Atmospheric Pollution Research

Publications
Seasonal AhR activity expressed as TCDD Eq BIO (pg m -3 ) for the 50% effect level (24 
Linear regression between the PAH exposure marker Σ B[a]P TEQ (ng m -3 ) and AhR activity (TCDD Eq BIO , pg m -3 ), as determined from passive sampling at urban capitals, regional 
The decline in the CAFLUX derived TCDD equivalent air concentrations (pg m -3 ) at 
There has been relatively little bioanalytical effect based monitoring conducted using samples derived from polyurethane foam (PUF) passive air samplers. Combining these techniques may provide a more convenient and cost effective means of monitoring the potential for biological effects resulting from exposure to complex mixtures in a range of scenarios. Seasonal polycyclic aromatic hydrocarbon (PAH) levels were monitored at sites around Australia using direct chemical analysis. In addition, both indirect acting genotoxicity (umuC assay) and aryl hydrocarbon receptor (AhR) activity (chemically activated fluorescent gene expression [CAFLUX assay]), which are effects potentially relevant to subsequent carcinogenesis for these compounds, were measured. The levels of PAHs as well as genotoxicity and AhR activity were all higher in winter compared to summer and for sites in urban capital cities compared to other locations. Statistically significant relationships were found between the levels of PAHs and both genotoxicity and AhR activity. The dominant contributors to the total AhR activity, were found to be for compounds which are not resistant to H(2)SO(4)/silica gel treatment and were relatively rapidly metabolised that is consistent with a PAH type response. Relative potency estimates for individual PAHs determined for the first time on the CAFLUX assay were used to estimate the proportion of total AhR activity (≤ 3.0%) accounted by PAHs monitored. Observed responses are thus largely due to non-quantified AhR active compounds.
 
Location of sampling sites (stars) and major industrial sources (circles) in Jeddah, Saudi Arabia. 
This paper presents the first comprehensive investigation of PM2.5 and PM10 composition and sources in Saudi Arabia. We conducted a multi-week multiple sites sampling campaign in Jeddah between June and September, 2011, and analyzed samples by XRF. The overall mean mass concentration was 28.4 ± 25.4 μg/m(3) for PM2.5 and 87.3 ± 47.3 μg/m(3) for PM10, with significant temporal and spatial variability. The average ratio of PM2.5/PM10 was 0.33. Chemical composition data were modeled using factor analysis with varimax orthogonal rotation to determine five and four particle source categories contributing significant amount of for PM2.5 and PM10 mass, respectively. In both PM2.5 and PM10 sources were (1) heavy oil combustion characterized by high Ni and V; (2) resuspended soil characterized by high concentrations of Ca, Fe, Al, and Si; and (3) marine aerosol. The two other sources in PM2.5 were (4) Cu/Zn source; (5) traffic source identified by presence of Pb, Br, and Se; while in PM10 it was a mixed industrial source. To estimate the mass contributions of each individual source category, the CAPs mass concentration was regressed against the factor scores. Cumulatively, resuspended soil and oil combustion contributed 77 and 82% mass of PM2.5 and PM10, respectively.
 
One of the aims of the HOPE project is to develop a method to assess healthiness and energy efficiency of buildings and strategies to optimise their performance towards health and energy efficiency. In this project, a building is defined "Healthy and Energy Efficient" if: it does not cause or aggravate illnesses of the building occupants; it assures a high level of comfort for the building occupants; it minimises the use of non-renewable energy taking into account available technology. Health and comfort performance criteria for buildings have been defined. They include a set of measurable parameters related to indoor air pollutants or physical characteristics of the indoor environment. Compliance with this set is expected to assure, with a high degree of confidence, the provision of acceptable performance of buildings and zones within them. Target values of the selected parameters have been set taking as reference the WHO air quality guidelines, when available. Health classification of buildings is based on health risk analysis from checklist data, if necessary integrated with measurements, and health- and comfort-related questionnaire data. To this purpose, health hazards have been divided into 3 groups and acute building-related symptoms are evaluated through the Building Symptom Index (BSI), a numeric indicator that considers the frequency of symptoms related to the Sick Building Syndrome (SBS) perceived by the occupants. In the end, the building is ranked according to the health status in one of 3 classes. Energy efficiency is evaluated by means of the energy index, i.e. the yearly total energy use per gross heated floor area, and the building is ranked accordingly. Combining the health and energy evaluation, the buildings are finally classified as optimal, medium, or low.
 
Location of the interest area and topography therein. Stars, diamonds, and the dot indicate the surface meteorological sites, PM2.5 sites and a MET-tower. The polygon marks the NAA.
HYSPLIT backward meteorological trajectory simulations for March 8 2009 from DP at three heights for each simulation.  
NME and NMB as function of average concentration. The solid (dashed) lines indicate the performance goals (criteria).  
NMB as function of average speciation concentration. The solid (dashed) lines indicate the performance goals (criteria).  
Weather Research and Forecasting model inline coupled with a chemistry package PM2.5 forecasts were assessed using fixed-site PM2.5 concentration and specification, and mobile PM2.5 concentration and temperature measurements from the Fairbanks winter 2008/09 field campaign. Performance differs with concentrations, varies among months and sites, and best results are achieved for PM2.5 concentrations between 15 and 50 µg/m³. On average over half-a-year and all sites, 24 h-average PM2.5 concentrations have a fractional bias and error, and a normalized mean bias and error of 22%, 67%, 13% and 71%, respectively. The skill scores derived from the mobile measurements indicate that high data density increases the representativeness of the observations and enhances the evaluation of spatial details. The model performed well for organic carbon and acceptably for sulfate, but underestimated ammonium significantly. PM2.5 concentrations measured by two different devices at the same site indicate that measurement errors at extremely low temperatures and humidities explain up to 24% of the normalized mean error. Some discrepancies can be attributed clearly to errors in emissions, chemical boundary conditions and meteorology.
 
This paper focuses on the problem of poor air quality in Jing-Jin-Ji region of China. According to the data of the report on the state of the environment in Jing-Jin-Ji region in 2013–2016, the fractional order accumulation GM(1,1) model was adopted to predict the number of the lightly polluted day and the annual average concentration of PM 2.5 in the Jing-Jin-Ji region in 2020. The forecasting results show that it will be difficult for the Jing-Jin-Ji region to achieve the goal of more than 80% lightly polluted day in 2020. The Jing-Jin-Ji region should adjust measures according to the local conditions, to achieve a significant improvement of air quality.
 
Laser desorption/ionization time-of-flight mass spectra and size of 4 individual SOA particles after 1 hours photooxidation (Aerosol diameter: (a): 0.702 μm, (b): 0.989 μm, (c): 1.774 μm, (d): 0.856 μm). 
A laboratory study was performed to investigate the composition of products formed from OH-initiated oxidation of aromatic hydrocarbon 1,3,5-trimethylbenzene. The experiments were conducted by irradiating 1,3,5-trimethylbenzene/CH3ONO/NO/air mixtures in smog chamber. The chemical composition of gas and particle-phase products have been investigated with gas chromatography/mass spectrometry (GC/MS) and the aerosol laser time-of-flight mass spectrometer (ALTOFMS), respectively. Experimental results showed that 3,5-dimethyl benzaldehyde, 2,4,6-trimethylphenol, 2-methyl-4-oxo-2-pentenal and 3,5-dimethyl-2-furanone were the predominant products in both the gas and particle-phases. However, there were some differences between detected gas-phase products and those of particle-phase, for example, oxalic acid, 2-methyl-2-hydroxy-3,4-dioxo-pentanal, and 2,3,5-trimethyl-3-nitrophenol were only existing in the particle-phase. The possible reaction mechanisms leading to these products are also proposed. Compared to offline methods such as GC-MS measurement, the ALTOFMS detection can analyze real-time the secondary organic aerosol (SOA) successfully and provide more information on the products. Thus, ALTOFMS is a useful tool to reveal the formation and transformation processes of SOA particles in smog chambers.
 
Ammonia (NH3) gas-aging of secondary organic aerosol (SOA) results in the formation of organonitrogen compound is an important class of brown carbon. The particulate products of aged 1,3,5-trimethylbenzene (135-TMB) SOA in the presence of NH3 were measured by UV-Vis spectrophotometer, attenuated total reflectance-Fourier transform infrared (ATR-FTIR), and aerosol laser time-of-flight mass spectrometer (ALTOFMS) in present study. Experimental results indicated that NH3 has significant promotion effect on 135-TMB SOA formation. Organic ammonium salts, such as ammonium methyl glyoxylate, ammonium 3,5-dimethylbenzoate, which are formed from NH3 reactions with gaseous organic acids were detected as the principal particulate products of NH3-aged 135-TMB SOA. 4-methyl-imidazole-2-acetaldehyde, 4-methyl- 1H-imidazole and other imidazole products via the heterogeneous reactions between NH3 and dialdehydes of 135-TMB SOA were newly measured. The formation of imidazole products suggests that some ambient particles contained organonitrogen compounds may be come from this mechanism. The results of this study may provide valuable information for discussing SOA aging mechanisms and new route for NH3 deposition. © 2017 Turkish National Committee for Air Pollution Research and Control.
 
