Due to higher rate of global spread by human interventions during the recent COVID-19 crisis [1], a significant reduction in PM2.5, PM10, NO2, SO2 and CO levels were observed in many major cities of the world like New York, Madrid, Sydney, Paris, Wuhan, Nanjing, Bangalore, and Delhi [2]. The prominent findings from many studies showed that the lockdown epochs of COVID-19 decreased the demands and hence the anthropogenic emissions which lead to cutting down of concentrations of pollutants in atmosphere. The huge reduction in transport demands, industrial work and energy demands due to lockdown played vital role in the reduction of the above-mentioned pollution constituents [3]. Similarly in Italy the 20 % reduction of GHG emissions and total carbon footprint were observed [4]. Similarly, the significant reductions in pollutants were also observed over India. The AOD over India reduced during the period of lockdown [5]. Significant reduction was observed in PM2.5, NO2 and CO by from Ambient Air Quality Monitoring stations established in different sites in Lucknow and Delhi, India [6]. In Delhi the maximum reduction was observed in PM10 (60%) followed by NO2 (52.5 %), PM2.5 (39%) and CO (30.35%). The reduction of 54%, 49%, 43%, 37% and 31% in National AQI of Central, Eastern, Southern, Western and Northern part of Delhi had been observed [7-10].

Apart from Delhi, in the other cities and outskirts having high pollution concentration (Gaziabad, Noida, Jind, Faridabad, Muzaffarnagar, Gurgaon, Bulandshahar, Panchgaon and Bhiwadi) the AQI was reduced during the lockdown specially in phase 1 and 2 [7,10]. The satellite observations also depict the 46% reduction in average NO2 and 27% improvement in AQI in eight cities (Delhi, Ahmedabad, Kolkata, Chennai, Bengaluru, Mumbai, Hyderabad and Pune) [11].

The evidence from previous studies also shows that the quarrying activities of stones in stone quarrying areas are also stopped due to Lockdown which led to the reduction in Dust Aerosols (PM10) in Dwarka River Basin of Jharkhand and West Bengal [12]. Apart from this reduction, due to crop-waste burning, an increase in concentration of SO2 was also noticed. Also, due to convective dust plumes, an increase in PM10 concentrations were also observed. In this paper, we report the results obtained from a special observational program of primary and secondary pollutant mass concentration and their association with crop-waste burning (anthropogenic) and dust plume emissions (natural) that occurred during the pristine environmental conditions such as the COVID-19 lockdown period.

The PM mass concentration and SO2 concentration data were collected by utilizing the Air Quality Monitoring Station (AQMS) at AUH, and also from National Data Centers being maintained by a Continuous Ambient Air Quality Monitoring Station which is installed in Campus of Amity University Haryana in collaboration of SAFAR/MAPAN project by Indian Institute of Tropical Meteorology, Pune. The processed data of Fire-Points for India was taken from the Processed Imageries of MODIS by National Centre for Medium Range Weather Forecasting (NCMRWF), New Delhi.

The results of the above observations are confirmed from the DREAM Dust Model which is developed by Barcelona Super Computing Center, Spain. The trajectories and dispersion results are generated by Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) Model developed by the National Ocean and Atmospheric Administration’s (NOAA) Air Resource Laboratory (ARL), Boulder, Colorado, USA.

The Crop-Waste Burning

In the early stages of the experimental period (May of 2020), we observed a significant increase in SO2 concentration from morning till noon time over the experimental location i.e., Amity University Haryana (AUH) from the analysis of Air Quality Monitoring Station (AQMS) data. It was quite surprising to notice a higher concentration of SO2 when the region was completely kept under lockdown by the Government.  The period of increase in concentration of SO2 i.e., morning till noon gave a hint that probably the peaks are due to crop-waste (stubble) burning since SO2 is one of the major pollutants that emit from crop-waste burning [13]. Moreover, the time of increase in this concentration coincides with the time of crop-waste burning in the region of Haryana State. On 2nd and 3rd May, the SO2 concentrations peaked up to 9 to 10 µg/m3 in mainly morning and evening hours of the day, whereas in other times, it varied from around 5-7 µg/m3 (Figures 1 & 2).

