Rural versus urban gaseous inorganic reactive nitrogen in the Indo-Gangetic plains (IGP) of India

The present study reports on the abundance of reactive nitrogen (NH3 and NO2) at two sites, i.e. Okhla (urban site) in Delhi and Mai (rural site), located in the nearby state: Uttar Pradesh. The measurements were carried out during the period from October, 2012 to September, 2013 on a monthly basis. The average concentrations of NH3 at Okhla and Mai have been recorded as 40.4 ± 16.8 and 51.57 ± 22.8 μg m−3, respectively. The average concentrations of NO2 have been recorded as 24.4 ± 13.5 and 18.8 ± 12.6 μg m−3 at Okhla and Mai, respectively. Results show that the seasonal variation at Mai was more prominent where NH3 concentrations varied at 72.0 μg m−3 during the winter, 47.2 μg m−3 during the summer and 30.7 μg m−3 during the monsoon season, whereas at Okhla the average NH3 concentrations were almost equal during different seasons, namely 44.2 μg m−3 during the winter, 42.5 μg m−3 during the summer and 38.9 μg m−3 during the monsoon season. This is probably due to significant differences in crops and in the fertilizer amounts applied across the seasons in rural areas, while urban areas have almost constant sources throughout the year. Winter concentrations were highest at both sites, followed by summer and then the monsoon season. The average NO2 concentrations were recorded as 39.6 μg m−3, 24.5 μg m−3and 10.4 μg m−3 during the winter, summer and monsoon season at Okhla, whereas the average NO2 concentrations were recorded as 27.5 μg m−3, 17.2 μg m−3 and 4.1 μg m−3 during the winter, summer and monsoon season, respectively. NO2 emissions at Okhla may be attributed to various urban activities, such as vehicular traffic and industries, while NO2 emissions at Mai may be attributed to biomass burning as a major source. However, NO2 concentrations from vehicular traffic and nearby industries cannot be ignored at Mai.


Introduction
Nitrogen (N) is an important constituent of human life and the earth system. N 2 gas constitutes 78.02% of the atmosphere. Like the saying, 'water, water everywhere, but not a drop to drink,' nitrogen surrounds us everywhere, but it cannot be directly utilized by plants and animals due to its inert nature. Plants and other organisms can utilize it only after its conversion to reactive nitrogen (Nr), which includes all chemically, biologically, photochemically and radioactively active compounds in the environment (Galloway et al 1995). Most of atmospheric N is contributed by N 2 , with a minor contribution from N 2 O (Gradel and Crutzen 1993), whereas traces of N are contributed by other N species such as NH 3 and NO x , etc. In the atmosphere, Nr is mainly released as NH 3 , NO x , with small quantities of other gases such as HONO or organic N (Hertel et al 2012).

Environmental Research Letters
Environ. Res. Lett. 9 (2014) 125004 (9pp) doi: 10.1088/1748-9326/9/12/125004 Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Due to anthropogenic activities, the natural biogeochemical cycle of nitrogen has been perturbed. Activities such as fertilizer application, animal husbandry, fossil fuel combustion, etc have contributed a huge amount of Nr species into the atmosphere (Galloway et al 2008), resulting in their increased depositions onto the earth's surface. According to research reports, elevated N deposition may lead to environmental problems such as eutrophication (Vitousek et al 1997, Liu et al 2011, biodiversity loss (Phoenix et al 2006, Clark and Tilman 2008, Song et al 2011, Li et al 2012a, acidification of soil (Guo et al 2010) and increased N 2 O emissions (Li et al 2012b). Nr species have an adverse impact on air quality (Chan and Yao 2008, Kulshrestha et al 2009, Shen et al 2011, Aber et al 1989, Bobbink et al 1998, Fangmeier et al 1994 and plants (Liu et al 2011, Sutton et al 2011. According to Galloway et al (1995), Nr is contributed by human activities at a rate of 140 Tg N per year globally in which 20 Tg, 80 Tg and 40 Tg annually are contributed by energy production, fertilizer production and the cultivation of legumes and other crops, respectively. An approximate 55-60% fraction of the total anthropogenically fixed N is redistributed back to the atmosphere as reduced N (NH y , NH 3 + NH 4 ) and oxidized N (NO x ) (Galloway et al 1995). The total (wet + dry) deposition of inorganic N in North America and Europe ranges at about 3-32 75 kg N ha −1 year −1 and 1-75 kg N ha −1 year −1 , respectively (Nadelhoffer 2001). These concentration ranges are considered to be higher than the N deposition during the pre-industrial period (Holland et al 1999).
