Impact of Lockdown on Air Pollutants during COVID-19 at Patna, India

Many countries shut their borders, imposed nationwide lockdown, and restricted several anthropogenic activities to arrest the spread of COVID-19. In the present study, the concentration of several air pollutants(PM10, PM2.5, NO2, NH3, SO2, CO and O3) during different phases of lockdown from monitoring stations of Patna was analyzed to assess the effect of lockdown restriction on air quality. Reduction in PM2.5, NH3, NO2, PM10 and CO concentration was observed by 59.79%, 58.2%, 49.49%, 39.57% and 24.04%, respectively during the lockdown period. National Air Quality Index(NAQI) value in the year 2020 had been observed to lower by 57.88% compared to the year 2019, during the same period. A more significant fall in the concentration of air pollutants was observed during the early phase of post-lockdown compared to the late stages of post-lockdown. The study reflects the significance of restriction on anthropogenic activities in improving air quality and provides clues for future action plans for improving air quality.


INTRODUCTION
The world has recently faced a global pandemic COVID-19, first detected in Wuhan, Central China, in December 2019 (He et al., 2020;Zhang et al., 2020;Zhu et al., 2020).The virus has gradually affected all nations since December 2019 (Rajbhandari et al., 2020;Wang et al., 2020).Until 13 th December 2020, nearly 70 million confirmed COVID cases, and 1.6 million deaths owing to COVID-19 was reported (WHO, 2020a).In response to the increase in the number of cases and contagion nature of COVID-19, the Government of India has imposed a 14-hour Janta Curfew on 22 nd March 2020, followed by a complete nation-wide lockdown from 24 th March to 30 th June 2020 in four phases.During that time, all commercial, industrial and educational activities were closed to limit the spread of infection.The unlock phase was slow, gradual, and completed in seven stages, and restricted commercial and transportation activities were permitted in non-containment zones.Lockdown has slowed down the pace of living, but it has positive outcomes regarding environmental issues including air quality across many cities in India (Agarwal et al., 2020;Gautam, 2020;Kumari and Toshniwal, 2020;Mahato et al., 2020;Sharma et al., 2020;Singh and Chauhan, 2020;Sri vastava et al., 2020) and in other countries also (Hashim et al., 2021;Baldasano, 2020;Dantas et al., 2020;Mesas-Carrascosa et al., 2020;Tobías et al., 2020;Wang et al., 2020;Zhu et al., 2020).
Patna, the capital of Bihar, is among the most populous and developing city in the Indo-Gangetic Plain.It has emerged as the second-largest economic center of eastern India after Kolkata and was ranked 32 nd among the top 100 polluted cities in the world and 24 th in India as per The World Air Quality report, 2020 (IQAir, 2020).The increasing population, economic and urban development have enhanced the burden on different sectors (industrial, transportation, domestic, constructional, biomass burning, etc.) as significant sources of air pollutants at Patna (Guttikunda et al., 2019).The literature review reveals that vehicular emissions (Jain et al., 2019;Sharma et al., 2016;Pandey and Venkataraman, 2014), use of biofuels in traditional cookstoves (Singh et al., 2021;Sen et al., 2018;Saud et al., 2012), municipal solid waste and agricultural biomass burning and 388 functional brick kilns (Joshi, 2019) were the major contributors to PM 2.5 and PM 10 in the Indo-Gangetic Plains (Kumar et al., 2020a;Arif et al., 2018).Several researchers reported a high concentration of air pollutants over the Indo-Gangetic plain (Mhawish et al., 2020;Mishra and Kulshrestha, 2020;Ojha et al., 2020;Shastri et al., 2017;Acharya and Sreekesh, 2013).
Among air pollutants, PM 2.5 and PM 10 constitute significant concern at Patna as these parameters often exceed the standard limits prescribed by NAAQS.Under the business-as-usual (BAU) scenario, the concentration level of PM 2.5 is expected to increase by 28%, from 104.4 μg/m 3 in 2018 to 134.0 μg/m 3 by 2030 (PCAAP, 2019).An increase in emission load for PM 2.5 in the city is also attributed to natural phenomena like climatic and meteorological conditions, the presence of very soft alluvial soil and atmospheric reactions generating secondary particulate matters (PCAAP, 2019).
During this lockdown, it has been noticed that the air pollution levels came within or below the permissible limit set by the CPCB of India (Gour et al., 2015).Most of the recent literature published in air quality during the COVID-19 pandemic lockdown in various Indian cities is illustrated in Table 1.The majority of the studies conducted (Bedi et al., 2020;Gautam, 2020;Kotnala et al., 2020;Kumar et al., 2020b;Mahato et al., 2020;Singh and Chauhan, 2020) revealed early trends as they were confined to the lockdown period only and ignored the post-lockdown scenario.Our area of interest, Patna, is part of the Indo-Gangetic Plain where air pollution is at its peak in the last quarter of the year (Mishra and Kulshrestha, 2020;Ojha et al., 2020).Moreover, the study conducted by Bedi et al. 2020 andNavinya et al. 2020 focused on various Indian cities at a time.However, only a single monitoring station was considered per city, which is insufficient for spatial analysis.The obtained data cannot truly depict the air quality of the entire city.Masum and Pal (2020) reported that changes in air quality due to the COVID-19 pandemic do not follow a similar trend in different regions; hence, a regional study is essential to determine the pattern of changes.Therefore, this paper aimed to study the impact of complete lockdown on the air quality of Patna by comparing air quality parameters during pre-lockdown, lockdown and postlockdown phases in the year 2020 with those of the corresponding period in 2019.In our study, Correlational analysis GIS-based spatial maps was also used to provide in-depth insights to the authorities concerned to formulate a better management plan to improve the anthropogenic air pollutant exposure in Patna and provide a baseline for other important cities of the Indo-Gangetic Plain.

