risk assessment of heavy metals and their source distribution in waters of a contaminated industrial...

RESEARCH ARTICLE

Risk assessment of heavy metals and their source distributionin waters of a contaminated industrial site

A. Keshav Krishna & K. Rama Mohan

Received: 14 June 2013 /Accepted: 11 November 2013 /Published online: 26 November 2013# Springer-Verlag Berlin Heidelberg 2013

Abstract Industrially contaminated sites with hazardous ma-terials are a priority and urgent problem all over the world.Appropriate risk assessment is required to determine healthrisks associated with contaminated sites. The present studywas conducted to investigate distribution of potentially haz-ardous, heavy metal (As, Cd, Cr, Cu, Ni, Pb and Zn) concen-trations in surface and groundwater samples collected duringsummer (pre-monsoon) and winter (post-monsoon) seasonsfrom an industrially contaminated site, Hyderabad, India, withpotential source of metal contamination because of industrialeffluents and usage of pesticides in agriculture. Heavy metal(HM) concentrations were analysed by using inductivelycoupled plasma-mass spectrometer and were compared withpermissible limits set by the World Health Organisation. Dataobtained was treated using multivariate statistical approacheslike R-mode factor analysis (FA), principal component analy-sis, cluster analysis, geoaccumulation index, enrichment fac-tor, contamination factor and the degree of contamination.

Health risk assessment like chronic daily intake (CDI) andhazard quotient (HQ) were also calculated. Relatively highlevels were noted in surface water with average concentrationsduring summer and winter seasons showing 16.13 and 11.83for As, 7.91 and 1.64 for Cd, 88.33 and 32.90 for Cr, 58.11and 28.26 for Cu, 53.62 and 69.96 for Ni, 173.8 and 118.6 forPb, and 2,943 and 1,889 μg/L for Zn. While in groundwater,the mean metal levels during two seasons were 18.18 and 3.76for As, 1.67 and 0.40 for Cd, 29.40 and 5.15 for Cr, 17.03 and4.19 for Cu, 25.4 and 6.09 for Ni, 81.7 and 2.87 for Pb and953 and 989 μg/L for Zn, respectively. FA identified twofactors with cumulative loadings of F1—60.82 % and F2—76.55 % for pre-monsoon surface water and F1—48.75 % andF2—67.55 % for groundwater. Whereas, three factors withcumulative loadings of F1—39.13 %, F2—66.60 % and F3—81.01 % for post-monsoon surface water and F1—50.31 %,F2—66.18 % and F3—81.54 % for groundwater. The healthrisk assessment like CDI and HQ indices with increased levelsof hazardous elements in the surface and groundwater weresafe for drinking purposes provided some water treatmentmethodologies are adopted.

Keywords Heavymetals . Surface water . Groundwater .

Multivariate analysis . PCA . CA . CDI . HQ

Introduction

Water contamination poses serious health risks. The contam-ination of water is directly proportional to the degree ofenvironmental degradation which can threat the very basisof human survival. Globally fresh water resources are imper-illed not only by over exploitation but also by ecologicaldegradation. Discharge of untreated effluents, waste and agri-cultural run-off from fields are some of the ways in which thewater is contaminated. Developed countries suffer from

Responsible editor: Philippe Garrigues

Electronic supplementary material The online version of this article(doi:10.1007/s11356-013-2359-5) contains supplementary material,which is available to authorized users.

A. K. Krishna :K. R. MohanCSIR-National Geophysical Research Institute, Habsiguda,Hyderabad 500007, India

A. K. Krishna (*)CSIR-National Geophysical Research Institute, Habsiguda,Hyderabad 500606, Indiae-mail: [email protected]

A. K. Krishnae-mail: [email protected]

A. K. Krishnae-mail: [email protected]

Environ Sci Pollut Res (2014) 21:3653–3669DOI 10.1007/s11356-013-2359-5

http://dx.doi.org/10.1007/s11356-013-2359-5

problems of chemical discharge into the water sources, mainlygroundwater, while developing countries face problems ofagricultural runoff in water sources (Kalra 2012). Some stud-ies have shown that fertilisers in water supplies may causecancer. In china, research on population exposed to nitrates intheir drinking water, suggested links between nitrate contam-ination and stomach–lever cancer where the chemicals used inwater pipes may contaminate the drinking water after it hasbeen treated. The tar, asphalt, iron, coal, PVC, concrete,asbestos and Pb are all potential sources of post-treatmentcontamination. Furthermore, it has been recognised that urbanstorm water runoff as a major source of pollutants whichcontains very high levels of heavy metals (HM) and organics(Zhu et al. 2008).

Recently, investigation of water contamination withHM has become a prime focus and rapidly increasingproblem all over the world. The continuous discharge ofoxygen demanding substances, toxic wastes, suspendedsolids and coloured wastes into the streams has beendeteriorating the quality of natural water. Both industrialactivities and growing demand have been posing pressureon the environment and the natural water sources, despitethe fact that stringent regulation and new technologieswhich are used in reducing the environmental impactespecially in the industrialised countries. The HM inwater could be derived from both natural (weathering ofbed rocks) and anthropogenic (mining, industries andagriculture activities) sources. In due course of time dis-charge of effluents into aquatic systems effects livingorganisms in drawing water. Furthermore, HM may con-taminate the surface and groundwater resulting incontamination of drinking water quality (Ahmed et al.2006; Muhammad et al. 2010; Alkarkhi et al. 2008). Incontinuation to the above discussion, it is a fact that HMtoo are considered to be as severe pollutants owing totheir toxicity and bio accumulative nature and it is in thissense, health risk assessment has assumed a major role inthe characterisation and remediation of contaminated siteswith implementation of new environmental code of con-ducts which defines contaminant threshold concentrations(Sante et al. 2009). Generally, various multivariate statis-tical analysis like metal-metal correlation, principal com-ponent analysis (PCA), cluster analysis (CA) are used forinterpretation of large, complex data with a view to un-derstand better, the water quality in the study area. Thestatistical analysis provides a valuable tool for reliablemanagement of water resources to pollution problemsand thereby allows identifying possible factors that influ-ence water system (Wunderlin et al. 2001).

In the present study area, keeping in view the HM contam-ination in drinking water due to vast industrial activity, riskassessment of HM and their source distribution in waters of acontaminated area and their impact on human health was

conducted. This study aims at determination of HM concen-trations in surface and groundwater during two seasons and itspotential health risk assessment. Multivariate statistical anal-ysis was used to identify the sources of contamination. Therisk assessment in surface and groundwater was assessedusing the geoaccumulation index (Igeo), enrichment factor(EF), contamination factor (Cf), degree of contamination(Cdeg) and health risk parameters like chronic daily intake(CDI) and hazard quotient (HQ). This study would providebaseline data regarding distribution and accumulation of toxicmetals in the surface water and groundwater, which wouldhelp in combating the contamination around the industrial andresidential zone by identifying major pollution sources. Itwould also help in designing a strategy to control the emissionand spread of pollutants in the study environment.

Materials and methods

Study area

Katedan Industrial Development Area (KIDA) is one of thethirteen industrial development areas developed on the out-skirts of Hyderabad city by Andhra Pradesh Industrial Infra-structure Corporation in 1965 is situated south of Hyderabadcity on Hyderabad-Bangalore National Highway (NH 7).There are about 300 industries dealing with dyeing, edibleoil production, battery manufacturing, metal plating, metalalloys, plastic products, chemicals, etc. located in the area.These industries could be categorised under small-, medium-and large-scale industries. It is observed that most of theindustries, directly release their effluent into the streams andthe solid waste generated is randomly dumped on open landalong roads and lakes. The industrial effluents contain appre-ciable amounts of inorganic and organic chemicals and theirbi-products. Most of the industries which are in small-scalesector are not having any sewer lines and they dischargeindustrial effluents in unlined channels and streams, therebycausing enormous contamination of water and soil. Indiscrim-inate dumping of domestic and industrial wastes in the studyarea (Govil et al. 2012) has become a routine practice withwhich, the levels of toxic elements in soil and water havedrastically exceeded the permissible limits prescribed byWorld Health Organisation (WHO 1996) guidelines.

