Comprehensive study on impact assessment of lockdown on overall ambient air quality amid COVID-19 in Delhi and its NCR, India.

Indian government announced the complete lockdown from 25 March, 2020 for all outdoor activities across the country due to containment of COVID-19. This study is an attempt to assess the impacts of lockdown on ambient air quality in five cities of Indian National Capital Region including Delhi, Gurugram, Noida, Ghaziabad and Faridabad. In this context, the data of air pollutants (PM10, PM2.5, NOx, NO, NO2, SO2, NH3, SO2, CO, and C6H6) from 36 locations of the study area were analyzed from 1st March to 1st May, 2020. The results showed that PM10 and PM2.5 level decreased upto 55-65 %. NOx and NO have shown maximum reduction (∼ 50-78 %). Similarly, consistent and significant reduction in other air pollutants such as SO2 (∼33 %), CO (∼45 %), NH3 (∼27 %) and C6H6 (∼53 %) has been observed. During lockdown Air Quality Index (AQI) shows improvement as its value significantly decreased (∼ 45 %-68 %). An interesting feature observed that during first week of lockdown O3 decreased but later it increased by ∼19-27%. The study suggests that this pandemic gives lessons for interventions for urban air pollution mitigation in controlling the health impact due to urban air pollution.

Introduction

Different countries across the world are facing the health challenges due to spread of Novel corona virus (COVID-19). On 24th March, when the COVID 19 cases hits the value of 500, India announced the complete lockdown in country. The lockdown norms are very strict in Indian cities which ban the stepping out of people from their homes (Ministry of Home Affairs, 2020). Despite of so many economical challenges, the positive thing observed during this pandemic situation is the improvement in air quality (Sharma et al., 2020; Kerimray et al., 2020). It is well known that the rapid urbanization, industrialization, and commercialization in Delhi, the capital of India results in deteriorating its ambient air quality. World Health Organization (WHO) has declared Delhi as one of the most polluted city in the world (WHO, 2016). In Delhi and its National Capital Regions (NCR), the growth in motorized vehicles and industrial activities were considered as the major emission sources of air pollution and results in the emission of particulate matter, nitrogen oxide, sulfur oxides, carbon monoxide, ammonia, ozone, volatile organic compounds, polycyclic aromatic hydrocarbons etc (Singh and Peshin, 2014; Gulia et al., 2018; Garg and Gupta, 2019; Sharma et al., 2014). In Delhi, the high levels of particulate matter and nitrogen oxides are always became a severe threat for public health due to its exceeding value from the permissible limit of ambient air quality standards setup by Central Pollution Control Board (CPCB) of India (Gour et al., 2015). During this lockdown, it has been noticed that the air pollution levels even in the most polluted city came within or below the permissible limit of ambient air quality standards setup by CPCB of India (Central Pollution Control Board (CPCB, 2009). Recently some studies were carried out to understand the impact of lockdown control measures on air pollutions (Li et al., 2020; Otmani et al., 2020; Dantas et al., 2020; Sharma et al., 2020; Kerimray et al., 2020) but they are fragmented and explore only few air pollutants. In this study, we try to investigate the effects of lockdown on a total of ten pollutants (PM10, PM2.5, NOx, NO, NO2, NH3, SO2, CO, C6H6, O3). Secondly, the hypothesis of this study is that there is little chance of influx and outflux of air pollutants since the meteorological condition was almost same for week before lockdown and during lockdown period. Since, during this lockdown period, all industrial, vehicular and commercial activities were restricted, it provides ideal condition for investigating their direct effect on extent of pollution reduction and their role in adding the air pollution loads.

Materials and methods

Study area

The study area covers National Capital Region (NCR) of Delhi, which includes five cities namely Delhi, Uttar Pradesh Province (Noida and Ghaziabad), Haryana province (Gurugram and Faridabad) of India (Fig. 1 ). These cities are some of the few most developed and densely populated cities in India and witnessing urbanization and industrialization at accelerated pace coupled with increased socio-economic activities. However, rapid development is accompanied by increased air pollution problem. Thus, NCR region facing severe air pollution challenge and Delhi has been identified as the most polluted city. Table S1 shows the economic activities and major industries in the study area.