1 Study Area
The area under investigation is Patna, the capital and largest city of Bihar situated 15 km along the bank of River Ganga between 25°56 N to 25°69 N latitude and 85°02 E to 85°25 E longitude in the Indo-Gangetic Plains.The city experiences a subtropical climate with a hot summer, a cold winter, and heavy rainfall during the monsoon season.Relative humidity can reach up to 100% during the summer season.At present, the Bihar State Pollution Control Board (BSPCB) monitors the air quality of Patna with the help of 6 Continuous Air Moni-toring Stations (CAMS) distributed throughout the city (Fig. 1).Reduction in NAQI was about 54% in Central Delhi followed by Eastern (49%), Southern (43%), Western (37%) and Northern (31%) regions of the city.When compared with 2019, significant improvement in air pollutants was observed with reduction of 60% (PM 10 ), 39% (PM 2.5 ), 53% (NO 2 ) and 30% for CO.
India (Gautam, 2020) 31 st March to 5 th April from 2016-2020 Aerosol Optical Depth The level of aerosol is found to be lowest in the last 20 years owing to 50% reduction during the lockdown period.

2 Data Collection and Methodology
To study the effect of lockdown on the air quality of Patna, primary data from 6 different monitoring stations was collected from the CPCB online portal (https:// app.cpcbccr.com/AQI_India/)and the BSPCB online portal (http://bspcb.bih.nic.in/environment-monitoring-data.html).The concentration of 7 different air pollutants was considered, including PM 2.5 , PM 10 , NO 2 , NH 3 , SO 2 , O 3 and CO.For contaminants such as PM 2.5 , PM 10 , NO 2 , NH 3 and SO 2 , the daily average (24h) subindex value and CO and O 3 daily maximum (8h) average sub-index value had been taken.This was further analysed to see the change in their mean concentration between different phases.The fluctuation in AQI value and mean concentration have been studied in four phases in the year 2020 from pre-lockdown (1 st January to 24 th March) to lockdown (25 th March to 31 st May) to early phases of post-lockdown (1 st June to 31 st August) to the late phase of post-lockdown (1 st September to 15 th Dec-ember).The method adopted to calculate NAQI is tabulated in (CPCB, 2014) briefly outlined here.AQI uses PM 2.5 , PM 10 , NO 2 , SO 2 , CO and O 3 as criteria pollutants, where the selection of parameters depends on AQI objective(s), data availability, averaging period and monitoring frequency (CPCB, 2014).Of all the air pollutants, the calculation of AQI requires a concentration of a minimum of three pollutants, with at least one being either PM 2.5 or PM 10 .The maximum sub-index AQI i of the corresponding pollutant is the overall AQI, where the sub-index AQI i of each pollutant can be calculated using Eq.(1).