Figure7ProposedagingmechanismofketoaldehydeproductsfromOH–initiatedoxidationof135–TMB. 
Secondary organic aerosol (SOA) from the photooxidation of aromatic compounds is a very complex mixture containing products with a different chemical nature that are dependent on aging processes. In this study, we focus on the chemical characterization of major products that are formed from the OH-initiated oxidation of 1,3,5-trimethylbenzene and subsequent aging through OH-initiated reactions in the presence of NOX. The chemical composition of aged particles were measured in real-time by an aerosol laser time of flight mass spectrometer (ALTOFMS) coupled with Fuzzy C-Means (FCM) clustering algorithm. Experimental results demonstrated that methyl glyoxylic acid, 2-methyl-4-oxo-2-pentenoic acid, 3,5-dimethylbenzoic acid, 2-methyl-2,3-dihydroxyl-4-oxopentanoic acid, dimethyl-nitrophenol, 3,4-dimethyl-2-hydroxy-3-oxo-pentandioic acid, 2,4-dimethyl-2,3-dihydroxy-6-oxo-4-heptenoic acid, 2,4-dimethyl-4-hydroxy-2,3-epoxy-heptylic acid, 2,4-dimethyl-2,3,4-trihydroxy-5,6-dioxo-heptylic acid, and oligomer components were the predominant products in the aging particles. The possible reaction mechanisms leading to these aged products were also discussed and proposed.
 
This review summarizes up-to-date scientific literature concerning unintentionally produced polychlorinated biphenyls (UP-PCBs), including information on their known or suspected formation pathways, occurrence in air globally, and properties relating to atmospheric persistence and transport. Prior to the listing of PCBs as an original “dirty dozen” persistent organic pollutant (POP) under the Stockholm Convention on Persistent Organic Pollutants, they were already widely regulated, and some monitoring in air was occurring due to environmental and health concerns. So far, the focus of monitoring has been for dominant congeners found in technical PCB formulations, such as Aroclors. However, recent research has shown that processes such as dye/pigment manufacturing, and industrial thermal processes have resulted in UP-PCB emission and detection globally. It is especially concerning that UP-PCBs make up a significant proportion of ∑PCBs (typically from a few percent to as high as 85%), and this contribution continues to increase. Among identified UP-PCBs, PCB-11 is the dominant congener detected in air. Three key recommendations from this review include i.) to include UP-PCBs, such as PCB-11 and PCB-209, as indicator congeners in air monitoring and research programs; ii.) to apply PUF disk passive air samplers as simple and cost-effective tools for generating new information on global air; and iii.) to identify and quantify the ongoing emission sources of UP-PCBs to air. The new information will raise awareness to the growing problem of UP-PCBs and could inform science and policy strategies for assessing and managing this emerging class of chemicals.
 
The main aim of this work is characterization of atmospheric aerosol using 11 stage cascade impactor. The first investigation of size-segregated sub-urban aerosols from the continental part of the Balkan peninsula in 11 fractions in the range of 0.0085 < Dp < 16 μm was performed from March 2012 to December 2013. Aerosols were measured at the Zeleno Brdo observatory (ϕ = 44°48’; λ = 20°28′–243 m above sea level), the highest landmark on the eastern side of Belgrade. Zeleno Brdo is surrounded by wooded vegetation and comprises of both southern facing rural and north-west orientated urban areas. About 70% of total aerosols are fine particles, belonging especially to the PM0.53–1.06 fraction which is found to be more pronounced in winter period. In this work, we applied tests of probability function models for three distributions: normal, log-normal and three-parameter Weibull, by comparing expected and observed values. We found that these models offer the possibility to determine whether the dominant emission source was the vicinity or distance of the measuring point. Results of this test could be a significant supplement to existing multivariate mathematical models for source apportionment, providing accurate estimation of the origin of emission sources and offering information on their position relative to the investigated area (local, regional or remote). In addition, the dependence of particle concentrations for each fraction investigated versus meteorological parameters was determined.
 
In this paper, we present validation of the cloud aerosol lidar with orthogonal polarization (CALIOP) level 1 and level 2 tropospheric aerosols products in coincidence to Micro Pulse Lidar (MPL) observations under different sky conditions and different seasons during March 2016–December 2018 over a tropical coastal station Kattankulathur (12.83oN, 80.04oE). In total, 33 simultaneous profiles during clear-sky (18 cases) and cloudy (15 cases; 7 mid-tropospheric (MT) clouds and 8 cirrus clouds cases) conditions are observed, out of which 5, 13, 6, and 9 cases are during winter, pre-monsoon, southwest (SW) monsoon and northeast (NE) monsoon seasons, respectively. CALIOP underestimates MPL within the boundary layer and overestimates in the free troposphere during both the clear-sky and cirrus conditions. In the clear-sky conditions, the mean bias of the level 1 aerosol products is found to be −14 ± 29% (18 ± 19%) below 3.0 km (over 3–10 km) while it is −39 ± 26% (33 ± 24%) below 3.0 km (over 3–8 km) underlying the cirrus clouds. The comparison underlying the thick MT clouds is not reliable. However, a better comparison is observed in the cases of thin MT clouds. A strong seasonal variation in the vertical distribution of the aerosol loadings near the surface with maximum (minimum) loading ∼0.01 ± 0.005 km⁻¹sr⁻¹ (∼0.004 ± 0.002 km⁻¹sr⁻¹) is observed during the winter (SW monsoon) season. The mean bias of CALIOP level 2 aerosol profiles over the altitude 0.3–5 km is found to be 14 ± 28%, 15 ± 44%, −14 ± 25%, and 24 ± 34% during the winter, pre-monsoon, SW monsoon, and NE monsoon seasons, respectively.
 
The widely used dispersion modelling system CALMET/CALPUFF has been applied in order to evaluate its ability to simulate dry and wet depositions at regional scales (up to 1000 km from a source) in the specific case of radionuclides released in the atmosphere, during the 1986 Chernobyl Nuclear Power Plant accident. The 137Cs cumulative deposition data sampled at 410 sites on the entire territory of Ukraine after the accident have been used for model verification. As meteorological input for feeding the CALMET pre-processor, we used a dataset of time series recorded in 211 surface stations, 194 precipitation stations and 14 upper air stations. Two different schemes for the emissions source have been adopted both available from scientific literature on pollutants release during the Chernobyl accident. This work shows that the CALMET/CALPUFF system is able to reproduce the large-scale features of the measured 137Cs deposition pattern, which are the main traces on the territory of Ukraine. However, the fine structure of depositions, which are mainly due to precipitations, are poorly caught. The simulated deposition pattern appears excessively smoothed and an explanation for that is provided. Besides, we have found that the resistant model for dry deposition velocity of 137Cs aerosol particles significantly underestimates depositions and the closest agreement with measurements is achieved with constant deposition velocity of 0.005 m/s. Finally, the strong dependence of the simulated contamination pattern on the emission source parameterization is confirmed.
 
In this work we report the results of HYSPLIT numerical simulations of the Chernobyl Cs-137 atmospheric transport, dispersion, and deposition on the regional scale (~1000 km from the source) with the main focus on the analysis of the deposition processes. In the simulations we used three different gridded datasets of input meteorology and three release scenarios previously published in the literature. The resistance method and the predefined constant value of the deposition velocity (0.005 m/s) were applied to calculate dry depositions whereas an approach based on a scavenging coefficient was used for both wet in-cloud and below-cloud removals. The results were statistically evaluated against the measurements of Cs-137 total depositions on the territory of Ukraine. Our simulations show considerable dependence of the HYSPLIT-predicted accumulated deposition pattern on both the input meteorology and source parameterizations. The best performance of HYSPLIT was obtained with the ERA Interim reanalysis data and the source model of Talerko (2005) and the constant deposition velocity. This simulation reproduced fairly well the spatial structure of the Cs-137 contamination on the territory of Ukraine with good evaluation statistics. However, not all significant local maxima of the contamination pattern were captured clearly. Our simulations also show that dry removal processes account for approximately 50% of the total depositions in Ukraine. Both wet in-cloud and below-cloud removal mechanisms had roughly equal influence on the total amount of Cs-137 radionuclides deposited on the territory of Ukraine.
 
