Assessment of Air Pollution before, during and after the COVID-19 Pandemic Lockdown in Nanjing, China

: A unique illness, the coronavirus disease 2019 (COVID-19), emerged in Wuhan, People’s Republic of China, in December 2019. To reduce the spread of the virus, strict lockdown policies and control measures were put in place all over the world. Due to these enforced limitations, a drastic drop in air pollution and an improvement in air quality were observed. The present study used six air pollutants (PM 10 , PM 2.5 , SO 2 , NO 2 , CO and O 3 ) to observe trends before, during and after the COVID-19 lockdown period in Nanjing, China. The data were divided into six phases: P1–P3, pre-lockdown (1 October–31 December 2019), lockdown (1 January–31 March 2020), after lockdown (1 April–30 June 2020), P4–P6: the same dates as the lockdown but during 2017, 2018 and 2019. The results indicate that compared with the pre-lockdown phase, the PM 10 and PM 2.5 average concentrations decreased by –27.71% and –5.09%. Compared with the previous three years, 2017–2019, the reductions in PM 10 and PM 2.5 were –37.99% and –33.56%, respectively. Among other pollutants, concentrations of SO 2 (–32.90%), NO 2 (–34.66%) and CO (–16.85%) also decreased during the lockdown, while the concentration of O 3 increased by approximately 25.45%. Moreover, compared with the pre- and during lockdown phases, PM 10 , PM 2.5 and NO 2 showed decreasing trends while SO 2 , CO and O 3 concentrations increased. These ﬁndings present a road map for upcoming studies and provide a new path for policymakers to create policies to improve air quality. The positive correlation between NO 2 and CO was due to the positive correlation between particulate matter (PM) and CO. It suggests that O 3 levels increased with the reduction in the other pollutants, especially NO 2 and CO.


Introduction
A new type of virus, a novel coronavirus disease, was first identified at the end of December 2019 in the capital city of China's Hubei Province, Wuhan. The World Health Organization (WHO) officially called it the coronavirus disease 2019 (COVID-19) [1]. The initial symptoms of COVID-19 are similar to those of SARS and MERS [2]. After the first confirmed case, the number of infected people increased rapidly all over the country, and within just a few days, due to the spread of COVID-19, it turned into a national disaster [3,4]. At the end of June 2020, in 215 countries around the world, more than 10.6 million people had been infected and 514,315 people had died as a result of the spread of COVID-19. The WHO confirmed COVID-19 as worldwide public health emergency due to its large-scale spread and contagious effects [2]. The prevention and treatment of the virus' infection has become a significant task [5]. By early August 2020, more than 19.46 million confirmed cases of COVID-19 were observed in over 200 countries and territories around the world, according to the WHO [4]. satellite data set. The authors found a significant decrease in the concentrations of the air pollutants during the lockdown. Goldberg et al. [36] reported that meteorological patterns have a major impact on air pollutants and observed a decrease in air pollutants in spring 2020 due to these meteorological variations. Li et al. [37] developed a regression model for predicting the mortality rate during COVID-19 and studied the impact of air quality on the mortality rate during COVID- 19.
The present study estimated the levels of six air pollutants (i.e., PM 10 , PM 2.5 , SO 2 , NO 2 , CO and O 3 ) in Nanjing, the capital city of Jiangsu Province, using data obtained from nine ground-based monitoring stations. This study determined the impact of the lockdown during COVID-19 on air pollution in Nanjing and compared the results with the previous 3 years to discover significant changes in air pollution. However, recent studies have shown that especially in urban areas and regions, forced limitations and lockdown measures contribute to improvements in air quality for limited time periods (e.g., [38]). Most of the studies used only a few air pollutants to assess the impacts of the COVID-19 lockdown on air quality [39]. Thus, studies assessing the impact of the COVID-19 lockdown on all six air pollutants before, during and after the lockdown period as well as a comparison of these air pollutants with the previous three years (2017-2019) for a deeper analysis, are lacking. Therefore, our study provides a more comprehensive analysis of this scenario, and the results show a remarkable decline in the selected pollutants during the lockdown period in comparison to the previous three years. The objectives of this study were (1) to examine the air pollutant concentrations before, during and after the lockdown period in Nanjing; (2) to explore and quantify the air-quality status due to the nationwide lockdown and strict limitations amid the lockdown phase; (3) to explore the air pollutant concentrations during the same window of time in the last three years. The current study provides credible and useful outcomes, and insights for the scientific community and policymakers on the impact of nationwide lockdowns on air pollution. Further, the study provides a roadmap for upgrading and improving Nanjing's air quality in the future.

