Estimation of Ambient Air Quality in Delhi

For decades, it has been well known that air pollution causes adverse health effects on humans (Thurston et al., 1989, Shandilya and Khare, 2012). Since the late 1980s, epidemiological studies were published relating the mortality and morbidity to the ambient levels of fine particles without showing the expected results (Borja-Aburto et al., 1997). Most emphasis was then given to the particles with an aerodynamic diameter smaller than 10μm (PM10) particularly to the class of the fine particulates. A special study carried out by United Nations Sources (UNECE, 1979) shows that the frequency of finding a particle of diameter range 2-4m is 27.5%, which is higher in an overall range of 1-12m, so the danger of the fine particulates of diameter ranges 2-4m is higher than the danger by any particle which may lie in the range of 4-12m. Therefore, there is a rising awareness about further fine fraction i.e., particles with an aerodynamic diameter smaller than 2.5μm (PM2.5). Aerosols, which originate directly from the sources, are termed as “Primary aerosols”. There may be the conversion of gaseous matter into the particulate due to many Physico-Chemical reactions. Such types of particulates are known as “Secondary Aerosols” (Perkins, 1974, Shandilya et al., 2009).


Introduction
For decades, it has been well known that air pollution causes adverse health effects on humans (Thurston et al., 1989, Shandilya and. Since the late 1980s, epidemiological studies were published relating the mortality and morbidity to the ambient levels of fine particles without showing the expected results (Borja-Aburto et al., 1997). Most emphasis was then given to the particles with an aerodynamic diameter smaller than 10µm (PM 10 ) particularly to the class of the fine particulates. A special study carried out by United Nations Sources (UNECE, 1979) shows that the frequency of finding a particle of diameter range 2-4m is 27.5%, which is higher in an overall range of 1-12m, so the danger of the fine particulates of diameter ranges 2-4m is higher than the danger by any particle which may lie in the range of 4-12m. Therefore, there is a rising awareness about further fine fraction i.e., particles with an aerodynamic diameter smaller than 2.5µm (PM 2.5 ). Aerosols, which originate directly from the sources, are termed as "Primary aerosols". There may be the conversion of gaseous matter into the particulate due to many Physico-Chemical reactions. Such types of particulates are known as "Secondary Aerosols" (Perkins, 1974. Delhi, the Capital and third largest city of India, ranks third in population among other Indian cities, estimated among 8.5 million (Crowther et al., 1990). Rapid urbanization and the unprecedented industrial and economic development during the last three decades have increased the vehicular population of Delhi by several folds. Consequently, harmful emission such as Pb, PAH, SO 2, NOx, particulate matter and carbon monoxide have also registered a sharp increase in these intervening years. As per the WHO report, Delhi is the fourth most populated city in the world. A 1997 study by the Center for Science and Environment (CSE) revealed that at least one person died prematurely every hour in Delhi in 1995 strictly because of Suspended Particulate Matter (SPM). A 1998 study by All India Institute of Medical Sciences (AIIMS), New Delhi, showed how emergency visits and deaths due to respiratory and heart problems are the highest when particulate levels peak during the winter (Down to Earth, 2000). It is estimated that in Delhi, the incidence of chronic bronchitis is 6-14 times more than the average elsewhere in India because of the thermal power plants and the vehicular pollution (Rao & Rao, 1989).
The objective of this study was to monitor NOx, SO 2 , PM 10, PM 2.5 , and Non Respirable Suspended Particulate (NRSP) levels in South Delhi at different land use patterns; to compare it with the levels found in the crowded residential site of East Delhi; and to compare the observed SPM,NOx, and SO 2 levels with the prescribed CPCB limits (except PM 2.5 compared with USEPA primary standards), which serves as the guidelines for the environmental planning of the region. Another objective was to find out the spatial and diurnal variation of NRSP, the Organic Matter level in NRSP, and the 12 hour variation of NOx and SO 2 . The final objective was the determination of Air Pollution Index at different places.

Materials and methods
A background to the Study Area is outlined in detail in Shandilya et al. (2007). In order to assess ambient pollution level, particulate samples were taken with the help of respirable dust sampler and fine particulate sampler. For the high volume Measurement of RSP, air was drawn through a 20.3X25.4cm Whatman GF/A glass fiber filter at a flow rate which was kept around 1-1.1m 3 /min. Detailed Procedures of Equipment (APM 460DX) are specified in operating manuals, available from equipment manufacturers (Envirotech Instruments Private Limited). The gases were determined with APM 411. The entire weighing procedure was done under controlled conditions (room temperature) in a balance room at Envirotech Instruments Pvt. Ltd., New Delhi using a Dhona-100DS microbalance. For PM 2.5 monitoring, ambient air enters APM550 system through an omni directional inlet designed to provide a clean aerodynamic cut-point for particles >10μm.

