a thesis about air pollution

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Front Public Health

Environmental and Health Impacts of Air Pollution: A Review

Ioannis manisalidis.

1 Delphis S.A., Kifisia, Greece

2 Laboratory of Hygiene and Environmental Protection, Faculty of Medicine, Democritus University of Thrace, Alexandroupolis, Greece

Elisavet Stavropoulou

3 Centre Hospitalier Universitaire Vaudois (CHUV), Service de Médicine Interne, Lausanne, Switzerland

Agathangelos Stavropoulos

4 School of Social and Political Sciences, University of Glasgow, Glasgow, United Kingdom

Eugenia Bezirtzoglou

One of our era's greatest scourges is air pollution, on account not only of its impact on climate change but also its impact on public and individual health due to increasing morbidity and mortality. There are many pollutants that are major factors in disease in humans. Among them, Particulate Matter (PM), particles of variable but very small diameter, penetrate the respiratory system via inhalation, causing respiratory and cardiovascular diseases, reproductive and central nervous system dysfunctions, and cancer. Despite the fact that ozone in the stratosphere plays a protective role against ultraviolet irradiation, it is harmful when in high concentration at ground level, also affecting the respiratory and cardiovascular system. Furthermore, nitrogen oxide, sulfur dioxide, Volatile Organic Compounds (VOCs), dioxins, and polycyclic aromatic hydrocarbons (PAHs) are all considered air pollutants that are harmful to humans. Carbon monoxide can even provoke direct poisoning when breathed in at high levels. Heavy metals such as lead, when absorbed into the human body, can lead to direct poisoning or chronic intoxication, depending on exposure. Diseases occurring from the aforementioned substances include principally respiratory problems such as Chronic Obstructive Pulmonary Disease (COPD), asthma, bronchiolitis, and also lung cancer, cardiovascular events, central nervous system dysfunctions, and cutaneous diseases. Last but not least, climate change resulting from environmental pollution affects the geographical distribution of many infectious diseases, as do natural disasters. The only way to tackle this problem is through public awareness coupled with a multidisciplinary approach by scientific experts; national and international organizations must address the emergence of this threat and propose sustainable solutions.

Approach to the Problem

The interactions between humans and their physical surroundings have been extensively studied, as multiple human activities influence the environment. The environment is a coupling of the biotic (living organisms and microorganisms) and the abiotic (hydrosphere, lithosphere, and atmosphere).

Pollution is defined as the introduction into the environment of substances harmful to humans and other living organisms. Pollutants are harmful solids, liquids, or gases produced in higher than usual concentrations that reduce the quality of our environment.

Human activities have an adverse effect on the environment by polluting the water we drink, the air we breathe, and the soil in which plants grow. Although the industrial revolution was a great success in terms of technology, society, and the provision of multiple services, it also introduced the production of huge quantities of pollutants emitted into the air that are harmful to human health. Without any doubt, the global environmental pollution is considered an international public health issue with multiple facets. Social, economic, and legislative concerns and lifestyle habits are related to this major problem. Clearly, urbanization and industrialization are reaching unprecedented and upsetting proportions worldwide in our era. Anthropogenic air pollution is one of the biggest public health hazards worldwide, given that it accounts for about 9 million deaths per year ( 1 ).

Without a doubt, all of the aforementioned are closely associated with climate change, and in the event of danger, the consequences can be severe for mankind ( 2 ). Climate changes and the effects of global planetary warming seriously affect multiple ecosystems, causing problems such as food safety issues, ice and iceberg melting, animal extinction, and damage to plants ( 3 , 4 ).

Air pollution has various health effects. The health of susceptible and sensitive individuals can be impacted even on low air pollution days. Short-term exposure to air pollutants is closely related to COPD (Chronic Obstructive Pulmonary Disease), cough, shortness of breath, wheezing, asthma, respiratory disease, and high rates of hospitalization (a measurement of morbidity).

The long-term effects associated with air pollution are chronic asthma, pulmonary insufficiency, cardiovascular diseases, and cardiovascular mortality. According to a Swedish cohort study, diabetes seems to be induced after long-term air pollution exposure ( 5 ). Moreover, air pollution seems to have various malign health effects in early human life, such as respiratory, cardiovascular, mental, and perinatal disorders ( 3 ), leading to infant mortality or chronic disease in adult age ( 6 ).

National reports have mentioned the increased risk of morbidity and mortality ( 1 ). These studies were conducted in many places around the world and show a correlation between daily ranges of particulate matter (PM) concentration and daily mortality. Climate shifts and global planetary warming ( 3 ) could aggravate the situation. Besides, increased hospitalization (an index of morbidity) has been registered among the elderly and susceptible individuals for specific reasons. Fine and ultrafine particulate matter seems to be associated with more serious illnesses ( 6 ), as it can invade the deepest parts of the airways and more easily reach the bloodstream.

Air pollution mainly affects those living in large urban areas, where road emissions contribute the most to the degradation of air quality. There is also a danger of industrial accidents, where the spread of a toxic fog can be fatal to the populations of the surrounding areas. The dispersion of pollutants is determined by many parameters, most notably atmospheric stability and wind ( 6 ).

In developing countries ( 7 ), the problem is more serious due to overpopulation and uncontrolled urbanization along with the development of industrialization. This leads to poor air quality, especially in countries with social disparities and a lack of information on sustainable management of the environment. The use of fuels such as wood fuel or solid fuel for domestic needs due to low incomes exposes people to bad-quality, polluted air at home. It is of note that three billion people around the world are using the above sources of energy for their daily heating and cooking needs ( 8 ). In developing countries, the women of the household seem to carry the highest risk for disease development due to their longer duration exposure to the indoor air pollution ( 8 , 9 ). Due to its fast industrial development and overpopulation, China is one of the Asian countries confronting serious air pollution problems ( 10 , 11 ). The lung cancer mortality observed in China is associated with fine particles ( 12 ). As stated already, long-term exposure is associated with deleterious effects on the cardiovascular system ( 3 , 5 ). However, it is interesting to note that cardiovascular diseases have mostly been observed in developed and high-income countries rather than in the developing low-income countries exposed highly to air pollution ( 13 ). Extreme air pollution is recorded in India, where the air quality reaches hazardous levels. New Delhi is one of the more polluted cities in India. Flights in and out of New Delhi International Airport are often canceled due to the reduced visibility associated with air pollution. Pollution is occurring both in urban and rural areas in India due to the fast industrialization, urbanization, and rise in use of motorcycle transportation. Nevertheless, biomass combustion associated with heating and cooking needs and practices is a major source of household air pollution in India and in Nepal ( 14 , 15 ). There is spatial heterogeneity in India, as areas with diverse climatological conditions and population and education levels generate different indoor air qualities, with higher PM 2.5 observed in North Indian states (557–601 μg/m 3 ) compared to the Southern States (183–214 μg/m 3 ) ( 16 , 17 ). The cold climate of the North Indian areas may be the main reason for this, as longer periods at home and more heating are necessary compared to in the tropical climate of Southern India. Household air pollution in India is associated with major health effects, especially in women and young children, who stay indoors for longer periods. Chronic obstructive respiratory disease (CORD) and lung cancer are mostly observed in women, while acute lower respiratory disease is seen in young children under 5 years of age ( 18 ).

Accumulation of air pollution, especially sulfur dioxide and smoke, reaching 1,500 mg/m3, resulted in an increase in the number of deaths (4,000 deaths) in December 1952 in London and in 1963 in New York City (400 deaths) ( 19 ). An association of pollution with mortality was reported on the basis of monitoring of outdoor pollution in six US metropolitan cities ( 20 ). In every case, it seems that mortality was closely related to the levels of fine, inhalable, and sulfate particles more than with the levels of total particulate pollution, aerosol acidity, sulfur dioxide, or nitrogen dioxide ( 20 ).

Furthermore, extremely high levels of pollution are reported in Mexico City and Rio de Janeiro, followed by Milan, Ankara, Melbourne, Tokyo, and Moscow ( 19 ).

Based on the magnitude of the public health impact, it is certain that different kinds of interventions should be taken into account. Success and effectiveness in controlling air pollution, specifically at the local level, have been reported. Adequate technological means are applied considering the source and the nature of the emission as well as its impact on health and the environment. The importance of point sources and non-point sources of air pollution control is reported by Schwela and Köth-Jahr ( 21 ). Without a doubt, a detailed emission inventory must record all sources in a given area. Beyond considering the above sources and their nature, topography and meteorology should also be considered, as stated previously. Assessment of the control policies and methods is often extrapolated from the local to the regional and then to the global scale. Air pollution may be dispersed and transported from one region to another area located far away. Air pollution management means the reduction to acceptable levels or possible elimination of air pollutants whose presence in the air affects our health or the environmental ecosystem. Private and governmental entities and authorities implement actions to ensure the air quality ( 22 ). Air quality standards and guidelines were adopted for the different pollutants by the WHO and EPA as a tool for the management of air quality ( 1 , 23 ). These standards have to be compared to the emissions inventory standards by causal analysis and dispersion modeling in order to reveal the problematic areas ( 24 ). Inventories are generally based on a combination of direct measurements and emissions modeling ( 24 ).

As an example, we state here the control measures at the source through the use of catalytic converters in cars. These are devices that turn the pollutants and toxic gases produced from combustion engines into less-toxic pollutants by catalysis through redox reactions ( 25 ). In Greece, the use of private cars was restricted by tracking their license plates in order to reduce traffic congestion during rush hour ( 25 ).

Concerning industrial emissions, collectors and closed systems can keep the air pollution to the minimal standards imposed by legislation ( 26 ).

Current strategies to improve air quality require an estimation of the economic value of the benefits gained from proposed programs. These proposed programs by public authorities, and directives are issued with guidelines to be respected.

In Europe, air quality limit values AQLVs (Air Quality Limit Values) are issued for setting off planning claims ( 27 ). In the USA, the NAAQS (National Ambient Air Quality Standards) establish the national air quality limit values ( 27 ). While both standards and directives are based on different mechanisms, significant success has been achieved in the reduction of overall emissions and associated health and environmental effects ( 27 ). The European Directive identifies geographical areas of risk exposure as monitoring/assessment zones to record the emission sources and levels of air pollution ( 27 ), whereas the USA establishes global geographical air quality criteria according to the severity of their air quality problem and records all sources of the pollutants and their precursors ( 27 ).

In this vein, funds have been financing, directly or indirectly, projects related to air quality along with the technical infrastructure to maintain good air quality. These plans focus on an inventory of databases from air quality environmental planning awareness campaigns. Moreover, pollution measures of air emissions may be taken for vehicles, machines, and industries in urban areas.

Technological innovation can only be successful if it is able to meet the needs of society. In this sense, technology must reflect the decision-making practices and procedures of those involved in risk assessment and evaluation and act as a facilitator in providing information and assessments to enable decision makers to make the best decisions possible. Summarizing the aforementioned in order to design an effective air quality control strategy, several aspects must be considered: environmental factors and ambient air quality conditions, engineering factors and air pollutant characteristics, and finally, economic operating costs for technological improvement and administrative and legal costs. Considering the economic factor, competitiveness through neoliberal concepts is offering a solution to environmental problems ( 22 ).

The development of environmental governance, along with technological progress, has initiated the deployment of a dialogue. Environmental politics has created objections and points of opposition between different political parties, scientists, media, and governmental and non-governmental organizations ( 22 ). Radical environmental activism actions and movements have been created ( 22 ). The rise of the new information and communication technologies (ICTs) are many times examined as to whether and in which way they have influenced means of communication and social movements such as activism ( 28 ). Since the 1990s, the term “digital activism” has been used increasingly and in many different disciplines ( 29 ). Nowadays, multiple digital technologies can be used to produce a digital activism outcome on environmental issues. More specifically, devices with online capabilities such as computers or mobile phones are being used as a way to pursue change in political and social affairs ( 30 ).

In the present paper, we focus on the sources of environmental pollution in relation to public health and propose some solutions and interventions that may be of interest to environmental legislators and decision makers.

Sources of Exposure

It is known that the majority of environmental pollutants are emitted through large-scale human activities such as the use of industrial machinery, power-producing stations, combustion engines, and cars. Because these activities are performed at such a large scale, they are by far the major contributors to air pollution, with cars estimated to be responsible for approximately 80% of today's pollution ( 31 ). Some other human activities are also influencing our environment to a lesser extent, such as field cultivation techniques, gas stations, fuel tanks heaters, and cleaning procedures ( 32 ), as well as several natural sources, such as volcanic and soil eruptions and forest fires.

The classification of air pollutants is based mainly on the sources producing pollution. Therefore, it is worth mentioning the four main sources, following the classification system: Major sources, Area sources, Mobile sources, and Natural sources.

Major sources include the emission of pollutants from power stations, refineries, and petrochemicals, the chemical and fertilizer industries, metallurgical and other industrial plants, and, finally, municipal incineration.

Indoor area sources include domestic cleaning activities, dry cleaners, printing shops, and petrol stations.

Mobile sources include automobiles, cars, railways, airways, and other types of vehicles.

Finally, natural sources include, as stated previously, physical disasters ( 33 ) such as forest fire, volcanic erosion, dust storms, and agricultural burning.

However, many classification systems have been proposed. Another type of classification is a grouping according to the recipient of the pollution, as follows:

Air pollution is determined as the presence of pollutants in the air in large quantities for long periods. Air pollutants are dispersed particles, hydrocarbons, CO, CO 2 , NO, NO 2 , SO 3 , etc.

Water pollution is organic and inorganic charge and biological charge ( 10 ) at high levels that affect the water quality ( 34 , 35 ).

Soil pollution occurs through the release of chemicals or the disposal of wastes, such as heavy metals, hydrocarbons, and pesticides.

Air pollution can influence the quality of soil and water bodies by polluting precipitation, falling into water and soil environments ( 34 , 36 ). Notably, the chemistry of the soil can be amended due to acid precipitation by affecting plants, cultures, and water quality ( 37 ). Moreover, movement of heavy metals is favored by soil acidity, and metals are so then moving into the watery environment. It is known that heavy metals such as aluminum are noxious to wildlife and fishes. Soil quality seems to be of importance, as soils with low calcium carbonate levels are at increased jeopardy from acid rain. Over and above rain, snow and particulate matter drip into watery ' bodies ( 36 , 38 ).

Lastly, pollution is classified following type of origin:

Radioactive and nuclear pollution , releasing radioactive and nuclear pollutants into water, air, and soil during nuclear explosions and accidents, from nuclear weapons, and through handling or disposal of radioactive sewage.

Radioactive materials can contaminate surface water bodies and, being noxious to the environment, plants, animals, and humans. It is known that several radioactive substances such as radium and uranium concentrate in the bones and can cause cancers ( 38 , 39 ).

Noise pollution is produced by machines, vehicles, traffic noises, and musical installations that are harmful to our hearing.

The World Health Organization introduced the term DALYs. The DALYs for a disease or health condition is defined as the sum of the Years of Life Lost (YLL) due to premature mortality in the population and the Years Lost due to Disability (YLD) for people living with the health condition or its consequences ( 39 ). In Europe, air pollution is the main cause of disability-adjusted life years lost (DALYs), followed by noise pollution. The potential relationships of noise and air pollution with health have been studied ( 40 ). The study found that DALYs related to noise were more important than those related to air pollution, as the effects of environmental noise on cardiovascular disease were independent of air pollution ( 40 ). Environmental noise should be counted as an independent public health risk ( 40 ).

Environmental pollution occurs when changes in the physical, chemical, or biological constituents of the environment (air masses, temperature, climate, etc.) are produced.

Pollutants harm our environment either by increasing levels above normal or by introducing harmful toxic substances. Primary pollutants are directly produced from the above sources, and secondary pollutants are emitted as by-products of the primary ones. Pollutants can be biodegradable or non-biodegradable and of natural origin or anthropogenic, as stated previously. Moreover, their origin can be a unique source (point-source) or dispersed sources.

Pollutants have differences in physical and chemical properties, explaining the discrepancy in their capacity for producing toxic effects. As an example, we state here that aerosol compounds ( 41 – 43 ) have a greater toxicity than gaseous compounds due to their tiny size (solid or liquid) in the atmosphere; they have a greater penetration capacity. Gaseous compounds are eliminated more easily by our respiratory system ( 41 ). These particles are able to damage lungs and can even enter the bloodstream ( 41 ), leading to the premature deaths of millions of people yearly. Moreover, the aerosol acidity ([H+]) seems to considerably enhance the production of secondary organic aerosols (SOA), but this last aspect is not supported by other scientific teams ( 38 ).

Climate and Pollution

Air pollution and climate change are closely related. Climate is the other side of the same coin that reduces the quality of our Earth ( 44 ). Pollutants such as black carbon, methane, tropospheric ozone, and aerosols affect the amount of incoming sunlight. As a result, the temperature of the Earth is increasing, resulting in the melting of ice, icebergs, and glaciers.

In this vein, climatic changes will affect the incidence and prevalence of both residual and imported infections in Europe. Climate and weather affect the duration, timing, and intensity of outbreaks strongly and change the map of infectious diseases in the globe ( 45 ). Mosquito-transmitted parasitic or viral diseases are extremely climate-sensitive, as warming firstly shortens the pathogen incubation period and secondly shifts the geographic map of the vector. Similarly, water-warming following climate changes leads to a high incidence of waterborne infections. Recently, in Europe, eradicated diseases seem to be emerging due to the migration of population, for example, cholera, poliomyelitis, tick-borne encephalitis, and malaria ( 46 ).

The spread of epidemics is associated with natural climate disasters and storms, which seem to occur more frequently nowadays ( 47 ). Malnutrition and disequilibration of the immune system are also associated with the emerging infections affecting public health ( 48 ).

The Chikungunya virus “took the airplane” from the Indian Ocean to Europe, as outbreaks of the disease were registered in Italy ( 49 ) as well as autochthonous cases in France ( 50 ).

An increase in cryptosporidiosis in the United Kingdom and in the Czech Republic seems to have occurred following flooding ( 36 , 51 ).

As stated previously, aerosols compounds are tiny in size and considerably affect the climate. They are able to dissipate sunlight (the albedo phenomenon) by dispersing a quarter of the sun's rays back to space and have cooled the global temperature over the last 30 years ( 52 ).

Air Pollutants

The World Health Organization (WHO) reports on six major air pollutants, namely particle pollution, ground-level ozone, carbon monoxide, sulfur oxides, nitrogen oxides, and lead. Air pollution can have a disastrous effect on all components of the environment, including groundwater, soil, and air. Additionally, it poses a serious threat to living organisms. In this vein, our interest is mainly to focus on these pollutants, as they are related to more extensive and severe problems in human health and environmental impact. Acid rain, global warming, the greenhouse effect, and climate changes have an important ecological impact on air pollution ( 53 ).

Particulate Matter (PM) and Health

Studies have shown a relationship between particulate matter (PM) and adverse health effects, focusing on either short-term (acute) or long-term (chronic) PM exposure.

Particulate matter (PM) is usually formed in the atmosphere as a result of chemical reactions between the different pollutants. The penetration of particles is closely dependent on their size ( 53 ). Particulate Matter (PM) was defined as a term for particles by the United States Environmental Protection Agency ( 54 ). Particulate matter (PM) pollution includes particles with diameters of 10 micrometers (μm) or smaller, called PM 10 , and extremely fine particles with diameters that are generally 2.5 micrometers (μm) and smaller.

Particulate matter contains tiny liquid or solid droplets that can be inhaled and cause serious health effects ( 55 ). Particles <10 μm in diameter (PM 10 ) after inhalation can invade the lungs and even reach the bloodstream. Fine particles, PM 2.5 , pose a greater risk to health ( 6 , 56 ) ( Table 1 ).

Penetrability according to particle size.

Multiple epidemiological studies have been performed on the health effects of PM. A positive relation was shown between both short-term and long-term exposures of PM 2.5 and acute nasopharyngitis ( 56 ). In addition, long-term exposure to PM for years was found to be related to cardiovascular diseases and infant mortality.

Those studies depend on PM 2.5 monitors and are restricted in terms of study area or city area due to a lack of spatially resolved daily PM 2.5 concentration data and, in this way, are not representative of the entire population. Following a recent epidemiological study by the Department of Environmental Health at Harvard School of Public Health (Boston, MA) ( 57 ), it was reported that, as PM 2.5 concentrations vary spatially, an exposure error (Berkson error) seems to be produced, and the relative magnitudes of the short- and long-term effects are not yet completely elucidated. The team developed a PM 2.5 exposure model based on remote sensing data for assessing short- and long-term human exposures ( 57 ). This model permits spatial resolution in short-term effects plus the assessment of long-term effects in the whole population.

Moreover, respiratory diseases and affection of the immune system are registered as long-term chronic effects ( 58 ). It is worth noting that people with asthma, pneumonia, diabetes, and respiratory and cardiovascular diseases are especially susceptible and vulnerable to the effects of PM. PM 2.5 , followed by PM 10 , are strongly associated with diverse respiratory system diseases ( 59 ), as their size permits them to pierce interior spaces ( 60 ). The particles produce toxic effects according to their chemical and physical properties. The components of PM 10 and PM 2.5 can be organic (polycyclic aromatic hydrocarbons, dioxins, benzene, 1-3 butadiene) or inorganic (carbon, chlorides, nitrates, sulfates, metals) in nature ( 55 ).

Particulate Matter (PM) is divided into four main categories according to type and size ( 61 ) ( Table 2 ).

Types and sizes of particulate Matter (PM).

Gas contaminants include PM in aerial masses.

Particulate contaminants include contaminants such as smog, soot, tobacco smoke, oil smoke, fly ash, and cement dust.

Biological Contaminants are microorganisms (bacteria, viruses, fungi, mold, and bacterial spores), cat allergens, house dust and allergens, and pollen.

Types of Dust include suspended atmospheric dust, settling dust, and heavy dust.

Finally, another fact is that the half-lives of PM 10 and PM 2.5 particles in the atmosphere is extended due to their tiny dimensions; this permits their long-lasting suspension in the atmosphere and even their transfer and spread to distant destinations where people and the environment may be exposed to the same magnitude of pollution ( 53 ). They are able to change the nutrient balance in watery ecosystems, damage forests and crops, and acidify water bodies.

As stated, PM 2.5 , due to their tiny size, are causing more serious health effects. These aforementioned fine particles are the main cause of the “haze” formation in different metropolitan areas ( 12 , 13 , 61 ).

Ozone Impact in the Atmosphere

Ozone (O 3 ) is a gas formed from oxygen under high voltage electric discharge ( 62 ). It is a strong oxidant, 52% stronger than chlorine. It arises in the stratosphere, but it could also arise following chain reactions of photochemical smog in the troposphere ( 63 ).

Ozone can travel to distant areas from its initial source, moving with air masses ( 64 ). It is surprising that ozone levels over cities are low in contrast to the increased amounts occuring in urban areas, which could become harmful for cultures, forests, and vegetation ( 65 ) as it is reducing carbon assimilation ( 66 ). Ozone reduces growth and yield ( 47 , 48 ) and affects the plant microflora due to its antimicrobial capacity ( 67 , 68 ). In this regard, ozone acts upon other natural ecosystems, with microflora ( 69 , 70 ) and animal species changing their species composition ( 71 ). Ozone increases DNA damage in epidermal keratinocytes and leads to impaired cellular function ( 72 ).

Ground-level ozone (GLO) is generated through a chemical reaction between oxides of nitrogen and VOCs emitted from natural sources and/or following anthropogenic activities.

Ozone uptake usually occurs by inhalation. Ozone affects the upper layers of the skin and the tear ducts ( 73 ). A study of short-term exposure of mice to high levels of ozone showed malondialdehyde formation in the upper skin (epidermis) but also depletion in vitamins C and E. It is likely that ozone levels are not interfering with the skin barrier function and integrity to predispose to skin disease ( 74 ).

Due to the low water-solubility of ozone, inhaled ozone has the capacity to penetrate deeply into the lungs ( 75 ).

Toxic effects induced by ozone are registered in urban areas all over the world, causing biochemical, morphologic, functional, and immunological disorders ( 76 ).

The European project (APHEA2) focuses on the acute effects of ambient ozone concentrations on mortality ( 77 ). Daily ozone concentrations compared to the daily number of deaths were reported from different European cities for a 3-year period. During the warm period of the year, an observed increase in ozone concentration was associated with an increase in the daily number of deaths (0.33%), in the number of respiratory deaths (1.13%), and in the number of cardiovascular deaths (0.45%). No effect was observed during wintertime.

Carbon Monoxide (CO)

Carbon monoxide is produced by fossil fuel when combustion is incomplete. The symptoms of poisoning due to inhaling carbon monoxide include headache, dizziness, weakness, nausea, vomiting, and, finally, loss of consciousness.

The affinity of carbon monoxide to hemoglobin is much greater than that of oxygen. In this vein, serious poisoning may occur in people exposed to high levels of carbon monoxide for a long period of time. Due to the loss of oxygen as a result of the competitive binding of carbon monoxide, hypoxia, ischemia, and cardiovascular disease are observed.

Carbon monoxide affects the greenhouses gases that are tightly connected to global warming and climate. This should lead to an increase in soil and water temperatures, and extreme weather conditions or storms may occur ( 68 ).

However, in laboratory and field experiments, it has been seen to produce increased plant growth ( 78 ).

Nitrogen Oxide (NO 2 )

Nitrogen oxide is a traffic-related pollutant, as it is emitted from automobile motor engines ( 79 , 80 ). It is an irritant of the respiratory system as it penetrates deep in the lung, inducing respiratory diseases, coughing, wheezing, dyspnea, bronchospasm, and even pulmonary edema when inhaled at high levels. It seems that concentrations over 0.2 ppm produce these adverse effects in humans, while concentrations higher than 2.0 ppm affect T-lymphocytes, particularly the CD8+ cells and NK cells that produce our immune response ( 81 ).It is reported that long-term exposure to high levels of nitrogen dioxide can be responsible for chronic lung disease. Long-term exposure to NO 2 can impair the sense of smell ( 81 ).

However, systems other than respiratory ones can be involved, as symptoms such as eye, throat, and nose irritation have been registered ( 81 ).

High levels of nitrogen dioxide are deleterious to crops and vegetation, as they have been observed to reduce crop yield and plant growth efficiency. Moreover, NO 2 can reduce visibility and discolor fabrics ( 81 ).

Sulfur Dioxide (SO 2 )

Sulfur dioxide is a harmful gas that is emitted mainly from fossil fuel consumption or industrial activities. The annual standard for SO 2 is 0.03 ppm ( 82 ). It affects human, animal, and plant life. Susceptible people as those with lung disease, old people, and children, who present a higher risk of damage. The major health problems associated with sulfur dioxide emissions in industrialized areas are respiratory irritation, bronchitis, mucus production, and bronchospasm, as it is a sensory irritant and penetrates deep into the lung converted into bisulfite and interacting with sensory receptors, causing bronchoconstriction. Moreover, skin redness, damage to the eyes (lacrimation and corneal opacity) and mucous membranes, and worsening of pre-existing cardiovascular disease have been observed ( 81 ).

Environmental adverse effects, such as acidification of soil and acid rain, seem to be associated with sulfur dioxide emissions ( 83 ).

Lead is a heavy metal used in different industrial plants and emitted from some petrol motor engines, batteries, radiators, waste incinerators, and waste waters ( 84 ).

Moreover, major sources of lead pollution in the air are metals, ore, and piston-engine aircraft. Lead poisoning is a threat to public health due to its deleterious effects upon humans, animals, and the environment, especially in the developing countries.

Exposure to lead can occur through inhalation, ingestion, and dermal absorption. Trans- placental transport of lead was also reported, as lead passes through the placenta unencumbered ( 85 ). The younger the fetus is, the more harmful the toxic effects. Lead toxicity affects the fetal nervous system; edema or swelling of the brain is observed ( 86 ). Lead, when inhaled, accumulates in the blood, soft tissue, liver, lung, bones, and cardiovascular, nervous, and reproductive systems. Moreover, loss of concentration and memory, as well as muscle and joint pain, were observed in adults ( 85 , 86 ).

Children and newborns ( 87 ) are extremely susceptible even to minimal doses of lead, as it is a neurotoxicant and causes learning disabilities, impairment of memory, hyperactivity, and even mental retardation.

Elevated amounts of lead in the environment are harmful to plants and crop growth. Neurological effects are observed in vertebrates and animals in association with high lead levels ( 88 ).

Polycyclic Aromatic Hydrocarbons(PAHs)

The distribution of PAHs is ubiquitous in the environment, as the atmosphere is the most important means of their dispersal. They are found in coal and in tar sediments. Moreover, they are generated through incomplete combustion of organic matter as in the cases of forest fires, incineration, and engines ( 89 ). PAH compounds, such as benzopyrene, acenaphthylene, anthracene, and fluoranthene are recognized as toxic, mutagenic, and carcinogenic substances. They are an important risk factor for lung cancer ( 89 ).

Volatile Organic Compounds(VOCs)

Volatile organic compounds (VOCs), such as toluene, benzene, ethylbenzene, and xylene ( 90 ), have been found to be associated with cancer in humans ( 91 ). The use of new products and materials has actually resulted in increased concentrations of VOCs. VOCs pollute indoor air ( 90 ) and may have adverse effects on human health ( 91 ). Short-term and long-term adverse effects on human health are observed. VOCs are responsible for indoor air smells. Short-term exposure is found to cause irritation of eyes, nose, throat, and mucosal membranes, while those of long duration exposure include toxic reactions ( 92 ). Predictable assessment of the toxic effects of complex VOC mixtures is difficult to estimate, as these pollutants can have synergic, antagonistic, or indifferent effects ( 91 , 93 ).

Dioxins originate from industrial processes but also come from natural processes, such as forest fires and volcanic eruptions. They accumulate in foods such as meat and dairy products, fish and shellfish, and especially in the fatty tissue of animals ( 94 ).

Short-period exhibition to high dioxin concentrations may result in dark spots and lesions on the skin ( 94 ). Long-term exposure to dioxins can cause developmental problems, impairment of the immune, endocrine and nervous systems, reproductive infertility, and cancer ( 94 ).

Without any doubt, fossil fuel consumption is responsible for a sizeable part of air contamination. This contamination may be anthropogenic, as in agricultural and industrial processes or transportation, while contamination from natural sources is also possible. Interestingly, it is of note that the air quality standards established through the European Air Quality Directive are somewhat looser than the WHO guidelines, which are stricter ( 95 ).

Effect of Air Pollution on Health

The most common air pollutants are ground-level ozone and Particulates Matter (PM). Air pollution is distinguished into two main types:

Outdoor pollution is the ambient air pollution.

Indoor pollution is the pollution generated by household combustion of fuels.

People exposed to high concentrations of air pollutants experience disease symptoms and states of greater and lesser seriousness. These effects are grouped into short- and long-term effects affecting health.

Susceptible populations that need to be aware of health protection measures include old people, children, and people with diabetes and predisposing heart or lung disease, especially asthma.

As extensively stated previously, according to a recent epidemiological study from Harvard School of Public Health, the relative magnitudes of the short- and long-term effects have not been completely clarified ( 57 ) due to the different epidemiological methodologies and to the exposure errors. New models are proposed for assessing short- and long-term human exposure data more successfully ( 57 ). Thus, in the present section, we report the more common short- and long-term health effects but also general concerns for both types of effects, as these effects are often dependent on environmental conditions, dose, and individual susceptibility.

Short-term effects are temporary and range from simple discomfort, such as irritation of the eyes, nose, skin, throat, wheezing, coughing and chest tightness, and breathing difficulties, to more serious states, such as asthma, pneumonia, bronchitis, and lung and heart problems. Short-term exposure to air pollution can also cause headaches, nausea, and dizziness.

These problems can be aggravated by extended long-term exposure to the pollutants, which is harmful to the neurological, reproductive, and respiratory systems and causes cancer and even, rarely, deaths.

The long-term effects are chronic, lasting for years or the whole life and can even lead to death. Furthermore, the toxicity of several air pollutants may also induce a variety of cancers in the long term ( 96 ).

As stated already, respiratory disorders are closely associated with the inhalation of air pollutants. These pollutants will invade through the airways and will accumulate at the cells. Damage to target cells should be related to the pollutant component involved and its source and dose. Health effects are also closely dependent on country, area, season, and time. An extended exposure duration to the pollutant should incline to long-term health effects in relation also to the above factors.

Particulate Matter (PMs), dust, benzene, and O 3 cause serious damage to the respiratory system ( 97 ). Moreover, there is a supplementary risk in case of existing respiratory disease such as asthma ( 98 ). Long-term effects are more frequent in people with a predisposing disease state. When the trachea is contaminated by pollutants, voice alterations may be remarked after acute exposure. Chronic obstructive pulmonary disease (COPD) may be induced following air pollution, increasing morbidity and mortality ( 99 ). Long-term effects from traffic, industrial air pollution, and combustion of fuels are the major factors for COPD risk ( 99 ).

Multiple cardiovascular effects have been observed after exposure to air pollutants ( 100 ). Changes occurred in blood cells after long-term exposure may affect cardiac functionality. Coronary arteriosclerosis was reported following long-term exposure to traffic emissions ( 101 ), while short-term exposure is related to hypertension, stroke, myocardial infracts, and heart insufficiency. Ventricle hypertrophy is reported to occur in humans after long-time exposure to nitrogen oxide (NO 2 ) ( 102 , 103 ).

Neurological effects have been observed in adults and children after extended-term exposure to air pollutants.

Psychological complications, autism, retinopathy, fetal growth, and low birth weight seem to be related to long-term air pollution ( 83 ). The etiologic agent of the neurodegenerative diseases (Alzheimer's and Parkinson's) is not yet known, although it is believed that extended exposure to air pollution seems to be a factor. Specifically, pesticides and metals are cited as etiological factors, together with diet. The mechanisms in the development of neurodegenerative disease include oxidative stress, protein aggregation, inflammation, and mitochondrial impairment in neurons ( 104 ) ( Figure 1 ).

Impact of air pollutants on the brain.

Brain inflammation was observed in dogs living in a highly polluted area in Mexico for a long period ( 105 ). In human adults, markers of systemic inflammation (IL-6 and fibrinogen) were found to be increased as an immediate response to PNC on the IL-6 level, possibly leading to the production of acute-phase proteins ( 106 ). The progression of atherosclerosis and oxidative stress seem to be the mechanisms involved in the neurological disturbances caused by long-term air pollution. Inflammation comes secondary to the oxidative stress and seems to be involved in the impairment of developmental maturation, affecting multiple organs ( 105 , 107 ). Similarly, other factors seem to be involved in the developmental maturation, which define the vulnerability to long-term air pollution. These include birthweight, maternal smoking, genetic background and socioeconomic environment, as well as education level.

However, diet, starting from breast-feeding, is another determinant factor. Diet is the main source of antioxidants, which play a key role in our protection against air pollutants ( 108 ). Antioxidants are free radical scavengers and limit the interaction of free radicals in the brain ( 108 ). Similarly, genetic background may result in a differential susceptibility toward the oxidative stress pathway ( 60 ). For example, antioxidant supplementation with vitamins C and E appears to modulate the effect of ozone in asthmatic children homozygous for the GSTM1 null allele ( 61 ). Inflammatory cytokines released in the periphery (e.g., respiratory epithelia) upregulate the innate immune Toll-like receptor 2. Such activation and the subsequent events leading to neurodegeneration have recently been observed in lung lavage in mice exposed to ambient Los Angeles (CA, USA) particulate matter ( 61 ). In children, neurodevelopmental morbidities were observed after lead exposure. These children developed aggressive and delinquent behavior, reduced intelligence, learning difficulties, and hyperactivity ( 109 ). No level of lead exposure seems to be “safe,” and the scientific community has asked the Centers for Disease Control and Prevention (CDC) to reduce the current screening guideline of 10 μg/dl ( 109 ).

It is important to state that impact on the immune system, causing dysfunction and neuroinflammation ( 104 ), is related to poor air quality. Yet, increases in serum levels of immunoglobulins (IgA, IgM) and the complement component C3 are observed ( 106 ). Another issue is that antigen presentation is affected by air pollutants, as there is an upregulation of costimulatory molecules such as CD80 and CD86 on macrophages ( 110 ).

As is known, skin is our shield against ultraviolet radiation (UVR) and other pollutants, as it is the most exterior layer of our body. Traffic-related pollutants, such as PAHs, VOCs, oxides, and PM, may cause pigmented spots on our skin ( 111 ). On the one hand, as already stated, when pollutants penetrate through the skin or are inhaled, damage to the organs is observed, as some of these pollutants are mutagenic and carcinogenic, and, specifically, they affect the liver and lung. On the other hand, air pollutants (and those in the troposphere) reduce the adverse effects of ultraviolet radiation UVR in polluted urban areas ( 111 ). Air pollutants absorbed by the human skin may contribute to skin aging, psoriasis, acne, urticaria, eczema, and atopic dermatitis ( 111 ), usually caused by exposure to oxides and photochemical smoke ( 111 ). Exposure to PM and cigarette smoking act as skin-aging agents, causing spots, dyschromia, and wrinkles. Lastly, pollutants have been associated with skin cancer ( 111 ).

Higher morbidity is reported to fetuses and children when exposed to the above dangers. Impairment in fetal growth, low birth weight, and autism have been reported ( 112 ).

Another exterior organ that may be affected is the eye. Contamination usually comes from suspended pollutants and may result in asymptomatic eye outcomes, irritation ( 112 ), retinopathy, or dry eye syndrome ( 113 , 114 ).

Environmental Impact of Air Pollution

Air pollution is harming not only human health but also the environment ( 115 ) in which we live. The most important environmental effects are as follows.

Acid rain is wet (rain, fog, snow) or dry (particulates and gas) precipitation containing toxic amounts of nitric and sulfuric acids. They are able to acidify the water and soil environments, damage trees and plantations, and even damage buildings and outdoor sculptures, constructions, and statues.

Haze is produced when fine particles are dispersed in the air and reduce the transparency of the atmosphere. It is caused by gas emissions in the air coming from industrial facilities, power plants, automobiles, and trucks.

Ozone , as discussed previously, occurs both at ground level and in the upper level (stratosphere) of the Earth's atmosphere. Stratospheric ozone is protecting us from the Sun's harmful ultraviolet (UV) rays. In contrast, ground-level ozone is harmful to human health and is a pollutant. Unfortunately, stratospheric ozone is gradually damaged by ozone-depleting substances (i.e., chemicals, pesticides, and aerosols). If this protecting stratospheric ozone layer is thinned, then UV radiation can reach our Earth, with harmful effects for human life (skin cancer) ( 116 ) and crops ( 117 ). In plants, ozone penetrates through the stomata, inducing them to close, which blocks CO 2 transfer and induces a reduction in photosynthesis ( 118 ).

Global climate change is an important issue that concerns mankind. As is known, the “greenhouse effect” keeps the Earth's temperature stable. Unhappily, anthropogenic activities have destroyed this protecting temperature effect by producing large amounts of greenhouse gases, and global warming is mounting, with harmful effects on human health, animals, forests, wildlife, agriculture, and the water environment. A report states that global warming is adding to the health risks of poor people ( 119 ).

People living in poorly constructed buildings in warm-climate countries are at high risk for heat-related health problems as temperatures mount ( 119 ).

Wildlife is burdened by toxic pollutants coming from the air, soil, or the water ecosystem and, in this way, animals can develop health problems when exposed to high levels of pollutants. Reproductive failure and birth effects have been reported.

Eutrophication is occurring when elevated concentrations of nutrients (especially nitrogen) stimulate the blooming of aquatic algae, which can cause a disequilibration in the diversity of fish and their deaths.

Without a doubt, there is a critical concentration of pollution that an ecosystem can tolerate without being destroyed, which is associated with the ecosystem's capacity to neutralize acidity. The Canada Acid Rain Program established this load at 20 kg/ha/yr ( 120 ).

