ARTICLE IN PRESS Ecotoxicology and Environmental Safety 66 (2007) 258–266 www.elsevier.com/locate/ecoenv Heavy metal contamination of soil and vegetables in suburban areas of Varanasi, India Rajesh Kumar Sharmaa, Madhoolika Agrawala,, Fiona Marshallb a Ecology Research Laboratory, Department of Botany, Banaras Hindu University, Varanasi 221005, India b SPRU, The Freeman Centre, University of Sussex, Brighton BN1 9QE, UK Received 7 July 2005; received in revised form 1 November 2005; accepted 23 November 2005 Available online 7 February 2006 Abstract Heavy metal contamination of soil resulting from wastewater irrigation is a cause of serious concern due to the potential health impacts of consuming contaminated produce. In this study an assessment is made of the impact of wastewater irrigation on heavy metal contamination of Beta vulgaris (palak); this is a highly nutritious leafy vegetable that is widely cultivated and consumed in urban India, particularly by the poor. A field study was conducted at three major sites that were irrigated by either treated or untreated wastewater in the suburban areas of Varanasi, India according to normal practice. Samples of irrigation water, soil, and the edible portion of the palak (Beta vulgaris L. var All green H1) were collected monthly during the summer and winter seasons and were analyzed for Cd, Cu, Zn, Pb, Cr, Mn, and Ni. Heavy metals in irrigation water were below the internationally recommended (WHO) maximum permissible limits set for agricultural use for all heavy metals except Cd at all the sites. Similarly, the mean heavy metal concentrations in soil were below the Indian standards for all heavy metals, but the maximum value of Cd recorded during January was higher than the standard. However, in the edible portion of B. vulgaris, the Cd concentration was higher than the permissible limits of the Indian standard during summer, whereas Pb and Ni concentrations were higher in both summer and winter seasons. Results of linear regression analysis computed to assess the relationship between individual heavy metal concentration in the vegetable samples and in soil showed that Zn in soil had a positive significant relationship with vegetable contamination during winter. Concentrations of Cd, Cu, and Mn in soil and plant showed significant positive relationships only during summer. Concentration of Cr and Pb during winter season and Zn and Ni during summer season showed significant negative relationships between soil and plant contamination. The study concludes that the use of treated and untreated wastewater for irrigation has increased the contamination of Cd, Pb, and Ni in edible portion of vegetables causing potential health risk in the long term from this practice. The study also points to the fact that adherence to standards for heavy metal contamination of soil and irrigation water does not ensure safe food. r 2006 Elsevier Inc. All rights reserved. Keywords: Wastewater; Irrigation; Heavy metals; Accumulation; Contamination; Beta vulgaris 1. Introduction It is known that serious systemic health problems can develop as a result of excessive accumulation of dietary heavy metals such as Cd, Cr, and Pb in the human body (Oliver, 1997). Heavy metals are extremely persistent in the environment; they are nonbiodegradable and nonthermodegradable and thus readily accumulate to toxic levels. Heavy metals can accumulate in the soil at toxic levels due Corresponding author. Fax: +91 542 2368156. E-mail address: madhoo58@yahoo.com (M. Agrawal). 0147-6513/$ - see front matter r 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ecoenv.2005.11.007 to the long-term application of wastewater (Bohn et al., 1985). One important dietary uptake pathway could be through crops irrigated with contaminated wastewater. Soils irrigated by wastewater accumulate heavy metals such as Cd, Zn, Cr, Ni, Pb, and Mn in surface soil. When the capacity of the soil to retain heavy metals is reduced due to repeated use of wastewater, soil can release heavy metals into ground water or soil solution available for plant uptake. In suburban areas, the use of industrial or municipal wastewater is common practice in many parts of the world (Feigin et al., 1991; Urie, 1986), including India (Singh ARTICLE IN PRESS R. Kumar Sharma et al. / Ecotoxicology and Environmental Safety 66 (2007) 258–266 et al., 2004). Access to adequate water for irrigation is a matter of increasing concern in India. To face the growing demand for irrigation water, nonconventional resources are often used. Important sources of heavy metals in wastewater are urban and industrial effluents, deterioration of sewerage pipe and treatment works, and the wear of household plumbing fixtures. Wastewater irrigation is known to contribute significantly to the heavy metal content of soils (Mapanda et al., 2005; Nan et al., 2002; Nyamangara and Mzezewa, 1999; Singh et al., 2004). Other sources of heavy metal contamination of agricultural soil are sewage sludge, fertilizers, and pesticides (Alloway