SI-AQI
Where C i is the concentration of pollutant 'i'; B HI and B LO are breakpoint concentrations higher and lower than C i and I HI and I LO are corresponding AQI values equiva- lent to B HI and B LO respectively.The health breakpoint of each pollutant is given in Table 2.

1 Variation in Air Pollutants Concentration
The minimum, maximum and mean value of the daily average concentration of different air pollutants have been summarized in Table 3.The results revealed particulate matter (PM 2.5 and PM 10 ) as the dominant and indicatory air pollutant during the entire study period.The concentration of PM 10 ranged from 17.33 μg/m 3 to 382.12 μg/m 3 while PM 2.5 was observed between 9.00 μg/m 3 to 236.38 μg/m 3 during the study period.The dominance of particulate matter among the air pollutants in the study area might be due to various anthropogenic activities like re-suspension of road dust, construction activities, open burning of solid waste, emission from vehicles and brick kilns etc. Patna is situated along the bank of river Ganga and ¾ th of its soil is alluvial and makes the soil soft and prone to the formation of dust which is blown into the city from the sandy riverbank adding to the particulate matter level as outside contributors (PCAAP, 2019).Various atmospheric reactions producing secondary particulate matter also add to the level of PM 2.5 and PM 10 in the study area.Fig. 2 represents the variation in daily average (24h) concentration of PM 10 , PM 2.5 , NO 2 , SO 2 , NH 3 and daily maximum (8h) average of O 3 and CO between 1 st January 2020 and 15 th December 2020 in the city of Patna.The result revealed a reduction in the daily average value of PM 10 , PM 2.5 , NO 2 , SO 2 , NH 3 and CO from pre-lockdown to lockdown and from lockdown to early phases of post-lockdown.The daily average PM 10 value was beyond the NAAQS standard limit of 100 μg/m 3 during the pre-lockdown.The daily average concentration of PM 10 and PM 2.5 has been reduced by 39.57% and 59.79%, respectively, from prelockdown to lockdown (Table 3).A higher decrease in PM 2.5 value compared to PM 10 could be due to reduction in transport activities as the contribution of vehicular emissions and emissions from the combustion of gasoline or other such fuels to PM 2.5 is much more pronounced than that of PM 10 (PCAAP, 2019; ARAI and TERI, 2018; Guttikunda and Jawahar, 2014).Inter estingly, the scenario gets reversed when the percent change between lockdown and the early stages of post-lockdown are analyzed.PM 10 declined by 49.49%, while PM 2.5 dropped only by 36.44%, which might be due to the relaxation in  inter-state and intra-state travel in the early post-lockdown period as vehicular emission contributes more to PM 2.5 (Kumar et al., 2020b).Further, negligible construction, industrial activities, waste burning reduction, landfills, and intermittent rainfall might be responsible for a significant drop in PM 10 level during the early post-lockdown period (CARB 2021; Sahoo et al., 2021;Sathe et al., 2021;Kumar et al., 2020a).A decline in the average concentration of NO 2 and NH 3 was also observed.The NO 2 value can be seen at an alarming level during prelockdown and has decreased by 52.52% from pre-lockdown to lockdown and 18.74% from lockdown to initial unlock stages, mainly due to a reduction in the transport sector.The NH 3 daily average concentration has been reduced by 58.20% from pre-lockdown to lockdown.