Fig. 1

Map showing study area locations in Delhi NCR.

Data collection

The air pollutants and meteorological data were collected for the period of 1st March to 1st May, 2020. The data were collected for 5 cities from 36 monitoring stations, twenty from Delhi and four each from Gurugram, Faridabad, Ghaziabad, and Noida. These monitoring locations have been shown in Fig. 1. The air pollution monitoring in India including those taken under this study is maintained by CPCB through air quality monitoring network Continuous Ambient Air Quality Monitoring System (CAAQMS). In order to understand the comprehensively impact of measures taken during lockdown on ambient air quality, ten most relevant pollutants were considered in this study. These air pollutants include particulate matter (PM10, and PM2.5), nitrogen oxides (NOx), nitrogen dioxide (NO2), and nitric oxide (NO), sulfur dioxide (SO2), ozone (O3), carbon mono-oxide (CO), ammonia (NH3), and benzene (C6H6), The data analyzed for meteorological parameters includes relative humidity (RH), ambient temperature (AT), wind speed (WS), and solar radiation (SR). The data for both meteorological parameters as well as air pollutants were collected for one hour interval (from 1st March to 1st May, 2020) for all 36 monitoring stations. These datasets were further converted into 24 h average and statistically analyzed. In this work AQI was determined using the input values of three criteria pollutants i.e., PM10, PM2.5 and NO2. The method for the calculation of overall AQI has been primarily based on the development of sub-indices for each pollutant and after then, the aggregation of sub-indices as per the CPCB (CPCB, 2015).

Results

Effect of lockdown on different air pollutant levels (8 days before and 8 days during lockdown)

In this study, it has been observed that from 17 March to 1 April, 2020 the meteorological parameters (AT, RH, SR, and WS) exhibit very little variations. For identifying the variations among meteorological parameters, one way ANOVA test has been performed at 0.05 level of significance (Table S2). These results show that there is no significant variation in meteorology from 17 March to 1 April. So, due to minimal interference of meteorology during 17 March to 1 April, it has been ideal to identify the exact effect of lockdown during this period. For further analysis, these days were divided into two categories: 1) 8 days before lockdown (17 March-24 March) and 2) 8 days during lockdown (25 March-1 April).

From obtained results, it was found that the concentration of all ten pollutants decreased significantly due to several restrictive measures taken by government authority during lockdown period. These reductions in pollutant levels have been given in Table 1 and shown in Fig. 2 (a–j). During these days, a significant reduction in PM10 levels was observed (52.92%–65.36%). Similar trends were also observed for PM2.5. reduction (57.28%–68.32%). Our study revealed that among nitrogen oxides, the highest reduction (upto 78.6 %) was observed for NO at all study locations except at Noida, where reduction rate for NO2 and NOx was found higher. The reductions in NO levels were as in the range of 49.76%–78.59% with maximum reductions at Delhi and minimum at Noida. Similarly, the reductions in NOx levels were in the range of 46.7%–66.12% with highest at Delhi and lowest in Faridabad. The reduction in NO2 concentrations were found to be from 17.04%–65.18 %. The reduction in SO2 levels were ranged from 12.55 % to -56.23 %. During the initial eight days of lockdown a significant reduction in CO, Benzene, NH3 levels were observed and ranged as 33.72 %–55.70 %, 32.67 %–69.93 %, and 14.34%–28.71% respectively. A significant reduction in O3 levels was also observed (9.37%–27.16%) during initial days of lockdown at all locations except at Faridabad where it marginally increased (1.67 %, from 38.98 μg/m3 to 39.63 μg/m3). Similarly, the AQI level, before the lockdown was observed in moderate to poor categories. However, within 8 days of lockdown, AQI value reduced by ∼45 %-68 % and came down under satisfactory category. The maximum reduction in AQI during lockdown was observed in Ghaziabad (∼68 %) followed by Delhi (∼61 %), Gurugram (∼59 %), Noida (∼58 %), and Faridabad (∼45 %). The detailed reduction and values of AQI along with color coding on weekly basis has been shown in Table 2 .

Table 1

Variation in average air pollutant concentrations 8 days before and during lockdown.

