Introduction

Land use changes, anthropogenic emissions from transport and industries and climate variability deeply affect the environmental quality globally (Stavrakou et al. 2019). In the context of the worldwide lockdown enforced due to the COVID-19 pandemic, there is an increased interest in studying changing air quality through satellite remote sensing (Collivignarelli et al. 2020; Dantas et al. 2020; Li et al. 2020; Nakada and Urban 2020; Sharma et al. 2020). Lockdowns were imposed in several countries to impede the progress of the spread of the coronavirus pandemic within the communities. This resulted in cleaner air quality over the US, China, India and Europe. The concentration of pollutants markedly decreased over urban areas, with varying differences among primary and secondary pollutants (Bao and Zhang 2020; Huang et al. 2020; Liu et al. 2020; He et al. 2020). This was mainly achieved due to decrease in traffic and industrial activities. Satellite remote sensing is now using multi-platform system and new algorithm in the processing chain that enables better resolution of emission inventories. Significant decrease in NO2, CO2, SO2, CH4 and aerosols levels have been observed globally during the lockdown period. Real-time observations of background air quality precisely measure ‘hotspots of air pollution’ through the Copernicus Sentinel-5 Precursor Tropospheric Monitoring Instrument (S5p/TROPOMI). A 14-days average is a better depiction of the emission profile over a region, due to pollutants being affected by the short-term weather changes and cloud cover.

Various satellite images, from NASA, Copernicus Sentinel-5P satellite, from the European Union Copernicus Programme, showed many cities throughout the globe witnessing a significant drop in air pollutants ranging between 30 and 60% owing to countrywide lockdowns. Satellite remote sensing is now using new algorithm in the processing chain as well as multi-platform system that enables emission inventories over a particular region with greater accuracy and speed. The most recent ESA supported project ICOVAC (Impacts of COVID-19 lockdown measures on Air quality and Climate) also helped in assessing the effects of the imposed COVID-19 lockdown and in evaluating the air quality impact from other factors.

Most important real-time observations of air quality, particularly background NO2 data was precisely measured by the Copernicus Sentinel-5 Precursor Tropospheric Monitoring Instrument (S5p/TROPOMI) which has been developed by the European Space Agency (ESA). Various studies have extensively used the output to assess tropospheric NO2 levels over polluted areas, both before the lockdown and during the lockdown (Haddout and Priya 2020; Nakada and Urban 2020; Tobías et al. 2020). TROPOMI instrument provides extent of the backscattered earthshine radiation in the UV, visible, near and short-wave IR spectral ranges at the exceptional spatial resolution of 3.5 × 7 km2 (Muller et al. 2019; Veefkind et al. 2012). The resolution further improved to 3.5 × 5.5 km2 from August 2019 onwards (Van Roozendael et al. 2019) which helped in more precise mapping of air quality. A 14-days average is taken as a better representation of the emission profile over a region, as the concentration of the pollutants are affected by the short-term weather changes and cloud cover. Air quality forecasts by NASA use near real-time (NRT) data from NASA’s ‘Land, Atmosphere Near real-time Capacity for EOS’ (LANCE) for monitoring a wide range of natural and anthropogenic pressure functions (NASA 2020a). The Table 1 below summarizes the satellites used by NASA for monitoring of different air quality parameters.

Table 1 Description of various platforms used by NASA for assessing different air quality parameters;
Full size table

CH4 emissions during COVID-19

In the atmosphere, CO2 is more abundant and more commonly associated with increase in temperature, and considered key contributing factor that causes global warming (Florides and Christodoulides 2009; Skytt et al. 2020). On the other hand, CH4 is approximately 30 times more powerful as a heat-trapping gas and plays a major role in climate radiative forcing. Methane has both natural and anthropogenic sources of emission. Natural sources include emission from agricultural lands, wetlands, marshy areas and enteric fermentation in ruminant animals. Anthropogenic sources include emission from fossil fuel extraction, landfill and various other anthropogenic activities (Janssens-Maenhout et al. 2019; Heimann et al. 2020). Most of the CH4 emissions, almost to the tune of 60%, happen in the northern hemisphere, however the extended lifetime to inter-hemispheric transport lessens the north to south (N:S) latitudinal gradient of CH4 upto 100 ppb (Heimann et al. 2020).

The Copernicus Sentinel-5P satellite, with its instrument TROPOMI, mapped CH4 emissions during the lockdown period. CH4 enhancements through oil and natural gas producing regions within the US were monitored through TROPOMI (de Gouw et al. 2020). Measurements of CH4 have been found to be in good agreement with those observed from the GOSAT instrument (Hu et al. 2018). Given its importance in the global warming, Canadian company GHGSat used mosaics of multiple satellite images (Varon et al. 2019) in association with Sentinel-5P team at SRON (Netherlands Institute for Space Research) to identify the hotspots and emission inventory of CH4 throughout the lockdown.

