Temporal variations in arsenic concentration in the groundwater of Murshidabad District, West Bengal, India

ORIGINAL ARTICLE

Temporal variations in arsenic concentration in the groundwaterof Murshidabad District, West Bengal, India

S. H. Farooq • D. Chandrasekharam • S. Norra •

Z. Berner • E. Eiche • P. Thambidurai • D. Stuben

Received: 20 July 2009 / Accepted: 5 March 2010

� Springer-Verlag 2010

Abstract Systematic investigations on seasonal varia-

tions in arsenic (As) concentrations in groundwater in both

space and time are scarce for most parts of West Bengal

(India). Hence, this study has been undertaken to investi-

gate the extent of As pollution and its temporal variability

in parts of Murshidabad district (West Bengal, India).

Water samples from 35 wells were collected during pre-

monsoon, monsoon and post-monsoon seasons and ana-

lyzed for various elements. Based on the Indian permissible

limit for As (50 lg/L) in the drinking water, water samples

were classified into contaminated and uncontaminated

category. 18 wells were reported as uncontaminated (on

average 12 lg/L As) and 12 wells were found contami-

nated (129 lg/L As) throughout the year, while 5 wells

could be classified as either contaminated or uncontami-

nated depending on when they were sampled. Although the

number of wells that alternate between the contaminated

and uncontaminated classification is relatively small

(14%), distinct seasonal variation in As concentrations

occur in all wells. This suggests that investigations con-

ducted within the study area for the purpose of assessing

the health risk posed by As in groundwater should not rely

on a single round of water samples. In comparison to other

areas, As is mainly released to the groundwater due to

reductive dissolution of Fe-oxyhydroxides, a process,

which is probably enhanced by anthropogenic input of

organic carbon. The seasonal variation in As concentra-

tions appear to be caused mainly by dilution effects during

monsoon and post-monsoon. The relatively high concen-

trations of Mn (mean 0.9 mg/L), well above the WHO limit

(0.4 mg/L), also cause great concern and necessitate fur-

ther investigations.

Keywords Arsenic � Groundwater contamination �Seasonal variations

Introduction

In many regions of the world, especially India (Das et al.

1994), Bangladesh (Ahmed et al. 2004), Vietnam (Berg

et al. 2001), China (Sun 2004) etc., As concentrations in

groundwater are significantly higher than the limit of

10 lg/L set by World Health Organization (WHO 2006).

Arsenic contamination in groundwater of West Bengal and

Bangladesh is well documented by various workers

(Bhattacharya et al. 1997; Chakraborti et al. 1996;

Chandrasekharam et al. 2001; Norra et al. 2005; Stuben

et al. 2003; van Geen et al. 2008) and is described as the

world’s biggest arsenic calamity (Chatterjee et al. 1995).

The presence of naturally elevated levels of As in

groundwater is confirmed in seven Indian states, namely

West Bengal, Bihar, Uttar Pradesh, Assam, Jharkhand,

Chattisgarh and Madhya Pradesh (Das et al. 1996; Chak-

raborti et al. 2003; Ahamed et al. 2006; Paul and Kar 2004;

Bhattacharjee et al. 2005; Acharyya et al. 2005; Chakrab-

orti et al. 1999). In West Bengal, investigations suggest

that nine districts, namely Malda, Murshidabad, Burdwan,

Nadia, Hoogly, Howrah, Kolkata, 24 Parganas (North) and

Electronic supplementary material The online version of thisarticle (doi:10.1007/s12665-010-0516-4) contains supplementarymaterial, which is available to authorized users.

S. H. Farooq (&) � D. Chandrasekharam � P. Thambidurai

Department of Earth Sciences, Indian Institute of Technology

Bombay, Powai, Mumbai 400076, India

e-mail: [email protected]

S. H. Farooq � S. Norra � Z. Berner � E. Eiche � D. Stuben

Institute of Mineralogy and Geochemistry, Karlsruhe Institute

of Technology, Karlsruhe 76131, Germany

123

Environ Earth Sci

DOI 10.1007/s12665-010-0516-4

24 Parganas (South), are the most severely affected; and

nearly 50 million people are at risk due to consumption of As

contaminated water (http://soesju.org/arsenic/wb1.htm).

However, the processes causing the high As concentrations

in the affected aquifers are not fully understood yet, but it is

estimated that about 200 million people in Asia are exposed

to As contaminated drinking water (Sun et al. 2006).

The presence of As in groundwater is associated with a

number of adverse effects on human health. The USEPA

considers As to be a human carcinogen (US Environmental

Protection Agency, 1996). Long time exposure of As may

cause various diseases such as melanosis, keratosis, non-

petting oedema, gangrene, leucomelanosis, circulatory

system problems and an increased cancer risk, especially of

skin, bladder and lungs (Arnold et al. 1990; Chen et al.

