The Effects of Air Quality and the Impact of Climate Conditions on the First COVID-19 Wave in Wuhan and Four European Metropolitan Regions

Abstract

This paper investigates the impact of air quality and climate variability during the first wave of COVID-19 associated with accelerated transmission and lethality in Wuhan in China and four European metropolises (Milan, Madrid, London, and Bucharest). For the period 1 January–15 June 2020, including the COVID-19 pre-lockdown, lockdown, and beyond periods, this study used a synergy of in situ and derived satellite time-series data analyses, investigating the daily average inhalable gaseous pollutants ozone (O₃), nitrogen dioxide (NO₂), and particulate matter in two size fractions (PM2.5 and PM10) together with the Air Quality Index (AQI), total Aerosol Optical Depth (AOD) at 550 nm, and climate variables (air temperature at 2 m height, relative humidity, wind speed, and Planetary Boundary Layer height). Applied statistical methods and cross-correlation tests involving multiple datasets of the main air pollutants (inhalable PM2.5 and PM10 and NO₂), AQI, and aerosol loading AOD revealed a direct positive correlation with the spread and severity of COVID-19. Like in other cities worldwide, during the first-wave COVID-19 lockdown, due to the implemented restrictions on human-related emissions, there was a significant decrease in most air pollutant concentrations (PM2.5, PM10, and NO₂), AQI, and AOD but a high increase in ground-level O₃ in all selected metropolises. Also, this study found negative correlations of daily new COVID-19 cases (DNCs) with surface ozone level, air temperature at 2 m height, Planetary Boundary PBL heights, and wind speed intensity and positive correlations with relative humidity. The findings highlight the differential impacts of pandemic lockdowns on air quality in the investigated metropolises.

1. Introduction

The novel coronavirus SARS-CoV-2 is responsible for the COVID-19 global disease outbreak, which was still ongoing worldwide in the summer of 2024. It started as an epidemic event in Wuhan, China, on 8 December 2019 [1] and evolved as a pandemic, declared by March 2020 [2]. Early studies estimated the risk of COVID-19 importation to Europe and worldwide by air travel from infected areas in China [3,4]. This study considers the effects of urban air pollution in synergy with climate conditions on coronavirus (COVID-19) disease incidence and mortality in Wuhan, China, and four European metropolises (Milan, Madrid, London, and Bucharest) from 1 January to 15 June 2020. From 1 January 2020 to 15 June 2020, 9,716,819 confirmed positive COVID-19 cases were reported, including 491,900 fatalities, with exponentially increasing numbers from more than 200 countries. This severe pneumococcal disease associated with COVID-19 rapidly spread in the winter–spring season of 2020 from Wuhan City, China [5,6,7,8], to other parts of the world, including Italy, Spain, the UK, and Romania in Europe. As a direct consequence, to limit social contact and flatten the epidemic curve during the first COVID-19 wave, several lockdown measures were implemented from 23 January 2020 to 8 April 2020 [9] in Wuhan, China; 8 March 2020 to 18 May 2020 in Italy [10]; 16 March to 8 June in Spain, 19 March 2020 to 10 May 2020 in the UK; and 15 March 2020 to 15 May 2020 in Romania. Like in other countries, the sanitary actions implemented to limit and prevent a high increase in COVID-19 viral infections improved the air quality of several urban regions worldwide [11,12,13]. Being a highly invasive pneumococcal contagious disease caused by the SARS-CoV-2 pathogen, COVID-19 has some similarities with previous outbreaks of coronaviruses (CoVs), Severe Acute Respiratory Syndrome (SARS)-CoV and Middle East Respiratory Syndrome (MERS)-CoV, but with some differences in the phenotypic and genomic structure, which can impact its pathogenesis [14,15,16]. Several epidemiological and toxicological studies found a high correlation with urban air pollution attributed to traffic-related and anthropogenic sources of pollutants, which can produce airway inflammation and hyper-responsiveness, contributing to an increased incidence and severity of cardiorespiratory diseases [17,18,19,20]. Due to their specific characteristics of genotoxicity, ecotoxicity, and oxidative potential, it was found that short-term and long-term exposure to particulate matter (PM2.5 and PM10) and gaseous pollutants (O₃ and NO₂) can be associated with increased susceptibility to lethality and morbidity from cardiorespiratory illnesses [21,22,23,24]. Advanced medical studies of the mechanisms associated with airway disease attributed to outdoor and indoor air pollutants considered alterations in lung and cardiac autonomic function, airway inflammation, blood pressure changes, and systemic inflammation, attributed to the epigenetic alteration of genes [25,26]. Also, recent studies have reported an association between increased levels of outdoor air pollutants such as PM2.5, PM10, O₃, and NO₂ and COVID-19 incidence and mortality, which can be explained by air pollutants’ role in immune system dysregulation and the increased susceptibility to SARS-CoV-2 viral infection of individuals due to a declining host defense system [27].

The COVID-19 pandemic posed major challenges, especially to societies and healthcare systems in large urban areas worldwide. Extensive research was conducted during several waves of COVID-19 to elucidate the impacts of several environmental, clinical, demographic, and socioeconomic factors on incidence and lethality variability in different countries and towns for development programs and design strategies in future viral pandemics.

In addition to local air pollution sources and meteorological factors, at the regional scale, long-range transport plays an important role in the surface levels of particulate matter and the function of gaseous pollutant concentrations in the geomorphology of observational sites. Among urban air pollutants, the focus is currently mainly on the Air Quality Index (AQI) associated with particulate matter in two size fractions (PM2.5 µm and PM10 µm) and the main gaseous air pollutants O₃ and NO₂. It is well recognized that O₃ and NO₂ are among the most threatening air pollutants in terms of harmful effects on human health, including increases in mortality, morbidity, and many respiratory and cardiovascular diseases. These air pollutants frequently occur in high concentrations in densely populated metropolitan areas in China and Europe [28]. Aerosol-related air pollution in cities can be quantified, in optical terms, through the parameter of the satellite-derived total Aerosol Optical Depth (AOD) at 550 nm, which is a critical indicator in understanding atmospheric physics and regional air quality and an important variable for assessing the aerosol load of the atmospheric column, sensitive to multiple air pollutants in the lower atmospheric system.