Climate, topography and drainage

The study area under investigation falls in semi arid zone.Usually, this region receives its first spell of rainfall from pre-monsoon conventional showers in the month of May but itsoccurrence is erratic. The area receives an average rainfall of900 mm from the southwest and northeast monsoon. Thehighest precipitation generally occurs during the southwest

3654 Environ Sci Pollut Res (2014) 21:3653–3669

monsoon during June–September. The intensity and amountof rainfall is unpredictable during the northeast monsoonperiod of October and November. The period between Januaryand May is the main dry season. However, there will berainfall followed due to the retreating monsoon by cyclonicstorms that commonly originate in the Bay of Bengal. Meanmonthly temperature ranges from 18 to 35 °C. Maximumtemperature rises to 43 °C during mid summers. The temper-ature gradually rises from the month of January and reachesmaximum in May and decreases gradually from June andreaches minimum in the month of November and December.The high temperature prevailing in this region causes heavyevapo-transpiration losses from surface and sub-surface waterbodies. The groundwater resources are depleted with the onsetof summer. Figure 1 shows the extent of the study area(KIDA) and the surrounding residential areas within the topo-graphic watershed. The area is gently sloping towards northwith the industrial area located in the highest elevated areas tothe south. The highest elevation recorded in the study area is570 m above mean sea level (msl) and the lowest being 520 mabove msl. The industrial area is separated from the down-stream residential areas by the railway and an interstate high-way. The drainage of the area consists of fiveman-made lakes,(cheruvu) viz. Noor Mohammad cheruvu, Chilan cheruvu,Ura cheruvu, Narsabaigunta and Devullama cheruvu that areinterconnected by small streams/creeks.

Soil, geology and hydrogeology

The soil cover is of well-developed residual soil of weatheredgranite. The soil is yellowish to brown–reddish sand withvarying content of silt. The soil is fairly permeable and theinfiltration rate can absorb most of the rain except for moreintensive rains, which can cause considerable surface flow anderosion. Lithological units consist of granites and pegmatite ofigneous origin belonging to the Archaean age. The granitesare pink and grey in colour, hard massive to foliated and welljointed. Epidote and Quartz veins cross cut the granite atvarious places. The groundwater in the Achaean granitesoccurs in the weathered and fractured zones under the watertable and in semi-confined conditions. The depth ofweathering and fractured zones dominantly controls the oc-currence and the movement of groundwater in these rocks.These rocks possess negligible primary porosity but are ren-dered with a porosity and permeability due to secondaryporosity by deep fracturing and weathering, which locallyform potential aquifers. Water level fluctuates annually inalmost all the bore wells. The average depth to groundwateris about 8 m.Water levels usually rise in post-monsoon seasonwith the water table fluctuating between 1 and 15 m belowground level and during pre-monsoon season water levelsusually decline and the water table fluctuates between 10and 25 m.

Sampling and preparation

Eighty-five water samples (33 surface water and 52 ground-water) were collected from the entire study area (Fig. 2) whichinclude effluent discharges from industries, water samplesalong the course of lakes, from open streams and groundwatersamples from bore wells, hand pumps and dug wells. Sampleswere also collected from the out lets of the industries aftereffluents had mixed with contaminants or less contaminatedsurface water.

The study of the groundwater was to be based upon infor-mation from the existing wells in the area. A well inventorystudy was done collecting data on type of well, use and depthof well. A total number of 119 wells were registered out ofwhich 98 bore wells were equipped with electric submersiblepumps, 14 bore wells equipped with hand pumps and 7 dugwells. In the present study, we choose randomly 52 ground-water samples covering the entire study area, keeping thesource of probable contamination. All the wells equipped withsubmersible pumps are mainly used for industrial purpose butmany are also used for domestic purpose, whereas all wellsequipped with hand pumps and dug wells are for domesticpurpose.

Water samples were collected in one liter capacity poly-thene bottles from representative bore wells/dug wells/handpumps distributed throughout the study area which are underuse during both seasons, i.e. summer (pre- monsoon) andwinter (post-monsoon) during the years 2008 and 2009. Priorto collection the bottles were thoroughly washed with dilutedacid and then with distilled water in the lab. Before filling thesample the bottle was rinsed to avoid any possible contami-nation. pH and TDS were measured onsite and samples werecollected in 1 L double-cap polythene bottles. Borehole watersamples were collected after discharging first few litres of thewater sample. The water samples were then filtered using 40micron pore size filter paper and acidified (2 mL of 2 % AR-grade HNO3) to pH 2 for chemical analysis to be carried outby inductively coupled plasma-mass spectrometry (ICP-MS).

Instrumentation

All the surface and groundwater samples were analysed fortoxic HM (As, Cd, Cr, Cu, Ni, Pb and Zn) by using ICP-MS.The instrument used was Plasma Quad (VG Elemental Ltd.,Winsford, Cheshine, UK). Sample solution was introducedinto the ICP-MS instrument by conventional pneumaticnebulisation using peristaltic pump with a solution uptake rateof about 1 mL/min. The system was operated in the massscanning mode after tuning the ion lenses (Balaram et al.1992). The water reference sample solution was prepared intriplicate by adding 1 mL of 1 μg/mL indium solution to 9 mLof the reference sample. Calibration curves were preparedusing multielement standard solution after dilution to

Environ Sci Pollut Res (2014) 21:3653–3669 3655

microgrammes per litre levels. Reference water samples1643b from National Institute of Standards and Technology(NIST, USA) were used to check the reliability of calibrationcurve. All results were obtained in the multielelement modeand the samples were prepared in triplicates and analysedtwice. The values obtained by ICP-MS are in close agreementwith recommended values, the precision are better than 2 %and show comparable accuracy (Table 1). Lower limits ofdetection for all elements were better than 1 μg/L. The preci-sion obtained in most cases was better than 5 % RSD withcomparable accuracy (Balaram 1993).

Data analysis and multivariate statistical methods

Surface and groundwater data was subjected through multi-variate statistical methods in terms of distribution and corre-lation among the studied parameters. The position of watersampling was recorded using an Atrax model GPS receiversystem. SPSS 10.0 software package (SPSS 1995) was usedfor statistical analysis of the metal data. Basic statistical pa-rameters such as range, mean, standard deviation, kurtosis andskewness were computed (as shown in Tables 2 and 3) alongwith correlation analysis (Tables 4 and 5), while multivariatestatistics in terms of PCA and CA were also carried out(Acosta et al. 2010). PCA was carried out using varimaxnormalised rotation on the data-set and the CA was appliedto the sample concentrations using dendogram method. Prin-cipal components provide information on the most meaning-ful parameters which describe whole data set affording datareduction with minimum loss of original information. PCA isa powerful technique for pattern recognition that attempts toexplain the variance of a large set of inter-correlated variablesand transforming into a smaller set of independent(uncorrelated) variables (principal components). FA furtherreduce the contribution of less significant variables obtainedfrom PCA and the new group of variables known asvarifactors are extracted through rotating the axis defined byPCA (Davis 1986; Shrestha and Kazama 2007; KeshavKrishna et al. 2009). PCA of the normalised variables (surfaceand groundwater data set) was performed to extract significantPC's and to further reduce the contribution of variables withminor significance.

CA on the other side organises a set of variables into two ormore mutually exclusive unknown groups/clusters based oncombination of internal variables. The idea of CA is to findout a system of organising variables where each cluster sharescommon properties. Highly contaminated sites with toxiccontaminants share critical properties such as high, acute andchronic toxicity, high environmental persistence often hashigh mobility leading to contamination of groundwater andhigh lipophilicity leading to bioaccumulation in food web.With a view to understand the HM dynamics, the present

study was carried out on surface and groundwater contamina-tion using various indices, including Igeo, EF, Cf and Cdeg.

The Igeo enables to estimate contamination comparingpresent and preindustrial metal concentration. This methodhas been used by Muller 1969 for several trace metal studiesin Europe. It is computed using the following equation;

Igeo ¼ Log Cn=1:5 Bnð Þ ð1Þ

In the present study, we applied the modified calculationbased on the equation given in Krzysztof et al. 2004. WhereCn denoted the concentration of a given element in the watertested, while Bn denoted the concentration of elements in theEarth's crust (Taylor andMcLennan 1995). Muller divided theI geo into six classes, they are (I geo≤0) practically un-contaminated, (0<I geo<1) uncontaminated to moderatelycontaminated, (1<Igeo<2) moderately contaminated, (2<I geo<3) moderately to heavily contaminated, (3<I geo<4)heavily contaminated and (5≤Igeo) extremely contaminated.

