was 3 times higher than the average value from the 12 background sites (11.2–46.5 µg m⁻³). The PM_2.5 concentrations are generally higher in east-central China than in the other parts of the country due to their relatively large particulate matter (PM) emissions and the unfavourable meteorological conditions for pollution dispersion. A distinct seasonal variability in PM_2.5 is observed, with highs in the winter and lows during the summer at urban sites. Inconsistent seasonal trends were observed at the background sites. Bimodal and unimodal diurnal variation patterns were identified at both urban and background sites. The chemical compositions of PM_2.5 were analysed at six paired urban and background sites located within the most polluted urban agglomerations – North China Plain (NCP), Yangtze River delta (YRD), Pearl River delta (PRD), North-east China region (NECR), South-west China region (SWCR) – and the cleanest region of China – the Tibetan Autonomous Region (TAR). The major PM_2.5 constituents across all the urban sites are organic matter (OM, 26.0 %), SO₄²⁻ (17.7 %), mineral dust (11.8 %), NO₃⁻ (9.8 %), NH₄⁺ (6.6 %), elemental carbon (EC) (6.0 %), Cl⁻ (1.2 %) at 45 % RH and unaccounted matter (20.7 %). Similar chemical compositions of PM_2.5 were observed at background sites but were associated with higher fractions of OM (33.2 %) and lower fractions of NO₃⁻ (8.6 %) and EC (4.1 %). Significant variations of the chemical species were observed among the sites. At the urban sites, the OM ranged from 12.6 µg m⁻³ (Lhasa) to 23.3 µg m⁻³ (Shenyang), the SO₄²⁻ ranged from 0.8 µg m⁻³ (Lhasa) to 19.7 µg m⁻³ (Chongqing), the NO₃⁻ ranged from 0.5 µg m⁻³ (Lhasa) to 11.9 µg m⁻³ (Shanghai) and the EC ranged from 1.4 µg m⁻³ (Lhasa) to 7.1 µg m⁻³ (Guangzhou). The PM_2.5 chemical species at the background sites exhibited larger spatial heterogeneities than those at urban sites, suggesting different contributions from regional anthropogenic or natural emissions and from long-range transport to background areas. Notable seasonal variations of PM_2.5-polluted days were observed, especially for the megacities in east-central China, resulting in frequent heavy pollution episodes occurring during the winter. The evolution of the PM_2.5 chemical compositions on polluted days was consistent for the urban and nearby background sites, where the sum of sulfate, nitrate and ammonia typically constituted much higher fractions (31–57 %) of PM_2.5 mass, suggesting fine-particle pollution in the most polluted areas of China assumes a regional tendency, and the importance of addressing the emission reduction of secondary aerosol precursors including SO₂ and NO_x. Furthermore, distinct differences in the evolution of [NO₃⁻]∕[SO₄²⁻] ratio and OC∕EC ratio on polluted days imply that mobile sources and stationary (coal combustion) sources are likely more important in Guangzhou and Shenyang, respectively, whereas in Beijing it is mobile emission and residential sources. As for Chongqing, the higher oxidation capacity than the other three cities suggested it should pay more attention to the emission reduction of secondary aerosol precursors. This analysis reveals the spatial and seasonal variabilities of the urban and background aerosol concentrations on a national scale and provides insights into their sources, processes and lifetimes.

Epidemiological studies have consistently linked exposure to PM_2.5 with adverse health effects. The oxidative potential (OP) of aerosol particles has been widely suggested as a measure of their potential toxicity. Several acellular chemical assays are now readily employed to measure OP; however, uncertainty remains regarding the atmospheric conditions and specific chemical components of PM_2.5 that drive OP. A limited number of studies have simultaneously utilised multiple OP assays with a wide range of concurrent measurements and investigated the seasonality of PM_2.5 OP. In this work, filter samples were collected in winter 2016 and summer 2017 during the atmospheric pollution and human health in a Chinese megacity campaign (APHH-Beijing), and PM_2.5 OP was analysed using four acellular methods: ascorbic acid (AA), dithiothreitol (DTT), 2,7-dichlorofluorescin/hydrogen peroxidase (DCFH) and electron paramagnetic resonance spectroscopy (EPR). Each assay reflects different oxidising properties of PM_2.5, including particle-bound reactive oxygen species (DCFH), superoxide radical production (EPR) and catalytic redox chemistry (DTT/AA), and a combination of these four assays provided a detailed overall picture of the oxidising properties of PM_2.5 at a central site in Beijing. Positive correlations of OP (normalised per volume of air) of all four assays with overall PM_2.5 mass were observed, with stronger correlations in winter compared to summer. In contrast, when OP assay values were normalised for particle mass, days with higher PM_2.5 mass concentrations (µg m⁻³) were found to have lower mass-normalised OP values as measured by AA and DTT. This finding supports that total PM_2.5 mass concentrations alone may not always be the best indicator for particle toxicity. Univariate analysis of OP values and an extensive range of additional measurements, 107 in total, including PM_2.5 composition, gas-phase composition and meteorological data, provided detailed insight into the chemical components and atmospheric processes that determine PM_2.5 OP variability. Multivariate statistical analyses highlighted associations of OP assay responses with varying chemical components in PM_2.5 for both mass- and volume-normalised data. AA and DTT assays were well predicted by a small set of measurements in multiple linear regression (MLR) models and indicated fossil fuel combustion, vehicle emissions and biogenic secondary organic aerosol (SOA) as influential particle sources in the assay response. Mass MLR models of OP associated with compositional source profiles predicted OP almost as well as volume MLR models, illustrating the influence of mass composition on both particle-level OP and total volume OP. Univariate and multivariate analysis showed that different assays cover different chemical spaces, and through comparison of mass- and volume-normalised data we demonstrate that mass-normalised OP provides a more nuanced picture of compositional drivers and sources of OP compared to volume-normalised analysis. This study constitutes one of the most extensive and comprehensive composition datasets currently available and provides a unique opportunity to explore chemical variations in PM_2.5 and how they affect both PM_2.5 OP and the concentrations of particle-bound reactive oxygen species.