As a novelty, this study used analyses of AOD spatiotemporal variability with a high impact on urban air quality and atmospheric process dynamics in the selected metropolises [29]. Also, the spatiotemporal variability of meteorological parameters (air temperature at 2 m height, air relative humidity, air pressure, wind speed intensity, and Planetary Boundary Layer height) and their cumulative effects at urban and regional scales may have great significant impacts on the persistence of viral infections in aerosols and viral infection transmission. Particularly, the daily Planetary Boundary Layer height (PBL) characteristics are significantly related to the dispersion and transport of PM and gaseous pollutants affecting air quality atmospheric process dynamics and the spatiotemporal distribution of air pollutants’ concentrations [30].

It has been recognized that urban air pollution can act as a coronavirus carrier, promoting its spreading together with the air-associated risk factors of disease development in older people [31,32], with a history of smoking [33,34], hypertension, and heart disease [35], with chronic lung disease or moderate-to-severe asthma [36].

The goal of this paper was to provide scientific evidence on the influence of ground surface air pollution and air quality, together with climate conditions, on the fast diffusion of the COVID-19 pandemic disease in the selected investigated metropolitan regions, and to assess the impact of lockdown periods on air quality improvement under the changing circumstances of climate factors in the urban agglomerated areas.

To quantify the effects of outdoor air pollution on COVID-19 fast diffusion and fatality, this study used the time-series trend analysis of the daily mean air quality parameters PM2.5, PM10, O₃, and NO₂ concentrations data and the total AOD at 550 nm, together with the daily average AQI. Also, climate variability analysis was considered at the metropolitan scale (PBL heights, air temperature, relative humidity, and wind speed intensity) over the 1 January–15 June 2020 period. This research period was split into two time windows: (1) the COVID-19 pre-lockdown period; and (2) the lockdown and beyond period, before the vaccination program implementation.

The specific objectives of this study were as follows: (1) to identify the spatiotemporal patterns of the first wave of COVID-19 with air quality, main air pollutants, and climate variables in Wuhan, China, and four European metropolises (Milan, Madrid, London, and Bucharest); (2) to investigate the temporal variation in the values of the total AOD at 550 nm on the metropolitan scale; (3) to quantitatively investigate the correlation between the total AOD at 550 nm and the COVID-19-related daily new cases, DNCs; (4) to detect how the first implemented lockdown influenced the air quality, main air pollutant concentrations, and the total AOD at 550 nm variation in a different context; (5) and to discuss the implication of study results from the perspective of future pandemic events. The findings can provide informative data for epidemiologic studies and air quality improvement.

2. Materials and Methods

2.1. Air Quality

Global climate warming and urban pollution-related pressure on human health and the environment have placed air pollution as an important issue of policy decision-making. It is well recognized that outdoor air pollution, which is a major environmental risk to human health that consists of various natural (mineral dust, biomass combustion, etc.) and anthropogenic (traffic-related, construction, power generation, etc.) sources, is controlled by several local and regional atmospheric processes like as emission, transport, and deposition.

Airborne biological particles, known as bioaerosols, contain living and dead microorganisms (fungi, viruses, bacteria, and their excretions like endotoxins, glucans, mycotoxins, fungal spores, and plant pollen), which are released from the biosphere into the atmosphere, significantly affecting human health as allergens and pathogens [37,38]. The aerial survival rates of bioaerosols are increased by resistance to solar ultra-violet radiation and association with particulate matter. SARS-CoV-2 coronavirus, with an average measured size of 0.1 µm, in the range of 60–140 nm, can be attached to particulate materials less than 0.1 µm as the carrier (droplet or particle), which can also be clustered with other PM [39,40]. Exposure to ambient near-ground hazardous atmospheric pollutants such as primary PM, Black Carbon (BC), O₃, NO₂, carbon monoxide (CO), volatile organic compounds (VOCs), and various heavy metals in urban areas is also associated with a wide range of adverse environmental effects and climate changes including extreme climate events. Understanding the behavior of particulate matter and gaseous pollutants in the lower atmospheric system is very useful for the assessment of the impact on air quality and the oxidation capacity of the atmosphere with a relatively long lifetime in urban crowded environments [41,42]. During the last several decades, air pollution has become a major environmental issue and health hazard, especially for the general population in metropolitan regions. Increases in outdoor air exposure to high concentrations of particulate matter and gaseous pollutants directly and indirectly affect people’s health outcomes [43,44,45,46,47,48].

