Machine learning-based country-level annual air pollutants exploration using Sentinel-5P and Google Earth Engine

Climatic condition is triggering human health emergencies and earth’s surface changes. Anthropogenic activities, such as built-up expansion, transportation development, industrial works, and some extreme phases, are the main reason for climate change and global warming. Air pollutants are increased gradually due to anthropogenic activities and triggering the earth’s health. Nitrogen Dioxide (NO2), Carbon Monoxide (CO), and Aerosol Optical Depth (AOD) are truthfully important for air quality measurement because those air pollutants are more harmful to the environment and human’s health. Earth observational Sentinel-5P is applied for monitoring the air pollutant and chemical conditions in the atmosphere from 2018 to 2021. The cloud computing-based Google Earth Engine (GEE) platform is applied for monitoring those air pollutants and chemical components in the atmosphere. The NO2 variation indicates high during the time because of the anthropogenic activities. Carbon Monoxide (CO) is also located high between two 1-month different maps. The 2020 and 2021 results indicate AQI change is high where 2018 and 2019 indicates low AQI throughout the year. The Kolkata have seven AQI monitoring station where high nitrogen dioxide recorded 102 (2018), 48 (2019), 26 (2020) and 98 (2021), where Delhi AQI stations recorded 99 (2018), 49 (2019), 37 (2020), and 107 (2021). Delhi, Kolkata, Mumbai, Pune, and Chennai recorded huge fluctuations of air pollutants during the study periods, where ~ 50–60% NO2 was recorded as high in the recent time. The AOD was noticed high in Uttar Pradesh in 2020. These results indicate that air pollutant investigation is much necessary for future planning and management otherwise; our planet earth is mostly affected by the anthropogenic and climatic conditions where maybe life does not exist.


Literature review and research motivation.
The clean air is furthermost significant for the human life because of oxygen and others essential gasses are generating considerable healthier life. The flora and fauna are similarly affected through several air pollutant, that's why clean air is necessary 23,24 . The air quality maintain is the key focus of the researchers, scientists and policy-makers for sustainable planning and development of the human life in addition to the environment. One study of China indicated that the reduction of human mobility is outlandishly related to the reduction of air pollution of the 44 Chinese cities where Intercity Migration Index (IMI) was used for calculating the results 25 . The research observed that the COVID-19 lockdown also influencing factor for reducing the air pollutants in different portions of the earth surface where megacities are the indicators for human activities and air pollution related scenarios [26][27][28][29][30][31] . One study observed that the mean concentration of the NO 2 level was reduced in the European cities where ~ 53% of the NO 2 level are fluctuated 32 . Malaysia also affected by air pollutant and COVID-19 lockdowns were directly affected the air pollutant like PM 2.5 , NO 2 , SO 2 , and CO concentration 13 .The research studies were carryout six air pollutants in China (120 cities) from January 23, 2020 to February 29, 2020 33 . The air pollution station data can't appropriate for identification of the distribution of country-level air quality, therefore satellite-based remote sensing data were used for monitoring the air pollutants and chemical concentration of the air 8,34 . Air pollutant measurement is more essential for environmental impact assessment; large-scale air pollutant measurement is a vital due to the large area. Therefore, GEE platform is help out for this situation (https:// devel opers. google. com/ earth-engine/ datas ets/ tags/ airquali ty). Numerous algorithms were applied for calculating the air quality of an area. In India, different types of air pollutants are calculated through those notified algorithms. Earth observational Satellite established RS expertise suggestions an operative explanation on behalf of the continuing spatio-temporal observing of the air quality above the numerous measures. Machine learning approaches are widely applied for air pollutant measurement of the earth's surface while some previous research outcomes are mentioned novel approaches for air pollutant measurement 35 .
