Current Issue

Delhi, the capital of India is situated with an altitude of 216 meters above sea level (latitude of 28°24′17′′N to 28°53′N and the longitude of 76°20′37′′E to 77°20′ 37′′E). The study area, Indira Gandhi International (IGI) airport is one of the busiest airports in South Asia. It is located 23 km south of Delhi (Fig. 1). Delhi has a single 40 km long corridor, which is oriented in the east -west direction for both incoming and outgoing flights. The number of flights operating per day is about 700 (Indian Express, 2011). The IGI airport is surrounded by major residential areas of Pahladpur and Palam in the north, Dwarka in the north-west, Vasant Vihar in the east, Mahipalpur and Vasant Kunj in the south east and Brijwasan is situated south of the IGI airport.

Fig. 1. 
Map of Delhi, India showing location of sampling area-the International Airport.

There are no large industries and heavy traffic roads in the vicinity of the IGI airport. However, it is served by subsidiary road emerging from National highway (NH-8) at a distance of 4-5 km from the airport. Thus, the direct impact of vehicular emission from the highway is minimal on the sampling locations. Three sampling locations were selected in the area surrounding the airport and one background location was chosen in a remote area. Site (LP) was selected near the landing point of the airport, site (T/I) was located near the taxi/idle point (south of the runway), while site (T-O) was situated near the take-off point of the airport. The background (BG) sampling location was located at a remote area at a distance of 10 km from the airport.

Sampling was done from December 2005 to February 2006, from the immediate periphery of the airport, using a stainless steel auger upto a depth of 5 cm. Soil samples were spread out and air-dried in dark, homogenized and then they were sieved through 2 mm diameter sieve. Representative samples were obtained after quartering and coning. Approximately 15 to 20 soil samples were collected from different spots from each sampling location (LP, T/I and T-O) in order to make one composite sample. Soil sampling was done twice monthly and per site 6 samples were collected for a period of 3 months. Thus, in the 3 month sampling period 24 samples were analyzed for 4 sites. Results are reported as an average of 6 samples per site.

Soil texture analysis showed sand as the predominant component (greater than 50%) of the airport soil. The range of organic carbon (OC) (Table 1) in the investigated soil samples was 0.92-1.33%. The pH values of soil were in the sub-alkaline range (7.73-7.94). The concentrations of the metals investigated in the airport and background soils are presented in Table 2.

Concentration of 8 metals (Cd, Fe, Cr, Pb, Zn, Cu, Ni and Mn) was determined in airport soils (sites LP, T/I, and T-O) and background soil (BG) in the <2mm fraction. The concentrations of metals measured in soil samples of the airport and background sites are presented in Table 2. The overall concentration of the airport was considered by averaging the values of LP, T/I and T-O sites. Results of two-way Anova illustrated that there were significant differences between all the four sampling sites with respect to the metal concentration (p≤0.05). On an average, for most of the metals (except for Fe and Mn), the airport sites were observed to have ~1.5 times higher concentration than BG.