The combustion of fossil fuel is primarily responsible for the anthropogenic emission of oxides of nitrogen (NO x ). A significant amount of NO x is emitted by petrol and by dieselbased internal combustion engines from the transport sector. Around 30-40% of global nitrogen oxide emissions and 85% of regional emissions in North America are due to the combustion supply (Logan 1983). NO x is emitted into the atmosphere from both natural and anthropogenic sources such as fossil fuel combustion (Gadi et al 2003), biomass burning (Lobert et al 1990), the oxidation of atmospheric ammonia and lightning (Seinfeld and Pandis 1998). A higher concentration of NO x may damage human health and vegetation (Kinney 2008). Oxides of N form HNO 3 and PAN in the atmosphere, which are important Nr species from an air quality viewpoint. The growing energy demand has resulted in the increased emission of NO x from coal combustion in thermal power plants and from petroleum combustion in the transport sector. In addition, biomass burning in traditional cooking and heating is also a significant source of NO x in the Indian region (Singh et al 2014).
India has the second highest population in the world. Around 70% of the population lives in villages, taking care of agriculture to meet food supply demands. In order to get a higher yield of agriculture and of the food product, the increased practice of fertilizer application has added an extra load of nutrients, especially the Nr species. Most N is applied as urea or DAP fertilizers, which are considered significant sources of NH 3 . Urea and ammonia-based fertilizers contribute atmospheric NH 3 due to volatilization. The elevated ambient temperature in tropical countries such as India plays an important role in NH 3 /NH 4 equilibrium because of the greater volatization of NH 3 due to its higher vapour pressure (Asman 1992). The emission factor for the percent loss of N of urea and DAP are 15 and 5, respectively (Parashar et al 1998). Other sources, such as decomposing excrement, inadvertent losses during the production of fertilizer and burning biomass, also contribute atmospheric NH 3 . In addition, shifting food habits from vegetarian to non-vegetarian (meat, milk and eggs diet, livestock production) are also responsible for increased atmospheric Nr.
Due to the population explosion and the exponential growth in agriculture, power generation and transport sectors in India, the emission of NH 3 and NO x species are increasing and are expected to rise further, having an adverse impact on human health and on plants. Considering the implications of these two species (NH 3 and NO x ) in the changing N cycle, the present study was carried out in Indo-Gangetic plains at two sites of different characteristics (urban and rural) to study the emissions and atmospheric levels in relation to their sources and their role in the meteorological parameters.

Monitoring sites
Two sites that have different land uses, land covers, sources and population densities in the Northern Indian plain were selected. The urban site has been referred to as Okhla, whereas the rural site has been referred to as Mai. Basic information regarding both sites has been given in table 1.
The location of both sites has been shown in figure 1. The map given in figure 1 also gives an idea of the land use, land cover and housing density in the surrounding areas of the sites. As the major objective of the study was to observe reactive nitrogen chemistry at typical urban and rural sites, the Okhla and Mai sites were chosen. Okhla is a residential and industrial urban area that has a high population density with a mixed population of four wheelers and two wheelers, construction activities, congested road and streets indicating a very close livelihood due to which sanitation issue was always a concern during sampling. In contrast, the rural area of Mai has agricultural dominance, rural activities and a lesser population density, which has a lesser influence on industrial activities and transport pollution.
2.1.1. Okhla. Okhla was selected as an urban site, which is located in the South Delhi district. The site is dominated by heavy vehicular traffic and industries. Okhla had a population density of 10 935 inhabitants per square kilometre (28 320/sq mi), with a population growth rate of 20.59% during the decade of 2001-2011. The South Delhi district had a population of 2 733 752 in 2011, which is roughly equal to the total population of Jamaica or the US state of Nevada. The Okhla area is congested in terms of high-rise buildings, roads and other infrastructure. In the vicinity of the sampling site, there is a cremation ground, which can also contribute emissions of pollutants. Such waste also can be potential source of Nr. Table 2 gives the number of meat shops and their meat production in the Okhla locality.   For the collection of gaseous NH 3 , 25 mM H 2 SO 4 was used as the absorbing solution, while NO 2 was collected using the arsenite method in which sodium hydroxide and sodium arsenitemixture are used as an absorbing reagent. The details of the methods are given elsewhere (Singh et al 2014).