This study presents the source apportionment of PM_2.5 performed by positive matrix factorization (PMF) on data presented here which were collected at urban (Institute of Atmospheric Physics – IAP) and rural (Pinggu – PG) sites in Beijing as part of the Atmospheric Pollution and Human Health in a Chinese megacity (APHH-Beijing) field campaigns. The campaigns were carried out from 9 November to 11 December 2016 and from 22 May to 24 June 2017. The PMF analysis included both organic and inorganic species, and a seven-factor output provided the most reasonable solution for the PM_2.5 source apportionment. These factors are interpreted as traffic emissions, biomass burning, road dust, soil dust, coal combustion, oil combustion, and secondary inorganics. Major contributors to PM_2.5 mass were secondary inorganics (IAP: 22 %; PG: 24 %), biomass burning (IAP: 36 %; PG: 30 %), and coal combustion (IAP: 20 %; PG: 21 %) sources during the winter period at both sites. Secondary inorganics (48 %), road dust (20 %), and coal combustion (17 %) showed the highest contribution during summer at PG, while PM_2.5 particles were mainly composed of soil dust (35 %) and secondary inorganics (40 %) at IAP. Despite this, factors that were resolved based on metal signatures were not fully resolved and indicate a mixing of two or more sources. PMF results were also compared with sources resolved from another receptor model (i.e. chemical mass balance – CMB) and PMF performed on other measurements (i.e. online and offline aerosol mass spectrometry, AMS) and showed good agreement for some but not all sources. The biomass burning factor in PMF may contain aged aerosols as a good correlation was observed between biomass burning and oxygenated fractions (r²= 0.6–0.7) from AMS. The PMF failed to resolve some sources identified by the CMB and AMS and appears to overestimate the dust sources. A comparison with earlier PMF source apportionment studies from the Beijing area highlights the very divergent findings from application of this method.

Fine particles were sampled from 9 November to 11 December 2016 and 22 May to 24 June 2017 as part of the Atmospheric Pollution and Human Health in a Chinese Megacity (APHH-China) field campaigns in urban Beijing, China. Inorganic ions, trace elements, organic carbon (OC), elemental carbon (EC), and organic compounds, including biomarkers, hopanes, polycyclic aromatic hydrocarbons (PAHs), n-alkanes, and fatty acids, were determined for source apportionment in this study. Carbonaceous components contributed on average 47.2 % and 35.2 % of total reconstructed PM_2.5 during the winter and summer campaigns, respectively. Secondary inorganic ions (sulfate, nitrate, ammonium; SNA) accounted for 35.0 % and 45.2 % of total PM_2.5 in winter and summer. Other components including inorganic ions (K⁺, Na⁺, Cl⁻), geological minerals, and trace metals only contributed 13.2 % and 12.4 % of PM_2.5 during the winter and summer campaigns. Fine OC was explained by seven primary sources (industrial and residential coal burning, biomass burning, gasoline and diesel vehicles, cooking, and vegetative detritus) based on a chemical mass balance (CMB) receptor model. It explained an average of 75.7 % and 56.1 % of fine OC in winter and summer, respectively. Other (unexplained) OC was compared with the secondary OC (SOC) estimated by the EC-tracer method, with correlation coefficients (R²) of 0.58 and 0.73 and slopes of 1.16 and 0.80 in winter and summer, respectively. This suggests that the unexplained OC by the CMB model was mostly associated with SOC. PM_2.5 apportioned by the CMB model showed that the SNA and secondary organic matter were the two highest contributors to PM_2.5. After these, coal combustion and biomass burning were also significant sources of PM_2.5 in winter. The CMB results were also compared with results from the positive matrix factorization (PMF) analysis of co-located aerosol mass spectrometer (AMS) data. The CMB model was found to resolve more primary organic aerosol (OA) sources than AMS-PMF, but the latter could apportion secondary OA sources. The AMS-PMF results for major components, such as coal combustion OC and oxidized OC, correlated well with the results from the CMB model. However, discrepancies and poor agreements were found for other OC sources, such as biomass burning and cooking, some of which were not identified in AMS-PMF factors.