EF was calculated using the modified formula given byLoska et al. 2004. This method is based on standardisation ofan element tested against a reference element. Referenceenvironment adopted here was the average concentration ofelements in Earth's crust as similar to Igeo. This aimed toenable a comparison of the two factors Igeo and EF. Equation(2) as suggested by Buat-Menard and Chesselet 1979 wasutilised,

EF ¼ Cn sampleð Þ=Cref sampleð Þf g=Bn backgroundð Þ=Bref backgroundð Þf g

ð2Þ

Where Cn (sample) is the content of the examined element,C ref (sample) is the content of the reference element in theexamined environment, Bn (background) is the content of theexamined element, B ref (background) is the reference elementin the reference environment. Sutherland (2000) divided EFinto five groups, they are: (EF<2) deficiency to minimalenrichment, (EF=2–5) moderate enrichment, (EF=5–20) sig-nificant enrichment, (EF=20–40) very high enrichment and(EF>40) extremely high enrichment. For determination ofwater contamination, EF and contamination degree are used.In the present study the modified form of the method forcalculation of Cf by Hakanson (1980) was applied. The Cf

is computed from the following eq. (3),

C f ¼ Co=Cn ð3Þ

Where Co is the mean content of metals of at least fivesampling sites,Cn is the concentration of elements in the Earth'scrust. Hakanson (1980) divided the Cf into four categories: (Cf

<1) low Cf indicating low contamination, (1≤Cf≤3) moderateCf, (3≤Cf≤6) considerable Cf and (6≥Cf ) very high Cf. TheCf indicates contamination of only one element. The sum of the


Fig. 1 Location map of the studyarea


Cf of all the elements yields the contamination degree (Cdeg) ofthe environment investigated. It can be divided into four cate-gories as: (Cdeg<8) low Cdeg, (8≤Cdeg<16) moderate Cdeg,(16≤Cdeg<32) considerable Cdeg and (32≥Cdeg) very highCdeg.

Human health risk assessment

The identification and characterisation of associated humanhealth risks are the need of the hour to be addressed based on

environmental geochemical studies. Health risk assessment isgenerally based on a quantification of risk level in relation totwo types of adverse effects: chronic (non-carcinogenic) andcarcinogenic. HM enter into human body through severalpathways through ingestion, dermal contact (direct exposure)with soil, vapour and particulate inhalation from the soilsource (ATSDR 1999; Iqbal et al. 2011). Chronic risk levelestimated was expressed as maximum hazard quotient(HQmax) calculated for a group of evaluated elements and ashazard index (HI) calculated as a sum of HQ of all evaluatedelements in every sample (HI=∑HQi). Characterisation of thechronic risk level consists of threshold effects (tolerancechemical level) and is based on the presumption and manifes-tation of adverse chronic effects until the threshold, i.e. thelifetime daily exposure level tolerated by human beings theso-called reference dose (RfD), is exceeded. The characteri-sation of carcinogenic risk level consists of a concept ofnonthreshold effects – that is, no dose is safe and risk-freeand each level of exposure can generate a carcinogenic re-sponse (USEPA 1989). Here in the present study, health riskfrom increased concentrations of HM in the surface andgroundwater was assessed in relation to its chronic as wellas carcinogenic effects, based on the evaluation of averagedaily dose estimates and defined toxicity values for toxic HM(USEPA 1999) according to the following relationships. Thechronic risk level was calculated as health risk assessmentusing CDI and HQ indices. The CDI through water ingestionwas calculated using the equation by USEPA (1992) below:

CDI ¼ C � DI=BW ð4Þ

where C , DI and BW represent the concentration of HM inwater (microgrammes per litre), average daily intake rate (2 L/day) and body weight (72 kg), respectively (USEPA 2005).Conversely, the chronic risk level was calculated (HQ) fornon-carcinogenic risk using following equation by USEPA(1999):

HQ ¼ CDI=RfD ð5Þ

where according to USEPA, the oral toxicity RfD values are0.0003 mg/kg-day for As, 0.0005 mg/kg-day for Cd, 0.0015for Cr, 0.037 mg/kg-day for Cu, 0.02 mg/kg-day for Ni,0.0036 mg/kg-day for Pb and 0.3 mg/kg-day for Zn, respec-tively. The scale of chronic risk level (HQ) based on averagedaily intake (CDI) and reference dose (milligrammes perkilogramme-day) is classified based on the ratio of CDI/RfDindicating≤1 (no risk) if>1≤5 (low risk), if>5≤10 (mediumrisk) and if>10 (high risk).

Table 1 Trace element data for water reference sample NIST 1643b byICP-MS

Analyte ICP-MS valuea

(ng/mL)Recommended value(ng/mL)

Average (RSD%)

As 50.29 49.0 1.81

Cd 19.89 20.0 0.39

Cr 18.72 18.6 0.45

Cu 21.69 21.9 0.68

Ni 47.87 49.0 1.67

Pb 24.02 23.7 0.94

Zn 66.89 66.0 0.94

a Average of six values

Fig. 2 Sample location map of the study area


Results and discussion

Physicochemical parameters and descriptive statistics

The physicochemical parameters like pH, TDS and EC werealso taken for interpretation. biochemical reactions are sensi-tive to variations of pH, TDS and EC. pH value being recip-rocal of log of hydrogen ion concentration is dissociated to aslight extent into positively charged hydrogen ion (H+) andnegative hydroxyl ions (OH−) which represents the balance ofa series of equilibrium existing in the water. The pH valuesvaried between 1.4 and 8.3 with mean pH of 3.3 for surfacewater and 6.4 to 8.1 and 6.4 to 7.8 in groundwater during preand post-monsoon seasons respectively. All the water samplesshow acidic nature for surface water and neutral to basic andalkaline values for groundwater. TDS values of groundwaterin the study area vary between 447 and 2,828 and 512 and 4,361 mg/L during pre- and post-monsoon, respectively . Thedesirable limit of TDS in drinking water is 500 mg/L (ISI1983). Portability of water decreases when the concentrationof TDS exceeds this limit and may cause gastrointestinalirritation. The high concentration of TDS is due to the com-bined effect of sewage contamination and industrial effluents.The electrical conductivity (EC) of water is directly propor-tional to the salt concentration and vice-versa. Hence, the

conductivity measurement provides an indication of ionicconcentrations. The EC of groundwater in the study areavaries between 1,010 and 7,600, 1,000 and 7,300 μS/cm at25 °C during both seasons, respectively.

The statistical summary of the distribution parameters forthe selected metals (As, Cd, Cr, Cu, Ni, Pb and Zn) in surfacewater during pre- and post-monsoon is given in Tables 2 and 3.During pre-monsoon, considerably elevated mean levels wereshown by Zn (2,943 μg/L), Pb (173.8 μg/L), Cr (88.33 μg/L),Cu (58.1 μg/L), Ni (53.6 μg/L), As (16.1 μg/L) and Cd(7.9 μg/L), while during post-monsoon the concentrationswere Zn (1,889 μg/L), Pb (118.6 μg/L), Cr (32.90 μg/L), Cu(28.26 μg/L), Ni (69.96 μg/L), As (11.83 μg/L) and Cd(1.64 μg/L). The quality of surface water is a very sensitiveissue. Anthropogenic influences (urban, industrial, agriculturalactivities and increasing consumption of water resources) aswell as natural processes (changes in precipitation inputs, ero-sion, andweathering of crustal materials) degrade surface watersand impair their use for drinking, industrial, agriculture or otherpurposes. In the present study area, the surface water concen-trations are quite equally distributed in the streams and the fivelakes during both the seasons. The dilution factor occurringduring the post-monsoon is probably very high as it reducesthe concentrations 2- to 8-fold depending on the metals duringthe monsoon period. On an average basis in surface water the

Table 3 Statistical parameters for selected metal distribution in groundwater (micrograms per litre; n =52) during pre- and post-monsoon

Parameters Pre-monsoon Post-monsoon

WHO Min Max Mean SD Skewness Min Max Mean SD Skewness

As 10.0 1.01 61.90 18.18 19.20 1.08 1.09 12.21 3.76 2.07 1.99

Cd 3.0 0.20 6.90 1.67 1.55 2.02 0.04 5.85 0.40 1.03 4.37

Cr 50.0 6.40 217.4 29.40 31.63 4.31 0.47 10.40 5.15 3.50 -0.18

Cu 2,000 1.40 200.1 17.03 29.14 5.30 0.92 17.60 4.19 3.03 2.02

Ni 20.0 1.20 76.20 25.44 19.02 0.88 0.11 37.90 6.09 7.27 2.38

Pb 10.0 0.40 453.8 81.72 101.6 1.58 0.09 45.10 2.87 7.17 5.05

Zn 3,000 34.0 5,818 953 1,091 2.40 15.70 22,036 989 4,241 4.92

Table 2 Statistical parameters for selected metal distribution in surface water (micrograms per litre; n=33) during pre- and post-monsoon