Results and discussion
3.1. Ammonia 3.1.1. Concentrations at the urban and rural sites. The average concentrations of NH 3 were recorded as 40.7 μg m −3 at the Okhla site and 51.6 μg m −3 at Mai. Table 4 gives descriptive statistics of the NH 3 at the Okhla and Mai sites.
At the Okhla site, NH 3 concentrations varied from 20 to 126 μg m −3 , with an average of 40.7 ± 16.8 μg m −3 , although the daily average concentration of NH 3 did not show a large variation during each respective month. However, relatively higher average concentrations were recorded during the months of December (50.8 μg m −3 ), January (49 μg m −3 ) and May (50 μg m −3 ). At Okhla, the average NH 3 concentration did not show much difference during the different seasons. During winter (November-January), summer (May-June) and the monsoon (July-September) season, the average NH 3 concentrations were recorded as 44.3, 42.5 and 38.9 μg m −3 , respectively. Such a seasonal pattern of NH 3 is very similar to the pattern reported by Singh and Kulshrestha (2012) at the JNU site of Delhi city.
At Mai, the NH 3 concentration varied from 13 to 127.8, with an average of 51.6 ± 22.8 μg m −3 . At this site NH 3 concentration had more pronounced seasonality. The average concentrations of NH3 during winter, summer and the monsoon season were recorded as 72.0, 47.2 and 30.7 μg m −3 , respectively. The difference in NH 3 levels at the urban and the rural site may be attributed to the difference in source types and their strength. At Okhla, the possible sources of NH 3 are solid municipal waste, tissue waste, traffic, etc, which remain constantly active throughout the year, while at Mai, the major sources of NH 3 are agricultural activities and cooking. During the winter season, fertilizer application and other agricultural activities are at full swing, which emit NH 3 . Gupta et al (2003) also observed the highest NH 3 during the winter season due to maximum vegetation  In general, N fertilizer application, increased livestock production, solid waste generation and high population density are the possible sources of NH 3 in India. Generally, 240 kg per ha per year DAP in basal dressing and 480 kg per ha per year urea in top dressing are applied in the field. However, the total fertilizer amount depends on the crop and season. According to the statistics, consumption of major fertilizers has increased in recent years. From 1999-2000 to 2011-2012, urea consumption increased by 44%, whereas DAP consumption increased by approximately 61%. According to data from the Fertilizer Association of India, out of the total national consumption, North India consumes around 38% of urea and 40% of DAP. However, less than 30% of the applied N fertilizer is absorbed by the crops. More than 20% is evaporated as ammonia emissions (Cai 1997). Around 20% of urea-N is lost into the atmosphere immediately after its application in the soil (Whitehead and Raistrick 1990) Factors such as soil moisture and temperature, soil texture and the form of nitrogen affect the intensity of NH 3 evaporation, controlling the efficiency of the fertilizer. Fertilizer N use efficiency can vary from one environment to another (Fennand Hossner 1985). In addition, the changing food habits of people also contribute to the increasing levels of atmospheric NH 3. However, NH 3 concentration is likely only representative of a small area as it is mainly influenced by low-level local sources, whereas for NO x , more remote sources also can play a role.
According to estimates, around 1175 Gg NH 3 is contributed by different kinds of fertilizer in India (Parashar et al 1998), with urea as the highest contributor. As shown in figure 2, around 94% NH 3 is contributed by urea fertilizer.