and Ayres, 1993; Ross, 1994). A number of previous studies from developing countries have reported heavy metal contamination in wastewater (Cao and Hu, 2000; Mapanda et al., 2005; Nyamangara and Mzezewa, 1999; Singh et al., 2004) and wastewaterirrigated soil (Cao and Hu, 2000; Mapanda et al., 2005; Nan et al., 2002; Nyamangara and Mzezewa, 1999; Singh et al., 2004). However, there are very few empirical data from India for heavy metal contamination of soil and irrigation water and its transfer to vegetable crops. The present study was carried out around Varanasi, India, where wastewater has been commonly used for irrigation of periurban crops for several decades. The objective of the present report is to quantify spatial and temporal variations in the total concentrations of Cd, Zn, Cr, Mn, Cu, Pb, and Ni in irrigation water, soil, and Beta vulgaris (var. All green H1, a leafy vegetable used for iron supplement, grown widely in the suburban region of Varanasi) from sites having long-term use of wastewater for irrigation of agricultural land. The concentrations in soil, crops, and water are compared with established safe limits. This provides a basis for guiding further activities aimed at preventing excessive exposure of humans through monitoring and control of irrigation water and/or amelioration of uptake to crops. 2. Materials and methods The field study took place in the urban fringe area of Varanasi city, situated in the Eastern Gangetic plain (251180 N latitude and 831 010 E longitude) of northern India. A reconnaissance survey was conducted in urban and suburban areas of Varanasi to identify the locations where wastewater from industries and municipal/domestic sewage is currently used for irrigation of vegetable crops. Interviews with government authorities and farmers helped to identify areas where wastewater irrigation has been common practice for many years, and where the irrigation water can be clearly sourced to treated or untreated industrial effluent. Three major vegetables production sites, Dinapur, Shivpur, and Lohta, were selected (Fig. 1). The soil properties of these sites were evaluated for pH, conductivity, organic matter, and NO3–N levels. The most common vegetable grown at all the study sites is palak (Beta vulgaris L. var. All green H1) throughout the year. Other seasonally produced vegetables are radishes (Raphanus sativus), tomatoes (Lycopersicum esculentum), lady’s finger (Abelmoschus esculentus), cauliflower (Brassica oleracea var. acephala), cabbage (Brassica oleracea var. capitata), and carrots (Daucus carota). These vegetable crops are mainly produced for home consumption and sale to residential areas of urban and suburban region of Varanasi. 259 At Dinapur, a sewage treatment plant (DSTP) of 80 MLD capacity was installed in 1986, where wastewater containing sewage and effluents discharged from various industries such as fabric printing, batteries, and paint, in urban areas of Varanasi is treated before discharge into the Ganges River. Farmers use this treated wastewater directly for irrigation via channels off the main outlet towards the fields. At Dinapur, five microsites were identified as D1, D2, D3, D4, and D5 (Fig. 1). Water samples were taken from D1 and D4 sites, while water, soil, and vegetable samples were taken from D2, D3, and D5 sites. Shivpur is a major industrial and residential area of Varanasi, having several small-scale industries such as fabric printing and dyeing, food processing, and electrical cables and paint manufacturing industries. Industrial effluents, in addition to untreated domestic sewage, flow through open drains in some areas and are directly used for irrigation of vegetables. Irrigation is conducted by lifting the water by pumps or by directly diverting from the drain. Three microsites, S1, S2, and S3, were identified in Shivpur area (Fig. 1). Water and soil samples were taken from all the sites, whereas vegetable samples were taken from the S2 site only. The Lohta site receives untreated sewage and industrial effluents from more than 100 industries located upstream of the drain at Chandpur, Maruadih, Lahartara, and Lohta industrial estates. The main industries are fabric printing, newspaper printing, and various manufacturing industries including iron gates, window nets, table fans, bicycle tires, and heavy agricultural equipment. The site also receives treated wastewater from the treatment plant of a large diesel locomotive works that manufactures diesel engines. Four microsites, L1, L2, L3, and L4, were identified in Lohta area. Water and soil samples were collected from all the sites, whereas vegetable samples were collected from the L4 site only (Fig. 1) 3. Sampling and analysis Water and soil samples were collected from the sites monthly from August 2003 to June 2004. However, vegetable and soil samples were taken randomly across the field during winter (December–January) and summer (April–May) seasons. Three replicate polyethylene bottles (acid-washed) of capacity 100 mL were immersed