Study Area
Nanjing, with its 2000 years of history, is the capital of Jiangsu Province and the economic hub of the Yangtze River Delta. Nanjing is one of the most important and rapidly developing cities in China, with a large population and well-developed industry and manufacturing sectors [40]. In recent years, Nanjing has been faced with extreme and severe air pollution owing to its rapid economic growth and non-stop energy consumption [41]. Nanjing lies between the geo-coordinates 31 • 14 to 32 • 37 and 118 • 22 to 119 • 14 , and occupies an area of approximately 6596 km 2 with 11 districts that can be classified into three regions: urban, suburban and rural. Geographically, the city is located within the Yangtze River Delta. As of 2019, the population of Nanjing was approximately 8.34 million with a density of 8500 people/km 2 in the central area. The city experiences a subtropical monsoon climate with four distinct seasons. The annual average temperature ranges between 16 • C and 38 • C in the summer and is −8 • C in the winter, and the annual precipitation is approximately 1090 mm [42]. In China, Nanjing is one of four garden cities with a per capita green area of 13.7 km 2 , although there are many large areas of industry and enterprise. The leading industries in Nanjing are electronics, automobiles, petrochemicals, and steel smelting. To improve and rectify these polluting sites and enterprises, the administration has taken a number of measures. Additionally, the city is bordered by highlands and mountains on three sides and by water on the other, which can result in the accumulation of air pollutants in urban areas.
Presently in Nanjing, there are nine monitoring stations operating and recording air pollutants. The locations of these stations are shown in Figure 1.

Environmental Data and Study Period
In order to investigate air quality in Nanjing, data on the daily average concentration for the six air pollutants, including particulate matter with a diameter of ≤10 µm (PM10), particulate matter with a diameter of ≤2.5 µm (PM2.5), sulfur dioxide (SO2), nitrogen dioxide (NO2), carbon monoxide (CO) and ozone (O3), were obtained from the Nanjing Environmental Monitoring Center, after being collected from the nine monitoring stations. The data for all concentrations had at least 347 daily values, which meets the annual concentration calculation criteria of the Ambient Air Quality Standards of China, which is 324 daily values. These nine monitoring stations were designed to signify areas with different characteristics and features according to the Technical Regulation for Selection of Ambient Air Quality Monitoring Stations of China. Among these nine national stations, Xianlin represents a cultural area, Pukou represents a suburban area, Caochangmen is an instructive and educational area, Xuanwuhu covers an environmental and ecological park area, Ruijinlu is connected with a residential and housing area, Zhonghuamen represents a transportation and traffic area, Maigaoqiao covers a commercial and industrial area, Aotizhongxin represents a built-up and new city area, and Shanxilu represents a commercial and market area. The Nanjing Environmental Monitoring Center provided the geographical location information for these monitoring sites. To assess the effect of the lockdown before, during and after the COVID-19 period, the data were divided into six periods in 2017-2020: P1-P3, before lockdown (1 October-31 December 2019), lockdown (1 January-31 March 2020), after lockdown (1 April-30 June 2020), and P4-P6, which include the same dates as the lockdown but in the previous three years (2017-2019).