The satellite imageries are applied for observer the air pollutant from early 1970s through the application of the Geostationary Operational Environmental Satellite (GOES), Advanced Very High Resolution Radiometer (AVHRR), and Landsat. In addition, Meteorological OPerational satellite (MetOP), Aura, and Sentinel-5 precursor (Sentinel-5P) were between the additional satellite-based datasets that have been commonly applied on behalf of air quality observing subsequently 1978. Regarding the literature, numerous investigators have deliberate observing, examining and reclamation of the air pollutants like Aerosol optical depth, SO 2 , NO 2 , CO, PM 2.5 , PM 10 , CH 4 , and O 3 applying these RS-based satellites. During COVID-19, worldwide air pollutants are also observed to monitor the situation before, during and after COVID-19 lockdown. The GEE platform is widely used for calculating the air pollutant using Sentinel-5P and MCD19A2 data during April 2018 and 2021 in major Indian cities 36 . Sentinel-5P also used for non-linear relationship between daily and annual air pollutant like CO, NO 2 , O 3 , and SO 2 37 . In turkey, Sentinel-5P is used for air pollutants measurement and MODIS data is used for AOD variation analysis in GEE cloud computing platform during January 2019 to September 2020 38 . The widely applied MODIS imageries were applied for calculating air pollutant in India from 2018 to 2021, where GEE platform notified formulas were applied. In India many researchers are doing the air quality measurement for environmental impact assessment and human activities analysis 10,39,40 . The air pollution is more harmful for human life because worldwide 5 millions of deaths were recorded due to air pollution and related activities. The air pollution increased temperature variation, respiratory disease, lung cancer, asthma and skin related problem 30,41 . Therefore, the research paper was used GEE cloud computing platform for delineation of the air pollutants and chemical concentration in India during 2018 to 2021. The GEE is used for hassle free large data analysis where data access, pre-processing, processing and management were done in cloud platform 42 . Research objectives. The air pollutant studies are located human intervention and industrials works are more harmful for development of air pollutant in India, where rural areas are located low air pollution. Meg-

Study area
The population pressure, industrial works, and anthropogenic activities like transportation development, industrial power plants, green space dynamics, and unexpected urbanization have been influencing the environment [44][45][46] . India is mostly affected country where, anthropogenic activities and human health-related problems are increase gradually, which is triggering factors for health issues, thermal variation, and air pollutants-related disease like asthma, respiratory disease, lung cancer, and skin-related problem. The most affected air pollutants are NO 2 , CO, O 3 , SO 2 , and CH 4 , which are affected by the air and pollutant in the environment (Fig. 1). India has mostly witnessed air pollutants-related activities in the last decades where COVID-19 times mostly influenced the air pollutants and chemical concentration of the air mass 9,29,36,47,48 . The 2020 and 2021 country-wide lockdown is the main key factor for visualization of the actual anthropogenic activities-based air pollution and related activities 36,47,48 . Because lockdown influences the air pollutant and healthy ecosystem, therefore pre, during, and post COVID-19 air pollutants measurements are the affective investigation for human activities and climate-derived scenarios.  www.nature.com/scientificreports/ Based on the Central Pollution Control Board, the Ministry of Environment, Forests, and Climate Change is monitoring the daily air pollutants over India (https:// app. cpcbc cr. com/ AQI_ India/) where PM 2.5 . PM 10 , NO 2 , NH 3 , SO 2 , CO, and O 3 are measured. The station data and satellite images like MODIS and Sentinel-5P were applied for monitoring the air pollutant and chemical properties in India. The decadal maps like 2018, 2019, 2020, and 2021 datasets were much helpful for understanding the air pollution and anthropogenic activities related information, which is helpful for decision making, planning purpose, future disaster management, and research purpose. The rural areas are used biomass cake, trash, and fuel-wood for cooking purposes whereas 100 million rural households are used stoves and Turbo stoves for daily purposes. Traffic congestion, greenhouse gases emission, burning of paddy residues after harvest are more affected by air pollutants and chemical properties. Some websites are mentioned the air quality of India like India Air Quality Index (AQI) (https:// www. aqi. in/ dashb oard/ india), and World Air Quality (https:// www. iqair. com/ in-en/ india). This research results will be helpful for future research and development where air quality investigation is more useful.