Landing location (LP) showed relatively high concentration for most of the metals (Fig. 2). Hence, emissions from aircraft during landing and take-off may be responsible for such enrichment at LP relative to T-O. This may be due to the fact that particulate emissions are strongly dependent on the power settings of the engine (Kalivoda and Feller, 1995). Particulate emissions are higher at low engine powers because combustion efficiency is low. During landing operations, engine is set at low power settings at which emissions are maximized due to incomplete combustion (Kesgin, 2006). Hence, the highest concentration at LP site should be due to its location near the landing point. Emissions tend to decrease as power setting increases, i.e. during take-off, which supported the minimum metal concentration at the sampling site near take-off point. In addition, LP is located downwind of the runway, receiving more depositional load, while T-O is in the upwind direction. Hence, lesser metal concentration at T-O site should be justified. Among the metals studied in the airport soils, Cd was found to have the minimum concentration (2.07-2.35 μg g^-1) (Table 2), while Fe had maximum concentration (4318-4441 μg g^-1). The average concentration loadings of the airport soil was Cd<Cu<Pb<Ni<Zn<Cr<Mn<Fe. The content of metals in soil is strongly correlated with grain size, if metal input remains constant (Sharma et al., 1999). In soils, Fe-oxides are the main absorbers of heavy metals (Xia et al., 2007). High values of Fe concentration observed in the study area are suspected to be associated with the Fe accumulation because of desilification. Other studies have also reported that high Fe concentrations are in soil due to natural geochemical processes (Salonen et al., 2007; Micó et al., 2006). High Mn concentration should be attributed to the geology of the area, due to the occurrence of Mn rich sedimentary rocks throughout the south western part of Delhi. This observation is in agreement with studies of Kaur and Rani (2006). The pH values in the airport soils were in the sub-alkaline range of 7.73-7.94. Thus, the metals are expected to remain largely immobile after deposition. The metals will be specifically adsorbed to Fe-oxides and clay minerals which become deprotonated due to high pH-values and tend to form stable hydroxo-metal complexes with metals under such conditions (Gocht et al., 2001).

Fig. 2. 
Variation of metal concentration levels in different sampling sites (except Fe & Mn).

Metal pollution from aircraft may rise through emissions from combustion of fuel additives (WHO, 2005). Pb, Mn, Fe, Zn, and Cu are commonly used as additives in aviation fuels (Collins et al., 2002) to boost the octane number, improve combustion, suppress smoke, and reduce NO_X emissions. Ni and Zn emissions may increase due to Ni-Zn alkaline storage batteries (Dell, 1996). High Cr concentration may be due to abrasion of primers applied to aircraft surfaces to prevent corrosion (a human carcinogen) (Carlton, 2003; Dell, 1996). Currently, aviation gasoline is the one of the fuels containing lead additive, which is used to boost the octane of aviation gasoline (US EPA, 2008).

The metal concentrations observed in this study were compared with these of literature in many other airport soils (Table 3). High Cd concentrations found in the present study have also been reported around Zagreb airport, Croatia; Tribhuban airport, Kathmandu and Thessaloniki airport, Greece; and Queen Alia International airport, Jordan. High Cd concentration could arise due to wear and tear of aircraft wheels and brakes (Czarnowska, 1999; de Miguel et al., 1997). Moreover, from Ni-Cd alkaline storage batteries are used as an emergency flight power source (Dell, 1996).

In India, as soil standards are not available for metals, the values in this study were compared with Dutch target values (Table 5). All the metal concentrations were below the guidance values, except Cd, Cr and Ni. Cd concentration in this study area was ~3 times higher than the Dutch standards.

Pearson correlation test was made to assess the relationship between 8 metals, OC, and grain sizes. The results of this analysis are reported in Table 4. OC did not show significant correlation with Cd, Cr, and Ni. This suggests that aerial deposition patterns (rather than an equilibrium situation between organic matter and other phases) probably have influenced these metal concentrations. Therefore, it supports the hypothesis that the metal burden of Cd, Cr and Ni the airport soils result from atmospheric concentration (Stalikas et al., 1997). This could be ascribed to the absence of metal accumulation in humic material and the occurrence of airborne carbonaceous particles formed by combustion (Mielke et al., 2004).

Significant correlation of OC with clay fraction suggests that grain size influences OC content. Thus, the lower the grain size, the higher the OC content. Significant correlation of OC with Fe, Mn, and Zn implies that soil properties influence these metal concentration and that these metals are probably derived from natural weathering. Zinc tends to form complexes with organic matter and mostly associates with organic matter through adsorption and/incorporation into the biologically resistant humic component in soils (Achterberg et al., 1997). In addition, the strong positive correlation with Fe and Mn indicates the probable adsorption of these metals on to the oxyhydroxides of Fe and Mn (Stalikas et al., 1997). The correlation of these metals suggests that these metals may have the same partitioning to the soil and mineralogical background. Nevertheless, good correlation among Cu, Pb and Zn suggests that significant part of these metals also contributed from automobiles and airplane exhaust. Likewise, there is a good correlation between for Pb and Cu with OC (Table 4), suggesting the high affinity of these metals for organic ligands (Balasoiu, 2001).