The other major source of NH 3 is the livestock population, which contributes around 1433 Gg of NH 3 . The livestock population is dominated by cattle, which emit a major fraction (∼73%) of atmospheric NH 3 (Parashar et al 1998). It is important to mention that solid waste and human excreta, etc are also important sources of ambient NH 3 at Okhla. As mentioned earlier, the site is highly populated and has poor sanitation arrangements. In addition, according to recent reports non-agricultural sources, such as vehicles equipped with catalytic converters, also contribute NH 3 (Bari et al 2003). Such non-agricultural emissions influence NH 3 mixing ratios at urban locations and nearby roads (Sutton    different months at Okhla and Mai. At Okhla, daytime concentrations of NH 3 varied from 4.9 to 139.1 μg m −3 , with an average of 38.7 μg m −3 , while nighttime concentrations varied from 3.7 to 114.2, with an average of 42.6 μg m −3 . The high concentrations during the nighttime can be attributed to atmospheric stability due to the trapping of gaseous NH 3 near ground level at both sites (Burkhardt et al 1998, Cadle et al 1982, Singh et al 2001, Singh and Kulshrestha 2012. Figure 3(b) shows the average concentrations of NH 3 during the day and night times at Mai. Day concentrations of NH 3 varied from 9.9 to 148.8 μg m −3 , with an average of 49.8, while nighttime concentrations varied from 5.9 to 145.7, with an average of 53.4 μg m −3 . Unlike Okhla, where NH 3 concentrations were always recorded higher during the nighttime, at Mai, daytime concentrations were higher during the December, March, May and June months. Daytime high concentrations of NH 3 are supported by Behera et al (2013) due to large evaporative emissions from various sources. It is possible that daytime higher NH 3 concentrations are observed due to evaporation from fertilized fields since these months are typical months during which fertilizer is applied to most of the crops. In addition, agriculture residue burning during the summer season may also give rise to daytime higher concentrations of NH 3 at Mai. To assess the difference between the daytime and nighttime levels of gaseous NH 3 , a paired t-test (two-tailed) was performed. The test showed that there was no significant diurnal difference at a 95% confidence level (P > 0.05) for both sites.

Comparison of concentrations of gaseous NH 3 with
other studies. Table 5 gives a comparison of NH 3 , concentrations at various sites worldwide. In general, NH 3 levels at both sites are much higher than at temperate sites. This is probably due to the tropical climate. The high temperature of tropics helps in the higher evaporation of NH 3 from agriculture fields, municipal waste dump yards and other sources. In addition, NH 3 , being an alkaline gas with an abundance of alkaline dust, also favours its gaseous existence (Singh and Kulshrestha 2012), while most temperate regions have an acidified atmosphere in which the dominance of sulphuric acid consumes gaseous NH 3 , forming NH 4 + and SO 4 2 − aerosols (Kulshrestha 2013).

Nitrogen dioxide (NO 2 )
3.2.1. Concentrations of NO 2 at Okhla and Mai. Table 6 gives descriptive statistics of NO 2 at Okhla and Mai. The average concentration of NO 2 at Okhla was recorded as 24.4 μg m −3 , ranging from 5 to 63.6 μg m −3 . At Mai, NO 2 varied from 2.5 to 40.6, with an average of 18 μg m −3 . Figure 4 shows the monthly variation of NO 2 at Okhla and Mai. The seasonal mean NO 2 concentrations at Okhla were recorded as 39.6, 24.5 and 10.4 μg m −3 during the winter, summer and monsoon season, respectively. At Mai, the mean NO 2 concentrations during the winter, summer and monsoon season were recorded as 27.5, 17.2 and 4 μg m −3 , respectively. The major source of NO 2 in a rural area is biomass burning (wood, crop residue and dung cakes) during cooking and heating. (Singh et al 2014). Generally, around 621 tons of biomass are burnt by a typical village of around 200 families in North India, as reported by Singh et al (2014). Okhla, since it is an urban site, had higher NO 2 than Mai. This is probably because the Okhla area is dominated by vehicular and industrial activities, which contribute significant amounts of NO 2 in air. Gurjar et al (2004) observed that NO   The possible reason for high NO 2 concentrations during the winter season is the formation of inversions in which pollutants are trapped in the boundary layer, whereas during the summer period, a higher mixing height allows dispersion in free tropospheric air, causing the dilution of the pollutants (Reddy et al 2012). In addition, higher photochemical activity also helps in breaking down the pollutants into other simpler compounds. Higher concentrations of pollutants during the winter season is a typical feature in the India region. Gupta et al (2003) have also reported higher concentrations of NO 2 during winters. A similar observation was reported by Kelly et al (1989) and Barbiaux et al (1992). However, the average concentrations observed in the present study are much higher than those reported in other studies for rural areas. Higher concentrations of NO 2 at Mai are probably due to NO 2 emissions from biomass burning during cooking and heating. In rural India, domestic cooking and heating is based on biomass burning (Singh et al 2014). In addition, the burning of rice and wheat straw in the industrial units located at ∼13 km from the sampling site could also contribute NO 2 at Mai. Also, local vehicular traffic can also give rise to NO 2 in the atmosphere. NOx emissions into the atmosphere are dominated by the combustion of fossils fuels and biomass, which represent 75% of total emissions, with more than 50% of that from fossil fuels alone. NO x emissions are therefore strongly influenced by anthropogenic activities. Ammonia oxidation is a potential tropospheric source of NOx. The oxidation of ammonia in the atmosphere is initiated by the reaction with OH This reaction may play a very important role in controlling NH 3 concentrations, especially in tropics where OH concentrations are higher in the atmosphere (McConnell 1973).