Parameters Pre-monsoon Post-monsoon

WHO Min Max Mean SD Skewness Min Max Mean SD Skewness

As 10.0 1.09 48.04 16.13 13.81 0.53 5.50 17.10 11.83 2.30 -0.57

Cd 3.0 0.04 63.50 7.91 13.41 2.93 0.50 6.20 1.64 1.49 2.38

Cr 50.0 5.27 770.8 88.33 135.3 3.46 6.70 107.10 32.90 21.04 1.80

Cu 2,000 1.11 230.7 58.11 65.81 1.07 13.20 256.4 28.26 41.45 5.53

Ni 20.0 0.11 214.3 53.62 57.14 1.11 26.50 128.8 69.96 19.22 0.41

Pb 10.0 0.22 1,207 173.8 227.1 2.24 25.30 877.2 118.6 158.1 3.74

Zn 3,000 15.70 21,879 2,943 4,212 2.64 400.9 13,117 1,889 2,758 3.07


metal levels showed following order Zn>Pb>Cr>Cu>Ni>Cd>As, whereas during post-monsoon concentrations orderwas Zn>Pb>Cu>Ni>Cr>As>Cd. All the concentrationsare exceeding the drinking water limit except for Cu, Crand Zn during the post-monsoon seeing the maximumvalues (Table 2). The elevated levels of these metals areassociated with most of the anthropogenic activities, particu-larly the industrial and agriculture activities, which are at thepeak during pre- monsoon when the water level is highest inthe lakes. For example, taking arsenic contamination, organiceffluents discharged by the industries can complex with arse-nic to form non-degradable metal complexes and they in turnenter the groundwater and migrate along natural hydrologicalgradient. Furthermore, arsenic contamination in Katedancomes mainly from paint, pharmaceutical, fertiliser and pesti-cides industries. Similarly, high chromium values were foundclose to industries manufacturing steel and paints. High valueof Cd with 63 μg/L was found near by the industrymanufacturing chemicals and pesticides.

However, during post-monsoon there is dilution due toclimatic variations and the anthropogenic activities are verylimited and hence the contribution of the metals is considerablydiminished. When illustrating and discussing the groundwaterdata from the study area, the area is divided into the industrial

area where the contamination is taking place, and the down-stream residential area towards north of National Highway 7.

Groundwater flow and contamination

All the wells in the Katedan area show that the static watertable is on an average rising during the monsoon from a depthof 16 to 5 m below surface level. For some wells, the depth togroundwater table raises from 33 to 7 m below surface level.The wells vary in depth from 30 to 250 m with an averagedepth of 65 m. The well casing is drilled into the decomposedrock zone to a depth of approximately 10–20 m and theborehole is open at the fractured granite and there is noparticular sealing of the upper part of the casing to preventcontaminated surface water to enter the well. Single wellpumping tests have been done on eight wells, in the industrialarea. The wells with high static water table, which are drawingonweathered granitic zone, had 75 times higher transmissivitythan the wells with static water table at 10 m. Assuming 0.1porosity in the weathered zone the natural groundwater flowfrom the industrial area may amount to 20–50 m3/day.Groundwater flow in the deeper granite depends on fracturesand it is known that some of the deeper wells are yielding 2–5 m3/h for several hours each day. In comparison with deeperstatic water level, the three wells tested did not yield much.Groundwater in industrial area will flow towards the NoorMohammad lake and some is expected to flow directly to-wards the downstream residential area. There will also begroundwater flow from the lake towards the nearer down-stream area. In the flatter areas downstream, the groundwaterflow will be very slow in comparison to the industrial areabecause of the very low gradient. However, the real situationat Katedan is different, the industries and residents areutilising all the available groundwater. Insignificant ground-water is flowing out of the area. A certain groundwater draw-down is therefore permanent most of the year. It is alsoobserved from the wells how the groundwater regime isrecharged back to a natural state (according to static waterlevel measurements) during the monsoon period. Figure 3shown visualises what is believed to be the groundwater

Table 5 Correlation coefficient (r) matrix of selected metals in ground-water during pre-monsoon (above the diagonal) and post-monsoon (be-low the diagonal; n=52)

Element As Cd Cr Cu Ni Pb Zn

As 1.00 0.32 0.61* 0.13 0.65* 0.35 0.52

Cd 0.15 1.00 0.59* 0.11 0.14 0.68* 0.58*

Cr −0.09 −0.06 1.00 0.06 0.15 0.45 0.82*

Cu 0.35 0.23 0.06 1.00 0.30 0.19 0.08

Ni 0.60 0.46 −0.09 0.52 1.00 0.23 0.24

Pb 0.24 0.64* −0.03 0.38 0.73* 1.00 0.57*

Zn 0.25 0.63* 0.09 0.34 0.73* 0.95*, ** 1.00

n number of groundwater samples

*Level of significance=0.05; **level of significance=0.01

Table 4 Correlation coefficient(r) matrix of selected metals insurface water during pre-monsoon (above the diagonal)and post-monsoon (below thediagonal; n =33)

n number of surface water samples

*Level of significance=0.05;**level of significance=0.01

Element As Cd Cr Cu Ni Pb Zn

As 1.00 0.58* 0.60* 0.92* 0.92* 0.69* 0.49

Cd 0.26 1.00 0.52* 0.64* 0.45 0.51 0.58*

Cr 0.14 −0.16 1.00 0.64* 0.52* 0.55* 0.52

Cu 0.44 0.19 −0.04 1.00 0.86*, a 0.69* 0.54

Ni 0.42 −0.15 0.48 0.37 1.00 0.65* 0.46

Pb 0.42 0.64* 0.12 0.30 0.13 1.00 0.44

Zn 0.24 0.92*, ** −0.12 0.10 −0.15 0.57* 1.00


situation in the study area. Most of the groundwater fromupstream and throughout the industrial area is pumped outand discharged to the Noor Mohammad lake via the streams.It is also expected that some of the deeper wells close to theNoor Mohammad Lake draw water from the lake. In some ofthe wells recirculation of the groundwater may take place. Anidealised sketch in Fig. 4 shows how the contaminant sources,surface soil and the streams with effluent and sediments arespreading contamination to the groundwater. The streams areassumed to be the main source as they carry water throughoutthe year. The wells extracting as much water as possible areimposing a steep gradient and increasing the flow. Some clean

water may also reach wells originating from infiltrating rainwater at an unpolluted area. The extracted groundwater at theindustrial site is always contaminated.

Inter-element relationship and statistical analysis

The inter-metal relationships can provide interesting informa-tion on element sources and pathways. Correlation analysisperformed between all variables (As, Cd, Cr, Cu, Ni, Pb andZn) are closely associated with each other. Tables 4 and 5summarise the HM correlation co-efficient in surface waterand groundwater for both pre and post-monsoon seasons

Fig. 3 Idealized sketch ofcontaminated groundwater andclean groundwater

Fig. 4 Contaminated and clean groundwater flow from source to well


respectively. In surface water some of the HM showed posi-tive correlation like Cu–As (r =0.92), Ni–As (r =0.92), Ni–Cu(r =0.86) and Zn–Cd (r =0.58) in pre-monsoon whereas, inpost-monsoon only few metals like Pb, Zn and Cd showedpositive correlation Pb–Cd (r =0.64), Zn–Cd (r =0.92) andZn–Pb (r =0.57). In groundwater during pre-monsoon, a pos-itive correlation was observed between Ni and As (r =0.65),Pb and Cd (r =0.67), Zn and Cd (r =0.58) and Zn and Cr (r =0.82) (Table 5). Whereas, in post-monsoon, a strong andsignificant positive correlation was shown by Ni, As, Cd, Znand Pb, with Ni–As (r =0.60), Zn–Cd ( r =0.63), Zn–Ni (r =0.73) and Zn–Pb (r =0.95). High and significant correlationsbetween these metals indicate that contaminants in the studyarea (KIDA) waters have similar source which originates fromindustrial activities. The correlation matrix observed was fur-ther supported by CA for both surface water and groundwater(Fig. 5a, b). Dendogram obtained for surface water during pre-

monsoon showed two clusters with As–Ni, Cd–Zn, As–Cu,As–Pb and Cd–Cr showing significant coefficients 0.873,0.746, 0.839, 0.665 and 0.607. During post-monsoon therewas significant correlation between Cd–Zn and Cd–Pb, withcoefficients 0.916 and 0.712, respectvely. Similarly forgroundwater, dendogram obtained during pre-monsoonshowed significant correlation between Cr–Zn, Cd–Pb, As–Ni and Cd–Crwith coefficients 0.817, 0.679, 0.651 and 0.614,respectively. During post-monsoon, Pb–Zn, Ni–Pb and Ni–Cdshowed strong correlation with coefficients 0.949, 0.802 and0.688, respectviely. The distribution of selected toxic HM like(As, Cd, Cu, Ni, Pb and Zn) were strongly influencing thestudy area with a common point source from industries there-by indicating anthropogenic influence which is confirmed andsupported by correlation and CA.