Sampling sites and details
The air quality monitoring in Delhi for present investigation was carried out from October 19, 2000to November 29, 2000, during which PM 10 , PM 2.5 , and non-respirable suspended particulate concentrations were measured at four monitoring sites. In order to obtain a clear picture of a source-based pollution, it was required that the monitoring must be carried out at different places of different characteristics in New Delhi. It was important to consider residential, industrial, and commercial areas for carrying out such source differentiation. Based on the above considerations, four sites were chosen for setting up the sampling sites, namely Envirotech Instruments Pvt. Ltd, Okhla, II Floor Roof (Industrial Area); Indian Institute of Technology, Delhi, V Block IV Floor Roof (Residential Area); Vasant Vihar Bus Depot II Floor Roof (Commercial Area); Shakarpur a Residence, III Floor Roof (Residential Area). More details and figures of sampling locations can be obtained from Shandilya et al., (2007.
Out of the four sites chosen, two were in residential areas, one each in industrial and commercial areas. The criterion that was taken into account was the presence of local sources. The Shakarpur site was of particular importance, since the results obtained were used to compare the SPM level in South Delhi. Simultaneously, when we analyzed the respirable suspended particulates in respect of heavy metals, it was interesting to know the concentration near prominent sources like the Indraprastha power station. The site was located just on the bank of the Yamuna, when on the other bank was the Indraprastha power station. But the site www.intechopen.com location was mainly a residential area. Except for Shakarpur, where the Indraprastha power station is situated nearby, the sources of the local emissions were considered.
SPM was collected on a filter paper weighed before and after exposure. It was dried in an oven at 80 o C for 45 minutes before each weighing. The size classification of PM was achieved through a cyclone installed in a Hi-Vol sampler, which separates the respirable PM 10 and non-respirable fraction in the case of the Hi-Vol sampler. The particles sized <10μm were collected on the filter, and >10μm were collected in a separate sampling cup, also known as a dust collector. Flow rates were monitored before and after every sampling period with manometer, which was factory calibrated with an accuracy of 1.3% full scale. The flow rate checks were made before and after each sampling event to ensure that filter holders were not leaking and to determine that fitter clogging had not occurred.
These gases were monitored with the help of APM411 assembly attached with APM460DX (Envirotech Instruments Pvt. Ltd.) using the impingers. The modified Jacob-Hochheiser method (or arsenate method) have determined nitrogen dioxide (IS-5182, 1975). For the 24 hour field sampling, it is preferable to use this method. The West-Gaeke spectrophotometric method is the standard method for SO 2 monitoring of 0.0005-5 ppm SO 2 in the ambient air.

Air quality index determination
Based on the monitoring data of various pollutants at different sites on different days, an index called "Air Quality Index" was calculated using the following equation, thereby assessing the present air quality status of Delhi. As per the EPA guidelines, pollutants such as O 3 , CO, SO 2 , NO 2 , PM 10 , and PM 2.5 are used for calculating the AQI for an area. In the present study, the pollutants used for calculating the AQI were NO 2 , SO 2 , and RSP. AQI was calculated for individual pollutants simultaneously at a place, and the maximum value corresponding to one particular pollutant was reported as the AQI of that area. The same procedure was adopted at all sites on each day, with AQI finally reported along with the responsible pollutants.

Results and discussion
For all sites, we have calculated the mean, standard deviation, and maximum, and we also calculated the geometric means. The geometric mean is the basis of preparing the air quality standards. The mean, standard deviation, maximum value, and geometric mean values are listed as below:

General discussion
In general, for PM 2.5 the secondary formation is more relevant than the primary emissions, but Wilson and Suh (1997) stated that some natural particles also occur in size range <2.5µm. A typical study (Pekkanen et al., 1997) reports that a considerable fraction of the road dust consisted of particles<2.5μm in diameter. The primary particles are emitted by a source as particles and are dispersed in the atmosphere without any major chemical transformation (Shandilya and Kumar, 2010a, 2010b. PM 10 mainly consisted of three distinct components: Primary vehicular emissions, secondary pollutants, and windblown dust. Specifically, major contributors to excessive PM 10 levels were combinations of area-wide emitters such as fugitive dust from roads, construction, and www.intechopen.com agriculture; smoke from residential wood combustion or prescribed burning; directly emitted exhaust from motor vehicle engines; and secondary sulfates, nitrates, and organics formed from gaseous ammonia, sulfur dioxide, nitrogen oxides, and reactive organic gas emissions (Sharma et al., 2004. The emissions inventories for these "nontraditional" types of sources were found to be much less accurate than those obtained for "traditional" ducted emissions from the industrial sources. The coarse fraction was mainly composed of the re-suspended dust. Farming operations (e.g., agriculture tillage, harvesting, and travel on unpaved roads) account for large portions of the primary PM 10 in many emissions inventories. PM 10 non-attainment areas are suspected of having RWC as a major contributor. The prescribed burning of agricultural fields and forest slash is suspected of being a major contributor to PM 10 in other areas. PM 10 is considered more a local and urban scale problem, whereas PM 2.5 has a much longer residence time and is considered a regional or trans-boundary problem. The spatial and temporal distribution of fine particle levels may vary substantially.
Natural background levels, excluding all anthropogenic sources in the US and elsewhere, were 4-11-µgm -3 and 1-5-µgm -3 for PM 10 and PM 2.5 , respectively. PM 10 mean concentrations were typically from 10-80-µgm -3 around the world. 10-20-µgm -3 is found at remote sites with no local sources (i.e., even in clean air). In urban areas, 60-220-μg/m 3 is typical, and in a heavily polluted area, levels may approach 2000-μg/m 3 . More than 50-µgm -3 was found at heavily polluted sites with high traffic density and/or local sources in the surroundings, where heavily polluted urban areas may reach 100-µgm -3 on average. In the heavily polluted cities of southern Europe, Latin America, and Asia, average PM 10 concentrations have reached 100µgm -3 , and more. PM 2.5 mean concentrations were between 7-80-µgm -3 . PM 10 trends were decreasing in the last decade, for emission and ambient levels, in the developed and developing world. The ratio of PM 2.5 to PM 10 at the urban sites usually averaged 0.55-0.6.
Many studies indicated that TSP and PM 10 concentration in the ambient air would be affected by various meteorological factors such as wind speed, wind direction, solar radiation, relative humidity, rainfall, and source conditions. Apart from the emission factors of the vehicles, SPM concentrations on the street level would be mainly affected by the mechanical turbulence (created by moving vehicles and wind) and by thermal turbulence produced by the hot vehicle exhaust gas (Shandilya and Kumar, 2009) .
24-hour average concentrations of the FSP, RSP, and TSP were calculated (See Methodology for a sample calculation) from samples collected at four stations at Southern Delhi and East Delhi and from samples taken from Satna. Our study is also not limited by the fact that, during winter, mixing depth in the early morning is lower than in the summer, thereby concentrating primary crustal aerosol emissions (e.g., from road dust). Since our study did not take place in the summer or winter, this condition is not applicable and the purpose of our study is not forfeited.