Environmental Data and Study Period
In order to investigate air quality in Nanjing, data on the daily average concentration for the six air pollutants, including particulate matter with a diameter of ≤10 µm (PM 10 ), particulate matter with a diameter of ≤2.5 µm (PM 2.5 ), sulfur dioxide (SO 2 ), nitrogen dioxide (NO 2 ), carbon monoxide (CO) and ozone (O 3 ), were obtained from the Nanjing Environmental Monitoring Center, after being collected from the nine monitoring stations. The data for all concentrations had at least 347 daily values, which meets the annual concentration calculation criteria of the Ambient Air Quality Standards of China, which is 324 daily values. These nine monitoring stations were designed to signify areas with different characteristics and features according to the Technical Regulation for Selection of Ambient Air Quality Monitoring Stations of China. Among these nine national stations, Xianlin represents a cultural area, Pukou represents a suburban area, Caochangmen is an instructive and educational area, Xuanwuhu covers an environmental and ecological park area, Ruijinlu is connected with a residential and housing area, Zhonghuamen represents a transportation and traffic area, Maigaoqiao covers a commercial and industrial area, Aotizhongxin represents a built-up and new city area, and Shanxilu represents a commercial and market area. The Nanjing Environmental Monitoring Center provided the geographical location information for these monitoring sites. To assess the effect of the lockdown before, during and after the COVID-19 period, the data were divided into six periods in 2017-2020: P1-P3, before lockdown (1 October-31 December 2019), lockdown (1 January-31 March 2020), after lockdown (1 April-30 June 2020), and P4-P6, which include the same dates as the lockdown but in the previous three years (2017-2019).

Statistical Analysis and Data Visualization
Summary statistics were performed for different variables during the study period. The current study reports variations in pollutant concentrations before, during and after the lockdown phases and compares the previous three years (2017-2019) with the same time frame as that of the lockdown. The net difference and percentage change in concentrations of the selected pollutants were also investigated. To better understand the impact of the nationwide lockdown on air pollution, this work also reports the temporal evolution in daily average concentrations of the pollutants before, during and after the lockdown periods. Meanwhile, violin plots were generated to represent the concentrations of selected air pollutants during the study period. To assess the relationships among air pollutants, linear regression analysis was performed for different air pollutants during the lockdown period. The current study area map was created with ArcGIS 10.2.2, while other plots were created with Python.

Changes in Different Pollutants before, during and after the Lockdown Phases
Changes in different pollutants (PM 10 , PM 2.5 , SO 2 , NO 2 , CO and O 3 ) before, during and after the lockdown periods are shown in Figure 2. Overall, an increasing trend in PM 10 was observed before the lockdown phase. During lockdown, a diminishing trend in the concentration of PM 10 was observed, and it continued to decrease after the lockdown phase with similar variabilities. The results suggest that the significant reduction in the concentration of PM 10 could be due to the limited vehicular, industrial and construction activities. A remarkable reduction in vehicular and transportation activities could also decrease air pollution [43]. Before the lockdown phase, PM 2.5 demonstrated an increasing trend; at first, during lockdown it showed an increasing trend and later decreased, while after lockdown, the concentration of PM 2.5 decreased significantly. The results suggest that, overall, during the lockdown phase, PM 2.5 showed a constantly decreasing trend.

Statistical Analysis and Data Visualization
Summary statistics were performed for different variables during the study period. The current study reports variations in pollutant concentrations before, during and after the lockdown phases and compares the previous three years (2017-2019) with the same time frame as that of the lockdown. The net difference and percentage change in concentrations of the selected pollutants were also investigated. To better understand the impact of the nationwide lockdown on air pollution, this work also reports the temporal evolution in daily average concentrations of the pollutants before, during and after the lockdown periods. Meanwhile, violin plots were generated to represent the concentrations of selected air pollutants during the study period. To assess the relationships among air pollutants, linear regression analysis was performed for different air pollutants during the lockdown period. The current study area map was created with ArcGIS 10.2.2, while other plots were created with Python.