In this study, the carbonaceous aerosols were separated into organic carbon (OC), water soluble OC (WSOC), water soluble humic-like substances (HULIS), water soluble non-HULIS, water insoluble OC (WINSOC), and soot carbon (SC). Then the abundance and ¹⁴C content of carbon fractions were determined to assess the contribution of contemporary versus fossil sources. The results showed that the OC fraction is the major carbon fraction, the OC sub-fractions such as WSOC, HULIS, non-HULIS, and WINSOC accounted for 39.5%–66.7%, 24.8%–43.2%, 11.9%–23.5%, and 31.3%–55.2% of total carbon (TC), respectively. The ¹⁴C data show that the contribution of contemporary carbon (fc) to TC is 38.3%–63.5%, all carbon fractions were mixtures of carbon derived from contemporary and fossil sources. The fc values of OC (38.8%–63.9%) more than that (25.7%–46.6%) of SC in the same sample, indicated OC generally contained more contemporary carbon than SC. The WSOC and HULIS had relatively higher fc levels than the corresponding WINSOC and the fc values of WINSOC all less than 45%, indicated that the insoluble organic fractions in PM2.5 samples were dominated by fossil fuel related sources. These results suggested that the contemporary and fossil contributions for different carbon fractions are different. In addition, the seasonal and spatial variations of contemporary and fossil contributions to carbon fractions indicated that the PM2.5 samples collected during summer at background site (Maofengshan, MFS) and winter at urban site (Wushan, WS) have relative high contribution from contemporary sources, suggested the differences of primary and secondary sources at the two sites.
 
Air pollution is a major threat to the health of the inhabitants of most large cities in the world including Delhi, the second most populated city. Delhi with 25 million people is among the most polluted. Meteorology as well as increasing anthropogenic emissions play important roles in Delhi's reduced air quality. Winter is marked by thick fog, low wind speeds, and low boundary layer heights that worsen air quality. In such an environment, this study monitored PM2.5, PM10, NOX (NO), CO, and O3. To improve air quality, the Delhi Government implemented an Odd-Even (OE) Scheme based on license plate numbers to reduce traffic emissions. The improvements in air quality were not obvious due to the variability in the meteorological conditions during that period. Measured concentrations were compared among the Pre-OE, During-OE, and Post-OE periods and only a limited number of statistically significant differences were observed when all three periods were compared on a 24-hour or hourly bases. Passenger cars represented only a small fraction of the on-road traffic and exceptions were granted for many vehicles. For the limited duration of this trial, variations in dispersion conditions drove much of the observed period-to-period variability. The effect of the vehicle removal strategy to improve air quality was unobservable. © 2018 Turkish National Committee for Air Pollution Research and Control
 
Over the past few years, burning agricultural or crop residues has resulted in serious air pollution, particularly in northern Thailand. In this study, constant endeavours were made to achieve the first-hand emission factors (EFs) of water-soluble ionic species (WSIS), organic carbon, elemental carbon, and selected metals emitted from the combustion of 17 plant species. The average EF values of organic carbon (OC) and elemental carbon (EC) ranged from 280 g kg⁻¹ to 1441 g kg⁻¹ and from 32 g kg⁻¹ to 142 g kg⁻¹, respectively. Sugarcane burning shows the highest EFs of toxic metals, such as V, Cr, Co, Ni, and Pb, which is concerning because Thailand is the world's second largest sugar exporter. The comparatively high percentage contribution of Cl⁻ coupled with extremely low levels of K⁺ resulted in questions on its reliability as a chemical tracer for maritime aerosols and biomass burning particles, respectively. OC/EC ratios varied from 4.68 (Leucaena leucocephala) to 11.4 (Terminalia catappa) with the average value of 8.88 ± 1.72. Further attempts to investigate the most suitable diagnostic binary ratios of metals representing biomass burning were conducted using 2D plots of Zn/Mn, Fe/Zn, Fe/Cu, Ni/Co, Mn/Cr, Ni/V, Al/Si, and Pb/As. In this study, Mn/Cr versus Ni/V and Pb/As versus Al/Si can be used as diagnostic binary ratios to classify biomass-burning-related aerosols.
 
With the spread of the COVID-19 virus globally, cities worldwide have implemented unprecedented social distancing policies to mitigate infection rates. Many studies have demonstrated that improved air quality and reduced carbon emissions have resulted from the COVID-19 pandemic. Yet, questions remain regarding changes in atmospheric CO2 concentrations because of the complex cycles involving the interaction of CO2 with the natural environment. In this study, we compared the changes in urban CO2 enhancement (△CO2) reflecting the contribution of local CO2 emissions to the atmospheric CO2 in urban areas, according to the intensity of social distancing policies implemented during the COVID-19 pandemic in Seoul, South Korea. We used data from three CO2 ground observation sites in the central area of Seoul and outside the urban area of Seoul. By comparing the urban CO2 concentration in Seoul with that of the background area using two different methods, considering both vertical and horizontal differences in CO2 concentration, we quantified the △CO2 of the pre-COVID-19 period and two COVID-19 periods, during which intensive social distancing policies with different intensities were implemented (Level 1, Level 2.5). During the pre-COVID-19 period, the average △CO2 calculated using the two methods was 24.82 ppm, and it decreased significantly to 16.42 and 14.36 ppm during the Level 1 and Level 2.5 periods, respectively. In addition, the urban contribution of Seoul to atmospheric CO2 concentration decreased from 5.27% during the pre-COVID-19 period to 3.54% and 3.19% during the Level 1 and Level 2.5 periods, respectively. The results indicate that the social distancing policies implemented in Seoul resulted in reduced local CO2 emissions, leading to a reduction in atmospheric CO2 concentration. Interestingly, it also shows that the extent of atmospheric CO2 concentration reduction can be greatly affected by the intensity of policies. Our study suggests that changes in human activity could reduce the urban direct contribution to the background CO2 concentration helping to further mitigate climate change.
 
Changes in primary emissions due to the COVID-19 lockdowns in Europe for the year 2020 have been estimated by considering fully open-access and near-real-time measured activity data from a wide range of information sources and with simple computational techniques. The estimates consist on a dataset of reduction factors that are both time- and country-dependent and provided for the following source categories: energy industry (power plants), manufacturing industry, road traffic, aviation, shipping and other stationary combustion activities such as residential and commercial-institutional activities. Inspired in other authors’ estimates for COVID reductions, the advantage of this methodology is that there is no use of machine learning, making this procedure more accessible to the general scientific community. We have followed a fast methodology that takes advantage of observed relationships between variables (e.g. temperature and energy demand) without needing special algorithms for finding those relationships. The comparison of our estimates with others from other authors indicate a reasonable agreement and pointing out that emissions dropped by a 17% on average in Europe, with large differences between sectors of activities and spatial heterogeneity. The most affected sector was aviation, with a spatial-averaged variation of −63% in emissions since the implementation of first restrictions with respect business-as-usual values. 2020 emission changes with respect to business-as-usual values in countries ranges from a −13% in Norway and Poland to a more than −20% in several Mediterranean countries as well as the United Kingdom. Two main periods of emission reductions have been identified.
 
Comparing predicted and measured PM 10 , PM 2.5 and TPN concentrations for the lockdown period at Marylebone Road.
Difference in Nanoparticle numbers (particles/cm 3 ) between 2019 and 2020 during the COVID-19 lockdown.
Comparison of fitted and cross-validated models for predicting PM 10 , PM 2.5 and TPN.
The main aim of the COVID-19 lockdown was to curtail the person-to-person transmission of COVID-19. However, it also acted as an air quality intervention. The effect of the lockdown has been extensively analysed on NO2, O3, PM10 and PM2.5, however, little has been done on how total (TPN) and nanoparticle numbers (NPN) have been affected by the lockdown. This paper quantifies the effect of the lockdown on TPN and NPN in the UK, and compares how the effect varies between rural, urban background and traffic sites. Furthermore, the effect on particle numbers is compared with particle mass concentrations, mainly PM10 and PM2.5. Two approaches are used: (a) comparing measured levels of the pollutants in 2019 with 2020 during the lockdown periods; and (b) comparing the predictions of machine learning with measured concentrations using business as usual (BAU) scenario during the lockdown period. P100 (particle size ≤100 nm) increased by 39% at Chilbolton Observatory (CHO) and decreased by 13% and 14% at London Honor Oak Park (LHO) and London Marylebone Road (LMR), respectively. Particles from 101 to 200 nm (P200) showed a similar trend to P100, however, average levels of particles 201–605 nm (P605) decreased at all sites. TPN, PM10 and PM2.5 concentrations decreased at LMR and LHO sites. Estimated PM10, PM2.5 and TPN decreased at all three sites, however, the amount of change varied from site to site. Pollutant concentrations increased back the to pre-pandemic levels, suggesting more sustainable interventions for permanent air quality improvement.
 