Air quality is considered a key environmental factor in COVID-19 infections. Several studies have reported a high association of COVID-19 infection rates with urban air pollution during days exceeding the threshold limits set for PM2.5, PM10, O₃, or NO₂ [49,50]. A relevant conclusion of these published studies is that cities with poor air quality and high-density inhabitants have an increased probability of high COVID-19 infections, which are mainly attributed to air pollution rather than human-to-human transmission. Particulate matter (PM) air pollution is very complex, covering a large size range, consisting of a multi-component matrix originating from different anthropogenic sources (traffic-related, power generation, etc.) and natural sources (biomass combustion, dust, etc.), being subject to several transport and removal atmospheric processes. In metropolitan agglomerated areas like Wuhan, Milan, Madrid, London, and Bucharest, the selected cities in this paper, the PM concentration is normally dominated by various size fractions (the ultrafine particles PM0.1 with diameter < 0.1 µm; fine particles PM2.5 with diameter ≤ 0.2.5 µm; coarse particles PM10 with diameter > 0.2.5 µm and ≤ 10 µm) [51,52]. PM is a heterogeneous complex mixture of suspended particles of different sizes, chemistry, shapes, and great spatio-temporal variability. Particulate matter with an aerodynamic diameter of 2.5 µm, mainly originating from combustion processes, is considered the most toxic for the respiratory and cardiovascular system, with strong inflammatory potential [53,54] and associated with increased morbidity and lethality. While coarse PM10 deposits mainly in the upper and large conducting airways, being used as an air quality indicator, fine PM or PM2.5 μm deposits throughout the lower respiratory tract, particularly in small airways and alveoli [55]. Ultrafine PM0.1 μm might be deposited in both the upper as well as in the lower respiratory tract [56,57]. Due to chemical composition and sometimes association with fungi, bacteria, and viruses, PM may present toxic heavy metals on its surface, which can increase its toxicity [58,59]. Also, epidemiological studies have reported PM’s capacity for infectious disease spread, serving as a transmission carrier for numerous and diverse types of viruses, including influenza viruses, lethal coronavirus viruses like as Severe Acute Respiratory Syndrome virus (SARS-CoV), and Middle East Respiratory Syndrome Coronavirus (MERS-CoV) [60,61]. It has been demonstrated that chronic or short-term exposure to ambient air pollutants in metropolitan regions may have a significant role in the diffusion of SARS-CoV-2 pathogens, being associated with changes in the physiology of the respiratory system (reduction in lung function and respiratory symptoms including pain on deep inspiration, cough, and shortness of breath) [62].

Ambient surface ozone (O₃), a secondary air pollutant, is a well-known respiratory irritant of great concern due to its toxicity to the human cardiorespiratory system and its adverse health effects proven by several environmental epidemiological studies [63,64]. Ozone gas occurs both in the Earth’s stratosphere and at the surface level. If stratospheric ozone is considered to be “good” protection from ultraviolet rays, in the troposphere and at the surface level it is a secondary air pollutant produced by a series of complex photochemical reactions, which involve solar radiation and ozone-precursors [65,66]. As a major greenhouse gas, O₃ has a significant contribution to climate change [67]. The surface formation of O₃ is dependent on its precursors’ relative concentrations in the atmosphere, as well as on local and regional meteorological conditions. Variability in the ground air temperature, wind speed, and direction, relative humidity, precipitation, Planetary Boundary Layer height, and solar surface irradiance associated with climate change has a high potential to affect the generation, distribution, and deposition of O₃.

Ozone is responsible for the inflammatory response of the cardiorespiratory system, being a potential oxidizer and pulmonary irritant [65].

Nitrogen dioxide (NO₂), a well-recognized traffic emissions tracer, has been associated with multiple adverse health outcomes. NO₂, an atmospheric pollutant gas that exacerbates bronchitis symptoms, is responsible for decreased lung function development, respiratory inflammation, responsiveness, infections, and symptoms [68,69]. The anthropogenic NO₂ in the ambient air is mainly from the combustion processes primarily from road transport (41%), energy production (22%), and energy use from households and commercial activities (13%), as well as from industry (13%) [70]. A multi-city analyses conducted in China and Europe provided evidence supporting the long-term and short-term consistent association between NO₂ and increased morbidity and mortality risk [71]. The pattern trends of the ground O₃ and NO₂ are strongly anti-correlated, showing that O₃ is strongly depressed by high NO₂ concentrations [72]. It is expected that climate change will increase the number of high ozone days in urban areas around the globe, with associated adverse impacts on respiratory health and an increased risk of respiratory infection [73].

2.2. Study Test Metropolitan Areas

This study comparatively analyzed the impact of urban air pollution under climate variability on the coronavirus disease (COVID-19) pandemic incidence and lethality in Wuhan in central China and four European metropolises (Milan in Italy, Madrid in Spain, London in the UK, and Bucharest in Romania), which have been selected to provide spatial representativeness and higher air pollution gradients for PM2.5, PM10, O₃, and NO₂ (Figure 1).

The Wuhan metropolitan region, centered at 30.58° N; 114.27° E, the capital of Hubei Province, is located in southeastern China, at the confluence of the Han and Yangtze rivers, on the eastern margin of the Jianghan Plain and the southern foot of Dabie Mountain. Wuhan is the ninth most densely populated city in China, having a high level of air pollution, favoring the chronic exposure of inhabitants and increased risk of cardiorespiratory diseases. The altitude of the study test area varies from 19.2 m to 873.7 m, with a mean value below 50 m. The city center is low and flat, surrounded by low mountains and hills. The climate is temperate, with relatively cold winters and hot, muggy, and rainy summers, while in winter, cold air can stagnate on the ground. The annual mean temperature is in the range of 15.8 °C to 17.5 °C, and the annual mean precipitation is in the range of 1150 mm to 1450 mm. The annual mean wind speed intensity ranges from 1.56 m/s to 2.73 m/s with the main directions of NNE and NE. Wuhan has a population of more than 13.65 million residents, and a total surface area of over 57,800 km². It has 13 administrative districts (Jiangan, Jianghan, Qiaokou, Hanyang, Wuchang, Qingshan, Hongshan, Dongxihu, Hannan, Caidian, Jiangxia, Huangpi, and Xinzhou). Due to last year’s fast urban growth and rapid industrialization, Wuhan was affected by increased levels of air pollution. The main components of aerosol sources are PM2.5, PM10, NO₂, and O₃ [74].

Milan (45.47° N; 9.22° E), a metropolitan test site is located in the Po Valley region, in the northern part of the Lombardy region, one of the most important Italian towns owing to its key role in the Italian economy, is also one of the most densely populated cities, with 1.4 million inhabitants distributed over 181 km². The whole Milan metropolitan area of 3632 km² has a population of about 3.24 million inhabitants. Milan is considered one of the most polluted cities in Europe [75]. The area is characterized by adverse climatic conditions such as frequent thermal inversion during anticyclonic conditions and persistent fog during the fall and winter seasons, which causes the accumulation of high levels of air pollutants’ concentrations enhancing the chronic exposure of inhabitants to increased risk.