Hence, air pollution has deleterious effects on both soil and water ( 121 ). Concerning PM as an air pollutant, its impact on crop yield and food productivity has been reported. Its impact on watery bodies is associated with the survival of living organisms and fishes and their productivity potential ( 121 ).

An impairment in photosynthetic rhythm and metabolism is observed in plants exposed to the effects of ozone ( 121 ).

Sulfur and nitrogen oxides are involved in the formation of acid rain and are harmful to plants and marine organisms.

Last but not least, as mentioned above, the toxicity associated with lead and other metals is the main threat to our ecosystems (air, water, and soil) and living creatures ( 121 ).

In 2018, during the first WHO Global Conference on Air Pollution and Health, the WHO's General Director, Dr. Tedros Adhanom Ghebreyesus, called air pollution a “silent public health emergency” and “the new tobacco” ( 122 ).

Undoubtedly, children are particularly vulnerable to air pollution, especially during their development. Air pollution has adverse effects on our lives in many different respects.

Diseases associated with air pollution have not only an important economic impact but also a societal impact due to absences from productive work and school.

Despite the difficulty of eradicating the problem of anthropogenic environmental pollution, a successful solution could be envisaged as a tight collaboration of authorities, bodies, and doctors to regularize the situation. Governments should spread sufficient information and educate people and should involve professionals in these issues so as to control the emergence of the problem successfully.

Technologies to reduce air pollution at the source must be established and should be used in all industries and power plants. The Kyoto Protocol of 1997 set as a major target the reduction of GHG emissions to below 5% by 2012 ( 123 ). This was followed by the Copenhagen summit, 2009 ( 124 ), and then the Durban summit of 2011 ( 125 ), where it was decided to keep to the same line of action. The Kyoto protocol and the subsequent ones were ratified by many countries. Among the pioneers who adopted this important protocol for the world's environmental and climate “health” was China ( 3 ). As is known, China is a fast-developing economy and its GDP (Gross Domestic Product) is expected to be very high by 2050, which is defined as the year of dissolution of the protocol for the decrease in gas emissions.

A more recent international agreement of crucial importance for climate change is the Paris Agreement of 2015, issued by the UNFCCC (United Nations Climate Change Committee). This latest agreement was ratified by a plethora of UN (United Nations) countries as well as the countries of the European Union ( 126 ). In this vein, parties should promote actions and measures to enhance numerous aspects around the subject. Boosting education, training, public awareness, and public participation are some of the relevant actions for maximizing the opportunities to achieve the targets and goals on the crucial matter of climate change and environmental pollution ( 126 ). Without any doubt, technological improvements makes our world easier and it seems difficult to reduce the harmful impact caused by gas emissions, we could limit its use by seeking reliable approaches.

Synopsizing, a global prevention policy should be designed in order to combat anthropogenic air pollution as a complement to the correct handling of the adverse health effects associated with air pollution. Sustainable development practices should be applied, together with information coming from research in order to handle the problem effectively.

At this point, international cooperation in terms of research, development, administration policy, monitoring, and politics is vital for effective pollution control. Legislation concerning air pollution must be aligned and updated, and policy makers should propose the design of a powerful tool of environmental and health protection. As a result, the main proposal of this essay is that we should focus on fostering local structures to promote experience and practice and extrapolate these to the international level through developing effective policies for sustainable management of ecosystems.

Author Contributions

All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication.

Conflict of Interest

IM is employed by the company Delphis S.A. The remaining authors declare that the present review paper was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

REVIEW article

Environmental and health impacts of air pollution: a review.

• 1 Delphis S.A., Kifisia, Greece
• 2 Laboratory of Hygiene and Environmental Protection, Faculty of Medicine, Democritus University of Thrace, Alexandroupolis, Greece
• 3 Centre Hospitalier Universitaire Vaudois (CHUV), Service de Médicine Interne, Lausanne, Switzerland
• 4 School of Social and Political Sciences, University of Glasgow, Glasgow, United Kingdom

One of our era's greatest scourges is air pollution, on account not only of its impact on climate change but also its impact on public and individual health due to increasing morbidity and mortality. There are many pollutants that are major factors in disease in humans. Among them, Particulate Matter (PM), particles of variable but very small diameter, penetrate the respiratory system via inhalation, causing respiratory and cardiovascular diseases, reproductive and central nervous system dysfunctions, and cancer. Despite the fact that ozone in the stratosphere plays a protective role against ultraviolet irradiation, it is harmful when in high concentration at ground level, also affecting the respiratory and cardiovascular system. Furthermore, nitrogen oxide, sulfur dioxide, Volatile Organic Compounds (VOCs), dioxins, and polycyclic aromatic hydrocarbons (PAHs) are all considered air pollutants that are harmful to humans. Carbon monoxide can even provoke direct poisoning when breathed in at high levels. Heavy metals such as lead, when absorbed into the human body, can lead to direct poisoning or chronic intoxication, depending on exposure. Diseases occurring from the aforementioned substances include principally respiratory problems such as Chronic Obstructive Pulmonary Disease (COPD), asthma, bronchiolitis, and also lung cancer, cardiovascular events, central nervous system dysfunctions, and cutaneous diseases. Last but not least, climate change resulting from environmental pollution affects the geographical distribution of many infectious diseases, as do natural disasters. The only way to tackle this problem is through public awareness coupled with a multidisciplinary approach by scientific experts; national and international organizations must address the emergence of this threat and propose sustainable solutions.

Approach to the Problem

The interactions between humans and their physical surroundings have been extensively studied, as multiple human activities influence the environment. The environment is a coupling of the biotic (living organisms and microorganisms) and the abiotic (hydrosphere, lithosphere, and atmosphere).

Pollution is defined as the introduction into the environment of substances harmful to humans and other living organisms. Pollutants are harmful solids, liquids, or gases produced in higher than usual concentrations that reduce the quality of our environment.

Human activities have an adverse effect on the environment by polluting the water we drink, the air we breathe, and the soil in which plants grow. Although the industrial revolution was a great success in terms of technology, society, and the provision of multiple services, it also introduced the production of huge quantities of pollutants emitted into the air that are harmful to human health. Without any doubt, the global environmental pollution is considered an international public health issue with multiple facets. Social, economic, and legislative concerns and lifestyle habits are related to this major problem. Clearly, urbanization and industrialization are reaching unprecedented and upsetting proportions worldwide in our era. Anthropogenic air pollution is one of the biggest public health hazards worldwide, given that it accounts for about 9 million deaths per year ( 1 ).

Without a doubt, all of the aforementioned are closely associated with climate change, and in the event of danger, the consequences can be severe for mankind ( 2 ). Climate changes and the effects of global planetary warming seriously affect multiple ecosystems, causing problems such as food safety issues, ice and iceberg melting, animal extinction, and damage to plants ( 3 , 4 ).

Air pollution has various health effects. The health of susceptible and sensitive individuals can be impacted even on low air pollution days. Short-term exposure to air pollutants is closely related to COPD (Chronic Obstructive Pulmonary Disease), cough, shortness of breath, wheezing, asthma, respiratory disease, and high rates of hospitalization (a measurement of morbidity).

The long-term effects associated with air pollution are chronic asthma, pulmonary insufficiency, cardiovascular diseases, and cardiovascular mortality. According to a Swedish cohort study, diabetes seems to be induced after long-term air pollution exposure ( 5 ). Moreover, air pollution seems to have various malign health effects in early human life, such as respiratory, cardiovascular, mental, and perinatal disorders ( 3 ), leading to infant mortality or chronic disease in adult age ( 6 ).

National reports have mentioned the increased risk of morbidity and mortality ( 1 ). These studies were conducted in many places around the world and show a correlation between daily ranges of particulate matter (PM) concentration and daily mortality. Climate shifts and global planetary warming ( 3 ) could aggravate the situation. Besides, increased hospitalization (an index of morbidity) has been registered among the elderly and susceptible individuals for specific reasons. Fine and ultrafine particulate matter seems to be associated with more serious illnesses ( 6 ), as it can invade the deepest parts of the airways and more easily reach the bloodstream.

Air pollution mainly affects those living in large urban areas, where road emissions contribute the most to the degradation of air quality. There is also a danger of industrial accidents, where the spread of a toxic fog can be fatal to the populations of the surrounding areas. The dispersion of pollutants is determined by many parameters, most notably atmospheric stability and wind ( 6 ).

In developing countries ( 7 ), the problem is more serious due to overpopulation and uncontrolled urbanization along with the development of industrialization. This leads to poor air quality, especially in countries with social disparities and a lack of information on sustainable management of the environment. The use of fuels such as wood fuel or solid fuel for domestic needs due to low incomes exposes people to bad-quality, polluted air at home. It is of note that three billion people around the world are using the above sources of energy for their daily heating and cooking needs ( 8 ). In developing countries, the women of the household seem to carry the highest risk for disease development due to their longer duration exposure to the indoor air pollution ( 8 , 9 ). Due to its fast industrial development and overpopulation, China is one of the Asian countries confronting serious air pollution problems ( 10 , 11 ). The lung cancer mortality observed in China is associated with fine particles ( 12 ). As stated already, long-term exposure is associated with deleterious effects on the cardiovascular system ( 3 , 5 ). However, it is interesting to note that cardiovascular diseases have mostly been observed in developed and high-income countries rather than in the developing low-income countries exposed highly to air pollution ( 13 ). Extreme air pollution is recorded in India, where the air quality reaches hazardous levels. New Delhi is one of the more polluted cities in India. Flights in and out of New Delhi International Airport are often canceled due to the reduced visibility associated with air pollution. Pollution is occurring both in urban and rural areas in India due to the fast industrialization, urbanization, and rise in use of motorcycle transportation. Nevertheless, biomass combustion associated with heating and cooking needs and practices is a major source of household air pollution in India and in Nepal ( 14 , 15 ). There is spatial heterogeneity in India, as areas with diverse climatological conditions and population and education levels generate different indoor air qualities, with higher PM 2.5 observed in North Indian states (557–601 μg/m 3 ) compared to the Southern States (183–214 μg/m 3 ) ( 16 , 17 ). The cold climate of the North Indian areas may be the main reason for this, as longer periods at home and more heating are necessary compared to in the tropical climate of Southern India. Household air pollution in India is associated with major health effects, especially in women and young children, who stay indoors for longer periods. Chronic obstructive respiratory disease (CORD) and lung cancer are mostly observed in women, while acute lower respiratory disease is seen in young children under 5 years of age ( 18 ).

Accumulation of air pollution, especially sulfur dioxide and smoke, reaching 1,500 mg/m3, resulted in an increase in the number of deaths (4,000 deaths) in December 1952 in London and in 1963 in New York City (400 deaths) ( 19 ). An association of pollution with mortality was reported on the basis of monitoring of outdoor pollution in six US metropolitan cities ( 20 ). In every case, it seems that mortality was closely related to the levels of fine, inhalable, and sulfate particles more than with the levels of total particulate pollution, aerosol acidity, sulfur dioxide, or nitrogen dioxide ( 20 ).

Furthermore, extremely high levels of pollution are reported in Mexico City and Rio de Janeiro, followed by Milan, Ankara, Melbourne, Tokyo, and Moscow ( 19 ).

Based on the magnitude of the public health impact, it is certain that different kinds of interventions should be taken into account. Success and effectiveness in controlling air pollution, specifically at the local level, have been reported. Adequate technological means are applied considering the source and the nature of the emission as well as its impact on health and the environment. The importance of point sources and non-point sources of air pollution control is reported by Schwela and Köth-Jahr ( 21 ). Without a doubt, a detailed emission inventory must record all sources in a given area. Beyond considering the above sources and their nature, topography and meteorology should also be considered, as stated previously. Assessment of the control policies and methods is often extrapolated from the local to the regional and then to the global scale. Air pollution may be dispersed and transported from one region to another area located far away. Air pollution management means the reduction to acceptable levels or possible elimination of air pollutants whose presence in the air affects our health or the environmental ecosystem. Private and governmental entities and authorities implement actions to ensure the air quality ( 22 ). Air quality standards and guidelines were adopted for the different pollutants by the WHO and EPA as a tool for the management of air quality ( 1 , 23 ). These standards have to be compared to the emissions inventory standards by causal analysis and dispersion modeling in order to reveal the problematic areas ( 24 ). Inventories are generally based on a combination of direct measurements and emissions modeling ( 24 ).

As an example, we state here the control measures at the source through the use of catalytic converters in cars. These are devices that turn the pollutants and toxic gases produced from combustion engines into less-toxic pollutants by catalysis through redox reactions ( 25 ). In Greece, the use of private cars was restricted by tracking their license plates in order to reduce traffic congestion during rush hour ( 25 ).

Concerning industrial emissions, collectors and closed systems can keep the air pollution to the minimal standards imposed by legislation ( 26 ).

Current strategies to improve air quality require an estimation of the economic value of the benefits gained from proposed programs. These proposed programs by public authorities, and directives are issued with guidelines to be respected.

In Europe, air quality limit values AQLVs (Air Quality Limit Values) are issued for setting off planning claims ( 27 ). In the USA, the NAAQS (National Ambient Air Quality Standards) establish the national air quality limit values ( 27 ). While both standards and directives are based on different mechanisms, significant success has been achieved in the reduction of overall emissions and associated health and environmental effects ( 27 ). The European Directive identifies geographical areas of risk exposure as monitoring/assessment zones to record the emission sources and levels of air pollution ( 27 ), whereas the USA establishes global geographical air quality criteria according to the severity of their air quality problem and records all sources of the pollutants and their precursors ( 27 ).

In this vein, funds have been financing, directly or indirectly, projects related to air quality along with the technical infrastructure to maintain good air quality. These plans focus on an inventory of databases from air quality environmental planning awareness campaigns. Moreover, pollution measures of air emissions may be taken for vehicles, machines, and industries in urban areas.

Technological innovation can only be successful if it is able to meet the needs of society. In this sense, technology must reflect the decision-making practices and procedures of those involved in risk assessment and evaluation and act as a facilitator in providing information and assessments to enable decision makers to make the best decisions possible. Summarizing the aforementioned in order to design an effective air quality control strategy, several aspects must be considered: environmental factors and ambient air quality conditions, engineering factors and air pollutant characteristics, and finally, economic operating costs for technological improvement and administrative and legal costs. Considering the economic factor, competitiveness through neoliberal concepts is offering a solution to environmental problems ( 22 ).

The development of environmental governance, along with technological progress, has initiated the deployment of a dialogue. Environmental politics has created objections and points of opposition between different political parties, scientists, media, and governmental and non-governmental organizations ( 22 ). Radical environmental activism actions and movements have been created ( 22 ). The rise of the new information and communication technologies (ICTs) are many times examined as to whether and in which way they have influenced means of communication and social movements such as activism ( 28 ). Since the 1990s, the term “digital activism” has been used increasingly and in many different disciplines ( 29 ). Nowadays, multiple digital technologies can be used to produce a digital activism outcome on environmental issues. More specifically, devices with online capabilities such as computers or mobile phones are being used as a way to pursue change in political and social affairs ( 30 ).

In the present paper, we focus on the sources of environmental pollution in relation to public health and propose some solutions and interventions that may be of interest to environmental legislators and decision makers.

Sources of Exposure

It is known that the majority of environmental pollutants are emitted through large-scale human activities such as the use of industrial machinery, power-producing stations, combustion engines, and cars. Because these activities are performed at such a large scale, they are by far the major contributors to air pollution, with cars estimated to be responsible for approximately 80% of today's pollution ( 31 ). Some other human activities are also influencing our environment to a lesser extent, such as field cultivation techniques, gas stations, fuel tanks heaters, and cleaning procedures ( 32 ), as well as several natural sources, such as volcanic and soil eruptions and forest fires.

The classification of air pollutants is based mainly on the sources producing pollution. Therefore, it is worth mentioning the four main sources, following the classification system: Major sources, Area sources, Mobile sources, and Natural sources.

Major sources include the emission of pollutants from power stations, refineries, and petrochemicals, the chemical and fertilizer industries, metallurgical and other industrial plants, and, finally, municipal incineration.

Indoor area sources include domestic cleaning activities, dry cleaners, printing shops, and petrol stations.

Mobile sources include automobiles, cars, railways, airways, and other types of vehicles.

Finally, natural sources include, as stated previously, physical disasters ( 33 ) such as forest fire, volcanic erosion, dust storms, and agricultural burning.

However, many classification systems have been proposed. Another type of classification is a grouping according to the recipient of the pollution, as follows:

Air pollution is determined as the presence of pollutants in the air in large quantities for long periods. Air pollutants are dispersed particles, hydrocarbons, CO, CO 2 , NO, NO 2 , SO 3 , etc.

Water pollution is organic and inorganic charge and biological charge ( 10 ) at high levels that affect the water quality ( 34 , 35 ).

Soil pollution occurs through the release of chemicals or the disposal of wastes, such as heavy metals, hydrocarbons, and pesticides.

Air pollution can influence the quality of soil and water bodies by polluting precipitation, falling into water and soil environments ( 34 , 36 ). Notably, the chemistry of the soil can be amended due to acid precipitation by affecting plants, cultures, and water quality ( 37 ). Moreover, movement of heavy metals is favored by soil acidity, and metals are so then moving into the watery environment. It is known that heavy metals such as aluminum are noxious to wildlife and fishes. Soil quality seems to be of importance, as soils with low calcium carbonate levels are at increased jeopardy from acid rain. Over and above rain, snow and particulate matter drip into watery ' bodies ( 36 , 38 ).

Lastly, pollution is classified following type of origin:

Radioactive and nuclear pollution , releasing radioactive and nuclear pollutants into water, air, and soil during nuclear explosions and accidents, from nuclear weapons, and through handling or disposal of radioactive sewage.

Radioactive materials can contaminate surface water bodies and, being noxious to the environment, plants, animals, and humans. It is known that several radioactive substances such as radium and uranium concentrate in the bones and can cause cancers ( 38 , 39 ).

Noise pollution is produced by machines, vehicles, traffic noises, and musical installations that are harmful to our hearing.

The World Health Organization introduced the term DALYs. The DALYs for a disease or health condition is defined as the sum of the Years of Life Lost (YLL) due to premature mortality in the population and the Years Lost due to Disability (YLD) for people living with the health condition or its consequences ( 39 ). In Europe, air pollution is the main cause of disability-adjusted life years lost (DALYs), followed by noise pollution. The potential relationships of noise and air pollution with health have been studied ( 40 ). The study found that DALYs related to noise were more important than those related to air pollution, as the effects of environmental noise on cardiovascular disease were independent of air pollution ( 40 ). Environmental noise should be counted as an independent public health risk ( 40 ).

Environmental pollution occurs when changes in the physical, chemical, or biological constituents of the environment (air masses, temperature, climate, etc.) are produced.

Pollutants harm our environment either by increasing levels above normal or by introducing harmful toxic substances. Primary pollutants are directly produced from the above sources, and secondary pollutants are emitted as by-products of the primary ones. Pollutants can be biodegradable or non-biodegradable and of natural origin or anthropogenic, as stated previously. Moreover, their origin can be a unique source (point-source) or dispersed sources.

Pollutants have differences in physical and chemical properties, explaining the discrepancy in their capacity for producing toxic effects. As an example, we state here that aerosol compounds ( 41 – 43 ) have a greater toxicity than gaseous compounds due to their tiny size (solid or liquid) in the atmosphere; they have a greater penetration capacity. Gaseous compounds are eliminated more easily by our respiratory system ( 41 ). These particles are able to damage lungs and can even enter the bloodstream ( 41 ), leading to the premature deaths of millions of people yearly. Moreover, the aerosol acidity ([H+]) seems to considerably enhance the production of secondary organic aerosols (SOA), but this last aspect is not supported by other scientific teams ( 38 ).

Climate and Pollution

Air pollution and climate change are closely related. Climate is the other side of the same coin that reduces the quality of our Earth ( 44 ). Pollutants such as black carbon, methane, tropospheric ozone, and aerosols affect the amount of incoming sunlight. As a result, the temperature of the Earth is increasing, resulting in the melting of ice, icebergs, and glaciers.

In this vein, climatic changes will affect the incidence and prevalence of both residual and imported infections in Europe. Climate and weather affect the duration, timing, and intensity of outbreaks strongly and change the map of infectious diseases in the globe ( 45 ). Mosquito-transmitted parasitic or viral diseases are extremely climate-sensitive, as warming firstly shortens the pathogen incubation period and secondly shifts the geographic map of the vector. Similarly, water-warming following climate changes leads to a high incidence of waterborne infections. Recently, in Europe, eradicated diseases seem to be emerging due to the migration of population, for example, cholera, poliomyelitis, tick-borne encephalitis, and malaria ( 46 ).

The spread of epidemics is associated with natural climate disasters and storms, which seem to occur more frequently nowadays ( 47 ). Malnutrition and disequilibration of the immune system are also associated with the emerging infections affecting public health ( 48 ).

The Chikungunya virus “took the airplane” from the Indian Ocean to Europe, as outbreaks of the disease were registered in Italy ( 49 ) as well as autochthonous cases in France ( 50 ).

An increase in cryptosporidiosis in the United Kingdom and in the Czech Republic seems to have occurred following flooding ( 36 , 51 ).

As stated previously, aerosols compounds are tiny in size and considerably affect the climate. They are able to dissipate sunlight (the albedo phenomenon) by dispersing a quarter of the sun's rays back to space and have cooled the global temperature over the last 30 years ( 52 ).

Air Pollutants

The World Health Organization (WHO) reports on six major air pollutants, namely particle pollution, ground-level ozone, carbon monoxide, sulfur oxides, nitrogen oxides, and lead. Air pollution can have a disastrous effect on all components of the environment, including groundwater, soil, and air. Additionally, it poses a serious threat to living organisms. In this vein, our interest is mainly to focus on these pollutants, as they are related to more extensive and severe problems in human health and environmental impact. Acid rain, global warming, the greenhouse effect, and climate changes have an important ecological impact on air pollution ( 53 ).

Particulate Matter (PM) and Health

Studies have shown a relationship between particulate matter (PM) and adverse health effects, focusing on either short-term (acute) or long-term (chronic) PM exposure.

Particulate matter (PM) is usually formed in the atmosphere as a result of chemical reactions between the different pollutants. The penetration of particles is closely dependent on their size ( 53 ). Particulate Matter (PM) was defined as a term for particles by the United States Environmental Protection Agency ( 54 ). Particulate matter (PM) pollution includes particles with diameters of 10 micrometers (μm) or smaller, called PM 10 , and extremely fine particles with diameters that are generally 2.5 micrometers (μm) and smaller.

Particulate matter contains tiny liquid or solid droplets that can be inhaled and cause serious health effects ( 55 ). Particles <10 μm in diameter (PM 10 ) after inhalation can invade the lungs and even reach the bloodstream. Fine particles, PM 2.5 , pose a greater risk to health ( 6 , 56 ) ( Table 1 ).

Table 1 . Penetrability according to particle size.

Multiple epidemiological studies have been performed on the health effects of PM. A positive relation was shown between both short-term and long-term exposures of PM 2.5 and acute nasopharyngitis ( 56 ). In addition, long-term exposure to PM for years was found to be related to cardiovascular diseases and infant mortality.

Those studies depend on PM 2.5 monitors and are restricted in terms of study area or city area due to a lack of spatially resolved daily PM 2.5 concentration data and, in this way, are not representative of the entire population. Following a recent epidemiological study by the Department of Environmental Health at Harvard School of Public Health (Boston, MA) ( 57 ), it was reported that, as PM 2.5 concentrations vary spatially, an exposure error (Berkson error) seems to be produced, and the relative magnitudes of the short- and long-term effects are not yet completely elucidated. The team developed a PM 2.5 exposure model based on remote sensing data for assessing short- and long-term human exposures ( 57 ). This model permits spatial resolution in short-term effects plus the assessment of long-term effects in the whole population.

Moreover, respiratory diseases and affection of the immune system are registered as long-term chronic effects ( 58 ). It is worth noting that people with asthma, pneumonia, diabetes, and respiratory and cardiovascular diseases are especially susceptible and vulnerable to the effects of PM. PM 2.5 , followed by PM 10 , are strongly associated with diverse respiratory system diseases ( 59 ), as their size permits them to pierce interior spaces ( 60 ). The particles produce toxic effects according to their chemical and physical properties. The components of PM 10 and PM 2.5 can be organic (polycyclic aromatic hydrocarbons, dioxins, benzene, 1-3 butadiene) or inorganic (carbon, chlorides, nitrates, sulfates, metals) in nature ( 55 ).

Particulate Matter (PM) is divided into four main categories according to type and size ( 61 ) ( Table 2 ).

Table 2 . Types and sizes of particulate Matter (PM).

Gas contaminants include PM in aerial masses.

Particulate contaminants include contaminants such as smog, soot, tobacco smoke, oil smoke, fly ash, and cement dust.

Biological Contaminants are microorganisms (bacteria, viruses, fungi, mold, and bacterial spores), cat allergens, house dust and allergens, and pollen.

Types of Dust include suspended atmospheric dust, settling dust, and heavy dust.

Finally, another fact is that the half-lives of PM 10 and PM 2.5 particles in the atmosphere is extended due to their tiny dimensions; this permits their long-lasting suspension in the atmosphere and even their transfer and spread to distant destinations where people and the environment may be exposed to the same magnitude of pollution ( 53 ). They are able to change the nutrient balance in watery ecosystems, damage forests and crops, and acidify water bodies.

As stated, PM 2.5 , due to their tiny size, are causing more serious health effects. These aforementioned fine particles are the main cause of the “haze” formation in different metropolitan areas ( 12 , 13 , 61 ).

Ozone Impact in the Atmosphere

Ozone (O 3 ) is a gas formed from oxygen under high voltage electric discharge ( 62 ). It is a strong oxidant, 52% stronger than chlorine. It arises in the stratosphere, but it could also arise following chain reactions of photochemical smog in the troposphere ( 63 ).

Ozone can travel to distant areas from its initial source, moving with air masses ( 64 ). It is surprising that ozone levels over cities are low in contrast to the increased amounts occuring in urban areas, which could become harmful for cultures, forests, and vegetation ( 65 ) as it is reducing carbon assimilation ( 66 ). Ozone reduces growth and yield ( 47 , 48 ) and affects the plant microflora due to its antimicrobial capacity ( 67 , 68 ). In this regard, ozone acts upon other natural ecosystems, with microflora ( 69 , 70 ) and animal species changing their species composition ( 71 ). Ozone increases DNA damage in epidermal keratinocytes and leads to impaired cellular function ( 72 ).

Ground-level ozone (GLO) is generated through a chemical reaction between oxides of nitrogen and VOCs emitted from natural sources and/or following anthropogenic activities.

Ozone uptake usually occurs by inhalation. Ozone affects the upper layers of the skin and the tear ducts ( 73 ). A study of short-term exposure of mice to high levels of ozone showed malondialdehyde formation in the upper skin (epidermis) but also depletion in vitamins C and E. It is likely that ozone levels are not interfering with the skin barrier function and integrity to predispose to skin disease ( 74 ).

Due to the low water-solubility of ozone, inhaled ozone has the capacity to penetrate deeply into the lungs ( 75 ).

Toxic effects induced by ozone are registered in urban areas all over the world, causing biochemical, morphologic, functional, and immunological disorders ( 76 ).

The European project (APHEA2) focuses on the acute effects of ambient ozone concentrations on mortality ( 77 ). Daily ozone concentrations compared to the daily number of deaths were reported from different European cities for a 3-year period. During the warm period of the year, an observed increase in ozone concentration was associated with an increase in the daily number of deaths (0.33%), in the number of respiratory deaths (1.13%), and in the number of cardiovascular deaths (0.45%). No effect was observed during wintertime.

Carbon Monoxide (CO)

Carbon monoxide is produced by fossil fuel when combustion is incomplete. The symptoms of poisoning due to inhaling carbon monoxide include headache, dizziness, weakness, nausea, vomiting, and, finally, loss of consciousness.

The affinity of carbon monoxide to hemoglobin is much greater than that of oxygen. In this vein, serious poisoning may occur in people exposed to high levels of carbon monoxide for a long period of time. Due to the loss of oxygen as a result of the competitive binding of carbon monoxide, hypoxia, ischemia, and cardiovascular disease are observed.

Carbon monoxide affects the greenhouses gases that are tightly connected to global warming and climate. This should lead to an increase in soil and water temperatures, and extreme weather conditions or storms may occur ( 68 ).

However, in laboratory and field experiments, it has been seen to produce increased plant growth ( 78 ).

Nitrogen Oxide (NO 2 )

Nitrogen oxide is a traffic-related pollutant, as it is emitted from automobile motor engines ( 79 , 80 ). It is an irritant of the respiratory system as it penetrates deep in the lung, inducing respiratory diseases, coughing, wheezing, dyspnea, bronchospasm, and even pulmonary edema when inhaled at high levels. It seems that concentrations over 0.2 ppm produce these adverse effects in humans, while concentrations higher than 2.0 ppm affect T-lymphocytes, particularly the CD8+ cells and NK cells that produce our immune response ( 81 ).It is reported that long-term exposure to high levels of nitrogen dioxide can be responsible for chronic lung disease. Long-term exposure to NO 2 can impair the sense of smell ( 81 ).

However, systems other than respiratory ones can be involved, as symptoms such as eye, throat, and nose irritation have been registered ( 81 ).

High levels of nitrogen dioxide are deleterious to crops and vegetation, as they have been observed to reduce crop yield and plant growth efficiency. Moreover, NO 2 can reduce visibility and discolor fabrics ( 81 ).

Sulfur Dioxide (SO 2 )

Sulfur dioxide is a harmful gas that is emitted mainly from fossil fuel consumption or industrial activities. The annual standard for SO 2 is 0.03 ppm ( 82 ). It affects human, animal, and plant life. Susceptible people as those with lung disease, old people, and children, who present a higher risk of damage. The major health problems associated with sulfur dioxide emissions in industrialized areas are respiratory irritation, bronchitis, mucus production, and bronchospasm, as it is a sensory irritant and penetrates deep into the lung converted into bisulfite and interacting with sensory receptors, causing bronchoconstriction. Moreover, skin redness, damage to the eyes (lacrimation and corneal opacity) and mucous membranes, and worsening of pre-existing cardiovascular disease have been observed ( 81 ).

Environmental adverse effects, such as acidification of soil and acid rain, seem to be associated with sulfur dioxide emissions ( 83 ).

Lead is a heavy metal used in different industrial plants and emitted from some petrol motor engines, batteries, radiators, waste incinerators, and waste waters ( 84 ).

Moreover, major sources of lead pollution in the air are metals, ore, and piston-engine aircraft. Lead poisoning is a threat to public health due to its deleterious effects upon humans, animals, and the environment, especially in the developing countries.

Exposure to lead can occur through inhalation, ingestion, and dermal absorption. Trans- placental transport of lead was also reported, as lead passes through the placenta unencumbered ( 85 ). The younger the fetus is, the more harmful the toxic effects. Lead toxicity affects the fetal nervous system; edema or swelling of the brain is observed ( 86 ). Lead, when inhaled, accumulates in the blood, soft tissue, liver, lung, bones, and cardiovascular, nervous, and reproductive systems. Moreover, loss of concentration and memory, as well as muscle and joint pain, were observed in adults ( 85 , 86 ).

Children and newborns ( 87 ) are extremely susceptible even to minimal doses of lead, as it is a neurotoxicant and causes learning disabilities, impairment of memory, hyperactivity, and even mental retardation.

Elevated amounts of lead in the environment are harmful to plants and crop growth. Neurological effects are observed in vertebrates and animals in association with high lead levels ( 88 ).

Polycyclic Aromatic Hydrocarbons(PAHs)

The distribution of PAHs is ubiquitous in the environment, as the atmosphere is the most important means of their dispersal. They are found in coal and in tar sediments. Moreover, they are generated through incomplete combustion of organic matter as in the cases of forest fires, incineration, and engines ( 89 ). PAH compounds, such as benzopyrene, acenaphthylene, anthracene, and fluoranthene are recognized as toxic, mutagenic, and carcinogenic substances. They are an important risk factor for lung cancer ( 89 ).

Volatile Organic Compounds(VOCs)

Volatile organic compounds (VOCs), such as toluene, benzene, ethylbenzene, and xylene ( 90 ), have been found to be associated with cancer in humans ( 91 ). The use of new products and materials has actually resulted in increased concentrations of VOCs. VOCs pollute indoor air ( 90 ) and may have adverse effects on human health ( 91 ). Short-term and long-term adverse effects on human health are observed. VOCs are responsible for indoor air smells. Short-term exposure is found to cause irritation of eyes, nose, throat, and mucosal membranes, while those of long duration exposure include toxic reactions ( 92 ). Predictable assessment of the toxic effects of complex VOC mixtures is difficult to estimate, as these pollutants can have synergic, antagonistic, or indifferent effects ( 91 , 93 ).

Dioxins originate from industrial processes but also come from natural processes, such as forest fires and volcanic eruptions. They accumulate in foods such as meat and dairy products, fish and shellfish, and especially in the fatty tissue of animals ( 94 ).

Short-period exhibition to high dioxin concentrations may result in dark spots and lesions on the skin ( 94 ). Long-term exposure to dioxins can cause developmental problems, impairment of the immune, endocrine and nervous systems, reproductive infertility, and cancer ( 94 ).

Without any doubt, fossil fuel consumption is responsible for a sizeable part of air contamination. This contamination may be anthropogenic, as in agricultural and industrial processes or transportation, while contamination from natural sources is also possible. Interestingly, it is of note that the air quality standards established through the European Air Quality Directive are somewhat looser than the WHO guidelines, which are stricter ( 95 ).

Effect of Air Pollution on Health

The most common air pollutants are ground-level ozone and Particulates Matter (PM). Air pollution is distinguished into two main types:

Outdoor pollution is the ambient air pollution.

Indoor pollution is the pollution generated by household combustion of fuels.

People exposed to high concentrations of air pollutants experience disease symptoms and states of greater and lesser seriousness. These effects are grouped into short- and long-term effects affecting health.

Susceptible populations that need to be aware of health protection measures include old people, children, and people with diabetes and predisposing heart or lung disease, especially asthma.

As extensively stated previously, according to a recent epidemiological study from Harvard School of Public Health, the relative magnitudes of the short- and long-term effects have not been completely clarified ( 57 ) due to the different epidemiological methodologies and to the exposure errors. New models are proposed for assessing short- and long-term human exposure data more successfully ( 57 ). Thus, in the present section, we report the more common short- and long-term health effects but also general concerns for both types of effects, as these effects are often dependent on environmental conditions, dose, and individual susceptibility.

Short-term effects are temporary and range from simple discomfort, such as irritation of the eyes, nose, skin, throat, wheezing, coughing and chest tightness, and breathing difficulties, to more serious states, such as asthma, pneumonia, bronchitis, and lung and heart problems. Short-term exposure to air pollution can also cause headaches, nausea, and dizziness.

These problems can be aggravated by extended long-term exposure to the pollutants, which is harmful to the neurological, reproductive, and respiratory systems and causes cancer and even, rarely, deaths.

The long-term effects are chronic, lasting for years or the whole life and can even lead to death. Furthermore, the toxicity of several air pollutants may also induce a variety of cancers in the long term ( 96 ).

As stated already, respiratory disorders are closely associated with the inhalation of air pollutants. These pollutants will invade through the airways and will accumulate at the cells. Damage to target cells should be related to the pollutant component involved and its source and dose. Health effects are also closely dependent on country, area, season, and time. An extended exposure duration to the pollutant should incline to long-term health effects in relation also to the above factors.

Particulate Matter (PMs), dust, benzene, and O 3 cause serious damage to the respiratory system ( 97 ). Moreover, there is a supplementary risk in case of existing respiratory disease such as asthma ( 98 ). Long-term effects are more frequent in people with a predisposing disease state. When the trachea is contaminated by pollutants, voice alterations may be remarked after acute exposure. Chronic obstructive pulmonary disease (COPD) may be induced following air pollution, increasing morbidity and mortality ( 99 ). Long-term effects from traffic, industrial air pollution, and combustion of fuels are the major factors for COPD risk ( 99 ).

Multiple cardiovascular effects have been observed after exposure to air pollutants ( 100 ). Changes occurred in blood cells after long-term exposure may affect cardiac functionality. Coronary arteriosclerosis was reported following long-term exposure to traffic emissions ( 101 ), while short-term exposure is related to hypertension, stroke, myocardial infracts, and heart insufficiency. Ventricle hypertrophy is reported to occur in humans after long-time exposure to nitrogen oxide (NO 2 ) ( 102 , 103 ).

Neurological effects have been observed in adults and children after extended-term exposure to air pollutants.

Psychological complications, autism, retinopathy, fetal growth, and low birth weight seem to be related to long-term air pollution ( 83 ). The etiologic agent of the neurodegenerative diseases (Alzheimer's and Parkinson's) is not yet known, although it is believed that extended exposure to air pollution seems to be a factor. Specifically, pesticides and metals are cited as etiological factors, together with diet. The mechanisms in the development of neurodegenerative disease include oxidative stress, protein aggregation, inflammation, and mitochondrial impairment in neurons ( 104 ) ( Figure 1 ).

Figure 1 . Impact of air pollutants on the brain.

Brain inflammation was observed in dogs living in a highly polluted area in Mexico for a long period ( 105 ). In human adults, markers of systemic inflammation (IL-6 and fibrinogen) were found to be increased as an immediate response to PNC on the IL-6 level, possibly leading to the production of acute-phase proteins ( 106 ). The progression of atherosclerosis and oxidative stress seem to be the mechanisms involved in the neurological disturbances caused by long-term air pollution. Inflammation comes secondary to the oxidative stress and seems to be involved in the impairment of developmental maturation, affecting multiple organs ( 105 , 107 ). Similarly, other factors seem to be involved in the developmental maturation, which define the vulnerability to long-term air pollution. These include birthweight, maternal smoking, genetic background and socioeconomic environment, as well as education level.

However, diet, starting from breast-feeding, is another determinant factor. Diet is the main source of antioxidants, which play a key role in our protection against air pollutants ( 108 ). Antioxidants are free radical scavengers and limit the interaction of free radicals in the brain ( 108 ). Similarly, genetic background may result in a differential susceptibility toward the oxidative stress pathway ( 60 ). For example, antioxidant supplementation with vitamins C and E appears to modulate the effect of ozone in asthmatic children homozygous for the GSTM1 null allele ( 61 ). Inflammatory cytokines released in the periphery (e.g., respiratory epithelia) upregulate the innate immune Toll-like receptor 2. Such activation and the subsequent events leading to neurodegeneration have recently been observed in lung lavage in mice exposed to ambient Los Angeles (CA, USA) particulate matter ( 61 ). In children, neurodevelopmental morbidities were observed after lead exposure. These children developed aggressive and delinquent behavior, reduced intelligence, learning difficulties, and hyperactivity ( 109 ). No level of lead exposure seems to be “safe,” and the scientific community has asked the Centers for Disease Control and Prevention (CDC) to reduce the current screening guideline of 10 μg/dl ( 109 ).

It is important to state that impact on the immune system, causing dysfunction and neuroinflammation ( 104 ), is related to poor air quality. Yet, increases in serum levels of immunoglobulins (IgA, IgM) and the complement component C3 are observed ( 106 ). Another issue is that antigen presentation is affected by air pollutants, as there is an upregulation of costimulatory molecules such as CD80 and CD86 on macrophages ( 110 ).