Although the anthropogenic source of NH 3 originates mainly from agricultural activities, including the use of soil fertilizer, domestic animal waste and the use of ammoniabased fertilizers, the reduction of NH 3 can be attri buted to industrial and traffic emissions in the urban areas (Farren et al., 2020;Li et al., 2020;Wang et al., 2015).No significant change in SO 2 concentration has been observed during the entire study period except in early post-lockdown ( June to September) that suggests vehicular restriction might have not much impact on SO 2 concentration (Lokhandwala and Gautam, 2020).Decrease in SO 2 concentration during the early post-lockdown period is due to substantial rainfall during this period (Fig. 2).Several researchers reported a reduction in the concentration of SO 2 during the rainy season, probably due to the washout effect (Ngarambe et al., 2021;Xue et al., 2020).The result revealed that the average concentration of CO had declined by 24.04% during the lockdown, and it had further decreased by 40.54% during the early postlockdown period.A decrease in CO concentration during the lockdown period might be due to restrictions on combustion engines, heating furnaces and automobile exhaust.Further decrease in CO concentration during the early post-lockdown period may be due to the precipitation washout effect (Ngarambe et al., 2021;Yoo et al., 2014).Strikingly, the mean average concentration of O 3 has increased by 73.75% from pre-lockdown to lockdown, which is due to a decrease in NO x and VOCs concentration because of the complete restriction on industrial activity and vehicular movement (Filonchyk and Hurynovich, 2020;Kumari and Toshniwal, 2020;Mahato et al., 2020).Sharma et al. (2020) reported that a decrease in the concentration of particulate matter could result in more sunlight penetration, thereby increasing the photochemical reaction that can uplift O 3 production.
The pollutant concentration data also revealed the effect of sequential unlock phases on air pollutants.PM 2.5 has experienced a maximum increase of about 235.01%from the early stages of post-lockdown to the late stages, followed by PM 10 (176.57%),NO 2 (137.29%),SO 2 (105.18%),CO (94.94%),NH 3 (87.36%)and O 3 (4.06%).An increase in the concentration of these pollutants during sequential unlock phases might have resulted from resumption of anthropogenic origin pollutants and changing weather and meteorological conditions (Mahato et al., 2020;Sharma et al., 2020;Srivastava et al., 2020).
The daily, weekly, and monthly variations of different air pollutants are conditionally formatted (Fig. 3).The higher value has been colored by red that decreases to green at the lower values.PM 10 , PM 2.5 , NO 2 , NH 3 and CO remained high during the pre-lockdown ( January to March) and late post-lockdown (October to December).During the lockdown, each of them experiences a reduction in their daily average concentration.The concentration of O 3 on the other hand, was observed to be high in May.The result suggests the reduction in transportation and industrial activities has a predominant impact on the concentration of particulate matter, NO 2 , and NH 3 , as their concentration started dropping with the implementation of the lockdown (near the end of March).CO and SO 2 have begun to decline at the tail end of April, suggesting restrictions on vehicular emissions might not have much to do with these pollutants, specifically SO 2 , and incomplete combustion of fossil fuel from biomass burning might have contributed to CO.The concentration of NH 3 shows an unusual increase towards the end of August (Fig. 3).This increase was explicitly marked in the data collected from the Danapur site, which might be due to some agriculturally based activities (CGWB, 2015) in addition to vehicular emissions.
The concentration of pollutants showed a further decrease in their daily average concentration during the early stages of the post-lockdown (1 st June to 31 st Aug ust).The effect of lockdown coupled with substantial monsoon rainfall, the contribution of clean air from the Bay of Ben- gal and Arabian Sea, and little input from brick kilns and biomass burning might be the reason behind such a drop (Arif et al., 2018).The diagrams also reveal a gradual increase in the deterioration of air quality from October onwards due to the relaxation in lockdown, the emergence of the winter season, prevailing westerly wind, and other anthropogenic factors that sum up to degrade air quality during the late phase of lockdown.

