Skip to main content
NCBI home page
Elsevier - PMC COVID-19 Collection logoLink to Elsevier - PMC COVID-19 Collection
. 2021 Mar 1;128:102418. doi: 10.1016/j.apgeog.2021.102418

Impact of the COVID-19 lockdown on air quality in the Delhi Metropolitan Region

Shouraseni Sen Roy a,∗, Robert C Balling Jr b
PMCID: PMC9753935  PMID: 36536956

Abstract

With the rapid spread of COVID-19 related cases globally, national governments took different lockdown approaches to limit the spread of the virus. Among them, the Government of India imposed a complete nationwide lockdown starting on March 25, 2020. This presented a unique opportunity to explore how a complete standstill in regular daily activities might impact the local environment. In this study, we have analyzed the change in the air quality levels stemming from the reduced anthropogenic activities in one of the most polluted cities in the world, the Delhi Metropolitan Region (DMR). We analyzed station-level changes in particulate matter, PM10 and PM2.5, across the DMR between April 2019 and 2020. The results of our study showed widespread reduction in the levels of both pollutants, with substantial spatial variations. The largest decreases in particulate matter were associated with industrial and commercial areas. Highest levels of PM10 and PM2.5 were observed near sunrise with little change in the time of maximum between 2019 and 2020. The results of our study highlight the role of anthropogenic activities on the air quality at the local level.

Keywords: COVID-19, PM10, PM2.5, Delhi metropolitan region, Harmonics, Diurnal

1. Introduction

Within a few months in 2020, COVID-19 had massively impacted the entire globe far beyond the direct effects on human health. These impacts vary at different spatial scales as reflected in the number of infections and mortality, as well as the indirect impacts on the economy and environment. Many local and national governments implemented various degrees of lockdown or “shelter at home” policies to contain the spread of this global pandemic. This has resulted in a massive reduction in global economic activity thereby significantly lowering air pollution levels in many areas of the world including China (Wu et al., 2020) and India (Patel et al., 2020). Specifically, substantial declines in ground level nitrogen dioxide, ozone, and particulate matter were reported within the first two weeks of the lockdown in 27 countries as revealed from the analysis of satellite data during February and March 2020 (Venter et al., 2020). However, due to the coarse spatial and temporal resolution of the satellite data, their study was limited in revealing the local level patterns in air pollution. This is particularly relevant in parts of Asia, such as in India, where most of the high pollution levels are concentrated in the large urban areas. Therefore, in the present study we have analyzed the change in levels of two pollutants, PM10 and PM2.5, over a one month period between April 2019 and 2020 in the Delhi Metropolitan Region (DMR), India (Fig. 1 ).

Fig. 1.

Fig. 1

Open in a new tab

Distribution of air pollution monitoring stations in the Delhi Metropolitan Region (DMR).

The Government of India imposed lockdown on all of its 1.3 billion citizens on March 25, 2020 to limit the spread of the COVID-19; the lockdown ended after 55 days on May 19, 2020. As a result, there was widespread shutdown of factories, road, and air traffic across the DMR. The shutdown was so severe that it triggered a massive migration of wage laborers walking for several 100s of kilometers across northern India to their villages caused by no available jobs and the cancellation of long distance transportation such as railways and bus systems. The initial results from satellite image analysis of aerosol optical depths revealed a significant drop across the northern plains resulting from the closure of coal-fired heavy industries and reduction in traffic across large urban areas like the DMR (Patel et al., 2020). In addition, there have been widespread reports in mainstream media outlets about the reduction in pollution levels in the DMR dropping from unhealthy and hazardous to good (Business Today, 2020; Ellis-Peterson et al., 2020). These trends have also been documented in recently published studies. For instance, Shehzad et al. (2020) analyzed Sentinel – 5 P satellite images and reported a significant improvement in air quality and a 40–50% decrease in atmospheric nitrogen dioxide levels in the large cities of India, including DMR and Mumbai. More detailed analysis of station level data for the DMR during pre and post lockdown periods revealed significant decreases in levels of particulate matter within days of the lockdown (Kumari & Toshniwal, 2020; Mahato et al., 2020). Additionally, detailed hourly analysis of air pollutants revealed a decline in the levels of particulate matter during nighttime and peak traffic hours (Singh et al., 2020).

This is particularly significant in view of the DMR and its suburbs being ranked as the most polluted city in the world for at least the last five years. Specifically, the pollutants with consistently high levels in the DMR are PM10, PM2.5, NO2, SO2, and O3 (Aneja et al., 2001; Balachandran et al., 2000; Gurjar et al., 2004; Kandikar 2007; Nagar et al., 2019; O'Shea et al., 2015; Sahu et al., 2011). Furthermore, India has recorded an increasing trend in population-weighted mean concentrations of PM2.5, with a noticeable increase since 2010 (Bhakta et al., 2019; Gurjar et al., 2016). These elevated concentrations of PM2.5 exposures have been attributed to annual premature death estimates of 272,000 for chronic obstructive pulmonary disease (COPD), 110,600 for ischemic heart disease, and 14,800 for lung cancer (Chowdhury & Dey, 2016). Specifically, for the DMR a reduction in life expectancy due to exposure to PM2.5 was 6.3 ± 2.2 years greater than the same for overall Indian population (3.4 ± 1.1 years) (Ghude et al., 2016).