To further shed more light and help in understanding thecomplex nature and apportionment among the metals in

Fig. 5 a , b Dendrogram definingthe hierarchical CA of toxicmetals in Surface water andgroundwater (pre- andpost-monsoon seasons)


surface and groundwater, multivariate analysis by means ofPCA/FA was carried out. By FA complex linear correlationbetween metal concentration in surface and groundwater dur-ing pre monsoon and post-monsoon was determined, whichenabled interpretation of metal correlation in the study area.Elements belonging to a given factor were defined by factormatrix after variance rotation, with those having strong corre-lations grouped into factors. Considering the influence theyexerted from surface water into the groundwater table bydetermining the distribution of elements in the study area(KIDA), the following multielement factor were divided intofactors with strong scattered anthropogenic influence. Thedistribution pattern of individual association of element insurface and groundwater was determined by principal com-ponent method (results shown in Tables 6 and 7).

Surface water

In surface water, the FA, based on eigenvalues and varimaxrotation, two factors for pre-monsoon and three factors forpost-monsoon explained most of the variability with totalvariance explained was 76.55 and 81.01 % for the two sea-sons. Factor 1 for pre-monsoon contributed 60.82 % to thetotal variance with a high loading on As (r =0.876), Cu (r =0.913), Ni (r =0.778), Cd (r =0.669), Cr (r =0.680), Pb (r =0.665) and Zn (r =0.835) as shown in the Table 6. Thesignificant loadings shown by As, Cu, Pb and Zn explain thecommon point source of these elements. As is mainly emittedby the copper producing industries, during lead and Zn pro-duction and in agriculture activity by spraying of pesticides.Furthermore, lead can end up in water and soils throughcorrosion of leaded pipelines in a water transporting systemand through corrosion of leaded paints and pollute surfacewaters. The high contamination of Pb (r =0.830) and Zn (r =

0.827) in factor 1 during post-monsoon surface water with39.13 % of total variance could be influenced by the surfacewater polluted with Zn, due to the presence of large quantitiesof Zn in the waste water of industrial plants where Zn is notpurified satisfactorily. Zn may also increase the acidity ofwater which correlated with acidic pH in the study area.Whereas, factor 2 contributed only 15.72 % during pre-monsoon with loading on Cd (r =0.533). However, significantloadings of (r =0.871) Cd in factor 1 of post-monsoon andfactor 1 of pre-monsoon Cd (r =0.669) indicates the source ofcadmium emission during the production of artificial phos-phate fertilisers, where part of the cadmium ends up in soilsafter the fertiliser is applied on farm and rest of the cadmiumends up in surface waters, when waste from fertiliser produc-tion is dumped by production companies. Factor 3 with14.41 % of total variance in post-monsoon of surface waterwith loading of Cr (r =0.705) could have been influenced bychromium in soils of the study area which strongly attaches tothe soil particles and is released into the surface waters. Inwater, Cr is absorbed on to sediment which becomes immo-bile and a small part of the chromium that ends up in waterwill eventually dissolve. Ni values in the pre- and post-monsoon surface water show values with mean concentration53.6 and 69.9 μg/L whereas, in groundwater during pre-monsoon it showed a value, 76.2 μg/L .There is decrease inthe Ni concentration during post-monsoon. Ni and Ni com-pounds have many industrial and commercial uses. Most Ni isused for the production of stainless steel and other Ni alloyswith high corrosion and temperature resistance. Ni metal andits alloys are used widely in the metallurgical, chemical andfood processing industries, especially as catalysts and pig-ments. Furthermore, Ni is easily accumulated in the biota,particularly in the phytoplankton or other aquatic plants,which are sensitive bioindicators of water pollution. It can

Table 6 Factor loading for selected heavy metal in surface water (n =33)

Heavy metals Pre-monsoon Post-monsoon

Factor 1 Factor 2 Factor 1 Factor 2 Factor 3

As 0.876 −0.381 0.600 0.501 −0.209Cd 0.669 0.533 0.871 −0.394 −0.107Cr 0.680 0.361 0.003 0.628 0.705

Cu 0.913 −0.195 0.470 0.444 −0.619Ni 0.778 −0.557 0.160 0.867 0.056

Pb 0.665 −0.034 0.830 0.043 0.187

Zn 0.835 −0.438 0.827 −0.415 0.189

Eigen value 4.258 1.101 2.740 1.922 1.009

Loading % 60.824 15.727 39.139 27.461 14.410

Cumulative % 60.824 76.551 39.139 66.600 81.010

Values of dominant heavy metals in each factor are set in italics

n number of surface water samples

Table 7 Factor loading for selected heavy metal in Groundwater (n =52)

Heavy metals Pre-monsoon Post-monsoon

Factor 1 Factor 2 Factor 1 Factor 2 Factor 3

As 0.745 0.400 0.512 −0.704 0.131

Cd 0.758 −0.340 0.699 0.361 −0.264Cr 0.837 −0.249 −0.033 0.409 0.883

Cu 0.239 0.556 0.575 −0.343 0.395

Ni 0.484 0.764 0.893 −0.250 0.025

Pb 0.745 −0.183 0.910 0.275 −0.100Zn 0.858 −0.232 0.902 0.323 −0.012Eigen value 3.413 1.316 3.522 1.153 1.037

Loading % 48.754 18.805 50.310 16.477 14.762

Cumulative % 48.754 67.559 50.310 66.187 81.549

Values of dominant heavy metals in each factor are set in italics

n number of groundwater samples


be deposited in the sediment by such processes as precipita-tion, complexation and adsorption on clay particles and viauptake by biota. Nevertheless, inhalation also is an importantroute of occupational exposure to Ni in relation to health risks.The source of Ni in the study area would be from some steelmanufacturing and food processing units.

Groundwater

Groundwater contamination can originate above or below thesurface of the Earth. Infiltration of polluted surface watercauses contamination below the surface of the Earth. Basedon the eigenvalues and varimax rotation shown in Table 7, two

Igeo-SW (Pre-monsoon)

-3.0

-2.0

-1.0

0.0

1.0

2.0

3.0

As Cd Cr Cu Ni Pb Zn

Elements

Geo

accu

mul

atio

n In

dex

(Ige

o) Min Max Mean

Igeo SW (Post Monsoon)

-2.5-2.0-1.5-1.0-0.50.00.51.01.52.02.5


Elements

Geo

accu

mul

atio

n In

dex

(Ige

o)

Min Max Mean

Igeo GW (Post-Monsoon)

-3.5

-3.0

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5


Elements

Geo

accu

mul

atio

n In

dex

(Ige

o)

Min Max Mean

Igeo GW (Pre-Monsoon)

-4.0

-3.0

-2.0

-1.0

0.0

1.0

2.0

3.0


Elements

Geo

accu

mul

atio

n In

dex

(Ige

o)

Min Max Mean

EF SW (Pre-monsoon)

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

As Cd Cr Ni Pb Zn

Elements

Enr

ichm

ent F

acto

r (E

F)

Min Max Mean

EF SW(Post-Monsoon)

0.00

5.00

10.00

15.00

20.00

25.00

30.00

As Cd Cr Ni Pb Zn

Elements

Enr

ichm

ent F

acto

r (E

F)

Min Max Mean

EF GW (Pre-Monsoon)

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

As Cd Cr Ni Pb Zn

Elements

Enr

ichm

entF

acto

r (E

F)

Min Max Mean

EF GW (Post-Monsoon)