Comparison of SPM level in Delhi
The sources prevalent at VV are Diesel Buses, CNG Buses, Traffic, and Mechanical Disturbances, while at IIT the sources are Traffic, Vegetative Burning, construction work, and residential sources like cooking. At Okhla, the sources are Vegetative Burning, Diesel Generators, Mechanical Disturbances, Traffic, and Industrial Activities. Concentrations of www.intechopen.com fine particles are close to main roads. Road transport and diesel exhaust are major sources of fine particles in urban areas Kumar, 2009, Kumar et al., 2011). PM 10 and PM 2.5 concentrations are elevated close to main roads compared to background sites. More monitoring details can be found in Shandilya et al (2006.

Variation of PM 2.5
With a minimum of one daily sample at a site, 24-hour PM 2.5 concentration at the urbanindustrial Delhi site varies from 85.1-157.3µgm -3 . 24-hour PM 2.5 concentration at the urbancommercial Delhi site varies from 78.6-198.1µgm -3 , and 24-hour PM 2.5 concentration at the urban-residential Delhi sites varies from 84. 6-198.4µgm -3 . The value at Delhi is well above 24-hour PM 2.5 USNAAQS (65µgm -3 ). More details of this monitoring can be found in . Variation of PM 2.5 value is highest on SKP, which suggests that the contribution from different sources varies to a great extent. The Standard Deviation values suggest that PM 2.5 source emission rate at VV varies greater than the emission rate at any place in South Delhi. The large variation is most likely due to the differences in air mass, emission characteristics, and meteorology.

Peak day composition
The maximum PM 2.5 concentration in Delhi, 198.4-μg/m 3 , was at SKP on November 26, 2000, which may be due to a wash out caused by the rain. The lowest PM 2.5 value of 78.6μg/m 3 was observed at VV bus depot (November, 24, 2000). When there was a strike and there was less traffic on the road, there was a significant decrease in PM 2.5 fraction.

Variation of PM 10
With a minimum of one daily sample at a site, 24-hour PM 10 concentration at the urbanindustrial Delhi site varies from 205.0-320.3-µgm -3 . 24-hour PM 10 concentration at the urbancommercial Delhi site varies from 210.0-322.6-µgm -3 , and 24-hour PM 10 concentration at the urban-residential Delhi sites varies from 174.5-321.2-µgm -3 . The PM 10 concentration value was always greater than the 24-hour NAAQS (150-μg/m 3 in industrial and commercial areas and 100-μg/m 3 in residential areas).

PM 2.5 /PM 10 ratio
The most common ratio of PM 2.5 /PM 10 is 0.60, but the prevalence of the local sources, with either a dominating fine or coarse fraction, may shift this ratio. PM 2.5 /PM 10 concentration at the urban-industrial Delhi site varies from 0.27-0.56 with an average±standard deviation of 0.47±0.11. PM 2.5 /RSP concentration at the urban-commercial Delhi site varies from 0.25-0.94 with an average±standard deviation of 0.59±0.26. PM 2.5 /RSP concentration at the urbanresidential South Delhi sites varies from 0.43-0.91 with an average±standard deviation of 0.69±0.20. PM 2.5 /RSP concentration at the urban-residential East Delhi sites varies from 0.62-0.95 with an average±standard deviation of 0.77±0.18. This shows that at Okhla, PM 2.5 contribution to PM 10 is 47%, which increases to 59% at VV, further increases to 69% at IIT and achieves its maximum (77%) at SKP. PM 2.5 /PM 10 ratio standard deviation is highest (0.26) at the VV site, suggesting the heterogeneous contribution of secondary particle formation sources. It has been observed that a part of PM 10 is PM 2.5 but not as much as in the case of IIT in South Delhi. The standard deviation (PM 2.5 /RSP) low value at Okhla (0.11) shows that the contribution of PM 2.5 /RSP is more consistent compared to other sites, and PM 2.5 sources contributing to RSP remains the same. The ratios of PM 2.5 /PM 10 are generally between 20 and 30%, with abnormal values from industrial particulate (both high-50% in Cynon-and low-17% in Port Talbot). The highest contribution of PM 2.5 /PM 10 strongly shows that industrial sources are predominant in the SKP area. Comparison with higher ratios generally found in large urban areas (Birmingham-QUARG report) indicates that the finer traffic particulate stick together or deposit on surfaces less rapidly than the coarse fraction of PM 10 , so they tend to build up in larger urban areas or still conditions. It has been observed that a part of PM 10 is PM 2.5 , but not as much as in the case of SKP. It shows that the condensation products of the chemical reactions primarily contribute PM 10 fraction that originated in this area. Data (PM 2.5 /PM 10 ) shows that PM 10 fraction that originated in industrial areas contains the condensation products of the chemical reactions and some primary particulate.