As is known, skin is our shield against ultraviolet radiation (UVR) and other pollutants, as it is the most exterior layer of our body. Traffic-related pollutants, such as PAHs, VOCs, oxides, and PM, may cause pigmented spots on our skin ( 111 ). On the one hand, as already stated, when pollutants penetrate through the skin or are inhaled, damage to the organs is observed, as some of these pollutants are mutagenic and carcinogenic, and, specifically, they affect the liver and lung. On the other hand, air pollutants (and those in the troposphere) reduce the adverse effects of ultraviolet radiation UVR in polluted urban areas ( 111 ). Air pollutants absorbed by the human skin may contribute to skin aging, psoriasis, acne, urticaria, eczema, and atopic dermatitis ( 111 ), usually caused by exposure to oxides and photochemical smoke ( 111 ). Exposure to PM and cigarette smoking act as skin-aging agents, causing spots, dyschromia, and wrinkles. Lastly, pollutants have been associated with skin cancer ( 111 ).

Higher morbidity is reported to fetuses and children when exposed to the above dangers. Impairment in fetal growth, low birth weight, and autism have been reported ( 112 ).

Another exterior organ that may be affected is the eye. Contamination usually comes from suspended pollutants and may result in asymptomatic eye outcomes, irritation ( 112 ), retinopathy, or dry eye syndrome ( 113 , 114 ).

Environmental Impact of Air Pollution

Air pollution is harming not only human health but also the environment ( 115 ) in which we live. The most important environmental effects are as follows.

Acid rain is wet (rain, fog, snow) or dry (particulates and gas) precipitation containing toxic amounts of nitric and sulfuric acids. They are able to acidify the water and soil environments, damage trees and plantations, and even damage buildings and outdoor sculptures, constructions, and statues.

Haze is produced when fine particles are dispersed in the air and reduce the transparency of the atmosphere. It is caused by gas emissions in the air coming from industrial facilities, power plants, automobiles, and trucks.

Ozone , as discussed previously, occurs both at ground level and in the upper level (stratosphere) of the Earth's atmosphere. Stratospheric ozone is protecting us from the Sun's harmful ultraviolet (UV) rays. In contrast, ground-level ozone is harmful to human health and is a pollutant. Unfortunately, stratospheric ozone is gradually damaged by ozone-depleting substances (i.e., chemicals, pesticides, and aerosols). If this protecting stratospheric ozone layer is thinned, then UV radiation can reach our Earth, with harmful effects for human life (skin cancer) ( 116 ) and crops ( 117 ). In plants, ozone penetrates through the stomata, inducing them to close, which blocks CO 2 transfer and induces a reduction in photosynthesis ( 118 ).

Global climate change is an important issue that concerns mankind. As is known, the “greenhouse effect” keeps the Earth's temperature stable. Unhappily, anthropogenic activities have destroyed this protecting temperature effect by producing large amounts of greenhouse gases, and global warming is mounting, with harmful effects on human health, animals, forests, wildlife, agriculture, and the water environment. A report states that global warming is adding to the health risks of poor people ( 119 ).

People living in poorly constructed buildings in warm-climate countries are at high risk for heat-related health problems as temperatures mount ( 119 ).

Wildlife is burdened by toxic pollutants coming from the air, soil, or the water ecosystem and, in this way, animals can develop health problems when exposed to high levels of pollutants. Reproductive failure and birth effects have been reported.

Eutrophication is occurring when elevated concentrations of nutrients (especially nitrogen) stimulate the blooming of aquatic algae, which can cause a disequilibration in the diversity of fish and their deaths.

Without a doubt, there is a critical concentration of pollution that an ecosystem can tolerate without being destroyed, which is associated with the ecosystem's capacity to neutralize acidity. The Canada Acid Rain Program established this load at 20 kg/ha/yr ( 120 ).

Hence, air pollution has deleterious effects on both soil and water ( 121 ). Concerning PM as an air pollutant, its impact on crop yield and food productivity has been reported. Its impact on watery bodies is associated with the survival of living organisms and fishes and their productivity potential ( 121 ).

An impairment in photosynthetic rhythm and metabolism is observed in plants exposed to the effects of ozone ( 121 ).

Sulfur and nitrogen oxides are involved in the formation of acid rain and are harmful to plants and marine organisms.

Last but not least, as mentioned above, the toxicity associated with lead and other metals is the main threat to our ecosystems (air, water, and soil) and living creatures ( 121 ).

In 2018, during the first WHO Global Conference on Air Pollution and Health, the WHO's General Director, Dr. Tedros Adhanom Ghebreyesus, called air pollution a “silent public health emergency” and “the new tobacco” ( 122 ).

Undoubtedly, children are particularly vulnerable to air pollution, especially during their development. Air pollution has adverse effects on our lives in many different respects.

Diseases associated with air pollution have not only an important economic impact but also a societal impact due to absences from productive work and school.

Despite the difficulty of eradicating the problem of anthropogenic environmental pollution, a successful solution could be envisaged as a tight collaboration of authorities, bodies, and doctors to regularize the situation. Governments should spread sufficient information and educate people and should involve professionals in these issues so as to control the emergence of the problem successfully.

Technologies to reduce air pollution at the source must be established and should be used in all industries and power plants. The Kyoto Protocol of 1997 set as a major target the reduction of GHG emissions to below 5% by 2012 ( 123 ). This was followed by the Copenhagen summit, 2009 ( 124 ), and then the Durban summit of 2011 ( 125 ), where it was decided to keep to the same line of action. The Kyoto protocol and the subsequent ones were ratified by many countries. Among the pioneers who adopted this important protocol for the world's environmental and climate “health” was China ( 3 ). As is known, China is a fast-developing economy and its GDP (Gross Domestic Product) is expected to be very high by 2050, which is defined as the year of dissolution of the protocol for the decrease in gas emissions.

A more recent international agreement of crucial importance for climate change is the Paris Agreement of 2015, issued by the UNFCCC (United Nations Climate Change Committee). This latest agreement was ratified by a plethora of UN (United Nations) countries as well as the countries of the European Union ( 126 ). In this vein, parties should promote actions and measures to enhance numerous aspects around the subject. Boosting education, training, public awareness, and public participation are some of the relevant actions for maximizing the opportunities to achieve the targets and goals on the crucial matter of climate change and environmental pollution ( 126 ). Without any doubt, technological improvements makes our world easier and it seems difficult to reduce the harmful impact caused by gas emissions, we could limit its use by seeking reliable approaches.

Synopsizing, a global prevention policy should be designed in order to combat anthropogenic air pollution as a complement to the correct handling of the adverse health effects associated with air pollution. Sustainable development practices should be applied, together with information coming from research in order to handle the problem effectively.

At this point, international cooperation in terms of research, development, administration policy, monitoring, and politics is vital for effective pollution control. Legislation concerning air pollution must be aligned and updated, and policy makers should propose the design of a powerful tool of environmental and health protection. As a result, the main proposal of this essay is that we should focus on fostering local structures to promote experience and practice and extrapolate these to the international level through developing effective policies for sustainable management of ecosystems.

Author Contributions

All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication.

Conflict of Interest

IM is employed by the company Delphis S.A.

The remaining authors declare that the present review paper was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

1. WHO. Air Pollution . WHO. Available online at: http://www.who.int/airpollution/en/ (accessed October 5, 2019).

Google Scholar

2. Moores FC. Climate change and air pollution: exploring the synergies and potential for mitigation in industrializing countries. Sustainability . (2009) 1:43–54. doi: 10.3390/su1010043

CrossRef Full Text | Google Scholar

3. USGCRP (2009). Global Climate Change Impacts in the United States. In: Karl TR, Melillo JM, Peterson TC, editors. Climate Change Impacts by Sectors: Ecosystems . New York, NY: United States Global Change Research Program. Cambridge University Press.

4. Marlon JR, Bloodhart B, Ballew MT, Rolfe-Redding J, Roser-Renouf C, Leiserowitz A, et al. (2019). How hope and doubt affect climate change mobilization. Front. Commun. 4:20. doi: 10.3389/fcomm.2019.00020

5. Eze IC, Schaffner E, Fischer E, Schikowski T, Adam M, Imboden M, et al. Long- term air pollution exposure and diabetes in a population-based Swiss cohort. Environ Int . (2014) 70:95–105. doi: 10.1016/j.envint.2014.05.014

PubMed Abstract | CrossRef Full Text | Google Scholar

6. Kelishadi R, Poursafa P. Air pollution and non-respiratory health hazards for children. Arch Med Sci . (2010) 6:483–95. doi: 10.5114/aoms.2010.14458

7. Manucci PM, Franchini M. Health effects of ambient air pollution in developing countries. Int J Environ Res Public Health . (2017) 14:1048. doi: 10.3390/ijerph14091048

8. Burden of Disease from Ambient and Household Air Pollution . Available online: http://who.int/phe/health_topics/outdoorair/databases/en/ (accessed August 15, 2017).

9. Hashim D, Boffetta P. Occupational and environmental exposures and cancers in developing countries. Ann Glob Health . (2014) 80:393–411. doi: 10.1016/j.aogh.2014.10.002

10. Guo Y, Zeng H, Zheng R, Li S, Pereira G, Liu Q, et al. The burden of lung cancer mortality attributable to fine particles in China. Total Environ Sci . (2017) 579:1460–6. doi: 10.1016/j.scitotenv.2016.11.147

11. Hou Q, An XQ, Wang Y, Guo JP. An evaluation of resident exposure to respirable particulate matter and health economic loss in Beijing during Beijing 2008 Olympic Games. Sci Total Environ . (2010) 408:4026–32. doi: 10.1016/j.scitotenv.2009.12.030

12. Kan H, Chen R, Tong S. Ambient air pollution, climate change, and population health in China. Environ Int . (2012) 42:10–9. doi: 10.1016/j.envint.2011.03.003

13. Burroughs Peña MS, Rollins A. Environmental exposures and cardiovascular disease: a challenge for health and development in low- and middle-income countries. Cardiol Clin . (2017) 35:71–86. doi: 10.1016/j.ccl.2016.09.001

14. Kankaria A, Nongkynrih B, Gupta S. Indoor air pollution in india: implications on health and its control. Indian J Comm Med . 39:203–7. doi: 10.4103/0970-0218.143019

15. Parajuli I, Lee H, Shrestha KR. Indoor air quality and ventilation assessment of rural mountainous households of Nepal. Int J Sust Built Env . (2016) 5:301–11. doi: 10.1016/j.ijsbe.2016.08.003

16. Saud T, Gautam R, Mandal TK, Gadi R, Singh DP, Sharma SK. Emission estimates of organic and elemental carbon from household biomass fuel used over the Indo-Gangetic Plain (IGP), India. Atmos Environ . (2012) 61:212–20. doi: 10.1016/j.atmosenv.2012.07.030

17. Singh DP, Gadi R, Mandal TK, Saud T, Saxena M, Sharma SK. Emissions estimates of PAH from biomass fuels used in rural sector of Indo-Gangetic Plains of India. Atmos Environ . (2013) 68:120–6. doi: 10.1016/j.atmosenv.2012.11.042

18. Dherani M, Pope D, Mascarenhas M, Smith KR, Weber M BN. Indoor air pollution from unprocessed solid fuel use and pneumonia risk in children aged under five years: a systematic review and meta-analysis. Bull World Health Organ . (2008) 86:390–4. doi: 10.2471/BLT.07.044529

19. Kassomenos P, Kelessis A, Petrakakis M, Zoumakis N, Christides T, Paschalidou AK. Air Quality assessment in a heavily-polluted urban Mediterranean environment through Air Quality indices. Ecol Indic . (2012) 18:259–68. doi: 10.1016/j.ecolind.2011.11.021

20. Dockery DW, Pope CA, Xu X, Spengler JD, Ware JH, Fay ME, et al. An association between air pollution and mortality in six U.S. cities. N Engl J Med . (1993) 329:1753–9. doi: 10.1056/NEJM199312093292401

21. Schwela DH, Köth-Jahr I. Leitfaden für die Aufstellung von Luftreinhalteplänen [Guidelines for the Implementation of Clean Air Implementation Plans]. Landesumweltamt des Landes Nordrhein Westfalen. State Environmental Service of the State of North Rhine-Westphalia (1994).

22. Newlands M. Environmental Activism, Environmental Politics, and Representation: The Framing of the British Environmental Activist Movement . Ph.D. thesis. University of East London, United Kingdom (2015).

23. NEPIS (National Service Center for Environmental Publications) US EPA (Environmental Protection Agency) (2017). Available online at: https://www.epa.gov/clean-air-act-overview/air-pollution-current-and-future-challenges (accessed August 15, 2017).

24. NRC (National Research Council). Available online at: https://www.nap.edu/read/10728/chapter/1,2014 (accessed September 17, 2019).

25. Bull A. Traffic Congestion: The Problem and How to Deal With It . Santiago: Nationes Unidas, Cepal (2003).

26. Spiegel J, Maystre LY. Environmental Pollution Control, Part VII - The Environment, Chapter 55, Encyclopedia of Occupational Health and Safety . Available online at: http://www.ilocis.org/documents/chpt55e.htm (accessed September 17, 2019).

27. European Community Reports. Assessment of the Effectiveness of European Air Quality Policies and Measures: Case Study 2; Comparison of the EU and US Air Quality Standards and Planning Requirements. (2004). Available online at: https://ec.europa.eu/environment/archives/cafe/activities/pdf/case_study2.pdf (accessed September 22, 2019).

28. Gibson R, Ward S. Parties in the digital age; a review. J Represent Democracy . (2009) 45:87–100. doi: 10.1080/00344890802710888

29. Kaun A, Uldam J. Digital activism: after the hype. New Media Soc. (2017) 20:2099–106. doi: 10.1177/14614448177319

30. Sivitanides M, Shah V. The era of digital activism. In: 2011 Conference for Information Systems Applied Research(CONISAR) Proceedings Wilmington North Carolina, USA . Available online at: https://www.arifyildirim.com/ilt510/marcos.sivitanides.vivek.shah.pdf (accessed September 22, 2019).

31. Möller L, Schuetzle D, Autrup H. Future research needs associated with the assessment of potential human health risks from exposure to toxic ambient air pollutants. Environ Health Perspect . (1994) 102(Suppl. 4):193–210. doi: 10.1289/ehp.94102s4193

32. Jacobson MZ, Jacobson PMZ. Atmospheric Pollution: History, Science, and Regulation. Cambridge University Press (2002). p. 206. doi: 10.1256/wea.243.02

33. Stover RH. Flooding of soil for disease control. In: Mulder D, editor. Chapter 3. Developments in Agricultural and Managed Forest Ecology . Elsevier (1979). p. 19–28. Available online at: http://www.sciencedirect.com/science/article/pii/B9780444416926500094 doi: 10.1016/B978-0-444-41692-6.50009-4 (accessed July 1, 2019).

34. Maipa V, Alamanos Y, Bezirtzoglou E. Seasonal fluctuation of bacterial indicators in coastal waters. Microb Ecol Health Dis . (2001) 13:143–6. doi: 10.1080/089106001750462687

35. Bezirtzoglou E, Dimitriou D, Panagiou A. Occurrence of Clostridium perfringens in river water by using a new procedure. Anaerobe . (1996) 2:169–73. doi: 10.1006/anae.1996.0022

36. Kjellstrom T, Lodh M, McMichael T, Ranmuthugala G, Shrestha R, Kingsland S. Air and Water Pollution: Burden and Strategies for Control. DCP, Chapter 43. 817–32 p. Available online at: https://www.dcp-3.org/sites/default/files/dcp2/DCP43.pdf (accessed September 17, 2017).

37. Pathak RK, Wang T, Ho KF, Lee SC. Characteristics of summertime PM2.5 organic and elemental carbon in four major Chinese cities: implications of high acidity for water- soluble organic carbon (WSOC). Atmos Environ . (2011) 45:318–25. doi: 10.1016/j.atmosenv.2010.10.021

38. Bonavigo L, Zucchetti M, Mankolli H. Water radioactive pollution and related environmental aspects. J Int Env Appl Sci . (2009) 4:357–63

39. World Health Organization (WHO). Preventing Disease Through Healthy Environments: Towards an Estimate of the Environmental Burden of Disease . 1106 p. Available online at: https://www.who.int/quantifying_ehimpacts/publications/preventingdisease.pdf (accessed September 22, 2019).

40. Stansfeld SA. Noise effects on health in the context of air pollution exposure. Int J Environ Res Public Health . (2015) 12:12735–60. doi: 10.3390/ijerph121012735

41. Ethical Unicorn. Everything You Need To Know About Aerosols & Air Pollution. (2019). Available online at: https://ethicalunicorn.com/2019/04/29/everything-you-need-to-know-about-aerosols-air-pollution/ (accessed October 4, 2019).

42. Colbeck I, Lazaridis M. Aerosols and environmental pollution. Sci Nat . (2009) 97:117–31. doi: 10.1007/s00114-009-0594-x

43. Incecik S, Gertler A, Kassomenos P. Aerosols and air quality. Sci Total Env . (2014) 355, 488–9. doi: 10.1016/j.scitotenv.2014.04.012

44. D'Amato G, Pawankar R, Vitale C, Maurizia L. Climate change and air pollution: effects on respiratory allergy. Allergy Asthma Immunol Res . (2016) 8:391–5. doi: 10.4168/aair.2016.8.5.391

45. Bezirtzoglou C, Dekas K, Charvalos E. Climate changes, environment and infection: facts, scenarios and growing awareness from the public health community within Europe. Anaerobe . (2011) 17:337–40. doi: 10.1016/j.anaerobe.2011.05.016

46. Castelli F, Sulis G. Migration and infectious diseases. Clin Microbiol Infect . (2017) 23:283–9. doi: 10.1016/j.cmi.2017.03.012

47. Watson JT, Gayer M, Connolly MA. Epidemics after natural disasters. Emerg Infect Dis . (2007) 13:1–5. doi: 10.3201/eid1301.060779

48. Fenn B. Malnutrition in Humanitarian Emergencies . Available online at: https://www.who.int/diseasecontrol_emergencies/publications/idhe_2009_london_malnutrition_fenn.pdf . (accessed August 15, 2017).

49. Lindh E, Argentini C, Remoli ME, Fortuna C, Faggioni G, Benedetti E, et al. The Italian 2017 outbreak Chikungunya virus belongs to an emerging Aedes albopictus –adapted virus cluster introduced from the Indian subcontinent. Open Forum Infect Dis. (2019) 6:ofy321. doi: 10.1093/ofid/ofy321

50. Calba C, Guerbois-Galla M, Franke F, Jeannin C, Auzet-Caillaud M, Grard G, Pigaglio L, Decoppet A, et al. Preliminary report of an autochthonous chikungunya outbreak in France, July to September 2017. Eur Surveill . (2017) 22:17-00647. doi: 10.2807/1560-7917.ES.2017.22.39.17-00647

51. Menne B, Murray V. Floods in the WHO European Region: Health Effects and Their Prevention . Copenhagen: WHO; Weltgesundheits organisation, Regionalbüro für Europa (2013). Available online at: http://www.euro.who.int/data/assets/pdf_file/0020/189020/e96853.pdf (accessed 15 August 2017).

52. Schneider SH. The greenhouse effect: science and policy. Science . (1989) 243:771–81. doi: 10.1126/science.243.4892.771

53. Wilson WE, Suh HH. Fine particles and coarse particles: concentration relationships relevant to epidemiologic studies. J Air Waste Manag Assoc . (1997) 47:1238–49. doi: 10.1080/10473289.1997.10464074

54. US EPA (US Environmental Protection Agency) (2018). Available online at: https://www.epa.gov/pm-pollution/particulate-matter-pm-basics (accessed September 22, 2018).

55. Cheung K, Daher N, Kam W, Shafer MM, Ning Z, Schauer JJ, et al. Spatial and temporal variation of chemical composition and mass closure of ambient coarse particulate matter (PM10–2.5) in the Los Angeles area. Atmos Environ . (2011) 45:2651–62. doi: 10.1016/j.atmosenv.2011.02.066

56. Zhang L, Yang Y, Li Y, Qian ZM, Xiao W, Wang X, et al. Short-term and long-term effects of PM2.5 on acute nasopharyngitis in 10 communities of Guangdong, China. Sci Total Env. (2019) 688:136–42. doi: 10.1016/j.scitotenv.2019.05.470.

57. Kloog I, Ridgway B, Koutrakis P, Coull BA, Schwartz JD. Long- and short-term exposure to PM2.5 and mortality using novel exposure models, Epidemiology . (2013) 24:555–61. doi: 10.1097/EDE.0b013e318294beaa

58. New Hampshire Department of Environmental Services. Current and Forecasted Air Quality in New Hampshire . Environmental Fact Sheet (2019). Available online at: https://www.des.nh.gov/organization/commissioner/pip/factsheets/ard/documents/ard-16.pdf (accessed September 22, 2019).

59. Kappos AD, Bruckmann P, Eikmann T, Englert N, Heinrich U, Höppe P, et al. Health effects of particles in ambient air. Int J Hyg Environ Health . (2004) 207:399–407. doi: 10.1078/1438-4639-00306

60. Boschi N (Ed.). Defining an educational framework for indoor air sciences education. In: Education and Training in Indoor Air Sciences . Luxembourg: Springer Science & Business Media (2012). 245 p.

61. Heal MR, Kumar P, Harrison RM. Particles, air quality, policy and health. Chem Soc Rev . (2012) 41:6606–30. doi: 10.1039/c2cs35076a

62. Bezirtzoglou E, Alexopoulos A. Ozone history and ecosystems: a goliath from impacts to advance industrial benefits and interests, to environmental and therapeutical strategies. In: Ozone Depletion, Chemistry and Impacts. (2009). p. 135–45.

63. Villányi V, Turk B, Franc B, Csintalan Z. Ozone Pollution and its Bioindication. In: Villányi V, editor. Air Pollution . London: Intech Open (2010). doi: 10.5772/10047

64. Massachusetts Department of Public Health. Massachusetts State Health Assessment . Boston, MA (2017). Available online at: https://www.mass.gov/files/documents/2017/11/03/2017%20MA%20SHA%20final%20compressed.pdf (accessed October 30, 2017).

65. Lorenzini G, Saitanis C. Ozone: A Novel Plant “Pathogen.” In: Sanitá di Toppi L, Pawlik-Skowrońska B, editors. Abiotic Stresses in Plant Springer Link (2003). p. 205–29. doi: 10.1007/978-94-017-0255-3_8

66. Fares S, Vargas R, Detto M, Goldstein AH, Karlik J, Paoletti E, et al. Tropospheric ozone reduces carbon assimilation in trees: estimates from analysis of continuous flux measurements. Glob Change Biol . (2013) 19:2427–43. doi: 10.1111/gcb.12222

67. Harmens H, Mills G, Hayes F, Jones L, Norris D, Fuhrer J. Air Pollution and Vegetation . ICP Vegetation Annual Report 2006/2007. (2012)

68. Emberson LD, Pleijel H, Ainsworth EA, den Berg M, Ren W, Osborne S, et al. Ozone effects on crops and consideration in crop models. Eur J Agron . (2018) 100:19–34. doi: 10.1016/j.eja.2018.06.002

69. Alexopoulos A, Plessas S, Ceciu S, Lazar V, Mantzourani I, Voidarou C, et al. Evaluation of ozone efficacy on the reduction of microbial population of fresh cut lettuce ( Lactuca sativa ) and green bell pepper ( Capsicum annuum ). Food Control . (2013) 30:491–6. doi: 10.1016/j.foodcont.2012.09.018

70. Alexopoulos A, Plessas S, Kourkoutas Y, Stefanis C, Vavias S, Voidarou C, et al. Experimental effect of ozone upon the microbial flora of commercially produced dairy fermented products. Int J Food Microbiol . (2017) 246:5–11. doi: 10.1016/j.ijfoodmicro.2017.01.018

71. Maggio A, Fagnano M. Ozone damages to mediterranean crops: physiological responses. Ital J Agron . (2008) 13–20. doi: 10.4081/ija.2008.13

72. McCarthy JT, Pelle E, Dong K, Brahmbhatt K, Yarosh D, Pernodet N. Effects of ozone in normal human epidermal keratinocytes. Exp Dermatol . (2013) 22:360–1. doi: 10.1111/exd.12125

73. WHO. Health Risks of Ozone From Long-Range Transboundary Air Pollution . Available online at: http://www.euro.who.int/data/assets/pdf_file/0005/78647/E91843.pdf (accessed August 15, 2019).

74. Thiele JJ, Traber MG, Tsang K, Cross CE, Packer L. In vivo exposure to ozone depletes vitamins C and E and induces lipid peroxidation in epidermal layers of murine skin. Free Radic Biol Med. (1997) 23:365–91. doi: 10.1016/S0891-5849(96)00617-X

75. Hatch GE, Slade R, Harris LP, McDonnell WF, Devlin RB, Koren HS, et al. Ozone dose and effect in humans and rats. A comparison using oxygen- 18 labeling and bronchoalveolar lavage. Am J Respir Crit Care Med . (1994) 150:676–83. doi: 10.1164/ajrccm.150.3.8087337

76. Lippmann M. Health effects of ozone. A critical review. JAPCA . (1989) 39:672–95. doi: 10.1080/08940630.1989.10466554

77. Gryparis A, Forsberg B, Katsouyanni K, Analitis A, Touloumi G, Schwartz J, et al. Acute effects of ozone on mortality from the “air pollution and health: a European approach” project. Am J Respir Crit Care Med . (2004) 170:1080–7. doi: 10.1164/rccm.200403-333OC

78. Soon W, Baliunas SL, Robinson AB, Robinson ZW. Environmental effects of increased atmospheric carbon dioxide. Climate Res . (1999) 13:149–64 doi: 10.1260/0958305991499694

79. Richmont-Bryant J, Owen RC, Graham S, Snyder M, McDow S, Oakes M, et al. Estimation of on-road NO2 concentrations, NO2/NOX ratios, and related roadway gradients from near-road monitoring data. Air Qual Atm Health . (2017) 10:611–25. doi: 10.1007/s11869-016-0455-7

80. Hesterberg TW, Bunn WB, McClellan RO, Hamade AK, Long CM, Valberg PA. Critical review of the human data on short-term nitrogen dioxide (NO 2 ) exposures: evidence for NO2 no-effect levels. Crit Rev Toxicol . (2009) 39:743–81. doi: 10.3109/10408440903294945

81. Chen T-M, Gokhale J, Shofer S, Kuschner WG. Outdoor air pollution: nitrogen dioxide, sulfur dioxide, and carbon monoxide health effects. Am J Med Sci . (2007) 333:249–56. doi: 10.1097/MAJ.0b013e31803b900f

82. US EPA. Table of Historical SO 2 NAAQS, Sulfur US EPA . Available online at: https://www3.epa.gov/ttn/naaqs/standards/so2/s_so2_history.html (accessed October 5, 2019).

83. WHO Regional Office of Europe (2000). Available online at: https://euro.who.int/_data/assets/pdf_file/0020/123086/AQG2ndEd_7_4Sulfuroxide.pdf

84. Pruss-Ustun A, Fewrell L, Landrigan PJ, Ayuso-Mateos JL. Lead exposure. Comparative Quantification of Health Risks . World Health Organization. p. 1495–1542. Available online at: https://www.who.int/publications/cra/chapters/volume2/1495-1542.pdf?ua=1

PubMed Abstract | Google Scholar

85. Goyer RA. Transplacental transport of lead. Environ Health Perspect . (1990) 89:101–5. doi: 10.1289/ehp.9089101

86. National Institute of Environmental Health Sciences (NIH). Lead and Your Health . (2013). 1–4 p. Available online at: https://www.niehs.nih.gov/health/materials/lead_and_your_health_508.pdf (accessed September 17, 2019).

87. Farhat A, Mohammadzadeh A, Balali-Mood M, Aghajanpoor-Pasha M, Ravanshad Y. Correlation of blood lead level in mothers and exclusively breastfed infants: a study on infants aged less than six months. Asia Pac J Med Toxicol . (2013) 2:150–2.

88. Assi MA, Hezmee MNM, Haron AW, Sabri MYM, Rajion MA. The detrimental effects of lead on human and animal health. Vet World . (2016) 9:660–71. doi: 10.14202/vetworld.2016.660-671

89. Abdel-Shafy HI, Mansour MSM. A review on polycyclic aromatic hydrocarbons: source, environmental impact, effect on human health and remediation. Egypt J Pet . (2016) 25:107–23. doi: 10.1016/j.ejpe.2015.03.011

90. Kumar A, Singh BP, Punia M, Singh D, Kumar K, Jain VK. Assessment of indoor air concentrations of VOCs and their associated health risks in the library of Jawaharlal Nehru University, New Delhi. Environ Sci Pollut Res Int . (2014) 21:2240–8. doi: 10.1007/s11356-013-2150-7

91. Molhave L, Clausen G, Berglund B, Ceaurriz J, Kettrup A, Lindvall T, et al. Total Volatile Organic Compounds (TVOC) in Indoor Air Quality Investigations. Indoor Air . 7:225–240. doi: 10.1111/j.1600-0668.1997.00002.x

92. Gibb T. Indoor Air Quality May be Hazardous to Your Health . MSU Extension. Available online at: https://www.canr.msu.edu/news/indoor_air_quality_may_be_hazardous_to_your_health (accessed October 5, 2019).

93. Ebersviller S, Lichtveld K, Sexton KG, Zavala J, Lin Y-H, Jaspers I, et al. Gaseous VOCs rapidly modify particulate matter and its biological effects – Part 1: simple VOCs and model PM. Atmos Chem Phys Discuss . (2012) 12:5065–105. doi: 10.5194/acpd-12-5065-2012

94. WHO (World Health Organization). Dioxins and Their Effects on Human Health. Available online at: https://www.who.int/news-room/fact-sheets/detail/dioxins-and-their-effects-on-human-health (accessed October 5, 2019).

95. EEA (European Environmental Agency). Air Quality Standards to the European Union and WHO . Available online at: https://www.eea.europa.eu/themes/data-and-maps/figures/air-quality-standards-under-the

96. Nakano T, Otsuki T. [Environmental air pollutants and the risk of cancer]. (Japanese). Gan To Kagaku Ryoho . (2013) 40:1441–5.

97. Kurt OK, Zhang J, Pinkerton KE. Pulmonary health effects of air pollution. Curr Opin Pulm Med . (2016) 22:138–43. doi: 10.1097/MCP.0000000000000248

98. Guarnieri M, Balmes JR. Outdoor air pollution and asthma. Lancet . (2014) 383:1581–92. doi: 10.1016/S0140-6736(14)60617-6

99. Jiang X-Q, Mei X-D, Feng D. Air pollution and chronic airway diseases: what should people know and do? J Thorac Dis . (2016) 8:E31–40.

100. Bourdrel T, Bind M-A, Béjot Y, Morel O, Argacha J-F. Cardiovascular effects of air pollution. Arch Cardiovasc Dis . (2017) 110:634–42. doi: 10.1016/j.acvd.2017.05.003

101. Hoffmann B, Moebus S, Möhlenkamp S, Stang A, Lehmann N, Dragano N, et al. Residential exposure to traffic is associated with coronary atherosclerosis. Circulation . (2007) 116:489–496. doi: 10.1161/CIRCULATIONAHA.107.693622

102. Katholi RE, Couri DM. Left ventricular hypertrophy: major risk factor in patients with hypertension: update and practical clinical applications. Int J Hypertens . (2011) 2011:495349. doi: 10.4061/2011/495349

103. Leary PJ, Kaufman JD, Barr RG, Bluemke DA, Curl CL, Hough CL, et al. Traffic- related air pollution and the right ventricle. the multi-ethnic study of atherosclerosis. Am J Respir Crit Care Med . (2014) 189:1093–100. doi: 10.1164/rccm.201312-2298OC

104. Genc S, Zadeoglulari Z, Fuss SH, Genc K. The adverse effects of air pollution on the nervous system. J Toxicol . (2012) 2012:782462. doi: 10.1155/2012/782462

105. Calderon-Garciduenas L, Azzarelli B, Acuna H, et al. Air pollution and brain damage. Toxicol Pathol. (2002) 30:373–89. doi: 10.1080/01926230252929954

106. Rückerl R, Greven S, Ljungman P, Aalto P, Antoniades C, Bellander T, et al. Air pollution and inflammation (interleukin-6, C-reactive protein, fibrinogen) in myocardial infarction survivors. Environ Health Perspect . (2007) 115:1072–80. doi: 10.1289/ehp.10021

107. Peters A, Veronesi B, Calderón-Garcidueñas L, Gehr P, Chen LC, Geiser M, et al. Translocation and potential neurological effects of fine and ultrafine particles a critical update. Part Fibre Toxicol . (2006) 3:13–8. doi: 10.1186/1743-8977-3-13

108. Kelly FJ. Dietary antioxidants and environmental stress. Proc Nutr Soc . (2004) 63:579–85. doi: 10.1079/PNS2004388

109. Bellinger DC. Very low lead exposures and children's neurodevelopment. Curr Opin Pediatr . (2008) 20:172–7. doi: 10.1097/MOP.0b013e3282f4f97b

110. Balbo P, Silvestri M, Rossi GA, Crimi E, Burastero SE. Differential role of CD80 and CD86 on alveolar macrophages in the presentation of allergen to T lymphocytes in asthma. Clin Exp Allergy J Br Soc Allergy Clin Immunol . (2001) 31:625–36. doi: 10.1046/j.1365-2222.2001.01068.x

111. Drakaki E, Dessinioti C, Antoniou C. Air pollution and the skin. Front Environ Sci Eng China . (2014) 15:2–8. doi: 10.3389/fenvs.2014.00011

112. Weisskopf MG, Kioumourtzoglou M-A, Roberts AL. Air pollution and autism spectrum disorders: causal or confounded? Curr Environ Health Rep . (2015) 2:430–9. doi: 10.1007/s40572-015-0073-9

113. Mo Z, Fu Q, Lyu D, Zhang L, Qin Z, Tang Q, et al. Impacts of air pollution on dry eye disease among residents in Hangzhou, China: a case-crossover study. Environ Pollut . (2019) 246:183–9. doi: 10.1016/j.envpol.2018.11.109

114. Klopfer J. Effects of environmental air pollution on the eye. J Am Optom Assoc . (1989) 60:773–8.

115. Ashfaq A, Sharma P. Environmental effects of air pollution and application of engineered methods to combat the problem. J Indust Pollut Control . (2012) 29.

116. Madronich S, de Gruijl F. Skin cancer and UV radiation. Nature . (1993) 366:23–9. doi: 10.1038/366023a0

117. Teramura A. Effects of UV-B radiation on the growth and yield of crop plants. Physiol Plant . (2006) 58:415–27. doi: 10.1111/j.1399-3054.1983.tb04203.x

118. Singh E, Tiwari S, Agrawal M. Effects of elevated ozone on photosynthesis and stomatal conductance of two soybean varieties: a case study to assess impacts of one component of predicted global climate change. Plant Biol Stuttg Ger . (2009) 11(Suppl. 1):101–8. doi: 10.1111/j.1438-8677.2009.00263.x

119. Manderson L. How global Warming is Adding to the Health Risks of Poor People . The Conversation. University of the Witwatersrand. Available online at: http://theconversation.com/how-global-warming-is-adding-to-the-health-risks-of-poor-people-109520 (accessed October 5, 2019).

120. Ministers of Energy and Environment. Federal/Provincial/Territorial Ministers of Energy and Environment (Canada), editor. The Canada-Wide Acid Rain Strategy for Post-2000 . Halifax: The Ministers (1999). 11 p.

121. Zuhara S, Isaifan R. The impact of criteria air pollutants on soil and water: a review. (2018) 278–84. doi: 10.30799/jespr.133.18040205

122. WHO. First WHO Global Conference on Air Pollution and Health. (2018). Available online at: https://www.who.int/airpollution/events/conference/en/ (accessed October 6, 2019).

123. What is the Kyoto Protocol? UNFCCC . Available online at: https://unfccc.int/kyoto__protocol (accessed October 6, 2019).

124. CopenhagenClimate Change Conference (UNFCCC) . Available online at: https://unfccc.int/process-and-meetings/conferences/past-conferences/copenhagen-climate-change-conference-december-2009/copenhagen-climate-change-conference-december-2009 (accessed October 6, 2019).

125. Durban Climate Change Conference,. UNFCCC (2011). Available online at: https://unfccc.int/process-and-meetings/conferences/past-conferences/copenhagen-climate-change-conference-december-2009/copenhagen-climate-change-conference-december-2009 (accessed October 6, 2019).

126. Paris Climate Change Agreement,. (2016). Available online at: https://unfccc.int/process-and-meetings/the-paris-agreement/the-paris-agreement

Keywords: air pollution, environment, health, public health, gas emission, policy

Citation: Manisalidis I, Stavropoulou E, Stavropoulos A and Bezirtzoglou E (2020) Environmental and Health Impacts of Air Pollution: A Review. Front. Public Health 8:14. doi: 10.3389/fpubh.2020.00014

Received: 17 October 2019; Accepted: 17 January 2020; Published: 20 February 2020.

Reviewed by:

Copyright © 2020 Manisalidis, Stavropoulou, Stavropoulos and Bezirtzoglou. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Ioannis Manisalidis, giannismanisal@gmail.com ; Elisavet Stavropoulou, elisabeth.stavropoulou@gmail.com

† These authors have contributed equally to this work

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

The Nonlinear Effects of Air Pollution on Health: Evidence from Wildfire Smoke

We estimate how acute air pollution exposure from wildfire smoke impacts human health in the U.S., allowing for nonlinear effects. Wildfire smoke is pervasive and produces air quality shocks of varying intensity, depending on wind patterns and plume thickness. Using administrative Medicare records for 2007–2019, we estimate that wildfire smoke accounts for 18% of ambient PM2.5 concentrations, 0.42% of deaths, and 0.69% of emergency room visits among adults aged 65 and over. Smaller pollution shocks have outsized health impacts, indicating significant health benefits from improving air quality, even in areas meeting current regulatory standards.

We thank Yifan Wang for excellent research assistance and Judson Boomhower, Marshall Burke, Alex Hollingsworth, Edson Severnini, Nikolaos Zirogiannis, and seminar participants at Cornell University, McGill University, the NBER EEE meeting, the Occasional Workshop in Environmental and Resource Economics, Princeton University, the Property and Environment Research Center, the University of Illinois, and the University of Texas at Austin for helpful comments. The research reported in this publication was supported by the National Institute on Aging of the National Institutes of Health under award numbers P01AG005842, R01AG053350, and R01AG073365. The content is solely the authors’ responsibility and does not necessarily represent the official views of the National Institutes of Health. Earlier versions of the paper were circulated under the titles “Blowing Smoke: Health Impacts of Wildfire Plume Dynamics” and “A Causal Concentration–Response Function for Air Pollution: Evidence from Wildfire Smoke.” The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.

MARC RIS BibTeΧ

Download Citation Data

More from NBER

In addition to working papers , the NBER disseminates affiliates’ latest findings through a range of free periodicals — the NBER Reporter , the NBER Digest , the Bulletin on Retirement and Disability , the Bulletin on Health , and the Bulletin on Entrepreneurship  — as well as online conference reports , video lectures , and interviews .