The concentration of CH4 is reported high in some regions in year 2020 as compared to year 2018, for example over a coal mine in the Shanxi province as shown in Fig. 1.

Fig. 1
figure 1

(Source: ESA 2020)

TROPOMI CH4 measurements over a coal mine in the Shanxi province, China

Full size image

The satellite images from NASA Giovanni portal (http://giovanni.gsfc.nasa.gov/) showed an increase in CH4 concentration globally as well as in Indian subcontinent and South America during March to May 2020 (lockdown period) as compared to the March–May 2019 period as shown in Fig. 2. According to Paris-based data firm Karryos, CH4 leaks rose 32% in the first eight months of 2020 globally. The increase in the concentration of CH4 is prominently led by drops in carbon emissions due to worldwide lockdowns. It is also due to the reduction in carbon emissions from oil and gas sectors. Karryos, that examined satellite data to estimate the extent of CH4 emissions, elaborated that around 100 CH4 leak episodes happened throughout the world. In some CH4 hotspots in Russia, Algeria and Turkmenistan, the concentration of CH4 rose upto 40% as compared to previous year for the same time period. In addition to those three countries, the US, Iran and Iraq were the three largest CH4 emitters in the first two-thirds of 2020. In Iraq, Karryos reported largest leak, that accounts 400 tons of CH4 per hour. The leak prolonged upto 150 miles into Saudi Arabia. However, in the US, the largest leak accounts 150 tons of CH4 per hour, that is equal to 10 coal-fired plants working at maximum capacity resulting in increase in methane emission globally and in various subcontinents during the lockdown period (Ecowatch 2020).

Fig. 2
figure 2

(Source: Giovanni online data system, established and maintained by the NASA GES DISC)

Time average map of methane, mole fraction in air (Daytime/Ascending, AIRS-only) in ppbv over different parts of the world—the left-side images are from March, 2019 to May, 2019 while the right-side images are from March, 2020 to May, 2020

Full size image

Changes in NO2 levels

Nitrogen dioxide is a short-lived pollutant, and its levels in the atmosphere can indicate level of economic activities, comparable to past emissions scenario and economic slowdowns. Changes in NO2 levels are responsive to changes in meteorological conditions. There are many factors, such as sunshine hours, changes in daily wind speed and temperature, that affect the residence time and the dispersal of gas in the atmosphere. In the global NO2 emission basket, fossil fuel combustion contributes 65% and the remaining 35% is contributed by forest fires, lightning and soils (Smith 2020). It is also true that various structural measures and emission regulations helped in improving the air quality, particularly the long-term declines in NO2 concentrations. Oxides of nitrogen are precursors of secondary inorganic aerosols formation that has major implications for climate and human health.

A collaborative study done by atmospheric experts from the European Space Agency (ESA), Royal Netherlands Meteorological Institute (KNMI) and Royal Belgian Institute for Space Aeronomy (BIRA-IASB), determined how COVID-19 pandemic had an impact on NO2 pollution (Bauwens et al. 2020). A sharp decrease in NO2 levels, by about 40–55% was observed in many cities in India and China (Figs. 3 and 4). An unexpected decrease in NO2 pollution in the range of 50–70% was observed in many cities of China due to strict lockdowns enforced in February 2020 (Fig. 5). There was a significant decrease in NO2 pollution in Western Europe and the United States in the range of 20–38%. Cities in Europe such as Madrid, Barcelona and Milan observed a decline of NO2 pollution in the range of 30–40% as compared to 2019 for the same period (Bauwens et al. 2020). During the lockdown, a reduction of 30% in the NO2 column was observed over North America and Europe in spring season. The column reduced by almost 50% in various parts of the Asia over the same period (Berman and Ebisu 2020). Furthermore, levels of particulate matter having size smaller than 2.5 microns were also reduced upto 35% over northern China (Smith 2020).

