The rising emissions of reactive nitrogen (Nr) from transport, industrial, and agricultural sectors in India have resulted in its consequent interactions with the removal mechanism of the atmospheric dust. This study, therefore, reports the fluxes of reactive nitrogen along with other inorganic species through dustfall over six sites of Delhi–National Capital Region (NCR) characterized by the changing dynamics of its different land-use pattern. The highest Nr fluxes were observed at site SMA Industrial estate (SMA; NO3− = 16.45 ± 10.17 mg m−2 day−1, NH4+ = 16.33 ± 16.00 mg m−2 day−1) and lowest at site Chuchchakwas village (CV; NO3− = 1.24 ± 0.16 mg m−2 day−1, NH4+ = 0 mg m−2 day−1). Sites Mukherjee Nagar (MN), Peeragarhi Chowk (PC), Jawaharlal Nehru University (JNU), and Noida Phase II (N-II), on the other hand, showed 3.59 ± 1.00, 3.39 ± 0.61, 2.98 ± 0.84, and 3.36 ± 0.78 mg m−2 day−1 of NO3− fluxes and 0.30 ± 0.06, 0.22 ± 0.04, 0.21 ± 0.04, and 0.22 ± 0.05 mg m−2 day−1 of NH4+ fluxes, respectively. The fraction of the total ions in the water soluble extract of the dustfall was also noticed to be the highest at the SMA site (22.2%) and lowest at the CV site (1.5%) with MN, PC, JNU, and N-II showing 3.5%, 3.7%, 2.9%, and 3.9% of their respective contributions. Relative abundances of Ca2+ and SO42− in the dustfall substantiated the stoichiometric reactions involved in Nr scavenging. The role of Ca2+ in the spatiotemporal variability of Nr fluxes was established with the help of neutralization ratios and regression plots. Morphological and particle size analysis further confirmed the anthropogenic-induced crustal interferences in the summertime dustfall fluxes of Nr species.
In response to the rising demand of food and energy security across the globe, the reactive nitrogen species (Nr) have undergone a rapid accumulation as NO3− and NH4+ the in environment. This is clearly reflected in the growing sources and altered transport pathways of global nitrogen cycling that has resulted in a tenfold rise in Nr formation (Vitousek et al. 1997; Galloway et al. 2004; Phoenix et al. 2006). However, by the virtue of its direct emission pathways, much of the Nr created has been dispersed heterogeneously as NOx and NH3 into the atmosphere. As a key driver to air pollution chemistry and climate changes, these Nr species have become central to the photochemical reactions during its residence time in the atmosphere. This has eventually resulted in an imbalance arising between its emissions and deposition fluxes, creating a spatial distribution of inorganic Nr even in the remotest regions of the world. (Zellweger et al. 2002; Galloway et al. 2004; Vet et al. 2014).
Owing to the rising emissions and increasing interactions of Nr with the atmospheric transport and removal processes, there has been an exceedance in its deposition fluxes beyond the critical threshold levels (Dentener et al. 2006). This is clearly evident in the increased conversion rates of NOx into HNO3 and plant available nitrogen (PAN) that have gone up from <5% to 24% h−1 in the urban plumes (Berkowitz et al. 2004). Thus, with the emission pattern changes between the years 2000 and 2007, an increasing median deposition values of Nr has been observed over developing Asia (+13.6%) and Africa (+19%), unlike the declining trend over the developed nations of Europe (−2.7%) and North America (−4.3%; Tørseth et al. 2012; International Joint Commission 2010).
Abundance of mineral dust (CaCO3) over the Indian atmosphere has significantly affected the deposition fluxes of the acidic gaseous pollutants by providing surface reactive sites to their heterogeneous chemistry (Usher et al. 2003; Kulshrestha 2013). NOy in the atmosphere can, therefore, be readily scavenged as NO3− by the strong base cations (Ca2+) of the mineral dust that eventually favor its partitioning toward the coarse-mode particle range (D > 1.2 µm; Kulshrestha et al. 1998; Metzger et al. 2006; Kumar et al. 2008):
The Nr partitioning in the fine-mode fraction (D < 1.2 µm), on the other hand, is known to be favored under NH3 excess condition (Kulshrestha et al. 1998; Metzger et al. 2006; Squizatto et al. 2013). But owing to the thermodynamic instability of the NH4NO3 salt formed, the evaporated aerosol precursor gases (NH3 and NO3−) are condensed on the preexisting larger aerosol particles that ultimately results in its partitioning toward the coarse-mode range (Wexler and Seinfeld 1990):
While the coarse mineral mass with sizes above 10 μm tends to deposit near its source, the fine mineral dust, on the other hand, has the tendency to be transported over the oceans and across the continents (Salam et al. 2003). This has resulted in dustfall becoming an important mechanism in controlling the fate of gaseous pollutants through sedimentation and impaction over the Indian regions (Morselli et al. 1999; Yun et al. 2002; Kulshrestha 2013).