Peak day composition
The maximum PM 10 concentration in Delhi, South Delhi and VV 322.6-μg/m 3 was on November 17, 2000 during our study; the lowest PM 10 of 174.5-μg/m 3 was observed at Shakarpur. This can be due to some change of wind pattern since there is a reduction in PM 2.5 concentration (171.8-μg/m 3 ), which eradicates the possibility of contribution from secondary particles and is further confirmed by an increase in NRSP fraction. Maximum PM 10 concentration at Okhla, 320.3-μg/m 3 , was on October 29, 2000 next to Deepawali. This high concentration can be attributed to crackers; this is well confirmed by the reduction of PM 2.5 concentration (85.1-μg/m 3 ), which is primarily caused by vehicles and is further confirmed by decline in the contribution of PM 2.5 to PM 10 suggested by the value of PM 2.5 /PM 10 (0.27, second lowest in all). Maximum PM 10 concentration at IIT, 274.7-μg/m 3 , occurred on November 3, 2000, the day when a celebration was held, so there was a great rush of vehicles in IIT. The increase in PM 10 was due to the resuspension of dust and vehicle exhaust. It was further confirmed by the increase in NRSP fraction on that day and an increase in PM 2.5 . The PM 2.5 value on the second day remained on the same level (168.7-μgm -3 ), showing that the residence time PM 2.5 was much higher, which was not the case with PM 10 and NRSP. Maximum PM 10 concentration at SKP, 321.2-μg/m 3 , took place on November 26, 2000, which may have been due to the wash out caused by the rain.

Variation of TSP
With a minimum of one daily sample at a site, 24-hour TSP concentration at the urbanindustrial Delhi site varies from 553.4-827.6-µgm -3 . 24-hr TSP concentration at the urbancommercial Delhi site varies from 453.2-717.2-µgm -3 and 24-hr TSP concentration at the urban-residential Delhi sites varies from 345.2-601.3-µgm -3 . The TSP concentration value was always greater than the 24-hr NAAQS of TSP i.e., 200-μg/m 3 in the residential area. The TSP concentration value was always greater than the NAAQS of TSP i.e., 500-μg/m 3 in the industrial and commercial area while it crossed the limit (NAAQS, 500-μg/m 3 ) only once on November 17, 2000 at the VV commercial area.

RSP/TSP ratio
RSP/TSP at the urban-industrial Delhi site varies from 0.27-0.56 with an average±standard deviation of 0.39±0.11. RSP/TSP concentration at the urban-commercial Delhi site varies from 0.42-0.64µgm -3 with an average±standard deviation of 0.50±0.09. RSP/TSP concentration at the urban-residential South Delhi sites varies from 0.37-0.46 with an average±standard deviation of 0.41±0.04. RSP/TSP concentration at the urban-residential East Delhi sites varies from 0.40-0.66 with an average±standard deviation of 0.53±0.11. This shows that at Okhla, RSP contribution to TSP is 39%, increasing to 50% at VV, decreasing to 41% at IIT, and again achieving its highest of 53% at SKP. The standard deviation (RSP/TSP) low value at IIT (0.04) shows that the contribution of RSP to TSP is consistent, while it varies equally at SKP and at Okhla. Therefore, the RSP emitting sources in the SKP area were contributing equally throughout the study, while these sources' emission pattern was changed at the other places.

Peak day composition
The Our results could be compared with the results of Kanpur, a study carried out by Sharma et al. (2003). They found at the background site, IIT Kanpur that PM 10 levels were 62±11.22μg/m 3 and at Agricultural University, the residential area, PM 10 levels were 71.46-335.83µg/m 3 . At the commercial area Naveen Market, PM 10 levels were 114.22-634.68-μg/m 3 , and at the industrial area, Lajpat Nagar, PM 10 levels were 221.77-650.73-µg/m 3 . On comparison, it can be concluded that PM 10 levels in all areas (industrial, commercial, residential) found the same range at both places (Delhi and Kanpur).
Our results could also be compared with the results of Jaipur, a study carried out by Das et al. (1998). They found at the industrial site that TSP levels were 384.61-μg/m 3 and, in another study by Gajraj et al. (submitted) in the Vishwakarma industrial area, mean PM 10 concentration during May 1999 was 242-μg/m 3 ; during June 1999, the mean concentration was PM 10 (208.5-μg/m 3 ). Our TSP levels at Okhla are certainly higher than TSP levels in Jaipur, as obtained by Das et al. (1998).