1996; Karim 2000). A review of the health problems

associated with consumption of As is given in a report by

WHO (2006). Apart from clinical symptoms, a number of

social and societal problems are aggravating the situation

(Bhattacharya et al. 2002; Ahmed et al. 2007; Nriagu et al.

2008). Dissolution of marriages and avoidance of arseni-

cosis patients are reported in many areas.

The severity of the As problem in West Bengal and

Bangladesh (jointly called as Bengal delta plain) has

prompted many governmental and non-governmental

organizations to come forward and act immediately. As a

result, millions of wells were sampled and, depending upon

the results of analysis, areas were labeled as As contami-

nated or safe. These samples were, however, collected

without taking into consideration that As concentrations

may vary throughout the year due to seasonal effects (pre-

monsoon, monsoon and post-monsoon). Only a few studies

were conducted on seasonal variations of As concentrations

in West Bengal and Bangladesh, some of them (McArthur

et al. 2004; Kinniburgh and Smedley 2001; Cheng et al.

2005; Wagner et al. 2005) suggest absence of any signifi-

cant change in As concentration, while others (Savarimu-

thu et al. 2006; Yakota et al. 2001) show very pronounced

seasonal variations. Dhar et al. (2008) suggested that the

temporal variability of dissolved As concentrations is a

function of water age, which is closely related to specific

geomorphic units in which groundwater occurs. However,

additional factors might also play a role.

This study has been undertaken to (1) investigate whe-

ther As concentrations in groundwater fluctuate in response

to seasons in the study area, (2) assess if such fluctuations

are sufficient to cause an area to be classified as uncon-

taminated during one season but contaminated during the

rest of the year. Areas that experience large seasonal

fluctuations in As concentrations might require more

extensive monitoring programs before the groundwater can

be labeled as safe or contaminated for the affected

residents.

Sampling and analytical methods

Study area

The study area is located in the eastern part of Murshidabad

district (latitude: N24�1400–N24�1700, longitude: E88�3100–88�4300), West Bengal, India and occurs within the delta

plain of the meandering river Padda (a tributary of the river

Ganga). In the east, the area is confined by the international

boundary with Bangladesh (Fig. 1). The general ground-

water flow direction in West Bengal is north-west to south-

east (Das et al. 1996). Three interconnected aquifer systems

exist in the Bengal delta plain. The shallowest aquifer

extends up to 12–15 m below the groundwater level and is

unconfined in most parts of the delta. The shallow aquifer is

mainly composed of fine to medium grained sands with

occasional intercalation of clay lenses. The intermediate

(35–46 m) and deep aquifers (70–150 m) are reported to be

at shallower depth in the Murshidabad district (Stuben et al.

2003) where the study area is situated. For all domestic

purposes groundwater is mainly pumped from tube-wells

placed mostly in the shallow aquifer and occasionally in the

intermediate aquifer.

Water sampling and analysis

To investigate the magnitude of seasonal variability in As

concentrations, water samples were collected from 35 wells

(33 tube wells and 2 open wells) during the years 2005–

2006 on a random basis. Each well was sampled once for

all the three seasons: pre-monsoon (March 2005), monsoon

(August 2005) and post-monsoon (January 2006). These

sampled wells (mostly shallow tube wells, B30 m; as

shown in supplementary data sheet 1) are mainly used for

drinking water purposes (Fig. 1). Out of the 35 sampled

wells, 21 are actively used domestic wells, 11 are wells of

public utility placed near schools, bus stops etc., two are

open wells and one is used as irrigation well. The depth of

these wells varies from 3 to 98 m (24 wells are up to

20 m deep (shallow aquifer), 4 wells are between 21 and

40 m deep (mainly intermediate aquifer), 7 wells are

41–100 m deep (deep aquifer) which means that water

from all three aquifers is exploited. The two sampled open

dug wells (well number 7 and 31) are 5 and 3 m deep,

respectively. Information about the depth of wells was

acquired through personal communication with the well

owners. Each well site was geo-referenced using a Garmin

Global Positioning System (GPSMAP-76S) instrument.

Water samples from all wells include the collection of: (1)

filtered samples (0.45 lm cellulose nitrate filter) for analyses

of major anions; and (2) filtered (0.45 lm cellulose nitrate

filter) and acidified (with 5 mL 14 M ultrapure HNO3/L)

samples for major cation and trace element analyses. All

Environ Earth Sci

123

samples were tightly sealed and stored at low temperature

until further analyses.

The tube wells, which were not in active use, were

purged for a minimum of three casing volumes before

sampling, while actively used ones were purged for a

minimum of 2 min. pH values, temperature and electric

conductivity (EC) were measured in the field instantly after

collecting the samples. The pH value was measured with a

portable pH-meter (Orion 261S) using a combination

electrode (pH C2401–7) and EC by a conductivity meter

(Orion 250A?) with an operating range between 0 and

500 mS/cm.