.2 Spatial Variation
Fig. 4 depicts the spatial variation of different air pollutants in the study area.PM 2.5 showed a significant reduction during the lockdown phase at nearly every sampling site in the city.During the first week of lockdown, it was reduced to 47.22% from the pre-lockdown level.NO 2 , on the other hand, has recorded maximum reduction of 63.18% within seven days since lockdown, suggesting restrictions on transport and other anthropogenic activities have played an essential role in lowering their concentrations.The concentration of NO 2 and CO was found to be greater at site 3 during the lockdown phase than that of the pre-lockdown phase, which might be due to biomass burning (Beig et al., 2020;Biswal et al., 2020;Ravindra et al., 2020) for household purposes by the population residing in two densely populated slum areas located near this monitoring site.An increase in NO 2 levels during the early phases of post-lockdown, especially in the western and central pockets of the study area (Fig. 4) might be due to relaxation in transport activ-ity.Towards the end of the year, an increase in air pollutants was noticed owing to various anthropogenic and natural factors.Spatial variation revealed that pollutants load prevailed mainly in the city's central region during the entire study period.The result is confined to the pollutants including PM 2.5 , NO 2 , and CO; the rest do not vary very much or are excluded due to the unavailability of data for some phases.

3 Statistical Correlation between Air Pollutants
Correlation between different air pollutants was computed for the study period (Fig. 5).The correlation analysis revealed the daily average concentration of PM 10 to be very strongly correlated with PM 2.5 (r = 0.87), which suggests they both contribute to the air from common sources.PM 10 and PM 2.5 showed moderate correlation with NH 3 (r = 0.47; r = 0.46), indicating a favorable role of ammonia conversion from gas to particle phase in particulate matter formation (Wang et al., 2015).The correlation of particulate matter with NO 2 ( r 10 = 0.74, r 2.5 = 0.63) and SO 2 ( r 10 = 0.50, r 2.5 = 0.46) suggests the presence of secondary pollutants generated by the photochemical reaction of gaseous pollutants mainly SO 2 and NO 2 (CARB, 2021;Botkin and Keller, 2000;Xu et al., 2020a).O 3 showed a negative correlation with all other pollutants, predominantly with PM 2.5 , and NH 3 during our study period and similar findings were also observed by Kumari and Toshniwal, 2020;Mahato et al., 2020;Sharma et al., 2020;Xu et al., 2020b.

4 Comparison of Air Pollutants
between 2019 and 2020 A comprehensive comparison of the concentration of five air pollutants (PM 2.5 , NO 2 , SO 2 , CO, O 3 ) during pre-lockdown, lockdown and different phases of post-lockdown in the year 2020 was made with corresponding period in the year 2019 (Fig. 6).The result revealed that a significant impact of lockdown is noticeable when the deviation was evaluated from lockdown to early stages of post-lockdown (from 1 st June 2020 to 31 st August 2020).In 2020, all the air pollutants showed a negative % change where the maximum reduction was observed for SO 2 that has decreased from 8.79 μg/m 3 to 4.05 μg/m 3 (53.95%) in 2020, followed by CO (38.57%),PM 2.5 (36.38%),NO 2 (20.82%) and O 3 (19.37%).Compared with the previous year's data, the monthly average concentration of NO 2 ,  SO 2 , CO was indeed higher during the same time (Fig. 6).Nearing the end of the study period, air pollutants have remarkably shown an increase in their concentration for both years.However, air pollutants have come across a more tremendous % increase in the year 2020 compared to 2019.The monthly average concentration of PM 2.5 has been maximum during the late phase of post-lockdown, which was noted at 88.70 μg/m 3 in 2020 and 139.35 μg/m 3 in 2019.Similarly, NO 2 has increased from 17.26 μg/m 3 to 43.42 μg/m 3 in the year 2020 and from 21.53 μg/m 3 to 31.23 μg/m 3 in the corresponding period of the year 2019.43.18% and 113.70% increase can also be noticeable for CO in 2019 and 2020 respectively (Fig. 6).An increase in the concentration of these pollutants during this phase might be due to the lowering of mixing layer height hindering the complete dispersion of pollutants (Guttikunda and Jawahar, 2014), winter inversion (Mahato et al., 2020;Mishra and Kulshrestha, 2020).Increase in biomass burning during winter that contributes 18-30% of the total concentration level during winter (PCAAP, 2019), burning of crackers during festive and marriage occasions at the tail end of the year (Mahato et al., 2020;Chauhan and Singh, 2017) also increase the concentration of pollutants.Besides these factors, high biomass residue burning in Northwest countries and the Northern regions of India sums up with dust aerosol from the west during winter and premonsoon (Srivastava et al., 2020;Arif et al., 2018), resum ption of brick manufacturing during winter and premonsoon period and westerly air mass coming from Indo-Gangetic Plain resulted in a comparatively elevated level of PM 2.5 and PM 10 during winters (Agarwal et al., 2020;Singh and Chauhan, 2020;Arif et al., 2018).