Explanation of spatial variation
The terminal settling velocity is an important factor for the spatial variation of SPM (Willeke & Baron, 1993). For very fine particles, such as a particle of 0.1-µm d ae setting velocity, is 8.65x10 -5 -cms -1 for a particle of 1-µm 3.48x10 -3 and for a particle of 10-µm in size 3.06x10 -1cms -1 . Primary particles (Aitken particles <0.1-µm) and large particles (e.g.>10-µm) are expected to have a bigger spatial variability than the particles in the accumulation mode (0.1-1-µm). Coagulation (for very fine particles <0.01-µm) and gravitational settling (for particles>1-µm) are the underlying mechanisms, which cause spatial heterogeneity. The distribution of PM 2.5 might be uniform in a situation where the secondary formation is important; in cities, however, with large emissions from heavy duties (diesel exhaust), PM 2.5 can also exhibit significant spatial variation. Secondary aerosols that contribute to the accumulation mode (0.1-2-µm) show quite homogeneous spatial distribution. In general, small-scale spatial variations of PM 2.5 were described to be smaller than the spatial variations of PM 10 (Tuch et al., 1997). The spatial variation was small for PM 2.5 but larger for PM 10 . Our results are in agreement with Monn (2001). With respect to the small-scale spatial variation in the urban areas, the largest variations occurred in the ultrafine (0.1-µm) and coarse mode (PM 10-2.5 ) (Burton et al., 1996). Ultrafine particle number counts have large spatial variations, and they are not well correlated to mass data, so spatial variation is much more reflected in PM 10 than in PM 2.5 .
The spatial variation was small for PM 2.5 but larger for PM 10 . Spatial correlations for PM 2.5 are found to be lower than 0.9 and around 0.6 for PM 10 . The contribution of coarse particle fraction (PM 10-2.5 ) on PM 10 can be important. Conclusively, we can say that SPM is complex with respect to its sources in Delhi. With the exception of the geological material, PM 2.5 source contribution estimates are not similar to those for PM 10 . A deep insight in the source contribution could be obtained if we consider emission factors of SPM from different sources.

Oxides of sulfur and nitrogen
In order to understand the sources, behavior, and mechanism of particle formation in the atmosphere, it is important to measure the precursors like SOx and NOx; this primarily contributes to the formation of the secondary particulate, which again constitutes a big fraction of total particulate matter in the atmosphere. This type of precursor characterization for the continental, marine background, and urban influenced aerosols has been reported from various sites throughout the world