A number of studies have established the relative abundance of Ca2+ over NH4+ as the major neutralizing agent in the coarse-mode fraction over the semiarid tracts of India (Kulshrestha et al. 1996, 1998; Rastogi and Sarin 2005; Rengarajan et al. 2007; Satsangi et al. 2013; Kumar and Sarin 2010). With the shifting precipitation pattern of the Indo-Gangetic Plain and its expanding sectors of agricultural and industrial activities, there has been a growing significance of dustfall providing ultimate sinks to its excess Nr acidity. As Delhi is considered to be the epitome of the growing population and pollution problems of the Gangetic Plains, the present study was, therefore, carried out for substantiating the available estimates of Nr deposition fluxes. Thus, the morphological and ionic compositions of the dustfall fluxes were determined and variations in the Nr dustfall fluxes were evaluated. The results were used for deciphering the interaction mechanism of Nr with the mineral dust for the purpose of strengthening our understanding concerning its tropospheric reactions and removal processes under dry weather conditions.
2.1. Sampling location
Delhi is located between 28°24′17″–28°53′00″N and 76°50′24″–77°20′37″E at the periphery of the Gangetic Plains with the Great Thar Desert to the west. Based on the distribution of pollution sources along with the changing dynamics of urban–rural composition, the study was carried out at the following six sites of Delhi–National Capital Region (NCR) region as shown in Figure 1.
2.1.1. Site Mukherjee Nagar (MN)
The site is located in the highly populated district of North Delhi. The area is characterized by the unplanned sprawling of residential colonies with a number of unauthorized colonies spurting along its fringe. Owing to illegal encroachment of street vendors to its already narrow lanes, the site often subjected to traffic jams especially during the evenings.
2.1.2. Site SMA Industrial estate (SMA)
It is an industrial estate located at a kilometer distance from the bustling Azadpur Mandi in the Northern District Zone of Delhi. Clutters of small-scale industries here are involved in re-rolling, annealing, and pickling of stainless steel products and enamel wares. The area is also characterized by its poor road condition with the regular movement of diesel-fueled trucks leading to its total breakage.
2.1.3. Site Peeragarhi Chowk (PC)
It is a well-recognized transport hub located at the major intersection connecting the outer ring road and Rohtak Road [National Highway 10 (NH-10)]. The area therefore witnesses a regular inflow of heavy traffic composed of diesel-fueled, heavy-duty goods and passenger vehicles that are involved in providing interstate connectivity.
2.1.4. Site Jawaharlal Nehru University (JNU)
It is located within the premises of Jawaharlal Nehru University campus in the south district zone of Delhi. The site is characterized by the presence of mini-forest covering with no major industrial areas surrounding it. Except for the few vehicles used by the students and the faculty for their daily purpose, the area is relatively free from traffic emissions, thereby making it a typical urban forest in its land-use pattern.
2.1.5. Site Noida Phase II (N-II)
Located in the Gautam Buddha district of Delhi–NCR, the site is a typical representative of semiurban, land-use pattern with urbanization spreading rapidly into the already fertile hinterlands of the western Uttar Pradesh. The site is characterized by frequent power cuts and failures that have resulted in an extensive use of diesel-fueled, heavy-duty generators for restoring normalcy in the daily functional activities of its people.
2.1.6. Site Chuchchakwas village (CV)
The site is a small village located in the Jhajjar district of Haryana state, constituting the National Capital Territory of Delhi. Owing to the changing demography and socioeconomic patterns of the area, there has been a great reduction of its agricultural fields into fallow lands. Its easy accessibility via roads from Dadri and Beri has also led to the site being subjected to the road dust emissions from its unpaved surfaces.