Changes in Different Pollutants before, during and after the Lockdown Phases
Changes in different pollutants (PM10, PM2.5, SO2, NO2, CO and O3) before, during and after the lockdown periods are shown in Figure 2. Overall, an increasing trend in PM10 was observed before the lockdown phase. During lockdown, a diminishing trend in the concentration of PM10 was observed, and it continued to decrease after the lockdown phase with similar variabilities. The results suggest that the significant reduction in the concentration of PM10 could be due to the limited vehicular, industrial and construction activities. A remarkable reduction in vehicular and transportation activities could also decrease air pollution [43]. Before the lockdown phase, PM2.5 demonstrated an increasing trend; at first, during lockdown it showed an increasing trend and later decreased, while after lockdown, the concentration of PM2.5 decreased significantly. The results suggest that, overall, during the lockdown phase, PM2.5 showed a constantly decreasing trend.  Before lockdown, SO 2 concentrations showed an increasing trend, while during and after lockdown, SO 2 concentrations decreased. The results suggest that the decrease in the concentration of SO 2 was due to the reduced thermal, transportation and industrial activities. The main sources of SO 2 are the burning of fuels such as coal and diesel used in thermal, transportation and industrial enterprises [44]. The stoppage of industrial, vehicular, and heating activities were responsible for the reduction in SO 2 emissions during the lockdown phase in China [10]. Moreover, before the lockdown phase, just as for PM, a rapidly increasing trend in NO 2 was observed, while during lockdown, NO 2 showed a declining trend. After lockdown, NO 2 presented a substantial diminishing tendency.
Before the lockdown phase, CO demonstrated an increasing trend, while during the lockdown phase, within just a few days, CO showed an increasing trend, then a consistent decline. After lockdown, CO presented an increasing trend with some fluctuations compared with the lockdown phase. Furthermore, a significant decreasing trend in O 3 was observed before the lockdown phase, while during lockdown, O 3 showed an increasing tendency. A rapidly growing trend was observed after the lockdown phase owing to the enhancement of solar radiation in the summer. In the summer, emissions of isoprene and biomass burning are responsible for the increment in O 3 [45]. The results suggest that the decrease in all air pollutants during COVID-19 was due to the low level of traffic, decreased human mobility and the closure of all industrial and construction activities [46].

Spatial Concentration Pattern of PM 10
Owing to the COVID-19 pandemic, a lockdown was imposed in many countries to control and minimize the movements of people. In China, the lockdown was implemented in January 2020. This type of shutdown and strict control policies reduce the movement of people and vehicles and restrict industries, markets, universities and construction activities, among others, leading to a drastic change and substantial improvement in air quality in many countries and regions. Our results demonstrate that during the COVID-19 pandemic lockdown phase in China, Nanjing experienced a significant change in PM 10 and a reduction of approximately −27.71% compared with the pre-lockdown phase, due to the restricted traffic and shutdown of industries. Compared with the pre-and during lockdown phases, an approximately −23.65% reduction in PM 10 concentration was recorded. The average concentration of PM 10 and the variations are shown in Figure 3 and Tables 1-3. The main sources of PM 10 concentrations are road dust and soil resuspension, agronomy, transportation and fugitive dust emissions generated by industrial and manufacturing activities [13]. During the COVID-19 lockdown in 2020, these activities diminished significantly, which resulted in a drastic drop and reduction in air quality, especially in cities that mostly experience heavy pollution due to transportation, manufacturing and industrial processes [47]. For instance, it was reported by Shakoor et al. (2020) [23] that an approximately 27% reduction in PM 10 concentration was found in China during the COVID-19 pandemic lockdown period. According to Bao and Zhang 2020 [10], during the COVID-19 pandemic, air pollution decreased by approximately 7.80% in 44 cities in northern China.
Our study shows, compared with 2019 (for the same time frame), the concentration of PM 10 was significantly reduced (36.90%) during the COVID-19 lockdown period. As compared with the previous three years 2017-2019 (same dates of lockdown), a −37.99% decline in the concentration of PM 10 was observed. According to Chen et al. (2020b) [20], the concentration of PM 10 during the COVID-19 pandemic lockdown period decreased in comparison to the previous three years (2017-2019). The results show the variations between the lockdown and post-lockdown periods. During the lockdown period, the reduction in the concentration of PM 10 was observed to be −9.02%. After the lockdown phase, compared to the last three years (2017-2019), a significant reduction in PM 10 concentration was observed, −43.58%. The results suggest that an improvement in PM 10 concentration was evident because of the COVID-19 lockdown. In summary, this significant reduction and improvement in the concentration of PM 10 demonstrates that the COVID-19 pandemic lockdown had a remarkable impact on the reduction of PM 10