The study focuses on the variability in aerosols, air-pollutants, and associated meteorological characteristics over peninsular India and neighboring ocean regions during both lockdown (25 March-31 May) and unlock (June–September) periods in 2020. During first lockdown phase, 40–60% reduction in aerosol loading observed over most parts except central-west India and the southern BOB. HYSPLIT model-based ‘back trajectory’ suggests the increase in anthropogenic aerosols over BOB due to long-range transport from an elevated layer. During second lockdown phase, an increase in AOD found over the eastern parts of India might come from the continuous operation of coal-fired thermal power plants and coal mines, besides domestic emissions. A surge in sea salt aerosols over BOB during the last two lockdown phases could be due to the Amphan super cyclone. During unlock phases, only natural aerosols show significant variation over the coastal and neighboring ocean regions with the simultaneous presence of intense summer monsoon. However, anthropogenic aerosols increased over the inland areas, which is most prominent in second and fourth unlock phases and could be attributed to the long-range transport. Temporal analysis of particulate matters and gaseous pollutants over coastal cities suggests Ahmedabad to be the most polluted one. The highest fall in NO2 was observed over Mumbai, whereas O3 concentration was appreciably enhanced over Ahmedabad, Visakhapatnam, and Chennai. Besides low anthropogenic emissions, long-range transport and prevailing meteorological conditions also played a significant role in governing the observed changes in aerosol loading and air pollutant concentration.
 
A regional air quality model system (RAQMS) driven by the Weather Research and Forecasting model (WRF) is applied to investigate the distribution and evolution of mineral dust and anthropogenic aerosols over China in April 2020, when air quality was improved due to reduced human activity during the COVID-19 epidemic, whereas dust storms began to attack China and deteriorated air quality. A dust deflation model was developed and improved mineral dust prediction. Model validation demonstrated that RAQMS is able to reproduce PM10, PM2.5 and aerosol components reasonably well. China suffered from three dust events in April 2020, with the maximum hourly PM10 concentrations exceeding 700 μg m⁻³ in downwind cities over the North China Plain (NCP). Mineral dust dominated PM10 mass (>80%) over the Gobi deserts in north and west China, while it comprised approximately 30–50% of PM10 over wide areas of east China. The domain and monthly mean dust mass fractions in PM10 were estimated to be 47% and 43% over the North China Plain and east China, respectively. On average, mineral dust contributed up to 22% and 21% of PM2.5 mass over the North China Plain and east China in April 2020, respectively. Sulfate and nitrate produced by heterogeneous chemical reactions on dust surface accounted for approximately 9% and 13% of secondary inorganic aerosols (SIA) concentration over the North China Plain and east China, respectively. The results from this study demonstrated that mineral dust made an important contribution to particulate matter mass during the COVID-19 epidemic in spring 2020 over China.
 
a) J NO2 mean diurnal variation for January (dotted gray), February (dashed blue), 1-13 804 March (light blue and diamonds), 14-31 March (dashed red), and April (dashed green and stars. b) 805 Mean J NO2 under sunny conditions before (Jan-13 March, blue line) and after (14 March-April, red 806 line) the COVID-19 lockdown. 807 808 809 810 811 812
The COVID-19 lockdown presented a peculiar opportunity to study a shift in the photochemical regime of ozone production in Quito (Ecuador) before and after mobility restrictions. Primary precursors such as NO and CO dropped dramatically as early as 13 March 2020, due to school closures, but ambient ozone did not change. In this work we use a chemical box model in order to estimate regimes of ozone production before and after the lockdown. We constrain the model with observations in Quito (ozone, NOx, CO, and meteorology) and with estimations of traffic-associated VOCs that are tightly linked to CO. To this end, we use the closest observational data of VOC/CO ratios at an urban area that shares with Quito conditions of high altitude and is located in the tropics, namely Mexico City. A shift in the chemical regime after mobility restrictions was evaluated in light of the magnitude of radical losses to nitric acid and to hydrogen peroxide. With reduced NOx in the morning rush hour (lockdown conditions), ozone production rates at 08:30–10:30 increased from 4.2 to 17 to 9.7–23 ppbv h⁻¹, respectively. To test further the observed shift in chemical regime, ozone production was recalculated with post-lockdown NOx levels, but setting VOCs to pre-lockdown conditions. This change tripled ozone production rates in the mid-morning and stayed higher throughout the day. In light of these findings, practical scenarios that present the potential for ozone accumulation in the ambient air are discussed.
 
The diverse climate types and the complex anthropogenic source emissions in China lead to the great regional differences of air pollution mechanisms. The COVID-19 lockdown has given us a precious opportunity to understand the effect of weather conditions and anthropogenic sources on the distribution of air pollutants in different climate zones. In this study, to understand the impact of meteorological and socio-economic factors on air pollution during COVID-19 lockdown, we divided 358 Chinese cities into eight climate regions. Temporal, spatial and diurnal variations of six major air pollutants from January 1 to April 18, 2020 were analyzed. The differences in the characteristics of air pollutants in different climate zones were obvious. PM2.5 reduced by 59.0%–64.2% in cold regions (North-East China (NEC) and North-Western (NW)), while O3 surged by 99.0%–99.9% in warm regions (Central South (CS) and Southern Coast (SC)). Diurnal variations of atmospheric pollutants were also more prominent in cold regions. Moreover, PM2.5, PM10, CO and SO2 showed more prominent reductions (20.5%–64.2%) in heating regions (NEC, NW, NCP and MG) than no-heating regions (0.8%–48%). Climate has less influence on NO2, which dropped by 41.2%–57.1% countrywide during the lockdown. The influences of weather conditions on the atmospheric pollutants in different climate zones were different. The wind speed was not the primary reason for the differences in air pollutants in different climate zones. Temperature, precipitation, and air pollution emissions led to prominent regional differences in air pollutants throughout the eight climates. The effect of temperature on PM, SO2, CO, and NO2 varied obviously with the latitude, at which condition temperature was negatively correlated to PM, SO2, CO, and NO2 in the north but positively in the south. The temperature was positively correlated to ozone in different climate zones, and the correlation was the highest in NEC and the lowest in SC. The rainfall has a strong removal effect on atmospheric pollutants in the climate regions with more precipitation, but it increases the pollutant concentrations in the climate regions with less precipitation. In regions with more emission sources, air pollutants experienced more significant variations and returned to pre-lockdown levels earlier.
 
The declaration of COVID-19 pandemic by the WHO initiated a series of lockdowns globally that varied in stringency and duration; however, the spatiotemporal effects of these lockdowns on air quality remain understudied. This study evaluates the global impact of lockdowns on air pollutants using tropospheric and ground-level indicators over a five-month period. Moreover, the relationship between air pollution and COVID-19 cases and mortalities was examined. Changes in the global tropospheric (NO2, aerosols, and O3) and ground-level (PM2.5, PM10, NO2, and O3) pollutants were observed, and the maximum air quality improvement was observed immediately after lockdown. Except for a few countries, a decline in air pollutants correlated with a reduction in Land Surface Temperature (LST). Notably, regions with higher tropospheric NO2 and aerosol concentrations were also COVID-19 hotspots. Our analysis showed moderate positive correlation for NO2 with COVID-19 cases (R² = 0.33; r = 0.57, P = 0.006) and mortalities (R² = 0.40; r = 0.63, P = 0.015), while O3 showed a weak-moderate positive correlation with COVID-19 cases (R² = 0.22; r = 0.47, P = 0.003) and mortalities (R² = 0.12; r = 0.35, P = 0.012). However, PM2.5, and PM10 showed no significant correlation with either COVID-19 cases or mortality. This study reveals that humans living under adverse air pollution conditions are at higher risk of COVID-19 infection and mortality.
 
To avoid the spread of COVID-19, China implemented strict prevention and control measures, resulting in dramatic variations in air quality. Here, we applied a machine learning algorithm (random forest model) to eliminate meteorological effects and characterize the high-resolution variation characteristics of air quality induced by COVID-19 in Beijing, Wuhan, and Urumqi. Our RF model estimates showed that the highest decrease in deweathered PM2.5 in Wuhan (−43.6%) and Beijing (−14.0%) was at traffic stations during lockdown period (February 1- March 15, 2020), while it was at industry stations in Urumqi (−54.2%). Deweathered NO2 decreased significantly in each city (∼30%–50%), whereas accompanied by a notable increase in O3. The diurnal patterns show that the morning peaks of traffic-related NO2 and CO almost disappeared. Additionally, our results suggested that meteorological effects offset some of the reduction in pollutant concentrations. Adverse meteorological conditions played a leading role in the variation in PM2.5 concentration in Beijing, which contributed to +33.5%. The true effect of lockdown reduced the PM2.5 concentrations in Wuhan, Beijing, and Urumqi by approximately 14.6%, 17.0%, and 34.0%, respectively. In summary, lockdown is the most important driver of the decline in pollutant concentrations, but the reduction of SO2 and CO is limited and they are mainly influenced by changing trends. This study provides insights into quantifying variations in air quality due to the lockdown by considering meteorological variability, which varies greatly from city to city, and provides a reference for changes in city scale pollutant concentrations during the lockdown.
 