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

The impact of data imputation on air quality prediction problem

Roles Conceptualization, Data curation, Formal analysis, Methodology, Visualization, Writing – original draft

Affiliations Faculty of Mathematics and Computer Science, University of Science, Ho Chi Minh City, Vietnam, Vietnam National University Ho Chi Minh City, Ho Chi Minh City, Vietnam, Faculty of Information Technology, HUTECH University, Ho Chi Minh City, Vietnam

Roles Conceptualization, Investigation, Methodology, Writing – review & editing

Affiliation SimulaMet, Oslo, Norway

Roles Formal analysis

Affiliation National Institute of Information and Communications Technology, Tokyo, Japan

Roles Formal analysis, Methodology

Affiliations Vietnam National University Ho Chi Minh City, Ho Chi Minh City, Vietnam, University of Information Technology, Ho Chi Minh City, Vietnam

Roles Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliations Faculty of Mathematics and Computer Science, University of Science, Ho Chi Minh City, Vietnam, Vietnam National University Ho Chi Minh City, Ho Chi Minh City, Vietnam

• Van Hua, 
• Thu Nguyen, 
• Minh-Son Dao, 
• Hien D. Nguyen, 
• Binh T. Nguyen

• Published: September 12, 2024
• https://doi.org/10.1371/journal.pone.0306303
• Reader Comments

With rising environmental concerns, accurate air quality predictions have become paramount as they help in planning preventive measures and policies for potential health hazards and environmental problems caused by poor air quality. Most of the time, air quality data are time series data. However, due to various reasons, we often encounter missing values in datasets collected during data preparation and aggregation steps. The inability to analyze and handle missing data will significantly hinder the data analysis process. To address this issue, this paper offers an extensive review of air quality prediction and missing data imputation techniques for time series, particularly in relation to environmental challenges. In addition, we empirically assess eight imputation methods, including mean, median, kNNI, MICE, SAITS, BRITS, MRNN, and Transformer, to scrutinize their impact on air quality data. The evaluation is conducted using diverse air quality datasets gathered from numerous cities globally. Based on these evaluations, we offer practical recommendations for practitioners dealing with missing data in time series scenarios for environmental data.

Citation: Hua V, Nguyen T, Dao M-S, Nguyen HD, Nguyen BT (2024) The impact of data imputation on air quality prediction problem. PLoS ONE 19(9): e0306303. https://doi.org/10.1371/journal.pone.0306303

Editor: Abid Rashid Gill, Islamia University of Bahawalpur, PAKISTAN

Received: February 14, 2024; Accepted: June 15, 2024; Published: September 12, 2024

Copyright: © 2024 Hua et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: - The dataset "Frankfurt (German)" is available at the following: URL: https://www.kaggle.com/datasets/avibagul80/air-quality-dataset Author: Avinash Bagul - University of Aberdeen - The dataset "Beijing (China)" is available upon request from the authors of the following article: Du W, Côté D, Liu Y. Saits: Self-attention-based imputation for time series, Expert Systems with Applications. 2023;219:119619. URL: https://www.sciencedirect.com/science/article/abs/pii/S0957417423001203 DOI: https://doi.org/10.1016/j.eswa.2023.119619 Dataset link: https://archive.ics.uci.edu/ml/datasets/Beijing+Multi-Site+Air-Quality+Data Corresponding authors: Yan Liu Email: [email protected] - The dataset " Northern Taiwan (Taiwan)" is available at the following: URL: https://www.kaggle.com/datasets/nelsonchu/air-quality-in-northern-taiwan Author: Open Government Data License, version 1.0 http://data.gov.tw/license - The dataset "Dalat (Vietnam)" is available upon request from the authors of the following article: Dao MS, Dang TH, Nguyen-Tai TL, Nguyen TB, Dang-Nguyen DT. Overview of MediaEval 2022 Urban Air: Urban Life and Air Pollution. In: Proc. of the MediaEval 2022 Workshop; 2023. p. 13–15. URL: https://ceur-ws.org/Vol-3583/paper4.pdf Corresponding authors: Minh-Son Dao Email: [email protected] Dataset link: https://github.com/BinhMisfit/air-pollution-datasets/tree/main/Dalat-air-quality-dataset - The dataset "Cau Giay District (Hanoi, Vietnam)" is available upon request from the authors of the following article: Ton-Thien MA, Nguyen CT, Le QM, Duong DQ, Dao MS, Nguyen BT. Air Pollution Forecasting Using Multimodal Data. In: International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems. Springer; 2023. p. 360–371. URL: https://link.springer.com/chapter/10.1007/978-3-031-36822-6_31 DOI: https://doi.org/10.1007/978-3-031-36822-6_31 Corresponding authors: Binh T. Nguyen Email: [email protected] Dataset link: https://github.com/BinhMisfit/air-pollution-datasets/tree/main/Hanoi-air-quality-dataset - The dataset "Minh Khai District (Hanoi, Vietnam)" is available upon request from the authors of the following article: Ton-Thien MA, Nguyen CT, Le QM, Duong DQ, Dao MS, Nguyen BT. Air Pollution Forecasting Using Multimodal Data. In: International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems. Springer; 2023. p. 360–371. URL: https://link.springer.com/chapter/10.1007/978-3-031-36822-6_31 DOI: https://doi.org/10.1007/978-3-031-36822-6_31 Corresponding authors: Binh T. Nguyen Email: [email protected] Dataset link: https://github.com/BinhMisfit/air-pollution-datasets/tree/main/Hanoi-air-quality-dataset .

Funding: This research is funded by Vietnam National University Ho Chi Minh City (VNU-HCM) under grant number DS2023-18-01. When working on this research paper, Ms. Van Hua was a Master student at the University of Science, Vietnam National University Ho Chi Minh City. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

1 Introduction

Air helps sustain human life, so air tracking and understanding its quality is essential for our health. Air pollutants can pose significant threats to public health, and sources of air pollution can come from nature, such as smoke from volcano eruptions or forest fires, methane from animals’ process of digesting food, or radon gas from radioactive decay in the earth’s crust. In addition, pollution can also come from manufacturing activities such as industry and agriculture. They emit CO 2 , CO , SO 2 , NO 2 , and other organic substances at extremely high concentrations, polluting the air. Besides, burning fossil fuels yields climate change and air pollution. Therefore, air quality has still been a concern in recent years. Consequently, environmental researchers mine air quality data to uncover potential value and information from these data, thereby capturing user behavior, estimating disease causes, discovering gases, detecting individual actions to reduce greenhouse gases, acid rain, etc., and then advising management agencies and local governments to plan related policies. By using machine learning techniques, the local air quality data can be analyzed using sensors that gather real-time humidity and temperature readings. Duong et al. (2021) effectively extracted pertinent features from a dataset. They applied machine learning models to forecast AQI (Air Quality Indexing) values and levels at any user-specified location in Ho Chi Minh City [ 1 ]. The dataset includes data on six atmospheric pollutants: SO 2 , NO 2 , PM 10, PM 2.5, CO , and O 3 , collected by volunteers who traversed predetermined routes to provide ground-truth AQI levels.

Many air quality data are in the form of time series and can contain missing values due to corrupted sensors, loss of electricity, etc. In such cases, data imputation, i.e., filling in missing values with some reasonable value according to some criteria, is a conventional practice to resolve the issue. The quality of imputation can significantly impact the downstream classification or prediction task. One can characterize missing data into three types: missing completely at random (MCAR), where the missing values are independent of any other values; missing at random (MAR), where missing values depend only on observed values; and missing not at random (MNAR), where missing values depends on both observed and unobserved values [ 2 ]. There are many methods to deal with missing values based on the missing data mechanism. This work focuses on the MAR case, as it is prevalent for sensor data related to the environment [ 3 ]. Furthermore, most air quality observation data are time series data. Dealing with missing values in time series data is often difficult, time-consuming, and labor-intensive. In addition, the missing data can significantly affect the processing and analysis of data. Therefore, handling missing values in time-series air quality data is necessary.

People can reveal critical enhancements regarding performance and running time by examining newly introduced approaches for data imputation. Multiple imputations can be further applied with these imputation methods to reduce the uncertainty by repeating the imputation procedure numerous times and averaging the results. Combining the imputation methods with forecasting models often results in a two-step process where imputation and forecasting models are separated. By doing this, the missingness is effectively explored in the forecasting model, thus leading to suboptimal analysis results. In addition, many imputation methods also have other requirements that may not be satisfied in real applications; for example, many of them work on data with low missing rates only, assume the data is missing randomly or completely at random, or can not handle time series data with varying lengths [ 4 ]. Moreover, training and applying these imputation methods are usually computationally expensive.

Various imputation techniques have been proposed to fill in missing values, each using a distinct set of assumptions, algorithms, and performance metrics. Choosing a relevant imputation method can significantly influence the subsequent analysis and the reliability of the results. To give a thorough comparative analysis of various missing data imputation methods for time series air quality of data [ 5 ], we compare several conventional but often used imputation techniques (mean, median, kNNI, and MICE) with several recently developed imputation techniques for temporal data (SAITS, BRITS, MRNN, and Transformer) to examine their impact on air quality data from various places. On the other hand, the rate of missing data can also impact the problem-solving strategy we use since missing values can be handled in the step of data preprocessing. Various works have conducted experiments under a variety of missing rates. For example, [ 6 ] conducts experiments with missing rates from 5% to 50%. Nevertheless, in some other papers, the missing values rate could range from 1% to 80% of data [ 7 – 9 ], or start from 10% to 50% [ 10 ].

While some work [ 11 ] has been done to compare the performance of classical and newly developed time series imputation techniques such as BRITS [ 12 ], SAITS [ 13 ] for health care data, such practical comparisons for air quality has not been conducted yet. In addition, while there have been several works that examine the effects of imputation on air quality [ 14 – 16 ], most of them do not cover state-of-the-art imputation methods for time series that have been developed in recent years. In addition, up to our knowledge, while there have been some surveys on air quality prediction [ 17 , 18 ] or missing data imputation for air quality data [ 19 , 20 ], there has not been any work that reviews both problems and systematically compares state-of-the-art imputation algorithms for air quality data. This motivates us to review recent studies related to the air quality prediction problem, along with missing values handling methods or techniques on time series data with a concentration on air quality data. In addition, we also empirically evaluate various time series imputation techniques, including classical and state-of-the-art methods for air quality data. In summary, the contribution of our work can be described as follows:

• We review existing techniques for air quality prediction and missing data imputation.
• We conduct experiments on various air quality datasets to compare the performance of various time series imputation methods using various measures.
• We provide analysis and evaluation of the performance of techniques.
• We provide practitioners practical guidance on how to deal with missing data in air quality data.

The structure of the paper can be organized as follows. Firstly, Section 1 gives an overview of the current research related to data and missing values and describes the research problem in Section 2. Afterward, we review the prediction methods for air quality data in Section 3. Besides, we also review related techniques imputing missing values in time series data from conventional to modern data in Section 4. Next, in Section 5, we present methods for imputing the missing values in this paper. Experiments compare and evaluate the results and imputation time of the methods and the accuracy of prediction models on air pollutant values and AQI levels on the different datasets in Section 6. Then, we discuss the related problems to impute missing values in Section 7. Finally, the paper ends with our conclusion and future works in Section 8.

2 Problem formulation

Most urban areas worldwide, including Vietnam, are facing increasing air pollution. Among them, the problem of air pollution due to dust is still the most prominent. In some large cities like Hanoi, the number of days with PM 10 and PM 2.5 dust pollution levels exceeding the limits is relatively high. The problem is how to reduce the impact of air pollution on human health. Therefore, to solve the above problem, experts believe that if air pollution is informed early in the form of prediction, it can help people proactively plan their lives, especially on days when air pollution is high, minimizing the effects of air pollution on health. Thereby, people will know and choose how to protect their health and that of their family members. Many countries predict air quality from three to five days in advance based on air and meteorological data (such as temperature, humidity, wind direction, and topography) from air monitoring stations. However, in implementing the problem of collecting data through sensors, the possibility of data loss of information occurs very often and unavoidably. Through this paper, we also present ways to handle missing data and how it will affect the problem of air pollution prediction or similar time series problems.

Missing data can exist in various ways, for example, at individual points or over intervals, where one sensor loses data for a period of time. In this section, we introduce preliminary definitions and formalize the problem of air quality imputation. Air quality data is generally collected from a set of sensors over different periods. We focus on the time series data with missing values. Some notations are defined to describe this problem.

The performance of each imputation model is computed by considering the indicating mask. The missing values in the matrix X will be imputed using traditional imputation techniques (i.e., Mean, Median, MICE, kNNI) and recently developed imputation techniques (i.e., SAITS, BRITS, MRNN, Transformer). In what follows, we will review the current approaches in detail.

3 Air quality prediction: Existing techniques

In the current studies, there is a wealth of research on air quality prediction due to its importance in informing about the pollution level that will allow policy-makers to adopt measures for reducing its impact [ 21 , 22 ]. Methods for air quality prediction can be classified into statistical, machine learning [ 23 , 24 ], and deep learning approaches.

3.1 Statistical methods

3.1.1 vector auto-regression (var)..

One of the most popular statistical models for forecasting multivariate time series is the Vector Auto-Regression (VAR). It is considered an extension of the univariate autoregressive model. The findings of [ 25 ] have revealed that the VAR model is particularly valuable in capturing the dynamic characteristics of economic and financial time series, making it a powerful tool for describing their behavior and making forecasts. In [ 26 ], VAR was used to forecast daily concentrations of air pollutants (i.e., CO , NO 2 , and SO 3 ) in Tehran city for the next 24h. For such a task, the authors have considered the correlations between air pollutants to get more accurate forecasts. Experimental results have indicated the high efficiency of the proposed method.

3.1.2 Autoregressive Integrated Moving (ARIMA).

Aside from VAR, ARIMA algorithms [ 27 ] were applied to forecast air quality. In [ 28 ], authors proposed a hybrid method named ARIMAX by combining the advantage of ARIMA and numerical modeling to forecast real-time air pollutants in Hong Kong (i.e., PM 2.5, O 3 , and NO 2 ). By employing experimental analysis, the proposed method significantly improves the quality of forecast results in multiple evaluation metrics. Similarly, the findings in [ 29 ] have shown the prominent role of ARIMA in forecasting PM10 in Dakar, Senegal. Accordingly, the proposed method combines system observations with multi-agent real-time simulation and evaluates with several simulations.

3.2 Traditional machine-learning methods

Some traditional machine learning algorithms used for air quality prediction can be Support Vector Regression (SVR), Random Forest (RF), and Linear Regression (LR).

3.2.1 Support Vector Regression (SVR).

SVR models were used to forecast PM 2.5 and PM 10 in London [ 30 ]. In that paper, the experimental results indicate the SVR’s efficiency in forecasting air quality parameters (i.e., PM2.5 and PM10). A nonlinear dynamic model based on the SVR technique was proposed to forecast AQI in Oviedo, Spain [ 31 ]. Accordingly, the proposed model first analyzed the relationship between primary and secondary pollutants. Then, it derived vital factors influencing the air quality and recommended potential enhancements for health and lifestyle. Zhu et al. [ 32 ] investigated an application of the SVR algorithm with a quasi-linear kernel for air quality prediction. For such a task, the paper designed a gated linear network to construct the multiple piecewise linear model, and it could be developed through the pre-training of a Winner-Take-All (WTA) autoencoder. This approach could outperform other state-of-the-art methods in the case of complex air quality prediction problems. It is due to the WTA strategy reducing the risk of overfitting and choosing appropriate sparsity parameters.

3.2.2 Random Forest (RF).

Regarding RF [ 33 , 34 ] proposed a parallel approach combined with Spark to forecast PM 2.5 in Beijing. The experimental results revealed the efficiency and scalability of the proposed method in the case of big data. Later, RF was used to select the most important features to improve the quality of real-time air quality prediction [ 35 ]. Concretely, the proposed method provides highly accurate predictions of three air pollutants (i.e., PM 2.5, NO 2 , SO 2 ) and outperforms other state-of-the-art methods.

3.2.3 Linear regression.

Linear regression is also a state-of-the-art model for air quality prediction. Indeed, many linear regression models have been proposed to predict AQI levels in New Delhi [ 36 ]; In Catalonia [ 37 ], authors combined factors including the effect of the surface reflectance capacity of urban surfaces with solar radiation and elevation to predict AQI level in Catalonia. The dataset is collected from 75 different air quality monitoring stations. A clustering technique was applied to cluster these stations based on their similarity. Meanwhile, Multiple Linear Regression (MLR) was used to replicate the annual mean values of AQI in Catalonia. Experimental results illustrated that the proposed model provided highly accurate predictions of AQI. Djuric et al. [ 38 ] proposed a multiple linear regression to forecast air pollution indices (i.e., SO 2 , NO 2 , PM 10, O 3 , and CO ) in Belgrade, Serbia. They collected the training and testing sets from the winters of 2011 and 2012/2013, respectively. In addition, the findings show that the proposed model can be scaled up to forecast long-term air quality.

3.3 Deep learning techniques

3.3.1 long short-term memory..

Apart from the traditional machine learning approaches, most deep learning methods, such as Long short-term memory (LSTM), have shown their superiority over many machine learning techniques. Even though in [ 39 ], the LSTM model has outperformed MLP and RNN models in predicting PM 10 and SO 2 in the Basaksehir district of Istanbul province. In [ 40 ], authors have proposed a bidirectional LSTM (Bi-LSTM) model by considering both past and future information to forecast PM2.5 of three cities in Korea and five cities in China. Accordingly, its performance is superior to GRU and LSTM in terms of the air quality forecast for these cities. Concretely, with short-term prediction, these models have similar performances. Meanwhile, with long-term prediction, Bi-LSTM outperformed GRU and LSTM. Wang and colleagues [ 41 ] developed a combination of the CT (chi-square test) with the LSTM model to analyze the relationship between air pollution variables. The paper identified the factors influencing air quality by using CT for such a task. Then, the AQI level was predicted by the LSTM model using a dataset collected at Shijiazhuang in the Hebei Province of China. In comparison to other competitive methods (i.e., SVR, MLP, BP neural network, Simple RNN), the proposed method provides an accuracy of 93.7% (the highest one). In addition, the proposed method outperforms the baseline methods in terms of MAE, MSE, and RMSE metrics. In [ 42 ], a GRU layer has been added to the LSTM structure to improve the accuracy of the air quality prediction problem. Experimental results with a dataset collected in Delhi show the outperformance of the proposed approach compared to other competitive methods of linear regression, GRU, kNN, and SVM in terms of MAE and R 2 .

3.3.2 Recurrent Neural Networks (RNN).

In [ 43 ], the authors have proposed an RNN algorithm to predict PM2.5 in Japan by employing a dynamic method to pre-train the model based on multi-step-ahead time series prediction. [ 44 ] apply RNN to predict PM 10, O 3 , SO 2 , CO and NO 2 . The dataset is collected from different sensors with intervals of 1 hour. In addition, the authors applied fine-tuning to find the best hyperparameters of neural network structure and optimization function. Moreover, the investigated model can be applied to predict similar pollutants in other neighboring areas.

3.3.3 Gated Recurrent Unit (GRU).

In the current literature, many research results indicate that existing models are best at short-term forecasts. Meanwhile, improving existing approaches to forecasting long-term air quality is necessary. [ 45 ] proposed an algorithm that is considered an enhanced version of GRU (named BiAGRU) by combining bidirectional gated recurrent unit integrated with an attention mechanism. By means of experimental analysis, the proposed model is superior to many traditional machine learning models and modern deep learning models.

Referring to [ 46 ], a model based on Gated Recurrent Units (GRUs) has been proposed to forecast NO 2 pollutant concentration. The proposed model is assessed and fine-tuned for such a task concerning the number of features, look-backs, neurons, and epochs. Also, in Beijing [ 47 ], authors introduced a model based on spatiotemporal CRUs combined with a Geographic Self-Organizing Map (GeoSOM). Concretely, all monitor stations were clustered using time-series features and geographical coordinates. Later, GRU models were proposed for clusters, and Gaussian vector weights were used to weigh different models in predicting the target sequence. Experimental results showed the technique’s efficiency compared to several state-of-the-art ones regarding MAE, MRE, and R2 metrics.

Since existing models do not fully consider the temporal dependencies, spatial correlations, and feature correlations hidden in a given dataset, in [ 48 ], authors examined these correlations by introducing a spatiotemporal deep learning model named Conv1D-LSTM based on 1-D convolutional neural network and LSTM for spatial and temporal correlation feature extraction. In addition, a fully connected network exploits these features for the air quality prediction problem. Furthermore, missing data have been imputed to enhance the quality of air quality prediction. The proposed method outperformed other well-known baseline methods through experimental analysis.

3.4 Data fusion

Besides techniques for the air quality prediction problems as mentioned above, there exist further works solving the problem by using data fusion. In this section, we will provide a brief discussion of these approaches.

3.4.1 Multimodal data.

Air pollution is one of the most worrying issues facing the world today. So, forecasting of particulate matter (PM) is necessary nowadays. Ton et al. [ 49 ] pointed out that combining meteorological features and timestamp information in Hanoi air quality datasets improved the results of PM10 and PM2.5 forecasting. The authors extracted two new features, which were weekend and working hour , from the “Date Time” recorded variable. Then, encoding the time into a vector of 0, 1 to include two new variables, weekend and working hour . First, with the variable weekend , the time vector from Monday to Friday was encoded as 0. On the other hand, during Saturday and Sunday weekends, the time vector was 1. Second, with working hour , the time vector was 1 in the range 7 AM to 7 PM, whereas this was 0. According to the authors, the time steps of the two new variables weekend and working hour were synchronized with other weather and air quality variables. It was highly efficient in 68% of the cases compared to other methods by conducting five deep learning models: MLP, 1D-CNN, LSTM, Bi-LSTM, and Stacked LSTM. Besides, in the long-term forecast of PM concentrations, the Vanilla LSTM model with combined features performed better than the other.

Similarly, to predict the PM 2.5 air pollution level in the short- and medium-term, Tejima and colleagues [ 50 ] also proposed a framework that looks for hidden associations between traffic factors and air pollution. The six steps in their framework can be defined as follows: (1) Use any machine learning algorithm to extract features from the traffic images, (2) Create a new dataset by combining the extracted features dataset and air pollution dataset using time, (3) Use fuzzy rules to convert this new dataset into an uncertain temporal database, and (4) Use uncertain periodic-frequent pattern mining techniques to uncover hidden relationships between various traffic factors and air pollution, (5) estimate air pollution level from a given image using transfer learning on a pre-trained model, and (6) predict air pollution level using estimated air pollution level and mined patterns dataset. Experimental results show that their method can accurately estimate and predict air pollution levels, ranging from 77% to 98%.

3.4.2 Neighbor stations.

Currently, air pollution and urban life influence human health. Therefore, environmental and data science experts always try to find the most accurate way to predict and provide timely warnings to humans. Specifically, Dao et al. [ 51 ] use methods of data imputation for the UrbanAir dataset to predict air pollution at a place without a station by using neighbor stations and predict air pollution of Dalat and discover the correlation/association between air pollution and human activities. The authors divided the article into two tasks that need to be performed: Subtask 1 only used environmental data to predict air pollution and only used traffic data in Subtask 2. Subtask 2 accepts training a prediction model using environmental and traffic data, but only traffic data is used to predict air pollution. While Subtask 2 only accepts AQI levels, Subtask 1 requires predicting both the exact value and AQI level of each pollutant concentration. The paper encouraged researchers to develop a generic framework to discover a correlation among different traffic factors, weather, and air pollution in a locality. By using these correlations, the authors improved the accuracy of AQI prediction and understood the mutual impact between urban life and air pollution.

Besides, Nguyen et al. [ 52 ] also introduced a dataset containing data about personal life and the surrounding environment, collected periodically along predetermined routes in Ho Chi Minh City, Vietnam. They also introduced self-developed devices and system architectures for data collection, storage, access, and visualization. There were interesting research topics and applications, including understanding the correlation between human health, air pollution, and traffic congestion.

3.4.3 Images.

Human health is mostly impacted by air pollution. Over time, there has been an increase in the number of patients and disease reports related to air pollution. By using lifelog data and urban nature similarity, a method was introduced in [ 53 , 54 ] that could predict AQI at a local and individual scale with a few images taken from smartphones and open AQI and weather datasets. Various public datasets pertaining to weather, air pollution, and images are used to develop and evaluate image retrieval and prediction model techniques. The outcomes support their hypothesis regarding the strong correlation between the AQI and snapshots of the surrounding area.

3.4.4 Variable selection.

Currently, several statistical and machine learning methods are used to uncover useful information and patterns for enormous datasets. The common model selection (variable selection) methods include Neural Networks (NN) and RF. The statistical methods like the Least Absolute Shrinkage And Selection Operator (LASSO) [ 55 ] and principal component analysis (PCA). The authors [ 56 ] have proposed combining NN with LASSO or RF for even better results. In addition, they tested these new methods along with classical techniques (ordinary least square and feed-forward NN) using Monte Carlo simulation and real-world air quality data from Italy. The study found that the combined methods achieved lower errors, suggesting they outperform the traditional approaches.

Many methods have been proposed to improve the performance of air quality prediction. However, most investigated methods are based on complete datasets. Therefore, we need to impute missing values to reinforce the prediction models’ performance.

4 Data imputation: Recent techniques

Various statistical and machine learning methods [ 57 – 59 ] have been developed to overcome the problem of missing data for time series, to fill in the missing values in the data, or in other words, imputing the missing values. However, methods have limitations in handling data with high missing rates or changes in available variables. In addition, the performances of these methods vary widely according to the type of data, noise levels, or other factors and show a high dependence on correlations within the data.

In this section, we want to provide an overview of the relationships among the given imputation techniques and comparisons and then discuss them individually.

4.1 Conventional methods

4.1.1 ignoring..

Ignoring [ 60 , 61 ] is a method that completely ignores missing values when conducting the analysis process. Although this is a simple method, if the rate of missing data is high enough to influence the analysis outcomes, it is highly dangerous.

4.1.2 Deletion.

An approach of removing/deleting missing observations from raw data is called Deletion [ 62 , 63 ]. It is also a frequently used method when the missing values of the data are not high, and removing missing values will not affect the analysis results. Nevertheless, when the missing data rate is high, deleting missing values makes the data incomplete and unsuitable for some other analysis applications.

4.1.3 Mean/Median/Mode imputation.

Mean/Median/Mode are simple methods. There, the missing value for a continuous variable is imputed by the mean/median of the observed values. When the missing values for a categorical variable are replaced by the Mode of the observed values, these approaches are quick to compute and simple to implement. Mean/Median/Mode Imputation methods [ 64 ] are a solution for better analysis results when they solve the issue of handling missing data values, whereas Ignoring and Deletion methods are thought to provide poor results in the analysis or data mining process when the missing data rate is high. Furthermore, the limitation of these methods is that the bias created by multiple values on the data has the same value, even if the data are MCAR. As a result, it may bias the estimation of skewed distributions.

4.1.4 Regression imputation.

There are two steps in Regression imputation [ 65 , 66 ]. The first is to estimate a linear regression model using the target variable’s observed values along with the explanatory variables. After that, one can use the model to predict values for the missing cases in the target variable. Missing values of the variable are replaced based on these predictions. There are two types. First is deterministic regression imputation. It means missing values are replaced with the exact prediction of the regression model. The second is stochastic regression imputation, which adds an additional random error term to the predicted value imputed by deterministic regression imputation. Regression imputation is the improvement over Mean/Median/Mode imputation. Besides, it has disadvantages, including the assumptions of error distribution and linear relationship, which are relatively strict and give poor results for heteroscedastic data.

4.1.5 Last Observation Carried Forward (LOCF).

Last Observation Carried Forward [ 67 , 68 ] fills in missing values by using the last observed value of the given features in each sample; if there is no previous observation, 0 will be filled in. LOCF assumes that the missing data is constant or follows a gradual change. However, if the missing values are not stationary or the sensor readings exhibit abrupt changes, this method may introduce bias and inaccuracies.

4.1.6 Multivariate Imputation by Chained Equations (MICE).

Multiple imputation offers numerous benefits compared to the single imputation methods mentioned above. MICE [ 69 , 70 ] is one of the most popular multiple imputation techniques. The process uses an iterative set of regression models to impute missing data from a dataset. It imputes missing values in the dataset’s variables by focusing on one variable at a time. Once the focus is placed on one variable, MICE uses all the other variables in the dataset to predict missingness in that variable. The prediction is based on a regression model, with the form of the model depending on the nature of the focus variable. MICE methods perform better and are more reliable for data with a limited sample size.

On the other hand, MICE has several benefits, such as results in unbiased estimates, being easily interpreted in a Bayesian context, and having a large number of workable algorithms built into the MICE framework. It is worth noting that MICE is especially helpful when missing values are associated with the target variable in a way that causes leakage. Users can also state what they believe to be the likely distribution of the missing value using MICE. However, MICE comes at a high computational cost.

4.1.7 First five last three logistic regression imputation (FTLRI).

Chen et al. [ 71 ] proposed an interesting approach for data imputation, namely FTLRI, for time-series air quality data. The paper is based on the traditional logistic regression and a presented “first Five & last Three” model. These techniques could explain relationships among disparate attributes and then derive highly relevant data, for both time and attributes, to the missing data, respectively. The results showed that FTLRI has a significant advantage over the compared imputation approaches, particularly in short-term and long-term time-series air quality data. Furthermore, FTLRI can perform better on datasets with relatively high missing rates (about 40%) since it only selects highly relevant data to the missing values instead of relying on all other data like other methods.

4.1.8 Autoregressive Distributed Lag (ARDL).

Selecting criteria is considered an important issue in the Autoregressive Distributed Lag (ARDL) model. El et al. [ 72 ] proposed the use of four imputation methods (k-Nearest Neighbors, Expectation-Maximization, Classification, and Regression Tree, and Random Forest) for handling the missing values. Their goal was to improve the accuracy of the model with the optimal order of lags. They compared these methods using real economic data related to foreign direct investment (FDI) in Libya. Their findings suggest that the Expectation-Maximization method performed best compared to the others.

Next, Mohamed et al. [ 73 ] introduced a new imputation technique called EPK. Using the Monte Carlo simulation, they evaluated the effectiveness of nine different imputation methods, including EPK. The simulations focused on a specific type of statistical model (binary logistic regression) when the missingness mechanism is MAR. Additionally, they tested the methods on real data from social network advertising. The results from both simulations and real-world applications showed that EPK outperformed other imputation methods regardless of where the missing data occurred (independent variables only, dependent variable only, or both).

4.2 Machine-learning approaches

The recent methods for imputing missing data in time series led to more accurate and improved imputed data than traditional approaches. Choosing an appropriate imputation method for a specific type of missing data significantly impacts the performance of data imputation.

4.2.1 k-Nearest Neighbor Imputation (kNNI).

kNNI method [ 14 , 69 ] uses the k-nearest neighbor to identify similar samples with normalized Euclidean distances or some other type of distance and impute the missing values with the average value of its neighbors. The k-nearest neighbor method can impute continuous variables (by using the mean or weighted mean among the k-nearest neighbors) and categorical variables (by using the Mode among the k-nearest neighbors). Both quantitative and qualitative features are handled by kNNI with ease. However, it performs computationally intensively for large data since it searches through all the datasets and requires the specification of hyper-parameters that can greatly affect the results.

4.2.2 MissForest.

The Random Forest (RF) algorithm can also applied for multivariate time series data, employing an average of the corresponding full values. Using proximity data points, this algorithm then iteratively improves the imputation of missing data. Generally, missForest is a technique that was proposed by [ 74 ] based on Random Forests. The article showed that RF intrinsically constitutes a multiple imputation scheme by averaging many unpruned classification or regression trees. The imputation error can be estimated without a test set using Random Forest’s built-in out-of-bag error estimates. Furthermore, missForest performs better than K-nearest neighbors and other imputation techniques, giving outstanding results for data containing non-linear relations and/or complex interactions. Additionally, it works well with data containing both qualitative and quantitative features. When using missForest, there is no need to tune parameters, do categorical encoding, or standardize the data. MissForest can be utilized to achieve good imputation results even in high-dimensional datasets with a large number of variables compared to the sample size.

4.3 Deep neural networks

In addition, many deep learning techniques have been developed to solve imputation for missing values in time series data.

4.3.1 GRU-D.

Chen et al. [ 75 ] proposed the GRU-D model, which is a deep learning model based on Gated Recurrent Unit (GRU) that takes two representations of missing patterns, i.e., masking and time interval, and effectively incorporates them into a deep model architecture so that it does not only captures the long-term temporal dependencies in time series but also utilizes the missing patterns to achieve better prediction results.

4.3.2 Deep auto-encoder.

One method that can be used for data imputation is the auto-encoder structure. It extracts features from low-dimensional layers using the encoder and decoder structure, and the decoder recovers missing values. As such, it can function as a methodological feature. [ 76 ] presented one technique using deep autoencoders for spatiotemporal challenges involving imputing missing data. The proposed method for capturing temporal and spatial patterns was a convolution bidirectional LSTM. Additionally, the authors analyzed an autoencoder’s latent feature representation in spatiotemporal data and illustrated its performance for missing data imputation. The experimental result illustrated that the convolution recurrent neural network outperforms state-of-the-art methods.

4.3.3 MultiLayer Perceptron (MLP).

Next, [ 77 ] estimated the missing values of a variable in multivariate time series data using a MultiLayer Perceptron. To achieve the best prediction performance for the specified time series, an automated technique was employed to identify the optimal MLP model architecture, filling in a long continuous gap instead of relying on isolated, randomly missing observations. The findings demonstrated that using MLP to fill a big gap produces better outcomes, especially when the data behaves nonlinearly.

4.3.4 Raindrop.

Raindrop [ 78 ] is a Graph Neural Network-based algorithm embedding irregularly sampled and multivariate time series. It is inspired by how raindrops hit a surface at varying time intervals and create ripple effects propagating throughout the surface. Raindrop helps handle missing data with irregular time series. It represents every sample as a separate sensor graph and models time-varying dependencies between sensors with a novel message-passing operator. It estimates the latent sensor graph structure and leverages the structure together with nearby observations to predict misaligned readouts. This model can be interpreted as a graph neural network that sends messages over graphs that are optimized for capturing time-varying dependencies among sensors. Another typical work comes from Festag et al. [ 79 ], where the authors developed a system based on Generative Adversarial Networks that consist of recurrent encoders and decoders with attention mechanisms and can learn the distribution of intervals from multivariate time series conditioned on the periods before and, if available, periods after the values that are to be predicted.

Therefore, it is worthwhile to understand the data types with missing values and propose an effective and robust strategy to fill time-series air quality data with missing values.

5 Data imputation for air quality prediction

A flowchart for the setup of the training process for the Machine Learning and Deep Learning framework proposed in this work is shown in Fig 1 .

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0306303.g001

In experiments, to obtain a thorough comparison, we compare some classical by widening traditional imputation methods with some recently developed imputation methods:

• Mean Imputation [ 80 ]: The missing values are replaced with the mean value of the corresponding features.
• Median Imputation [ 81 ]: It is similar to Mean imputation, but the median is utilized instead of the mean.
• Multivariate Imputation by Chained Equations (MICE) [ 69 ]: MICE imputes missing values in the variables of the dataset by focusing on one variable at a time. Once the focus is placed on one variable, MICE uses all the other variables in the dataset to predict missingness in that variable. The prediction is based on a regression model, with the form of the model depending on the nature of the focus variable.
• k-Nearest Neighbor Imputation (kNNI) [ 69 ]: It uses the k-nearest neighbor method to identify similar samples and impute the missing values with the average value of its neighbors. The k-nearest neighbor method can impute continuous variables (the mean or weighted mean among the k-nearest neighbors) and categorical variables (the mode among the k-nearest neighbors).

The main deep learning methods researched for time series imputation are SAITS [ 13 ], BRITS [ 12 ], MRNN [ 82 ], and Tranformer [ 83 , 84 ]. All of them are deep learning approaches published recently for time series imputation.

• Self-attention-based imputation for time series (SAITS) [ 13 ]: a self-attention mechanism for missing value imputation in multivariate time series. Typically, it is trained by a joint-optimization approach. SAITS learns missing values from a weighted combination of two diagonally masked self-attention (DMSA) blocks. DMSA explicitly captures both the temporal dependencies and feature correlations between time steps, which improves imputation accuracy and training speed. Meanwhile, the weighted-combination design enables SAITS to dynamically assign weights to the learned representations from two DMSA blocks according to the attention map and the missingness information.
• Bidirectional Recurrent Imputation for Time Series (BRITS) [ 12 ]: a method for filling the missing values for multiple correlated time series. It learns the missing values in a bidirectional recurrent dynamical system without any specific assumption. The imputed values are treated as variables of the RNN graph and can be effectively updated during the backpropagation.
• Multi-directional Recurrent Neural Network (MRNN) [ 82 ]: is a neural network architecture including two blocks (interpolation and imputation) trained simultaneously. The interpolation process operates within data streams, while the imputation process operates across data streams. The interpolater uses a Bi-directional Recurrent Neural Network (Bi-RNN) to interpolate missing values within each channel along the time dimension. Afterward, using a simple, fully connected neural network, the imputer can compute an estimate for each time step along all channels.
• Transformer [ 83 , 84 ]: Transformer is a self-attention-based model. It uses transformer architecture in an unsupervised manner to perform missing value imputation. Unlike the existing transformer architectures, this model only uses the encoder part of the transformer due to computational benefits. It is a joint-optimization training approach of imputation and reconstruction for self-attention models to perform missing value imputation for multivariate time series.

In the following sections, we will compare different data imputation techniques with various datasets related to air quality prediction.

6.1 Experimental setup

The efficacy of the missing data imputation methods depends heavily on the problem domain, for example, sample size, types of variables, and missingness mechanisms.

We evaluated the methods mentioned in Section 5 on six real datasets. These datasets include cases with small, moderate, and large sample sizes: Air quality in Frankfurt, Germany (Available on: https://www.kaggle.com/datasets/avibagul80/air-quality-dataset ); Beijing Multi-Site Air Quality (Available on: https://archive.ics.uci.edu/dataset/501/beijing+multi+site+air+quality+data ) [ 13 , 85 ]; Air quality in Northern Taiwan (Available on: https://www.kaggle.com/datasets/nelsonchu/air-quality-in-northern-taiwan ); Air Quality in Dalat, Vietnam (Available on: https://github.com/BinhMisfit/air-pollution-datasets/tree/main/Dalat-air-quality-dataset ) [ 51 ] and Air quality dataset in Minh Khai district and Cau Giay district in Hanoi, Vietnam (Available on: https://github.com/BinhMisfit/air-pollution-datasets/tree/main/Hanoi-air-quality-dataset ) [ 49 ].

The descriptions of the six datasets used in this work and their preprocessing details are elaborated on below:

The first dataset is a time-series air quality dataset with categorical contextual information (time and weather); the air pollution PM2.5 values were collected from sense-boxes installed in Frankfurt, Germany. The dataset has been read from 14 different sensors in close spatial proximity. The dataset was efficiently labeled and can be used as a gold-standard dataset for unsupervised problems. Similarly, the test set of this data takes data from 20% original dataset, 20% of the remaining 80% of the original dataset is used for validation, and the remaining is for training. We also chose every 30 minutes of data and every 1-hour consecutive step to generate time series data samples.

The second dataset is Beijing Multi-Site Air-Quality. It includes hourly air pollutant data from 12 monitoring sites in Beijing. This dataset collected data from 01/03/2013 to 28/02/2017 (48 months in total). For each monitoring site, 12 continuous time series variables are measured ( e . g ., PM 2.5, PM 10, SO 2). The test set of the third dataset takes data from 20% original dataset, 20% of the remaining 80% of the original dataset is used for validation, and the remaining is for training. The validation set contains data from the following 05/11/2016. The training set takes from 18/12/2015. In addition, we take every one-hour data to generate a time series of data samples for every 24 consecutive steps.

The third dataset is from the Environmental Protection Administration, Executive Yuan, R.O.C. (Taiwan). It was only collected in Northern Taiwan in 2015, containing air quality and meteorological monitoring data. Besides, this data included 25 air pollution stations and 21 features. The test set takes data from 20% original dataset, 20% of the remaining 80% of the original dataset is used for validation, and the remaining is for training. Specifically, the training set takes place on 15/01/2015, the validation set takes place on 01/09/2015, and the remaining part is used as a test set. We selected every 1-hour data and every 12 consecutive steps in experiments to generate time series data samples.