Due to the paucity of data, most of the above mentioned station level studies in the DMR have examined the variations for select pollutants in limited number of stations, ranging between two and five. The results from these studies revealed significant variations in the levels of various pollutants and diurnal cycles across the study area. All of these studies highlighted the excessive readings and increase in the levels of pollutants in the DMR, mainly resulting from traffic congestion and industrial activities. Thus, with the implementation of the strict measures associated with the lockdown, the drop in air pollution levels is distinctly visible and merits detailed analysis. Therefore, the two specific objectives of the study are:

  • 1.

    Determine the local-level spatial patterns of change in the levels of two major pollutants, PM10 and PM2.5, between April 2019 and 2020.

  • 2.

    Analyze the changes in the time of maximum for peak levels in PM10 and PM2.5 between April 2019 and 2020.

2. Data and methods

Station-level hourly data for PM10 and PM2.5 were obtained from the Central Pollution Control Board (CPCB), collected as part of the Air Quality Monitoring Program in India. The data were collected for April 2019 and 2020 to assess the change in pollution levels due to the lockdown. There are 40 pollution monitoring stations spread across the DMR, out of which there were continuous data available for 34 stations for PM10 and 31 stations for PM2.5 (Fig. 1). We analyzed only two pollutants PM10 and PM2.5 because these were the only two pollutants with complete data across all the stations. The data for the other pollutants were incomplete and thus not suitable for analysis. Furthermore, these two pollutants have consistently been the biggest challenge for air quality scientists and policymakers and therefore the results of the study are particularly relevant. In order to test the spatial distribution of the station network across the study area, the nearest-neighbor statistics were calculated as the ratio between the observed mean distance among the station locations and the expected mean distance given a random distribution (Clark & Evans, 1954). The ratios of 1.03 for PM10 and 0.90 for PM2.5 show that both station networks have a random distribution.

We analyzed the changes in the levels of PM10 and PM2.5 for the monthly average values by calculating the differences at the station level. The percentage change in the levels of the pollutants at the station level were visualized using spatial interpolation, simple kriging. Kriging is a surface interpolation method utilized to visualize spatial variation through a variogram, thus minimizing prediction errors (Oliver & Webster, 1990; Sen Roy, 2006a). The final interpolated surface is calculated by incorporating the spatial and statistical relationships among the different variables by using the following equation:

Z(s) = μ + ε(s)

Where μ is a known constant utilized to interpolate the resulting surface, s denotes the location being predicted, and ε(s) is the error term (Sen Roy, 2006b). This method was preferred because of its accuracy in surface interpolation and lower root mean square error.

Harmonic analysis was used to determine the time of maximum concentration of PM10 and PM2.5 levels based on the average hourly values over the course of a month. To fit a trigonometric wave with one maxima and one minima, the harmonic equation for any station with 24 values takes the form:

f(x)=X‾+∑r=1N/2Arcos(rθ−Φr)

where f(x) is the estimated value in each interval, X‾ is the average value over the N = 24 intervals, A r is the amplitude of the rth harmonic wave, r is the frequency or number of times the harmonic wave is repeated over the fundamental period (in this case r = 1), θ is derived as 2πx/N where x signifies the intervals over the fundamental period, and Φris the phase angle of the rth harmonic reinterpreted as the time of maximum. The basic form is explained below:

f(x)=X‾+∑r=1N/2[arsin(2πrx/P)+brcos(2πrx/P)]

where the Fourier coefficients, a r and b r, are calculated as:

ar=∑x=1N2N[f(x)sin(2πrx/P)]

and

br=∑x=1N2N[f(x)cos(2πrx/P)]

The amplitude, A r, is calculated as (a r 2 + b r 2 ) 0.5, the phase angle, Φr, equals tan−1 (a r /b r), and the portion of variance denoted by the rth harmonic wave, V r, is calculated as A r 2 /2s 2 wherein s is the standard deviation of the N values (see Nelson, 1983, p. 190). Given that the harmonic wave is fitted to PM10 or PM2.5 averages in 24 hourly intervals, the explained variance levels had to be > 0.16 to be statistically significant at the 0.05 level of confidence. This method has been successfully used to study diurnal patterns in air pollution variables in previous studies (Liu & Sen Roy, 2014; O'Shea et al., 2015).