one by one at an interval of 15 s in an open drain that was being used for irrigation purposes and immediately after filling, 1 mL of concentrated HNO3 was added to the water to avoid microbial utilization of heavy metals. The bottles were brought back to the laboratory and digestion was completed within a week. Soil samples were collected in triplicate from the field receiving wastewater regularly for irrigation. A monolith of 10  10  15 cm was dug for the collection of soil. Soil samples were air dried, crushed, and passed through 2-mm-mesh sieve and stored at ambient temperature before analysis of soil properties and concentrations of heavy metals. Vegetable samples were also collected in triplicate (2 kg each) from the same field simultaneously. Care was taken to get samples of the same varieties and age group from different selected sites. Replicate samples were washed using clean water and then separately oven-dried at 80 1C till constant weight was achieved. The samples were then crushed separately through a steel grinder and the crushed material was passed through 2-mm sieve. The sieved samples were kept at ambient temperature before analysis. The soil pH was measured in suspension of 1:5 soil to water using a glass electrode (Orion, EA, 940 USA). In the ARTICLE IN PRESS 260 R. Kumar Sharma et al. / Ecotoxicology and Environmental Safety 66 (2007) 258–266 Fig. 1. Map of study sites in suburban region of Varanasi. same suspension, conductivity and concentration of NO3–N were measured using conductivity meter (Systronics, India) and NO3–N electrode (Orion, EA 940 USA), respectively. The organic carbon was determined using Walkley and Black’s method (Allison, 1986). The irrigation water samples (50 mL) were digested with 10 mL of concentrated HNO3 at 80 1C until the solution became transparent (APHA, 1985). The solution was filtered through Whatman No. 42 filter paper and the solution was diluted to 50 mL with distilled water. For soil and vegetable samples, 0.5 g of dried samples were digested with 15 mL of HNO3, H2SO4, and HClO4 in 5:1:1 ratio at 80 1C until a transparent solution was obtained (Allen et al., 1986). The solution was filtered through Whatman No. 42 filter paper and the solution was diluted to 50 mL with distilled water. The concentrations of heavy metal in filtrate of water, soil, and plant samples were determined with an atomic absorption spectrophotometer (Perkin-Elmer model 2130, USA) fitted with a specific lamp of a particular metal using appropriate drift blanks. Quality control measures were taken to assess contamination and reliability of data. Blank and drift standards (Sisco Research Laboratories Pvt. Ltd., India) were run after five determinations to calibrate the instrument. The coefficient of variation of replicate analysis was determined for different determinations and for precision of analysis. Variations were found to be less than 10%. Precision and accuracy of analysis were also ensured through repeated analysis of samples against National Institute of Standards and Technology standard reference material (SRM 1570) for all the heavy metals. The results were found to be within 72% of the certified value. The data were subjected to two-way analysis of variance (ANOVA). Mean, median, and range was also used to assess the contamination levels of heavy metals in soil and irrigation water. Linear regression analysis was used to determine the relationship between heavy metal concentrations in soil and plant during different seasons. 4. Results Soil pH and conductivity were higher during summer than in the winter season at all sites (Table 1). Organic ARTICLE IN PRESS R. Kumar Sharma et al. / Ecotoxicology and Environmental Safety 66 (2007) 258–266 261 Table 1 Physicochemical properties of soil collected from different study sites during winter and summer Study sites Dinapur Shivpur Lohta Organic matter (g kg1) Electrical conductivity (mS cm1) pH NO3–N (mg kg1) Summer Winter Summer Winter Summer Winter Summer Winter 95.8670.60 85.1770.60 94.5770.35 89.5770.75 80.6970.40 89.1470.20 0.3070.00 0.3270.01 0.3770.00 0.2070.00 0.2370.00 0.3670.00 8.5170.00 8.6370.00 8.6370.01 8.0270.02 8.0270.01 8.3670.01 17.1770.32 25.7770.09 34.1070.81 16.0670.06 15.3770.03 28.1070.06 Table 2 Results of two-way ANOVA for heavy metal concentrations in wastewater, soil, and Beta vulgaris L. plants in Varanasi, India Variations Wastewater Site Month Sites  months Soil Site Month Sites  months B. vulgaris Site Season Sites  seasons df Cd Zn Cr Mn Cu Pb Ni 2 5 10 2.527NS 9.335 1.888 10.397 6.894 3.124 10.755 5.476 2.202 1.572NS 6.454 3.358 4.585 4.140 4.015 6.639 98.508 5.047 2.220NS 19.167 5.687 2 5 10 22.201 331.219 86.738 16.595 4.491 4.365 25.894 5.726 10.678 23.212 12.959 18.618 7.817 4.101 3.116 7.671 29.570 15.616 2.771NS 40.623 17.967 2 1 2 32.436 361.920 18.206 208.044 17698.037 585.698 101.992 678.574 162.235 2166.173 5694.45 701.027 333.411 275.490 206.332 234.395 701.548 97.723 55.104 154.574 12.553 Note: NS, not significant; df, degree of freedom.  