The fourth dataset is Dalat Air Quality. Urban Air provides a streaming dataset from CCTV and air station networks installed in Dalat City, Vietnam. The system runs 24 × 7 and has several real problems, such as sudden camera/station turn-on/switch-off, noise, and outliers. There are ten air pollution stations (i.e., sensors 01-10), three attached to weather stations (i.e., sensor01, sensor02, sensor03), and fourteen CCTV cameras. Furthermore, the test set of the dataset takes data from 20% original dataset, 20% of the remaining 80% of the original dataset is used for validation, and the remaining is for training. We generate time series data samples by selecting every 1-hour data and every 24 consecutive steps.

The two final datasets were collected hourly at two monitoring stations in Hanoi: Cau Giay district and Minh Khai district. For example, Cau Giay dataset with observation time from 25/2/2019 to 25/11/2020, and Minh Khai dataset from 01/1/2019 to 25/11/2020 record measured features including PM 10, PM 2.5, SO 2 , O 3 , NO 2 , NO , NO x , CO , Temperature, Humidity, Wind speed, Rain, Wind direction, Atmospheric Pressure, Solar Radiation. Besides, 20% of the original dataset is utilized for the test set, 20% of the remaining 80% is used for validation, and the remaining is used for training. Then, we choose one hour’s worth of data per 24 consecutive steps to create time series data samples.

After the preprocessing step, general information about the datasets is described in Table 1 .

https://doi.org/10.1371/journal.pone.0306303.t001

We generate artificial missingness to evaluate all imputation methods used. It is important to note that normalization is applied in the preprocessing of all datasets. For each dataset, the missing ratio p is varied from 10% to 80% with increments of 10% for each dataset to evaluate the models at different missing ratios. For p ∈ {10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%}, we train the model to fill the missing values and then calculate the imputation accuracy. However, with the high missing rates in the original Dalat dataset (greater than 40%) for this dataset, we generate extra artificial missing values with missing rates of 10%–30% only.

Besides, the specific information about the architectures of SAITS, BRITS, MRNN, and Transformer models in this paper can be described in Table 2 .

https://doi.org/10.1371/journal.pone.0306303.t002

For multiple imputation cases with K imputations, we have K values for MAE and RMSE per dataset, and we use the average to evaluate the model performance. We designed each experiment 10 times. We report mean MAE and RMSE, along with their running time, as the performance metrics.

In this paper, all the models were trained/tested on a computer with the following configurations: Intel(R) Xeon,(R) Gold 6254 CPU @3.10GHz/512GB RAM.

Detailed experimental results and the running time of imputation methods on six datasets are recorded in Tables 3 – 8 and described in Figs 2 – 7 .

https://doi.org/10.1371/journal.pone.0306303.g002

https://doi.org/10.1371/journal.pone.0306303.g003

https://doi.org/10.1371/journal.pone.0306303.g004

Due to the high missing rates in the original Dalat dataset greater than 40%, we generate extra artificial missing values with missing rates of 10%–30% only.

https://doi.org/10.1371/journal.pone.0306303.g005

https://doi.org/10.1371/journal.pone.0306303.g006

https://doi.org/10.1371/journal.pone.0306303.g007

https://doi.org/10.1371/journal.pone.0306303.t003

https://doi.org/10.1371/journal.pone.0306303.t004

https://doi.org/10.1371/journal.pone.0306303.t005

https://doi.org/10.1371/journal.pone.0306303.t006

https://doi.org/10.1371/journal.pone.0306303.t007

https://doi.org/10.1371/journal.pone.0306303.t008

6.2 Results on different datasets

6.2.1 frankfurt dataset..

Table 3 presents the results of the traditional imputation method compared with other imputation methods regarding the experiment accuracy and running time on the Frankfurt air quality dataset. One can see that when the missing rate of the dataset increases from 10% to 80%, Mean and Median methods have an almost constant MAE error (fluctuating around 0.787–0.788); the MRNN model gives the highest error (greater than 0.904). Moreover, the Transformer model has an MAE error from 0.651 to 0.77; the kNNI model alone has the smallest MAE and RMSE errors among the remaining machine learning models, such as MICE, even lower than the currently used neural-network models, such as SAITS and BRITS. In general, in this dataset, the MAE and RMSE errors of the kNNI model both give the lowest and most stable results among the remaining missing data models when the missing rate of the original data is 0%, and the artificial missing data rate gradually increases from 10% to 80%. As depicted in Fig 2c , it is worth noting that the running time of the models used in this dataset mainly increases when the missing rate of the dataset changes from 10% to 80%. On the other hand, compared to traditional models or basic machine learning models, although models based on neural networks have a long calculation time, MRNN gives relatively positive results and is the most effective among the models using neural networks in terms of Mean, Median, and MICE.

On the other hand, when the original dataset is not missing and the data size is larger than one million records (for example, Frankfurt air quality data), kNNI is considered the model that gives the best results with an artificial missing rate of 10%–80% and time execution time gradually increases from 62.684 × 10 3 milliseconds to 941.832 × 10 3 milliseconds, followed by SAITS with computation time decreasing from 843.803 × 10 3 milliseconds to 391.645 × 10 3 milliseconds.

6.2.2 Beijing dataset.

Table 4 depicts the results of imputation methods compared with other imputation methods on the Beijing air quality dataset. In this dataset, the SAITS model also gives the lowest MAE measure compared to other machine learning models; the model error varies from 0.142 to 0.349, followed by BRITS. Meanwhile, the traditional data-filling models vary from 0.724 to 0.885 as the missing ratio gradually increases from 10% to 80%. On the other hand, the MAE error of the SAITS model when the artificial missing rate of data changes from 10% to 40% is lower than that of the MICE model; on the contrary, the RMSE error of MICE is lower than that of SAITS, and the lowest in the remaining used models such as BRITS, kNNI, Transformer, Mean, Median, and MRNN. The artificial missing rate of data ranges from 50% to 70%, the MAE and RMSE errors of the SAITS model are stable again, and the experimental results obtained are the smallest among the models. The remaining models give a relatively large error, with MRNN having the largest error. When the missing data rate is at 80%, the kNNI model gives the best MAE and RMSE errors compared to traditional data filling or machine learning models. In addition, Fig 3c shows the computational time of models such as Median, MRNN, MICE, SAITS, and BRITS remains almost constant when the missing rate increases from 10% to 80%. Next, kNNI is a model with large fluctuations in calculation time, gradually increasing as the missing rate increases. Moreover, the calculation time of the Transformer gradually decreases and changes sharply as the missing ratio increases. Compared to traditional machine learning models, MRNN has the most stable and fastest calculation time compared to the remaining models in this dataset.

6.2.3 Taiwan dataset.

Table 5 shows that the MAE error between traditional models and current models using neural networks grows larger as the missing rate of input temporal data increases on the Northern Taiwan air quality dataset. Specifically, SAITS is the model with the lowest error (ranging from 0.121 to 0.284), followed by BRITS (error only from 0.136 to 0.29) in this data. Meanwhile, methods such as Mean, Median, kNNI, or MICE give errors when filling in missing values that deviate greatly from the original value. Besides, when the artificial missing rate of the data changes from 10% to 60%, the MAE and RMSE errors of the SAITS model give the lowest results. However, when the missing data rate reaches 70%, the SAITS model’s MAE error is the smallest, but the RMSE error is higher than that of the kNNI model. When the missing rate is from 80%, the MAE and RMSE error results of kNNI are the lowest among the models, followed by BRITS.

On the one hand, in Fig 4c , we can see the computational time of machine learning models like kNNI is the smallest after traditional missing data filling models like Mean and Median. Besides, MRNN is the model with the least computational time among machine learning models, followed by MICE. The remaining models have fluctuating and irregular calculation times, the highest when the missing data rate is 10%–30%, and the lowest when the missing data rate is 40%–50%. By comparing the performance of the Mean, Median, kNNI, MICE, SAITS, BRITS, MRNN, and Transformer models, we see that SAITS is the best missing data imputation model on Northern Taiwan and Beijing air quality dataset with an artificial missing rate under 70%. One can see that when the original missing rate of the dataset is less than 30% (or 10%). The dataset only has a few hundred thousand records. SAITS seems to be the model with the lowest error, and model execution time also gradually decreased (from 4223.126 × 10 3 milliseconds to 1117.453 × 10 3 milliseconds with the Northern Taiwan dataset and from 403.223 × 10 3 milliseconds to 259.380 × 10 3 milliseconds with the Beijing dataset) as the missing rate of the dataset increased.

6.2.4 Dalat dataset.

Table 6 presents similar experimental results on the Dalat air quality dataset [ 51 ]. In this table, traditional methods such as Mean and Median give a constant MAE measure (about 0.83–0.86) when the missing rate of data changes from 10% to 30% and almost the result of these measures is the largest compared to the remaining missing data filling models. Meanwhile, the neural network models used in this dataset, such as SAITS and BRITS, give optimal results, which are not much different from traditional machine learning models such as kNNI and MICE (for MAE measurement, the shortest). However, when the artificial missingness ratio of the data is at 10%, MICE gives relatively low MAE and RMSE errors among the models. Furthermore, when increasing the artificial missing rate to 20%, although the MAE error of MICE is the lowest, the RMSE error of the BRITS model is the smallest. Next, when continuing to increase the artificial missing rate of the model to 30%, the experimental results, MICE is the model with the smallest MAE, and kNNI is the model with the lowest RMSE of all. On the other hand, the calculation time of most models increases when the data’s artificial missing rate increases. Accordingly, MRNN is the model with the fastest computation time, followed by SAITS, Transformer, BRITS, and MICE in Fig 5c .

6.2.5 Cau Giay dataset.

The experimental results on the Cau Giay District air quality dataset [ 49 ] are presented in Table 7 . One can see that the MAE errors of MICE showed the best results among the used models (only from 0.241–0.424) when the artificial missing rate of the data gradually increased from 10% to 80%. The second best is SAITS (from 0.289–0.578), and the third one is BRITS (from 0.337–0.666). However, the RMSE error of SAITS is the lowest with artificial missing rates of 10%–30% and 50%–60%. Meanwhile, when the missing rate is 40% and increases to 70%–80%, MICE almost always gives relatively good results compared to the remaining models. Besides, the MAE and RMSE errors of the models become larger when the missing rate changes from 10% to 80%, especially the MAE and RMSE errors of the two methods Mean and Median are large, only fluctuating around 0.8. In addition, the running time of machine learning and neural network models is significantly slower than Mean imputation (0.379–0.509 milliseconds) and Median imputation (0.933–0.582 milliseconds). Also, the slowest is MICE, with a relatively large running time, ranging from 71.192 × 10 3 to 71.527 × 10 3 milliseconds.

6.2.6 Minh Khai dataset.

Table 8 presents experimental results on the Minh Khai air quality dataset [ 49 ]. Although the experimental time of MICE is the largest, the model gives the most optimal MAE and RMSE error results among the models used, followed by SAITS, BRITS, kNNI, and Transformer. The MAE and RMSE errors of the Mean and Median methods hardly change much when increasing the missing data rate from 10% to 80%; MAE and RMSE errors of the remaining models gradually increase as the missing data rate increases.

Furthermore, one can also see that when the original missing rate of data is less than 5%, MICE is the method that gives the most optimal results among the missing data imputation methods used in this article (specifically with the Cau Giay district and Minh Khai district air quality dataset). Besides, the running time of MICE is high and increases the fastest when the missing rate of data gradually increases with the Minh Khai dataset. On the other hand, the Cau Giay dataset does not change much over time. However, when considering the performance of filling in missing values using neural networks, SAITS is the model with the most optimal performance, followed by BRITS. We knew that Transformer is a deep-learning model designed to solve many problems. However, in this study, we can see that Transformer hardly promotes its strengths, and experimental results on different data all give much larger MAE and RMSE errors than SAITS and BRITS.

6.3 The impact of different numbers of layers

Based on the experimental results presented above, although MICE gives a better MAE measure than SAITS in some cases (specifically in datasets in Vietnam), the running time of MICE is many times longer than that of SAITS. Therefore, we propose SAITS as the model to fill in missing values for the air quality data for the multivariate time series of those tested in this paper. We now perform another test and compare the results when performing missing data filling on air quality data of SAITS with the Transformer model with the different number of layers cases (i.e., two layers, four layers, and six layers) and the BRITS model. We then propose the best model to fill in the last missing value before predicting air quality for the following year.

The result of evaluating Transformer and SAITS with two-layer, four-layer, and six-layer for both datasets is presented accordingly in Fig 8 . From there, one can see that SAITS with two, four, and six layers do not show as clearly as Transformer with two, four, and six layers in the six datasets, including Frankfurt, Beijing, and Taiwan. However, for the Dalat, Cau Giay, and Minh Khai datasets, we can see the results of both SAITS and Transformer with two, four, and six layers clearly shown. As a result, SAITS with two layers performs better than Transformers with two, four, and six layers. Besides, the RMSE performance of SAITS and Transformer with layers of all datasets in Fig 9 is unclear and changes frequently. In contrast, the running time of SAITS and Transformer with two layers in Fig 10 for both datasets is also a more effective model with four layers and six layers.

https://doi.org/10.1371/journal.pone.0306303.g008

https://doi.org/10.1371/journal.pone.0306303.g009

https://doi.org/10.1371/journal.pone.0306303.g010

6.4 Predict air quality in Vietnam combined AQI indexes

Based on the results and evaluating the performance of the methods with different numbers of layers above, we propose the SAITS model with two layers as the most optimal missing value estimation technique with a small sample size on the three air quality datasets in Vietnam.

We started predicting factors affecting air pollution in the following 24 hours based on Vector Auto-Regression (VAR) and related features. Next, we analyzed daily air pollution levels across countries on three datasets in Vietnam, which were mentioned based on the Air Quality Index (AQI). The AQI level of each pollutant is calculated according to the instruction (Available on: https://www.airnow.gov/sites/default/files/2020-05/aqi-technical-assistance-document-sept2018.pdf ) and divided into seven levels (i.e., Good, Moderate, Unhealthy for Sensitive Groups, Unhealthy, Very Unhealthy, Hazardous, and Extreme Hazardous). It only includes six air pollutants (i.e., PM 2.5, PM 10, CO , NO 2 , SO 2 , and O 3 ) that are required to predict their values and related AQI levels. This index tells us how clean or polluted the air is and what it means to human health. The higher the AQI index value, the greater the level of air pollution and the more negative impact it has on human health.

Additionally, we also perform the outlier detection for the datasets using Mahalanobis distance, where points with large Mahalanobis distance (greater than 97.5%) are considered to be outliers. However, we didn’t remove the outliers from the datasets because deleting them will make the data become irregularly sampled time series data, and VAR is not designed for that type of data.

We analyze indexes such as SO 2 ; CO ; PM 10; PM 2.5; O 3 recorded by sensors in Dalat, Vietnam. The concentration of PM 2.5 collected at some stations is large (ranging from 50–100 μg / m 3 ), but the air quality collected from stations in this place is at an acceptable average level, some stations exceeding the threshold range from 100–106 μg / m 3 but at a poor level and affecting sensitive groups of people. (Similar to the concentration of fine particles PM 10). Meanwhile, the levels of CO , SO 2 , and O 3 in the air are low, only from 1.09–10.7 μg / m 3 , completely at a good level and do not have much impact on human health. In general, from the end of 2019 to the beginning of 2020, the concentration of these indicators increased and posed serious harm to human health. Besides, some other factors can interfere with the fluctuation of air pollution in Dalat City, known as the city of thousands of flowers, and attract many people, especially young people, to visit and relax during festivals or on the weekend. Next, Dalat is geographically located in a mountain valley and has cool weather that may impact air pollution and some activities.

Next, we also analyzed factors such as PM 10, PM 2_5, CO , SO 2 , O 3 recorded by two districts of Cau Giay and Minh Khai in Hanoi, Vietnam. Fine dust concentrations PM 10 and PM 2.5 in the Cau Giay district are mostly at moderate levels according to the AQI categories table, and they do not cause serious effects on human health. However, on some days when the concentration of these indicators is recorded at a high level, greater than 100 μg / m 3 , and even on some days, the concentration of fine dust in some time frames exceeds 300 μg / m 3 , seriously affecting the health of the people of the Capital. In addition, the number of motorbikes and vehicles circulating in Hanoi is quite large, which is also the cause of air pollution here; CO concentration is greater than 1000 μg / m 3 . For example, from 22/2/2023-23/2/2023, CO concentration was recorded above 10, 000 μg / m 3 at an alarming level. In addition, the concentration of SO 2 is relatively low, possibly because this area does not have many manufacturing plants, so the amount of toxic chemicals such as SO 2 released into the environment is small, at a good level, and does not affect the human health. From the end of 03/2019 to mid-04/2019, the concentration of O 3 gradually increased, from 100–350 μg / m 3 , changing from a level that is not harmful to human health to a seriously harmful level. Besides, the concentration of fine dust particles PM 10, PM 2.5 in the air recorded in the Minh Khai district dataset is also quite high, which is not good for human health. Some days, the fine dust concentration of these particles exceeds 479 μg / m 3 and 255 μg / m 3 , respectively PM 10 and PM 2.5. Furthermore, the amount of SO 2 and O 3 in the Minh Khai district dataset is quite low, at a normal level according to the AQI categories scale, so it does not affect human health. However, the concentration of CO in the air is very high; most days, the concentration is recorded to be greater than 1000 μg / m 3 , which is considered “Exceeding AQI.” Recommendations for hazard classification should be implemented. In summary, Hanoi is known to domestic and foreign friends as the Capital of Vietnam, a place to work and welcome heads of state. Based on the air pollution problem in the Cau Giay and Minh Khai districts, we need to take many measures to reduce air pollution and improve the environment to attract tourists, creating economic development and international cooperation conditions.

7 Discussion

There are many estimation techniques to handle missing data. A lot of them focus on the missing values of the time series. However, these techniques might not be able to capture time information and produce reliable imputation results if timestamps are missing. Therefore, expanding on current techniques to impute missing timestamps may be suitable for handling such situations. Research on deep learning modeling is a being cared area. Numerous novel deep-learning models with practical applications have been proposed in recent years. There is a growing number of research papers on deep learning for imputing missing data, and these new approaches seem promising. However, in practical applications, the validity and strength of these models must be carefully assessed. In many cases, reliable and reproducible code is frequently unavailable or incomplete. Compared to conventional statistical methods, the number of hyperparameters for deep learning models typically requires much more tuning. In some applications, the hyperparameter search on big data may be prohibitively expensive due to the required training time or memory size. However, our research showed that for data with small, moderate, or even large sample sizes (i.e., when the sample size is less than 30,000), the stability and convergence of the deep learning models needed to be revised.

Numerous factors, including the sample size of the data, the distribution of the variables, the number of missing values in the data, the correlation structure of the data, and potential missingness mechanisms, affect how effective different missing data imputation techniques are.

While this work presents a thorough investigation of time series imputation techniques and provides practical implications for practitioners, it also has some limitations. For example, different types of geographic locations, such as mountains and plains, may affect air quality. Collecting more datasets and examining the patterns for various types of geographic locations may help draw out common patterns and insights to improve the imputation quality. However, due to limited data available, we have not achieved that goal yet. In addition, this work so far has only concentrated on examining the effect of missing data for the missing at random pattern. However, it is also possible that air quality data is missing, not at random. These will be topics for our future research.

8 Conclusions

We have presented an investigation of the impact of data imputation techniques on the air quality prediction problem. In general, SAITS gives the lowest error, and the difference in running time is negligible compared to the missing value imputation efficiency that SAITS provides when the missing rate increases. Besides, BRITS is the model that gives the second-best error among deep learning models, only after SAITS. kNNI running time can increase significantly as the missing rate increases. However, this may not be true for other methods. Also, kNNI proves itself to remain a promising imputer for the dataset. In addition, for three datasets (Northern Taiwan, Beijing, and Frankfurt) which have a large sample size, at the high missing rate of 80%, kNNI outperforms other techniques, including the state-of-the-art such as SAITS, BRITS, and Transformer. Meanwhile, other cases were varied using limited sample size and ratios of missing data. The experiment results show the conventional method, MICE, outperforms the recently proposed deep learning methods, such as SAITS and BRITS, in these experiments.

The outcomes of this article can open a new direction for predicting air pollution. However, as discussed, it also has some limitations, such as the experiments concentrated only on missingness at random, and the paper has not been able to draw insights by grouping datasets based on geological characteristics due to limited data. Therefore, in the future, we will collect more datasets to examine the imputation quality based on types of geographic locations or some other characteristics and consider the imputation effects for nonrandomly missing data. Also, we will develop an ensemble technique combining the latest missing techniques and SAITS to enhance the imputation performance. Moreover, the process can integrate the knowledge in a particular domain to support extracting helpful information from datasets [ 86 ]. In addition, in the future, we plan to empirically evaluate the performance of imputation techniques for other types of environmental data, such as imbalanced missing data. Last but not least, we want to investigate if methods to combine datasets such as ComImp [ 87 ] can be used to combine air quality datasets to improve the imputation and prediction quality.

Acknowledgments

We want to thank the University of Science, Vietnam National University in Ho Chi Minh City, and AISIA Research Lab in Vietnam for supporting us throughout this paper.

• 1. Duong DQ, Le QM, Nguyen-Tai TL, Nguyen HD, Dao MS, Nguyen BT. An effective AQI estimation using sensor data and stacking mechanism. In: Proceedings of the 20th International Conference on New Trends in Intelligent Software Methodologies, Tools and Techniques (SoMeT 21). vol. 337. IOS Press; 2021. p. 405–418.
• 2. Vu MA, Nguyen T, Do TT, Phan N, Halvorsen P, Riegler MA, et al. Conditional expectation for missing data imputation. arXiv preprint arXiv:230200911. 2023;.
• 3. Shaadan N, Rahim N. Imputation analysis for time series air quality (PM10) data set: A comparison of several methods. In: Journal of Physics: Conference Series. vol. 1366. IOP Publishing; 2019. p. 012107.
• View Article
• Google Scholar
• 6. Sridevi S, Rajaram S, Parthiban C, SibiArasan S, Swadhikar C. Imputation for the analysis of missing values and prediction of time series data. In: 2011 international conference on recent trends in information Technology (ICRTIT). IEEE; 2011. p. 1158–1163.
• 7. Sitaram D, Dalwani A, Narang A, Das M, Auradkar P. A measure of similarity of time series containing missing data using the mahalanobis distance. In: 2015 second international conference on advances in computing and communication engineering. IEEE; 2015. p. 622–627.
• 9. Dhevi AS. Imputing missing values using Inverse Distance Weighted Interpolation for time series data. In: 2014 Sixth international conference on advanced computing (ICoAC). IEEE; 2014. p. 255–259.
• 11. Le Lien P, Do TT, Nguyen T. Data imputation for multivariate time-series data. In: 2023 15th International Conference on Knowledge and Systems Engineering (KSE). IEEE; 2023. p. 1–6.
• PubMed/NCBI
• 19. Peña M, Ortega P, Orellana M. A novel imputation method for missing values in air pollutant time series data. In: 2019 IEEE Latin American Conference on Computational Intelligence (LA-CCI). IEEE; 2019. p. 1–6.
• 21. Nguyen DH, Nguyen-Tai TL, Nguyen MT, Nguyen TB, Dao MS. MNR-Air: An economic and dynamic crowdsourcing mechanism to collect personal lifelog and surrounding environment dataset. A case study in Ho Chi minh city, Vietnam. In: MultiMedia Modeling: 27th International Conference, MMM 2021, Prague, Czech Republic, June 22–24, 2021, Proceedings, Part II 27. Springer; 2021. p. 206–217.
• 23. Le DD, Tran AK, Dao MS, Nazmudeen MSH, Mai VT, Su NH. Federated Learning for Air Quality Index Prediction: An Overview. In: 2022 14th International Conference on Knowledge and Systems Engineering (KSE). IEEE; 2022. p. 1–8.
• 26. Gholamzadeh F, Bourbour S. Air pollution forecasting for Tehran city using vector auto regression. In: 2020 6th Iranian Conference on Signal Processing and Intelligent Systems (ICSPIS). IEEE; 2020. p. 1–5.
• 27. Wijesekara W, Liyanage L. Comparison of imputation methods for missing values in air pollution data: Case study on Sydney air quality index. In: Advances in Information and Communication: Proceedings of the 2020 Future of Information and Communication Conference (FICC), Volume 2. Springer; 2020. p. 257–269.
• 29. Ngom B, Diallo M, Seyc MR, Drame MS, Cambier C, Marilleau N. PM10 data assimilation on real-time agent-based simulation using machine learning models: case of dakar urban air pollution study. In: 2021 IEEE/ACM 25th International Symposium on Distributed Simulation and Real Time Applications (DS-RT). IEEE; 2021. p. 1–4.
• 30. Sotomayor-Olmedo A, Aceves-Fernandez MA, Gorrostieta-Hurtado E, Pedraza-Ortega JC, Vargas-Soto JE, Ramos-Arreguin JM, et al. Evaluating trends of airborne contaminants by using support vector regression techniques. In: CONIELECOMP 2011, 21st International Conference on Electrical Communications and Computers. IEEE; 2011. p. 137–141.
• 32. Zhu H, Hu J. Air quality forecasting using SVR with quasi-linear kernel. In: 2019 International Conference on Computer, Information and Telecommunication Systems (CITS). IEEE; 2019. p. 1–5.
• 33. Zhang C, Yuan D. Fast fine-grained air quality index level prediction using random forest algorithm on cluster computing of spark. In: 2015 IEEE 12th Intl Conf on Ubiquitous Intelligence and Computing and 2015 IEEE 12th Intl Conf on Autonomic and Trusted Computing and 2015 IEEE 15th Intl Conf on Scalable Computing and Communications and Its Associated Workshops (UIC-ATC-ScalCom). IEEE; 2015. p. 929–934.
• 34. Duong DQ, Le QM, Nguyen-Tai TL, Bo D, Nguyen D, Dao MS, et al. Multi-source machine learning for aqi estimation. In: 2020 IEEE International Conference on Big Data (Big Data). IEEE; 2020. p. 4567–4576.
• 35. Li J, Shao X, Zhao H. An online method based on random forest for air pollutant concentration forecasting. In: 2018 37th Chinese Control Conference (CCC). IEEE; 2018. p. 9641–9648.
• 36. Barthwal A, Acharya D. An internet of things system for sensing, analysis & forecasting urban air quality. In: 2018 IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT). IEEE; 2018. p. 1–6.
• 44. Lim YB, Aliyu I, Lim CG. Air pollution matter prediction using recurrent neural networks with sequential data. In: Proceedings of the 2019 3rd International Conference on Intelligent Systems, Metaheuristics & Swarm Intelligence; 2019. p. 40–44.
• 49. Ton-Thien MA, Nguyen CT, Le QM, Duong DQ, Dao MS, Nguyen BT. Air Pollution Forecasting Using Multimodal Data. In: International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems. Springer; 2023. p. 360–371.
• 50. Tejima K, Dao MS, Zettsu K. Mm-aqi: A novel framework to understand the associations between urban traffic, visual pollution, and air pollution. In: International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems. Springer; 2022. p. 597–608.
• 51. Dao MS, Dang TH, Nguyen-Tai TL, Nguyen TB, Dang-Nguyen DT. Overview of MediaEval 2022 Urban Air: Urban Life and Air Pollution. In: Proc. of the MediaEval 2022 Workshop; 2023. p. 13–15.
• 52. Nguyen-Tai TL, Nguyen DH, Nguyen MT, Nguyen TD, Dang TH, Dao MS. Mnr-hcm data: A personal lifelog and surrounding environment dataset in ho-chi-minh city, viet nam. In: Proceedings of the 2020 on Intelligent Cross-Data Analysis and Retrieval Workshop; 2020. p. 21–26.
• 53. La TV, Dao MS, Tejima K, Kiran RU, Zettsu K. Improving the awareness of sustainable smart cities by analyzing lifelog images and IoT air pollution data. In: 2021 IEEE International Conference on Big Data (Big Data). IEEE; 2021. p. 3589–3594.
• 54. Dao MS, Zettsu K, Rage UK. Image-2-aqi: Aware of the surrounding air qualification by a few images. In: Advances and Trends in Artificial Intelligence. From Theory to Practice: 34th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, IEA/AIE 2021, Kuala Lumpur, Malaysia, July 26–29, 2021, Proceedings, Part II 34. Springer; 2021. p. 335–346.
• 64. Aljuaid T, Sasi S. Proper imputation techniques for missing values in data sets. In: 2016 International Conference on Data Science and Engineering (ICDSE). IEEE; 2016. p. 1–5.
• 68. Zhou H, Yu KM, Lee MG, Han CC. The application of last observation carried forward method for missing data estimation in the context of industrial wireless sensor networks. In: 2018 IEEE Asia-Pacific Conference on Antennas and Propagation (APCAP). IEEE; 2018. p. 1–2.
• 69. Zainuddin A, Hairuddin MA, Yassin AIM, Abd Latiff ZI, Azhar A. Time Series Data and Recent Imputation Techniques for Missing Data: A Review. In: 2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST). IEEE; 2022. p. 346–350.
• 76. Asadi R, Regan A. A convolution recurrent autoencoder for spatio-temporal missing data imputation. arXiv preprint arXiv:190412413. 2019;.
• 78. Zhang X, Zeman M, Tsiligkaridis T, Zitnik M. Graph-guided network for irregularly sampled multivariate time series. arXiv preprint arXiv:211005357. 2021;.
• 85. Chen S. Beijing Multi-Site Air Quality; 2019. UCI Machine Learning Repository.
• 87. Nguyen T, Khadka R, Phan N, Yazidi A, Halvorsen P, Riegler MA. Combining datasets to improve model fitting. In: 2023 International Joint Conference on Neural Networks (IJCNN); 2022.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 02 September 2024

Transforming air pollution management in India with AI and machine learning technologies

• Kuldeep Singh Rautela 1 &
• Manish Kumar Goyal 1  

Scientific Reports volume  14 , Article number:  20412 ( 2024 ) Cite this article

133 Accesses

3 Altmetric

Metrics details

• Engineering
• Environmental chemistry

A comprehensive approach is essential in India's ongoing battle against air pollution, combining technological advancements, regulatory reinforcement, and widespread societal engagement. Bridging technological gaps involves deploying sophisticated pollution control technologies and addressing the rural–urban disparity through innovative solutions. The review found that integrating Artificial Intelligence and Machine Learning (AI&ML) in air quality forecasting demonstrates promising results with a remarkable model efficiency. In this study, initially, we compute the PM 2.5 concentration over India using a surface mass concentration of 5 key aerosols such as black carbon (BC), dust (DU), organic carbon (OC), sea salt (SS) and sulphates (SU), respectively. The study identifies several regions highly vulnerable to PM 2.5 pollution due to specific sources. The Indo-Gangetic Plains are notably impacted by high concentrations of BC, OC, and SU resulting from anthropogenic activities. Western India experiences higher DU concentrations due to its proximity to the Sahara Desert. Additionally, certain areas in northeast India show significant contributions of OC from biogenic activities. Moreover, an AI&ML model based on convolutional autoencoder architecture underwent rigorous training, testing, and validation to forecast PM 2.5 concentrations across India. The results reveal its exceptional precision in PM 2.5 prediction, as demonstrated by model evaluation metrics, including a Structural Similarity Index exceeding 0.60, Peak Signal-to-Noise Ratio ranging from 28–30 dB and Mean Square Error below 10 μg/m 3 . However, regulatory challenges persist, necessitating robust frameworks and consistent enforcement mechanisms, as evidenced by the complexities in predicting PM 2.5 concentrations. Implementing tailored regional pollution control strategies, integrating AI&ML technologies, strengthening regulatory frameworks, promoting sustainable practices, and encouraging international collaboration are essential policy measures to mitigate air pollution in India.

Similar content being viewed by others

Optimized air quality management based on air quality index prediction and air pollutants identification in representative cities in China

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

Disentangling drivers of air pollutant and health risk changes during the COVID-19 lockdown in China

Introduction.

Air pollution has emerged as a critical global environmental health issue, with 92% of the world's population exposed to pollutant levels exceeding air quality guidelines 1 , 2 . This widespread exposure poses significant health risks, including increased incidence of respiratory diseases, cardiovascular problems, and premature mortality 3 , 4 . In India specifically, ambient particulate matter (PM) exposure has been linked to an estimated 1.1 million premature deaths annually, with air pollution becoming the fourth leading cause of mortality nationwide 5 , 6 . The economic impact is also substantial, with the World Bank estimating that air pollution costs India 3–8% of its GDP due to healthcare expenses, reduced productivity, and premature deaths 7 .

Atmospheric aerosols, particularly black carbon, organic carbon, dust, sea salt, and sulfates, have been extensively researched in South and Southeast Asia over the past two decades 8 . However, the magnitude of these impacts is largely influenced by spatio-temporal variability and the composition of these aerosols 9 . Aerosols, including, are significant constituents of atmospheric PM and account for approximately 30–70% of the fine aerosol mass over urban areas in India 5 , 10 . However, in recent decades, this concern has increased notably, primarily attributed to the rapid surge in population, unplanned urban development, and the expansion of industries 11 , 12 . India, home to the world's largest population share at 17.76%, faces a significant environmental challenge, with many of its cities (eg; Delhi, Mumbai, Kolkata) ranking among the most polluted on the global scale 13 , 14 . An investigation based on World Health Organization (WHO) data from 2008–2013 brought attention to India's status among the most polluted nations 15 . India has faced alarming and extensive air pollution incidents in the last twenty years, prompting substantial concern among regulatory authorities. The Indo-Gangetic Plain (IGP) is highly susceptible to severe pollution incidents, notably prevalent in the post-monsoon and winter period 16 . Similarly, in many metropolitan cities across India, such as Delhi, air quality has deteriorated to hazardous levels. Concentrations of particulate matter (PM 2.5 and PM 10 ) have surged beyond 500 µg/m 3 , while nitrogen oxides (NO 2 ) have exceeded 10 µg/m 3 . Additionally, ozone (O 3 ) and sulfur dioxide (SO 2 ) levels have surpassed 5 µg/m 3 , alongside other pollutants 17 . The concentration of these pollutants, often surpassing 500 µg/m 3 , far exceeds WHO's safe annual limit of 10 µg/m 3 and India’s national ambient air quality standards (NAAQS) of 40 µg/m 3 during winters 18 . According to the Economic Times, 12.25 million vehicles are registered in Delhi, growing at a rate of 7% per annum, and they account for 67% of the total pollution 19 , 20 . Additionally, Coal-based thermal power plants and small-scale industries each contribute 12% to the pollution, including emissions from various industrial units followed by the agricultural and biomass burning in Delhi and surrounding areas 20 . This increased pollution level has raised considerable concern among authorities and stakeholders, prompting focused efforts towards addressing this critical issue 9 . The urgency of addressing air pollution in India is evident through compelling data illustrating its significant impact across various sectors.

AI & ML have become pivotal in addressing air pollution by harnessing big data analytics, utilising advanced computing systems, scalable storage, and parallel processing technologies 21 , 22 , 23 . These innovations enable comprehensive management and mitigation strategies for various air pollutants, bridging the gap between atmospheric and climate sciences through sophisticated data-driven approaches. Previous studies have proposed various AI&ML-based models as pivotal components for air pollution and aerosol transport 5 , 8 , 24 , 25 , 26 . Initially, researchers have introduced succinct and efficient statistical models for practical applications. These statistical models primarily encompass multiple linear regression (MLR) 27 and autoregression moving average (ARMA) 28 methods. The predominant use of linear hypotheses in developing statistical models contrasts with the inherent nonlinear properties exhibited by pollutant concentrations. Consequently, researchers have advocated for integrating data mining methods 29 and machine learning models 30 , 31 , 32 designed to accommodate nonlinear predictions in studying air pollutants. However, the notably nonlinear and non-stationary nature of pollutants poses challenges for achieving high prediction accuracy with these models. As a result, several studies have turned to various deep learning techniques 8 , 33 , 34 , 35 , 36 to enhance the prediction of air pollutant levels.

Despite numerous efforts to forecast concentration of major pollutants, comprehending the complex relationship among diverse influencing factors remains a persistently challenging task. Studies exploring the relevance of these factors in predicting pollutants have been scarce and constrained in scope 37 , 38 . Typically, researchers tend to utilize all accessible features and input them into prediction models. While it holds true that AI&ML models exhibit superior performance in scenarios with abundant data availability, the effectiveness of these models in pollutant prediction hinges on understanding and incorporating the most influential factors. Figure  1 illustrates the comprehensive AI/ML model development workflow for environmental or traffic-related predictions. The process includes data collection across various domains, preprocessing, algorithm selection, model development, training, testing, and validation. The process completes with prediction, incorporating a feedback loop for model refinement if needed, ensuring adaptability and continuous improvement in predictive accuracy.

Charting the sequential steps of AI and ML involvement in predicting air pollution concentrations.

Previous studies have conducted comparative analyses between AI&ML-based methodologies for forecasting concentrations of various pollutants. Initially, Mc Kendry 39 evaluated Artificial Neural Networks (ANN) with MLR for simulating the concentrations of PM 2.5 and PM 10 . Similarly, Dutta and Jinsart 40 compared the performances of decision tree and ANN algorithms in estimating PM 10 concentrations. Other comparisons include Turias et al. 41 pitting back-propagation based ANN against ARIMA for predicting the Sulfur Dioxide (SO 2 ), concentrations of Carbon Monoxide (CO) and Suspended Particulate Matter (SPM), over an industrialized region. Shang and He 42 formulated an innovative prediction method by coupling of ANN and Random forest (RF) to forecast hourly PM 2.5 concentrations. Bozdağ et al. 43 presented a comprehensive analysis for the simulation of PM 10 concentrations by comparing various modelling approaches—ANN, KNN (K-Nearest Neighbour Algorithm), SVM (Support Vector Machine) , LASSO (Least Absolute Shrinkage and Selection Operator), RF, and xGBoost.

This study systematically explores the consequences of severe air pollution in India, focusing on contributors like PM, Organic Aerosols (OAs), BC, Water-Soluble Brown Carbon (WS-BrC), and Volatile Organic Compounds (VOCs). Remediation techniques, including legislation, NAAQS, and an Air Quality Index (AQI), are inspected alongside the evolution of emission load studies and management strategies. Additionally, the study investigates the integration of AI&ML in mitigating and predicting air pollution. It details the application of AI&ML models and underscores the potential of deep learning algorithms, exemplified through a case study predicting PM 2.5 concentrations over India. Identifying challenges like technological barriers, regulatory hurdles, public awareness gaps, agricultural practices, urbanization impacts, cross-border pollution, climate change interlinkages, and socio-economic disparities, the study emphasizes the urgency of comprehensive solutions. Looking forward, the study discusses prospects involving emerging technologies and global collaborations. The study emphasizes the imperative to address air pollution in India holistically, leveraging AI&ML advancements, global cooperation, and technological innovations to formulate effective strategies for combatting the multifaceted challenges posed by air pollution in the region.

Results and discussion

Consequences of air pollution in india.

Air pollution in India specially in metropolitan cities has dire consequences for public health, stemming from increased levels of particulate matter, nitrogen oxides, and various pollutants. This increase pollution level is consistently linked to increased respiratory diseases, particularly asthma, chronic obstructive pulmonary disease (COPD), and bronchitis 7 , 44 . Children, with developing respiratory systems, are particularly vulnerable to irreversible health issues upon prolonged exposure, while the elderly, with compromised immune systems, face pre-eminent risks, including deep lung penetration, inflammation, and enduring damage caused by PM 2.5 . Beyond respiratory implications, air pollution has severe cardiovascular consequences, with nitrogen oxides significantly contributing to an increased risk of heart attacks and strokes, leading to heightened cardiovascular mortality with prolonged exposur 7 . The significant study conducted by the CPCB in Delhi highlighted robust correlations between air quality levels and negative health effects. Comparative analysis against a rural control population in West Bengal indicated a 1.7-fold higher occurrence of respiratory symptoms in Delhi, emphasizing the direct impact of air quality on public health 20 , 45 , 46 , 47 . Odds ratios for upper and lower respiratory symptoms were 1.59 and 1.67, respectively, emphasizing the profound impact of air pollution. The study also highlighted a significantly higher prevalence of current and physician-diagnosed asthma in Delhi, with lung function notably reduced in 40.3% of Delhi's participants compared to 20.1% in the control group 20 .