2.2. Sample collection and gravimetric analysis
Dust samples were collected during the summer season from April to June 2015 using surrogate surfaces. The 3-month sampling represents a quarter of the annual deposition. This period has special importance because of the dry weather conditions of the summer season that provide comparatively higher dust deposition fluxes than the rest of the season. This has been well substantiated by Ghosh et al. (2014) over the semitracts of India, where high particulate mass loading resulting from the strong winds and dry atmospheric condition were typically observed during April–May. Petri plates of 140-mm dimension were used for the collection of atmospheric dust, owing to its chemical inertness to the organic and inorganic components of the dustfall fluxes. For the purpose of providing significant aerodynamic resistance, the petri plates were kept on the roof top of the building and were changed after every 5 days of ambient exposure. Micrometeorological techniques such as eddy measurements being sensitive to the supermicron range particulates are commonly used in the estimation of deposition fluxes based on diffusion, impaction, and interception (Davidson et al. 1985; Ruijgrok et al. 1995). Surrogate surface approaches, on the other hand, gain credibility with the increasing particle sizes and are, therefore, best known for providing accurate measurements to the rapidly falling particles (Davidson et al. 1985; Hales et al. 1987). Wind tunnel studies by Goossens et al. (1994) have already established the dust collection efficiency of dry propylene surfaces to be higher than a moist filter pack. Saxena et al. (1992) have also shown maximum deposition rates for propylene dishes based on the dry deposition rates of SO42− and NO3− over the semiarid region of India. Therefore, for meeting the objectives of the present study in the best possible and feasible manner, propylene petri dishes were selected as surrogate surfaces in the passive collection of atmospheric dust.
Keeping in view the dry weather condition under which the sampling has been performed, provision of wind fetches and rain sheds were not deemed necessary, as their presence would have interfered with the dust deposition rates. Hence, the petri plates were kept exposed to the naturally falling dust and were, therefore, changed after every 5 days of ambient exposure instead of weeks. Roof tops were selected for avoiding any interference from local soil/surface or other secondary fugitive dust (Alahmr et al. 2012; Zhao et al. 2010). Crops or forests were avoided for sampling as their presence would have introduced canopy resistance to the passively collected dust samples. Impact of winds on the sample representativeness could also be ignored as the edges of petri dishes were high enough to check any rebouncing of deposited dust particles under normal wind conditions. There were no incidences of high winds, and the average wind velocity was observed to be 2.6 m s−1 during the sampling period. Representativeness of the samples could be further corroborated through duplicate sampling where the uncertainties associated with the dustfall and ionic fluxes were observed to be ≤10% at all the sites.
The mass concentrations of the collected dust were determined gravimetrically using the electronic microbalance (Mettler Toledo-ME 204). Each petri plate was weighed before and after sampling. The dustfall fluxes were calculated using the following equation (Katz 1969):
where M1 is the weight (mg) of the petri plates before dust collection, M2 is the weight (mg) of the petri plates after dust collection, A is the area (m2) of the petri plates, T is the duration (days) for which the petri plates were exposed for dust collection, and DF is the dustfall flux (mg m−2 day−1).
2.3. Chemical analysis of the dustfall fluxes
The water soluble fraction was extracted by washing the petri plate with 30 mL of ultrapure Milli-Q water followed by its ultrasonication for 30 min at a working frequency of 1 Hz. The extracts were filtered through 0.2-µm nylon syringe filter and were analyzed for its major cations (K+, NH4+, Na+, Ca2+, and Mg2+) and anions (F−, Cl−, NO3−, and SO42−) with the help of ion chromatography (Metrohm 883 basic plus model). Anions were analyzed using a Metrosep A SUPP 4 column of 250 × 4 mm2 dimension and eluent mixture of 1.8 mmol L−1 Na2CO3 + 1.7 mmol L−1 NaHCO3 at a flow rate of 1 mL min−1. Cations, on the other hand, were analyzed using a Metrosep C4 column of 100 × 4 mm2 dimension with the eluent mixture of 1.7 mmol L−1 nitric acid + 0.7 mmol L−1 dipicolinic acid along with the 50 mM of H2SO4 as a self-generating suppressor at a flow rate of 0.9 mL min−1. Calibration of the method and quantification of components were carried out using MERCK reference standards (CertiPUR). In the case of anions, 1-, 2-, and 5-ppm standards were used for achieving the calibration curve of each component, whereas the calibration curve of each cation was achieved by using 2-, 5-, and 10-ppm standards, respectively. The precision values were observed to be 0.23% for Na+, 1.23% for NH4+, 0.17% for K+, 0.37% for Ca2+, 0.7% for Mg2+, 2.18% for SO42−, 2.13% for NO3−, 1.6% for Cl−, and 0.79% for F−, thereby indicating good quality of data.