3. Results and discussion

The DMR is located in the northern interior of the Indian subcontinent, and thus experiences a typical continental climate. The comparative results from the two different years reveal substantially lower levels of in the levels of air pollution between the two years with distinct spatial variations across the DMR (Fig. 2 ). Typical of megacities of Asia, the DMR is densely populated (greater than 25,000 persons per km2), with an estimated total population of about 17 million. Majority of the DMR is urban, with 97.5% of the population identified as urban (Sen Roy et al., 2020). According to Jain et al. (2016), the net percent change in land use from 1977 to 2014 for urban built-up areas increased by 30.61%, along with a decrease by 22.75% for cultivated areas, 5.31% for dense forest, and 2.76% for wasteland. With the steep increase in population accompanied by car ownership and unplanned urbanization, the DMR has experienced steep increase in levels of air pollution over the years. The extremely hazardous levels of air pollution in the winter months has led to the forced closure of schools and steep increase respiratory diseases among the local population. The sources of air pollution are not only from local sources such as vehicular emissions and industrial activities, but more recently the burning of crops after harvesting in the agricultural fields in the neighboring states of Punjab and Haryana. Thus, the pollution levels noted in 2019 typifies that time of the year in the DMR over the last decade.

Fig. 2.

Fig. 2

Open in a new tab

Changes in average hourly levels of particulate matter between April 2019 and 2020 (a) PM10; (b) PM2.5.

As is evident from Fig. 2, the level for both pollutants in 2019 were higher than those observed in 2020 for all the stations. The percentage change for PM10 ranged between 20 and 70%, while the range of decline was wider for PM2.5 at 15–90%. At the local level, 21 out of 31 stations experienced greater than 50% decline for PM10. The largest declines were concentrated in the eastern half of the DMR, which also consistently experienced higher levels of particulate matter. Moreover, for the PM2.5 the largest percentage declines (greater than 50%) were observed for only 8 out of 34 stations located in the central and eastern parts of the DMR. The greater decline in central DMR is due to the lower levels of economic activity associated with the high density of office buildings and commercial areas in the core downtown area. Similarly, the higher rates of decline in the eastern DMR are associated with the closure of local factories and thermal power plants. Specifically, there are three thermal power plants located in the DMR, which include Indraprastha Power Station and Rajghat Power House in the east, and Badarpur Power Station in the northwest. Majority of the land use in the northern and western DMR is residential, and thus experienced relatively lower levels of decline in the pollutants. Two stations, located in Shadipur (west of central downtown) and Dilshad Garden (east) experienced greater than 85% decline in PM2.5 (Fig. 2b and Table 2 ). The lowest differences in average hourly levels of PM10 were located outside the central core of the DMR, such as Najafgarh (192 μg/m3 in 2019 vs 146.1ug/m3 in 2020), Ashok Vihar (168.2ug/m3 in 2019 vs 95.42μg/m3 in 2020), and Aya Nagar (155.7ug/m3 in 2019 vs 78.53μg/m3 in 2020) (Table 1, Fig. 2a). Among these three stations, two of them are predominantly residential, while Najafgarh consists of transitioning from rural to urban land uses mixed with industries. Similarly in the case of PM2.5, the lowest differences in average hourly levels were observed in predominantly residential areas, including Lodhi Road (57μg/m3 in 2019 vs 46.5ug/m3 in 2020), Aya Nagar (51.8ug/m3 in 2019 vs 39.1ug/m3 in 2020), and Punjabi Bagh (155.7ug/m3 in 2019 vs 73.4ug/m3 in 2020) (Table 2, Fig. 2b). Moreover, the hourly maximum values between the two years revealed higher values across most of the stations during 2019 for both PM10 and PM2.5, except in Punjabi Bagh (for both pollutants) and Pusa DPCC (PM2.5) in West Delhi (Table 1, Table 2). There was also a greater amount of variance in the hourly maximum levels for both of the pollutants during 2019 compared to 2020. However, the patterns were not as distinct in the case of hourly minimum values across the DMR. Our results are in conformity with the results of a previous study by Tiwari et al. (2015), who found substantial spatial variations in the correlation between observed levels of PM10 and PM2.5 at the seasonal scale and weekday vs weekends for a limited number of stations. Specifically, their analysis of hourly level observations of these two particulates revealed the mean coarse mode particulate mass concentration (PM10–2.5) as 113.6 ± 70.4 μg/m3 (varied from 13.6 to 630 μg/m3) constituting about 49% contribution of PM2.5 to PM10 concentrations. Moreover, Singh et al. (2013) found distinct seasonal patterns in the levels of PM10–2.5 across the DMR, ranging from highest during the dry summer months to lowest during the cold winter months. This is due to the greater load of coarser particles during the summer months as a result of dust transport form surrounding areas during the summer months compared to winter months.

Table 2.

Descriptive statistics for PM2.5 maximum, minimum, and average levels.