5 Aqi Interpretation
Fig. 7 shows changes in the monthly average AQI value from 1 st January to 15 th December for 2019 and 2020.Since the beginning of the year, the air quality has improved slightly compared to the previous year.However, compared to the AQI value for the same period in 2019,  the most significant drop in the AQI value in 2020 occurs from April to June.The AQI value ranged from 30 to 224 in 2020, peaking at 275 in 2019 with a 30.91% decline in the average AQI value during the lockdown period.The AQI value remained in a good to the moderately polluted zone during the lockdown and early postlockdown phases that were in the poor category last year (Fig. 7).Compared to 2019, the AQI value this year has declined by 57.88% from pre-lockdown to lockdown, which was recorded as 47.57%.Such a drop in the AQI values in the pandemic and early stages of post-pande mic phases again justifies our discussion that the imple mentation of lockdown was a powerful reason for the changes observed in the air quality this year.

CONCLUSION
The main focus of the study was to investigate the influence of emission reductions due to restrictions on transportation, industrial and various anthropogenic contributors of air pollutants on the air quality of Patna during the COVID-19 lockdown and also to examine the changes in air pollutants levels throughout the year.The study concludes that the implementation of lockdown certainly had a significant positive impact on improving the quality of the air in Patna.The pollutants that experienced maximum reduction during the lockdown period include particulate matter, NO 2 , and NH 3 .On the contrary, SO 2 had not declined by much and O 3 rose amid this period due to a drop in PM and NO 2 levels.A higher pollution load was reported by spatial analysis in the central part of the city.The results also revealed a notable contribution of biomass burning to NO 2 during the lockdown.A drop observed in the early stages of post-lockdown was due to the influence of the monsoon sea son and southerly wind.Together with the lockdown, the consequences have lowered the pollutant's concentration even further to a greater extent than the previous year.In addition, the relaxation in pandemic restrictions and the impact of climatic conditions once again paved the way for an increase in these pollutants towards the end of the year.The study eventually concluded that the problem determining air quality is mainly anthropogenic.However, lockdown is not a permanent solution to expect imp rove ment in water or air quality.Instead, it provides us with evidence that improvement in air quality can be achieved if proper measures are drafted/worked out in the existing regulatory plans and are implemented strictly by the concerned authorities in a phase-wise manner.

Fig. 1 .
Fig. 1.Map showing the study area along with monitoring stations.

Fig. 2 .
Fig. 2. Variation in criteria air pollutants level from pre-lockdown to post-lockdown in the year 2020.

Fig. 3 .
Fig. 3. Conditional formatting representing daily, weekly and monthly variation in air pollutants level in 2020.

Fig. 4 .
Fig. 4. Comparing the spatial distribution of air pollutants concentration during pre-lockdown, lockdown, early and late post-lockdown phases in Patna for the year 2020.

Fig. 6 .
Fig. 6.Comparative analysis of percentage change in air pollutants level between 2019 and 2020.

Fig. 7 .
Fig. 7. Variation in AQI value between the years 2019 and 2020 at Patna.

Table 1 .
Summarizing the recent studies conducted in various cities of India related to air quality during nationwide lockdown due to COVID-19 pandemic.

Table 2 .
National AQI classes, range, health impacts and health breakpoints for the seven pollutants (scale: 0-500).

Table 3 .
Summarizing daily average concentration and % change of air pollutants in different phases of the year 2020.