To assess the origin of metals and their potential link to anthropogenic activities, geoaccumulation index (Igeo) was calculated. It is a quantitative measure to evaluate metal pollution of the soil matrix. It was originally defined by Müller (1979) in order to determine contamination in sediments and soils by comparing current concentrations with pre-industrial levels. It can be calculated by the following equation:

where Cn is the measured concentration of the metal (n) in the <2 mm fraction of soil, Bn is the geochemical background value in average shale (Turekian and Wedepohl, 1961) of metal n, and 1.5 is the background matrix correction factor due to lithogenic effects. Müller distinguished seven classes of I-geo (class 0 to 6) (Müller, 1981) where the highest class (class 6) reflects 100-fold enrichment above the background values.

The results of I_geo values determined in this study are shown in Table 5. The I-geo values of Pb and Cd were 0.30 and 2.32, respectively (Table 5). It thus indicates contamination in the airport soil. According to the Müller classification, airport soils of Delhi belonged to the pollution class 1 and 2 with respect to Pb and Cd, respectively. Thus, the study area showed high Cd contamination.

Pollution load index (PLI) for a particular site has also been evaluated following the method proposed by Tomilson et al. (1980). This parameter can be expressed as:

PLI=(CF1×CF2×CF3×…×CFn)^1/n

where n is the number of metals (six metals were considered except Fe and Mn) and CF is the contamination factor. The contamination factor can be calculated from the following relation:

Background values were calculated using the average shale value (Turekian and Wedepohl, 1961). The PLI for the airport site was found to be 1.34, while that for BG site it was 0.48. Thus, it can be concluded that the airport site is ~3 times more contaminated than the background site.

Principal Component Analysis (PCA) is a source apportionment tool that can be used to reduce a set of original variables and to extract a small number of latent factors (principal components, PCs) for analyzing relationships among the observed variables. By determining the PCA component loadings, communalities, and eigen values, the variables related to specific geochemical or environmental sources or processes were identified. In PCA analysis, Varimax normalized rotation was applied to the axes which maximised the variance of the components.

To obtain more insight into the data structure, PCA was performed to identify the main sources influencing the metal concentration in the soil in the immediate periphery of the airport area. Results obtained by PCA are given in Table 6. Concentrations of 8 metals were chosen as active variables. The majority of the variance (85.6%) of the scaled data was explained by four eigen vectors known as principal components. The first principal component (PC1) explained 26.7% of total variance, the second (PC2), the third (PC3), and the fourth (PC4) principal component explained 23.4%, 21.1% and 14.4% of the total variances, respectively.

The first component (PC1) was predominately weighted in metals; Fe, Mn and Zn. These results showed that the metals (Fe, Mn, and Zn) in the airport soils were associated with metal industry and tire wear particles (Bennett et al., 2011; Councell et al., 2004). The second component (PC2) had high factor loadings for metals, Cu, Pb, and Zn. All these metals are anthropogenically derived. Pb can originate from industries like battery, paint particles, and fuels with additives such as anti-knocking agent (Lovestead and Bruno, 2009). The third component (PC3) had high factor loadings of Cr and Ni. The patterns derived by PC2 and PC3 are suspected to be derived from mobile sources like aircraft and automobile emissions, respectively, although it is yet difficult to critically distinguish from each other. The fourth component is heavily weighed in Cd, probably reflecting aircraft emission (from aircraft batteries) (Kulin, 1997). Overall, it can be inferred that the primary source of metals in the soils of airport area is of anthropogenic origin such as vehicular and aircraft emissions.