Po 0.01. Po 0.001.  Po 0.05. matter content also showed a similar trend. The value was highest at Dinapur during summer and lowest at Shivpur during winter (Table 1). NO3–N content did not vary significantly between winter and summer at Dinapur, but was significantly higher at Lohta and Shivpur during the summer as compared to the winter. The ANOVA showed significant variations in different soil properties due to sites and seasons. Site  season interaction was also significant except for organic matter. Heavy metal concentrations in irrigation water showed significant variations between sites and months and for the site  month interaction. However, the sitewise variation was not significant for Cd, Mn, and Ni (Table 2). The mean concentrations (mg L1) of heavy metals ranged from 0.01 to 0.02 for Cd, 0.09 to 0.23 for Zn, 0.03 to 0.09 for Cr, 0.12 to 0.15 for Mn, 0.04 to 0.11 for Cu, 0.08 to 0.10 for Pb, and 0.03 to 0.05 for Ni (Table 3). Among the heavy metals, Cu and Pb were higher at Dinapur, which received only treated wastewater, while sites receiving untreated wastewater (Lohta and Shivpur), Zn, Cr, and Mn were higher than at Dinapur (Table 3). ANOVA tests showed that variations in concentrations of heavy metals in soil were significant due to site, month, and site  month interaction, except for Ni, which did not show significant sitewise variation (Table 2). The concentration (mg kg1 dry soil) of heavy metals in soil of the study sites ranged between 0.55 and 8.85 for Cd, 14.23 and 387.78 for Zn, 13.40 and 679.89 for Cr, 0.36 and 339.36 for Mn, 2.55 and 203.45 for Cu, 0.46 and 44.50 for Pb, and 2.00 and 34.45 for Ni during different months (Table 4). The highest mean concentrations were recorded for Mn at all sites, followed by Zn, Cr, Cu, Pb, and Ni, and the minimum was observed for Cd. Chromium concentration was, however, second highest at Lohta. The highest mean concentrations of Zn, Cr, Cu, and Pb were recorded at Lohta and then Shivpur, followed by Dinapur. Mn concentration was in decreasing order from Lohta4Dinapur4Shivpur, but an opposite trend was found for Cd (Table 4). The concentrations of Cd, Zn, Cr, and Mn were higher during summer, whereas Cu, Pb, and Ni were higher during winter season in B. vulgaris receiving wastewater for irrigation (Fig. 2). Both sites and seasons individually and in combination showed significant effects on heavy metal concentrations in B. vulgaris (Table 2). In terms of absolute concentration, Mn was highest in B. vulgaris, followed by Zn, and then Cr, Cu, Pb, and Cd, and the lowest concentration was observed for Ni during the summer season. In the winter season, however, the trend from highest to lowest concentration was Mn then Cu, Pb, Cr, Zn, Ni, and Cd. The mean concentrations in B. vulgaris (mg kg1 dry weight) varied from 0.50 to 4.36 for Cd, 2.22 ARTICLE IN PRESS R. Kumar Sharma et al. / Ecotoxicology and Environmental Safety 66 (2007) 258–266 262 Table 3 Heavy metal concentrations (mg L1) in wastewater used for irrigation in suburban region of Varanasi, India Suburban area Cd Dinapur (n ¼ 84) Mean Median Range Lohta (n ¼ 60) Mean Median Range Shivpur (n ¼ 54) Mean Median Range Cr Mn Cu Pb Ni 0.09 0.07 0.00-0.056 0.03 0.02 0.00-0.82 0.12 0.09 0.00–1.08 0.11 0.05 0.00–1.54 0.10 0.06 0.00–0.37 0.05 0.03 0.00–0.37 0.17 0.09 0.02–1.25 0.09 0.06 0.00–0.30 0.15 0.14 0.01–0.81 0.06 0.03 0.00–0.49 0.08 0.05 0.00–0.29 0.05 0.04 0.00–0.19 0.01 0.01 0.00–0.03 0.23 0.15 0.01–0.86 0.04 0.04 0.00–0.14 0.15 0.10 0.00–0.67 0.04 0.03 0.00–0.20 0.08 0.05 0.00–0.31 0.03 0.01 0.00–0.14 0.01 2.0 0.1 0.2 0.2 0.5 0.2 0.02 0.01 0.00–0.10 0.02 0.01 ND–0.23 WQ criteria for irrigationa a Zn Source: Pescod (1992). Table 4 Heavy metal concentration (mg kg1 dry weight) in wastewater irrigated soil of three major vegetable production areas at suburban region of Varanasi, India Suburban area Cd Zn Cr Mn Cu Pb Ni Dinapur (n ¼ 54) Mean Median Range 2.80 1.65 0.63–8.30 43.56 33.70 14.23–121.85 30.67 29.75 13.40–98.65 156.96 155.81 0.36–286.15 20.35 19.00 2.55–52.60 15.57 13.23 5.46–35.00 13.37 10.25 2.00–34.45 Lohta (n ¼ 66) Mean Median Range 2.26 1.50 0.69–6.95 92.57 65.82 17.89–387.78 172.75 70.39 18.05–679.89 191.51 182.88 40.20–339.36 42.03 20.35 5.45–203.45 19.51 15.53 6.50–38.90 14.52 11.55 2.45–32.30 Shivpur (n ¼ 53) Mean Median Range 2.69 1.45 0.55–8.85 87.89 65.89 20.75–223.56 79.27 29.15 13.55–500.00 145.74 159.56 21.75–256.80 33.50 21.30 8.90–107.10 18.35 14.96 0.46–44.50 15.61 11.23 4.03–32.25 3–6 300–600 n/a n/a 135–270 250–500 75–150 0.01–0.7 10–300 5–3000 100–4000 2–100 2–200 10–1000 Permissible limits of indian standarda Range of heavy metal concentrations in uncontaminated soilb Note: n/a, not available. a Source: Awashthi (2000). b Source: Bowen (1966). to 41.51 for Zn, 5.37 to 27.83 for Cr, 7.53 to 117.94 for Mn. 10.95 to 28.58 for Cu, 3.09 to 15.74 for Pb, and 1.81 to 7.57 for Ni (Fig. 2). Linear regression analysis was used to evaluate the relationship between individual heavy metal concentration in soil and in the edible leafy portion of B. vulgaris. The trend of metal concentrations in the plants would be a function of trace metal content in the soil and their absorption in plant tissues. The relationships between the individual heavy metals in soil and B. vulgaris are presented in Figs. 3 and 4. Significant positive correlations were found for Zn and Cu during winter and Cd, Cr, and Mn during summer season. Negative significant correlations were, however, found for Cr and Pb during the winter season and Zn and Ni during the summer season. The relationships for Cr, Mn, and Ni during winter season and Cu and Pb during summer season were not significant (Figs. 3 and 4). 