CITY	Period	PM10 (μg/m3)	PM2.5 (μg/m3)	NO2 (μg/m3)	NOx (μg/m3)	NO (μg/m3)	O3 (μg/m3)	NH3 (μg/m3)	SO2 (μg/m3)	CO(mg/m3)	Benzene	AQI
											(μg/m3)
NOIDA	Before	151.85	70.5	36.99	30.06	9.05	39.16	37.08	14.53	1.65	0.46	137
	During	56.55	26.98	12.88	11.54	4.55	28.52	30.84	9.33	0.86	0.15	57
	Variation	−95.3	−43.52	−24.11	−18.51	−4.5	−10.64	−6.24	−5.2	−0.79	−0.31	−80
	Variation (%)	−62.76	−61.74	−65.18	−61.59	−49.76	−27.16	−16.83	−35.8	−48.04	−67.03	−58.39
GURUGRAM	Before	155.37	72.32	21.85	25.64	15.14	54.23	59.2	17.17	1.06	5.18	140
	During	53.82	26.83	14.54	11.02	5.77	49.15	46.28	15.02	0.55	3.49	57
	Variation	−101.55	−45.49	−7.31	−14.62	−9.37	−5.08	−12.93	−2.16	−0.51	−1.69	−83
	Variation (%)	−65.36	−62.9	−33.48	−57.01	−61.91	−9.37	−21.84	−12.55	−47.9	−32.67	−59.29
GHAZIABAD	Before	186.84	91.58	48.54	36.57	13.29	43.15	37.77	26.17	1.15	2.48	207
	During	66.86	32.95	19.33	13.78	4.41	31.43	32.35	11.46	0.76	1.3	67
	Variation	−119.97	−58.62	−29.2	−22.79	−8.88	−11.72	−5.42	−14.72	−0.39	−1.18	−140
	Variation (%)	−64.21	−64.02	−60.16	−62.33	−66.81	−27.16	−14.34	−56.23	−33.72	−47.67	−67.63
FARDIDABAD	Before	190.58	79.1	11.92	32.48	29.14	38.98	29.76	7.59	1.33	4.24	163
	During	89.73	25.06	9.89	17.29	13.11	39.63	45.3	4.88	0.59	2.25	90
	Variation	−100.85	−54.04	−2.03	−15.19	−16.02	0.65	15.54	−2.71	−0.74	−1.99	−73
	Variation (%)	−52.92	−68.32	−17.04	−46.77	−55	1.67	52.21	−35.74	−55.7	−46.93	−44.79
DELHI	Before	169.53	79.24	39.19	40.85	21.54	39.47	26.85	16.09	1.08	3.23	163
	During	63.42	33.85	16.42	13.84	4.61	34.5	19.14	11.36	0.63	0.97	63
	Variation	−106.11	−45.39	−22.77	−27.01	−16.93	−4.97	−7.71	−4.73	−0.45	−2.26	−100
	Variation (%)	−62.59	−57.28	−58.1	−66.12	−78.59	−12.58	−28.71	−29.42	−41.38	−69.93	−61.35

Fig. 2

Variations of ambient air pollutants: Eight days before and eight days during lockdown period in Delhi NCR.

Table 2

AQI variations in study regions from 1st March to 1st May, 2020.

Lockdown	Period	Phase	NOIDA	GURUGRAM	GHAZIABAD	FARIDABAD	DELHI
Before Lockdown	1March-8March	I	121	97	143	132	121
	9March-16March	II	120	119	129	147	130
	17March-24March	III	137	140	207	163	163
During Lockdown	25March-1April	I	57	57	67	90	63
	2April-9April	II	77	74	102	84	91
	10April-17April	III	128	112	148	127	125
	18April-25April	IV	105	89	123	113	108
	26April-1 May	V	106	101	111	137	103
AQI Reduction from Before Lockdown III to During Lockdown I			−58.39%	−59.28%	−67.63%	−44.78%	−61.34%
AQI indication
Good (0−50)		Minimal Impact
Satisfactory (51−100)		Minor breathing discomfort to sensitive people
Moderate (101−200)		Breathing discomfort to the people with lung, heart disease, children and older adults
Poor (201−300)		Breathing discomfort to people on prolonged exposure
Very Poor (301−400)		Respiratory illness to the people on prolonged exposure
Severe (>401)		Respiratory effects even on healthy people