Sources of sulfur oxides and nitrogen oxides
The natural sources of nitrogen oxides are the geothermal activity and bacterial action. Natural sources include the soil microbes, vegetation, biomass, slum, burning, and lightning. Natural sources include the biological emissions (e.g. vegetation and soil microbes) and abiotic processes (e.g., geogenic, lightning, and biomass burning). Globally, lightning strikes occur about 100-times/second (Borucki and Chameides, 1984). Lighting strikes are characterized by a highly ionized and high temperature channel, in which molecular nitrogen is dissociated and reacts with oxygen to form NO (Guenther et al, 2000).
The primary sources of NOx are the motor vehicles, electric utilities, and other industrial, commercial, and residential sources that burn fuels. Other sources include the chemical and nitric acid manufacture, the detonation of nitrate-containing explosives, and electric sources. Man made sources of nitrogen oxide include the automobile, railways (coal fired and diesel locos), airways (landing and takeoff operations), industrial processes like chemicals (nitric acid plants, processes where nitrogenous compounds are used), metal processing, fertilizer (nitrogenous fertilizers), and other miscellaneous activity like domestic combustion, forest fires and managed fires, industrial boilers, furnaces, and commercial combustion. Although most NO is produced by the natural bacterial action, it must be emphasized that the rate of emission from man-related sources has, by estimate, increased by one-fourth between 1966 and 1968. Certainly, man's greatest contribution to the family of nitrogen oxides arises from the combustion of fuels. The contribution to the total NOx can be categorized as Utilities 27%; Motor Vehicles 49%; Other Sources 5%; and Industrial/ Commercial/Residential 19%. Nearly 2/3 of NOx emissions originate at the Earth's surface from fossil fuel and biomass combustion. Vehicle exhausts are known to increase the inputs of NO 2 to the atmosphere (see, e.g., Hargreaves et al., 2000). The burning of natural gas (which produces negligible SOx, CO, and HC) does produce a very significant quantity of oxides of nitrogen.
Man caused nitrogen emissions as equivalent NO 2 are estimated as 53x10 6 TPA over the world, 95% of which are emitted in the Northern Hemisphere. Half of these are from the combustion of coal. Natural sources of NOx are estimated as between 15 and 17-times pollutant emissions. (Sawyer et al, 1998). The major source of NOx is the combustion when the fixation of atmospheric N occurs at the high flame temperature. The oxides are emitted mainly as NO, which is normally rapidly oxidized to NO 2 by atmospheric O 2 and O 3 . Motor vehicle exhaust contributes a sizable fraction of the total emissions of NOx. As well as the stationary combustion sources, the manufacture of HNO 3 and nitrate fertilizer is the source of NO & NO 2 . On a global basis, natural emissions of oxides of nitrogen (such as releases from soil and the ocean, lightning strikes, oxidation of ammonia, and stratospheric oxidation) all make substantial contributions to the total emissions. Garg et al (2001) has reported the sectoral shares of NOx emissions for India in 1995. According to them, a major part of NOx emissions is contributed from 28% by the nonenergy sources, 28% by electric power generation, 2% by road transport, 19% by Biomass burning, 4% by other transports, and 19% by industries. Delhi tops the table of top 10 NOx emitting districts in the terms of total emissions by emitting total NOx 66-Gg in 1990 and 82-Gg in 1995; in the terms of emissions per unit area, NOx/area (tons km -2 ) was 44 in 1990 and 55 in 1995. The largest NOx emission chunk in Delhi is from the oil sources.
The major atmospheric man made sources of sulphur dioxide (SO 2 ) are the burning of fossil fuel in stationary combustion and industrial process, viz. petroleum, chemical, metallurgical and mineral industries, domestic emission and mineral industries, domestic emission, and fires, while the natural source are the volcanoes (wherever these are present, but in India there are none), geothermal activity, the bacterial decomposition of organic matter, forest fires, and managed fires. Motor vehicles are a relatively minor source of SO 2 , since refined motor fuel normally has low sulfur content.
Ambient SO 2 results largely from the stationary sources (that burn the coal and oil, refineries, pulp, and paper mills) and from the nonferrous metal smelters. The sources of the pollution include the emissions from the on-road vehicles, non-road vehicles (like planes, ships, trains and industries), and small businesses and households where polluting products are used.
Nitrogen oxides (NOx) are produced when fossil fuels are burned in the motor vehicles, power plants, furnaces, and turbines. Other gasoline powered mobile sources (motorcycles, recreational vehicles, lawn, garden, and utility equipment) have high emissions on per quantity of the fuel consumed basis, but their contribution to the total emissions is small. Heavy-duty diesel vehicles are the dominant mobile source of the Oxides of nitrogen. Oxides of nitrogen emissions on a fuel-consumed basis are much greater from the diesel mobile sources than from the gas.
Another common source of SO 2 in the atmosphere is the metallurgical operations. Many ores, like Zn, Cu, and Pb are primarily sulfides. During the smelting of these ores, SO 2 is evolved in the stack concentrations of 5-10% (SO 2 ). Among miscellaneous operations releasing SO 2 into the atmosphere are the sulphuric acid plant, paper-manufacturing plants, open burning of the refuse, and municipal incinerators which contribute some amount of SO 2 to the atmosphere (Peavy et al, 1985).
In the entire sulphur entering atmosphere, 1/3 rd is by the anthropogenic activities, and 2/3 rd come from natural sources such as Hydrogen Sulfide (H2S) or SOx. The problem with the manmade sources is the distribution, rather than the amount. Delhi, in this instance, is being recognized as a city with a potentially high traffic load on its roads. Therefore, the combustion of fossil fuels like diesel and petrol can be considered as a major source of the SO 2 in Delhi. Besides this thermal power plant, which is located in the Southeast about 1-1.5 km far from the SKP station, it is also contributing SO 2 in the atmosphere because the coal is used as a main fuel for the heat and the power generation. However, SO 2 concentration in the air also depends upon the sulfur content of fuels that vary from less than 1% of good quality anthracite to over 4% for bituminous coal.
Most crude petroleum products contain less than 1% sulfur; a few contain up to 5%. In the coming years, Delhi is supposed to have less concentration of SO 2 in its atmosphere because the fuel with less S content has already been introduced into the market. Another common source of the sulfur dioxide in Delhi is metallurgical operation. Among the miscellaneous operations releasing releasing SO 2 into the atmosphere are sulfuric plants and paper manufacturing plants. The open burning of refuse and municipal incinerator also contribute some amount of SO 2 . Garg et al (2001) also reported the sectoral shares of SO 2 emissions for India in 1995. According to them, a major part of SO 2 emissions is contributed to 46% by non-energy sources, 36% by electric power generation, 8% by road transport, 6% by biomass burning, 3% by other transports, and 1% by industries. The details of industrial SO 2 emissions shows that 40%, 27%, 14%, 10%, 7%, and 2% are shared by non-energy industries, steel industry, fertilizer, cement, other industries, and refineries, respectively.
In Delhi, SO 2 /area (tkm -2 ) in 1990 was 47.9, in 1995 it was 46.7, and the average annual concentrations (μg/m 3 ) were 16 in 1990 and 24 in 1995. The largest contribution to SO 2 levels is by the transport sector. Our study results should be considered by keeping the points of the study done by Kamyotra et al (2000) and Jain (2001). These studies have pointed out two important points: by increasing the sampling period, the concentration shown becomes less (Kamyotra et al., 2000) and the diurnal pattern over 24 hours changes with time, local conditions, and activities (Jain, 2001).
The nighttime high values can be attributed to the phenomenon of the inversion. According to Kamyotra et al (2000), the lower values found during the day can be ascribed for the evaporation of the liquid, but this evaporation is absent during the night time which, in turn, increases the night time values, so high values obtained can be attributed to the combined effect of these facts.
On Okhla, the highest day SO 2 concentration variation (1.89-μgm -3 ) suggests that the source emission rate variation at Okhla is higher comparatively. On the VV site, day SO 2 concentration variation was lowest in all (0.64-μgm -3 ), suggesting that the source diversity and source emission rate variation are lower comparatively. On the IIT site, day SO 2 concentration variation is 0.83-μgm -3 , and on the SKP site day SO 2 concentration variation is 1.07-μgm -3 . In all of Delhi, day SO 2 concentration varies from 3.21-8.42-μgm -3 .
On SKP, Night SO 2 concentration variation was the highest in all (5.26-μgm -3 ), suggesting that the source emission rate variation at SKP is higher comparatively. On the IIT site, night-SO 2 concentration variation was the lowest in all (1.21-μgm -3 ), suggesting that the source diversity and source emission rate variation are lower comparatively. On the VV site, night SO 2 concentration variation is 4.38-μgm -3 , and on the Okhla site night SO 2 concentration variation is 2.32-μgm -3 . In all of Delhi, night SO 2 concentration varies from 12.96-36.84-μgm -3 . On SKP, 24-hour SO 2 concentration variation was highest in all (3.13-μgm -3 ), suggesting that the source emission rate variation at SKP is higher comparatively. On the IIT site, 24hour SO 2 concentration variation was the lowest in all (0.81-μgm -3 ), suggesting that the source diversity and source emission rate variation are lower comparatively. On the VV site, www.intechopen.com 24-hour SO 2 concentration variation is 2.15-μgm -3 , and on the Okhla site 24-hour SO 2 concentration variation is 1.76-μgm -3 . In all of Delhi, 24-hour SO 2 concentration varies from 8.45-21.70-μgm -3 .