In addition to respiratory effects, non-respiratory impacts were observed in the cities as compared to rural controls. The prevalence of hypertension was notably higher in cities (36% vs. 9.5% in controls), correlating positively with respirable suspended particulate matter (PM 10 ) levels in ambient air 48 . Chronic headaches, eye irritation, and skin irritation were significantly more pronounced in most of the cities. Community-based studies consistently affirm the association between air pollution and respiratory morbidity. Studies focusing on indoor air pollution reveal similar correlations with respiratory morbidity, extending to conditions such as attention-deficit hyperactivity disorder in children, increased blood levels of lead, and decreased serum concentration of vitamin D metabolites 49 . Beyond health impacts, the environmental consequences of air pollution are profound. Pollutants harm plants and animals, disrupt ecosystems, and lead to biodiversity loss 50 . The issue extends beyond health and the environment, impacting economics and society, straining healthcare, productivity, and social equity, demanding holistic strategies spanning economic, social, and environmental facets making it imperative, in this crisis, to understand the existing and potential remediation techniques 51 .

The economic and social ramifications are substantial, with healthcare costs soaring as the incidence of pollution-related illnesses rises 7 . Treating respiratory and cardiovascular diseases places a significant burden on the healthcare system, affecting both public and private healthcare expenditures 44 . Air pollution in India incurred an estimated economic toll of $95 billion in 2019, amounting to 3% of the country's GDP, attributable to decreased productivity, increased work absences, and premature fatalities 52 . The economic implications of air pollution extend beyond direct healthcare costs, affecting labor markets and overall productivity 53 . Social disparities are accentuated by air pollution, with vulnerable communities facing disproportionate exposure to pollutants. Factors such as socio-economic status, access to healthcare, and geographic location contribute to disparities in exposure and health outcomes 54 . Addressing these social dimensions is crucial for devising equitable solutions that prioritize environmental justice. As India grapples with the immediate consequences of air pollution, emerging challenges require attention. Also, climate change exacerbates existing issues, influencing weather patterns and contributing to the persistence of stagnant air masses that trap pollutants and their transportation mechanism 8 . The increasing frequency of extreme weather events further complicates pollution dynamics 55 . Moreover, the complex interplay of indoor and outdoor air pollution adds another layer of complexity, with indoor air pollution often stemming from household activities such as cooking with solid fuels, compounding the overall burden on public health 49 . However, government policies and initiatives take center stage in this exploration, with regulatory measures, such as emission standards and vehicle restrictions, scrutinized for their effectiveness and implementation challenges 12 . Sustainable urban planning, including the creation of green spaces and transportation planning for pollution reduction, is examined as a proactive approach to mitigate pollution at its source 56 . Technological solutions, ranging from air purifiers to pollution monitoring devices, are also evaluated 57 . The challenges of scalability, accessibility, and integration into existing infrastructure are dissected to discern the practicality and potential impact of these technologies. Emerging technologies and global collaborations are explored as potential catalysts for change 57 , 58 .

Contributors to air pollution in India

Air pollution in India is a complex issue with multiple sources and contributors, as highlighted by various studies conducted by Lalchandani et al. 59 , Tobler et al. 60 , Rai et al. 61 , Talukdar et al. 62 and Wang et al. 63 . The sources and contributors to air pollution can be broadly categorized into particulate matter (PM 2.5 and PM 10 ), organic aerosols (OAs) including black carbon (BC), water-soluble brown carbon (WS-BrC), and volatile organic compounds (VOCs). Each of these components plays a signifsicant role in the overall air quality of the region.

Particulate matter (PM)

Particulate matter is a key component of air pollution, and Lalchandani et al. 59 conducted studies using the Positive Matrix Factorization (PMF) model to identify and apportion different sources of PM. The sources identified included traffic-related emissions, dust transportation, solid-fuel burning emissions, and secondary factors 62 , 64 . Traffic-related emissions in metropolitan cities were found to be the significant contributor to the total concentration of PM, for example, at the IIT Delhi site, emphasizing the impact of vehicular activities on air quality. Additionally, solid fuel burning emissions, often associated with residential cooking and heating, were identified as a major contributor to PM, particularly at night 62 . Rai et al. 61 conducted source apportionment of elements in PM 10 and PM 2.5 , identifying nine source profiles/factors, including dust, non-exhaust sources, solid fuel combustion, and industrial/combustion aerosol plume events. The contribution of anthropogenic sources to elements associated with health risks, such as carcinogenic elements. The geographical origins of these sources were also determined, emphasizing the regional and local influences on element concentrations in atmosphere 65 .

Organic aerosols (OAs)

Organic aerosols are another crucial component of air pollution, and studies by Tobler et al. 60 and Lalchandani et al. 62 revealed three main components of OAs: solid fuel combustion OAs (SFC OAs), hydrocarbon-like OA (HOAs) from vehicular emissions, and oxygenated OAs (OOAs). Lalchandani et al. 65 further categorized these components into sub-factors, providing a detailed understanding of the OA composition. Emissions stemming from traffic emerged as the primary contributor to the overall OA mass, underscoring the profound influence of vehicular pollution 59 .

Black carbon (BC)

BC, a product of incomplete combustion, was studied by Using the Absorption Ångström Exponent (AAE) method, contributions from biomass burning and vehicular emissions were apportioned 66 . Vehicular emissions were found to be a dominant source of BC, contributing around 67.5% 62 , 67 . The distinction between BC and brown carbon (BrC), which absorbs light in the near-UV to visible region, was also discussed, highlighting the need to consider multiple light-absorbing aerosols in air quality assessments.

Water-soluble brown carbon (WS-BrC)

Rastogi et al. 68 performed a PMF analysis of WS-BrC spectra, identifying six factors representing specific sources of BrC. The study revealed diurnal variability in BrC absorption, with factors associated with different emission sources. The presence of secondary BrC was indicated, suggesting the importance of atmospheric processes in the formation of brown carbon. This finding adds another layer of complexity to the sources of light-absorbing aerosols in the atmosphere 69 .

Volatile organic compounds (VOCs)

Wang et al. 63 investigated the characteristics and sources of VOCs, identifying six factors related to traffic, solid fuel combustion, and secondary sources. Traffic-related emissions were found to be the dominant source of VOCs at the urban site, while at the suburban site (MRIIRS), contributions from secondary formation and solid fuel combustion were more significant. The study highlighted the major role of anthropogenic sources in VOC pollution 70 .

Current remediation techniques

India has faced escalating challenges in managing air pollution over the years, necessitating the implementation of diverse remediation techniques. Figure  2 illustrates the legislative evolution of air quality management in India across three eras: Pre-Internet (1905–89), Transition (1990–99), and Internet Era (2000 onwards). This timeline showcases key acts and regulations implemented over time to address air pollution. The bottom timeline highlights the progression of NAAQS in India, from monitoring just 3 pollutants in 1982 to 7 in 1994, and 12 in 2009. The latest phase (2019–24) involves a comprehensive review of air quality standards under the National Clean Air Programme (NCAP) in 2019, demonstrating India's ongoing commitment to improving air quality management.

Legalisation and Evaluation of NAAQS in India 12 .

Legislation and regulatory measures

India's legislative landscape has evolved significantly to address air pollution. The introduction of key acts such as the Air (Prevention and Control of Air Pollution) Act in 1981 and subsequent amendments empowered central and state pollution control boards to handle severe air pollution emergencies 71 . The Environment (Protection) Act of 1986 served as an umbrella act for environmental protection, while the Motor Vehicles Act has been periodically amended to regulate vehicular pollution 72 . Recent developments include the Motor Vehicles (Amendment) Bill of 2019, allowing the government to recall vehicles causing environmental harm 73 . The establishment of institutions like the National Green Tribunal (NGT) and the National Environment Tribunal reflects a commitment to environmental accountability 74 .

National ambient air quality standards (NAAQS) and air quality index (AQI)

The formulation and periodic revision of National Ambient Air Quality Standards (NAAQS) have been pivotal in regulating air quality 18 . Beginning in 1982, the Central Pollution Control Board (CPCB) introduced NAAQS, initially covering SO 2 , NO 2 , and SPM 47 . Subsequent amendments expanded the list to include RSPM, Pb, NH 3 , and CO 75 . The National Air Quality Index (NAQI) was introduced to enhance public awareness, categorizing air quality into six levels from 'Good' to 'Severe' 76 . This index, based on the concentration of eight pollutants, guides interventions for improved air quality.

Air pollution monitoring network

India's air quality monitoring network has witnessed substantial growth. The initiation of the National Ambient Air Quality Monitoring (NAAQM) Network in 1984, expanded to the National Air Quality Monitoring Programme (NAMP), marked a critical step 77 . The network, comprising both manual and Continuous Ambient Air Quality Monitoring System (CAAQMS) stations, now stands at 1082 locations 78 , 79 . Real-time monitoring, as exemplified by CAAQMS, provides valuable data for prompt decision-making. The introduction of the System of Air Quality and Weather Forecasting and Research (SAFAR) further enhances forecasting capabilities 80 .

Evolution of studies on emission load

Emission inventories, critical for formulating air pollution control policies, have evolved over time. Initiatives by CSIR-NEERI and CPCB in the late twentieth century laid the foundation 12 . Emission inventory data, collected through GIS, has become integral in mapping pollution sources and understanding spatial distribution 81 . The Air Pollution Knowledge Assessments (APnA) city program and organizations like TERI contribute to city-specific inventories 82 . The emphasis on utilizing secondary data streamlines the process, enabling the creation of comprehensive databases for national and urban pollution inventories. The secondary data refers to datasets that include emission loads from various sources such as vehicular emissions, industrial outputs, construction activities, residential heating, and biomass burning 83 .

Management strategies and control policies

India's air pollution management strategies encompass a multifaceted approach, with a blend of judicial interventions and executive actions.

Judicial interventions

The judiciary, particularly through petitions filed by M.C. Mehta, has been instrumental in setting guidelines and policies 84 . For instance, interventions in the Taj Trapezium Zone and the oversight of air quality management plans for non-attainment cities by the National Green Tribunal (NGT) are notable 74 . The judiciary has played a significant role in shaping policies for better governance and legislation.

Executive actions

Several executive measures contribute to air pollution control. The Auto Fuel Policy, initiated in 2003 and updated in 2014, addresses vehicular emissions 85 . Emphasis on alternative fuels, as seen in the National Auto Fuel Policy and the Pradhan Mantri Ujwala Yojana (PMUY) for subsidized LPG connections, aligns with cleaner fuel initiatives 86 . Stricter emission standards for thermal power plants and the push for Hybrid and Electric Vehicles (EVs) under schemes like Faster Adoption and Manufacturing of Hybrid & Electric Vehicles (FAMHE) contribute to pollution reductions 87 .

AI&ML Techniques for addressing and forecasting air pollution

Overview of ai&ml models.

Various AI&ML techniques, such as ANN, Fuzzy logic (FL), Support Vector Machine (SVM), Convolutional Neural Network (CNN), Recurrence Neural Network (RNN), Long Short-Term Memory (LSTM), Convolutional Autoencoder (CA) etc., are commonly used in previous studies to predict and forecast earth and atmospheric variables 8 , 25 , 88 , 89 , 90 , 91 (Table 1 ). AI&ML models have become pivotal in processing and simulating non-linear information, with a notable focus on ANNs 92 . ANNs emulate the human nervous system, comprising interconnected neurons that collectively address a spectrum of challenges, from function approximation to clustering and optimization 93 . The three-stage process involved in ANN modelling, encompassing design, training, and validation, underscores its versatility 92 . During the design phase, crucial parameters such as architecture, layers, neurons, and learning algorithms are thoroughly chosen 94 . Training involves iterative adjustments of synaptic weights to minimize errors, while validation gauges the network's generalization performance for unknown data.

Multilayer Perceptron (MLPs), a prominent type of ANN, have proven effective in predicting atmospheric pollution events. Typically featuring input, hidden, and output layers, MLPs can adapt to complex patterns by incorporating multiple hidden layers 92 . Configuring neurons in the hidden layers is of utmost importance, as an incorrect count can lead to over-fitting or under-fitting. Techniques like thumb rule and trial and error, network reduction offer solutions to optimize neuron numbers. FL, another AI technique, operates on a different paradigm by assigning truth values in a range. Developed from fuzzy set theory, it accommodates linguistic variables, making it adept at handling uncertainty in natural language statements. Fuzzy logic's three main phases—fuzzification, inference, and defuzzification—form a robust modelling system capable of addressing nuanced problems. SVM are popular for supervised learning, excelling in classification, prediction, density estimation, and pattern recognition. SVM seeks an optimal hyperplane to segregate data into predefined classes, with kernel functions playing a pivotal role in introducing non-linearity.

Deep Neural Networks (DNNs) represent an advanced version of ANNs, characterized by structural depth and scalability 8 . DNNs, with more than three layers, can automatically extract features from raw inputs, known as feature learning. Notable architectures within DNNs, such as CA, LSTM, CNNs and RNNs have demonstrated superior performance, especially in air pollution forecasting. The training of DNNs demands significant computational power, leading to advancements in processing capabilities and the development of sophisticated algorithms. Overcoming challenges like vanishing gradient and overfitting has prompted the application of advanced algorithms like SVM, RF, Greedy layer-wise, and Dropout. The application of these models extends across various domains due to their versatility and robust performance. The modelling of complex atmospheric variables such as air pollution forecasting, LSTM, CA, and CNNs emerge as particularly effective and popular architectures.

Application of AI&ML in addressing and forecasting air pollution

The application of AI&ML models, particularly ANNs, FL, SVM and DL models, have emerged as a crucial tool in addressing and forecasting air pollution. ANNs have helped in a transformative era in air pollution forecasting, with a diverse range of applications capturing the attention of researchers. Numerous studies attest to the success of ANNs in predicting both particulate and gaseous pollutants with desired accuracy over various spatio-temporal resolution. The early forays into air pollution forecasting by Mlakar et al. 95 marked a significant milestone, employing a trained nonlinear three-layered back propagation feed forward network. This model successfully predicted the concentration of SO 2 over a thermal power plant, showcasing the potential of ANNs. Subsequent research expanded the scope and sophistication of ANN applications. Similarly, Arena et al. 96 demonstrated the efficacy of multi-layer perceptron in predicting concentration of SO 2 over an industrial area, emphasizing the model's accuracy across diverse weather conditions. Sohn et al. 97 extended the ANN approach to model multiple pollutants, including NO, SO 2 , NO 2 , CO, O 3 , CH 4 and total hydrocarbons. The results indicated reasonable accuracy within a limited prediction range, highlighting the need for further optimization by incorporating additional weather-related input parameters. The application of ANNs in gaseous pollutants forecasting continued with studies by Slini et al. 98 and Kandya 99 both emphasizing the importance of optimizing input parameters for improved accuracy. Comparative assessments with other forecasting techniques consistently positioned ANNs as superior for gaseous pollutants. Chaloulakou et al. 100 found that ANN outperformed Multiple Linear Regression (MLR) in predicting ozone concentrations, showcasing the model's superior accuracy. Similar findings were reported by Mishra and Goyal 101 , compared Principal Component Analysis (PCA)-based ANN model with MLR for estimating the concentrations of NO 2 . In the realm of particulate matter forecasting, ANNs have proven equally effective. Fernando et al. 102 successfully used multi-layered MLP to predict PM 10 concentrations, considering parameters such as hourly meteorological data, particulate, matter with statistical indicators. Grivas and Chaloulakou 103 employed an ANN model for hourly PM 10 predictions, showcasing consistent accuracy even in the presence of noisy datasets. The versatility of ANNs extends to predicting roadside contributions to PM 10 concentrations, as demonstrated by Suleiman et al. 104 . Comparative studies with other models have affirmed the efficacy of ANNs in particulate matter forecasting. Zhang et al. 105 utilized BPANN to forecast the concentrations of PM 10 and found BPANN outperforming other models in predictive accuracy. Paschalidou et al. 106 evaluated the multi-layer perceptron-based ANN those models provided superior results compared to Radial Basis Function models, establishing the former's dominance in terms of forecasting capability. Contrasting trends were observed in certain studies, such as those by Mishra et al. 107 and Moisan et al. 108 , where alternative models outperformed ANN during extreme events. This highlights the nuanced nature of model performance, with specific conditions favouring different approaches. However, recent progress has witnessed researchers utilizing ensemble methods to improve both the stability and accuracy of ANN models. Liu et al. 109 combined Wavelet Packet Decomposition (WPD), Particle Swarm Optimization (PSO), and BPNN to create an ensemble model for PM 2.5 forecasting, demonstrating superior precision compared to individual models.

FL, renowned for its capacity to manage uncertainty, enhanced fault tolerance, and adeptness in handling highly complex nonlinear functions, has garnered extensive adoption in the realm of air pollution prediction. The advantages of FL are exemplified in various studies. For example, Chen et al. 110 innovatively introduced a novel fuzzy time series model specifically for O 3 prediction, showcasing its superior performance when compared to traditional fuzzy time series models. Jain and Khare 111 applied a neuro-fuzzy model to predicts the concentration of CO in Delhi, achieving accurate estimates at complex urban levels. Carbajal-Hernández et al. 112 predicts air quality in Mexico City by utilising FL model alongside autoregression model and signal processing. The introduction of a novel algorithm, the "Sigma operator," allowed for precise evaluation of air quality variables, showcasing the effectiveness of fuzzy-based models. Moreover, Al-Shammari et al. 113 , evaluates stochastic and FL-driven models to estimate the daily maximum concentrations of O 3 . The findings indicated that the FL-based model exhibited a marginal superiority over the statistical model particularly in instances of severe pollution events. Innovative approaches like the Fuzzy Inference Ensemble (FIE), as proposed by Bougoudis et al. 114 , demonstrated high accuracy in air pollution forecasting for Athens. Another significant application was presented by Song et al. 115 , where different probability density functions were employed to enhance particulate matter (PM) forecasting. They developed an adaptive neuro-fuzzy model, emphasizing the importance of density functions in addressing uncertainty associated with future PM trends. Furthermore, Wang et al. 116 presented a hybrid model for forecasting air pollution. This model merges uncertainty analysis with fuzzy time series, demonstrating precision in predicting PM and NO 2 concentrations. Behal and Singh 117 leveraged FL within an intelligent IoT sensor framework to monitor and simulate benzene, demonstrating satisfactory statistical efficacy in recent advancements. The versatility of fuzzy logic extends to unconventional pollutants as demonstrated by Arbabsiar et al. 118 , who modelled the leakage of CH 4 and H 2 S using a fuzzy inference technique. The suggested model demonstrated satisfactory performance when evaluating these contaminants.

Support Vector Machines (SVM), when combined with other machine learning algorithms, have been helpful in forecasting diverse types of pollutants. Feng et al. 119 compared SVM with other models for forecasting daily maximum concentrations of O 3 in Beijing, highlighting its stable and accurate performance. Yeganeh et al. 120 assessed the efficacy of a forecasting model utilizing SVM integrated with Partial Least Squares (PLS) for the prediction of CO concentrations, demonstrating positive outcomes. García Nieto et al. 121 conducted a comparative analysis of various prediction models for PM 10 concentrations, determining that the SVM method exhibited superior accuracy and robustness. Luna et al. 122 utilized Principal PCA in combination with SVM and ANN for the prediction of O 3 levels in Rio de Janeiro. Their study specifically investigated the influence of meteorological parameters on the concentrations of O 3 . Wang et al. 123 proposed hybrid adaptive forecasting models combining SVM and ANN for predicting PM 10 and SO 2 , demonstrating superior performance compared to individual models. FL and SVM in the forecasting air pollution levels have proven to be highly effective in addressing the complexities and uncertainties associated with predicting pollutant concentrations.

While still in its early stages, the potential of DNNs in this domain is evident from a review of various applications such as forecasting of variables in earth and atmospheric sciences. Early on, Freeman et al. (2018) employed a combination of Long Short-Term Memory (LSTM) and Recurrent Neural Networks (RNN) to predict ozone concentrations in an urban area. While showing strong predictability in 8 h average ozone concentrations, various model runs revealed overfitting concerns, underscoring the necessity for further refinement. Wang and Song 125 introduced an ensemble method using a deep LSTM network with fuzzy c-means clustering for air quality forecasting. This ensemble approach outperformed individual models, showcasing its efficacy in both short-term and long-term predictions. Zhou et al. 126 explored the application of LSTM and deep learning algorithms for multi-step ahead forecasting of PM 2.5 , PM 10 , and NO x . Their deep learning architecture, integrating dropout neurons and L 2 regularization, demonstrated exceptional capabilities in capturing variations in the processes of air pollutant generation. Recent research highlights the growing preference for employing deep neural networks to capture dynamic spatiotemporal features from historical air quality and climatological datasets. Fan et al. 91 introduced stacked LSTM (LSTME), spatiotemporal deep learning (STDL), time delay neural network (TDNN), autoregressive moving average (ARMA), and support vector regression (SVR) for modelling of air pollutants over different spatiotemporal resolutions. The inclusion of auxiliary inputs resulted in a model with exceptional performance, outshining other machine learning techniques. Soh et al. 127 proposed a STDL integrating ANN, CNN, and LSTM for PM 2.5 prediction. The model exhibited stability over extended time periods, with noise reduction achieved through Airbox sensor source models, further enhancing prediction accuracy. Qi et al. 128 presented a novel forecasting approach employing a fusion of Graph Convolutional and LSTM (GC-LSTM) neural networks, aiming to investigate spatial interdependence within air quality data. The spatial correlation modelling highlighted the consistency of the GC-LSTM model for short-term forecasting, suggesting potential improvements for long-term predictions with enhanced spatiotemporal considerations. Fan et al. 91 developed a LSTM-based deep–RNN for predicting PM 2.5 for different spatiotemporal frames showcasing superior specificity measures compared to baseline models. In a novel approach, Li et al. 129 and Zhang et al. 130 incorporated large-scale datasets of graphical images for air pollution estimation, utilizing CNN. The models, trained on images capturing various atmospheric conditions, demonstrated improved prediction accuracy, emphasizing the adaptability of deep learning to diverse data types. These models offer robust solutions, demonstrating superior performance in various studies and showcasing their potential to contribute significantly to the field of environmental monitoring and public health.

Performance analysis

The evaluation is based on the comparison of their performances using statistical measures such as RMSE and R 2 , widely accepted metrics in air pollution forecasting studies. Previous research, utilizing a range of datasets, has yielded disparate results 134 . While certain studies advocate for ensemble methods, others find negligible disparities in the overall accuracy of the outcomes. The efficacy of AI and ML-driven methodologies relies heavily on the precise curation of influential parameters, especially when addressing various pollutants such as PM, O 3 , NO 2 , SO 2 , and CO 29 . For example, for PM forecasting, critical elements such as precipitation, pressure, humidity, land utilization, wind speed and direction, traffic flow on roads, and population density exert significant influence. Similarly, different influential parameters are identified for SO 2 , NO 2 , O 3 , and CO, emphasizing the importance of tailoring models to specific pollutants. The precision of the methods is notably impacted by the direct correlation between these factors and forecasted levels of pollutants. Additionally, the efficacy of AI&ML models hinges upon variables including network structure, intricacy, learning algorithms, correspondence between input and output information, and the presence of data interference. A comprehensive analysis shows the varying performances of DNN, SVM, ANN, and Fuzzy techniques across different pollutants. DNNs emerge as particularly effective in forecasting PM concentrations, outperforming other techniques with R 2 and mean RMSE values of 0.96 and 7.27 μg/m 3 , respectively 91 , 126 , 133 . In O 3 prediction, SVM, FL and DNN exhibit superior accuracy, with DNNs once again leading with R 2 and mean RMSE values of 0.92 and 3.51 μg/m 3 , respectively 119 , 120 . SVM excels in forecasting NO 2 concentrations, although Fuzzy and DNN techniques also demonstrate reasonable accuracy 116 , 118 , 131 . Notably, the DNN approach consistently stands out, showcasing the best statistical performance for O 3 and CO categories. For CO, DNN achieves an exceptional RMSE of 0.69 × 10 –5  ppm and an R 2 of 0.95 119 , 120 , 124 , 125 . The overall analysis represents the superiority of DNN across all pollutants, with the lowest overall RMSE score of 5.68. However, despite DNN's dominance, it is crucial to note the underdeveloped application of ensemble methodologies based on DL models for the forecasting of air pollution 131 , 135 , 136 . These approaches, involving multiscale spatiotemporal predictions, have untapped potential to further advance the field, incorporating more explanatory variables to represent air pollution episodes with robust dynamical forcing. The DNN emerges as the leading AI&ML system for the forecasting and prediction of air pollution based on statistical evidence, the exploration of ensemble approaches presents an avenue for future developments in enhancing predictive accuracy.

Prediction of PM 2.5 concentrations

The study used a convolutional autoencoder (CA) for analysing PM 2.5 concentrations. The dataset was divided into training (70%), testing (20%), and validation (10%) sets, trained over 30 epochs (Fig.  3 ). This PM 2.5 -focused CA processes sequences of ten consecutive images, using acquired features to reconstruct subsequent images. The visual representation of the model's capabilities includes sequences of 10 input images, their corresponding 11 th ground truth, and the model's predictions (Fig.  4 ). The model demonstrates promising performance in predicting PM 2.5 concentration patterns across India. Comparing the actual 11 th image with the predicted one reveals that the model successfully captures the broad spatial distribution of PM 2.5 concentrations. Key findings show that the model accurately predicts high concentration areas in the northern regions, particularly in the IGP (Fig.  4 ). It also effectively represents lower concentrations in southern and eastern coastal areas. The model captures the general gradient from northwest to southeast quite effectively. The prediction tends to slightly overestimate PM 2.5 levels in the northwestern region. Additionally, some localized high-concentration areas in central India are not fully captured in the prediction. Furthermore, the model's prediction shows a smoother distribution compared to the more granular actual data. (Fig.  4 ). Performance evaluation employed established image quality metrics: Structural Similarity Index (SSIM), Peak Signal-to-Noise Ratio (PSNR) and Mean Squared Error (MSE) (Fig.  5 ). SSIM, which assesses image similarity, predominantly ranged from 0.50 to 0.70 during training, slightly lowering to 0.45 to 0.55 during testing, and stabilizing at 0.50 to 0.60 in validation. PSNR peaked at 25 to 30 dB during training, followed by 24 to 28 dB in testing, and 28 to 30 dB in validation. Lower MSE values (10 to 15 µg/m 3 in training, 10 to 20 µg/m 3 in testing, and 8 to 11 µg/m 3 in validation) signify improved accuracy at the pixel level.

RMSE loss during the training, testing and validation phase.

Example set for predicting the 11th image of PM 2.5 by providing a batch of 10 images of concentration and comparing with the 11th actual image. The maps were generated using Python in a Jupyter Notebook with Matplotlib (v3.3.4) and Basemap (v1.2.2) libraries ( https://matplotlib.org/ and https://matplotlib.org/basemap/ ).

Model evaluation parameters used for prediction the PM 2.5 concentrations.

These metrics offer insights into image quality, indicating some variation between training, testing, and validation, yet within acceptable ranges. Consistently higher SSIM and PSNR values and lower MSE values highlight the model's exceptional precision compared to benchmarks. The model's excellence traces back to its ability to capture complex spatio-temporal features through Autoencoder-based models and strategic integration of Conv2d, Batch Normalization, and Upsampling layers. The model outperforms prior methodologies in predicting PM 2.5 concentrations, achieving precise and high-quality predictions across phases. Attempting to forecast PM 2.5 levels for the next 4 days led to efficiency parameter decreases (SSIM, PSNR, MSE) with increased time frames, suggesting the need for more parameters for model efficiency improvement (Fig.  6 ). Predicting PM 2.5 concentrations remains challenging due to intricate spatiotemporal features, where DL models offer promise. Leveraging deep learning architectures and transfer learning, this study fine-tuned models, achieving promising PM 2.5 prediction results. Despite ongoing challenges in precise location predictions due to PM 2.5 's dynamic nature, the model demonstrated spatial distribution prediction abilities, evident in visual comparisons between predicted and actual PM 2.5 concentration maps.

Example set of predictions of PM 2.5 for next 4 days compared with their actual images. The maps were generated using Python in a Jupyter Notebook with Matplotlib (v3.3.4) and Basemap (v1.2.2) libraries ( https://matplotlib.org/ and https://matplotlib.org/basemap/ ).

Challenges and limitations

Technological barriers.

One of the primary challenges lies in overcoming technological barriers. While advanced pollution control technologies exist, their widespread adoption is hindered by factors such as high costs and limited access to cutting-edge solutions. Many regions, particularly in rural areas, lack the infrastructure necessary to deploy and maintain sophisticated air quality monitoring and purification systems. Bridging this technological divide is essential for comprehensive pollution control.

Regulatory and enforcement challenges

India grapples with the challenge of implementing and enforcing air quality regulations consistently. While the country has established regulatory frameworks to curb emissions from industries, vehicles, and other pollution sources, enforcement remains uneven. This inconsistency is often compounded by resource constraints, bureaucratic hurdles, and the need for stronger mechanisms to penalize non-compliance. Strengthening regulatory frameworks and enhancing enforcement mechanisms are critical steps in addressing this challenge.

Public awareness and participation

Creating widespread awareness and fostering public participation are essential components of any successful pollution control strategy. However, there is a considerable gap in public awareness regarding the causes and consequences of air pollution. Engaging citizens in proactive measures, such as adopting sustainable practices and reducing individual carbon footprints, requires comprehensive educational campaigns and community involvement. Overcoming societal inertia and instigating behavioral change are significant challenges in this regard.

Agricultural practices and crop burning

Agricultural practices, particularly the prevalent practice of crop burning, contribute significantly to air pollution. The burning of crop residues releases substantial amounts of particulate matter and pollutants into the air. Farmers resort to this practice due to a lack of viable alternatives and time constraints between harvest seasons. Developing and promoting sustainable agricultural practices, coupled with providing farmers with effective alternatives to crop burning, is a complex challenge that requires a holistic approach.

Urbanization and infrastructure development

Rapid urbanization and infrastructure development, while essential for economic growth, often contribute to increased pollution levels. The construction industry, in particular, releases pollutants into the air. Balancing the need for development with sustainable and environmentally conscious practices poses a significant challenge. Implementing green building technologies, stringent emission norms for construction activities, and incorporating urban planning strategies that prioritize air quality are vital steps in addressing this challenge.

Cross-border pollution

Air pollution knows no boundaries, and India contends with the impact of cross-border pollution. Transboundary movement of pollutants, especially during crop burning seasons, contributes to elevated pollution levels in various regions. Collaborative efforts with neighbouring countries are necessary to address this challenge effectively. Developing joint strategies, sharing data, and fostering regional cooperation are imperative for tackling the transboundary dimension of air pollution.

Climate change interlinkages

The interlinkages between air pollution and climate change present a complex challenge. Mitigating air pollution often aligns with climate action goals, but there are trade-offs and synergies that need careful consideration. Striking a balance between addressing immediate air quality concerns and contributing to long-term climate resilience requires integrated policies and strategic planning.

Socio-economic disparities

Air pollution disproportionately affects vulnerable communities, exacerbating existing socio-economic disparities. The challenge lies in designing interventions that address environmental concerns and promote social equity. Ensuring that pollution control measures do not inadvertently burden marginalized communities and providing equitable access to clean technologies are critical to overcoming this challenge.

Future prospects

India stands at the cusp of a pivotal moment in its battle against air pollution, with promising avenues emerging on both technological and collaborative fronts.

Emerging technoloagies

The integration of cutting-edge technologies offers hope for India's future in pollution control. Advancements in AI&ML, when coupled with sophisticated numerical weather prediction models, present a potent toolset for predicting and managing air pollution. These technologies can enhance real-time monitoring, improve predictive capabilities, and facilitate data-driven decision-making, allowing for more precise and targeted interventions. Additionally, the fusion of AI&ML with numerical weather prediction (NWP) models can refine pollution control strategies by providing a deeper understanding of atmospheric dynamics and pollutant dispersion patterns. Furthermore, exploring potential breakthroughs in sustainable energy sources offers a transformative pathway. Shifting from traditional, pollutant-intensive energy sources to sustainable alternatives is crucial for reducing the overall carbon footprint. Investments in research and development, coupled with policy incentives, can accelerate the adoption of clean and renewable energy solutions, fostering a paradigm shift in India's energy landscape.

Global collaborations

Recognizing that air pollution transcends national boundaries, India looks toward global collaborations as a key driver for progress. International efforts in combating air pollution gain significance as countries join forces to address shared challenges. Collaborative platforms provide opportunities for knowledge sharing, exchange of best practices, and collective research initiatives. India's participation in these global endeavours not only enriches its own understanding of air pollution dynamics but also contributes to the global pool of knowledge. By fostering partnerships with other nations, India can access expertise, technologies, and resources that augment its capacity to implement effective pollution control measures. Knowledge sharing and collaborative research initiatives form the cornerstone of global efforts. Platforms that facilitate the exchange of data, research findings, and innovative solutions enable nations to collectively tackle the intricate and interconnected challenges of air pollution. As India engages in these collaborative endeavours, it not only benefits from the collective wisdom of the global community but also contributes its unique insights and experiences, enriching the collective understanding of air pollution dynamics.

India's strategic focus on emerging technologies and global collaborations holds immense promise in navigating the future. By harnessing the power of advanced technologies and participating in international initiatives, India can chart a course toward a cleaner, more sustainable future where the skies are clear, and the air is a testament to the collective commitment to environmental well-being.

Materials and methods

Maintaining fresh air quality is a complex undertaking influenced by various factors over time. These elements encompass air pollutant emissions, deposition, weather patterns, traffic dynamics, and human activities, among others 8 , 64 . The complexity of these interrelated factors makes it challenging for traditional shallow models to offer precise portrayals of air quality attributes. Based on the above review, deep learning algorithms were found most suitable for predicting air quality variables without needing prior knowledge. This capability enhances the potential for more accurate predictions regarding air quality, signifying a valuable contribution to addressing the intricacies associated with sustaining optimal air quality levels.

The case study utilized MERRA-2 reanalysis data from the NASA GESDISC DATA ARCHIVE application 137 , 138 . This dataset, spanning from January 1, 2015, to December 31, 2022, features a spatial resolution of 0.5° × 0.625° and a temporal resolution of 1 h (Fig.  7 ). It includes five key variables: black carbon surface mass concentration (BCSMASS), dust surface mass concentration—PM 2.5 (DUSMASS25), organic carbon surface mass concentration (OCSMASS), sea salt surface mass concentration—PM 2.5 (SSSMASS25), and SO 4 surface mass concentration (SO 4 SMASS). These variables are analysed across three dimensions: latitude, longitude, and time. The concentration of the PM 2.5 (µg/m 3 ) for each grid cell was computed as 139 , 140 :

Surface PM 2.5 concentration over India during ( a ) Winters and ( b ) Summers; Maps were generated using R Studio (v4.3.3, https://www.rstudio.com/ ).

Convolutional Autoencoder model

Air quality monitoring and predicting PM 2.5 concentrations accurately stands crucial for public health and environmental management 8 . The case study explores an innovative approach employing an Autoencoder-based DL model for forecasting PM 2.5 concentrations from spatiotemporal data over India. The study begins by complexly handling the datasets, leveraging PyTorch's Dataset and data loader classes. The ATMriver Dataset class is crafted to capture the dataset, enabling sequential data handling 9 . The data, formatted into tensors and split into training, testing, and validation subsets in a ratio of 70, 20 and 10, respectively, undergoes a custom transformation via the tensor class, ensuring compatibility with the neural network model 8 , 141 . The core of this methodology lies in the architecture of the Autoencoder, a neural network comprising convolutional and transposed convolutional layers. Specifically, the model comprises convolutional layers (conv1, conv2, conv3) responsible for feature extraction and transposed convolutional layers (conv1_d, conv2_d, conv3_d) for data reconstruction (Fig.  8 ). Each convolutional layer is paired with batch normalization and dropout (set at 25%) to regularize the network and prevent overfitting. The use of five layers in this Autoencoder architecture allows for hierarchical feature extraction and reconstruction, enhancing the model's ability to learn complex representations. The learning rate, a critical hyperparameter governing the magnitude of parameter updates during optimization, is set to 0.0025 for the Adam optimizer. This value influences the convergence speed and stability of the training process. A higher learning rate might lead to faster convergence but risks overshooting the optimal parameters, while a lower rate might result in slower convergence. The chosen learning rate balances the trade-off between convergence speed and stability, aiming to facilitate efficient model training while preventing divergence or oscillation in the optimization process.

Convolution autoencoder architecture for PM 2.5 data processing with model features an encoding phase with three autoencoder stages, followed by a decoding phase with two transpose convolution stages; structure enables dimensionality reduction and subsequent reconstruction of PM 2.5 concentration maps.

To train the Autoencoder, a custom root mean squared error (RMSE) loss function is defined. This loss function quantifies the disparity between predicted and actual PM 2.5 concentrations, guiding the model toward more accurate predictions. The training process iterates through the dataset multiple times (epochs), optimizing the model parameters using the Adam optimizer. The evaluation phase of the model involves assessing its predictive capabilities on separate testing and validation sets. The model's outputs are compared against the original images PM 2.5 concentrations, and the RMSE loss is computed. The best-performing model, based on its performance on the testing set, is identified and saved for the prediction. Further the records and reports the losses incurred during training, testing, and validation across epochs, providing insights into the model's loss curve and performance stability. Additionally, the best model's loss metric is highlighted, signifying its capability to accurately predict PM 2.5 concentrations. The evaluation of the trained model's predictive capability in this study primarily relied on two widely accepted image quality metrics: Structural Similarity Index (SSIM) and Peak Signal-to-Noise Ratio (PSNR). The Structural Similarity Index (SSIM) serves as a measure to assess the similarity between the predicted and actual images 142 . SSIM evaluates the perceived change in structural information, including luminance, contrast, and structure, between the predicted and actual images. A higher SSIM score, closer to 1, indicates a greater similarity between the two images, implying better predictive performance of the model. Peak Signal-to-Noise Ratio (PSNR) is another commonly used metric for quantifying the quality of reconstructed or predicted images. PSNR measures the ratio between the maximum possible power of a signal and the power of corrupting noise that affects the fidelity of its representation. Higher PSNR values signify lower image distortion or higher image fidelity, implying better prediction accuracy in capturing the details of the actual images.

Addressing the complex challenges of air pollution in India necessitates a multifaceted and technically informed approach. The existing impediments, including technological barriers and limited access to advanced pollution control technologies, underline the urgency of bridging the technological divide, particularly in rural areas. While regulatory frameworks are in place, inconsistent enforcement due to resource constraints and bureaucratic hurdles requires strategic strengthening. Public awareness and participation, integral components of effective pollution control, demand targeted educational campaigns to instigate behavioural change. Agricultural practices, notably crop burning, pose a significant challenge, and resolving this requires not only viable alternatives but a holistic approach that integrates sustainable agricultural practices. Rapid urbanization and infrastructure development, while essential for economic growth, necessitate the incorporation of green building technologies, stringent emission norms, and urban planning strategies prioritizing air quality. Cross-border pollution adds a transboundary dimension, demanding collaborative efforts with neighbouring countries. The intricate interlinkages between air pollution and climate change underscore the need for carefully balanced policies that address immediate air quality concerns while contributing to long-term climate resilience. Moreover, the disproportionate impact of air pollution on vulnerable communities emphasizes the importance of interventions that promote social equity alongside environmental considerations. Looking towards the future, the convergence of emerging technologies offers a beacon of hope. The integration of AI&ML with numerical weather prediction models presents a potent toolset for real-time monitoring, precise predictive capabilities, and data-driven decision-making. This amalgamation not only enhances our understanding of atmospheric dynamics and pollutant dispersion patterns but also refines pollution control strategies. Exploring breakthroughs in sustainable energy sources becomes imperative for reducing the overall carbon footprint. Shifting from traditional, pollutant-intensive energy sources to clean and renewable alternatives require concerted efforts through research, development, and policy incentives.