Figure1: Variations observed in high resolution SO2 data on 2 May 2020.

Figure2: Variations observed in high resolution SO₂ data on 3 May 2020.

Also, the variations observed in SO₂ mass concentration for three consecutive days (9, 10 and 11 May 2020) clearly depict that the concentration was high during morning hours due to burning activity. The bare minimum concentration observed around 2 PM on all the days is ascribed to be due to break in the activity up to almost 3 PM and thereafter, the activity revived and increased gradually. On average, the concentration was found to be higher on 9 and 11 May 2020, as seen in (Figure 3). On 17 May, the concentration (Fig. 2) started rising at 06:00 AM from 7µg/m³ and reaches its peak at around 9:30 AM to 17µg/m³ and returns to 1-5µg/m³ by 2:00 PM. Similar phenomenal variation was seen on 19 May, but the peak is much higher than that observed on 17.

Figure 3: Variations observed in high resolution SO₂ data on 9, 10, & 11 May 2020.

May as from 5 µg/m³ to 45µg/m³ and lowered down to 1-5µg/m³ (Figure 4). The concentrations on 2 & 3 May are found to be low, which may be a sudden incident that took place far from the observation site as compared to 9, 10, 11, 17 & 19 May 2020 observations.

Figure 4: Variations observed in high resolution SO₂ data on 17 & 19 May 2020.

One more significant event was spotted on 19 May around 08:45 h with SO₂ concentration up to about 45 microgram per cubic meter during crop waste burning is NOx which didn’t show significant variations might be due to more Molecular Mass than SO2 since it is one of the major factors in the dispersion of pollutants and compound in the atmosphere [14]. The burning events were confirmed by the NASA-MODIS fire counts for India. The imagery clearly shows that during late April and early May, the intensity and no. of fire counts are very low in the region of Haryana and with date its started increasing around the observation site.

The imageries of 23 April 2020 taken for the previous 7 days, show the very less near the observation site. On 28 April, previous 7 days imageries show the increase in the region of Haryana, rural areas of Delhi and Western Uttar Pradesh. Whereas in Western Uttar Pradesh and Madhya Pradesh the burning incidents can be seen from Mid-April to early May since the crops are harvested earlier in MP as compared to Haryana and Punjab (Figure 5).

Figure 5: Fire Point Counts on 23 & 28 April 2020 for last 7 days.

Modis Satellite Retrievals

The 24 h mean MODIS fire point count imageries on 2, 3, 9, 10, 11 May 2020 clearly show high density intensity of burning and no. of fire point counts which provide explanation and observation of observed SO₂ mass concentration variations of the respective dates (Figure 6). Apart from this, from the imagery of 12 May 2020 it is quite clear that the intensity and No. of fire count were decreased in region of Western Uttar Pradesh and other agricultural region of the country for example Madhya Pradesh and concentrated in the region of Punjab and Haryana, and specially around the observational site.  This explains the high variations and peaks in the SO₂ concentrations. This is also due to the harvesting season difference between different regions of India (Fig. 6). The peaks on 17 & 19 May were also verified by the imagery 2020 it is quite clear that the intensity and No. of fire count were decreased in region of Western Uttar Pradesh and other agricultural region of the country for example Madhya Pradesh and concentrated in the region of Punjab and Haryana and specially around the observational site, this explains the high variations and peaks in the SO2 concentrations. This is also due to the harvesting season difference between different regions of India (Figure 6). The peaks of 17 & 19 May were also verified by the imagery but unfortunately the imagery data was not available to us during the research.

Figure 6: Fire Point Counts on 2, 3, 9, 10, 11 & 12 May 2020 for last 24 Hrs.