Materials and methodology
Data used. Air quality and chemical concentration analysis are more important for investigating the earth's surface phenomenon and public health emergencies due to anthropogenic activities. The Sentinel-5P is widely used for monitoring the air pollutant and Copernicus data is applied for aerosol optical depth analysis. The GEE cloud computing platform is applied for air quality measurement (https:// devel opers. google. com/ earthengine/ datas ets/ catal og/ ECMWF_ CAMS_ NRT) and Copernicus data is used for real time monitoring the AOD (https:// apps. ecmwf. int/ datas ets/ data/ cams-nreal time). The AOD data was derived from moderate resolution imaging spectrometer (MODIS) in the GEE platform from the year of 2018 to 2021, where pre, during and post COVID-19 air pollutants and AOD were calculated. The air quality monitoring stations data also used for validating the air pollutants and AOD values over the Indian megacities.
Data analysis and measurement. Air quality measurement is more important for the future disaster management and planning purposes because of high industrial works, anthropogenic activities, transportation development, and some unnecessary stages 15,28,34,39 . The earth's surface is mostly influenced by several air pollutants like NO 2 , CO, SO 2 , CH 4 , Aerosol, PM 2.5 , PM 10 , and O 3 15,27,39,49 The Sentinel-5P datasets are used for monitoring the air pollutant basically Nitrogen Dioxide and Carbon Monoxide 48,50 , where AOD is measured using 550 nm Copernicus near real-time monitoring data 13,48 . The code of the GEE is freely available, where the codes are modified based on the criteria and area of interest (AOI) 38,51,52 . European Space Agency (ESA) air quality monitoring data was used for air pollutant analysis (https:// www. senti nel-hub. com/).
In GEE platform Sentinel-5P level 2 (L2) datasets were converted into L3 exhausting the bin_spatial procedure of harpconvert toolbox 53 . The air pollutants like NO 2 and CO were calculated form filtered pixels with Quality Assurance (QA) standards like 75% and 50% for NO 2 and CO respectively 54 . The source of the NO 2 in the air is basically power plants, vehicles, industrial emissions, and anthropogenic activities 15,55 . The NO 2 is more harmful to the people and also increased environmental health emergencies 12 . The GEE is used for monitoring the air pollutant like NO 2 and CO, where ESA near real-time data was used for AOD in India for estimating, and investigating the environmental health during the study periods.

Results
Air quality assessment is a very important phenomenon for investigating the anthropogenic and climatic conditions in the earth's atmosphere. The recent time like 2018-2021 is the most affected anthropogenic activities, which is more increased the air pollution like Nitrogen Dioxide, Carbon Monoxide and chemical concentration like AOD using Sentinel-5P ESA datasets, MODIS and Copernicus data. This study is to identify the air pollutant and atmospheric chemical concentration in the recent year from 2018 to 2021 pre, during and post COVID-19 periods.
Aerosol optical depth measurement. The AOD is calculated using the GEE cloud computing platform, where the chemical composition is measured by earth observational satellite data with the near real-time monitoring system. The AOD map of India indicates the chemical properties are increased due to anthropogenic and industrial works. The block color indicates the low AOD area whereas the purple color indicates the high AOD area.