The Madrid (40.42° N, 3.70° W) metropolitan area with a surface of 8026 km² is located in the center of the Iberian Peninsula, on the high Castilian Central Plateau. It has a population of 6.98 million inhabitants, being the largest metropolitan area in Spain, and the third-largest city in the European Union. Its climate is Mediterranean with continental influences, characterized by hot summers and cool winters. The urban area is settled on an uneven plain approximately 700 m high, with the lowest altitudes of the basin located in the south and southeast, away from the mountains [76]. During summer, topographical and thermal atmospheric circulations develop on a regional scale resulting in severe O₃ pollution episodes. Road traffic-related air pollution is the main source of O₃ precursors in the basin, representing 65% of NO_x, 67% of CO, 87% of PM10, 85% of PM2.5, and 14% of the total VOC emissions per year, respectively.

The London (51.33° N, 0.42° W) study area, the largest European city and the second-greatest economic center globally with heavy traffic, is located in South East England and has a population of nearly 14.4 million inhabitants within an area of 1738 km². The UK is defined as having a temperate oceanic climate, with cool winters, warm summers, and precipitation fairly evenly distributed all year [77]. London’s topography is predominantly flat low land terrain with a mean elevation of 42 m, except the North Downs on the southern border, reaching up to 200 m above sea level, and the Chiltern Hills on the north-western border, reaching up to 160 m.

The Bucharest metropolitan (44.43° N, 26.09° E) area, located in the southeastern part of Romania and the southeastern part of Europe, is considered the greatest urban carbon emitter city in Romania, and among the most polluted metropolitan cities in Europe. It is placed in a flat area, with a total surface of 5080 km², and has about 1.88 million residents [78]. The metropolitan area of Bucharest is made up of 7 administrative counties. Bucharest’s traffic follows an increasing trend, mainly due to extended old car use. Another source of high levels of air pollution within the city, which sometimes exceed critical standard limits for Romania and the European Union, is heating based on fossil fuels such as coal and natural gas.

Despite general downward trends recorded in emissions over recent years, all European and Wuhan metropolises still present exceedances of the air quality legal limits according to the Directive 2008/50/EC in Europe, and, respectively, Chinese standards according to the Air Pollution Prevention and Control Action Plan (APPCAP) implemented in September 2013 by the Chinese authorities. During the lockdown periods, all of the investigated cities enforced policies to reduce anthropogenic aerosol emissions in urban areas towards mitigating the aerosol adverse health effects.

2.3. Data Used

To analyze the impact of the first COVID-19 wave on air quality in the selected metropolitan regions, this study used available in situ monitoring and reanalysis datasets’ information from different sources. This research used the COVID-19 window period from 1 January 2020 to 15 June 2020. Time-series datasets for outdoor air pollutants and climate parameters were provided by city monitoring networks and several satellite platforms. All COVID-19 incidence and lethality data, namely daily new cases—DNCs, and daily new deaths—DNDs, total COVID-19, and total death cases, were delivered by COVID-19 information webpages [79,80].

The daily derived total Aerosol Optical Depth at 550 nm data (MODIS Terra—AOD) products were obtained from NASA (National Aeronautics and Space Administration)—Giovanni portal (Geospatial Interactive Online Visualization and Analysis Infrastructure) [81]. This study used the daily mean time-series climate data (air temperature—T at 2 m height, air relative humidity—RH, air pressure—p, wind speed intensity—w and direction, Planetary Boundary Layer height—PBL) for the study period and selected metropolitan regions from MERRA-2 Version 2 (Modern-Era Retrospective Analysis for Research and Applications) [82] and C3S (Copernicus Climate Change Service) [83]. The daily mean time series at the ground level of the PM2.5, PM10, O₃, and NO₂ concentrations have been provided by local networks or the AQICN (World Air Quality Index) [84], and for Wuhan also from an online air quality monitoring and analysis platform [85]. To describe the urban air quality of the selected metropolitan areas, this paper considered the Global Air Quality Index (AQI), defined according to the classification of air quality and EU regulations [86], which is described by the following formula:

$$AQI=Max\left({\displaystyle \frac{O_3(24 \mathrm{h})}{100}} , {\displaystyle \frac{N{O}_2(24 \mathrm{h})}{90}} , {\displaystyle \frac{P{M}_{10}(24 \mathrm{h})}{50}} , {\displaystyle \frac{S{O}_2(24 \mathrm{h})}{125}} , {\displaystyle \frac{CO(24 \mathrm{h})}{\mathrm{10,000}}}\right)$$

(1)

where O₃ (24 h), PM10 (24 h), NO₂ (24 h), SO₂ (24 h), and CO (24 h) represent the daily mean values, respectively, of ozone, particulate matter in size 10 μm, nitrogen dioxide, sulfur dioxide, and carbon monoxide present in urban air. Based on the global criteria for the main air pollutants (O₃, PM10, NO₂, SO₂, CO), air quality is classified into six classes from very good to very poor, as presented in

Table 1. Air Quality Index (AQI) classification.