5. Discussion The results of the field study showed that continuous application of treated and untreated wastewater led to elevated levels of heavy metals in the soil and edible portion of B. vulgaris. The concentrations of all the heavy metals showed spatial and temporal variations, which may be ascribed to the variations in heavy metal sources and the quantity of heavy metals discharged through the sewage and effluents in irrigation water. The mean concentrations of Cd, Cr, Ni, and Mn in treated wastewater of Dinapur recorded during the present study were similar to those in the earlier report of Singh et ARTICLE IN PRESS R. Kumar Sharma et al. / Ecotoxicology and Environmental Safety 66 (2007) 258–266 263 Fig. 2. Heavy metal (Cd, Zn, Cr, Mn, Cu, Pb, and Ni) concentrations (mg kg1 dry weight) in Beta vulgaris L. var. all green H1 during summer and winter seasons at Dinapur, Shivpur, and Lohta sites. al. (2004). However, the mean concentrations of Pb and Cu were higher and that of Zn was lower than the reported values of Singh et al. (2004). The comparison of mean heavy metal concentrations in treated and untreated wastewater from the data of other countries suggests that the values of the present study were manyfold lower than the levels observed at Harare, Zimbabwe (Mapanda et al., 2005; Nyamangara and Mzezewa, 1999), but higher than those recorded at Wodonga and Canberra, Australia (Smith et al., 1996). Cao and Hu (2000) in China also reported manyfold higher Cu concentration (12 mg L1) in wastewater than the present study value (0.11 mg L1). The metal concentrations in wastewater used for irrigation are above the recommended maximum value for Cd at Lohta and Dinapur sites (Table 3). Cr concentration was close to the permissible level at Lohta only (Table 3), but lower at the other sites. The elevated levels of Cd in irrigation water at Dinapur and Lohta may be due to effluents discharged from various heavy-metal-based industries such as the fabric printing (Cu, Cd, Zn, Ni, and Cr), battery (Pb, Mn, As, and Cd), and paint industries (Pb, Cd, Zn, Cu, and Pb). The higher Cr content at Lohta site is due to effluent discharged from more than 50 industries where chromium and its compounds are used as color and pigment, plating, and alloys for metal surface treatment and as catalysts, situated in the catchment area of drains discharging Fig. 3. Relationships between heavy metal (Cd, Zn, Cr, and Mn) concentrations (mg kg1 dry weight) in Beta vulgaris L. var. all green H1 and soil during summer and winter seasons. P ¼ concentration of heavy metal in plant; S ¼ concentration of heavy metal in soil; NS ¼ not significant. wastewater at Lohta. The diesel locomotive works, where chromium plating is carried out on a large scale, also discharges water into the drain after treatment. Significant positive correlations were observed between Pb and Ni (R2 ¼ 0:43; Po0:01), Pb and Cr (R2 ¼ 0:39; Po0:01), Pb and Zn (R2 ¼ 0:27; Po0:05), Pb and Cd (R2 ¼ 0:65; Po0:01), Ni and Cd (R2 ¼ 0:51; Po0:01), and Ni and Zn (R2 ¼ 0:29; Po0:05) at Shivpur and between Pb and Ni (R2 ¼ 0:32; Po0:05), Pb and Cr (R2 ¼ 0:47; Po0:01), Pb and Zn (R2 ¼ 0:43; Po0:01), and Zn and Cr (R2 ¼ 0:56; Po0:01) at Lohta, suggesting that these heavy metals in irrigation water originate from common sources. At Dinapur no strong correlation was found except between Cd and Cu (R2 ¼ 0:43; Po0:01) and between Cd and Pb (R2 ¼ 0:23; Po0:05), suggesting that Cd, Cu, and Pb are more common in sewage water, which also receives effluents from some small-scale industries. It is evident that treatment of wastewater is not helpful in elimination of heavy metals at Dinapur. The range of concentrations of Zn, Cu, Pb, and Ni in soil observed during the present study was below the official Indian standard (Awashthi, 2000). The maximum observed value of Cd was higher than ARTICLE IN PRESS 264 R. Kumar Sharma et al. / Ecotoxicology and Environmental Safety 66 (2007) 258–266 Fig. 4. Relationships between heavy metal (Cu, Pb, and Ni) concentrations (mg kg1 dry weight) in Beta vulgaris L. var. all green H1 and soil during summer and winter seasons. P ¼ concentration of heavy metal in plant; S ¼ concentration of heavy metal in soil; NS ¼ not significant. the Indian standard during January. Cd, Zn, and Cu concentrations were, however, higher than the values reported for typical