Apart from this, the imagery for 12 May 2020 is quite clear that the intensity and number of fire count decreased in the region of Western Uttar Pradesh and other agricultural region of the country. For example, Madhya Pradesh and concentrated in the region of Punjab and Haryana and specially around the observational site. This explains the greater variations and peaks in the SO₂ concentration. This could also be partly due to the activities relating to the harvesting season between different regions of India (Figure 6). The peaks observed on 17 & 19 May may also verify by the imagery but unfortunately the imagery data was not available to us during the research.

Air Mass Trajectory Analysis

The HYSPLIT Air Mass Trajectories also verifies somehow the crop burning incidents. The 5 Hrs. backward trajectories of 10 May 2020 at 1500m AGL, 1200m AGL and 500 m AGL shows that the air mass reaching the observational site from the nearby agricultural areas. Similarly, the 6hrs back trajectories of 19 May 2020 at the same heights as on 10 May 2020 also shows the air mass reaching to the observational site is from agricultural areas (figure 7).

Figure 7: HYSPLIT air mass back-trajectories of 10 & 19 May 2020.

The Dust Plume Findings

Another event which increases the parameter of air quality was the dust plume which hit the observational site on 15 May 2020 at about 6 PM evening and continued up to morning of 16 May 2020. The time variation of PM₁, PM_2.5 and PM₁₀ mass concentration observed on 15 & 16 May 2020 are plotted in Fig. 8. The plots clearly show a significant increase in PM₁, PM_2.5 and PM₁₀ at 6 PM on 15 May and the peak concentrations continued till next day morning (Figure 8). The wind speed was also observed to be low during the period of the dust event. On 15 May 2020 the PM concentrations started increasing while the wind speed started decreasing at the same time became calm. The same low value continued till the next morning on 16 May 2020. The higher values of PM started decreasing around 9 AM and the wind speed started increasing at the same time (Figure 9). This evidences that the observed dust event was a dust plume (having increase in dust parameter and low wind speed) [15].

Figure 8: Variations in high resolution PM data observed on 15 & 16 May 2020.

Figure 9: Variations in high resolution Wind Speed data on 15 & 16 May 2020.

Dust Model Results

The DREAM (Dust Regonal Atmospheric Model) model forecast results of dust load of 16 May 2020 for 24 h and 30 h are depicted in (Figure 10). The results show that the whole Indian subcontinent was covered with the plume of dust, which verifies our analysis PM Concentrations and wind speed (Fig. 10). The HYSPLIT Air Mass trajectories of 15 May with starting time 10 PM and total run time of 6 h and 16 May with starting time 12 PM with total model run time of 12 & 24 h at the height of 500, 1200, & 1500 m AGL were also taken into consideration to assess the long- and short-range forecast of dust aerosols to the observational site.

The trajectories show that the air masses at different heights reaching observational sites are majorly coming from the western side and from the state of Rajasthan (near Thar Desert). The trajectories also show that the long-range transport of aerosols also originate from Pakistan on 24 Hr. back from 12 PM on 16 May 2020 (Fig. 11). The transport of dust plume can also be seen in DREAM Dust Model results which clearly coincide with the HYSPLIT Trajectories.

Figure 10: DREAM Dust Model Forecast result for Dust Load over Indian Sub-continent on 16 May 2020.

The HYSPLIT air mass trajectories of 15 May with starting time 10 PM and total run time of 6 h and 16 May with starting time 12 PM with total model run time of 12 & 24 h at the height of 500, 1200, & 1500 m AGL were also taken into consideration to assess the long and short range of dust aerosols to the observational site.

The trajectories show that the air masses at different heights reaching to observational sites are majorly coming from the western side and from the state of Rajasthan (near Thar Desert). The trajectories also shows that the long-range transport also originates from Pakistan on 24 Hr. back from 12 PM on 16 May 2020 (Figure 11). The transport of dust plume can also be seen in DREAM.

Figure 11: HYSPLIT AIR Mass Back-Trajectories of 15 &. 16 May 2020.

Dust Model results which clearly coincide with the HYSPLIT Trajectories. The results collectively show that the dust event was a short-lived dust plume but not a storm since locally originated due to the convective activity in low wind scenario. The dust may have originated from Sahara but at an instant the phenomenon is totally local.