AQI	<10	10–20	20–30	30–50	50–80	>80
Class	Very good	Good	Satisfactory	Sufficiently	Poor	Very poor

:

2.4. Statistical Methods

Cross-correlation analysis was used to evaluate the similarity between two time-series datasets of the ambient daily mean PM in two size fractions (PM2.5 and PM10) and the daily mean total AOD at 550 nm levels, climate observables (air temperature at 2 m height, air relative humidity, wind speed intensity, and air pressure), and daily new COVID-19 incidence and mortality in metropolitan areas. The dependence between pairs of the daily mean time-series datasets was determined in this study by statistical standard tools, Spearman rank-correlation, and rank-correlation non-parametric test coefficients, as well as linear regression analysis. Spearman’s r coefficient quantifies how well the relationship between two variables can be represented by a monotonic function, without any linearity assumption, whether they are linear or not. The normality of the daily mean time-series datasets was assessed through Kolmogorov–Smirnov tests of normality. Because the daily new COVID-19 cases (DNCs) and daily new COVID-19 deaths (DNDs) have a non-normal distribution, Spearman rank correlation was used to identify the linear correlation between the important variables: (1) air pollutants’ PM2.5, PM10 concentrations, total Aerosol Optical Depth at 550 nm, and climate variables; and (2) COVID-19 incidence and mortality rates. OriginPro 2021b software or Microsoft Windows was employed for data processing.

3. Results and Discussion

3.1. Air Pollutants Impacts on COVID-19 Disease in the Metropolitan Areas

3.1.1. Air Pollutants and Air Quality Index Variability

To quantify the short- and medium-term effects of air pollution on fast COVID-19 viral infection diffusion, and quantify the lockdown effect on air pollutants’ levels, the daily mean concentrations of PM2.5, PM10, O₃, and NO₂, were computed for two time periods: (1) pre-lockdown (from 1 January 2020–22 January 2020 for Wuhan in China, and from 1 January 2020–15 March 2020 for European cities); (2) the lockdown (23 January 2020–8 April 2020 and beyond until 15 June for Wuhan, and 15 March 2020–15 June 2020 for the lockdown and beyond for European cities).

For each city, the daily mean recorded concentrations of the investigated air pollutants PM2.5, PM10, NO₂, and O₃ concentrations were compared. The main objective of this study was to estimate the impact of a substantial reduction in traffic and industrial activity during the COVID-19 lockdown period on COVID-19 incidence.

Based on the data presented in

Table 2. The average daily mean air pollutant concentrations, AQI, and AOD levels for the selected metropolitan areas from 1 January 2020–15 June 2020, which covers two time periods (pre-lockdown; lockdown, and beyond).

Average Daily Mean Value	Period	Wuhan	Milan	Madrid	London	Bucharest
PM2.5	Pre-lockdown	(63.55 ± 24.21) In the range (22–104) (µg/m3)	(49.07 ± 17.82) In the range (10–100) (µg/m3)	(28.64 ± 12.24) In the range (10–52) (µg/m3)	(21.93 ± 10.01) In the range (9–44) (µg/m3)	-
	Lockdown and beyond	50.67 ± 15.56 In the range (17–88) (µg/m3)	(21.60 ± 12.78) In the range (4–65) (µg/m3)	(20.43 ± 10.57) In the range (8–65) (µg/m3)	(21.60 ± 8.69) In the range (10–54) (µg/m3)	-
PM10	Pre-lockdown	(147.09 ± 38.64) In the range (70–194) (µg/m3)	(118.91 ± 38.30) In the range (30–89) (µg/m3)	(71.44 ± 23.44) In the range (32–126) (µg/m3)	(50.76 ± 20.80) In the range (24–116) (µg/m3)	(23.41 ± 11.02) In the range (7–54) (µg/m3)
	Lockdown and beyond	(117.46 ± 28.55) In the range (47–184) (µg/m3)	(58.03 ± 24.99) In the range (13–51) (µg/m3)	(51.70 ± 17.28) In the range (23–109) (µg/m3)	(47.15 ± 15.13) In the range (20–15) (µg/m3)	(26.03 ± 17.76) In the range (7–140) (µg/m3)
O3	Pre-lockdown	(17.09 ± 9.38) In the range (2–32) (µg/m3)	(11.59 ± 9.19) In the range (2–32) (µg/m3)	(15.54 ± 7.56) In the range (1–30) (µg/m3)	(21.63 ± 8.43) In the range (0–33) (µg/m3)	(17.07 ± 6.08) In the range (3–29) (µg/m3)
	Lockdown and beyond	(48.39 ± 20.37) In the range (17–117) (µg/m3)	(37.09 ± 10.28) In the range (12–57) (µg/m3)	(31.89 ± 7.03) In the range (13–51) (µg/m3)	(31.49 ± 6.68) In the range (10–49) (µg/m3)	(26.59 ± 7.03) In the range (10–46) (µg/m3)
NO2	Pre-lockdown	(23.73 ± 6.53) In the range (15–44) (µg/m3)	(33.93 ± 7.96) In the range (17–57) (µg/m3)	(28.07 ± 10.03) In the range (14–62) (µg/m3)	(28.72 ± 7.49) In the range (9–42) (µg/m3)	(15.70 ± 5.69) In the range (6–30) (µg/m3)
	Lockdown and beyond	(18.27 ± 8.02) In the range (18–42) (µg/m3)	(19.96 ± 9.85) In the range (4–39) (µg/m3)	(13.07 ± 8.62) In the range (12–43) (µg/m3)	(21.41 ± 8.33) In the range (6–42) (µg/m3)	(9.11 ± 5.22) In the range (2–25) (µg/m3)
AQI	Pre-lockdown	(88.45 ± 34.25) In the range (42–142)	(48.35 ± 24.90) In the range (17–114)	(30.39 ± 10.48) In the range (16–62)	(24.95 ± 7.70) In the range (16–44)	(36.59 ± 15.10) In the range (14–71)
	Lockdown and beyond	(66.27 ± 22.46) In the range (20–128)	(29.91 ± 10.20) In the range (15–69)	(30.56 ± 6–59) In the range (16–50)	(26.81 ± 7.43) In the range (16–57)	(29.72 ± 6.26) In the range (19–48)
AOD	Pre-lockdown	(0.19 ± 0.08) In the range (0.08–0.33)	(0.24 ± 0.12) In the range (0.04–0.65)	(0.14 ± 0.07) In the range (0.05–0.43)	(0.12 ± 0.04) In the range (0.05–0.25)	(0.13 ± 0.05) In the range (0.6–0.29)
	Lockdown and beyond	(0.29 ± 0.14) In the range (0.08–0.80)	(0.29 ± 0.11) In the range (0.14–0.73)	(0.20 ± 0.11) In the range (0.07–0.61)	(0.20 ± 0.09) In the range (0.07–0.61)	(0.26 ± 0.14) In the range (0.06–0.68)