Surface ozone (O3) is a major air pollutant around the world. This study investigated O3 concentrations in nine cities during the Coronavirus disease 2019 (COVID-19) lockdown phases. A statistical model, named Generalized Additive Model (GAM), was also developed to assess different meteorological factors, estimate daily O3 release during COVID-19 lockdown and determine the relationship between the two. We found that: (1) Daily O3 significantly increased in all selected cities during the COVID-19 lockdown, presenting relative increases from −5.7% (in São Paulo) to 58.9% (in Guangzhou), with respect to the average value for the same period in the previous five years. (2) In the GAM model, the adjusted coefficient of determination (R²) ranged from 0.48 (Sao Paulo) to 0.84 (Rome), and it captured 51–85% of daily O3 variations. (3) Analyzing the expected O3 concentrations during the lockdown, using GAM fed by meteorological data, showed that O3 anomalies were dominantly controlled by meteorology. (4) The relevance of different meteorological variables depended on the cities. The positive O3 anomalies in Beijing, Wuhan, Guangzhou, and Delhi were mostly associated with low relative humidity and elevated maximum temperature. Low wind speed, elevated maximum temperature, and low relative humidity were the leading meteorological factors for O3 anomalies in London, Paris, and Rome. The two other cities had different leading factor combinations.
 
The Chinese government implemented strict emission reduction measures of air pollution between 2013 and 2017. However, from the winter of 2017 until February 2020, during the COVID-19 pandemic, the twenty explosive rise (ER) events of PM2.5 mass in twelve heavy aerosol pollution episodes (HPEs) still appeared in Beijing and its vicinity (BIV). To explore the controlling mechanism for the ER under the condition of drastically reduced emissions, the vertical structure of meteorological elements by L-band second-level sounding and aerosol properties by Lidar were investigated associating with the analysis of surface concentration in PM2.5 mass, its main precursor gases, as well as black carbon (BC) by seven-wavelength Aethalometer. The planetary boundary layer height (BLH) was also estimated together with an analysis of the unfavorable meteorological index (PLAM) that can quantify the impact of unfavorable meteorological conditions to cause the change of PM2.5 concentration. The results suggested that the ER reoccurrence's fundamental cause is that the emissions have not yet fallen sufficiently to a level to decouple HPEs from unfavorable meteorological conditions. During the ER period, the BLH dropped significantly. The fact that PM2.5, its precursor gases, and black carbon increased almost in a similar proportion, indicating that the boundary layer structure change caused by aerosol accumulation is the dominant reason for the ER phenomenon compared to the chemical conversion factor. The two-way feedback effect between the further worsened meteorological conditions and the accumulation of PM2.5 typically interpreted 54%–93% of the ER. An HPE starting 8 Feb. 2020 during the COVID-19 period underwent one of the worst meteorological conditions, quantified by PLAM, in BIV since 2013. However, with a similar level of unfavorable meteorological conditions, the average PM2.5 concentration during the HPE in 2020 was only about 66% of that of a similar HPE in 2016. It shows that the substantial reduction of emissions reduces the PM2.5 pollution level primarily as before when facing an equivalent level of unfavorable meteorological conditions. These results combined suggest that China's continuous efforts to reduce emissions proceed in the right direction and have achieved the desired results.
 
Countries in Northeast Asia have been regulating PM2.5 sources and studying their local and transboundary origins since PM2.5 causes severe impacts on public health and economic losses. However, the separation of local and transboundary impacts is not fully realized because it is impossible to change air pollutant emissions from multiple countries experimentally. Exceptionally, the early stage of the COVID-19 outbreak (January–March 2020) provided a cross-country experiment to separate each impact of PM2.5 sources identified in Seoul, a downwind area of China. We evaluated the contributions of PM2.5 sources compared to 2019 using dispersion normalized positive matrix factorization (DN-PMF) during three meteorological episodes. Episodes 1 and 2 revealed transboundary impacts and were related to reduced anthropogenic emissions and accumulated primary pollutants in Northeast China. Anthropogenic emissions, except for the residential sector, decreased, but primary air pollutants accumulated by residential coal combustion enhanced secondary aerosol formation. Thus, the contributions of sulfate and secondary nitrate increased in Seoul during episode 1 but then decreased maximally with other primary sources (biomass burning, district heating and incineration, industrial sources, and oil combustion) during episode 2 under meteorological conditions favorable to long-range transport. Local impact was demonstrated by atmospheric stagnation during episode 3. Meteorological condition unfavorable to local dispersion elevated the contributions of mobile and coal combustion and further contributed to PM2.5 high concentration events (HCE). Our study separates the local and transboundary impacts and highlights that cooperations in Northeast Asia on secondary aerosol formation and management of local sources are necessary.
 
Photochemical regime for ozone (O3) formation is complicated in the sense that reducing emission of nitrogen oxides (NOx) may increase O3 concentration. The lockdown due to COVID-19 pandemic affords a unique opportunity to use real observations to explore the O3 formation regime and the effectiveness of NOx emission control strategies. In this study, observations from ground networks during the lockdowns were used to assess spatial disparity of the Ratio of Ozone Formation (ROF) for nitrogen dioxide (NO2) reduction in the Greater Bay Area (GBA) of China. The health risk model from Air Quality Health Index (AQHI) system in Hong Kong was adopted to evaluate the risk tradeoffs between NO2 and O3. Results show that the levels of O3 increase and NO2 reduction were comparable due to high ROF values in urban areas of central GBA. The ozone reactivity to NO2 reduction gradually declined outwards from central GBA. Despite the O3 increases, the NOx emission controls reduced the Integrated Health Risk (IHR) of NO2 and O3 in most regions of the GBA. When risk coefficients from the AQHI in Canada or the global review were adopted in the risk analyses, the results are extremely encouraging because the controls of NOx emission reduced the IHR of NO2 and O3 almost everywhere in the GBA. Our results underscore the importance of using a risk-based method to assess the effectiveness of emission control measures and the overall health benefit from NOx emission controls in the GBA.
 
During the COVID-19 lockdown, only the most basic and necessary production activities were retained in China. Such strict measures have caused many inconveniences to the people and the economy, but also provided the research community with a rare opportunity to compare the effects of weather conditions and human activities on air quality in the region. Here, a comparative analysis of the impact of weather conditions and human activities on air quality in the Dongting and Poyang Lake Region (DPLR) is proposed during the COVID-19 pandemic based on a circulation-to-environment approach. T-mode objective circulation classification method was applied to explore the effects of weather conditions on the concentration of two typical pollutants (PM2.5 and ozone) affecting the DPLR. PM2.5 and ozone concentrations under the nine identified circulation patterns are discussed. Under the control of circulation type CT1, CT3 and CT6, the PM2.5 concentrations in this area are high, while under the control of CT2 and CT9, the ozone concentrations are high. By comparing the variation in PM2.5 and ozone concentrations in three important cities in the region (Wuhan, Changsha, and Nanchang) during the three stages (Before Controlling, During Controlling and After Controlling) of COVID-19 from January to April 2020 and the corresponding months of a reference period (2015–2019), it is found that after controlling the human activities, the PM2.5 concentration dropped by 37.45%, while the ozone concentration increased by 111.83%. Ozone concentration was mainly affected by the synoptic circulation pattern, while the PM2.5 concentration was more affected by human activities.
 
Exposure to air pollution can exacerbate the severe COVID-19 conditions, subsequently causing an increase in the death rate. In this study, we investigated the association between long-term exposure to air pollution and risks of COVID-19 hospitalization and mortality in Arak, Iran. Air pollution data was obtained from air quality monitoring stations located in Arak, including particulate matter (PM), nitrogen dioxide (NO2), sulfur dioxide (SO2), ozone (O3) and carbon monoxide (CO). Daily numbers of Covid-19 cases including hospital admissions (hospitalization) and deaths (mortality) were obtained from a national data registry recorded by Arak University of Medical Sciences. A Poisson regression model with natural spline functions was applied to set the effects of air pollution on COVID-19 hospitalization and mortality. The percent change of COVID-19 hospitalization per 10 μg/m³ increase in PM2.5 and PM10 were 8.5% (95% CI 7.6 to 11.5) and 4.8% (95% CI 3 to 6.5), respectively. An increase of 10 μg/m³ in PM2.5 resulting in 5.6% (95% CI: 3.1–8.3%) increase in COVID-19 mortality. The percent change of hospitalization (7.7%, 95% CI 2.2 to 13.3) and mortality (4.5%, 95% CI 0.3 to 9.5) were positively significant per one ppb increment in SO2, while NO2, O3 and CO were inversely associated with hospitalization and mortality. Our findings strongly suggesting that a small increase in long-term exposure to PM2.5, PM10 and SO2 elevating risks of hospitalization and mortality related to COVID-19.
 