Chemical properties of the atmosphere is more sensitive for environment in addition to the human life, therefore AOD measurement is more important. The AOD was calculated from MODIS datasets using Google Earth Engine cloud computing platform where different years mean datasets were added and calculated the values of AOD over India. Four different years' data were used for calculating the AOD like 2018, 2019, 2020, and 2021. Figure 2 indicates the AOD maps of different time periods of India, where mostly affected AOD is located in 2018 and 2021. Uttar Pradesh Bihar, West Bengal, Jharkhand and Punjab are mostly affected by chemical concentration where rest of the area are moderate to low AOD located (Fig. 2). The AOD is fluctuated due to COVID-19 country-wide lockdown, restrictions and anthropogenic activities like industrial works, fuel burning, transportation and many others aspect. 2019 is low AOD located where 2020 and 2021 are gradually increasing the air pollutants and AOD in India. The Red color indicates high AOD values where Blue color indicates the low values. During COVID-19 pandemic air quality is improved because of low fuel use, less transportation and industrial works, therefore the AOD for different years are located variation while pre and post COVID time AOD high observed. Those phenomena indicates that the air quality variation must be fluctuated due to the industrial works in India in addition to the globally. AOD is basically measuring the aerosols from urban haze, dust particles, sea salt and some particles, which is increase in top of the atmosphere. During COVID-19 www.nature.com/scientificreports/ lockdown, due to the low industrial works, and less transportation urban haze, dust particles and some particles are reduce and AOD is less observed in urban and surrounding areas in India. AOD calculation indicates those phenomena from 2018 to 2021. The earth phenomenon and environment also suffer from this decision. Only death, tears, and economic losses are the results of air pollution. The human mind is now traveling the epidemic like COVID-19, where this result is a new addition for thinking about the future. Human minds now need some surgeries to protect the human bringing. We all are birth because to help life and protect the unexpected thinks not needs any unexpected situation otherwise human life maybe lost the oxygen in future. Our motherland is now very tired, needs proper planning, management, development, and belief to protect, otherwise, the motherland earth needs hospitalization due to extreme health issues.
Air pollutant analysis. The earth's atmosphere is mostly affected by several activities, which is frequently occurring by humans' activities. Numerous techniques and datasets were applied for calculating and investigating the air quality in addition to the other chemical properties is generated. In this investigation, two frequently affected air pollutants like nitrogen dioxide and carbon monoxide. The air pollutant is gradually increased which shows that the human activities are truly destroying the earth's breathing. Life is doesn't exist without oxygen and breathing, now the earth is helpless and tried due to human activities. The study results indicate the actual condition of the earth's atmosphere. Due to the high industrial works and vehicle use, air pollution is gradually increased in Indian megacities where dust particles, and smoke particles are more effective in human health. COVID-19 lockdown is clam down those phenomena frequently therefore pre, during and post COVID-19 air pollutant, chemical properties of air and air quality measurement is more essential for environmental sustainability. www.nature.com/scientificreports/ Nitrogen dioxide (NO 2 ). The nitrogen dioxide is measured in the GEE cloud computing platform with the help of Sentinel-5P. Figure 3 indicates the NO 2 values where red color indicates high NO 2 values and block color indicate the low level of NO 2 . Frequently nitrogen dioxide is highly located in burning areas like the agricultural waste, fossil fuel burring location. The north Indian states like West Bengal, Jharkhand, Bihar, Odisha, Uttar Pradesh, Chhattisgarh, Madhya Pradesh, Haryana, and Punjab states are highly nitrogen dioxide located due to the agricultural waste and the fuel burring. However, the year of 2020 and 2021 were mostly nitrogen dioxide located in Delhi, Chhattisgarh, West Bengal, Haryana, and Punjab location. Table 1  . This scenarios and datasets indicate the actual condition and adaptation policies due to the anthropogenic activities and climate change condition. COVID-19 and related restrictions (lockdown, restricted transportation associability) are influences the air pollutants which indicates the climate change is depends on the anthropogenic activities and human can change the environmental issues for future life growing and sustainable healthy livelihood. Human can protect the environmental degradation and ecological diversification due to awareness, planning, management and development for sustainable life, otherwise climate change is lost the human life gradually. The NO 2 information indicates that during COVID-19, low NO 2 was observed, but pre phase NO 2 is high located.