Furthermore, global collaborations stand out as a key driver for progress, given the transboundary nature of air pollution. Participating in international efforts fosters knowledge sharing, exchange of best practices, and collective research initiatives. By engaging in these collaborative activities, India not only enriches its understanding of air pollution dynamics but contributes to the global pool of knowledge. Platforms facilitating data exchange, research findings, and innovative solutions enable nations to collectively tackle the complex challenges of air pollution. In navigating the future, India's strategic focus on emerging technologies and global collaborations holds immense promise. The careful harnessing of advanced technologies and participation in international initiatives can chart a course toward a cleaner, more sustainable future. The fusion of AI&ML with numerical weather prediction (NWP) models positions India to proactively manage air quality, with the skies serving as a testament to the collective commitment to environmental well-being. As India progresses, the synergy of technological advancements and global cooperation emerges as the cornerstone for effective, informed, and sustainable solutions to combat air pollution.

Data availability

Data will be made online on a reasonable request to the corresponding author.

Masood, A. & Ahmad, K. A review on emerging artificial intelligence (AI) techniques for air pollution forecasting: Fundamentals, application and performance. J. Clean. Prod. 322 , 129072 (2021).

Article   CAS   Google Scholar  

WHO. Atlas on Children’s Health and the Environment . (2017).

Jiang, X. Q., Mei, X. D. & Feng, D. Air pollution and chronic airway diseases: what should people know and do?. J. Thorac. Dis. 8 , E31–E40 (2016).

PubMed   PubMed Central   Google Scholar  

Hayes, R. B. et al. PM25 air pollution and cause-specific cardiovascular disease mortality. Int. J. Epidemiol. 49 , 25–35 (2020).

Article   PubMed   Google Scholar  

Shakya, D., Deshpande, V., Goyal, M. K. & Agarwal, M. PM2.5 air pollution prediction through deep learning using meteorological, vehicular, and emission data: A case study of New Delhi. India. J. Clean. Prod. 427 , 139278 (2023).

Brown, P. E. et al. Mortality Associated with Ambient PM2.5 Exposure in India: Results from the Million Death Study. Environ. Health Perspect. 130 , 097004 (2022).

Article   PubMed   PubMed Central   Google Scholar  

Pandey, A. et al. Health and economic impact of air pollution in the states of India: the Global Burden of Disease Study 2019. Lancet Planet. Heal. 5 , e25–e38 (2021).

Article   Google Scholar  

Rautela, K. S., Singh, S. & Goyal, M. K. Characterizing the spatio-temporal distribution, detection, and prediction of aerosol atmospheric rivers on a global scale. J. Environ. Manage. 351 , 119675 (2024).

Article   CAS   PubMed   Google Scholar  

Rautela, K. S., Singh, S. & Goyal, M. K. Resilience to Air Pollution: A Novel Approach for Detecting and Predicting Aerosol Atmospheric Rivers within Earth System Boundaries. Earth Syst. Environ. https://doi.org/10.1007/s41748-024-00421-0 (2024).

Chakraborty, S. et al. Extending the Atmospheric River Concept to Aerosols: Climate and Air Quality Impacts. Geophys. Res. Lett. 48 (9), e2020GL091827 (2021).

Kapoor, M. Managing Ambient Air Quality Using Ornamental Plants-An Alternative Approach. Univers. J. Plant Sci. 5 , 1–9 (2017).

Article   MathSciNet   Google Scholar  

Gulia, S. et al. Evolution of air pollution management policies and related research in India. Environ. Challenges 6 , 100431 (2022).

UN. Two ‘Population Billionaires’, China and India, Face Divergent Demographic Futures. Dep. Econ. Soc. Affiars 1–10 (2023).

IQAir. (Report) World Air Quality Report. 2020 World Air Qual. Rep. 1–35 (2020).

WHO. World Health Statistics . (2014).

Kumari, S., Verma, N., Lakhani, A. & Kumari, K. M. Severe haze events in the Indo-Gangetic Plain during post-monsoon: Synergetic effect of synoptic meteorology and crop residue burning emission. Sci. Total Environ. 768 , 145479 (2021).

Delhi Air Pollution: Real-time Air Quality Index. https://aqicn.org/city/delhi .

CPCB. National Ambient Air Quality Status & Trends 2019. Cent. Pollut. Control Board 53 , 1689–1699 (2020).

Google Scholar  

IITK. Comprehensive study on air pollution and green house Google Scholar. A Rep. Submitt. to Gov. NCT Delhi DPCC Delhi 1–334 (2016).

Rizwan, S., Nongkynrih, B. & Gupta, S. K. Air pollution in Delhi: Its Magnitude and Effects on Health. Indian J. Community Med. 38 , 4 (2013).

Bai, L., Wang, J., Ma, X. & Lu, H. Air Pollution Forecasts: An Overview. Int. J. Environ. Res. Public Health 15 , 780 (2018).

Masood, A. & Ahmad, K. A model for particulate matter (PM2.5) prediction for Delhi based on machine learning approaches. Procedia Comput. Sci. 167 , 2101–2110 (2020).

Mo, X., Zhang, L., Li, H. & Qu, Z. A Novel Air Quality Early-Warning System Based on Artificial Intelligence. Int. J. Environ. Res. Public Health 16 , 3505 (2019).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Krishan, M. et al. Air quality modelling using long short-term memory (LSTM) over NCT-Delhi. India. Air Qual. Atmos. Heal. 12 , 899–908 (2019).

Singh, S. & Goyal, M. K. An innovative approach to predict atmospheric rivers: Exploring convolutional autoencoder. Atmos. Res. 289 , 106754 (2023).

Singh, S. & Goyal, M. K. Enhancing climate resilience in businesses: The role of artificial intelligence. J. Clean. Prod. 418 , 138228 (2023).

Li, C., Hsu, N. C. & Tsay, S.-C. A study on the potential applications of satellite data in air quality monitoring and forecasting. Atmos. Environ. 45 , 3663–3675 (2011).

Article   ADS   CAS   Google Scholar  

Box, G. E. P., Jenkins, G. M., Reinsel, G. C. & Ljung, G. M. Journal of time series analysis book review time series analysis: forecasting and control, 5th edition, by. J. Time. Ser. Anal 37 , 709–711 (2016).

Siwek, K. & Osowski, S. Data mining methods for prediction of air pollution. Int. J. Appl. Math. Comput. Sci. 26 , 467–478 (2016).

Fu, M., Wang, W., Le, Z. & Khorram, M. S. Prediction of particular matter concentrations by developed feed-forward neural network with rolling mechanism and gray model. Neural Comput. Appl. 26 , 1789–1797 (2015).

Sekar, C., Gurjar, B. R., Ojha, C. S. P. & Goyal, M. K. Potential Assessment of Neural Network and Decision Tree Algorithms for Forecasting Ambient PM2.5 and CO Concentrations: Case Study. J. Hazardous, Toxic, Radioact. Waste 20 , (2016).

Akhtar, A., Masood, S., Gupta, C. & Masood, A. Prediction and Analysis of Pollution Levels in Delhi Using Multilayer Perceptron (Springer, 2018).

Book   Google Scholar  

Li, X., Peng, L., Hu, Y., Shao, J. & Chi, T. Deep learning architecture for air quality predictions. Environ. Sci. Pollut. Res. 23 , 22408–22417 (2016).

Huang, C.-J. & Kuo, P.-H. A Deep CNN-LSTM Model for Particulate Matter (PM2.5) Forecasting in Smart Cities. Sensors 18 , 2220 (2018).

Article   ADS   PubMed   PubMed Central   Google Scholar  

Ma, J., Ding, Y., Cheng, J. C. P., Jiang, F. & Wan, Z. A temporal-spatial interpolation and extrapolation method based on geographic Long Short-Term Memory neural network for PM2.5. J. Clean. Prod. 237 , 117729 (2019).

Chang, Y.-S. et al. An LSTM-based aggregated model for air pollution forecasting. Atmos. Pollut. Res. 11 , 1451–1463 (2020).

Biancofiore, F. et al. Recursive neural network model for analysis and forecast of PM10 and PM2.5. Atmos. Pollut. Res. 8 , 652–659 (2017).

Konovalov, I. B., Beekmann, M., Meleux, F., Dutot, A. & Foret, G. Combining deterministic and statistical approaches for PM10 forecasting in Europe. Atmos. Environ. 43 , 6425–6434 (2009).

McKendry, I. G. Evaluation of Artificial Neural Networks for Fine Particulate Pollution (PM 10 and PM 2.5) Forecasting. J. Air Waste Manage. Assoc. 52 , 1096–1101 (2002).

Dutta, A. & Jinsart, W. Air Pollution in Indian Cities and Comparison of MLR, ANN and CART Models for Predicting PM10 Concentrations in Guwahati. India. Asian J. Atmos. Environ. 15 , 68–93 (2021).

Turias, I. J., González, F. J., Martin, M. L. & Galindo, P. L. Prediction models of CO, SPM and SO2 concentrations in the Campo de Gibraltar Region, Spain: a multiple comparison strategy. Environ. Monit. Assess. 143 , 131–146 (2008).

Shang, Z. & He, J. Predicting Hourly <tex>$\mathbf{PM}_{2.5}$</tex> Concentrations Based on Random Forest and Ensemble Neural Network. in 2018 Chinese Automation Congress (CAC) 2341–2345 (IEEE, 2018). https://doi.org/10.1109/CAC.2018.8623175 .

Bozdağ, A., Dokuz, Y. & Gökçek, Ö. B. Spatial prediction of PM10 concentration using machine learning algorithms in Ankara. Turkey. Environ. Pollut. 263 , 114635 (2020).

Murray, C. J. L. et al. Five insights from the Global Burden of Disease Study 2019. Lancet 396 , 1135–1159 (2020).

Haque, M. & Singh, R. Air Pollution and Human Health in Kolkata, India: A Case Study. Climate 5 , 77 (2017).

Rajak, R. & Chattopadhyay, A. Short and Long Term Exposure to Ambient Air Pollution and Impact on Health in India: A Systematic Review. Int. J. Environ. Health Res. 30 , 593–617 (2020).

CPCB. Epidemiological Study on Effect of Air Pollution on Human Health (Adults) in Delhi CENTRAL POLLUTION CONTROL BOARD MINISTRY OF ENVIRONMENT & FORESTS. (2012).

‘India needs to address challenge of adult immunisation†TM - Elets eHealth. https://ehealth.eletsonline.com/2018/10/india-needs-to-address-challenge-of-adult-immunisation/ .

Raju, S., Siddharthan, T. & McCormack, M. C. Indoor Air Pollution and Respiratory Health. Clin. Chest Med. 41 , 825–843 (2020).

Manisalidis, I., Stavropoulou, E., Stavropoulos, A. & Bezirtzoglou, E. Environmental and Health Impacts of Air Pollution: A Review. Front. Public Heal. 8 , 14 (2020).

Fang, Z., Wu, P.-Y., Lin, Y.-N., Chang, T.-H. & Chiu, Y. Air Pollution’s Impact on the Economic, Social, Medical, and Industrial Injury Environments in China. Healthcare 9 , 261 (2021).

Economy and air pollution - Clean Air Fund. https://www.cleanairfund.org/theme/economics/ .

OECD. Climate-resilient Infrastructure. Policy Perspectives. OECD Environ. Policy Pap. 1–46 (2018).

EPA Research: Environmental Justice and Air Pollution | US EPA. https://www.epa.gov/ej-research/epa-research-environmental-justice-and-air-pollution .

Chakraborty, S., Fu, R., Massie, S. T. & Stephens, G. Relative influence of meteorological conditions and aerosols on the lifetime of mesoscale convective systems. Proc. Natl. Acad. Sci. 113 , 7426–7431 (2016).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Sturiale & Scuderi. The Role of Green Infrastructures in Urban Planning for Climate Change Adaptation. Climate 7 , 119 (2019).

Gulia, S. et al. Performance evaluation of air pollution control device at traffic intersections in Delhi. Int. J. Environ. Sci. Technol. 19 , 785–796 (2022).

Allioui, H. & Mourdi, Y. Exploring the Full Potentials of IoT for Better Financial Growth and Stability: A Comprehensive Survey. Sensors 23 , 8015 (2023).

Lalchandani, V. et al. Real-time characterization and source apportionment of fine particulate matter in the Delhi megacity area during late winter. Sci. Total Environ. 770 , 145324 (2021).

Tobler, A. et al. Chemical characterization of PM2.5 and source apportionment of organic aerosol in New Delhi. India. Sci. Total Environ. 745 , 140924 (2020).

Rai, P. et al. Real-time measurement and source apportionment of elements in Delhi’s atmosphere. Sci. Total Environ. 742 , 140332 (2020).

Talukdar, S. et al. Air Pollution in New Delhi during Late Winter: An Overview of a Group of Campaign Studies Focusing on Composition and Sources. Atmosphere (Basel). 12 , 1432 (2021).

Wang, T., Wei, K. & Ma, J. Atmospheric Rivers and Mei-yu Rainfall in China: A Case Study of Summer 2020. Adv. Atmos. Sci. https://doi.org/10.1007/s00376-021-1096-9 (2020).

Sarkar, S., Chauhan, A., Kumar, R. & Singh, R. P. Impact of Deadly Dust Storms (May 2018) on Air Quality, Meteorological, and Atmospheric Parameters Over the Northern Parts of India. GeoHealth 3 , 67–80 (2019).

Wei, W. et al. Comprehensive Assessment of Pollution Sources and Health Impacts in Suburban Area of Shanghai. Toxics 11 , 552 (2023).

Blanco-Donado, E. P. et al. Source identification and global implications of black carbon. Geosci. Front. 13 , 101149 (2022).

Mangaraj, P., Sahu, S. K., Beig, G. & Yadav, R. A comprehensive high-resolution gridded emission inventory of anthropogenic sources of air pollutants in Indian megacity Kolkata. SN Appl. Sci. 4 , 117 (2022).

Rastogi, N. et al. Diurnal variability in the spectral characteristics and sources of water-soluble brown carbon aerosols over Delhi. Sci. Total Environ. 794 , 148589 (2021).

Mukherjee, A. et al. Sources and atmospheric processing of brown carbon and HULIS in the Indo-Gangetic Plain: Insights from compositional analysis. Environ. Pollut. 267 , 115440 (2020).

Tripathi, N. et al. Characteristics of VOC Composition at Urban and Suburban Sites of New Delhi, India in Winter. J. Geophys. Res. Atmos. https://doi.org/10.1029/2021JD035342 (2022).

Act, A. (Prevention and C. of A. P. Air_Act_1981. (1981).

Environment (Protection) Act. The Environment (Protection) Act, 1986 Act No. 29 OF 1986. 1–9 (1986).

Bill, M. V. Amendment. THE GAZETTE OF INDIA EXTRAORDINARY. 1988 , 4–6 (2019).

Rengarajan, S., Palaniyappan, D., Ramachandran, P. & Ramachandran, R. National Green Tribunal of India—an observation from environmental judgements. Environ. Sci. Pollut. Res. 25 , 11313–11318 (2018).

CPCB. Pollution Control Acts, Rules & Notifications Issued Thereunder. Central Pollution Control Board, Ministry of Environment, Forest and Climate Change, Government of India. https://cpcb.nic.in/7thEditionPollutionControlLawSeries2021.pdf (2021).

CPCB. National Air Quality Index. Cent. Pollut. Control Board 1–58. https://app.cpcbccr.com/ccr_docs/About_AQI.pdf (2014).

National Ambient Air Quality Monitoring. Air Quality Trends and Action For Plan. Naaqms 5. http://cpcb.nic.in/upload/NewItems/NewItem_104_airquality17cities-package-.pdf (2006).

Roychowdhury, A. & Somvanshi, A. Breathing Space: How to track and report air pollution under the National Clean Air Programme. Cent. Sci. Environ. (New Delhi, 2020).

Roychowdhury, A., Somvanshi, A. & Kaur, S. Urban Lab-Centre for Science and Environment Analysis Status of air quality monitoring in India: Spatial spread, population coverage and data completeness. https://www.cseindia.org/Note-AQM-Network-analysis.pdf (2023).

Yadav, R. et al. COVID-19 lockdown and air quality of SAFAR-India metro cities. Urban Clim. 34 , 100729 (2020).

Lestari, P., Arrohman, M. K., Damayanti, S. & Klimont, Z. Emissions and spatial distribution of air pollutants from anthropogenic sources in Jakarta. Atmos. Pollut. Res. 13 , 101521 (2022).

Guttikunda, S. K., Nishadh, K. A. & Jawahar, P. Air pollution knowledge assessments (APnA) for 20 Indian cities. Urban Clim. 27 , 124–141 (2019).

Gargava, P. & Rajagopalan, V. Source apportionment studies in six Indian cities—drawing broad inferences for urban PM10 reductions. Air Qual. Atmos. Heal. 9 , 471–481 (2016).

M.C. Mehta And Anr vs Union Of India & Ors on 20 December, 1986. https://indiankanoon.org/doc/1486949/ .

AFVP 2025. Report of the Expert Committee on Auto Fuel Vision & Policy 2025. Press Inf. Bur. 221 , 174. https://cdn.climatepolicyradar.org/navigator/IND/2014/national-auto-fuel-policy-and-auto-fuel-vision-and-policy-2025_c53488e9acdfd8095d576abd64e15892.pdf (2014).

Sahu, V. et al. Assessment of a clean cooking fuel distribution scheme in rural households of India – “Pradhan Mantri Ujjwala Yojana (PMUY)”. Energy Sustain. Dev. 81 , 101492 (2024).

Das, P. K. & Bhat, M. Y. Global electric vehicle adoption: implementation and policy implications for India. Environ. Sci. Pollut. Res. 29 , 40612–40622 (2022).

Gimeno, L. et al. Major Mechanisms of Atmospheric Moisture Transport and Their Role in Extreme Precipitation Events. Annu. Rev. Environ. Resour. 41 , 117–141. https://doi.org/10.1146/annurev-environ-110615-085558 (2016).

Thayyib, P. V. et al. State-of-the-Art of Artificial Intelligence and Big Data Analytics Reviews in Five Different Domains: A Bibliometric Summary. Sustainability 15 , 4026 (2023).

Alzubaidi, L. et al. Review of deep learning: concepts, CNN architectures, challenges, applications, future directions. J. Big Data 8 , 53 (2021).

Fan, J. et al. A Spatiotemporal Prediction Framework for Air Pollution Based on Deep RNN. ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci. IV - 4/W2 , 15–22 (2017).

Rautela, K. S., Kumar, D., Gandhi, B. G. R., Kumar, A. & Dubey, A. K. Application of ANNs for the modeling of streamflow, sediment transport, and erosion rate of a high-altitude river system in Western Himalaya. Uttarakhand. RBRH https://doi.org/10.1590/2318-0331.272220220045 (2022).

Sofi, M. S. et al. Modeling the hydrological response of a snow-fed river in the Kashmir Himalayas through SWAT and Artificial Neural Network. Int. J. Environ. Sci. Technol. https://doi.org/10.1007/s13762-023-05170-7 (2023).

Karagulian, F. et al. Contributions to cities’ ambient particulate matter (PM): A systematic review of local source contributions at global level. Atmos. Environ. 120 , 475–483 (2015).

Mlakar, P., Božnar, M. & Lesjak, M. Neural Networks Predict Pollution. In Air Pollution Modeling and Its Application X (ed. Millán, M. M.) (Springer, 1994).

Arena, P., Fortuna, L., Gallo, A., Nunnari, G. & Xibilia, M. G. Air pollution estimation via neural networks. IFAC Proc. 28 , 787–792 (1995).

Sohn, S. H., Oh, S. C. & Yeo, Y.-K. Prediction of air pollutants by using an artificial neural network. Korean J. Chem. Eng. 16 , 382–387 (1999).

Slini, T., Karatzas, K. & Moussiopoulos, N. Correlation of air pollution and meteorological data using neural networks. Int. J. Environ. Pollut. 20 , 218 (2003).

Kandya, A. Forecasting the Tropospheric Ozone using Artificial Neural Network Modelling Approach: A Case Study of Megacity Madras. India. J. Civ. Environ. Eng. 01 , 2 (2013).

Chaloulakou, A., Saisana, M. & Spyrellis, N. Comparative assessment of neural networks and regression models for forecasting summertime ozone in Athens. Sci. Total Environ. 313 , 1–13 (2003).

Article   ADS   CAS   PubMed   Google Scholar  

Mishra, D. & Goyal, P. Development of artificial intelligence based NO 2 forecasting models at Taj Mahal. Agra. Atmos. Pollut. Res. 6 , 99–106 (2015).

Fernando, H. J. S. et al. Forecasting PM10 in metropolitan areas: Efficacy of neural networks. Environ. Pollut. 163 , 62–67 (2012).

Grivas, G. & Chaloulakou, A. Artificial neural network models for prediction of PM10 hourly concentrations, in the Greater Area of Athens. Greece. Atmos. Environ. 40 , 1216–1229 (2006).

Suleiman, A., Tight, M. R. & Quinn, A. D. Hybrid Neural Networks and Boosted Regression Tree Models for Predicting Roadside Particulate Matter. Environ. Model. Assess. 21 , 731–750 (2016).

Zhang, H., Liu, Y., Shi, R. & Yao, Q. Evaluation of PM 10 forecasting based on the artificial neural network model and intake fraction in an urban area: A case study in Taiyuan City. China. J. Air Waste Manage. Assoc. 63 , 755–763 (2013).

Paschalidou, A. K., Karakitsios, S., Kleanthous, S. & Kassomenos, P. A. Forecasting hourly PM10 concentration in Cyprus through artificial neural networks and multiple regression models: Implications to local environmental management. Environ. Sci. Pollut. Res. 18 , 316–327 (2011).

Mishra, D., Goyal, P. & Upadhyay, A. Artificial intelligence based approach to forecast PM 2.5 during haze episodes: A case study of Delhi India. Atmos. Environ. 102 , 239–248 (2015).

Moisan, S., Herrera, R. & Clements, A. A dynamic multiple equation approach for forecasting PM 2.5 pollution in Santiago. Chile. Int. J. Forecast. 34 , 566–581 (2018).

Liu, H., Jin, K. & Duan, Z. Air PM2.5 concentration multi-step forecasting using a new hybrid modeling method: Comparing cases for four cities in China. Atmos. Pollut. Res. 10 , 1588–1600 (2019).

Chen, J. et al. A comparison of linear regression, regularization, and machine learning algorithms to develop Europe-wide spatial models of fine particles and nitrogen dioxide. Environ. Int. 130 , 104934 (2019).

Jain, S. & Khare, M. Adaptive neuro-fuzzy modeling for prediction of ambient CO concentration at urban intersections and roadways. Air Qual. Atmos. Heal. 3 , 203–212 (2010).

Carbajal-Hernández, J. J., Sánchez-Fernández, L. P., Carrasco-Ochoa, J. A. & Martínez-Trinidad, J. F. Assessment and prediction of air quality using fuzzy logic and autoregressive models. Atmos. Environ. 60 , 37–50 (2012).

Article   ADS   Google Scholar  

Al-Shammari, E. T. Public warning systems for forecasting ambient ozone pollution in Kuwait. Environ. Syst. Res. 2 , 2 (2013).

Bougoudis, I., Demertzis, K. & Iliadis, L. HISYCOL a hybrid computational intelligence system for combined machine learning: the case of air pollution modeling in Athens. Neural Comput. Appl. 27 , 1191–1206 (2016).

Song, Y., Qin, S., Qu, J. & Liu, F. The forecasting research of early warning systems for atmospheric pollutants: A case in Yangtze River Delta region. Atmos. Environ. 118 , 58–69 (2015).

Wang, J., Li, H. & Lu, H. Application of a novel early warning system based on fuzzy time series in urban air quality forecasting in China. Appl. Soft Comput. 71 , 783–799 (2018).

Behal, V. & Singh, R. Personalised healthcare model for monitoring and prediction of airpollution: machine learning approach. J. Exp. Theor. Artif. Intell. 33 , 425–449 (2021).

Arbabsiar, M. H., Ebrahimi Farsangi, M. A. & Mansouri, H. Fuzzy logic modelling to predict the level of geotechnical risks in rock tunnel boring machine (TBM) tunnelling. Rud. Zb. 35 , 1–14 (2020).

Feng, Y., Zhang, W., Sun, D. & Zhang, L. Ozone concentration forecast method based on genetic algorithm optimized back propagation neural networks and support vector machine data classification. Atmos. Environ. 45 , 1979–1985 (2011).

Yeganeh, B., Motlagh, M. S. P., Rashidi, Y. & Kamalan, H. Prediction of CO concentrations based on a hybrid Partial Least Square and Support Vector Machine model. Atmos. Environ. 55 , 357–365 (2012).

García Nieto, P. J., Combarro, E. F., del Coz Díaz, J. J. & Montañés, E. A SVM-based regression model to study the air quality at local scale in Oviedo urban area (Northern Spain): A case study. Appl. Math. Comput. 219 , 8923–8937 (2013).

Luna, A. S., Paredes, M. L. L., de Oliveira, G. C. G. & Corrêa, S. M. Prediction of ozone concentration in tropospheric levels using artificial neural networks and support vector machine at Rio de Janeiro. Brazil. Atmos. Environ. 98 , 98–104 (2014).

Wang, P., Liu, Y., Qin, Z. & Zhang, G. A novel hybrid forecasting model for PM10 and SO2 daily concentrations. Sci. Total Environ. 505 , 1202–1212 (2015).

Freeman, B. S., Taylor, G., Gharabaghi, B. & Thé, J. Forecasting air quality time series using deep learning. J. Air Waste Manage. Assoc. 68 , 866–886 (2018).

Wang, J. & Song, G. A Deep Spatial-Temporal Ensemble Model for Air Quality Prediction. Neurocomputing 314 , 198–206 (2018).

Zhou, Y., Chang, F.-J., Chang, L.-C., Kao, I.-F. & Wang, Y.-S. Explore a deep learning multi-output neural network for regional multi-step-ahead air quality forecasts. J. Clean. Prod. 209 , 134–145 (2019).

Soh, P.-W., Chang, J.-W. & Huang, J.-W. Adaptive Deep Learning-Based Air Quality Prediction Model Using the Most Relevant Spatial-Temporal Relations. IEEE Access 6 , 38186–38199 (2018).

Qi, Y., Li, Q., Karimian, H. & Liu, D. A hybrid model for spatiotemporal forecasting of PM2.5 based on graph convolutional neural network and long short-term memory. Sci. Total Environ. 664 , 1–10 (2019).

Li, Y., Huang, J. & Luo, J. Using user generated online photos to estimate and monitor air pollution in major cities. Proceedings of the 7th International Conference on Internet Multimedia Computing and Service 1–5 (ACM, New York, NY, USA). https://doi.org/10.1145/2808492.2808564 . (2015).

Zhang, L., Nan, Z., Xu, Y. & Li, S. Hydrological impacts of land use change and climate variability in the headwater region of the Heihe River Basin, northwest China. PLoS One 11 , 1–25 (2016).

Li, X. et al. Long short-term memory neural network for air pollutant concentration predictions: Method development and evaluation. Environ. Pollut. 231 , 997–1004 (2017).

Zhang, C. et al. On Estimating Air Pollution from Photos Using Convolutional Neural Network. in Proceedings of the 24th ACM international conference on Multimedia 297–301 (ACM, New York, NY, USA). https://doi.org/10.1145/2964284.2967230 . (2016).

Qiu, Y. et al. Regional aerosol forecasts based on deep learning and numerical weather prediction. Npj Clim. Atmos. Sci. 6 , 71 (2023).

Goyal, M. K. & Rautela, K. S. Aerosol Atmospheric Rivers: Detection and Spatio-Temporal Patterns. https://doi.org/10.1007/978-3-031-66758-9_2 (2024).

Du, S., Li, T., Yang, Y. & Horng, S.-J. Deep Air Quality Forecasting Using Hybrid Deep Learning Framework. IEEE Trans. Knowl. Data Eng. 33 , 2412–2424 (2021).

Araujo, L. N., Belotti, J. T., Alves, T. A., de Tadano, Y. S. & Siqueira, H. Ensemble method based on Artificial Neural Networks to estimate air pollution health risks. Environ. Model. Softw. 123 , 104567 (2020).

Randles, C. A. et al. The MERRA-2 Aerosol Reanalysis, 1980 Onward. Part I: System Description and Data Assimilation Evaluation. J. Clim. 30 , 6823–6850 (2017).

Rautela, K. S., Singh, S. & Goyal, M. K. Aerosol atmospheric rivers: patterns, impacts, and societal insights. Environ. Sci. Pollut. Res. https://doi.org/10.1007/s11356-024-34625-8 (2024).

Buchard, V. et al. Evaluation of the surface PM2.5 in Version 1 of the NASA MERRA Aerosol Reanalysis over the United States. Atmos. Environ. 125 , 100–111 (2016).

Provençal, S., Buchard, V., da Silva, A. M., Leduc, R. & Barrette, N. Evaluation of PM surface concentrations simulated by Version 1 of NASA’s MERRA Aerosol Reanalysis over Europe. Atmos. Pollut. Res. 8 , 374–382 (2017).

Singh, S., Goyal, M. K. & Jha, S. Role of large-scale climate oscillations in precipitation extremes associated with atmospheric rivers: nonstationary framework. Hydrol. Sci. J. https://doi.org/10.1080/02626667.2022.2159412 (2023).

Cheggoju, N. & Satpute, V. R. Blind quality scalable video compression algorithm for low bit-rate coding. Multimed. Tools Appl. 81 , 33715–33730 (2022).

Download references

Acknowledgements

We would like to express our sincere gratitude to the Department of Civil Engineering, Indian Institute of Technology, Indore for their support and resources, which have been instrumental in the successful completion of the present study.

Author information

Authors and affiliations.

Department of Civil Engineering, Indian Institute of Technology Indore, Simrol, Indore, 453552, Madhya Pradesh, India

Kuldeep Singh Rautela & Manish Kumar Goyal

You can also search for this author in PubMed   Google Scholar

Contributions

Kuldeep Singh Rautela: Conception and design, material preparation, data collection and analysis, writing original draft -review and editing. Manish Kumar Goyal: Conception and design, provided critical feedback, writing original draft -review and editing, supervision.

Corresponding author

Correspondence to Manish Kumar Goyal .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ .

Reprints and permissions

About this article

Cite this article.

Rautela, K.S., Goyal, M.K. Transforming air pollution management in India with AI and machine learning technologies. Sci Rep 14 , 20412 (2024). https://doi.org/10.1038/s41598-024-71269-7

Download citation

Received : 11 March 2024

Accepted : 26 August 2024

Published : 02 September 2024

DOI : https://doi.org/10.1038/s41598-024-71269-7

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Air pollution
• Technological innovation
• Regulatory frameworks
• Societal engagement
• Global collaboration

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

• Skip to main content
• Accessibility information

• Enlighten Enlighten

Enlighten Theses

• Latest Additions
• Browse by Year
• Browse by Subject
• Browse by College/School
• Browse by Author
• Browse by Funder
• Login (Library staff only)

In this section

Estimating air pollution and its relationship with human health

Powell, Helen Louise (2012) Estimating air pollution and its relationship with human health. PhD thesis, University of Glasgow.

The health impact of short-term exposure to air pollution has been the focus of much recent research, the majority of which is based on time-series studies. A time-series study uses health, pollution and meteorological data from an extended urban area. Aggregate level data is used to describe the health of the population living with the region, this is typically a daily count of the number of mortality or morbidity events. Air pollution data is obtained from a number of fixed site monitors located throughout the study region. These monitors measure background pollution levels at a number of time intervals throughout the day and a daily average is typically calculated for each site. A number of pollutants are measured including, carbon monoxide (CO); nitrogen dioxide (NO2); particulate matter (PM2.5 and PM10), and; sulphur dioxide (SO2). These fixed site monitors also measure a number of meteorological covariates such as temperature, humidity and solar radiation. In this thesis I have presented extensions to the current methods which are used to estimate the association between air pollution exposure and the risks to human health. The comparisons of the efficacy of my approaches to those which are adopted by the majority of researchers, highlights some of the deficiencies of the standard approaches to modelling such data. The work presented here is centered around three specific themes, all of which focus on the air pollution component of the model. The first and second theme relate to what is used as a spatially representative measure of air pollution and allowing for uncertainty in what is an inherently unknown quantity, when estimating the associated health risks, respectively. For example the majority of air pollution and health studies only consider the health effects of a single pollutant rather than that of overall air quality. In addition to this, the single pollutant estimate is taken as the average concentration level across the network of monitors. This is unlikely to be the average concentration across the study region due to the likely non random placement of the monitoring network. To address these issues I proposed two methods for estimating a spatially representative measure of pollution. Both methods are based on hierarchical Bayesian methods, as this allows for the correct propagation of uncertainty, the first of which uses geostatistical methods and the second is a simple regression model which includes a time-varying coefficient for covariates which are fixed in space. I compared the two approaches in terms of their predictive accuracy using cross validation. The third theme considers the shape of the estimated concentration-response function between air pollution and health. Currently used modelling techniques make no constraints on such a function and can therefore produce unrealistic results, such as decreasing risks to health at high concentrations. I therefore proposed a model which imposes three constraints on the concentration-response function in order to produce a more sensible shaped curve and therefore eliminate such misinterpretations. The efficacy of this approach was assessed via a simulation study. All of the methods presented in this thesis are illustrated using data from the Greater London area.

Actions (login required)

Downloads per month over past year

View more statistics

The University of Glasgow is a registered Scottish charity: Registration Number SC004401

The impact of air pollutant transport on air quality and human health in global and regional model applications

Add to collection, downloadable content.

• March 19, 2019
• Affiliation: Gillings School of Global Public Health, Department of Environmental Sciences and Engineering
• As air pollution can travel long distance, change in emissions from one region influence air quality and associated premature mortality over others. This research uses ensemble-modeled concentrations of anthropogenic ozone (O3) and fine particulate matter (PM2.5) to quantify avoided premature mortality from 20% emission reductions of 6 regions (i.e. North America (NAM), Europe (EUR), South Asia (SAS), East Asia (EAS), Russia/Belarus/Ukraine (RBU) and the Middle East (MDE)) and 3 sectors (i.e. Power and Industry (PIN), Ground Transportation (TRN) and Residential (RES)) and evaluate the impact of interregional transport of precursor emissions from local (i.e. Kao-Ping air basin (KPAB)) and upwind air basin regions (i.e. North and Chu-Miao Air Basin (NCMAB), Central Air Basin (CTAB), Yun-Chia-Nan Air Basin (YCNAB), and Yi-Lan and Hua-Dong Air Basin (YLHDAB)) on O3 and PM2.5 air quality over KPAB. For health impact assessment, we estimate 290,000 (95% CI: 30,000, 600,000) premature O3-related deaths and 2.8 million (0.5 million, 4.6 million) PM2.5-related premature deaths globally for the baseline year 2010. Reducing emissions from MDE and RBU can avoid more O3-related deaths outside of these regions than within while reducing MDE emissions also avoids more PM2.5-related deaths outside of MDE than within. TRN emissions account for the greatest fraction (26-53% of global emission reduction) of O3-related premature deaths in most regions, except for EAS (58%) and RBU (38%) where PIN emissions dominate. For air quality impact assessment, anthropogenic emissions from upwind and local emissions can contribute 17% and 7% of daily maximum 8-hour O3 concentrations, respectively on the highest O3 day while 36.8% and 26.6% of 24-hour average PM2.5 concentrations, respectively during the high PM2.5 days over KPAB, indicating that the upwind emissions play a significant role in KPAB O3 and PM2.5 concentration. The most effective emission control strategy can be approached by reducing upwind anthropogenic NOX emission along with local VOC emission for O3 while upwind anthropogenic NOX emission along with local primary PM2.5 emission for PM2.5. The result highlights the importance of long-range air pollution transport and suggests that emission reductions can improve air quality and have associated health benefits downwind. Therefore, regional cooperation to reduce air pollution transported over long distances may be desirable.
• December 2018
• Community multiscale air quality model (CMAQ)
• fine particulate matter (PM2.5)
• Decoupled Direct Method (DDM)
• Ensemble model
• Environmental science
• Environmental management
• Environmental health
• Long range transport (LRT)
• https://doi.org/10.17615/xzek-2c74
• Dissertation
• In Copyright
• Napelenok, Sergey
• Turpin, Barbara
• Vizuete, William
• West, Jason
• Doctor of Philosophy
• University of North Carolina at Chapel Hill Graduate School

This work has no parents.

Select type of work

Master's papers.

Deposit your masters paper, project or other capstone work. Theses will be sent to the CDR automatically via ProQuest and do not need to be deposited.

Scholarly Articles and Book Chapters

Deposit a peer-reviewed article or book chapter. If you would like to deposit a poster, presentation, conference paper or white paper, use the “Scholarly Works” deposit form.

Undergraduate Honors Theses

Deposit your senior honors thesis.

Scholarly Journal, Newsletter or Book

Deposit a complete issue of a scholarly journal, newsletter or book. If you would like to deposit an article or book chapter, use the “Scholarly Articles and Book Chapters” deposit option.

Deposit your dataset. Datasets may be associated with an article or deposited separately.

Deposit your 3D objects, audio, images or video.

Poster, Presentation, Protocol or Paper

Deposit scholarly works such as posters, presentations, research protocols, conference papers or white papers. If you would like to deposit a peer-reviewed article or book chapter, use the “Scholarly Articles and Book Chapters” deposit option.

• Atmospheric Sciences
• Air Pollution

Causes, Consequences and Control of Air Pollution

• August 2013
• Conference: All India Seminar on Methodologies for Air Pollution Control
• At: Malviya National Institute of Technology, Jaipur, Rajasthan, India

• Rajasthan Technical University

Discover the world's research

• 25+ million members
• 160+ million publication pages
• 2.3+ billion citations

• Dwi Kurniawan
• Swagata Das
• Priya Mondai

• Syed Imran Hussain Shah
• Tahir Hussain Seehar
• Muhammad Raashid

• Asim Biswas
• Jon Warland
• Insha Anjum
• Myrna C Bigueja

• Joselito D Tabardillo

• Recruit researchers
• Join for free
• Login Email Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google Welcome back! Please log in. Email · Hint Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google No account? Sign up

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• My Bibliography
• Collections
• Citation manager

Save citation to file

Email citation, add to collections.

• Create a new collection
• Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed.

• Search in PubMed
• Search in NLM Catalog
• Add to Search

The Impact of Air Pollution Information on Individuals' Exercise Behavior: Empirical Study Using Wearable and Mobile Devices Data

Affiliations.

• 1 School of Business and Management, Royal Holloway, University of London, Egham, United Kingdom.
• 2 School of Computing, National University of Singapore, Singapore, Singapore.
• PMID: 39255029
• DOI: 10.2196/55207

Background: Physical exercise and exposure to air pollution have counteracting effects on individuals' health outcomes. Knowledge on individuals' real-time exercise behavior response to different pollution information sources remains inadequate.

Objective: This study aims to examine the extent to which individuals avoid polluted air during exercise activities in response to different air pollution information sources.