Spatial Concentration Pattern of PM 2.5
Particulate matter, also known as PM 2.5 with an aerodynamic diameter of 2.5 µm or less than 2.5 µm, is linked with numerous adverse human health effects, especially respiratory and cardiovascular infections. According to the Global Burden of Disease, particulate matter (PM 2.5 ) is the seventh and fourth main risk factor in the world and China, respectively [15,48]. During the COVID-19 lockdown phase in Nanjing, the PM 2.5 concentration decreased by −5.09% compared with the pre-lockdown phase. Compared with PM 10 , the reduction in the concentration of PM 2.5 was relatively lower. The average concentration of PM 2.5 and variations are presented in Figure 3 and Tables 1-3. Further, compared with the pre-and during lockdown phases, Nanjing experienced a significant decline of −33.41% after the lockdown. Similar decreasing trends in the concentrations of PM 2.5 in many cities across the world have been reported by Singh and Chauhan (2020) [8]. Vehicular emissions, industrial production and construction activities, among others, are the main sources of PM 2.5 concentrations [13]. The results suggest that during the COVID-19 lockdown, reduced human mobility and the stoppage of vehicular and industrial processes subsequently caused a considerable drop in fine particulate matter in the environment. Previously, Chen et al. (2020b) [20] also reported that during the COVID-19 period, PM 2.5 declined in many areas and regions across China.
During the lockdown period a noteworthy improvement and reduction in the PM 2.5 concentration (−34.81%) was found compared with 2019. The previous three years' (2017-2019) data showed a significant reduction in PM 2.5 (−33.56%). The results indicate that the reduction ratio in 2020 was greater compared with 2017-2019 in Nanjing. A similar pattern of PM 2.5 reduction was reported in the USA, using air-quality data between 2017 and 2020 and 2017 and 2019 in China, respectively [20,22]. In the Guangxi Region, compared with 2016-2019, during the lockdown period, an approximately −22.6% reduction in PM 2.5 was observed as reported by   [49]. However, the present study revealed that compared with the lockdown period, the concentration of PM 2.5 reduced by up to −31.63%, and after lockdown it reached −54.57%. Overall, compared with the previous three years, the present study concludes that the lockdown and strict polices caused a notable reduction and improvement in air quality.

Spatial Concentration Pattern of SO 2
The present study shows that due to the nationwide lockdown, a significant improvement in SO 2 was observed. Compared with the pre-and during lockdown phases, SO 2 showed an increasing trend. The results demonstrate that compared with the pre-lockdown phase, an approximately −32.90% reduction in the SO 2 concentration was observed. However, compared with the pre-and during lockdown phases, SO 2 increased by 5.95% in Nanjing. This increment in SO 2 concentration can be attributed to coal burning, power plants and industrial activities after lockdown [13]. The average concentration of SO 2 and variations are presented in Figure 3 and Tables 1-3. The main sources of SO 2 are coal and diesel used in thermal power, burning of fuels, transport and industry [44]. Due to the COVID-19 lockdown, limited industrial, vehicular and heating activities were responsible for the reduction in SO 2 emissions. Previously in China, a similar pattern in SO 2 reduction was reported [10,12]. In addition, it is evident from Shakoor et al.'s (2020) [23] study, that compared with the pre-lockdown phase, SO 2 was reduced by 18% during the lockdown phase in China.
Moreover, compared to the year before (2019) with the same time frame, a sizable reduction and improvement in the concentration of SO 2 was observed, that is, up to −43.77% during the lockdown phase in Nanjing. A reduction in the SO 2 concentration during this period was greater compared with the pre-lockdown phase for the same time frame in Nanjing. Similar results have been reported by Bao and Zhang (2020) [10] and Islam et al. (2020) [39] in China and Bangladesh, respectively. These studies demonstrate that compared to 2019, the SO 2 concentration decreased. Further, our study also found that compared with the previous three years with the same time period (2017-2019), SO 2 showed a sharp decline of up to −57.70% during the lockdown phase. It is evident from the study by Chen et al. (2020b) [20] that concentrations of SO 2 were reduced compared with 2017-2019 in many areas across China. Nanjing experienced a considerable increasing trend in SO 2 up to 31.92% after the lockdown. However, in the previous three years, the concentration of SO 2 decreased by −44.20%. According to   [49], SO 2 was found at its lowest level during the lockdown phase compared with the previous five years in the Guangxi Region, China. Overall, there was a decline in SO 2 concentration throughout the study period in Nanjing, except after the lockdown phase compared with the pre-and during lockdown phases.