The annual of air pollutants of CO (a, b), NO (c, d), NO 2 (e, f), O 3 (g, h), PM 10 (i, j), 220 PM 2.5 (k, l), and BC (m, n) in Augsburg between 2020 and 2010-2019. The left panel indicated 221 time series of the monthly mean concentrations in the periods studied. The right panel indicated 222 comparison of the total concentrations in the periods studied. 223 224 3.2. The best fitting model and its driving factors 225 3.2.1. The best fitting model 226 The splitting proportion of training data and test data were set as 50:50, 70:30, and 90:10 227 for four models. The fitting effects were compared and summarized in Table 1, based on R 2 , 228 MSE, RMSE, MAE, and MAPE for test data (Table S6). As a result, Random Forest showed 229 the best fitting effects than the other three models for all air pollutants. In addition, Random 230 Forest showed different fitting effects for each air pollutant by using different splitting 231 proportions. Based on the optimized fitting effects, 90:10 was chosen as the best splitting 232 proportion for CO, NO, NO 2 , O 3 , PM 10 , and PM 2.5 , with R 2 of 0.585, 0.526, 0.585, 0.747, 0.559, 233 and 0.613, respectively. While 70:30 splitting proportion was selected for BC with R 2 of 0.452. 234 235 Table 1. The fitting effects of four models based on R 2 , MSE, RMSE, MAE, and MAPE values 236 (Random Forest, RF; K-nearest Neighbors, KNN; Linear regression, LN; and Lasso regression, 237 Lasso). 238
Driving factor of different variables based on Random Forest data in Augsburg from 285 01/08/2016 to 30/09/2019. 286 287 It is interesting to observe the role of wind speed and wind direction on PM 10 , PM 2.5 , and 288 BC concentrations, which driving factors were higher than 0.15 and 0.2, respectively. We 289 further showed the distributions of PM 10 , PM 2.5 , and BC with wind speed and wind direction to 290 interpret their interplay in Table 2. The radius graphs represented the wind speed (m/s) and 291 direction (0-360 ○ ), while the color represented mass concentrations of air pollutants in Fig. 4. 292 The wind frequencies were predominant in the west (13.08%) and east (10.18%) directions 293 compared to north (5.85%) and south (5.63%) directions. But the PM 10 and PM 2.5 in west (18.68 294 and 11.97 µg/m 3 ) and east (18.77 and 12.04 µg/m 3 ) directions did not show consistent trends to 295 the north (18.73 and 12.02 µg/m 3 ) and south (18.64 and 11.94 µg/m 3 ) directions. The average 296 wind speeds in west (2.91 m/s) and south (2.91 m/s) directions were higher than north (2.90 297 m/s) and east (2.88 m/s) direction. The PM 10 were lower in west (18.68 µg/m 3 ) and south (18.64 298 µg/m 3 ) direction than in north (18.73 µg/m 3 ) and east direction (18.77 µg/m 3 ). This is also 299
The COVID-19 pandemic in Germany in 2020 brought many regulations to impede its transmission such as lockdown. Hence, in this study, we compared the annual air pollutants (CO, NO, NO2, O3, PM10, PM2.5, and BC) in Augsburg in 2020 to the record data in 2010–2019. The annual air pollutants in 2020 were significantly (p < 0.001) lower than that in 2010–2019 except O3, which was significantly (p = 0.02) higher than that in 2010–2019. In a depth perspective, we explored how lockdown impacted air pollutants in Augsburg. We simulated air pollutants based on the meteorological data, traffic density, and weekday and weekend/holiday by using four different models (i.e. Random Forest, K-nearest Neighbors, Linear Regression, and Lasso Regression). According to the best fitting effects, Random Forest was used to predict air pollutants during two lockdown periods (16/03/2020–19/04/2020, 1st lockdown and 02/11/2020–31/12/2020, 2nd lockdown) to explore how lockdown measures impacted air pollutants. Compared to the predicted values, the measured CO, NO2, and BC significantly reduced 18.21%, 21.75%, and 48.92% in the 1st lockdown as well as 7.67%, 32.28%, and 79.08% in the 2nd lockdown. It could be owing to the reduction of traffic and industrial activities. O3 significantly increased 15.62% in the 1st lockdown but decreased 40.39% in the 2nd lockdown, which may have relations with the fluctuations the NO titration effect and photochemistry effect. PM10 and PM2.5 were significantly increased 18.23% an 10.06% in the 1st lockdown but reduced 34.37% and 30.62% in the 2nd lockdown, which could be owing to their complex generation mechanisms.
 
The prevention and control measures in place during the coronavirus disease 2019 (COVID-19) pandemic served as perfect conditions for a natural experiment, which has provided an opportunity to investigate the extent to which environmental regulations can improve air quality in the short term. This article examines the relationship between anti-epidemic measures and air quality via a regression discontinuity design (RDD) based on the daily data from 326 prefecture-level cities in China. The empirical results indicate that, during the period of epidemic prevention and control in China, the air quality index (AQI) significantly decreased by 20.56, and the emission concentrations of the pollutants PM2.5, PM10, and NO2 decreased by 19.01, 20.20, and 2.13, respectively. It was further found that the continued operation of life-supporting industries during the epidemic, such as thermal power plants and heating industries, may be the reasons why there were no significant decreases in the concentrations of SO2 and CO. The O3 concentration, which is related to sunlight and the concentrations of NOx, was also not found to have changed significantly in the short term. Moreover, it was found that the more daily confirmed COVID-19 cases in an area, the greater the improvement of the air quality.
 
China's COVID-19 lockdown closed many of its air pollution sources and created an unprecedented clean-air scenario. However, air pollution remained significant, arousing questions about its causes and reawakening debate over the country's ultralow emission (ULE) policy. Here, we analyze the characteristics and causes of China's air pollution during this period to assess ULE's effectiveness. We find that the concentrations of fine particles (PM2.5, PM10), NO2, and CO in Wuhan decreased significantly during the lockdown. However, O3 and SO2 concentrations increased, though to different extents. The ULE policy promoted emission reductions, but emissions from coal-fired industries (e.g., thermal power and steel) remained the fundamental source of air pollution and appeared to have been underestimated in previous researches. In contrast, emissions from construction and transportation and the ULE's effectiveness may have been overestimated. The characteristics of air pollution in southern and northern cities support these conclusions. Therefore, it is necessary to strengthen the adjustment of China's energy and industrial structures, and identify more suitable mitigation measures for each region's characteristics. It is also necessary to strengthen the synergistic governance of pollutants and promote emission reduction technologies.
 
Continuous measurements of gaseous elemental mercury (GEM) were conducted in Qingdao from March 2020 to March 2021. The average concentration of GEM was (2.39 ± 1.07 ng/m³) with a variation range of 0.27–10.78 ng/m³. GEM exhibited a clear pattern of daily variation, with daily peaks occurring between 11:00–13:00. GEM concentrations were higher in winter (2.80 ± 1.28 ng/m³) than that in summer (2.18 ± 1.05 ng/m³). The high winter concentrations were related to coal-fired heating and the increased frequency of polluted weather in northern China. Principal component analysis showed that the main factors affecting GEM concentration were fossil fuel combustion, natural source release and atmospheric diffusion conditions. The anthropogenic emission sources were the main source of GEM in spring and winter, and natural sources of GEM was large in summer. The potential source contribution function suggested that North and Central China were the main potential sources of GEM, and there were large differences in the potential sources of GEM in different seasons. Comparing the GEM in the same time periods in 2018, 2020, and 2021, government policies, temporary lockdown measures for the COVID-19 epidemic, and urban village renovation led to a decreasing trend of GEM concentrations. This study contributes to a better understanding of the effects of long-range transport of air masses and anthropogenic emissions on atmospheric mercury in eastern coastal cities and offshore areas.
 
The COVID-19 virus outbreak has been declared a “global pandemic”. Therefore, “lockdown” was issued in affected countries to control the spread of the virus. To assess air pollution during and after lockdowns, this study selected pandemic hotspots in China (Wuhan), Japan (Tokyo), the Republic of Korea (Daegu), and India (Mumbai) and compared the Air Quality Index (AQI) in these areas for the past three years. The results indicated that air pollution levels were positively correlated with a reduction in pollutant levels during and after lockdowns in these cities. In Tokyo, low levels of air pollution, no significant change in the distribution of “good” and “moderate” days was observed during lockdown. In Daegu, mid-level air pollution, the percentage of “unhealthy” days (AQI>100) markedly reduced during lockdown; however, this reverted after lockdown was lifted. In Wuhan and Mumbai, high air pollution levels, the percentage of unhealthy days remarkably decreased during lockdown and continued to reduce after lockdown. It was found that PM2.5 was the critical pollutant for all cities because its sub-AQI was the largest of the six pollutant species for the majority of days. In addition, PM10 dominated the overall AQI for 2.2–9.6% of the period in Wuhan and Mumbai, and its sub-AQI reduced during lockdown. The mean sub-AQI for NO2, SO2, CO, and O3 was within the “good” category for all cities. In conclusion, the lockdown policy reduced air pollution in general and this reduction was more significant for regions with high air pollution levels.
 