, in comparison with the pre-lockdown period, during the lockdown and beyond lockdown, the particulate matter PM10 recorded a reduction of 20.1% for Wuhan, 52% for Milan, 27.9% for Madrid, and 7% for London, and an increase of 1.09% for Bucharest. The particulate matter PM2.5 decreased during the lockdown and beyond lockdown, with 57% for Milan, 29% for Madrid, 21% for Wuhan, and 2% for London. This study found low reductions in the primary anthropogenic emissions of PM in London and Wuhan.

The findings of this study highlight that, compared with the pre-lockdown period, during lockdown there was a significant improvement in air quality in all of the investigated metropolises. On average, PM25 reductions were estimated to be about 56% for Milan, 29% for Madrid, 21% for Wuhan, and 2% for London. Despite the drastic reduction in road traffic and some industries, the reduction in PM2.5 concentrations was less than expected, explained probably by the increased contributions from agricultural biomass and domestic burning, and climate conditions favoring high secondary aerosol formation.

During the first COVID-19 wave lockdown and beyond until 15 June 2020, due to reduced emissions from road dust, vehicle wear, and construction/demolition, in comparison with the pre-lockdown period, particulate matter PM10 concentrations decreased by 52% in Milan, 28% in Madrid, 21% in Wuhan, and 1.5% in London. This analysis indicates that the highest reductions in PM10 concentrations were recorded in the Milan and Madrid metropolitan areas, followed by Wuhan. The particulate matter concentration decrease was low in London, while there was a small increase in Bucharest. A possible explanation for Bucharest was an April Saharan dust intrusion episode. However, in addition to lockdown restrictions, the seasonal decreased trend of particulate matter must be considered. During and beyond lockdown, the average Air Quality Index AQI recorded a significant reduction with 38% in Milan, followed by Wuhan at 26%, and 18% in Bucharest, and an increase in AQI was found for Madrid at 5% and London at 7.5%. During and after lockdown, the average daily mean ground level O₃ concentrations increased in all of the selected metropolises, namely with a 3.24 factor for Milan, with a 2.84 factor for Wuhan, with a 2.05 factor for Madrid, with a 1.47 factor for London, and with a 1.57 factor for Bucharest. Again, the spring seasonal trend had its contribution to the ground-level ozone concentration increase. Also, during and beyond the lockdown, the average daily mean ground-level NO₂ concentrations decreased as follows: 44% for Wuhan, 42% for Milan, 54% for Madrid, 26% for London, and 42% for Bucharest. The decrease in NO₂ concentrations is partly attributed to seasonal variation but mostly to the reduction imposed on traffic and industrial sources. This means that COVID-19 had a significant positive impact on greenhouse gas (GHG) emissions across the selected metropolises.

The increase in ground-level ozone concentrations during and beyond the lockdown period in all of the analyzed metropolitan regions (Figure 2) is mainly explained by a significant reduction in related traffic NOx emissions and complex chemistry involved in O₃ formation in volatile organic compound (VOC)-NOx mixtures; as NOx concentrations are decreased, a greater number of OH radicals are available to react with VOCs, which leads to greater ozone formation.

Ozone is also eliminated due to its rapid reaction with NO, which, being less available to titrate the ozone, produces an increase in the atmosphere [87,88,89,90]. In addition to this process, spring seasonal variation might be considered. The highest increased ground-level ozone concentrations have been registered in the Wuhan metropolitan region. For effective emission control policies and a better understanding of different air pollution source contributions, in particular, O₃ production is very important to assess lockdown effects on air quality in large cities. Among the dominant sectors contributing to air pollution in cities, the urban transport, industrial, commercial, household, and institutional sectors were strongly impacted by the lockdown measures in China, Italy, Spain, the UK, and Romania. Our results are in good agreement with other published studies that reported decreases in PM10, PM2.5, and NO₂ levels and increases in O₃ concentrations in large metropolitan regions worldwide during the first COVID-19 outbreak [91,92,93,94,95,96,97]. Similar results based on FLEXPART-WRF and WRF-Chem modeling experiments were reported for the registered decreased levels of PM2.5 concentrations in Wuhan between pre-lockdown and lockdown [98].

According to previously published studies [99,100], the daily mean ground-level ozone concentrations were negatively correlated with the daily average concentrations of particulate matter PM2.5 and PM10 for all of the investigated cities, namely PM2.5 [(r = −0.64 with p < 0.01) for Milan; (r = −0.38 with p < 0.01) for Madrid; (r = −0.28 with p < 0.01) for London; (r = −0.48 with p < 0.01) for Wuhan] and PM10 [(r = −0.65 with p < 0.01) for Milan; r = (−0.48 with p < 0.01,) for Madrid; r = (−0.28 with p < 0.01) for London; (r = −0.37 with p < 0.01) for Bucharest; and (r = −0.29 with p < 0.01} for Wuhan]. Also, statistical Spearman correlation coefficients show a positive relationship of PM in both size fractions of the daily mean PM2.5 and PM10 with ground-level NO₂ concentrations as follows: for PM2.5 (r = 0.62 with p < 0.01) for Milan; (r = 0.37 with p < 0.01) for Madrid; (r = 0.18 with p = 0.05) for London; (r = 0.18 with p = 0.02) for Wuhan, and for PM10 (r = 0.69 with p < 0.01) for Milan; (r = 0.50 with p < 0.01) for Madrid; (r = 0.15 with p = 0.05) for London; (r = 0.25 with p < 0.01) for Bucharest; (r = 0.60 with p < 0.01) for Wuhan.