uncontaminated soil (Bowen, 1966). The comparison of the data from the present study with earlier findings of Singh et al. (2004) at Dinapur, Varanasi suggested that the range of concentrations of Cd and Cr in soil was higher; however, Zn, Mn, Cu, Pb, and Ni were lower than the previously reported range. This variation may be ascribed to the differences in sites, frequency of soil collection, and crops under cultivation between the two studies. There were significant variations in the concentrations of heavy metals in soil between sites and seasons. This trend may be correlated with the soil properties that influence heavy metal availability at different sites. The higher values of soil pH, conductivity, organic matter, and NO3–N content observed during summer as compared to winter may be ascribed to a higher frequency of irrigation during summer associated with high temperature and low soil moisture availability. Higher soil pH during summer may have maintained higher concentrations of heavy metals in the topsoil layer owing to higher evaporation. The concentration of Cu (2.55–203.45 mg kg1) in soil observed in the present study is higher than concentrations reported in Zimbabwe by Tandi et al. (2004) (2.50–133.3 mg kg1) and Mapanda et al. (2005) (7–145 mg kg1). The concentrations of Cd, Cr, Ni, and Zn in soil observed during the present study were also higher than those reported at Zimbabwe (Mapanda et al., 2005). The relatively low heavy metal contamination in soil as compared with water and crops, and a lack of significant correlations between the concentrations of heavy metals in soil and wastewater at various sites, may be ascribed to the continuous uptake of heavy metals by plants during their growth and development. This is supported by the data of heavy metal concentrations in the edible portion of B. vulgaris. Significant positive correlations were observed between the concentration of Zn and Cu in soil and plant tissue during winter and Cd, Cr, and Mn in soil and plant tissue during summer season. Absorption and accumulation of heavy metals in plant tissue depend upon many factors, which include temperature, moisture, organic matter, pH, and nutrient availability. Soil properties influencing heavy metal availability varied significantly between the sites. The Dinapur and Lohta sites showed higher organic matter content than the Shivpur site. The Dinapur and Lohta sites also showed similar contamination levels for Cd, Zn, and Mn during the summer, which may be at least partially explained by the high organic matter. Organic complexing molecules of low molecular weight (LMW) serve as carriers of micronutrients. LMW has been shown to increase Cd uptake (Chen and Aviad, 1990), whereas the presence of organic matter has been reported to increase the uptake of Zn in the wheat plant (Rupa et al., 2003). The pH of the soil was close to 8 at all sites. The alkaline range of soil (48.0) is known to restrict the mobilization of heavy metals, thus reducing their uptake. However, the high nutrient input from irrigation water at these sites could result in relatively high growth rates and relatively high uptake of heavy metals as a result. The field data support this argument in that a higher yield of B. vulgaris was recorded at all three sites that were irrigated with wastewater as compared to the area receiving uncontaminated irrigation water. The NO3–N content was similar at all sites, while the concentration in summer was consistently higher than in winter. Long-term application of N is known to be associated with an increase in plant uptake and bioavailability of heavy metals (Nambiar and Ghosh, 1984). In the present study it seems that many soil factors such as pH, organic matter, nitrogen bioavailability, soil moisture, and temperature have interacted to impact on uptake. The uptake and accumulation of Cd, Zn, Cr, and Mn in B. vulgaris were higher during the summer season, whereas Cu, Ni, and Pb accumulated more during the winter season. It may be expected that during the summer season the relatively high decomposition rate of organic matter is likely to release heavy metals in soil solution for possible uptake by plants (McGrath et al., 1994). The higher uptake of heavy metals such as Cd, Zn, Cr, and Mn during the summer season may be due to high transpiration rates as compared to the winter season due to high ambient temperature and low humidity (Ingwersen and Streck, 2005). From Fig. 2 it can be predicted that LMW organic molecules act as a good carrier for Cd, Cr, Mn, and Zn but ARTICLE IN PRESS R. Kumar Sharma et al. / Ecotoxicology and Environmental Safety 66 (2007) 258–266 a poor carrier for Cu, Pb, and Ni during the summer season. During the winter season the concentrations of Cu, Ni, and Pb were higher in irrigation water as compared to summer season, which may have had a direct impact on uptake. For Zn and Mn the present results agreed with the observation of Gadallah (1994) for sunflower that during the summer season, concentrations of Zn