Spatio-temporal variations in air pollutants during extended period of lockdown

To study the effect of extended period of lockdown on air pollution, all pollutants were analyzed for period from 1 March to 1 May, 2020. The daily variation of all the ten pollutants has been shown in Figs. S1–S3. The variations in metrological parameters have been shown in Fig. S4. In this study, during these two months of analysis, average mean ambient temperature at Delhi NCR region varied from 23.81 °C to 28.37 °C. The average relative humidity varied as 49.27%–55.19%. Solar radiations varied from 90.81 W/m2 to 208.89 W/m2. There were few events of rainfall, and dust-storms and it’s affected the pollution level. The rainfall during these days caused sharp decline in PM10, and PM2.5 levels at all locations. This decline was observed due to washout and settlement of all the particles dispersed in the ambient air. Also, in later days of lockdown a sudden increase in air pollutants were observed due to dust storm events in Delhi NCR. The backward Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) showing air parcel and deposition or dispersion of atmospheric pollutants has been shown in Fig. S5. These trajectories show the long range transport of air pollution into Delhi region from another location. After comparing the average levels of air pollutants before 24 days and after 24 days of lockdown, relatively low reduction was observed as compared to initial eight days of lockdown. The reduction for PM10 (∼ 27%–38%), PM2.5 (∼ 39%–47%), NO2 (∼ 15%–59%), NOx (∼54 % to 63 %), NO (∼60 % to 78 %), NH3 (∼3.5 % to 25 %), CO (∼26 % to 44 %), and Benzene (∼30 % to 51 %) has been observed during study duration (Figs. S1–S3). An increase of (∼1.73 % to 6.72 %) for SO2 and (∼15 % to 27 %) for O3 have been obtained.

Pearson correlation statistics was applied between meteorological parameters and air pollutants at 0.05 level of significant and shown in Table S3. Strong negative correlation of ambient temperature has been observed with NO2, NOx and NO pollutants, whereas a strong positive correlation (R2>0.64) has been observed with ozone at all study places. Faridabad and Gurugram show moderate negative correlation with SO2. Moderate positive correlation (R2 = 0.32−0.41) has been observed between ozone and solar radiations at all study locations except Noida. In all the cities except Delhi, relative humidity shows positive correlation with NO2, NOx, and NO; and negative correlation with ozone. The negative correlation of O3 with NO2 (R2= -0.33 to -0.57), NOx (R2= -0.45 to -0.56), and NO (R2= -0.28 to -0.49) has been obtained.

Air quality trend analysis from 2015–2020

The trend analysis of pollutants namely PM2.5, NO2, SO2, CO, O3, and Benzene has been carried out from year 2015–2020 and shown in Fig. S6. This analysis covers data for Delhi and its NCR for the period of 1st January-30th November for the years 2015–2020. The average concentration of all the pollutants for year 2015–2019 has been compared with the concentration of year 2020 (i.e., lockdown year). The concentration of PM2.5, NO2, SO2, CO, O3, and Benzene were decreased by nearly 42 %, 11 %, 18 %, 13 %, 6%, and 13 % respectively. The highest reduction obtained for particulate matter shows the effect of restricted activities during this lockdown period.

Discussion

The reduced levels of air pollutants even in the most polluted cities i.e., Delhi and NCR came under the permissible limits as prescribed by CPCB, India during this lockdown period. The concentration of PM10, PM2.5, NOx, NO, NO2, SO2 in the study area, were decreased by 61.6 %, 60.0 %, 58.6 % 62.3 %, 46.8 %, and 33.8 % respectively. These trends are similar to be observed by other studies across the world as shown in Table S4. The most significant reduction was observed for PM10, PM2.5 (∼ 55–65 %) and NOx especially NO level ∼50−78% as shown in radar diagrams (Fig. S7). Studies carried out across the world have shown similar results (Dantas et al., 2020; Nakada and Urban, 2020; Li et al., 2020; Bao and Zhang, 2020; Kerimray et al., 2020; Otmani et al., 2020; Abdullah et al., 2020). This could be explained by the fact that the containment measures taken by Indian Government related to the termination of transportation and industrial activities had its implications on emission release from both vehicular and industrial exhaust.