Station Year Maximum Standard Deviation Minimum Standard Deviation Average
Alipur 2019 447.5 108.302 8.75 4.55144 83.0603
Alipur 2020 149 38.1889 6.25 5.39188 47.5502
AshokVihar 2019 576.5 128.339 6 7.5641 95.3493
AshokVihar 2020 170 48.7817 5 5.67683 48.7782
Aya Nagar 2019 176.64 27.9336 0.13 7.65842 51.8133
Aya Nagar 2020 133.13 18.3072 0.16 3.08136 39.103
Bawana 2019 451 104.909 8.25 8.51754 103.071
Bawana 2020 279 66.855 8 6.92493 65.5673
CRRI_Mathura 2019 612.91 93.1076 0.08 6.39099 83.0624
CRRI_Mathura 2020 238.57 52.6926 0.82 5.58828 44.0428
DTU 2019 414.89 89.0331 12.37 5.69309 88.7776
DTU 2020 219.47 55.6613 3.62 5.80001 51.77
Karni Singh 2019 571 106.454 1 5.92663 65.8622
Karni Singh 2020 111.75 24.6156 2 5.78783 32.2407
Dwarka 2019 321.75 78.8187 3.75 7.60587 79.0549
Dwarka 2020 228.25 60.6285 5.75 4.69812 49.5281
IGI 2019 314.41 52.7348 2.24 5.73481 69.3283
IGI 2020 166.18 39.2024 0.6 3.40492 38.6598
IHBAS 2019 521.38 105.401 15 12.1524 111.183
IHBAS 2020 81.25 10.9892 10 0.15108 13.4519
ITO 2019 580 112.367 13 8.42679 88.8093
ITO 2020 329.5 84.6624 20 1.90715 74.5252
JhPuri 2019 806 143.623 1 9.89721 108.849
JhPuri 2020 184.25 46.0978 3.5 4.62052 53.7597
JLN 2019 294.25 65.4434 1.5 6.60945 65.0576
JLN 2020 179.25 42.377 1.25 4.89172 37.301
LodhiRoad 2019 220.37 37.1103 0.7 7.87626 56.9681
LodhiRoad 2020 157.77 35.5614 0.16 3.68312 46.4696
Major Dhyan 2019 741.5 141.564 8.25 7.25418 70.3656
Major Dhyan 2020 136.75 30.6355 3.5 4.1349 40.911
Mandir Marg 2019 341.25 55.2182 6.25 10.8533 76.6734
Mandir Marg 2020 153 31.0096 3.5 4.48487 36.8879
Mundka 2019 341 77.0833 4 7.28381 96.7941
Mundka 2020 276.25 68.01 4 5.59664 60.0536
NSIT 2019 607.36 134.288 6.57 7.24242 107.903
NSIT 2020 130.63 19.5396 21.96 4.09561 51.6994
Najafgarh 2019 268.5 68.7479 3 4.87394 68.406
Najafgarh 2020 191.5 57.9931 3.5 4.7637 47.9101
Narela 2019 542 113.72 10 6.85347 89.8291
Narela 2020 318.5 77.0327 4 12.0454 61.4989
North Campus 2019 381.93 83.2922 0.54 8.92598 93.7665
North Campus 2020 130.52 22.9001 0.24 1.94323 30.6803
Okhla 2019 288 65.7498 6.5 6.63966 73.1985
Okhla 2020 204 54.6407 7 4.95957 43.193
Patparganj 2019 320 71.1421 7 5.98171 67.463
Patparganj 2020 186.75 40.254 2.25 4.56474 36.5616
Punjabi Bagh 2019 328.5 75.4048 1 6.70212 73.3961
Punjabi Bagh 2020 715.25 205.564 2 4.98458 59.567
PusaDPCC 2019 301.5 62.3228 0.5 6.63007 61.599
PusaDPCC 2020 433.25 101.975 1 10.0395 41.749
RKP 2019 100 9.89105 3 19.6651 62.3571
RKP 2020 186 44.4902 1 6.07955 38.8317
Rohini 2019 523 118.413 7 7.25465 94.5361
Rohini 2020 376 74.5775 7 5.59098 60.8941
Shadipur 2019 754.18 176.328 7.62 10.9735 125.004
Shadipur 2020 88.75 15.9437 10 0.86196 18.426
SiriFort 2019 573 125.781 2.25 7.88153 76.1917
SiriFort 2020 459 83.9472 5 4.71554 42.7199
SoniaVihar 2019 383 89.404 5 10.0589 80.8302
SoniaVihar 2020 174.5 39.6314 5 5.06475 42.1439
SriAurobindo 2019 296.75 61.9354 1 5.94034 58.9903
SriAurobindo 2020 164.5 38.0119 3.5 4.24435 36.7326
VivekVihar 2019 592.75 146.715 2 9.02831 82.0469
VivekVihar 2020 228 61.9504 5 4.70387 45.9259
Wazirpur 2019 460.5 99.0444 12.5 10.2953 101.449
Wazirpur 2020 195.5 49.8522 7.5 4.98144 51.4931
Open in a new tab

Table 1.

Descriptive statistics for PM10 maximum, minimum, and average levels.