and Mn were higher as compared to other seasons. The mean concentrations of all the heavy metals in B. vulgaris at Dinapur were considerably higher than the values reported by Singh et al. (2004) in vegetable crops from the same area. This may be ascribed to the considerable variation in heavy metal uptake by different types of vegetables. In the present study, samples of B. vulgaris only were collected, whereas Singh et al. (2004) analyzed a range of vegetable crops having different heavy metal absorption potentials. In the present study plants were harvested at maturity; no such information is available for the study of Singh et al. (2004), but if harvest occurred earlier, clearly less heavy metal absorption would have taken place. The comparison of mean values of heavy metals recorded in the plant material during the present study with Indian permissible limits (Awashthi, 2000) showed that Pb and Ni concentrations were higher both during summer and winter seasons, however, Cd was higher only in the summer season at all the sites during 2004. Mean concentrations of Zn, Cu, and Mn were below the permissible limits at all the sites. Cr concentration at Lohta was also higher than the permissible limit during summer, whereas at Dinapur it was approaching the limit. This study clearly showed that consumption of B. vulgaris by the urban and suburban population might pose health hazards due to Cd, Pb, and Ni contamination. Significant negative correlations were observed between heavy metal concentrations in soil and plant for Cr and Pb during winter and Zn and Ni during summer. This response clearly indicates antagonism between the metals during absorption. The results of linear regression analysis further showed that during the winter season the interaction of Cd and Zn was synergistic. Similar synergism has been reported by Hinsely et al. (1984) in the sewage-sludge-amended corn plants. Nan et al. (2002) also showed synergism of Cd and Zn absorption in fields irrigated by wastewater. In summer, however, Cd and Zn interacted antagonistically, as concentrations of Cd in soil and plant tissue were positively correlated and for Zn negatively correlated (Fig. 3). During summer, presumably due to moisture limitation and high temperature, Cd interfered with the Zn uptake. This trend agreed with a previous report of Zhou et al. (1994), where interaction of Cd and Zn resulted in reduction of Zn uptake with consequent increase of Cd uptake and bioaccumulation in rice plants grown in pots. 6. Conclusion The study concludes that irrigation by treated or untreated wastewater has increased the heavy metal 265 concentrations in soil and plants of receiving area. Cadmium concentration in irrigation water was found above the permissible limit set by WHO for irrigation of agricultural land at Dinapur and Lohta sites. Heavy metal concentrations in plants show significant spatial and temporal variations. Cd, Pb, and Ni were above the Indian permissible limits, while Zn and Cu were within limits in the edible portion of B. vulgaris. Cr concentration was above the permissible limit at Lohta site, receiving effluents from Cr based industries. The results of the present study further suggest that Zn, Cr, and Mn concentrations in plants are influenced by seasonal variations, whereas Cd, Cu, Pb, and Ni did not show any change in uptake pattern due to seasonal variations. The consumption of Cd, Pb, and Ni-contaminated portion of B. vulgaris plants by suburban people of Varanasi may pose health hazards. An important issue is that the contamination levels were frequently higher than permissible limits in the plant tissue, at the same sites as water and soil samples that comply with established safe standards. This has important implications for policy in that programmes aimed at monitoring and controlling heavy metal concentrations in irrigation water sources will not necessarily result in acceptable levels in vegetables. Policies and programs need to be adapted so that local edaphic conditions and agricultural practices are taken into account, and appropriate local measures developed for ameliorating heavy metal uptake by crops for a given set of local conditions. These measures need to be regularly reviewed to take into account factors such as the accumulation of heavy metals in the topsoil over time. Acknowledgments The authors gratefully acknowledge that the present research work is an output of a collaborative research project entitled ‘‘Contaminated irrigation water and food safety for the urban and peri-urban poor: Appropriate measures for monitoring and control from field research in India and Zambia,’’ which is funded by Department for International Development (DFID), United Kingdom, for the benefit of developing countries (R8160). The present study work was not conducted on human or experimental animals where national or institutional guidelines are used for the protection of human subjects and animal welfare. In the present study, national and international norms of heavy metals