Peak day composition
The

Variation in NO 2
With a minimum of one 12-hour daily sample at a site, 12-hour per day NO 2 concentration at the urban-industrial Delhi site varies from 28. On VV, day NO 2 concentration variation was highest in all (5.53-μgm -3 ), which suggests that the source emission rate variation at VV is higher comparatively. On the Okhla site, day NO 2 concentration variation was the lowest in all (2.73-μgm -3 ), which suggests that the source diversity and source emission rate variation is lower comparatively. On the IIT site, day-NO 2 concentration variation is 3.06-μgm -3 , and on the SKP site, the day NO 2 concentration variation is 3.47-μgm -3 . In all of Delhi, the day NO 2 concentration varies from 21.00-46.92-μgm -3 .
On the VV site, night-NO 2 concentration variation was the highest in all (24.71-µg/m 3 ), which suggests that the source emission rate variation at SKP is higher comparatively. On the IIT site, night NO 2 concentration variation was the lowest in all (8.67-μgm -3 ), which suggests that the source diversity and source emission rate variation is lower comparatively. On the Okhla site night NO 2 concentration variation is 24.20-μgm -3 , and on the SKP site night NO 2 concentration variation is 19.36-μgm -3 . In all of Delhi, night NO 2 concentration varies from 47.13-162.02-μgm -3 .
On Okhla, 24-hour NO 2 concentration variation was highest in all (11.68-μgm -3 ), which suggests that the source emission rate variation at Okhla is higher comparatively. On the IIT site, 24-hour NO 2 concentration variation was the lowest in all (5.67-μgm -3 ), which suggests that the source diversity and source emission rate variation are lower comparatively. On the VV site, 24-hour NO 2 concentration variation is 11.07-μgm -3 , and on the SKP site, 24-hour NO 2 concentration variation is 11.10-μgm -3 . In all of Delhi, 24-hour NO 2 concentration varies from 35.12-104.47-μgm -3 .

Peak day composition
The As evident from the table 2, the concentration of NOx was the highest at SKP, followed by Okhla, VV, and IIT. The trend is attributed to the prevailing activities at the sites. 24-hour average NOx concentration is mainly contributed by the nighttime values. These nighttime higher NOx concentrations at the different sites also reflect the local source conditions.

SOx to NOx day ratio
The day ratio at an urban-industrial Delhi site varies from 0.11-0.28 and at the urbancommercial Delhi site varies from 0.11-0.16; at two urban-residential Delhi sites, it varies from 0.11-0.22. At SKP, variation of the ratio (0.01) is lowest, while it is highest at the Okhla (0.06). The variation of ratio at VV was 0.02, and at IIT it was 0.03, and in Delhi the ratio ranged from 0.11-0.28. The variation in Delhi (4%) shows that the sources liberating NOx are also releasing SOx. Summarizing our observations at Okhla in South Delhi showed that the day ratio was 0.18±0.06. IIT in South Delhi showed that the ratio was 0.18±0.03. VV in South Delhi showed that the ratio was 0.14±0.02. SKP in East Delhi showed that the ratio was 0.12±0.01, and overall the Delhi ratio was 0.16±0.04.