Sulfur dioxide (SO 2 ). The SO 2 is most influencing air pollutants which is affected the human bodies and increased health emergencies.

Ozone (O 3 ).
Ozone is more important factor for our environment where human activities and climate change are influence the ozone variation in the atmosphere. The Sentinel-5P is widely used for calculating the spacebased air pollutants where ozone is more effective air pollutant which is comes from the greenhouse gasses. The ozone is mostly located in 2020 and 2021, where 2018 and 2019 are low ozone located based on Sentinel-5P and GEE platform (Fig. 5). The mostly affected states were central, north and eastern Indian states. Table 3 indicates the variation of the ozone is different periods, where 10 AQI monitoring stations were located like Kolkata, Delhi, Patna, Chennai, Mumbai, Ahmedabad, Bengaluru, Lucknow, Muzaffarnagar and Guwahati area. This data indicates the ozone variation over the selected AQI station over India. High O 3 damage the human health like respiratory tract tissues, irritation and body inflammation, high coughing, asthma like symptoms and chest tightness. Therefore, O 3 measurement is more essential for environmental protection as well as healthy human life.
Carbon monoxide (CO). The same cases are notified in the carbon monoxide maps because this map also indicates the carbon sequences are high. Figure 6 indicates the CO map of the study area where red color indicates the high CO and black color indicates low CO. This map indicates the actual earth atmospheric condition in the study area, where the whole world may face the problems like social, political, economic, and medical emergencies. CO is basically generated due to transportation activities and industrial work. The initial year like 2018 is most CO located due to anthropogenic activities and industrial work, where 2019, 2020 low CO values are were recorded due to COVID-19 restriction and lockdown. The transportation and industrial works have been stop and the results are low CO. However, in the year of 2021, CO gradually increased in the different parts of India due to reopening the industrial works, vehicles and anthropogenic activities. People can change the environmental degradation for adopting some plan and management system otherwise climate change grass the entire life. Mainly vehicles like cars, tracks, fossil fuel burn and individual vehicles are the reason for high CO variation, therefore COVID-19 time measurement of CO is more important because lockdown reducing the CO while pre COVID-19 CO is high. Needs group vehicles use, solar power applies, natural gasses apply and plantations are the possible solution to reducing CO over India.  4 , and CO in the air. The Sentinel 5P is widely used for monitoring the space-based AQI in GEE platform. Figure 7 indicates the variation of AQI in different periods, where 2018, 2019, 2020 and 2021 yearly average data were taken to calculate the variation of AQI. The 2020 and 2021 results indicate AQI change is high where 2018 and 2019 indicates low AQI throughout the year. Different station data were used for monitoring the variation of different air pollutants in India, where NO 2 , SO 2 , O 3 , and CO are monitoring in different ten AQI station. Figure 8 indicates the variation of maximum and minimum air pollutant of different time periods in India. This variation located that the COVID-19 restrictions and lockdowns are mostly influences the fluctuation and low air pollutants where reopening of industrial works and anthropogenic activates are increased the air pollutants in India. Figure 9 indicates that the AQI monitoring station of most effective place Ballygunge, Kolkata, where AQI and others air pollutant parameters are observed (Fig. 9). Figure 10 indicates the box plot of different AQI station minimum and maximum data where four air pollutants are recorded and investigated. The study results will helpful for planning, development, awareness and future disaster management system. People awareness is necessary for unexpected development of air pollutants otherwise climate change and anthropogenic activities are hammering the ecological diversity and destroying the natural phenomena of earth. www.nature.com/scientificreports/

Discussion