Fig. 3
figure 3

(Source: ESA 2020, sentinel online)

Concentration of NO2 before lockdown (left) and during the lockdown (right) over India as measured by Sentinel-5P

Full size image
Fig. 4
figure 4

(Source: Le et al. 2020)

NO2 level averaged over the COVID-19 lockdown period for three weeks during Jan. 23 to Feb. 13, 2020 (left) as compared to similar timeframe in 2019, the circle in the map indicates Wuhan city, China

Full size image
Fig. 5
figure 5

Copernicus Sentinel-5P satellite data exhibiting sharp decreases in NO2 levels over China and India during 2020 as compared to 2019 (left)

Full size image

Based on 800 local ground-level monitoring stations in China, huge fall in levels of NO2 was observed which further led to a concurrent increase in concentration of other secondary pollutants like ozone (Shi and Brasseur 2020). Ozone is a known secondary pollutant which is formed in the troposphere when favorable conditions like intense sunlight and high temperatures catalyze chemical reactions. It was found that the concentration of ozone enhanced by a factor of 2.0 ± 0.2 with mean reduction in PM2.5 by 30% and NO2 levels by 40–60% in the Wuhan city during the lockdown period. Ozone is destroyed by nitrogen oxides, therefore, when nitrogen dioxide levels decreased, the ground-level ozone increased. This causes formation of nighttime NO3 radical with increasing oxidizing ability leading to the development of secondary particulate matter (Huang et al. 2020).

In Barcelona, Spain, a momentous drop was observed for black carbon and NO2 levels (− 45 to − 51%), which are primarily caused by traffic emissions (Fig. 6). The PM10 levels declined by 28–31%. However, as expected with lower NO2 levels and VOC-limited environment, O3 levels increased by 33–57% (Tobías et al. 2020) (Fig. 7).

Fig. 6
figure 6

(Source: ESA)

Copernicus Sentinel-5P satellite data showing major drop in NO2 levels over various cities of Europe—including Madrid, Rome and Paris

Full size image
Fig. 7
figure 7

(Source: Tobías et al. 2020)

Average concentration of ambient tropospheric NO2 as quantified by TROPOMI-ESA in the Iberian Peninsula during comparable time periods in 2019 and 2020

Full size image

Reduction in tropospheric NO2 concentration was reported over Southwest USA based on satellite images released by NASA for the period of March 25–April 25, 2015–2019 (average) as compared with average of March 25–April 25, 2020 (Fig. 8). San Francisco reported 22%, Los Angeles 31%, San Diego 25%, Phoenix 16% and Las Vegas 10% decrease in NO2 concentration due to lockdown. Similar decrease in tropospheric NO2 concentration was reported in California as show in the Fig. 7 (Fig. 9).

Fig. 8
figure 8

(Source: NASA 2020b)

Tropospheric NO2 column over Southwest USA during March 25–April 25, 2015–2019 (averaged) compared with March 25–April 25, 2020 (averaged)

Full size image
Fig. 9
figure 9

(Source: NASA 2020b)

Decrease in tropospheric NO2 concentration during COVID 19 lockdown in California

Full size image

Satellite images released by NASA showed a major decrease in tropospheric NO2 concentration in Florida, Northeast USA, Southwest USA and Indian subcontinent during lockdown in 2020 in wake of checking COVID-19 pandemic (Fig. 10). In Indian Subcontinent, when averaged tropospheric NO2 concentration of March 25–April 25, 2020 period was compared with 3-years average of March 25–April 25, 2017–2019 period, it showed a decrease in NO2 concentration in Delhi by 56%, Kolkata by 23%, Mumbai by 38%, Lahore by 45%, Dhaka by 45% and Karachi by 36% (Fig. 10d).

Fig. 10
figure 10figure 10

(Source: NASA 2020c)

a–d Satellite images showing decrease in tropospheric NO2 concentration in Florida, Northeast USA, Southwest USA and Indian subcontinent during lockdown in 2020

Full size image

NOx, which is the sum total of NO (nitrogen oxide) and NO2, along with VOCs (volatile organic compounds) are the main precursors of photochemical production of ground-level tropospheric ozone (O3). The high concentration of ozone is toxic to all forms of life—including crops and human beings. The rate of ozone formation within the troposphere varied depending upon height with maximum at ground level In the Northern Hemisphere, the production of O3 is low in winter season at mid- and high-latitudes due to reduction in sunshine hours and temperature. The production of O3 increases as spring and summer advances. Even though O3 does not decrease by the same proportion as NOx, a reduction in ground level O3 concentrations is expected with reducing NOx levels. However, evidence of increased tropospheric ozone levels has been reported during lockdown at several places (Korhale et al. 2020). The increase in concentration of ozone is due to the lower titration of ozone through NO because of the sharp decrease in the concentration of local NOx by road transportation. In an unpolluted region, the reaction of VOCs with OH– radicals dominate when the ratio of VOCs (including CO) to NOx is higher, consequently enhancing ozone production. Overall, the major effect of complete lockdown events on the concentrations of NOx, PM and O3 came from the major reduction in surface transportation. Under VOC-limited conditions and during the lockdown the main cause of the greater O3 concentrations in cities is due to the drop in NOx emissions to a lower ozone titration by NO. Secondly, as PM emissions were lesser, the higher solar radiation preferred production of ozone and caused rise of ozone precursors releases (Sicard et al. 2020). This is also due to the complex variations in emissions profiles during the lockdown and indicates how mass reductions of distinct contaminants can cause a rise in others, further triggering variations in wider tropospheric composition and reactivity. During the lockdowns, a drop of approximately 70% of the total NOx releases was observed, primarily due to the major drop in vehicular density on the roads. At the time of lockdown, with less NO in the ambient air to sustain the photo-stationary responses that terminate ozone molecules, the overall ozone formation went up. This shows that reducing the development of secondary contaminants such as ozone is challenging and requires strict measures to control it.