SOx to NOx night ratio
The night ratio at the urban-industrial Delhi site varies from 0.19-0.42, and at the urbancommercial Delhi site varies from 0.22-0.38, and at two urban-residential Delhi sites varies from 0.19-0.29. At SKP, the variation of the ratio (0.02) is the lowest, while it remains the highest at Okhla (0.08). The variation of ratio at VV was 0.06, and at IIT it was 0.03, and in Delhi the ratio ranges from 0.19-0.42. The variation in Delhi (6%) shows that sources liberating NOx are also releasing SOx. Summarizing our observations at Okhla in South Delhi showed that the night ratio was 0.24±0.08. IIT in South Delhi showed that the ratio was 0.25±0.03. VV in South Delhi showed that the ratio was 0.27±0.06. SKP in East Delhi showed that the ratio was 0.22±0.02, and overall in Delhi the ratio was 0.25±0.06.

Sox to NOx 24-hr ratio
The 24-hour ratio at the urban-industrial Delhi site varies from 0.18-0.35, and at the urbancommercial Delhi site varies from 0.20-0.30, and at two urban-residential Delhi sites varies from 0.17-0.26. The SKP and IIT variation of the ratio (0.02) is the lowest, while the highest is at Okhla (0.06). The variation of ratio at VV was 0.04, and overall in Delhi the ratio ranges from 0.17-0.35. The variation in Delhi (4%) shows that the sources liberating NOx are also releasing SOx.
Summarizing our observations at Okhla in South Delhi showed that the 24-hour ratio was 0.22±0.06. IIT in South Delhi showed that the ratio was 0.23±0.02. VV in South Delhi showed that the ratio was 0.24±0.04. SKP in East Delhi showed that the ratio was 0.20±0.02, and overall in Delhi the ratio was 0.22±0.04.
Many studies indicated that NOx and SOx concentration in ambient air would be affected by the various meteorological factors such as the wind speed, wind direction, solar radiation, relative humidity, rainfall, and source conditions. Our results could also be compared with the values obtained by the CPCB personnel at ITO during the study. These were taken from their site (http://www.envfor.nic.in/). Our values are in good agreement with the CPCB values. During comparison, it was kept in mind that these values are of those samples, which were taken for eight-hour duration as per CPCB rules. But these values are normalized for 24-hours to represent the daily mean value (as per CPCB personnel).
The SOx levels were always lesser than the 24-hour NAAQS of SOx, i.e., 80-μg/m 3 in the residential area and 120-μg/m 3 in the industrial area. The NOx levels were lesser than 24www.intechopen.com

Air quality index
In order to assess the degree of air pollution in Delhi, an index called air quality index (AQI) has been used as recommended by USEPA (EPA-454/R-99-010). More details are given in the literature review section, and the calculation method adopted is given in the methodology section. Calculated AQI has been presented in the table below. It can be concluded that PM is the critical parameter for the deterioration of ambient air quality in Delhi.

Comparison with other studies
In the industrial area of Jaipur, Gajraj et al (2000) found 24 observations of RSPM, where the air quality was unhealthy for 14 days and unhealthy for sensitive groups for the remaining days. For 32 observations of RSPM in Jaipur, Jain (2001) found that the air quality was good on six days, moderate on 20 days, and unhealthy for sensitive people on six days. She also found for 24 observations of SO2 that the air quality was good on all days.

Conclusions
The PM 2.5 value at Delhi is well above 24-hour PM 2.5 USNAAQS (65-µgm -3 ). The variation of PM 2.5 value is highest on SKP, which suggests that the contribution from different sources is varying to a great extent. The PM 10 concentration value was always greater than the 24-hour NAAQS (150-μg/m 3 in industrial and commercial areas and 100-μg/m 3 in residential areas).
The study shows that, at Okhla, PM 2.5 contribution to PM 10 is 47%; it increases to 59% at VV, further increases to 69% at IIT, and achieves its maximum (77%) at SKP. The highest contribution of PM 2.5 /PM 10 strongly shows that industrial sources are predominant in the SKP area. This shows that at Okhla RSP contribution to TSP is 39%; it increases to 50% at VV and decreases to 41% at IIT and again achieves its highest of 53% at SKP. The above discussion put forward that the secondary formation and diesel exhaust, which are prime important mechanisms of the formation of PM 2.5 , are predominant at IIT but the primary particulate sources also seem like prime important reasons for the formation of RSP. In all of Delhi, 24hour SO 2 concentration varies from 8.45-21.70-μgm -3 and 24-hour NO 2 concentration varies from 35.12-104.47-μgm -3 . The variation in Delhi shows that sources liberating NOx are also releasing SOx. The SOx levels were always lesser than the 24-hour NAAQS of SOx i.e. 80μg/m 3 in the residential area and 120-μg/m 3 in the industrial area. The NOx levels were lesser than the 24-hour NAAQS of NOx i.e. 80-μg/m 3 in the residential area and 120-μg/m 3 in the industrial area. The SO 2 and NO x data do not show any weekend or holiday effect. The variations observed are most likely due to differences in the air mass, emission characteristics, and the meteorology. The category of air quality based on AQI for SOx and NOx at all sites was always good for the public. The category of air quality based on AQI for PM at all sites was always unhealthy for the public. It can be concluded that PM is the critical parameter for the deterioration of ambient air quality in Delhi.

Acknowledgement
The authors are thankful to Envirotech Instruments Private Limited, Delhi for their sponsorship of the study. Thanks were due to all the persons, organizations, institutions, and companies who provided a secure location for the sampling equipment used in this www.intechopen.com