Spatial distribution of 380 major cities (red points) and 12 countries (orange areas) for the analysis of PM 2.5 , O 3 , and NO 2 concentrations. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
(a) The probability distribution of lockdown start and end dates in 577 global cities affected by COVID-19 (see Appendix for details) and (b) the distribution of the number of days in lockdown. In panel (a), the blue shaded area represents the global lockdown phase identified for this analysis (March to May 2020). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Daily mean PM 2.5 concentrations across all national stations from the January 1, 2020 (Day of the Year, DOY = 1) to the June 28, 2020 (DOY = 180). The red lines indicate the concentration of pollutants in 2020, while the blue lines indicate the average concentration of pollutants in the baseline (2015-2019), with standard deviations in the baseline in light blue. Gray rectangles show the duration of lockdown in each of the 12 study countries. The numbers in parentheses indicate the relative rate of change (ROC) of daily mean PM 2.5 concentration after the COVID-19 lockdown. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Lockdowns implemented in response to COVID-19 have caused an unprecedented reduction in global economic and transport activity. In this study, variation in the concentration of health-threatening air pollutants (PM2.5, NO2, and O3) pre- and post-lockdown was investigated at global, continental, and national scales. We analyzed ground-based data from >10,000 monitoring stations in 380 cities across the globe. Global-scale results during lockdown (March to May 2020) showed that concentrations of PM2.5 and NO2 decreased by 16.1% and 45.8%, respectively, compared to the baseline period (2015–2019). However, O3 concentration increased by 5.4%. At the continental scale, concentrations of PM2.5 and NO2 substantially dropped in 2020 across all continents during lockdown compared to the baseline, with a maximum reduction of 20.4% for PM2.5 in East Asia and 42.5% for NO2 in Europe. The maximum reduction in O3 was observed in North America (7.8%), followed by Asia (0.7%), while small increases were found in other continents. At the national scale, PM2.5 and NO2 concentrations decreased significantly during lockdown, but O3 concentration showed varying patterns among countries. We found maximum reductions of 50.8% for PM2.5 in India and 103.5% for NO2 in Spain. The maximum reduction in O3 (22.5%) was found in India. Improvements in air quality were temporary as pollution levels increased in cities since lockdowns were lifted. We posit that these unprecedented changes in air pollutants were mainly attributable to reductions in traffic and industrial activities. Column reductions could also be explained by meteorological variability and a decline in emissions caused by environmental policy regulations. Our results have implications for the continued implementation of strict air quality policies and emission control strategies to improve environmental and human health.
 
The COVID-19 disease caused by the SARS-CoV-2 virus first identified in December 2019 has resulted in millions of deaths so far around the world. Controlling the spread of the disease requires a good understanding of the factors (e.g. air pollutants) that influence virus transmission and the conditions under which it spreads. This study analyzed the relationships between COVID-19 cases and both short-term (6-month) and long-term (60-month) exposures to eight air pollutants (NO, NO2, NOx, CO, SO2, O3, PM2.5 and PM10) in Tehran city, Iran, by integrating geostatistical interpolation models, regression analysis, and an innovated COVID-19 incidence rate calculation (Q-index) that considered the spatial distributions of both population and air pollution. The results show that the higher COVID-19 incidence rate was significantly associated with the exposure to higher concentrations of CO, NO, and NOx during the short-term period; the higher COVID-19 incidence rate was significantly related to the exposure to higher concentrations of PM2.5 during the long-term period; while COVID-19 incidence rate was not significantly associated with the concentrations of O3, SO2, PM10 and NO2 in either period. This study indicates that exposure to air pollutants can effect an increase in the number of infected people by transmitting the virus through the air or by predisposing people to the disease over time. The Q-index calculation method developed in this study can be also used by other studies to calculate more accurate disease rates that consider the spatial distribution of both population and air pollution.
 
Atmospheric pollution studies have linked diminished human activity during the COVID-19 pandemic to improve air quality. This study was conducted during January to March (2019–2021) in 332 cities in China to examine the association between population migration and air quality, and examined the role of three city attributes (pollution level, city scale, and lockdown status) in this effect. This study assessed six air pollutants, namely CO, NO2, O3, PM10, PM2.5, and SO2, and measured meteorological data, with-in city migration (WCM) index, and inter-city migration (ICM) index. A linear mixed-effects model with an autoregressive distributed lag model was fitted to estimate the effect of the percent change in migration on air pollution, adjusting for potential confounding factors. In summary, lower migration was associated with decreased air pollution (other than O3). Pollution change in susceptibility is more likely to occur in NO2 decrease and O3 increase, but unsusceptibility is more likely to occur in CO and SO2, to city attributes from low migration. Cities that are less air polluted and population-dense may benefit more from decreasing PM10 and PM2.5. The associations between population migration and air pollution were stronger in cities with stringent traffic restrictions than in cities with no lockdowns. Based on city attributes, an insignificant difference was observed between the effects of ICM and WCM on air pollution. Findings from this study may gain knowledge about the potential interaction between migration and city attributes, which may help decision-makers adopt air-quality policies with city-specific targets and paths to pursue similar air quality improvements for public health but at a much lower economic cost than lockdowns.
 
The characteristics of black carbon (BC) aerosols, their sources, and their impact on atmospheric radiative forcing were extensively studied during the COVID-19 lockdown (28th March–31st May 2020) at a high-altitude rural site over the Western Ghats in southwest India. BC concentration and the contribution of BC originating from biomass burning (BCbb) estimated from the aethalometer model during the lockdown period were compared with the same periods in 2017 and 2018 and with the pre-lockdown period (1st February to March 20, 2020). BC concentrations were 44, 19, and 17% lower during the lockdown period compared with the pre-lockdown periods of 2020 and similar periods (28th March to 31st May) of 2017 and 2018, respectively. BCbb contributed ∼50% to total BC during the lockdown period of 2020 and compensated for the decrease in BC concentration due to lower traffic emissions. The characteristics of light-absorbing organic carbon (brown carbon; BrC) absorption at 370 nm were evaluated during the lockdown and the pre-lockdown periods of 2020, 2017, and 2018. The BrC was estimated to be the highest during the lockdown period of 2020. Finally, atmospheric radiative forcing was calculated using the mean BC concentration during the pre-lockdown, lockdown, and similar periods (28th March to 31st May) of 2017 and 2018.
 
The definition of haze day is still controversial, and there are different data sources for haze data in China. This study analyzes the uncertainty of haze data in China, which is a prerequisite for studying haze trends. The overall trends defined by different approaches were relatively consistent, but the average values were different. The annual number of haze days defined by visibility and relative humidity was higher than the number defined by manual observation. The correlation coefficient between the two datasets for the whole country was 0.92, and ranged from 0.7 to 0.9 for different regions. The similarity between the trends of haze days by different approaches was greater in the Yangtze River Delta region and the Pearl River Delta region than in the Southwest Region. Limited impact of relative humidity on the trends of haze days was detected in China and different regions. The influence of station screening was also investigated. Station screening had no effect on the annual average number of haze days and the long-term trend. Based on the above analysis, the haze data defined by visibility and relative humidity were chosen to study the spatial and temporal distributions for China. In terms of seasons, haze days were most frequent in winter and least frequent in summer. Except for the Northwest region, the number of haze days increased noticeably after 1980. Our results can further deepen the understanding of the characteristics of haze pollution, and support the study of the causes of haze pollution in China.
 
Atmospheric visibility (Vis) is used to evaluate air pollution over global land surface during 1973–2012. Over global more populated areas with major anthropogenic emissions and hygroscopic aerosol species, weaker surface wind is generally associated with more severe air pollution and Vis degradation, while an increase of relative humidity (RH) generally favors more efficient secondary formation and hygroscopic growth of aerosol and Vis impairment. The allocation of meteorological factor with the major aerosol component over global different areas can influence the formation and variation of aerosol and cloud, and their effects on the atmospheric boundary layer (ABL) evolution and air pollution. Over East Asia, India and Southeast Asia, the air pollution with a growing concern with Vis decrease can be associated to an increase in anthropogenic emission and a synchronous stabilization of ABL (characterized by a weakened, less variable surface wind and a decreased diurnal temperature range, DTR). Both the increase of aerosol loading and cloud cover could (i) increase daily minimum air temperature versus decrease daily maximum air temperature and result in a decrease of DTR, (ii) reduce solar radiation reaching the surface, stabilize the ABL and aggravate air pollution. While Vis increase over Europe could be related to more effective emission control and an intensification of surface wind. Also significantly, a close relationship is identified between RH and DTR; RH explains about 90% of the variance of DTR over Europe in summer; a drier weather is corresponding to a greater surface incoming short-wave flux and a greater DTR.
 
Clustering is an explorative data analysis technique used for investigating the underlying structure in the data. It described as the grouping of objects, where the objects share similar characteristics. Over the past 50 years, clustering has been widely applied to atmospheric science data in particular, climate and meteorological data. Since the 1980's, air pollution studies began employing clustering techniques, and has since been successful, and the aim of this paper is to provide a review of such studies. In particular, two well known and commonly used clustering methods i.e. k-means and hierarchical agglomerative, that have been applied in air pollution studies have been reviewed. Air pollution data from two sources i.e. ground-based monitoring stations and air mass trajectories depicting pollutant pathways, have been included. Research works that have focused on spatio-temporal characteristics of air pollutants, pollutant behavior in terms of source, transport pathways, apportionment and links to meteorological conditions, comprise much of the research works reviewed. A total of 100 research articles were included during the period of 1980–2019. The purpose of the clustering approach, the specific technique used and the data to which it was applied constitute much of the discussion presented in this review. Overall, the k-means technique has been extensively used among the studies, while average and Ward linkages were the most frequently applied hierarchical clustering techniques. Reviews of clustering techniques applied in air pollution studies are currently lacking and this paper aims to fill that gap. In addition, and to the best of the authors' knowledge, this is the first review dedicated to clustering applications in air pollution studies, and the first that covers the longest time span (1980–2019).
 