The risk of infection from pathogen-bearing particulate matter and other aerosols is very high in urban agglomerated regions. Furthermore, the main sources of airborne microbes and viruses come from local, regional, or transboundary sources. The first experimental evidence that SARS-CoV-2 RNA can be attached to outdoor particulate matter PM10 in defined climate conditions of atmospheric stability and high concentrations of pollutants has been provided for Bergamo city in Lombardy province in Northern Italy during the COVID-19 pandemic. This achievement was very important for COVID-19 science, suggesting a possible use of this test as an indicator of epidemic recurrence [101]. It seems that PM emissions from road traffic affected COVID-19 viral infection severity more than infection risk [102]. A crucial role in SARS-CoV-2 infection and mortality levels is played by pre-existing immune disorders attributed to long-term or short-term exposure to high air surface concentrations of particulate matter and gaseous pollutants [103,104]. According to scientific literature in the field, in addition to air pollution, there are several other factors responsible for the etiology and severity of COVID-19 symptoms, like existing comorbidities, immunity system, the patient’s age, sex, genetic and nutritional status, etc. The temporal pattern of daily mean NO₂ ground levels during 1 January–15 June 2020 is presented in Figure 3 for all analyzed metropolises. In the pre-lockdown period, the highest average daily NO₂ concentrations have been recorded in Milan and Madrid cities, followed by Wuhan, London, and Bucharest, being attributed to significant traffic-related pollution sources.

The temporal pattern of the Air Quality Index (AQI) in the pre-lockdown period and during the lockdown period for all analyzed metropolitan regions is shown in Figure 4. However, the improvement in urban air quality did not show consistent temporal patterns among investigated cities. This inconsistency is associated with differences in the emission intensities of air pollutants and the location of fixed and mobile air quality monitoring stations, prevailing local atmospheric dynamics, chemical processes of reactive air pollutants induced by atmospheric oxidants, and their removal by atmospheric processes. As Table 2 shows, the results indicate that due to the implementation of the reduction in air pollution sources, during the lockdown and beyond period, air quality was significantly improved in Wuhan and Milan, but not in the Madrid metropolis. Additionally, the pollutant concentrations during the same lockdown period in the prior five years (2015–2019) were assessed. Like in other studies, the comparison in this study with the average values for the lockdown period with the same period of 2015–2019, has shown a low improvement in air quality during the lockdown in all the investigated European metropolises (7.3% for PM2.5 in Madrid, 11.1% for PM2.5 in London, 21.2% for PM2.5 in Milan) [105].

As in other countries, a lockdown response to COVID-19, air quality was improved due to a reduction in industrial and general economic activity in the selected study areas in China and Europe, with a high impact on tropospheric and ground-level pollution with particulate matter and gaseous pollutants [106,107,108].

3.1.2. Aerosol Optical Depth Temporal Pattern in Metropolitan Areas

During the analyzed COVID-19 pre- and pandemic lockdown period, the Aerosol Optical Depth (AOD) derived from the MODIS Terra satellite data presents a clear annual variation with maxima in spring and summer (sometimes associated with the transboundary dust intrusions), and minima in autumn and winter. Despite the COVID-19 outbreak in the spring of 2020, the subsequent reduction in mobility and physical contact, and the decrease in international tourism, compared with the same time window (March) for the non-pandemic (2015–2019) period, during the lockdown period AOD levels recorded different variations in the studied metropolises. Changes in the total AOD at 550 nm during the first COVID-19 wave and associated lockdown are not significantly different from the long-term and year-to-year variability in AOD in the investigated cities; small changes indicate that registered reductions in anthropogenic aerosol emissions have a low effect in changing the net contribution of aerosol scattering and absorption to total aerosol extinction. Figure 5 presents the temporal patterns of the daily mean AOD in the investigated metropolises from 1 January 2019 to 15 June 2020. The higher levels of AOD recorded during February and March 2020 in Wuhan, and during March and April in European metropolises may explain the reported highest rates of daily COVID-19 incidence cases (DNCs) (Figure 6), as well as the total COVID-19 incidence cases (DNCs), and the total COVID-19 deaths (DNDs) presented in Figure 7 and Figure 8, respectively.

As COVID-19 transmission is more dependent on environmental and demographic variables, the adopted strict social restrictions and public health interventional policies in the Wuhan metropolis resulted in an abrupt decrease in the middle of February 2020 of daily new incidence cases of DNCs (Figure 6).

3.2. Air Pollution and Climate Variability Impact on the First COVID-19 Wave

Also, the relationships between the air pollutants (PM2.5, PM10, O₃, and NO₂) and meteorological variables (air temperature at 2 m height, air relative humidity, wind speed, and Planetary Boundary Layer height) were investigated during the first wave of COVID-19 lockdown and beyond until 15 June 2020 using Spearman rank correlation analysis. This analysis would improve the understanding of the mechanisms of air pollution episode evolution under diverse local and regional climate conditions and suggest potential ways of reducing air pollution in large metropolitan areas. Furthermore, correlation analyses between the main air pollutants and COVID-19 incidence cases were performed to help ascertain the emission sources responsible for the reduction in concentrations of air pollutants during the lockdown periods. While comparing major air quality parameters with COVID-19-related incidence cases during the first wave of COVID-19 lockdowns and beyond until 15 June 2020, we found negative correlations between surface ozone concentrations and positive correlations between surface nitrogen dioxide and COVID-19 incidence in all of the analyzed cities (

Table 3. Spearman rank correlation coefficients between daily COVID-19 incidence DNC cases, daily mean air quality, Aerosol Optical Depth, and climate variables for the selected metropolitan areas during lockdowns and beyond until 15 June 2020.