2.4. Morphological analysis of dustfall fluxes
The dustfall samples were analyzed for its morphological characteristic using scanning electron microscopy (SEM; Carl Zeiss EVO 40, Germany). For this purpose, the dust samples were mounted directly on the carbon tape pasted on the metallic stubs using a soft hair brush. The loaded sample was subjected to a thin film coating of gold deposition with the help of a Sputter loader (Sputter coater-Polaron SC7640). At 20 kV, the SEM images were obtained under different magnifications of 11-mm working distance. The images were further analyzed for its morphological features and size distribution with the help of ImageJ software. The software is used for converting a grayscale image into binary format with the help of brightness discrimination. This helps in providing background subtractions during particle sizing where each particle is characterized by its projected area in pixels. Since the area has to be calculated by the diameter of its equivalent circle, special corrections were applied to the effective diameter of the irregular particles.
3. Results and discussion
3.1. Ionic composition of the dustfall fluxes
The average summertime fluxes of the dustfall and its ionic species were determined for the sampling period of 90 days (April–June), as given in Table 1. It was observed that the mean values of dustfall ranged from 39 g m−2 at the N-II site to 104 g m−2 at the SMA site, owing to its different land-use activities. The Ca2+ showed the highest fluxes among the major cationic species with a mean value range from 0.34 (CV site) to 1.5 g m−2 (SMA site). The SO42−, on the other hand, was observed to be the highest among the major anionic species with a mean flux range from 0.28 (CV site) to 12 g m−2 (SMA site).
The percent contribution of the major ions to the dustfall composition was also represented with the changing dynamics of the pollution sources as shown in Figure 2. Heavy industrial emissions resulted in the highest contribution of ions (22%) to the dust fluxes at site SMA in the order of SO42− > Cl− > NO3− > Ca2+ = NH4+ > Na+ = K+ > F−. Minimum anthropogenic disturbances at site CV, on the other hand, resulted in the lowest contribution of its ions to dustfall fluxes (1.6%) with Cl− > Ca2+ > SO42− > NO3− > K+ > Na+ > Mg2+ = F− > NH4+. The order of ionic fluxes at site MN and PC was observed to be Ca2+ > SO42− > NO3− > Cl− > Na+ > K+ > Mg2+ = F− > NH4+ and Ca2+ > Cl− > SO42− > NO3− > Na+ > K+ > Mg2+ > NH4+ = F− that contributed 3.5% and 3.7% to its dustfall fluxes, respectively. Site JNU being a typical urban forest showed a relatively lower contribution of its ionic fluxes (2.9%) that was interestingly observed to be in the sequence of SO42− > Ca2+ > NO3− > Cl− > K+ > Na+ > F− > Mg2+ = NH4+. The anomaly of high SO42− fluxes could be attributed to the use of Portland cement in the untimely construction and renovation activities of the campus done during the sampling period. The dust fluxes at site N-II showed 3.9% of its fraction contribution to the ionic fluxes with Ca2+ > SO42− > NO3− > Cl− > Na+ = K+ > Mg2+ = F− > NH4+.
3.2. Nr scavenging by the dustfall fluxes
Neutralization ratios were calculated for ascertaining the scavenging capacity of the dustfall fluxes to SO42− and NO3− using the following equations for the equivalent concentration of respective species (Kulshrestha et al. 2003; Satsangi et al. 2013):
It was observed that Ca2+ showed the highest neutralization ratios among all the base cations with its values ranging from 0.25 (SMA site) to 1.80 (CV site), as given in Table 2. The lowest neutralization ratio, on the other hand, was observed for NH4+ at all the sites except SMA, where the heavy industrial emissions rich in ammonia provided a higher neutralization ratio of NH4+ (0.16) rather than Mg2+ (0.09) and K+ (0.01). The ratios clearly established the dominating role of Ca2+ and the near absence of NH4+ in the stoichiometric neutralization reaction of the dustfall fluxes.
A further analysis into the regression plots of NO3− + SO42− with Ca2+ revealed the significance of Ca2+ and SO42− in the scavenging of NO3− in the dust fluxes as shown in Figures 3a, 3c, 3e, 3g, 3i, and 3k. As most of the data values were observed below the 1:1 neutralization line of NO3− + SO42− versus Ca2+ plots, a complete scavenging of NO3− was confirmed by the excess Ca2+ in the dustfall fluxes. However, a low median equivalence ratio of NO3−/SO42− that varied from 0.25 at the SMA site to 0.80 at the N-II site showed a relative abundance of SO42− over NO3− in the dustfall fluxes. This becomes suggestive toward the strong affinity of SO42− rather than NO3− with Ca2+ in the dustfall fluxes that has eventually resulted in NO3− scavenging only after the complete neutralization of SO42− with Ca2+ (Metzger et al. 2006; Kulshrestha et al. 2009).