0.0

5.0

10.0

15.0

20.0

As Cd Cr Ni Pb Zn

Element

Enr

ichm

net F

acto

r (E

F) Min Max Mean

Fig. 6 Summary of geoaccumulation index (Igeo) and enrichment factor (EF) for toxic heavy metals in surface and groundwater during pre- and post-monsoon seasons


factors for pre-monsoon with 48.75 % total variance for factor1 and 18.805 % for factor 2 were observed with high loadingon As (r =0.745), Cd (r =0.758), Cr (r =0.837), Pb (r =0.745)and Zn (r =0.858) in factor 1 and loading on Cu (r =0.556), Ni(r =0.764) in factor 2. Factor 1 with similar high loadings ofAs, Cd, Cr, Pb and Zn in both surface water and groundwateris attributed to the anthropogenic influence of these tracemetals in groundwater. The data reveals that these trace ele-ments have migrated from surface water. As the study area is

surrounded by granitic rocks, the migration of trace elementsin garnitic terrain is through factures and joints, and is muchfaster than that in sedimentary formation. Similarly, factor 1 inpost-monsoon groundwater showed 50.31 % of total variancewith high loadings on Ni (r =0.893), Pb (r =0.910) and Zn(r =0.902). The low loadings of Ni in surface water of factor 1and high loadings of Ni in factor 1 of post-monsoon ground-water might have influenced by nickel released into the air bytrash incinerators located in some of the industries of the study

CF-SW (Pre-Monsoon)

0.010.020.030.040.050.060.070.080.090.0

100.0


Element

Con

tam

inat

ion

Fac

tor(

CF

)Min Max Mean

Element Min Max Mean

As 0.25 4.84 2.34Cd 0.03 21.17 4.09Cr 0.19 15.42 2.69Cu 0.00 0.12 0.04Ni 0.20 10.72 4.04Pb 0.06 90.08 27.24Zn 0.03 4.73 1.30

Deg.Con 0.76 147.08 41.74

CF-SW (Post-monsoon)

0.0

5.0

10.0

15.0

20.0

25.0

30.0


Element

Con

tam

inat

ion

Fact

or (

CF)

Min Max Mean


As 0.55 1.71 1.18Cd 0.17 2.07 0.55

Cr 0.13 2.14 0.66Cu 0.01 0.13 0.01Ni 1.33 6.44 3.50Pb 2.53 27.72 10.65Zn 0.13 3.19 0.55

Deg.Con 4.85 43.40 17.10

0.0

5.0

10.0

15.0

20.0

25.0 Min Max Mean

CF-GW (Pre-monsoon)

0.05.0

10.015.020.025.030.035.040.045.050.0


Element

Con

tam

inat

ion

Fact

or (

CF)

Min Max Mean


As 0.10 6.19 1.82Cd 0.06 2.28 0.55Cr 0.13 4.35 0.59Cu 0.00 0.10 0.01Ni 0.06 3.81 1.27Pb 0.04 45.38 8.17Zn 0.01 1.93 0.32

Deg.Con 0.40 64.04 12.73

CF-GW (Post-Monsoon)

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00


Elements

Con

tam

inat

ion

Fac

tor

(CF

)

Min Max Mean


As 0.11 1.22 0.38Cd 0.01 1.95 0.13Cr 0.01 0.21 0.10Cu 0.00 0.01 0.00Ni 0.01 1.90 0.30Pb 0.01 4.51 0.29Zn 0.01 7.34 0.33

Deg.Con 0.16 17.14 1.53

Fig. 7 Summary of Cf and Cdeg for toxic heavy metals surface and groundwater during pre- and post-monsoon seasons


area, which might have settled on the surface water when itbecomes part of waste water streams originated from indus-trial effluents. The large part of all Ni compounds that arereleased to the environment will absorb to sediment andbecome immobile. However, in acidic group nickel is boundto become more mobile and it will often rinse out to thegroundwater which is reflected in the post-monsoongroundwater.

The contamination levels, geochemical normalisation andCdeg were also assessed using Igeo, EF and Cdeg as suggestedby (Sutherland 2000; Reimann and de Caritat 2005; Vega et al.2009). The Igeo is the quantitative measure of the pollutionindex. It can be used as a reference to estimate the extent ofHM pollution in aquatic system. Figure 6 demonstrates themin, mean and maximum Igeo values of HM in surface andgroundwater. Among surface water during pre-monsoon theIgeo values ranged from −0.42 to 0.86 for As, −1.40 to 1.50 forCd, −0.55 to 1.36 for Cr, −2.51 to 0.76 for Cu, −0.51 to 1.21for Ni, −1.03 to 2.26 for Pb and −1.35 to 0.85 for Zn. Duringpost-monsoon Igeo values ranged form −0.08 to 0.41 for As,−0.60 to 0.49 for Cd, 0.70 to 0.57 for Cr, −2.00 to 0.72 for Cu,0.30 to 0.99 for Ni, 0.58 to 2.21 for Pb and −0.70 to 0.82 forZn. On the whole the maximum Igeo values status of surfacewater during pre-monsoon indicated that the HM As and Znfell under un-contaminated to moderately contaminated zonewhereas, Cd, Cr and Ni were moderately contaminated andlead was moderately to heavily contaminated. During post-monsoon As, Cd, Cr, Ni and Zn showed uncontaminated tomoderately contaminated and Pb showedmoderately to heavi-ly contaminated. Considering groundwater Igeo values for twoseasons the average Igeo values indicated 0.16 for As, −0.238for Cd, −0.19 for Cr, −2.13 for Cu, 0.13 for Ni. 0.45 for Pb and−0.57 for Zn during pre-monsoon and the average Igeo valuesfor post-monsoon indicate −0.30 for As, −0.702 for Cd, −0.81for Cr, −2.50 for Cu, −0.34 for Ni, −0.37 for Pb and −0.31 forZn. Overall, the elements As, Cd, Cr, Cu, Ni, Pb and Znshowed practically uncontaminated water during post-monsoon. This could be the dilution and dispersion of thesemetals by the time they are released into groundwater.

The EF was carried out to assess the anthropogenic intru-sions of the metals in surface and groundwater. EF valueswere interpreted as suggested by Sutherland (Iqbal et al.2011). Figure 6 demonstrates the minimum, mean and maxi-mum EF values of the heavy toxic metals. The mean EFvalues for pre-monsoon surface water indicated elementsCd, Cr, Ni and Zn with deficient to minimal enrichment andEF values greater than 5 for As and Pb showed significantenrichment. Whereas, the mean EF values during post-monsoon showed EF<2 for As, Cd, Cr, Ni and Zn indicatingdeficient to minimal enrichment and EF>5 for Pb with sig-nificant enrichment. Lead being immobile the large particleswill drop to the ground from vehicular exhaust and pollutesurface waters. Assessment of groundwater EF for pre-

monsoon with mean EF values>2 for As and Ni showedminimal enrichment, EF values >5 for Pb showed significantenrichment and EF<2 for Cd, Cr and Zn showed deficient tominimal enrichment. During post-monsoon the mean EFvalues<2 for Cd, Cr, Ni, Pb and Zn showed deficient tominimal enrichment and EF value>2for As showed minimalenrichment. Overall the mean EF values of As and Pb classi-fied the surface and groundwater as moderate enrichment inpost-monsoon to significant enrichment in pre-monsoon sea-son. Based on Igeo and EF results, the KIDA is extremelycontaminated due to many years of random dumping of haz-ardous waste and free discharge of effluents by the industries.The application of the index of geoaccumulation, EF and Cf

enabled us to find elevated contents of some toxic metals insurface and groundwater of KIDAwith As, Cd, Cr, Cu, Ni, Pband Zn. Simultaneously, metal depletion was observed inwater EF<1 was observed for Cd, Cr and Zn in pre- andpost-monsoon surface and groundwater.

The Cdeg is a cumulative index based on the contaminationby each measured element (Cf) in the water where the currentand pre-industrial concentrations are compared on one on onebasis. Contamination factor (Cf) and Cdeg was used to deter-mine the general contamination. It is considered as moreappropriate parameter to assess overall contamination. Figure 7demonstrates minimum, mean and maximum Cf values of theindividual elements in surface and groundwater. Surface waterin Katedan during pre-monsoon season is moderately contam-inated with As andCr with meanCf value of 2.34 and 2.69, Cdand Ni showed considerable contamination with Cf value 4.09and 4.04, whereas lead showed a Cf of 27.24 with very highCf. The contamination degree was found to be 41.74 whichindicate very high Cdeg. When seen during post-monsoonsurface water except Pb with Cf value 10.65 indicating veryhigh Cf. Other elements Cd, Cr, Cu and Zn showed low Cf,whereas, As and Ni showed Cf value 1.18 and 3.50 withmoderate contamination. The sum of the contamination deter-mined for each element yields the contamination degree. Thecontamination degree for post-monsoon surface water wasCdeg of 17.10 indicating considerable Cdeg.