Station Year Maximum Standard Deviation Minimum Standard Deviation Average
Alipur 2019 862 127.924 30.25 14.1712 228.103
Alipur 2020 401.5 71.1405 30.75 10.8133 134.402
AshokVihar 2019 571 74.2654 35 10.7427 168.204
AshokVihar 2020 316 59.4934 17 5.89851 95.4233
AnandVihar 2019 929.75 156.419 53 25.1023 297.312
AnandVihar 2020 244.25 48.0845 26 16.5048 99.7517
Aya Nagar 2019 659.19 134.415 2.72 17.2113 155.674
Aya Nagar 2020 267.64 43.9914 2.5 8.94845 78.5316
Bawana 2019 879 135.246 19 20.1892 293.964
Bawana 2020 510 88.3488 40 13.8379 157.693
CRRI_Mathura 2019 722.3 110.78 9.57 24.3945 224.479
CRRI_Mathura 2020 522.22 82.9647 10.39 8.18575 105.91
DTU 2019 1000 205.991 18.75 17.3706 261.663
DTU 2020 359 64.773 23 8.67686 124.096
Karni Singh 2019 727 94.1505 7.5 11.7732 209.521
Karni Singh 2020 327 54.2852 21 7.14694 94.3364
Dwarka 2019 928 204.873 17.25 30.203 282.942
Dwarka 2020 488.5 87.5935 29 5.06922 116.574
IGI 2019 946.88 173.898 9.21 16.6022 224.754
IGI 2020 294.79 49.5147 2.73 9.04335 88.1071
ITO 2019 888 133.592 24 14.957 180.733
ITO 2020 336 76.5181 11 14.7504 92.4485
JhPuri 2019 757 90.972 17 21.3399 270.208
JhPuri 2020 374 69.8787 25 10.2574 126.443
JLN 2019 572.5 61.3752 12.5 17.7128 227.902
JLN 2020 350.75 59.2342 22.25 7.00038 95.8382
LodhiRoad 2019 693.88 116.096 11.53 12.9554 175.874
LodhiRoad 2020 348.12 85.1159 0.41 12.3119 87.5506
Major Dhyan 2019 498.75 56.9316 8.5 18.5752 200.816
Major Dhyan 2020 318.25 52.341 16 7.2127 89.018
Mandir Marg 2019 595 72.964 37 21.2579 229.153
Mandir Marg 2020 373.75 75.3458 16 8.43321 93.2862
Mundka 2019 973 124.8 19.5 21.7858 337.353
Mundka 2020 418.75 60.6591 23.25 7.09626 134.142
Najafgarh 2019 818.5 158.865 12.25 10.2112 191.994
Najafgarh 2020 614.5 101.876 18 16.3333 146.059
Narela 2019 721 88.8081 32.5 11.4617 277.344
Narela 2020 516 102.888 34.5 18.6923 157.114
North Campus 2019 852.95 113.615 8.74 23.7951 289.572
North Campus 2020 291.06 53.8835 0.3 11.1233 90.9651
Okhla 2019 875 120.866 17.5 11.9767 194.985
Okhla 2020 442 78.2785 27.5 6.75701 100.779
Patparganj 2019 723.75 113.481 62.5 23.5294 206.623
Patparganj 2020 389.25 71.8545 19 4.7531 81.7944
Punjabi Bagh 2019 675 90.6604 16.25 18.5927 223.43
Punjabi Bagh 2020 790.75 207.079 27.5 8.80379 112.304
PusaDPCC 2019 791 101.408 12.25 27.7238 221.084
PusaDPCC 2020 567 129.393 15.75 17.7111 88.3698
RKP 2019 850.75 120.056 17.75 20.104 235.803
RKP 2020 428 85.9955 11.25 11.5326 90.7663
Rohini 2019 781 122.662 13.75 17.467 268.78
Rohini 2020 566.75 92.7613 35 8.4801 137.409
SiriFort 2019 986.25 195.796 18 46.049 301.46
SiriFort 2020 563.25 105.185 25 6.1976 94.1084
SoniaVihar 2019 497 33.011 16 23.0697 233.819
SoniaVihar 2020 324 53.0742 25 7.43589 101.207
SriAurobindo 2019 748 116.038 4 16.5317 194.134
SriAurobindo 2020 238.5 35.6937 15.5 6.02407 71.6146
VivekVihar 2019 962 138.55 51 16.2336 245.17
VivekVihar 2020 428 85.9023 50 3.17364 107.087
Wazirpur 2019 907 109.841 31 30.9399 305.448
Wazirpur 2020 331 62.0834 24 9.41813 106.565
Open in a new tab

In addition, we examined the change in monthly peak time of maximum for the two pollutants between the two years (Fig. 3 ). The peak time of maximum showed a gradual progression from after midnight to early morning hours for PM10 across most of the stations in a north to south direction during both years (Fig. 3a). In the case of PM2.5, the peak time of maximum occurrence occurred closer to the early morning hours, with a few hours earlier occurrence in the north relative to the south (Fig. 3b). The midnight to early morning maximum observed for both the pollutants can be attributed to the minimum variations in the convective available potential energy (CAPE) and other thermodynamic parameters in the early morning hours in the DMR (Ratnam et al., 2013). In addition there is relative lower atmospheric boundary layer and high traffic density in the early morning hours in the DMR. Similar results of higher levels of PM2.5 early morning and midnight were also found by Bhakta et al. (2019) from the analysis of 2 years of data at one station in the DMR. The results of their study also revealed a strong negative correlation between air temperatures and levels of PM2.5.

Fig. 3.

Fig. 3

Open in a new tab

Spatial distribution of peak time of maximum (a) Average PM10; (b) Average PM2.5. The symbols pointing north indicate time of maximum at midnight, those pointing south indicate time of maximum at noon, and those pointing west indicate maximum at 6 p.m., and so on.