Air pollution and the extreme climatic conditions are simultaneously related to each other's, because climate change has been triggering factor for air pollutant and chemical properties of the air. Sunlight is affected by air pollutant like methane, ozone, and aerosol, therefore those air pollutant measurements is necessary for investigating the climate change impacts on the earth's surface. The high voltage electric discharge has been changed the oxygen to ozone where ozonosphere damaged increased the ultraviolet rays in earth's surface. The climate change influences the air quality and pollutants where environment and ecological disturbances are noticed. India has witness of air pollutants in the different megacities where COVID-19 lockdowns were the affective and eye opening time where people can noticed the anthropogenic activities is the main reason for air pollution. Delhi, Kolkata, Mumbai, Hyderabad, Bangalore, Durgapur, Bokaro, and many others Indian cities are mostly recorded low air pollution. The different years Sentinel-5P and MODIS data were used for delineating the air pollutants and AOD in the years of 2018 to 2021. Not only India, others countries have noticeable change located during the COVID-19 country-wide lockdown and restriction [56][57][58][59] . Those conditions increase some awareness, knowledge; environment protraction related activities and planning are increase. COVID-19 triggering the low air pollution and positive air quality because of the countrywide lockdown, restrictions and transportation restrictions over India, therefore vehicles, fossil fuel use and industrial waste were not generated. Those conditions are triggering the ecological variation, environmental protection and improved air quality over India. Most of the megacities are observed low dust particles, air pollution and smoke particles. Therefore, this study can help to identifying the reason and benefits of lockdown or awareness to protect our environment. Fossil fuel also applied in the rural areas for cooking purpose but industrial areas, and vehicles are more used fossil fuel, due to the restriction, those triggering factors are reducing and increase healthy life with improved environmental condition. Nowadays many approaches were applied for air quality measurement and environmental impact assessment like machine learning 60 where symmetric mean average percentage error (SMAPE) multicascade space-time learning model (MCST-Tree) was applied for PM 2.5 distribution measurement. Another air pollutant like PM 2.5 , PM 10 , and O 3 was measured through multi-AP learning network 61 . These research outcomes are mentioned the air pollution is gradually impacts on the earth with huge health issues. The capital city Delhi areas are noticed 80% of pollution generated through transportation sector and this condition located during lockdown periods, where air quality was healthy located during lockdown by Automotive Research Association of India (ARAI) and the Energy and Resources Institute (TERI). Based on Central Pollution Control Board (CPCB) and Ministry of Environment and Forests (MoEF), motorised vehicles are generated ~ 70% of air pollutants, where industrials areas, and fossil fuel are one of the reasons for air pollution in India. The AOD was noticed high in Uttar Pradesh and neighbourhood location in 2018, where simultaneously fluctuation located in the previous four years. Nevertheless, the NO 2 variation located high due to agricultural waste burning and related activities. The main hotspot areas are West Bengal, Uttar Pradesh, Delhi, Haryana, Punjab, Odisha, and Bihar area. Same scenarios are located in different air pollutants where AQI is also fluctuated due to COVID-19 lockdown and related restrictions. Need proper planning, management, awareness, and development for sustainable future planning and healthy life otherwise overwhelming population pressure, www.nature.com/scientificreports/ transportation development and industrial works are affectively influencing the air quality and increase health emergencies in future.

Limitation and recommendation
Air quality measurement is more important factor for sustainable development of planning, human health, unplanned urban expansion and environmental degradation 8,13,62 . Many techniques are used for space-based air quality and pollutants measurement where Sentinel-5P is widely used for space-based air quality measurement and investigation 12,15,63 . But pixel data and station data can't overlap and accurate based on the air quality In India, agricultural waste burning, fossil fuel burning, industrial works and vehicles are the main air pollutants development criteria where industry and transport are the main reason for urban air pollution and agricultural waste burning is the main reason for rural air pollution. Proper planning, management and development strategies can help to protect the environment otherwise climate change and air pollution will increase the health emergencies, ecological diversity and environmental degradation. This study results might be helpful for the researchers, planners, policymakers, administrators, and others stakeholders for sustainable planning for protecting the environment. www.nature.com/scientificreports/ www.nature.com/scientificreports/  www.nature.com/scientificreports/