Daily Average Variable	Wuhan	Milan	Madrid	London	Bucharest
	DNC	DNC	DNC	DNC	DNC
Air Quality Index (AQI)	0.37 *	0.32 *	0.35 *	0.39 *	0.56 *
Aerosol Optical Depth (AOD)	0.31 *	0.27 *	0.42 *	0.37 *	0.14
O3 (Ozone) (µg/m3)	−0.57 *	−0.32 *	−0.42 *	−0.37 *	−0.29 *
NO2 (µg/m3)	0.45 *	0.42 *	0.65 *	0.12	0.49 *
T (air temperature at 2 m height) (°C)	−0.89 *	−0.39 *	−0.76 *	−0.73 *	−0.57 *
RH (relative humidity) (%)	0.15	0.39 *	0.53 *	0.64 *	0.36 *
w (wind intensity) (m/s)	−0.25 *	−0.24 *	−0.12	−0.49 *	−0.18
PBL (Planetary Boundary Layer height) (m)	−0.49 *	−0.35 *	−0.14	−0.11	−0.60 *

Note: p value: * p ≤ 0.05—significant values; without * indicate p ≥ 0.05—nonsignificant values.

).

The findings in Table 3 also reported positive correlations between COVID-19 incidence DNC cases, Air Quality Index, and Aerosol Optical Depth, and air relative humidity for all investigated metropolises, and negative correlations between the air temperature at the 2 m height, Planetary Boundary Layer height, and wind speed intensities for Wuhan, Milan, Madrid, London, and Bucharest. As

Table 4. The average daily mean of Planetary Boundary Layer height levels for the selected metropolitan areas from 1 January 2020 to 15 June 2020 (pre-lockdown; lockdown and beyond).

Average Daily Mean Value	Period	Wuhan	Milan	Madrid	London	Bucharest
PBL (m)	Pre-lockdown	453.59	445.82	686.83	1103.83	727.02
	Lockdown and beyond	585.8	1300.76	1768.04	1215.25	1749.23

shows, significant variation was detected in the PBL heights between the selected metropolitan areas, with lower levels of Planetary Boundary Layer height recorded particularly over Wuhan, Milan, and Madrid during the pre-lockdown period and the first wave of COVID-19 lockdown and beyond until the middle of June 2020, which may also explain the high COVID-19 incidence rates.

However, the findings of this study highlight the crucial role of urban aerosol loading and pollutant gases in synergy with climate variability on COVID-19 pandemic evolution in large metropolitan areas. Local and regional outdoor-specific climate conditions (air temperature, relative humidity, wind speed intensity and direction, and Planetary Boundary Layer height) can be top predictors of airborne coronavirus illness diffusion. Like other studies, this research suggests that air quality in large metropolitan areas can be improved by a reduction in traffic-related emissions [109,110,111,112,113,114,115,116,117,118,119,120,121,122].

As

Table 5. Summary of population size, total COVID-19 incidence DNCs, and total COVID-19 deaths DNDs for metropolitan areas between 1 January 2020 and 15 June 2020.

Metropolis	Wuhan	Milan	Madrid	London	Bucharest
Population size (million inhabitants)	13.65	3.24	6.98	14.4	1.88
Total COVID-19 cases (DNCs) during March 1 January 2020–15 June 2020	50,441	23,919	74,836	27,875	2420
Total COVID-19 deaths (DNDs) during March 1 January 2020–15 June 2020	3869	3465	6974	3919	108

shows, the greatest total COVID-19 incidence (DNCs) and total death cases were registered during the first COVID-19 wave in the Madrid metropolis, followed by the Wuhan metropolis, London, and Milan. The lowest values of COVID-19 incidence (DNCs) and mortality cases were recorded in the Bucharest metropolis.

An interesting study that explored the strategies and practices of the Chinese government risk communication adopted during the COVID-19 lockdown in Wuhan provided significant implications for effective health risk communication at the early stage of the epidemic response, mainly based on drastic epidemic prevention and control, which limited the viral disease transmission [123]. Other studies suggested that by substantially reducing air pollutant sources and human mobility within and between cities, lockdowns limited COVID-19 spreading [124,125,126]. Like other previous studies, our findings suggest that large agglomerated metropolitan areas will be a key factor influencing future transmission of viral infection events. Also, this study highlights the importance of updating urban policies related to the thresholds of air pollution exposures during pandemic events and adopting urgent strategies for inhabitants’ protection from harmful environmental stressors.

4. Conclusions

This paper focused on identifying key environmental factors, such as ambient air pollution and climate factors, that could increase the severity of the health outcomes of COVID-19 in Wuhan in central China, and a few densely European metropolises (Milan, Madrid, London, and Bucharest). In summary, this study used a comprehensive time-series analysis of the key air pollutants of particulate matter PM2.5, PM10, gaseous pollutants O₃ and NO₂, Air Quality Index, and Aerosol Optical Depth data together with climate and coronavirus data for the period 1 January–15 June 2020 to provide additional evidence on the possible impacts of ground air pollution concentrations on the fast diffusion of SARS-CoV-2 pathogens. The main air pollutants (inhalable PM2.5 and PM10, NO₂), AQI, and aerosol loading AOD revealed a direct impact and an association with COVID-19 spreading and severity. The results show that meteorological changes together with air pollutant emissions reduction in the investigated metropolises explain the total decrease in air pollutant concentrations during the lockdown period. Chronic or short-term exposure of cities’ inhabitants to high concentration levels of PM2.5, PM10, O₃, NO₂, or other air pollutants strongly affects the human immune system, the pathogenesis of severe respiratory infections, and fast transmission in European and Chinese towns). The different types of lockdowns implemented in every metropolis based on the severity of the COVID-19 pandemic resulted in a significant reduction in the main outdoor air pollutants, especially particulate matter and nitrogen dioxide, in all metropolises, especially in Wuhan and Milan. Also, there were recorded increasing trends in ground ozone levels, probably attributed to nonlinear chemistry associated with the reduction in traffic-related oxides of nitrogen (NO_X).