As seen in Fig. 3, the times of maximum PM10 or PM2.5 levels did not change appreciably between 2019 and 2020 at most stations. For each pollutant, the mean difference in the time of maximum across the station network was essentially equal to the standard error of the estimate in calculating the mean, thereby suggesting that the difference is not statistically significant. We also analyzed the change in the time of maximum for the monthly maximum and minimum levels, and the patterns were predominantly similar to that observed for the average monthly levels.

The different sources of air pollution in the DMR are well documented, which include industrial activities, transport, road side construction, and regional emission sources that contribute a significant fraction to aerosol mass loading in the region (Nagpure et al., 2013; Saxena et al., 2014, 2017; Sen et al., 2016; Sharma et al., 2016). Thus, the almost complete stop in the rush hour traffic and other anthropogenic activities, including industrial and construction, can be considered as the major factors for the substantial decline in levels of particulate matter at the local level. It is also noteworthy that the spatial and temporal patterns of particulate matter in the DMR are a result of anthropogenic activities across the wider densely populated northern plains. The advective transport of particulate matter across the northern plains and consequent dispersal of pollutants to the marine atmospheric boundary layer of the Bay of Bengal has been well documented (Lelieveld et al., 2001; Sudheer & Sarin, 2008). Since the lockdown was at the national level, the results of our study can be representative of the levels of air pollution across the wider region of the Indian subcontinent.

4. Conclusions

In the present study we have examined the impacts of lockdown in the DMR on the spatial patterns of air quality during April 2020. We analyzed two variables, PM2.5 and PM10 at the station level during April 2019 and 2020. The main findings of our study are summarized below:

  • 1.

    There was substantial decline in the levels of PM10 (20–70%) and PM2.5 (15–90%) in air quality across the DMR.

  • 2.

    Spatially, the highest decline for the particulate matter was observed over the downtown core area and the adjacent industrial areas in the east and west.

  • 3.

    The areas experiencing greater decline in the levels of particulate matter are associated with greater proportion of commercial land uses and economic activity in the form of offices and industries. Overall, the decline in PM10 was more widespread than PM2.5.

  • 4.

    The diurnal patterns of the time of maximum for average monthly levels occurred closer to midnight for PM10 and early morning hours for PM2.5, which were in conformity with the results of previous studies.

  • 5.

    The time of maximum values did not change significantly between 2019 and 2020.

The results of our study highlight some of the positive impacts of the lockdown during a one month period on the local environment. As evident from the results of previously published studies, elevated levels of particulate matter in the DMR have led to increased rates of premature mortality in the DMR. An analysis of the relative contribution of various sectors to the levels of PM2.5 in the DMR and its surrounding area revealed transportation as the leading sector, followed by residential (in the form of wood, coal, kerosene, cow dung used with poor combustion technology in informal settlements, and liquefied petroleum gas with less emission in almost all houses), power plants (coal as fuel), and industrial sectors (Jain et al., 2018; Sahu et al., 2011). Therefore, with the implementation of a complete lockdown in the DMR leading to a steep decline in anthropogenic activities in the transportation and industrial sector resulted in the substantial improvement in air quality. Moreover, both of these pollutants have been consistently above the national standards and persistently represented a major challenge for policymakers and air quality scientists. However, it is noteworthy that the steep declines in levels of pollutants observed in different parts of the DMR have come at substantial social and economic costs, which make them difficult to sustain in the long-term. Further analysis is required to examine the implementation of similar phased lockdowns without excessive negative socio-economic impacts to achieve a more sustained decrease in levels of air pollution.

This is particularly critical in view of the increased mortality, particularly an 11% increase in cardiovascular mortality as result of a 10 μgm−3 increase of PM2.5 (Bourdrel et al., 2017). Additionally, PM2.5 has been identified as the 5th risk factor of mortality, with 59% of those occurring in East and South Asia (Cohen et al., 2017). Therefore, it would be worthwhile to explore the impact of lower levels of particulate matter on the general health of the population in the DMR once the appropriate data are available for analysis.

Author statement

Shouraseni Sen Roy and Robert C. Balling Jr.: Conceptualization, Methodology, Data curation, Visualization, Investigation, Writing- Original draft preparation, Reviewing, and Editing.