NH4+ was plotted with NO3− + SO42− for all the sites except CV, where nondetectable (ND) values of NH4+ were maintained throughout the sampling period. As shown in Figures 3b, 3d, 3f, 3h, and 3j, a weak correlation of NH4+ with NO3− + SO42− confirmed the near absence of NH4+ in the stoichiometric neutralization reactions. The fluxes of NH4+ were thus considered to be limited by the predominance of a strong base cation (Ca2+) in the dustfall fluxes that resulted in most of the dataset values lying above 1:1 neutralization line in NH4+ versus NO3− + SO42− plots.
Estimated regression coefficients (β1) also confirmed the significant role of Ca2+ in the scavenging of NO3− + SO42− with values ranging from 0.3 to 1.36 (Table 3). As the coefficients were not found to be significant for NH4+, its role in the dustfall scavenging was again confirmed to be negligible in the dustfall scavenging.
3.3. Spatial distribution of the Nr fluxes
Figure 4 shows the average dustfall fluxes of NO3− as well as NH4+ at different sites. The lowest fluxes of NO3− (1.24 ± 0.16 mg m−2 day−1) and NH4+ (0 mg m−2 day−1) were observed at the rural site CV, owing to its minimum exposure to the anthropogenic disturbances. On the other hand, the highest fluxes of NO3− (16.45 ± 10.17 mg m−2 day−1) and NH4+ (16.33 ± 16 mg m−2 day−1) were observed at the SMA site, owing to the diesel emissions from trucks and industrial activities along with the losses from the industrial refrigeration units that contributed significantly to the emission of its precursor gases (NOx and NH3). Despite the regular emissions from the traffic flow, the fluxes of NO3− (3.39 ± 1 mg m−2 day−1) and NH4+ (0.22 ± 0.04 mg m−2 day−1) at site PC were observed to be relatively lower than the fluxes of NO3− (3.59 ± 1.00 mg m−2 day−1) and NH4+ (0.30 ± 0.06 mg m−2 day−1) at high density residential site MN. Such a disparity could be attributed to the bottleneck situation of traffic congestion at site MN arising from its heavily encroached narrow road conditions. Prevalent use of light-duty vehicles in addition to its poor sewerage system could also be implicated for the high NH4+ fluxes at the MN site as compared to the PC site (Reis et al. 2009). However, the NO3− (3.36 ± 0.78 mg m−2 day−1) and NH4+ (0.22 ± 0.05 mg m−2 day−1) fluxes at the semiurban N-II site were observed to be comparatively higher than the NO3− (2.98 ± 0.84 mg m−2 day−1) and NH4+ (0.21 ± 0.04 mg m−2 day−1) fluxes at the urban JNU site. This could be attributed to the frequent use of power generators over the suburban N-II site along with the biomass burning and biogenic emissions from its local soil (Anderson et al. 2003; Calvo et al. 2013). Since the alkaline nature of the dustfall fluxes are known for repelling the adsorption of NH3 onto the dust particles, the fluxes of NH4+ were, therefore, observed to be consistently lower than the fluxes of NO3− at all the sites, irrespective of their emission strength (Singh and Kulshrestha 2012).
Tropical regions are usually characterized by the presence of high dust loading in their atmosphere, unlike the temperate. This results in additional dry deposition fluxes being provided through mineral dust buffering reactions over the tropics that inevitably introduces disparity in the values of such regional comparisons. On comparison of our results with the other case studies (Table 4), a higher Nr deposition flux pattern was observed for our urban study area. The fluxes of NO3− over Pune showed comparatively higher values than the urban and suburban sites of the present study. The Nr fluxes at the rural site, on the other hand, were observed to be even lower than the Gopalpura rural site of Agra.