Spatial Concentration Pattern of NO 2
The results witnessed during the lockdown phase, show that owing to the COVID-19 pandemic and post-lockdown, Nanjing experienced a significant reduction in NO 2 concentration. The average concentration of NO 2 and variations are shown in Figure 3 and Tables 1-3. The study revealed that compared with the pre-lockdown phase, a −34.66% reduction and improvement in NO 2 concentration was observed. During the lockdown phase, compared with pre-and during lockdown phases, the concentration of NO 2 decreased by −22.70%. The reduction in NO 2 concentration was greater during the lockdown phase. A similar pattern of results was observed by NASA (2020) [21], and NO 2 emissions decreased by 30% in Central China. Shakoor et al. (2020) [23] reported that during the lockdown phase, a 38% decline in NO 2 concentration was found in China. Another study revealed that during lockdown, there was a decline in NO 2 concentration over the Yangtze River Delta [9]. NO 2 is mainly emitted from vehicles, power plants, road traffic and industrial and manufacturing activities in urban areas and regions. Due to the nationwide lockdown, all of these sectors closed, which resulted in the decline in pollutants such as NO 2 .
During the lockdown, Nanjing was found to experience a −27.36% drop in NO 2 concentration compared to the previous year, 2019. It was reported by   [49] that NO 2 showed a declining trend during the lockdown phase compared with the previous 5 years. Compared to the 2019 measurements for the same dates, a 34.7% drop in the concentration of NO 2 was found in North China. Compared with 2019, the concentration of surface and tropospheric NO 2 decreased by up to 20-30% in China [25]. It is evident from the study by Chen et al. (2020b) [20] that in many areas across China, the concentration of NO 2 decreased during 2017-2019. According to Berman and Ebisu 2020 [22], between 2017 and 2020, NO 2 emissions decreased by −26% during the lockdown phase in the USA. Moreover, compared to the lockdown phase, the concentration of NO 2 declined by up to −2.20% after the lockdown. Compared to the previous three years, for the same time frame (2017-2019), a significant decline of −33.89% in NO 2 concentration was observed after the lockdown phase. In summary, a drastic change and improvement in the concentration of NO 2 showed that the lockdown and strict policies had a significant impact on changing and decreasing the NO 2 concentrations in Nanjing.

Spatial Concentration Pattern of CO
During the lockdown phase, due to the COVID-19 pandemic in Nanjing, like other pollutants, CO showed a decreasing trend, while after lockdown the CO concentration increased. The average concentration of CO and variations are depicted in Figure 3 and Tables 1-3. The present study shows that compared to the pre-lockdown phase, the concentration of CO decreased by −16.85% during the lockdown. The results suggest that it this was mainly due to the restrictions on transportation and vehicles in Nanjing, while compared with the pre-and during lockdown phases, CO increased by 3.07%. CO is emitted from vehicular sources, burning of agricultural waste, biofuel burning and combustion of fossil fuels [50]. During the lockdown period, all of these sectors were closed, which resulted in a significant improvement in the concentration of CO. It was previously reported that during the lockdown period in China, the concentration of CO was reduced by 18% [23]. These findings are very similar to our results. Another study reported by Hanaoka and Masui (2020) [19] revealed that due to the COVID-19 pandemic lockdown period, 25% and 6% reductions in CO concentrations were observed in China and globally, respectively. Meanwhile, during the nationwide lockdown phase in the USA, the concentration of CO was reduced by 1.1%, while in the megacity Delhi, India, compared with the pre-lockdown phase, CO decreased by 30.35% during the lockdown phase [23,51].
Moreover, our results demonstrate that compared to 2019 and using the same dates, the concentration of CO decreased by −20.43% during the lockdown phase, while compared with the previous three years and the same time frame, CO was reduced by up to −25.75%, which is greater than for 2019 in Nanjing. Chen et al. (2020b) [20] found that in many areas across China, the concentration of CO was reduced during the lockdown compared to the previous three years. Compared with the same period from 2016 to 2019, there was a decline in CO concentration of −18.6% during the lockdown in South China. During the lockdown phase, the CO concentration was at its lowest ratio compared to the previous 5 years, and the concentration of CO decreased by 14.5% [49]. Further, compared with the lockdown phase, CO showed an increasing trend of up to 13.51% after the lockdown, while in comparison to the previous three years, concentrations of CO decreased by an average of −15.72% in Nanjing. Overall, during the study period, Nanjing was found to experience a significant reduction in the concentration of CO, except after the lockdown phase.