In this study, the long-term trend in atmospheric carbon monoxide (CO) concentration was analyzed using the CO levels measured (intermittently) at an air quality monitoring (AQM) station in Seoul, Korea, between the years 1987 and 2013. Temporal trends in CO were analyzed on an annual and seasonal basis in reference to other important air pollutants such as methane (CH4), particulate matter (PM10), sulfur dioxide (SO2), nitrogen monoxide (NO), nitrogen dioxide (NO2), mercury (Hg), and ozone (O3). The annual mean of CO for the entire period was 0.93 ± 0.22 ppm. CO levels were reduced by 83% from 3.25 ± 0.78 ppm (1987) to 0.51 ± 0.31 ppm (2013). Its relative reduction was compared over three periods chosen arbitrarily as period 1 (fast reduction, 1987-1988), period 2 (intermediate reduction, 1999-2000), and period 3 (slow reduction, 2004-2013). The concentrations of CO were strongly correlated with others (e.g., SO2, NO, NO2, O3, and Hg), suggesting the effects of similar source processes (e.g., fuel combustion). The reduction in its level was marginally consistent with the decreasing trend in the total CO column concentration in Seoul by the Measurements of Pollution in the Troposphere (MOPITT) satellite between 2000 and 2013, indicating decreasing anthropogenic CO emissions (despite increasing anthropogenic CO2 emissions). The rapid relative reduction of CO in period 1 and the subsequent slower but moderate reduction thereafter appear to reflect the effects of both enforcement of administrative regulations and advances in emissions control technologies. © 2017 Turkish National Committee for Air Pollution Research and Control.
 
NO X EFs for various fuel types used in the current work Sector Unit Fuel type CNG/NG LPG Wood Diesel Coal/Coke Kerosene FO 
Gridded NO X emission in India from all sources (a) 1991, (b) 2001, (c) growth during 1991-2001, (d) locations of thermal power stations in India (2001), (e) 2011, (f) growth during 2001-2011.
Gridded NO X emissions in India from bio-fuel combustion in 2001.
NO x estimation for Indian region as reported by different researchers Reference Estimated NO X in Gg (Year) Sources
The fossil fuel and bio-fuel burning in a developing country like India can have a significant impact on global climate. In the current work, we have set-up a more realistic, accurate and spatially distributed, all India, NOx emissions from different fuel combustion and industrial activities at 1 degrees x1 degrees grid resolution by incorporating the most recently available micro-level activity data as well as country specific emission factors (EFs) at high resolution. The emission scenarios and their trends are studied in a comprehensive way for approximately 593 districts (sub-region) in India. We have developed three scenarios to construct the possible range of past and present NOx emissions using Geographical Information System (GIS) based methodology. The total NOx emissions are estimated to be 2 952 Giga gram (Gg)/yr, 4 487 Gg/yr and 7 583 Gg/yr for three different base years, i.e., 1991, 2001 and 2011. NOx emission trends in India during 1990s and 2000s due to different major anthropogenic activities are estimated and their growth is discussed. A strong growth of NOx is found during 2000s as compared to 1990s. All major cities remain as top emitters of NOx. The present work depicts that the contribution of fossil fuel will gradually increase in coming years and will be around 91% by 2011. The present new gridded emission inventory will be very useful as an input to Chemical Transport Modeling study over Indian geography. (c) Author(s) 2012. This work is distributed under the Creative Commons Attribution 3.0 License.
 
The present study was to evaluate the frequency distribution of dust events by analysing on long-term data over the northern Persian Gulf (NPG). The study results accordingly revealed that the Persian Gulf (PG) on the northwest had been affected by the dust deposition transported more often by strong northwesterly winds, along with the atmospheric turbulence from some sources in Iraqi deserts. However, the eastern parts of the PG had been influenced by wind-induced dust at or nearby the stations in this study. Kish Island had accordingly experienced the highest frequency of dust events and Siri Island had seen the highest frequency of dust storms due to high-speed winds and vicinity to Saudi Arabian deserts. The largest number of the dust events, originating from external sources, had further occurred in summer and late spring (May to July), whereas the peak frequency of the dust storms from domestic sources was seen in spring. Spring was a transitional season and dust events could thus happen due to atmospheric instability. The dust events had also stricken in the warm months of the year though the atmospheric fronts in February had created many dust storm events. As well, the rising or falling trends of the dust storms depended on precipitation and drought intensity in this region. Finally, the study results demonstrated that atmospheric patterns such as shamal winds, precipitation rate, wind speed/direction, as well as the distance from dust sources could contribute to the dust storm events over the NPG.
 
We examined long-term aerosol optical depth (AOD) trends over 53 sites across the globe which comprise 49 sites from the Aerosol Robotic Network (AERONET) and 4 sites from the Sky radiometer Network (SKYNET) during 1995–2018. Most of these sites are located in remote and isolated aged-background regions, and few are in urban/semi-urban sites having averaged AOD ∼0.1 at 500 nm. These selected sites have a global distribution including tropical, mid-latitudes, high-latitudes and Polar regions. Among them, there are 14 high-altitude stations (∼1028–5050 m amsl), including Himalayan and Polar regions. The main objective of the present work is to evaluate the AOD trends over the aged-background sites across the globe. We found that significant number of sites located in North-South America, Europe, Arctic and Australia have statistically significant and negative trends varied from −6.3x to −1.0x AOD year−1. The negative trends over these sites could be attributed to reduction in anthropogenic emission. Furthermore, there are mixed trends of positive as well as negative over Asian and southern oceanic regions including Antarctica. Some of the trends are weak and statistically non-significant, probably due to non-availability of long-term ground based data. However, the AOD trends over these regions show increasing tendency with statistically significant trends of 8.0x to 4.7x AOD year−1. The present study has also many important aspects on global and regional climate change at high-mountain and aged-background sites in particular, where the satellite based measurements are inaccurate and biased due to extremely low AOD.
 
Multiple household-related driving factors of residential direct carbon emissions (RDCE) in China at regional level have not yet been sufficiently addressed or quantified. In this paper, a logarithmic mean Divisia index (LMDI) decomposition analysis was employed to examine the factors (e.g., the number of households, per capita household income, household size, urbanization, energy intensity, energy structure and emission coefficient) impacting the changes in RDCE in Shanxi province of China from 1995 to 2014. The results showed that the increase in RDCE mainly attributed to the growing per capita household income and the increasing number of households. Additionally, the expansion of urbanization also contributed marginally to the increase in emissions. However, the shrinking household size was a main inhibitory factor and the decline in energy intensity was also responsible for the diminishing emissions. Based on the results, four emission reduction measures and strategies were identified: (i) using market economic mechanism to regulate household consumption behaviors towards environment protection and low carbon development, as well as encouraging the use of energy-efficiency domestic appliances and less energy-intensive lifestyles; (ii) setting strict divorce processes to lower divorce rates and encouraging people to live with their children and parents; (iii) realizing green transformation development of urbanization; (iv) promoting a shift to renewable and clean energy in people's daily life and power generation, e.g., wind, solar, hydro, nuclear and biogas.
 
Currently, the increasing carbon emissions accompanied with rapid urbanization and industrial development is the central environmental problem in China. In this paper, the carbon emission abatement index (CACI) consisting of equal weighted efficiency and equity indexes is employed to figure out the allocation of carbon emission mitigation quotient when the inclination and capacity in each province are all taken into consideration. Then, the novel spatial panel data model combined with economic weight matrix is used to estimate the spatial dependence of the carbon emission reduction potential at the provincial level. Moreover, the driving forces are examined. The result suggests that the provinces with low marginal cost and per capita carbon emissions should take on more burdens regarding carbon emission reduction (e.g. Ningxia, Shanghai, Shanxi, Qinghai and Tianjin), while the provinces with low economic development and carbon intensity can be allocated relatively less loads (e.g. Sichuan, Jiangxi, Guangxi, Anhui and Hunan). Furthermore, the capacity of carbon emission mitigation is negatively affected by urbanization and positively affected by per capita GDP, population density and energy intensity. The coefficients of economic development and energy utilization efficiency are also statistically significant, and indicates mutual effects among regions.
 
Top-cited authors
Philip K Hopke
  • Clarkson University
Shiva Nagendra SM
  • Indian Institute of Technology Madras
Andreas Markwitz
  • GNS Science
Mukesh Khare
  • Indian Institute of Technology Delhi
Suresh Tiwari
  • Indian Institute of Tropical Meteorology