in irrigation water, soil, and vegetables are used to compare contamination levels found with standards prescribed for the protection of human health. References Allen, S.E., Grimshaw, H.M., Rowland, A.P., 1986. Chemical analysis. In: Moore, P.D., Chapman, S.B. (Eds.), Methods in Plant Ecology. Blackwell Scientific Publication, Oxford, London, pp. 285–344. Allison, L.E., 1986. Organic carbon. In: Klute, A. (Ed.), Methods of Soil Analysis, Part I. American Society of Agronomy, Madison, WI, pp. 1367–1381. ARTICLE IN PRESS 266 R. Kumar Sharma et al. / Ecotoxicology and Environmental Safety 66 (2007) 258–266 Alloway, B.J., Ayres, D.C., 1993. Chemical Principles of Environmental Pollution. Blackie Academic and Professional. An imprint of Chapman and Hall, Oxford, UK. APHA (American Public Health Association), 1985. Standard Methods for the Examination of Water and Wastewater. American Public Health Association, Washington, DC. Awashthi, S.K., 2000. Prevention of Food Adulteration Act no 37 of 1954. Central and State Rules as Amended for 1999, third ed. Ashoka Law House, New Delhi. Bohn, H.L., McNeal, B.L., O’ Connor, A.G., 1985. Soil Chemistry, second ed. Wiley-Inter Science Publications, New York, USA. Bowen, H.J.M., 1966. Trace Elements in Biochemistry. Academic Press, New York. Cao, Z.H., Hu, Z.Y., 2000. Copper contamination in paddy soils irrigated with wastewater. Chemos 41 (1–2), 3–6. Chen, Y., Aviad, T., 1990. Effect of humic substances on plant growth. In: McCarthy, R., Clapp, C.E., Malcolm, R.L. (Eds.), Humic Substances in Soil and Crop Sciences. American Society of Agronomy and Soil Science Society of America, Madison, WI, pp. 161–186. Feigin, A., Ravina, I., Shalhevet, J., 1991. Irrigation with Treated Sewage Effluent. Springer, Berlin. Gadallah, M.A.A., 1994. Effects of industrial and sewage wastewaters on the concentration of soluble carbon, nitrogen and some mineral elements in sunflower plants. J. Plant Nutr. 17 (8), 1369–1384. Hinsely, T.D., Redborg, K.E., Pietz, R.C., Ziegler, E.L., 1984. Cadmium and zinc uptake by corn (Zea mays L.) with repeated applications of sewage sludge. J. Agric. Food Chem. 32, 155–163. Ingwersen, J., Streck, T., 2005. A regional scale study on the crop uptake of cadmium from sandy soils: measurement and modeling. J. Environ. Qual. 34, 1026–1035. Mapanda, F., Mangwayana, E.N., Nyamangara, J., Giller, K.E., 2005. The effect of long-term irrigation using wastewater on heavy metal contents of soils under vegetables in Harare, Zimbabwe. Agric. Ecosyst. Environ. 107, 151–165. McGrath, S.P., Chang, A.C., Page, A.L., Wilter, E., 1994. Land application of sewage sludge: scientific perspectives of heavy metal loading limits in Europe and the United States. Environ. Rev. 2, 108–118. Nambiar, K.K.M., Ghosh, A.B., 1984. Highlights of Research of a Long Term Fertilizer Experiment in India (1971–1982). LTFE Research Bulletin No. 1. Indian Agric. Inst., New Delhi, pp. 1–100. Nan, Z., Li, J., Zhang, Cheng, G., 2002. Cadmium and zinc interaction and their transfer in soil–crop system under actual field conditions. Sci. Total Environ. 285, 187–195. Nyamangara, J., Mzezewa, J., 1999. The effects of long-term sewage sludge application on Zn, Cu, Ni and Pb levels in clay loam soil under pasture grass in Zimbabwe. Agric. Ecosyst. Environ. 73, 199–204. Oliver, M.A., 1997. Soil and human health: a review. Eur. J. Soil Sci. 48, 573–592. Pescod, M.B., 1992. Wastewater Treatment and Use in Agriculture. FAO Irrigation and Drainage Paper 47, Food and Agriculture Organization of the United Nations, Rome, 125pp. Ross, S.M., 1994. Sources and forms of potentially toxic metals in soil–plant systems. In: Ross, S.M. (Ed.), Toxic Metals in Soil–Plant Systems. Wiley, England, UK, pp. 3–26. Rupa, T.R., Sinivas, R.C., Subha, R.A., Singh, M., 2003. Effects of farmyard manure and phosphorus on Zn transformation and phytoavailability in two altisol of India. Bioresour. Technol. 87 (3), 279–288. Singh, K.P., Mohon, D., Sinha, S., Dalwani, R., 2004. Impact assessment of treated/untreated wastewater toxicants discharge by sewage treatment plants on health, agricultural, and environmental quality in wastewater disposal area. Chemos 55, 227–255. Smith, C.J., Hopmans, P., Cook, F.J., 1996. Accumulation of Cr, Pb, Cu, Ni and Cd in soil following irrigation with treated urban effluent in Australia. Environ. Pollut. 94 (3), 317–323. Tandi, N.K., Nyamangara, J., Bangira, C., 2004. Environmental and potential health effects of growing leafy vegetables on soil irrigated using sewage sludge and effluent: a case of Zn and Cu. J. Environ. Sci. Health B 39, 461–471. Urie, D.H., 1986. The status of wastewater irrigation of forest. In: Cole, D.W., Henry, C.L., Nutter, W.L. (Eds.), The Forest Alternative for Treatment and Utilization of Municipal and Industrial Wastes. University of Washington Press, Seattle, WA, USA, pp. 26–40. Zhou, Q., Wu, Y., Xiong, X., 1994. Compound pollution of Cd and Zn and its ecological effect on rice plant. Chin. J. Appl. Ecol. 5, 428–441 (in Chinese).