Spatial Concentration Pattern of O 3
In our study, analysis of the O 3 concentration revealed that there was an increasing trend in the concentration of O 3 during and after the lockdown phases in Nanjing. The average concentration of O 3 and variations are shown in Figure 3 and Tables 1-3. O 3 is a minor contaminant that is emitted from sunlight and nitrogen oxides (NOx), and volatile organic compounds (VOCs) [52]. During the lockdown phase, the concentration of O 3 increased by 25.45% compared with the pre-lockdown phase, while compared with the pre-and during lockdown phases, O 3 continuously increased by 72.10% in Nanjing. The increment ratio for the concentration of O 3 was greater after the lockdown than during the lockdown due to the winter-summer transection. A similar observation was reported by Chen et al. (2020b) [20], who found that during the lockdown phase in China, the concentration of O 3 increased. According to Islam et al. (2020) [39], O 3 increased during the lockdown phase in Bangladesh, while Mahato et al. (2020) [51] found that O 3 showed an increasing trend during the lockdown phase in Delhi, India.
Moreover, compared to 2019, the concentration of O 3 increased by 26.03%. Compared with the previous three years, the O 3 concentration increased by 10.61% in Nanjing. Chen et al. (2020b) [20] reported that there was a rising trend in O 3 concentration during the lockdown phase in China. Further, compared with the lockdown phase, Nanjing experienced a remarkable increment in the concentration of O 3 after the lockdown with an average increment of 54.64%, while in comparison with the previous three years, 2017-2019, for the same period, the concentration of O 3 continuously increased by an average of 71.05% in Nanjing. This increment in O 3 concentration was greater compared with the lockdown phase in 2020. A drastic change and notable improvement in air quality were observed due to the implementation of the lockdown amid the COVID-19 pandemic, except for O 3 in Nanjing.

Correlation between Different Air Pollutants
The correlations between different air pollutant concentrations during the lockdown period in Nanjing are shown in Figure 4. The daily (24 h) average concentrations of PM 10 and PM 2.5 were highly correlated (r = 0.82) with SO 2 (r = 0.26). There was a significant positive correlation between PM 10 and NO 2 (r = 0.41). The strong correlation between PM 10 and PM 2.5 suggests that it was due to their mutual pollution sources: vehicular and industrial emissions. The correlation analysis indicated that there was a strong correlation between PM 10 and CO during the lockdown phase in Nanjing (r = 0.52). PM 2.5 , SO 2 (r = 0.07) and NO 2 (r = 0.26) were highly correlated. This shows that PM 2.5 , SO 2 and NO 2 had a certain homology, while after the chemical reactions occurred in the atmosphere, SO 2 and NO 2 produced small particles. Combustion of fossil fuels contaminated with sulfur compounds result in the existence of SO 2 in the atmosphere. The daily average concentrations of PM 2.5 and CO were significantly correlated (r = 0.51) during the lockdown phase. Moreover, the (24 h) average concentration of SO 2 was significantly correlated with the (24 h) average concentration of NO 2 (r = 0.38) and CO (r = 0.27). The concentration of NO 2 was strongly correlated with the CO concentration (r = 0.39) and O 3 (r = 0.23). The positive correlation between NO 2 and CO was due to the positive correlation between particulate matter (PM) and CO. It suggests that O 3 levels increased with the reduction in the other pollutants, especially NO 2 and CO.  [20] reported that there was a rising trend in O3 concentration during the lockdown phase in China. Further, compared with the lockdown phase, Nanjing experienced a remarkable increment in the concentration of O3 after the lockdown with an average increment of 54.64%, while in comparison with the previous three years, 2017-2019, for the same period, the concentration of O3 continuously increased by an average of 71.05% in Nanjing. This increment in O3 concentration was greater compared with the lockdown phase in 2020. A drastic change and notable improvement in air quality were observed due to the implementation of the lockdown amid the COVID-19 pandemic, except for O3 in Nanjing.