References

  1. Aneja V.P., Agarwal A., Roelle P.A., Phillips S.B., Tong Q.S., Watkins N., Yablonsky R. Measurements and analysis of criteria pollutants in New Delhi, India. Environment International. 2001;27:35–42. doi: 10.1016/s0160-4120(01)00051-4. [DOI] [PubMed] [Google Scholar]
  2. Balachandran S., Meena B.R., Khillare P.S. Particle size distribution and its elemental composition in the ambient air of Delhi. Environment International. 2000;26:49–54. doi: 10.1016/s0160-4120(00)00077-5. [DOI] [PubMed] [Google Scholar]
  3. Bhakta R., Khillare P.S., Jyethi D.S. Atmospheric particulate matter variations and comparison of two forecasting models for two Indian megacities. Aerosol Science and Engineering. 2019;3(2):54–62. [Google Scholar]
  4. Bourdrel T., Bind M.A., Béjot Y., Morel O., Argacha J.F. Cardiovascular effects of air pollution. Archives of Cardiovascular Diseases. 2017;110:634–642. doi: 10.1016/j.acvd.2017.05.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Business Today Lockdown reduces Delhi pollution; air quality turns 'good' from 'hazardous.' Business Today. 2020. https://www.businesstoday.in/latest/trends/lockdown--reduces-delhi-pollution-air-quality-turns-good-from-hazardous/story/399397.html
  6. Chowdhury S., Dey S. Cause-specific premature death from ambient PM2.5 exposure in India: Estimate adjusted for baseline mortality. Environment International. 2016;91:283–290. doi: 10.1016/j.envint.2016.03.004. [DOI] [PubMed] [Google Scholar]
  7. Clark P.J., Evans F.C. Distance to nearest neighbor as a measure of spatial relationships in populations. Ecology. 1954;35:445–453. [Google Scholar]
  8. Cohen A.J., Brauer M., Burnett R., et al. Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: An analysis of data from the global burden of diseases study 2015. The Lancet. 2017;389:1907–1918. doi: 10.1016/S0140-6736(17)30505-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Ellis-Petersen H., Ratcliffe R., Cowie S., Daniels J.P., Kuo L. It's positively alpine!: Disbelief in big cities as air pollution falls. The Guardian. 2020. https://www.theguardian.com/environment/2020/apr/11/positively-alpine-disbelief-air-pollution-falls-lockdown-coronavirus
  10. Ghude S.D., Chate D.M., Jena C., Beig G., Kumar R., Barth M.C., Pfister G.G., Fadnavis S., Pithani P. Premature mortality in India due to PM2.5 and ozone exposure. Geophysical Research Letters. 2016;43:4650–4658. [Google Scholar]
  11. Gurjar B.R., Ravindra K., Nagpure A.S. Air pollution trends over India megacities and their local-to-global implications. Atmospheric Environment. 2016;142:475–495. [Google Scholar]
  12. Gurjar B.R., van Aardenne J.A., Lelieveld J., Mohan M. Emission estimates and trends (1990–2000) for mega city Delhi and implications. Atmospheric Environment. 2004;38:5663–5681. [Google Scholar]
  13. Jain M., Dawa D., Mehta R., Dimri A.P., Pandit M.K. Monitoring land use change and its drivers in Delhi, India using multi-temporal satellite data. Modeling Earth Systems and Environment. 2016;2(1):19. [Google Scholar]
  14. Jain S., Sharma S.K., Mandal T.K., Saxena M. Source apportionment of PM10 in Delhi, India using PCA/APCS, UNMIX and PMF. Particuology. 2018;37:107–118. [Google Scholar]
  15. Kandlikar M. Air pollution at a hotspot location in Delhi: Detecting trends, seasonal cycles and oscillations. Atmospheric Environment. 2007;41:5934–5947. [Google Scholar]
  16. Kumari P., Toshniwal D. Impact of lockdown measures during COVID-19 on air quality–A case study of India. International Journal of Environmental Health Research. 2020:1–8. doi: 10.1080/09603123.2020.1778646. [DOI] [PubMed] [Google Scholar]
  17. Lelieveld J.O., Crutzen P.J., Ramanathan V., Andreae M.O., Brenninkmeijer C.A.M., Campos T., Cass G.R., Dickerson R.R., Fischer H., De Gouw J.A., Hansel A. The Indian ocean experiment: Widespread air pollution from south and southeast Asia. Science. 2001;291(5506):1031–1036. doi: 10.1126/science.1057103. [DOI] [PubMed] [Google Scholar]
  18. Liu Z., Sen Roy S. Spatial patterns of seasonal level diurnal variations of ozone and respirable suspended particulates in Hong Kong. The Professional Geographer. 2014;67(1):17–27. doi: 10.1080/00330124.2014.886922. [DOI] [Google Scholar]
  19. Mahato S., Pal S., Ghosh K.G. Effect of lockdown amid COVID-19 pandemic on air quality of the megacity Delhi, India. The Science of the Total Environment. 2020 doi: 10.1016/j.scitotenv.2020.139086. 139086. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Nagar P.K., Sharma M., Das D. A new method for trend analyses in PM10 and impact of crop residue burning in Delhi, Kanpur and Jaipur, India. Urban Climate. 2019;27:193–203. [Google Scholar]
  21. Nagpure A.S., Sharma K., Gurjar B.R. Traffic induced emission estimates and trends (2000–2005) in megacity Delhi. Urban Climate. 2013;4:61–73. [Google Scholar]