Global CO2 emission reduction during the lockdown

Several countries throughout the world experienced significant drop in CO2 concentration of more than 40% than the previous year due to lockdown restrictions during 2020. As a result, the air quality improved greatly and the risk of several respiratory diseases, for example bronchitis, asthma and other lung diseases also decreased markedly (Watts 2020). In China, which is accountable for the world’s highest carbon emissions, a drop of around 250 Mt of carbon sources was observed, which is more than half of the annual output for the whole UK. Correspondingly, in Europe, the reduction of carbon sources of around 390 Mt as a result of lockdown was reported. In USA, CO2 emissions reduced to ~ 40% due to reduction in commuter vehicles, which is one of the major sources of CO2 emissions. A decrease of ~ 44% car sells and fossil fuel consumption in London also resulted in decreased emissions (Paital 2020). A different study by Sharma et al. (2020) in 22 cities of various regions of India for the period of 16 March–14 April 2020 described substantial decrease in AQI, which were up to 29% (East), 32% (West), 15% (Central), 44% (North) and 33% (South) as compared to several preceding years for the same monitoring period. Dutheil et al. (2020) obtained data from the TROPOMI which reported 6% reduction in CO2 concentration worldwide with 25% drop in China alone. They also clarified that mortalities might have also reduced due to decline in air pollution. Paital (2020) evaluated the influence of air pollutants such as CO2, NO2, PM and weather variables globally on the infection and rate of thinning out of corona virus. Air pollution was associated to an elevated risk of corona virus infection and consequently, early and complete lockdown (particularly in China and India) led to noteworthy drop in the concentration of CO2 as explained in above studies. However, such deviations in CO2 concentrations were temporary and are rising back after lockdown has been lifted (Watts 2020). Satellite images from NASA exhibited 25% reduction in carbon emissions in China approximating to 6% of the global emissions in Feb 2020 majorly due to quarantine imposed reduction in transportation activities (Isaifan 2020).

Changes in SO2 levels

There is a rapid increase in the emissions of SO2 in India during the past one decade due to increase in traffic network, coal-based power plants and biomass burning. The higher levels of emissions intensify the problem of haze over the several parts of the nation, especially during winter season when the air is relatively stable. However, their concentration dropped drastically due to the extensive restriction on social, commercial and industrial activities since the lockdown imposed on 25 March 2020. The rapid decline was primary due to restricted traffic movements and industrial activities worldwide. It is observed that concentrations of SO2 in polluted regions of India have dropped upto 40% during the second phase of the lockdown between April 2019 and April 2020. Using remote sensing data from the Copernicus Sentinel-5P satellite, from the European Union Copernicus Programme, a large reduction in concentrations of SO2 throughout the country is observed during the lockdown (Fig. 11).

Fig. 11
figure 11

(Source: Sentinel-5P satellite European Union Copernicus Programme)

Averaged SO2 concentrations in April 2019, compared to April 2020 and location of power plants in India, 1 Dobson Unit (DU) = 0.4462 mmol m−2.

Full size image

The darker shades of purple and red color represent the higher concentrations of SO2 in the atmosphere. The black dots in the picture indicate the sites of large coal-fired thermal power plants where SO2 emissions are relatively high. The darker emission zones are represented by three states namely Odisha, Jharkhand, and Chhattisgarh which have large number of coal-fired thermal power plants. NASA satellite images of Indian subcontinent for the years 2017–2019 compared with March 25–April 25, 2020 in three highlighted Area 1, Area 2 and Area 3 show decrease in SO2 emissions—concentration of SO2 in Area 2 decreased by 25%, and by 12% in Area 3 of power generation whereas Area 1 showed increase in SO2 concentration by 83% as shown in the Fig. 12.