The Cf for groundwater pre-monsoon samples showed Cf

values 0.55 for Cd, 0.59 for Cr, 0.01 for Cu and 0.32 for Znindicating low Cf. As with Cf1.82 showed moderate Cf andPb with Cf of 8.17 showed very high contamination. TheCdeg for pre-monsoon groundwater was Cdeg of 12.73 indi-cating moderate Cdeg. Whereas, post-monsoon groundwatersamples showed mean Cf values for As, Cd, Cr, Cu, Ni, Pband Zn as 0.38, 0.13, 0.10, 0.00, 0.30, 0.29 and 0.33 with lowCf indicating low contamination. Contamination degreeshowed low Cdeg with Cdeg value 1.53 for post-monsoongroundwater. Overall the mean Cf values for Pb classifiedthe surface and groundwater as moderately contaminatedduring pre-monsoon season and getting diluted during post-monsoon season.


Human health risk assessment

Information was gathered from the local people about drink-ing water, age, food habits, body weight and health problemsin the study area. It was observed that residents were usingsurface and groundwater for various domestic and drinkingpurposes. Therefore, the surface flowing water from streamsthat were used mostly for domestic purposes were also select-ed for HM risk assessment like CDI and HQ indices.

Chronic daily intake

The results of CDI values are summarised in Table 8.Theresults suggest that in study area where people have utilisedsurface water for domestic usage and groundwater for drink-ing were contaminated with HM. The CDI values for surface

water ranged from 0.07 to 1.34 for As, 0.002 to 1.76 for Cd,0.26 to 21.4 for Cr, 0.11 to 6.40 for Cu, 0.11 to 5.95 for Ni,0.018 to 33.5 for Pb and 2.50 to 393 μg kg−1 day−1 for Znduring pre-monsoon. Whereas, during post-monsoon CDIvalues were ranging from 25.3 to 877.2 for As, 400.9 to 13,117 for Cd, 0.06 to 0.48 for Cr, 0.01 to 0.17 for Cu, 0.19 to2.98 for Ni, 0.37 to 7.12 for Pb and 0.53 to 3.58 μg/kg for Zn,respectively. Therefore, the order of toxicity of HM meanconcentrations for surface water during pre- and post-monsoon were found in the order of Zn>Cr>Cu>Ni>As>Pb>Cd and Cd>As>Zn>Pb>Ni>Cr>Cu. The high CDIvalues during post-monsoon in surface water may be attribut-ed to the degree of health risk on human health. Furthermore,the CDI values obtained are only indication or pathfinder tothe toxicity of HM found in surface water but nevertheless,CDI values obtained cannot be considered as a tool to addressthe source of contamination. However, the potential source of

Table 8 Chronic daily intake indices for heavy metals

Heavymetals

Statistics Surface water Groundwater

Pre-monsoon

Post-monsoon

Pre-monsoon Post-monsoon

As Min 0.07 25.30 0.03 0.03

Max 1.34 877.2 1.72 0.34

Mean 0.65 154.5 0.50 0.10

SD 0.34 209.3 0.55 0.07

Cd Min 0.002 400.9 0.00 0.00

Max 1.76 13,117 0.19 0.16

Mean 0.34 2,282 0.05 0.01

SD 0.42 3,485 0.05 0.03

Cr Min 0.26 0.06 0.18 0.01

Max 21.4 0.48 6.04 0.29

Mean 3.73 0.31 0.82 0.14

SD 4.22 0.09 1.11 0.10

Cu Min 0.11 0.01 0.04 0.03

Max 6.40 0.17 5.56 0.49

Mean 2.47 0.05 0.47 0.12

SD 1.82 0.05 1.05 0.10

Ni Min 0.11 0.19 0.03 0.00

Max 5.95 2.98 2.12 1.05

Mean 2.24 0.92 0.71 0.17

SD 1.53 0.72 0.56 0.23

Pb Min 0.018 0.37 0.01 0.00

Max 33.5 7.12 12.6 1.25

Mean 7.56 1.04 2.27 0.08

SD 6.50 1.64 3.09 0.25

Zn Min 2.50 0.53 0.95 0.44

Max 393 3.58 161 612

Mean 108 1.89 26.4 27.4

SD 95.1 0.71 34.8 139

Table 9 Hazard quotient indices for heavy metals

Heavymetals

Statistics Surface water Groundwater

Pre-monsoon

Post-monsoon

Pre-monsoon Post-monsoon

As Min 23.3 50.9 0.001 0.001

Max 447 158 0.048 0.009

Mean 216 109 0.014 0.003

SD 120 24.2 0.015 0.002

Cd Min 4.4 27.8 0.000 0.000

Max 3,527 344 0.005 0.005

Mean 682 91.6 0.001 0.000

SD 946 90.5 0.001 0.001

Cr Min 7.1 5.0 0.005 0.000

Max 578 80.4 0.168 0.008

Mean 101 24.7 0.023 0.004

SD 136 18.1 0.031 0.003

Cu Min 0.8 2.6 0.001 0.001

Max 45.8 50.9 0.154 0.014

Mean 17.6 5.6 0.013 0.003

SD 13.5 10.9 0.029 0.003

Ni Min 5.7 36.8 0.001 0.000

Max 297 179 0.059 0.029

Mean 112 97.2 0.020 0.005

SD 81.8 30.8 0.015 0.006

Pb Min 0.5 19.5 0.000 0.000

Max 932 677 0.350 0.035

Mean 210 91.6 0.063 0.002

SD 214 153 0.086 0.007

Zn Min 8.3 37.1 0.026 0.012

Max 1,312 1,214 4.490 17.00

Mean 360.5 175 0.735 0.763

SD 348 301 0.968 3.864


contamination can be attributed to the release of industrialeffluents on to the surface and further, due to the pesticideusage in agriculture fields wherein it is left into the streams asrunoff from the fields. Similarly, the CDI values in ground-water used for drinking purpose ranged from 0.03 to 1.72,0.00 to 0.19, 0.18 to 6.04, 0.04 to 5.56, 0.03 to 2.12, 0.01 to12.6 and 0.95 to 161 μg/kg for As, Cd, Cr, Cu, Ni, Pb and Znduring pre-monsoon. Whereas, in post-monsoon the CDIvalues ranged from 0.03 to 0.34, 0.00 to 0.16, 0.01 to 0.29,0.03 to 0.49, 0.00 to 1.05, 0.00 to 1.25 and 0.44 to 612 μg/kgfor As, Cd, Cr, Cu, Ni, Pb and Zn respectively (Table 8). Theorder of toxicity for both pre-monsoon and post-monsoongroundwater was found in the order of Zn>Pb>Cr>Ni>As>Cu>Cd and Zn>Ni>Cr>Cu>As>Pb>Cd respectively.

Hazard quotient indices

Table 9 summarises the HQ indices of HM through regularconsumption of surface and groundwater for various purposesin the study area. The mean HQ index values for As, Cd, Cr,Cu, Ni, Pb and Zn for surface water were 216, 682, 101, 17.6,112, 210 and 360.5, while that of post-monsoon were 109,91.6, 24.7, 5.6, 97.2, 91.6 and 175, respectively. Similarly, forgroundwater pre-monsoon the mean HQ index values were0.014, 0.001, 0.023, 0.013, 0.020, 0.063 and 0.735 for As, Cd,Cr, Cu, Ni, Pb, Zn and post-monsoon samples showed 0.003,0.000, 0.004, 0.003, 0.005, 0.002 and 0.763, respectively.Therefore, the order of toxicity of HM mean concentrationsfor surface water during pre- and post-monsoon were found inthe order of Zn>Cr>Cu>Ni>As>Pb>Cd and Cd>As>Zn>Pb>Ni>Cr>Cu.