Correlation between Different Air Pollutants
The correlations between different air pollutant concentrations during the lockdown period in Nanjing are shown in Figure 4. The daily (24 h) average concentrations of PM10 and PM2.5 were highly correlated (r = 0.82) with SO2 (r = 0.26). There was a significant positive correlation between PM10 and NO2 (r = 0.41). The strong correlation between PM10 and PM2.5 suggests that it was due to their mutual pollution sources: vehicular and industrial emissions. The correlation analysis indicated that there was a strong correlation between PM10 and CO during the lockdown phase in Nanjing (r = 0.52). PM2.5, SO2 (r = 0.07) and NO2 (r = 0.26) were highly correlated. This shows that PM2.5, SO2 and NO2 had a certain homology, while after the chemical reactions occurred in the atmosphere, SO2 and NO2 produced small particles. Combustion of fossil fuels contaminated with sulfur compounds result in the existence of SO2 in the atmosphere. The daily average concentrations of PM2.5 and CO were significantly correlated (r = 0.51) during the lockdown phase. Moreover, the (24 h) average concentration of SO2 was significantly correlated with the (24 h) average concentration of NO2 (r = 0.38) and CO (r = 0.27). The concentration of NO2 was strongly correlated with the CO concentration (r = 0.39) and O3 (r = 0.23). The positive correlation between NO2 and CO was due to the positive correlation between particulate matter (PM) and CO. It suggests that O3 levels increased with the reduction in the other pollutants, especially NO2 and CO.

Conclusions
The capital of Jiangsu Province, Nanjing, is known for its extreme and severe pollution levels, and it is considered one of the most polluted cities in China and internationally. The COVID-19 pandemic lockdown and shutdown policies in different parts of the world have provided a great opportunity to study the impact of humans on the environment. Therefore, the results of this study may help us to understand how responsible we are for our air pollution problems and how to mitigate them. In the current study, the impact of the nationwide lockdown that was imposed in order to curb the rapid spread of COVID-19 in China, on the air quality of the capital city of Jiangsu Province, Nanjing, was evaluated based on six air pollutants. The results show that there was a significant reduction in these pollutants, except for O 3 during the lockdown phase in Nanjing. Compared to the previous three years (2017-2019), the concentrations of PM 10 and PM 2.5 were reduced by −37.99% and −33.56%, respectively, during the lockdown. Therefore, compared to the previous three years, PM 10 and PM 2.5 concentrations decreased by −43.58% and −54.57%, respectively, after the lockdown. Moreover, a significant reduction in other pollutants, such as SO 2 , NO 2 and CO concentrations, were also observed during the lockdown phase. SO 2 and CO showed a tendency to increase after the lockdown phase in Nanjing. During the lockdown, an approximately 25-35% improvement in air quality was observed. This study provides useful information and outcomes for future studies and creates a new paradigm for regulatory authorities and policymakers to create policies for the improvement of air pollution. Other studies, as discussed in Section 1, have mainly focused on either air quality assessment and monitoring over the last few years or the comparison between before and current air pollutants; however, our study performed three types of comparisons, i.e., before, after, and current trends as well as a comparison with data on air pollutants from the last 3 years. Another major advantage of this study is that it considered all of the pollutants, while other studies have focused on only a few pollutants. This helps in policy recommendations for government with regard to how trends in air quality are changing after COVID-19, and it helps air quality and monitoring stations to take the necessary steps if air quality does not improve. Furthermore, the outcomes of the current study will encourage the Chinese administration to deliberate on how to decrease industrial and vehicular contamination to reform and improve air quality, which will help to maintain better public health in China. The current study can be further extended to other regions and countries by finding new correlation factors between air pollutants and meteorological indicators.