Methods: We used data on individuals' exercise behaviors captured by wearable and mobile devices in 83 Chinese cities over a 2-year time span. In our data set, 35.99% (5896/16,379) of individuals were female and 64% (10,483/16,379) were male, and their ages predominantly ranged from 18 to 50 years. We further augmented the exercise behavior data with air pollution information that included city-hourly level measures of the Air Quality Index and particulate matter 2.5 concentration (in µg/m 3 ), and weather data that include city-hourly level measures of air temperature (ºC), dew point (ºC), wind speed (m/s), and wind direction (degrees). We used a linear panel fixed effect model to estimate individuals' exercise-aversion behaviors (ie, running exercise distance at individual-hour, city-hour, or city-day levels) and conducted robustness checks using the endogenous treatment effect model and regression discontinuity method. We examined if alternative air pollution information sources could moderate (ie, substitute or complement) the role of mainstream air pollution indicators.

Results: Our results show that individuals exhibit a reduction of running exercise behaviors by about 0.50 km (or 7.5%; P<.001) during instances of moderate to severe air pollution, and there is no evidence of reduced distances in instances of light air pollution. Furthermore, individuals' exercise-aversion behaviors in response to mainstream air pollution information are heightened by different alternative information sources, such as social connections and social media user-generated content about air pollution.

Conclusions: Our results highlight the complementary role of different alternative information sources of air pollution in inducing individuals' aversion behaviors and the importance of using different information channels to increase public awareness beyond official air pollution alerts.

Keywords: air pollution; econometric analysis; exercise activity; information sources; wearable and mobile devices.

©Yang Yang, Khim-Yong Goh, Hock Hai Teo, Sharon Swee-Lin Tan. Originally published in JMIR mHealth and uHealth (https://mhealth.jmir.org), 10.09.2024.

PubMed Disclaimer

• Search in MeSH

Related information

Linkout - more resources, full text sources.

• JMIR Publications
• MedlinePlus Consumer Health Information
• MedlinePlus Health Information
• Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.

A Study and Analysis of Air Quality Index and Related Health Impact on Public Health

ICICNIS 2020

13 Pages Posted: 20 Jan 2021

Pranav Shriram

MIT Academy of Engineering Alandi Pune

Srinivas Malladi

Koneru Lakshamaiah Education Foundation

Date Written: January 18, 2021

In 21st Century monitoring environmental value is a challenging and complex task but technology has changed now we can easily find the air quality index of any area. In the existing scenario, we will easily get air quality and environmental parameters on the internet. On the internet generalize statistical data is available which shows the quality of air. In a current scenario, the environmental parameter is impacting human health. In day today, the air quality index going worse and it impacts people's health. The people are facing different health issues like hair loss, asthma, lunges, and heart problems. It is important to know the environmental condition in our day-to-day traveling route. Recently air pollution is increased, the increase of harmful air particles is majorly affecting by the air quality index. Due to exposure to air pollutants, affects human health and causing many hazardous diseases like asthma and many more having a major impact on the lungs. The air pollution is impacting public health and creating multiple health-related problems, this causes a major medical cost every year derived from the disease. To travel safely with considering the health issues is a major concern in an urban area. In this paper, highlighting the impact of the air quality index on the human body, where the Air Quality Index is measured using the concerned information, this information will help to suggest a safe route where the air quality index is low so it can decrease the impact on human health. To find the safest path between source and destination we are using Dijkstra's algorithm. In this paper, we have studied the different research papers and made a comparative study to find the research gaps. The proposed model is a step up in the standard of living in regard to human health. The proposed model is consisting of three main components a) Air quality index b) health impact and c) safest path. In the first model the real-time data is collecting from the government agency or private agency these data will store in the database for analysis. The huge amount of data will handle by evaluation and analysis model in this model the data sanitization process will apply to get the more accurate data from sources it also calculates the different air particles and its ranges this data will transfer to the second model for identifying the health impact. The health impact model will calculate the average air quality index, time for traveling, and distance from the source to destination. This information will process to find short-term and long-term health impacts on public health. In the third safest path model, it will show the different nodes from source to destination at a particular distance. each node information will be stored in the form of a weighted graph in the database. Dijkstra’s algorithm is applying to find the safest path from source to destination. Dijkstra’s algorithm finds the node such that where its air quality value is less the algorithm will identify each node path in the graph such that the average traveling path consists of less air pollution.

Keywords: Air Quality Index, Air Pollution, Health Impact, Dijkstra’s Algorithm, Shortest Path, Navigation, Wireless Sensor Network, Graph Theory, Google Map

Suggested Citation: Suggested Citation

Pranav Shriram (Contact Author)

Mit academy of engineering alandi pune ( email ).

Dehu Phata, Alandi (D) Pune, MA 412105 India

Koneru Lakshamaiah Education Foundation ( email )

Green Fields, Vaddeswaram Guntur District Andhra Pradesh India

Do you have a job opening that you would like to promote on SSRN?

Paper statistics, related ejournals, information systems ejournal.

Subscribe to this fee journal for more curated articles on this topic

Environment & Human Health eJournal

Environment & public health ejournal, environmental pollution & prevention ejournal, world health day ejournal.

• Ways to Give
• Contact an Expert
• Explore WRI Perspectives

Filter Your Site Experience by Topic

Applying the filters below will filter all articles, data, insights and projects by the topic area you select.

• All Topics Remove filter
• Climate filter site by Climate
• Cities filter site by Cities
• Energy filter site by Energy
• Food filter site by Food
• Forests filter site by Forests
• Freshwater filter site by Freshwater
• Ocean filter site by Ocean
• Business filter site by Business
• Economics filter site by Economics
• Finance filter site by Finance
• Equity & Governance filter site by Equity & Governance

Search WRI.org

Not sure where to find something? Search all of the site's content.

What Happens When Extreme Heat and Air Pollution Collide

• Air Quality
• Urban Development
• Urban Efficiency & Climate

On July 22, the world experienced its hottest day in recorded history . The global average temperature reached 17.2 degrees C (62.9 degrees F), prompting U.N. Secretary-General António Guterres to issue a global call to action on extreme heat .

The problem of extreme heat, however, doesn’t exist in a vacuum: When temperatures rise, so too can air pollution levels, as the Intergovernmental Panel on Climate Change’s (IPCC) Sixth Assessment Report (AR6) , an in-depth assessment of the state of climate change authored and reviewed by hundreds of scientists and experts, recognized last year.

Mexico City is one of many urban areas around the globe where this interplay can take hold. Last spring, record temperatures and windless conditions led to a three-day severe pollution alert . The city also activated emergency measures such as limiting traffic to help bring down particulate emissions and ozone levels. It was a dark reminder of the past, harkening back to the 1990s when Mexico City was named the world’s most polluted city. Walking around outside during that time had the same impact as smoking two packs of cigarettes a day.

Since then, Mexico City has taken bold steps to clean the air by introducing measures like prioritizing clean fuels and hastening the shift to electric buses. As a result, the city’s residents are now living healthier and longer lives — on average, three years longer than in previous decades.

But Mexico City faces a new, dangerous threat: longer and more frequent heat waves supercharging its air pollution. And as extreme heat continues to worsen, especially in cities where it is exacerbated by the urban heat island effect , Mexico City and other cities around the world must develop integrated strategies to tackle these dual, correlated challenges.

The Connection Between Heat and Air Pollution

Throughout the thousands of pages of the IPCC’s AR6 report , the authors detailed some of the most alarming climate impacts, including the deeply intertwined relationship between global warming and poor air quality.

Put simply, air pollution levels spike when temperatures rise . This happens in a variety of ways. High temperatures can lead to more frequent droughts and more intense wildfires, both of which increase particulate matter (PM10 and PM2.5). Wildfires also release large amounts of black carbon, nitrogen oxides (NOx), carbon monoxide (CO) and other volatile organic compounds (VOCs). Heat also accelerates biological processes responsible for the degradation of organic waste and wastewater, releasing both air pollutants and greenhouse gases into the air.

Certain pollutants , however, actually feed on the heat. Ground-level (or tropospheric) ozone , an often overlooked but deadly pollutant, forms when VOCs, including methane, and NOx emissions from vehicles, industrial facilities, waste and agricultural burning and other sources chemically react through exposure to sunlight . Warmer temperatures accelerate these reactions, leading to increased ozone production, which manifests as a harmful haze. As a result, during hotter, dryer, less windy months — and especially during heat waves — ground-level ozone can reach dangerous levels in cities.

Countries around the world are seeing the correlation between high temperatures and high ozone levels. During a heat wave that spread across Europe in July 2022, the ground-level ozone in Portugal, Spain and Italy all registered at least double the 100 micrograms per cubic meter (µg/m³) deemed safe by the World Health Organization. That same summer, China also experienced elevated ozone levels during a heat wave . And a recent study made a broader connection between high ozone and high heat in China , based on ozone levels observed between 2014 and 2019.

Increased ground-level ozone can pose serious health risks, particularly to vulnerable populations like children, pregnant people and older adults. Ground-level ozone pollution also threatens critical ecosystems like forests by weakening their ability to respond to stresses like drought, cold and disease. It also damages crop production by reducing plants’ ability to turn sunlight into growth and contributes to rising global temperatures by reducing the ability of trees to absorb carbon dioxide.

A Growing Threat to Public Health

On its own, air pollution can risk lives and livelihoods. But when coupled with extreme heat, the results can be even more deadly. The combination of high temperatures and stagnant air created during heat waves makes people more vulnerable to severe health impacts and urban infrastructure more susceptible to degradation.

Air pollution and heat exposure can each have short and long-term impacts on the respiratory and cardiovascular systems. Ozone alone accounted for roughly 490,000 deaths globally in 2021, and long-term exposure to ozone contributed to roughly 13% of all Constructive Obstructive Pulmonary Disease (COPD) deaths around the world that same year. And one study attributed air pollution, including PM2.5 and ground-level ozone, to more than 7,000 adverse health outcomes in children, 10,000 deaths and 5,000 hospitalizations a year in Jakarta, Indonesia. Extreme heat accounts for roughly 489,000 deaths globally per year. And, during Europe’s 2022 heat wave alone, more than 60,000 heat-related deaths occurred. More research is needed to understand how those deaths could have also been impacted by exposure to air pollutants.

Studies show that risks to individual health are heightened when air pollution and high temperatures are simultaneously at play. For instance, recent research found that high temperatures can exacerbate physiological responses to short-term ozone exposure. According to a 2022 study, mortality risk on days with combined exposure increases by an estimated 21% . Another study on the effect of heat and ozone on respiratory hospitalizations in California found that lower-income neighborhoods and areas with high unemployment rates were disproportionately susceptible to the combined impacts of heat and ozone.

Children and the elderly are the most vulnerable populations facing this deadly combination. Air pollution is currently the second leading risk factor of death for children under 5 years old. Meanwhile, those aged 50 and older suffer at a higher rate from pre-existing conditions such as COPD, diabetes, stroke and heart disease, and are especially susceptible to high levels of tropospheric (ground-level) ozone. Low- and middle-income countries are also disproportionately affected by ozone, as they account for a significant piece of the total number of deaths attributed to ozone since 2010. As air quality worsens and our planet continues to get hotter, the world needs to take urgent action to prevent, and to treat the most vulnerable from, these impacts.

Solutions to a Deadly Combination

Working to weaken the relationship between heat and air quality is critical for reducing the effects of these combined threats. Tackling the emissions that warm our planet and reducing the pollutants that contaminate our air is critical for addressing the root causes of each problem.  But leaders can also take action to more immediately protect residents and build climate resilience.

Health preparedness

As we adjust to rising temperatures, it is vital that our medical systems are able to keep up with the growing number of people affected by heat and air pollution. During heat waves and high pollution events, cities must be prepared to handle an increased intake of people seeking medical attention, especially those with pre-existing conditions who are more vulnerable to respiratory and cardiovascular issues during extreme heat events. By increasing access to medical emergency rooms and live-saving medications, cities can strengthen emergency response capacity and bolster public health infrastructure. Bangkok’s air pollution clinic , dedicated solely to treating patients suffering from air pollution-related illnesses and educating the public about air quality safety, is a potential model for other cities to follow. The more capacity that public health systems have to treat patients suffering from air pollution and heat-related illnesses, the more lives will be saved.

Better air quality forecasting

Early warning systems for extreme weather are critical tools for preparing people for dangerous conditions, as Guterres noted in his call to action on extreme heat.  But access to information about air quality is also essential for navigating the spikes in pollution levels that accompany heat waves. Integrating air pollution forecasting into early warning systems is especially dire in low- and middle-income countries that often lack the data, capabilities and satellite modeling needed to generate their own air quality forecasts. WRI and the NASA Global Modelling and Assimilation Office have collaborated to give cities in lower-income countries access to air quality forecasts through a tool called  CanAIRy Alert . GEOS-CF bias-corrected forecasts are currently available for 121 sites in 21 cities around the world, helping decision-makers better predict increases in air pollution, identify solutions and prepare public health responses.

Integrated climate and clean air solutions

The impacts of air pollution and extreme heat are intertwined, so their solutions should also be connected. Reducing emissions — by mandating strict standards for industries, improving public transport and encouraging non-motorized transport, for example — can clean the air while helping curb the temperature increases associated with climate change. Ending dependency on fossil fuels and investing in renewable energy sources are also imperative and can help reduce both temperatures and air pollution levels.

In the short term, cities should develop emergency response plans to hazardous heat and air quality, which could include limiting cars allowed on the roads and shutting down high-polluting factories to temporarily reduce emissions during high pollution events. Cities can increase their longer-term resilience to both heat and air pollution through enhanced urban planning that could feature open ventilation corridors to more effectively disperse air pollution. They can also build green infrastructure like urban tree cover, which can interrupt the urban heat island effect by cooling cities while also absorbing air pollutants.

Building Momentum

Guterres’ call to action in response to the record-breaking July 2024 heat wave is a welcome, and essential, step forward.  As part of this mobilization, countries around the world must also consider the role that air pollution is playing. The combination of extreme heat and poor air quality is especially harmful to human health and our ecosystems, and the world must take swift action on both.

A better understanding of the interplay between high temperatures and air pollution is critical for implementing immediate and long-term solutions to the problem. Deeper knowledge about the connection, and more widespread and equitable access to data and tools, can lead to more effective preparations. Solutions to this dual threat should also consider the susceptibilities and vulnerabilities of different populations, like disproportionate health impacts, illnesses and hospitalizations. The next step is building global momentum — and taking collective action to maintain it.

Nina Saaty contributed to this article.

Relevant Work

Southeast asian cities have some of the most polluted air in the world. el niño is making it worse, 5 under-recognized impacts of air pollution, how climate change affects health and how countries can respond, 3 reasons why countries must take faster action to reduce short-lived climate pollutants, how you can help.

WRI relies on the generosity of donors like you to turn research into action. You can support our work by making a gift today or exploring other ways to give.

Stay Informed

World Resources Institute 10 G Street NE Suite 800 Washington DC 20002 +1 (202) 729-7600

© 2024 World Resources Institute

Envision a world where everyone can enjoy clean air, walkable cities, vibrant landscapes, nutritious food and affordable energy.

92 Air Pollution Essay Topic Ideas & Examples

🏆 best air pollution topic ideas & essay examples, 👍 good essay topics on air pollution, 💡 interesting topics to write about air pollution, ❓air pollution research questions.

• Air Pollution and Its Impact on Human Health Community needs assessment is a systematic process in which the health educator, the nurse and other health care professionals together with the members of the community determine the health problems & needs of the community […]
• Car Air Pollution Further, NO2 can prevent the flow of oxygen in the blood to other parts of the body like the brain. These toxic substances settle in the lungs and disrupt the normal flow of air in […]
• Nurse Associate’s Role in Air Pollution Prevention This paper analyzes current health promotion strategies in Somerset and the United Kingdom, obstacles to preventative health strategies, health screening programs, the impact of psycho-social, economic, and behavioral factors, epidemiology and genomics, vaccination and immunization […]
• Environmental Behavior and Air Pollution in Ohio Once people become aware of the harmful effects of air pollution on the environment and health, it is likely that they will adopt positive behaviors, reduce behaviors and activities that contribute to air pollution and […]
• The Public Perceptions of Air Pollution and Related Policies in London The primary questions for consideration are the public perceptions of air pollution and related policies in London and other cities of the United Kingdom, previous surveys regarding existing policies related to the environment or air […]
• Water & Air Pollution and Health Issues in Brazil The main environmental effects of pollution include the destruction of marine habitats, water scarcity, and anoxia. The conclusion is informative because the writer includes strategies to alleviate the problem of air and water pollution in […]
• Air Pollution Impacts on Weather and Climate Air pollution is rated to be the major cause of discomfort in the living creatures of the world for air is essential for the survival of every living creature.
• An Investigation of Green Roofs to Mitigate Air Pollution With Special Reference to Tehran, Iran Thus, the aim of the research is to inquire into the basic information on the concept of green roofs, to answer the research questions on different attributes of green roofs, methods used to construct green […]
• Air Pollution Sources, Effects and Ways of Minimizing This paper discusses the various sources of air pollution, the effects of air pollution, and ways of minimizing air pollution. Definitely, the destruction of the atmosphere is a serious issue of concern to many people, […]
• Preposition 23: Suspension of Air Pollution Control Act On the one hand, it was approved by the California Air Resources Board that considered it more realistic to suspend the implementation of this law due to the existing $ billion deficit leading to the […]
• Air Pollution in Beijing and Its Effects on Society It is worth noting that different regions/countries/cities in the world have different levels of air pollution depending on the intensity/presence of causing agents and the techniques applied in dealing with air pollution.
• Dealing With Air Pollution Polluted air contains nitrogen oxides and other toxic substances that dissolve in the atmosphere to return to the Earth in the form of acid rain, which is detrimental to the ecosystem.
• Smog and Air Pollution in Los Angeles The city is often covered with a yellow veil in the sky, so the problem of smog is an actual problem of the state.
• The Ecogeographical Impact of Air Pollution The weakness of the text is that the safety of NPs and their probable toxic effects on human health and the environment are not evaluated.
• Air Pollution and Impact of Transportation Emissions of greenhouse gases, air pollution, the release of ballast water, aquatic invasive species, and oil and chemical leaks are only some of the environmental problems that marine transportation continues to cause.
• Air Pollution and Lung Disease To design a study in order to explore the link between lung disease and air pollution, it would be possible to follow a four-step process started by identifying the level or unit of analysis.
• Air Pollution in China: Atmospheric Chemistry and Physics One of the most acute environmental problems in China is air pollution, which the authorities are trying to solve, but still, many people, factories, and active processes of globalization do not allow environmental programs to […]
• Air Pollution and Vulnerability to Covid-19 In other words, the findings will be used as one of the key arguments for showing that air pollution is detrimental to both individual and societal health.
• Fundamentals of Air Pollution The components of secondary air pollution include ozone and nitrogen oxides. Smog occurs when “car exhausts are exposed to direct sunlight”.
• Air Pollution: The Problem’ Review Indoor pollution and related conditions are a big burden to the already suffering world according to the reports of the world health organization that it’s the 8th most important risk factor and is perceived to […]
• Law, Property Rights, and Air Pollution In the law of torts, ‘harm’ is considered when there is physical invasion to a person there fore in the case there was no violation of this law as the secretary was not harmed by […]
• Air Pollution in Middle East: Saudi Arabia The rate of air pollution in the world has increased gradually since the advent of the industrial revolution in the early 1800s.
• Air Pollution and Health Issues in the US The industry of health care is closely connected to the industrial activities sector, which has the largest impact on the atmosphere through polluting the air, soil, and waters.
• Air Pollution and Ecological Perspectives of the Atmosphere The major contributors to CO2, one of the main pollutants in the atmosphere, are the burning of fossil fuels and deforestation.
• How China Cuts Its Air Pollution 5, which is the smallest and one of the most harmful polluting particles, were 54 percent lower in the last quarter of 2017 as compared to the same period in 2016, specifically in Beijing.
• Climate Change: Reducing Industrial Air Pollution One of the most effective measures of air quality in the USA is the Air Quality Index, which estimates air conditions by concentrations of such pollutants as particle solution, nitrogen and sulfur dioxide, carbon monoxide, […]
• Air Pollution in the United Arab Emirates’ Cities In the article called Evaluating the Potential Impact of Global Warming on the UAE Residential Buildings, the author focuses on the negative consequences of global warming on the situation in the United Arab Emirates.
• Climate Change, Air Pollution, Soil Degradation Then followed by outdoor air pollution, soil degradation which can also be called as soil contamination, global overpopulation, drinking water pollution, nuclear waste build-up, disappearing of the water supplies, indoor air pollution, depletion of the […]
• Air Pollution in Washington State and Healthy Living of People The problem of air pollution is closely related to the issue of the energy supply of the US. Due to the high level of air pollution in Washington state, there is a growing threat to […]
• Air Pollution as the Trigger of the Ecological Catastrophe The key data collection tool is a survey that is targeted at determining the main factors of air pollution, finding out the social opinion regarding the quality of air in different cities, and estimating the […]
• Air Pollution Impact on Children’s Health in the US In these parts of the country, the level of air pollution is much higher. Nevertheless, the growing number of vehicles in the United States contributes to air pollution.
• Air Pollution in Los Angeles The escalation of congestion in the city has worsened the problem of air pollution because of the volume of unhealthy air emitted in the atmosphere.
• Environmental Revolution: Air Pollution in China For instance, a case study of the current pollution levels in China reveals that the country is struggling with the management of hazy weather.
• The New York City Air Pollution As the reports say, the state of health of some of the New York residents has grown increasingly worse, mostly due to the air pollution and the diseases that it has triggered.
• Air Pollution Effects on the Health and Environment According to the National Ambient Air Quality Standards, there are six principal air pollutants, the excess of which critically affects the health, lifestyle, and welfare of the population. Still, to my mind, the priority should […]
• Environmental Justice and Air Pollution in Canada One of the best ways is to explain that air pollution is a major contributor to the burgeoning problem of global warming.
• Principles of Air Pollution Control and Analysis The increased attention to air quality is a recent development as people were previously not concerned about the quality of air in the atmosphere.
• New York City Air Pollution Problem One positive impact of technological advancements on the environment in New York is the ability to provide communication options that are friendly to the environment.
• China’s Air Pollution Problem The fact that we do not know the rate at which the economy is slowing down denotes that we cannot tell the rate at which air pollution in the country is reducing and those who […]
• China’s Air Pollution Is Not Unique China and the United States of America have adversely been mentioned to be the leading polluters of the atmosphere. The recent statistics indicate that the gap between the level of pollution by China and that […]
• Air Pollution: Human Influence on Environment For these reasons, the emission of aerosols in the air has become a major issue of concern allover the world and it is one of the many issues that need to be addressed and controlled […]
• Air Pollution Sources in Houston Though pollution is virtually everywhere, the paper focuses on Houston, one of the major cities is the US that have unacceptable levels of pollutants that pose health risks to the lives of people, plants, and […]
• Air Pollution: Public Health Impact In this regard, the paper explores various articles on opencast coals mining, aviation emissions, and geological storage of carbon dioxide and public health concerns in air pollution.
• Does Air Pollution in Schools Influence Student Performance? When the quality of the air is poor, allergens are likely to be present in the air. To this end, the paper has revealed that poor IAQ may cause a number of short and long-term […]
• Impact of Blowing Drums on Air Pollution According to the CSB report, BP failed to “implement or heed all the safety recommendations regarding the blowdown drums before the blast”.
• Air Pollution Effects on the Health in China The justification of the study is premised on the fact that China is one of the world’s largest coal producers and consumers, hence the need to evaluate the health implications of coal pollution on the […]
• Air Pollution and Its Consequences Air pollution refers to the infusion of chemicals, particles and biological matter that are hazardous and are the cause of discomfort to humanity and other living organisms into the atmosphere.
• Air Pollution by Automobiles This paper shall address specific automobile pollutants in relation to causes and public health, to draft possible recommendations to the obstacles, in order to manage the problem.
• Climate and Air Pollution The earth has a number of climatic systems that ensure the distribution of heat across the face of the earth. Global warming is the result of retention of heat by the earth’s atmosphere originally from […]
• A Discussion of Air Pollution & Related Health Implications on the Community The first task in the multidisciplinary team should be to identify the leading sources of air pollution within the community and the nature of the specific toxics or hazardous chemicals associated with the pollutants.
• Are Land Values Related to Ambient Air Pollution Levels?
• What Are the Main Causes of Air Pollution?
• How Pollution Affects the Environment?
• Could Air Pollution Have a Negative Impact on Water Quality?
• Does Air Pollution Affect Consumption Behavior?
• How Can Air Pollution Affect an Individual’s Health?
• What Are the Natural Causes of Air Pollution?
• Does Air Pollution Affect Global Warming?
• Is Air Pollution Getting Worse?
• How Are Air Pollution Harmful to Plants?
• When Did Air Pollution Start?
• How Do American Cities Handle Air Pollution?
• Does Expanding Regional Train Service Reduce Air Pollution?
• What Affects Air Pollution the Most?
• How Does Outdoor Air Pollution Affect the Quality of Indoor Air?
• Why Can Air Pollution Harm the Environment Dramatically?
• How Do Trees Prevent Air Pollution?
• What Are the Biggest Air Pollution Sources?
• Does Air Pollution Affect Animals?
• How Can We Prevent Air Pollution at Home?
• Where Is Air Pollution the Worst?
• How Does Air Pollution Affect Humans?
• What Are the Best Ways to Reduce Air Pollution?
• Is Air Pollution the Biggest Killer?
• How Does Acid Rain Affect Air Pollution?
• Extinction Research Topics
• Ocean Pollution Titles
• Hazardous Waste Essay Topics
• Noise Pollution Essay Titles
• Overpopulation Topics
• Climate Change Titles
• Recycling Research Ideas
• Deforestation Research Ideas
• Chicago (A-D)
• Chicago (N-B)

IvyPanda. (2024, March 2). 92 Air Pollution Essay Topic Ideas & Examples. https://ivypanda.com/essays/topic/air-pollution-essay-topics/

"92 Air Pollution Essay Topic Ideas & Examples." IvyPanda , 2 Mar. 2024, ivypanda.com/essays/topic/air-pollution-essay-topics/.

IvyPanda . (2024) '92 Air Pollution Essay Topic Ideas & Examples'. 2 March.

IvyPanda . 2024. "92 Air Pollution Essay Topic Ideas & Examples." March 2, 2024. https://ivypanda.com/essays/topic/air-pollution-essay-topics/.

1. IvyPanda . "92 Air Pollution Essay Topic Ideas & Examples." March 2, 2024. https://ivypanda.com/essays/topic/air-pollution-essay-topics/.

Bibliography

IvyPanda . "92 Air Pollution Essay Topic Ideas & Examples." March 2, 2024. https://ivypanda.com/essays/topic/air-pollution-essay-topics/.

IvyPanda uses cookies and similar technologies to enhance your experience, enabling functionalities such as:

• Basic site functions
• Ensuring secure, safe transactions
• Secure account login
• Remembering account, browser, and regional preferences
• Remembering privacy and security settings
• Analyzing site traffic and usage
• Personalized search, content, and recommendations
• Displaying relevant, targeted ads on and off IvyPanda

Please refer to IvyPanda's Cookies Policy and Privacy Policy for detailed information.

Certain technologies we use are essential for critical functions such as security and site integrity, account authentication, security and privacy preferences, internal site usage and maintenance data, and ensuring the site operates correctly for browsing and transactions.

Cookies and similar technologies are used to enhance your experience by:

• Remembering general and regional preferences
• Personalizing content, search, recommendations, and offers

Some functions, such as personalized recommendations, account preferences, or localization, may not work correctly without these technologies. For more details, please refer to IvyPanda's Cookies Policy .

To enable personalized advertising (such as interest-based ads), we may share your data with our marketing and advertising partners using cookies and other technologies. These partners may have their own information collected about you. Turning off the personalized advertising setting won't stop you from seeing IvyPanda ads, but it may make the ads you see less relevant or more repetitive.

Personalized advertising may be considered a "sale" or "sharing" of the information under California and other state privacy laws, and you may have the right to opt out. Turning off personalized advertising allows you to exercise your right to opt out. Learn more in IvyPanda's Cookies Policy and Privacy Policy .

Essay on Air Pollution for Students and Children

500+ words essay on air pollution.

Essay on Air Pollution – Earlier the air we breathe in use to be pure and fresh. But, due to increasing industrialization and concentration of poisonous gases in the environment the air is getting more and more toxic day by day. Also, these gases are the cause of many respiratory and other diseases . Moreover, the rapidly increasing human activities like the burning of fossil fuels, deforestation is the major cause of air pollution.

How Air Gets Polluted?

The fossil fuel , firewood, and other things that we burn produce oxides of carbons which got released into the atmosphere. Earlier there happens to be a large number of trees which can easily filter the air we breathe in. But with the increase in demand for land, the people started cutting down of trees which caused deforestation. That ultimately reduced the filtering capacity of the tree.

Moreover, during the last few decades, the numbers of fossil fuel burning vehicle increased rapidly which increased the number of pollutants in the air .

Causes Of Air Pollution

Its causes include burning of fossil fuel and firewood, smoke released from factories , volcanic eruptions, forest fires, bombardment, asteroids, CFCs (Chlorofluorocarbons), carbon oxides and many more.

Besides, there are some other air pollutants like industrial waste, agricultural waste, power plants, thermal nuclear plants, etc.

Greenhouse Effect

The greenhouse effect is also the cause of air pollution because air pollution produces the gases that greenhouse involves. Besides, it increases the temperature of earth surface so much that the polar caps are melting and most of the UV rays are easily penetrating the surface of the earth.

Get the huge list of more than 500 Essay Topics and Ideas

Effects Of Air Pollution On Health

Moreover, it increases the rate of aging of lungs, decreases lungs function, damage cells in the respiratory system.

Ways To Reduce Air Pollution

Although the level of air pollution has reached a critical point. But, there are still ways by which we can reduce the number of air pollutants from the air.

Reforestation- The quality of air can be improved by planting more and more trees as they clean and filter the air.

Policy for industries- Strict policy for industries related to the filter of gases should be introduced in the countries. So, we can minimize the toxins released from factories.

Use of eco-friendly fuel-  We have to adopt the usage of Eco-friendly fuels such as LPG (Liquefied Petroleum Gas), CNG (Compressed Natural Gas), bio-gas, and other eco-friendly fuels. So, we can reduce the amount of harmful toxic gases.

To sum it up, we can say that the air we breathe is getting more and more polluted day by day. The biggest contribution to the increase in air pollution is of fossil fuels which produce nitric and sulphuric oxides. But, humans have taken this problem seriously and are devotedly working to eradicate the problem that they have created.

Above all, many initiatives like plant trees, use of eco-friendly fuel are promoted worldwide.

{ “@context”: “https://schema.org”, “@type”: “FAQPage”, “mainEntity”: [{ “@type”: “Question”, “name”: “Mention five effect of air pollution on human health?”, “acceptedAnswer”: { “@type”: “Answer”, “text”: “The major risk factor related to human health are asthma, lung cancer, Alzheimer, psychological complications, and autism. Besides, there are other effects of air pollution on a person’s health.”} }, { “@type”: “Question”, “name”: “What is the effect of air pollution in the environment?”, “acceptedAnswer”: { “@type”: “Answer”, “text”:”Acid, rain, ozone depletion, greenhouse gases, smog are many other things are the cause of air pollution that affect the environment severely.”} }] }

Customize your course in 30 seconds

Which class are you in.

• Travelling Essay
• Picnic Essay
• Our Country Essay
• My Parents Essay
• Essay on Favourite Personality
• Essay on Memorable Day of My Life
• Essay on Knowledge is Power
• Essay on Gurpurab
• Essay on My Favourite Season
• Essay on Types of Sports

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Download the App

• Cathedral City
• Crime & Courts
• Environment
• Desert Hot Springs
• Indian Wells
• Palm Desert
• Palm Springs
• Rancho Mirage
• National Politics
• Homeless in the Valley

California has dramatically improved its air quality, but racial disparities persist

Black and Latino communities in California breathe the dirtiest air despite state policies that have successfully restricted vehicle emissions, according to a study.

The study, published in Science Advances on Wednesday, found racial disparities in exposure to vehicle air pollution from 2000 to 2019.

Although fine particulate matter in the air was reduced by 65% in the state during the 19-year period, there is still a disparity in air quality between white communities and communities of color.

The 2000 to 2019 time frame was studied because the state's regulatory agencies pursued aggressive policies to reduce emissions across the entire on-road vehicle fleet, according to the study.

“California has been remarkably effective at controlling pollution from on-road emission sources, from cars and light-duty trucks to heavy-duty vehicles,” said Joshua Apte, the study's senior author and an associate professor of civil and environmental engineering at UC Berkeley in a statement . “This is a tremendous win for public health, but our work isn’t done because there’s been no narrowing of the relative gap between the most exposed and least exposed racial and ethnic groups.”

Emissions from cars, or light-duty vehicles in general, are the biggest source of exposure for communities of color, according to the study.

Air pollution exposure disparities are larger by race or ethnicity because of historical racism and racist practices such as housing discrimination and highway relocation. The practices segregated cities and placed high-pollution sources near communities of color, according to the study.

"Highways are disproportionately concentrated in some neighborhoods and not others, and we have to spread that burden around more equally if we want to ultimately get rid of these disparities,” Apte said.

How does air pollution affect people's health?

Short -term exposure to fine particulate matter in the air can cause premature mortality, increased hospital admissions for heart or lung causes, asthma attacks, and other issues, according to the California Air Resources Board.

Long-term exposure to air pollution has been linked to premature death in people who have chronic heart or lung disease and reduced lung function growth in children, according to the board.

What has California done to reduce air pollution?

California requires cleaner fuels and advanced emissions controls for cars on the road.

California also plans to ban the sale of new gas cars by 2035 and require all vehicles to be electric or hydrogen-powered.

California is the nation's most populous state, with 39 million people, and accounts for 10% of the U.S. car market.

According to the Department of Energy , California had the most electric vehicles (EVs) in the United States, with approximately 1.2 million registrations as of December 2023.

Will electric vehicles help reduce air pollution?

Even if we transition to all-electric vehicles, air pollution will not be eliminated because tires and brakes will still emit emissions, Apte said.

"As long as vehicles and vehicle emissions are disparately concentrated in overburdened communities, those communities are going to have a higher share of the overall exposure,” he said. “California’s increasing focus on policies that can bring down emissions in places that are historically overburdened, such as accelerated retirement of dirty old vehicles that operate near ports and rail yards ,  is one way to address these disparities."

Wes Woods II covers West County for the Ventura County Star. Reach him at  [email protected] , 805-437-0262 or  @JournoWes .

Home — Essay Samples — Environment — Air Pollution — Air Pollution: Causes, Effects, And Solutions

Air Pollution: Causes, Effects, and Solutions

• Categories: Air Pollution Environmental Issues Pollution

About this sample

Words: 1158 |

Published: Feb 8, 2022

Words: 1158 | Pages: 2 | 6 min read

Table of contents

Causes and effects of air pollution, possible solutions to the problem of air pollution.

This essay explores the critical issue of air pollution, emphasizing human activities as the primary contributors, including fossil fuel combustion, industrial emissions, and transportation. It discusses the harmful effects of air pollution on human health, the environment, and global ecosystems, highlighting the urgency of addressing this global crisis. Offering practical solutions, such as adopting renewable energy sources and implementing cleaner technologies, the essay serves as a problem and solution essay example detailed , advocating for collective action to mitigate the adverse effects of air pollution and protect the planet for future generations.

Works Cited

• Begum, B. A., & Hill, J. A. (2019). Air Pollution and Public Health: A Primer. In Air Pollution and Health (pp. 3-22). Elsevier.
• Bhaskar, A., & Upadhyay, R. (2021). Air Pollution: Causes, Impacts and Control Measures. In Environmental Pollution and Control Measures (pp. 29-52). Springer.
• Chakraborty, S., & Pervez, S. (2019). Impact of Air Pollution on Human Health and Environment: An Overview. In Environmental Impact of Chemical Pollution (pp. 3-24). Elsevier.
• Dockery, D. W., & Pope III, C. A. (2020). Air Pollution and Health. In Air Pollution and Health (pp. 23-35). Elsevier.
• Garg, A., Martin, R. V., & Crounse, J. D. (2021). Air Pollution and Its Effects on Climate and Health. In Climate and Air Pollution (pp. 1-21). Springer.
• Hidy, G. M., & Pennell, W. T. (2020). Air Pollution: Chemicals and Particles in Ambient Air and Their Health Effects. In Encyclopedia of Environmental Health (pp. 22-31). Elsevier.
• Kampa, M., & Castanas, E. (2020). Human Health Effects of Air Pollution. Environmental Pollution, 151, 362-367.
• Lelieveld, J., Evans, J. S., Fnais, M., Giannadaki, D., & Pozzer, A. (2015). The Contribution of Outdoor Air Pollution Sources to Premature Mortality on a Global Scale. Nature, 525(7569), 367-371.
• Pruss-Ustun, A., Wolf, J., Corvalan, C., Bos, R., & Neira, M. (2016). Preventing Disease through Healthy Environments: A Global Assessment of the Burden of Disease from Environmental Risks. World Health Organization.
• World Health Organization. (2018). Ambient Air Pollution: A Global Assessment of Exposure and Burden of Disease. World Health Organization.

Cite this Essay

To export a reference to this article please select a referencing style below:

Let us write you an essay from scratch

• 450+ experts on 30 subjects ready to help
• Custom essay delivered in as few as 3 hours

Get high-quality help

Dr. Heisenberg

Verified writer

• Expert in: Environment

+ 120 experts online

By clicking “Check Writers’ Offers”, you agree to our terms of service and privacy policy . We’ll occasionally send you promo and account related email

No need to pay just yet!

Related Essays

3 pages / 1370 words

3 pages / 1393 words

2 pages / 718 words

3 pages / 1557 words

Remember! This is just a sample.

You can get your custom paper by one of our expert writers.

121 writers online

Still can’t find what you need?

Browse our vast selection of original essay samples, each expertly formatted and styled

Related Essays on Air Pollution

The air we breathe today is full of toxic and hazardous pollution. Pollution is a big problem to our health and the environment. Pollution is created by individuals, communities or industries that collect and dispose pollutants [...]

Environmental pollution is a pressing issue that affects the health and well-being of our planet and its inhabitants. Pollution can take many forms, including air pollution from industrial emissions and vehicle exhaust, water [...]

Traffic jams are the bane of urban existence, causing frustration, wasted time, and environmental damage. As cities continue to grow, the problem of traffic congestion becomes increasingly dire. This essay explores the causes [...]

As the world grapples with the escalating crisis of climate change, the term "carbon footprint" has become increasingly prevalent in both academic and public discourse. A carbon footprint refers to the total amount of [...]

Pakistan, a nation characterized by its rich cultural heritage and diverse landscapes, is grappling with a growing concern – the environmental challenges plaguing its big cities. As these urban centers continue to swell with [...]

Air pollution has continued to increase over recent decades, particularly in Vietnam. The World Health Organization’s cancer agency said, the air we breathe is laced with cancer-causing substances and is being officially [...]

Related Topics

By clicking “Send”, you agree to our Terms of service and Privacy statement . We will occasionally send you account related emails.

Where do you want us to send this sample?

By clicking “Continue”, you agree to our terms of service and privacy policy.

Be careful. This essay is not unique

This essay was donated by a student and is likely to have been used and submitted before

Download this Sample

Free samples may contain mistakes and not unique parts

Sorry, we could not paraphrase this essay. Our professional writers can rewrite it and get you a unique paper.

Please check your inbox.

We can write you a custom essay that will follow your exact instructions and meet the deadlines. Let's fix your grades together!

Get Your Personalized Essay in 3 Hours or Less!

We use cookies to personalyze your web-site experience. By continuing we’ll assume you board with our cookie policy .

• Instructions Followed To The Letter
• Deadlines Met At Every Stage
• Unique And Plagiarism Free