Conclusions

The study of HM concentrations in surface water and ground-water pollution due to uncontrolled industrial effluent dis-charge during two seasons for pre-monsoon and post-monsoon were found to be in the order of Zn>Pb>Cr>Cu>Ni>Cd>As and Zn>Pb>Cu>Ni>Cr>As>Cd. The results ofFA performed on seven HM, identified two factors for surfacewater and three factors for groundwater during both the sea-sons controlling their variability in waters of study area.Pollution around Katedan industrial area has increased dueto discharge of industrial effluents from paint, pharmaceutical,fertiliser, steel and chemical manufacturing industries and tosome extent from pesticides in agriculture fields. These fur-ther, migrate from surface water bodies into the ground waterthereby contaminating the aquifer. Nevertheless, the inter-metal correlation of selected HM in surface and groundwatershowed significant positive correlation between metal pairswhich were supported by the multivariate approaches like CAand PCA. Furthermore, PCA revealed that migration patterns

of HM released into the environment in the form of untreatedeffluents by the industries indicate the point source of pollu-tion. The health risk assessment like CDI and HQ indicesindicated that the groundwater is safe for drinking purposeprovided some water treatment methodologies are adopted.Igeo, EF and Cf exhibited moderate to high contamination forfew metals like Zn, Cr, Cu and As during pre- and post-monsoon seasons. Pb, Cd and Ni showed deficient to minimalenrichment and moderate to heavy contamination with overallconsiderable contamination.

Recommendations

The untreated effluents emerging from the industries must bemonitored for maintaining the standards prescribed by thepollution control board for various industries in the region.The drinking water in the study area poses chronic health riskcannot be ruled out in the near future, as it is seen that theconcentrations of toxic metals in surface water are higher thanthe groundwater. However, as some of the toxic HM exceededtheir safe levels, it is suggested that water without treatmentshould not be used by residents from the contaminated sites orelse use the alternative drinking water supplied by Govern-ment authorities. Furthermore, the present study would pro-vide a baseline data for assessment of contamination especial-ly for the industrial zone where industrial discharges and useof pesticides in agriculture cultivation can be attributed as thepotential sources of contamination. Furthermore, letting out ofthis contaminated water on to the surface appeared as themajor source of HM in surface and groundwater in this studyarea.

Acknowledgements The present work is a part of authors Ph.D. workand was instigated by Dr.P.K.Govil under Indo-Norwegian collaborativeprogramme. The authors thank Drs. P.K. Govil, Jan Erik Sorlie, and N.N.Murthy for their support, encouragement and scientific inputs. Thanks arealso due to Drs. V. Balaram and M. Satyanarayanan for providing ICP-MS analytical facility. The authors would also sincerely thank Prof.Mrinal K Sen, Director CSIR-National Geophysical Research Institute,for his permission to publish this paper. Last but not the least; thanks arealso due to the anonymous reviewers for their excellent review of themanuscript point wise for its successful publication.

References

Acosta JA, Faz A, Martinez SM (2010) Identification of heavy metalsources by multivariate analysis in a typical Mediterranean City (SESpain). Environ Monit Assess 169:519–530

Ahmed D, Fevzi Y, Tuna AL, Nedim O (2006) Heavy metals in water,sediment and tissues of Leuciscus cephalus from a stream in southwestern Turkey. Chemosphere 63:1451–1458

Alkarkhi AFM, Ismail N, Easa AM (2008) Assessment of arsenic andheavy metal contamination in cockies using multivariate statisticaltechniques. J Hazard Mater 150:783–789


ATSDR (1999) Toxicological profile for Arsenic, TP-92/02, Agency fortoxic substances and Diseases Registry, US. Departmental of Healthand Human services, Atlanta, G.A.9).

Balaram V (1993) Characterisation of trace elements in environmentalsamples by ICP-MS. At spectrosc 6:174–179

Balaram V, Manikyamba C, Ramesh SL, Anjiah KV (1992) Rare earthand trace element determination of iron formation reference samplesby ICP-Mass spectrometer. At Spectrosc 13:19–25

Buat-Menard P, Chesselet R (1979) Variable influence of the atmosphericflux on the trace metal chemistry of oceanic suspended matter, EarthPlanet. Sci Lett 42:398–411

Davis JC (1986) Statistica and data analysis in geology, 2nd edn. Wiley,New York

Govil PK, Sorlie JE, Sujatha D, Keshav Krishna A, Murthy NN, RamaMohan K (2012) Assessment of heavy metal pollution in lakesediments of Katedan Industrial Development Area, Hyderabad,India. Environ Earth Sci 66:121–128

Hakanson K (1980) An ecological risk index for aquatic pollution control,a sedimentological approach. Water Res 14:975–1001

Iqbal J, Munir H, Shah (2011) Distribution, correlation and risk assess-ment of selected metals in urban soils from Islamabad, Pakistan. JHazard Mater 192:887–898

ISI (1983) Drinking water standards, Table 1 Substances and character-istics affecting the acceptability of water for domestic use 18, 10500Indian Standard Institution, New Delhi.

Kalra RN (2012), Monsoon Contaminations, The Times of India,Mumbai, dated Sep 2–2012, Section, Special report, p. 5.

Keshav Krishna A, Satyanarayanan M, Govil PK (2009) Assessment ofheavy metal pollution in water using multivariate statistical tech-niques in an industrial area: a case study from Patancheru, MedakDistrict, Andhra Pradesh, India. J Hazard Mater 167:366–373

Krzysztof L,Wiechula D, Korns I (2004)Metal contamination of farmingsoils affected by industry. Environ Int 30:159–165

Loska K, Wiechula D, Korns I (2004) Metal contamination of farmingsoils affected by industry. Environ Int 30:159–165

Muhammad S, Shah MT, Khan S (2010) Arsenic health risk assessmentin drinking water and source apportionment using multivariatestatistical techniques in Kohistan Region, Northern Pakistan. FoodChem Toxicol 48:2855–2864

Muller G (1969) Index of geo-accumulation in sediments of the RaineRiver. GeoJournal 2:108–118

Reimann C, de Caritat P (2005) Distinguishing between natural andanthropogenic sources for elements in the environment: regionalgeochemical surveys versus enrichment factors. Sci Total Environ337:91–107

Sante MD, Mazzieri F, Pasqualini E (2009) Assessment of the sanitaryand environmental risks posed by a contaminated industrial site. JHazar Mater 171:524–534

Shrestha S, Kazama F (2007) Assessment of surface water quality usingmultivariate statistical techniques; a case study of the Fuji riverbasin, Japan. Environ Model Softw 22(4):464–475

Professional Statistics 6.1, 385, Marija J. Norusis/SPSS Inc.,Chicago.

Sutherland RA (2000) Bed sediment associated trace metals in an urbanstream, Oahu, Hawaii. Environ Geol 39:611–627

Taylor SR, McLennan SM (1995) The geochemical evolution of thecontinental crust. Rev Geophys 33:165–241

USEPA (1989) Risk assessment guidance for superfund (RAGS): volumeI- Human health evaluation manual (HHEM) part A: baseline riskassessment, Interim Final (EPA/540/1-89/002). United StatesEnvironmental Protection Agency, Office of Emergency andRemedial Response, Washington, DC.

USEPA (1992) Guidelines for Exposure Assessment, EPA/600/Z-92/001.Risk Assessment Forum, Washington, DC

USEPA (1999) Guidance for performing Aggregate Exposure and RiskAssessment office of Pesticide Programs, Washington, DC.

USEPA (2005) Guidelines for carcinogen risk assessment. United StatesEnvironmental Protection Agency, Risk Assessment Forum,Washington, DC (EPA/630/P-03/001F).

Vega FA, Covelo EF, Cerqueina B, Andrade ML (2009) Enrichment ofmarsh soils with heavy metals by effect of anthropic pollution. JHazard Mater 170:1056–1063

WHO, Guidelines for drinkingwater health criteria and other supportinginformation, vol. 2,World Health Organization, Geneva, 1996, p. 973.

Wunderlin DA, Diaz MP, Ame MV, Pesce SF, Hued AC, Bistoni MA(2001) Pattern recognition techniques for the evaluation of spatialand temporal variation in water quality. A case study: Suquia riverbasin (Cordoba, Argentina). Water Res 35:2881–2894

Zhu W, Bian B, Li L (2008) Heavy metal contaminations of road-deposited sediments in a medium size city of China. EnvironMonit Assess 147:171–181


SPSS® (Statistical Package for Social Studies) version 6.1, USA. (1995).

risk assessment of heavy metals and their source distribution in waters of a contaminated industrial...

Documents