The authors are thankful to the Director, Indian Institute of Tropical Meteorology (IITM-Pune), MoES, GoI for all types of data support and experimental facilities at Amity University Haryana. The authors appreciate the data analysis help rendered by M. Sateesh of National Center for Medium Range Weather Forecasting (NCMRWF), New Delhi, India. Thanks, are also due to the National Aeronautics and Space Administration (NASA) for MODIS fire count imageries, Barcelona Supercomputing Center for DREAM-Dust forecast results, and National Oceanic and Atmospheric Administration-Environmental Research Laboratory (NOAA-ERL) for HYSPLIT Model Data.

The corresponding author (Panuganti C. S. Devara) is responsible for conceptualization, methodology, project administration and supervision, and original manuscript preparation.  The principal author (Shubhansh Tiwari) has contributed to the data curation and analysis, draft-writing, and plotting. Pritam B. Sharma is responsible for the visualization of the paper. Gufran Beig has contributed to the formal analysis of data and review of the manuscript. B.S. Murthy has contributed to the data and validation.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The authors declare that they have no conflict of interest.

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

1. WHO COVID-19 Situation Report Number 51. 2020.
2. Wang, Q, Su M. A preliminary assessment of the impact of COVID-19 on environment – A case study of China. Science of the Total Environment. 2020.
3. Helm D. The Environmental impacts of the coronavirus. Environ resour econ (dordr). 2020; 1-18.
4. Rugani B & Dario C. Impact of COVID-19 outbreak measures of lockdown on the Italian Carbon Footprint. Science of the Total Environment, 2020.
5. Lokhandwala S. & Gautam P. Indirect impact of COVID-19 on environment: A brief study in Indian context. Environmental Research. 2020.
6. Srivastava, S., Kumar, A Bauddh, K. et al., 21-Day Lockdown in India Dramatically Reduced Air Pollution Indices in Lucknow and New Delhi, India. Bulletin of environmental contamination and toxicology. 2020; 105: 9-17.
7. Mahato S, Pal S & Ghosh KG. Effect of lockdown amid COVID-19 pandemic on air quality of the megacity Delhi, India. Science of the Total Environment. 2020.
8. Devara PCS, Kumar A, Sharma PB, Banerjee P, Khan AA, Tripathi A, Tiwari S & Beig G. Influence of Air Pollution on Coronavirus (COVID-19): Some Evidences from Studies at AUH, Gurugram, India. Open Research Communication in the Social Science Research Network (SSRN), USA. 2020.
9. Srivastava AK, Bhoyar PD, Kanawade VP, Devara PCS, Thomas A and Soni VK. Improved air quality during COVID-19 at an urban Megacity over the Indo-Gangetic Basin: From stringent to relaxed lockdown phases, Urban Climate. 2021; 36.
10. Sonbawne SM, Fadnavis S, Vijayakumar K, Devara PCS and Chavan P. Phase-Resolved Lockdown Features of Pollution Parameters Over an Urban and Adjoining Rural Region During COVID-19. Front. Environ. Sci. 2022.
11. Siddiqui A, Halder S, Chauhan P et al. COVID-19 Pandemic and City-Level Nitrogen Dioxide (NO2) Reduction for Urban Centres of India. J Indian Soc Remote Sens. 2020.
12. Mandal I & Pal S. COVID-19 pandemic persuaded lockdown effects on environment over stone quarrying and crushing areas. Science of the Total Environment. 2020.
13. Jain N, Bhatia A & Pathak H. Emission of air pollutants from crop residue burning in India. Aerosol and Air Quality Research. 2014; 14: 422-430.
14. Bolin B, Aspling G & Persson C. Residence time of atmospheric pollutants as dependent on source characteristics, atmospheric diffusion processes and sink mechanisms. Tellus. 1974; 26: 185-195.
15. Jquline S. Facts and File Dictionary of Weather and Climate. 2001.