Fig. 12
figure 12

(Source: NASA 2020c)

Tropospheric SO2 column for the averaged years during 2017–2019 (left) as compared with time period during 25 March–25 April, 2020 in the Indian subcontinent

Full size image

Measurement of reduction in aerosols through MODIS

Aerosols over Indian region were lowest in the past two decades during the lockdown period (NASA 2020a). NASA’s Terra and Aqua satellites (Sentinel-5P and AURA) use Moderate Resolution Imaging Spectroradiometer (MODIS) to measure both size distribution and optical depth (AOD) of ambient aerosol globally every hour. A marked reduction in the level of AOD was observed within few weeks of reduced anthropogenic activities. The images of Indian sub-continent taken during March 31–April 5 for 2016–2020 from NASA Earth Observatory exhibited depleting AOD extents over Indian region. In addition, the AOD average in 2020 (lockdown period) is much lower than the average during 2016–2019 (Fig. 13). It is clear from the images that light yellow pixels, tan pixels and dark brown pixels and illustrate the small to negligible, lower to moderate and high aerosol concentrations respectively (Gautam 2020). Reliability and accuracy level of various satellites used for monitoring air pollutants in the atmosphere is provided in the Table 2.

Fig. 13
figure 13

Aerosol optical thickness (Terra/Modis) over Indian sub-continent during March 31–April 5 in 2016, 2017, 2018, 2019 and 2020 (NASA 2020b)

Full size image
Table 2 Reliability and accuracy level of various satellites used for monitoring air pollutants in the atmosphere (Wang et al. 2020)
Full size table

Learning from the pandemic: stronger policy regime needed to control emissions

Despite global pledges to reduce the emissions, the air quality is witnessing significant peaks in stark contradiction to the worldwide efforts in reducing GHGs. The lockdown period also indicated failure to meet many binding targets of emission reductions by large emitters such as China and the US. The short-term emissions reductions indicated the need for several policy level challenges along with multiple forms of international cooperation and deeper engagements (Fig. 14). The pandemic and deteriorating air quality standards both have potentially distressing global implications which need quicker response framework at scale for any future crises (Klenert et al. 2020). There is also a need for rapid remediating interventions from both public and private sectors (Goulder 2020). The slower response in phasing out polluting industries and technologies locks local and regional economies into carbon-intensive futures that further reduce flexibility in developing future mitigation strategies.

Fig. 14
figure 14

Policy strategies needed to control emissions

Full size image

The lockdown imposed only a limited window to reduce the emissions not because of the structural measures, but due to a sudden and abrupt closure of the pollution sources. The global methane emissions maintained a steeply increasing trend, indicating the dominance of oil and gas industry and lower impacts of policies and regulations. The response of the industry has also been slow in cleaner technology investment.

Conclusion and future directions

The various satellite observations unequivocally show that Northern China, Western Europe, Indian-subcontinent and the US witnessed a substantial decline in air pollution during the lockdown period, as compared to the same time last year. It is evident that the sharp decreases in air pollution are mainly credited to the lockdown measures taken to contain the spread of the Covid‐19 pandemic. The period of lockdown witnessed sharp reductions in surface and air traffic movements and industrial activities. As several countries around the globe enforced lockdowns, a broad spectrum of indicators showed massive decline in electricity demand, transport use and industrial activities with shift in working hours and reduction in mobility and working patterns. However, such sudden reductions were not due to structural changes or quality enforcements, but due to restraining of the population’s mobility and shutting down of a large number of commercial and industrial units. The levels are again going up as the lockdown phase eased off. In most parts of the Europe, a similar decline in emissions, though limited to 20% was observed during the economic slowdown caused by the global recession (Castellanos and Boersma 2012). The available data shows that it is difficult to attribute the decline in emissions solely to the pandemic, given that there are numerous reasons of falling demand of fossil-fuel during March 2020, relative to the same month in preceding years. The lockdown period provided a unique opportunity to study the ambient environment and assess the effects of the cutback of various emission sources and to evaluate introduction of stricter air quality standards and regulatory policies. This requires for a global effort to frame more stringent regulatory standards and climate policies so that drastic improvements in air quality could be achieved. For accurate and early warning of poor air quality, satellite monitoring over a large area can play a significant role in emission surveillance, from the point of stricter regulation and structured control management. There is also a need of refining the modeling techniques for better prediction of air quality (Dentener et al. 2020). For strengthened scientific strategies to control emissions, further analysis is urgently needed over longer time periods to substantiate the changing emission profile as observed in column changes observed from various satellites. This information is crucial because emission scenarios give people and policy makers various options to lessen the risk of exposure of poor air by taking appropriate measures. Various satellite images are available for the users which are easy to visualize and are used for mitigating the effects of bad air quality.