The primarily anthropogenic source of NOx is high temperature combustion processes of fossil fuels, especially diesel, thermal power generation, industrial emissions, and automobile exhaust (Bhanarkar et al., 2005; Chen et al., 2007). The cities under study have cumulatively highest numbers of vehicles in the country and host a numbers of small and medium industries of chemical, petro-chemicals, metal casting, automobiles, steel, scrap metal, drugs & pharmaceuticals industries, dye and paint. The reduction of NO was higher than those observed for other pollutants. Similarly, consistent and significant reduction in other pollutants such as SO2 (33 %), CO (44.8 %), NH3 (26.6 %) and C6H6 (53 %) was also observed.

An interesting feature observed that during first week of lockdown O3 decreased but later it increased by ∼19−27%. During initial days of lockdown, a significant reduction in O3 levels was observed. NOx plays important role in O3 production. Too little of NOx in ambient atmosphere results in O3 loss (HO2+O3, OH + O3) rather than peroxy radicals cycling (e.g. HO2+NO) leading to net O3 chemical destruction and too much of NOx lead to radical termination by alternate route (e.g. OH + NO2) and consequently production of O3 decreases with increases in NOx (NOx > a few hundred pptv, urban and rural atmosphere which is case here in present study where in the NOX level always higher than thousand pptv). Intermediate level of NOx is efficient for net O3 production via cycling of HOx and NOx radicals.

The negative correlation of ambient temperature with nitrogen oxides and positive with ozone could lead to increased photochemical oxidation of nitrogen oxides contributes in increased ozone formation. The increased solar radiations and ambient temperature result in increasing the photochemical activities and therefore more formation of the ozone in ambient air. These results (positive correlation of ozone with ambient temperature and solar radiation) clearly show the role of photo-chemistry in the formation of tropospheric ozone in the atmosphere. Our results show that with the increase in temperature and solar radiations, nitrogen oxides react with oxidizing agents (such as volatile organic compounds) and results in the formation of ozone. Previously conducted studies (Lee et al., 2010; Ling et al., 2011; Briz-Redón et al., 2020) on ozone formation by photochemical oxidations were also show the similar trend observed in this study. However, we could be able to decipher the exact reasons of such anomalies i.e reduction during initial days and then increase during extended period.

It is vividly clear from this study that the major sources of air pollution in NCR region are local contribution of pollutants from vehicle exhaust, industries emission and dust from constructional activities. Since, there is a little chance of influx or outflux, as the variation in meteorological conditions was not significant as confirmed by ANOVA hypothesis testing. We treat it as one of the most important outcomes of this study to identify the source more convincingly. This ideal condition (complete lockdown) in real world rarely found to pin point the exact cause of air pollution.

Conclusion

Consequent to several strict measures, taken during the lockdown such as limiting public and private transport, closing industrial and commercial activities, the concentration of different pollutants (PM10, PM2.5, NOx, NO, NO2, SO2, NH3, CO, C6H6) reduced significantly. This is mainly attributed to a significant reduction in local emissions from vehicle exhaust and industrial production. This clearly demonstrates that with proper implementation of regulations, norms and standards, there is enough scope to maintain the ambient air quality. Interestingly, it has been observed that in initial days of lockdown O3 levels reduced due to almost constant meteorology, but in later days it increased as increase of solar radiations and ambient temperature which enhance the photochemical activity between volatile organic compounds and nitrogen oxides results in ozone formation. The overall results show that the atmospheric chemistry of pollutants, the emission sources, the meteorological conditions plays a major role in identifying the levels of ambient air pollutants. Since, the results shows that major cause of air pollution in Delhi and NCR is vehicle, industry, and constructional activities, hence it is important to take stringent measures for containing the air pollution and formulate standards and norms to suit present scenario.

Declaration of Competing Interest

The authors report no declarations of interest.

CRediT authorship contribution statement

Anchal Garg: Conceptualization, Methodology, Formal analysis, Investigation, Writing - original draft. Arvind Kumar: Conceptualization, Methodology, Investigation, Writing - review & editing. N.C. Gupta: Conceptualization, Writing - review & editing, Supervision, Visualization, Validation.