At lower NOx levels, NH 2 reacts with O 3 , forming the NH 2 O radical The NH 2 O radical reacts with oxygen and finally forms NO. Therefore, such a reaction sequence results in a source or a sink of nitrogen oxides, depending on ambient NOx concentrations (Delmas et al 1997).
3.2.2. Day-night variation of NO 2 at Okhla and Mai during different months. Figures 5(a) and (b) show the variation of NO 2 at the Okhla and Mai sites during different months. The daytime concentrations of NO 2 varied from 5.0 to 72.6, with an average of 25.9 μg m −3 , whereas nighttime concentrations of NO 2 varied from 5.0 to 66.2, with an average of 22.8 μg m −3 . At Mai, daytime concentrations varied from 2.3 to 56.9, with an average of 19.8 μg m −3 , while nighttime concentrations varied from 2.6 to 36.9, with an average of 17.7 μg m −3 . Daytime higher concentrations of NO 2 at both sites are quite obvious due to various sources, such as industries, traffic, etc, which are more active during the daytime. A paired t-test (two-tailed) result showed the significant diurnal variation at a 95% confidence level (P < 0.05). In a study reported by NEERI (2008) at Delhi, during the summer season, the daytime NO 2 concentration  was recorded as 40 μg m −3 , while at midnight, the NO 2 concentration was 30 μg m −3 . During the winter season, the daytime concentration of NO 2 was recorded at about 60 μg m −3 , which was higher than was recorded in summer and in the post-monsoon season. Contrary to the winter season, the NO 2 level increased gradually after sunset to about 80 μg m −3 by midnight. It is worth mentioning that unlike NH 3 levels, which are affected by local sources, NO 2 levels can be affected by more remote sources.

Conclusion
The present study reports the abundance of gaseous inorganic reactive nitrogen species (NH 3 and NO 2) at urban and rural sites in the Indo-Gangetic plains in India. The results showed significant differences in the NH 3 levels between the rural and urban sites due to different types of sources and their strength. High concentrations of gaseous NH 3 measured at the rural site can be attributed to fertilizers and biomass burning, whereas municipal waste, human excreta, vehicular traffic and tissue waste were found responsible for NH 3 emissions at the urban site. The seasonal variation of NH 3 at the rural site (Mai) was more significant, with NH 3 concentrations at 72 μg m −3 during the winter, 47.2 μg m −3 during the summer and 30.7 μg m −3 during the monsoon season due to agricultural patterns and the amount of fertilizer applied during different seasons. The average NH 3 levels during different seasons at Okhla were not very different, indicating that at the urban site, the sources of NH 3 are almost constant throughout the year. At Okhla, NH 3 concentrations were recorded as 44.2 μg m −3 during the winter, 42.5 μg m −3 during the summer and 38.9 μg m −3 during the monsoon season. Generally, at both sites, nighttime NH 3 was always higher than daytime NH 3 due to atmospheric conditions. The average concentrations of NO 2 were 24.4 μg m −3 and 18.8 μg m −3 at Okhla and Mai, respectively. Seasonal average NO 2 concentrations at Okhla were recorded as 39.6, 24.5 and 10.4 μg m −3 during the winter, summer and monsoon seasons, respectively. The average concentrations of NO 2 were recorded as 27.5, 17.2 and 4 μg m −3 during the winter, summer and monsoon seasons, respectively, at Mai. The lower NO 2 concentrations during the monsoon season at both sites can be due to the washout effect of Monsoon rains. Generally, daytime NO 2 concentrations were higher than nighttime NH 3 at both sites, as the NO 2 sources are very active during the daytime. The study reveals that due to biomass burning, the rural site (Mai) showed high levels of NO 2 . Such high levels of NO 2 are not generally expected at a rural site. The study reveals that both NH 3 and NO 2 are contributed by biomass burning at the rural site, suggesting that biomass burning emissions might have severe health effects in the long run, especially on females and children, who are most exposed in side houses where burning takes place. Therefore, there is a need to improve our understanding about Nr emissions and their chemistry at rural sites in order to prepare reliable estimates of Nr. This can be achieved by strengthening the monitoring of and modeling efforts about Nr in India. This will further help in understanding the potential health and environmental risks due to Nr.