  22. Nelson H. Addison-Wesley; Reading, Massachusetts: 1983. Harmonic analysis. [Google Scholar]
  23. Oliver M.A., Webster R. Kriging: A method of interpolation for geographical information systems. International Journal of Geographical Information Systems. 1990;4:313–332. [Google Scholar]
  24. O'Shea P.M., Sen Roy S., Singh R.B. Diurnal variations in the spatial patterns of air pollution across the Delhi metropolitan region. Theoretical and Applied Climatology. 2015;124(3):609–620. doi: 10.1007/s00704-015-1441-y. [DOI] [Google Scholar]
  25. Patel K. Airborne particle levels plummet in northern India. 2020. https://earthobservatory.nasa.gov/images/146596/airborne-particle-levels-plummet-in-northern-india
  26. Ratnam M.V., Santhi Y.D., Rajeevan M., Rao S.V.B. Diurnal variability of stability indices observed using radiosonde observations over a tropical station: Comparison with microwave radiometer measurements. Atmospheric Research. 2013;124:21–33. [Google Scholar]
  27. Sahu S.K., Beig G., Parkhi N.S. Emissions inventory of anthropogenic PM2. 5 and PM10 in Delhi during commonwealth games 2010. Atmospheric Environment. 2011;45(34):6180–6190. [Google Scholar]
  28. Saxena M., Sharma A., Sen A., Saxena P., Mandal T.K., Sharma S.K., Sharma C. Water soluble inorganic species of PM10 and PM2. 5 at an urban site of Delhi, India: Seasonal variability and sources. Atmospheric Research. 2017;184:112–125. [Google Scholar]
  29. Saxena M., Singh D.P., Saud T., Gadi R., Singh S., Sharma S., Mandal T.K. Study on particulate polycyclic aromatic hydrocarbons over Bay of Bengal. Atmospheric Research. 2014;145– 146:205–217. [Google Scholar]
  30. Sen Roy S. Impact of lunar cycle on the precipitation in India. Geophysical Research Letters. 2006;33(1) [Google Scholar]
  31. Sen Roy S. The impacts of ENSO, PDO, and local SSTs on winter precipitation in India. Physical Geography. 2006;27:464–474. [Google Scholar]
  32. Sen Roy S., Rahman A., Ahmed S., Shahfahad, Ahmad I.A. Alarming groundwater depletion in the Delhi metropolitan region: A long-term assessment. Environmental Monitoring and Assessment. 2020;192 doi: 10.1007/s10661-020-08585-8. [DOI] [PubMed] [Google Scholar]
  33. Sen A., Ahammed Y.N., Banerjee T., Chatterjee A., Choudhuri A.K., Das T., Deb N.C., Dhir A., Goel S., Khan A.H., Mandal T.K., Murari V., Rao P.S., Saxena M., Sharma S.K., Sharma A., Vachaspati C.V. Spatial variability in ambient atmospheric fine and coarse mode aerosols over indo-Gangetic plains, India and adjoining oceans during the onset of summer monsoons. Atmos. Poll. Res. 2016;7:521–532. [Google Scholar]
  34. Sharma S.K., Mandal T.K., Jain S., Saraswati S.A., Saxena M. Source apportionment of PM2.5 in Delhi, India using PMF model. Bulletin of Environmental Contamination and Toxicology. 2016;97:286–293. doi: 10.1007/s00128-016-1836-1. [DOI] [PubMed] [Google Scholar]
  35. Shehzad K., Sarfraz M., Shah S.G.M. The impact of COVID-19 as a necessary evil on air pollution in India during the lockdown. Environmental Pollution. 2020;266:115080. doi: 10.1016/j.envpol.2020.115080. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Singh V., Singh S., Biswal A., Kesarkar A.P., Mor S., Ravindra K. Diurnal and temporal changes in air pollution during COVID-19 strict lockdown over different regions of India. Environmental Pollution. 2020;266:115368. doi: 10.1016/j.envpol.2020.115368. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Singh K., Tiwari S., Jha A.K., Aggarwal S.G., Bisht D.S., Murty B.P., Khan Z.H., Gupta P.K. Mass-size distribution of PM 10 and its characterization of ionic species in fine (PM 2.5) and coarse (PM 10− 2.5) mode, New Delhi, India. Natural Hazards. 2013;68(2):775–789. [Google Scholar]
  38. Sudheer A.K., Sarin M.M. Carbonaceous aerosols in MABL of Bay of bengal: Influence of continental outflow. Atmospheric Environment. 2008;42(18):4089–4100. [Google Scholar]
  39. Tiwari S., Hopke P.K., Pipal A.S., Srivastava A.K., Bisht D.S., Tiwari S., Singh A.K., Soni V.K., Attri S.D. Intra-urban variability of particulate matter (PM2. 5 and PM10) and its relationship with optical properties of aerosols over Delhi, India. Atmospheric Research. 2015;166:223–232. [Google Scholar]
  40. Venter Z.S., Aunan K., Chowdhury S., Lelieveld J. COVID-19 lockdowns cause global air pollution declines with implications for public health risk. 2020. https://www.medrxiv.org/content/10.1101/2020.04.10.20060673v1.article-metrics [DOI] [PMC free article] [PubMed]
  41. Wu X., Nethery R.C., Sabath B.M., Braun D., Dominici F. medRxiv; 2020. Exposure to air pollution and COVID-19 mortality in the United States. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Applied Geography (Sevenoaks, England) are provided here courtesy of Elsevier

RESOURCES