3.4. Temporal variation of the Nr fluxes
The time series of the Nr fluxes were plotted for different sites, as shown in the Figure 5. The NO3− fluxes showed values ranging from 0.2 to 8.2 mg m−2 day−1at site MN, 0.04 to 156.5 mg m−2 day−1 at site SMA, 0.1 to 7.6 mg m−2 day−1 at site PC, 0.1 to 8.9 mg m−2 day−1 at site JNU, 0.02 to 7.6 mg m−2 day−1 at site N-II, and 0.4 to 2.1 mg m−2 day−1 at site CV. The NH4+ fluxes, on the other hand, showed a value range of 0–0.7 mg m−2 day−1 at site MN, 0–379.5 mg m−2 day−1 at site SMA, 0.2 ± 0.04 mg m−2 day−1 at site PC, 0.2 ± 0.04 mg m−2 day−1 at site JNU, and 0.03–0.5 mg m−2 day−1 at site N-II with ND values at site CV. Owing to the summertime meteorology of strong winds and unstable atmospheric conditions, fluxes of NO3− and NH4+ showed insignificant differences (α < 0.05) during most of the sampling period. However, frequent dust storm and rain events provided minor fluctuations in the time series data.
3.5. Morphological and particle size distribution of the dustfall
Morphological assessment of the dustfall samples were carried out with the help of SEM for the purpose of showing the diversity of the dust particulates. As shown in Figure 6, the most common shapes were observed to be irregular, tubular, platy, long and prismatic, porous, spherical, crystalline, rhombic, and elliptical. This could be attributed to the geology along with wind velocity and direction acting as the major controllers of the mineralogy and morphology of the airborne dust (Zarasvandi et al. 2011).
Some common images of single dust particles were also analyzed, as shown in Figure 7. The presence of large spherical or spheroidal and relatively compact or porous morphology of the particle confirmed soot particles. This clearly reflected the influence of anthropogenic sources of fossil fuel combustion and biomass burning. The prevalence of irregular granular aggregates in the sample also showed the dominant sources of ground dust and construction dust at the respective sites (Zhao et al. 2010). Tubular and rhomboidal shapes were also observed in the samples because of the calcite and illite minerals originating from the crustal sources. In addition to the above, irregular and spherical shapes also signified the presence of biogenically originated particles (Pachauri et al. 2013). Dust particles showing fractal agglomeration, on the other hand, confirmed the occurrence of Ca2+-dominated reactions in the atmospheric dust (Gao and Anderson 2001).
Image data were analyzed for a different size range of diameter from <2.5 to >10 μm. The data showed a modal distribution with the maximum frequency of the particle diameter being <2.5 μm (Table 5); 90%, 89%, 89%, 85%, 85%, and 82% of the dustfall fluxes were present in <2.5-μm particle size range at CV, JNU, N-II, SMA, MN, and PC sites, respectively. This gives a clear indication of the role of anthropogenic activities in the crustal-derived dust fluxes (Srivastava et al. 2007). Thus, the pollution of inhalable particles (dynamic diameter less than 10 μm, PM10) considered as harmful to the human health (Reddington et al. 2015) cannot be ruled out in the Delhi–NCR region.
The present study showed the variability in the dustfall fluxes of oxidized Nr (NO3−) and reduced Nr (NH4+) species based on the ionic composition of the dustfall fluxes and its emission sources. Neutralization ratios and ion balance regression plots revealed the significance of relative abundance of Ca2+ and SO42− in Nr scavenging. Dominance of Ca2+ in the stoichiometric neutralization reactions resulted in a low median equivalence ratio of NO3−/SO42− (<1) in the dustfall fluxes at all the sites, owing to its affinity for SO42− over NO3−. The NH4+ being a weak base cation showed an overall weak scavenging that could be attributed to the predominance of Ca2+ as a major base cation in the dustfall. Spatial distribution showed that the Nr fluxes ranged from 1.24 ± 0.16 (site CV) to 16.45 ± 10.17 mg m−2 day−1 (site SMA) for NO3− and from being ND (site CV) to 16.33 ± 16 mg m−2 day−1 (SMA) for NH4+ depending upon the variability of its different emissions sources. Summertime meteorology of strong winds and unstable atmospheric conditions provided a constant trend in the time series data of Nr fluxes. Minor fluctuations could be attributed because of the frequent occurrence of rain and dust storm events during the sampling period. The morphological analysis of the dust samples further confirmed the relative contribution of crustal sources at each site with a modal distribution in the size fractions arising from the presence of maximum frequency of the particles in <2.5-μm range.
We sincerely thank the financial support received from Department of Science and Technology, New Delhi, to conduct this research work and the University Grants Commission, India. Analytical assistance provided by Advance Instrumentation Research Facility (AIRF), JNU, New Delhi, is gratefully acknowledged.