Analysis for all major (cations) and trace elements was

done by high resolution ICP-MS (Axiom, Thermo/VG

Elemental, UK) at the Institute of Mineralogy and Geo-

chemistry, Karlsruhe Institute of Technology, Germany.

Sulfate (SO42-) concentrations were measured by spec-

trophotometer (Shimadzu UV-Visible spectrophotometer

160), alkalinity by titration and chloride (Cl-) by

Expandable Ionanalyzer 940A with a combination elec-

trode Orion ionplus 9817 BN.

Drinking water guidelines

The resulting dissolved As concentrations were compared

with the Permissible Indian Limit (PIL) and the WHO

drinking water standards. The Government of India has set

a provisional water quality standard of 50 lg/L of As in

drinking water, which is five times higher than the WHO

limit of 10 lg/L, the limit also set by various developed

countries. Effects of As consumption on human health are

not uniform and are controlled by many factors, such as

nutritional value of daily diet (Chen et al. 1988; National

Research Council 1999), standard of living etc.; thus, no

unambiguous threshold value can be defined as the

dividing line between ‘‘safe’’ and ‘‘unsafe’’. Further, WHO

guidelines for water quality apply to ‘‘finished water’’. For

most study wells, which were primarily domestic wells,

‘‘finished water’’ may not be equivalent to filtered water.

The data presented here may be slightly biased towards low

As concentrations as the samples were filtered to minimize

the colloids and to remove the sediments entrained during

pumping.

Results

In general, groundwater in the study area is nearly neutral

to mildly alkaline. The pH value ranges from 6.8 to 7.7 for

all three seasons (Table 1). During pre-monsoon season pH

values range from 7.2 to 7.7 (mean 7.4), while for monsoon

and post-monsoon the values range from 6.8 to 7.5 (mean

7.2) and 6.9–7.6 (mean 7.2), respectively.

Considerable temporal variations are observed for all

major ions (Table 1). Ca2? (55.3–276 mg/L) is the domi-

nating cation and HCO3- (231–958 mg/L) is the major

anionic species present in the groundwater irrespective of

seasons. Concentrations of other major cations such as Na?

(2.7–108 mg/L), K? (2.5–50.6 mg/L) and Mg2? (8.7–

123 mg/L) and anions such as Cl- (7.2–157 mg/L) and

SO42- (10.6–129 mg/L) also show a considerable vari-

ability. Mean SO42- concentrations for pre-monsoon,

monsoon and post-monsoon are 38.1, 36.2 and 34.7 mg/L,

respectively. As a whole, SO42- concentrations in the study

area are quite high (mean 36.3 mg/L) compared to other As

affected areas (Eiche et al. 2008; Smedley and Kinniburgh

2002), where they are frequently well below detection

limit. Chloride shows considerable seasonal variations in

mean concentrations (pre-monsoon 54 mg/L, monsoon

42 mg/L and post-monsoon 40.5 mg/L), as the mean Cl-

Fig. 1 Area map showing locations of water sample collection wells

Environ Earth Sci

123

concentration for pre-monsoon is 33% higher than for post-

monsoon.

The major ion composition, plotted on a piper diagram

(Fig. 2), indicates that the groundwater is mainly of

Ca–HCO3 type, which is typical for As affected water bodies

in young Asian deltaic aquifers (Eiche et al. 2008). The piper

diagram indicates that there are no major seasonal differ-

ences in the hydrochemical characteristics of wells. Data

clustering suggests that the groundwaters evolved mainly

from a rainfall derived recharge which percolated through

unsaturated sandy sediments and interacted with the sedi-

ments for a relatively short period of time.

Iron concentrations vary from 91 lg/L to 150 mg/L.

With respect to seasonal variations, most of the samples

show highest Fe concentrations during pre-monsoon. The

mean concentrations during pre-monsoon, monsoon and

post-monsoon are 36.0, 23.8 and 23.8 mg/L, respectively

(Table 1). Considerable seasonal variations in Mn con-

centrations are also observed, but are less prominent when

compared to Fe concentrations; and the mean seasonal

concentrations range between 0.8 and 1.1 mg/L. In general,

the mean Mn concentration show a relative decrease of

24% from pre-monsoon to post-monsoon, while in the case

of Fe a relative decrease of 34% is noticed from pre-

monsoon to post-monsoon.

Arsenic concentrations vary from less than 1 to 341 lg/L

(median 31.3 lg/L). 49% of the sampled wells shows As

concentrations above the limit of 50 lg/L either in one or all

the seasons throughout the year. The mean As concentrations

for pre-monsoon, monsoon and post-monsoon are 63.2, 59.2

and 54.9 lg/L, respectively. Like most of the elements

(except U and Na) mean As concentrations also show a clear

decreasing trend from pre-monsoon through monsoon to

post-monsoon (Table 1).

Discussion

To evaluate the seasonal variability, the water samples

were divided into three groups: (1) Group-L, having As

concentrations\50 lg/L throughout the year (2) Group-H,

those having As concentrations C50 lg/L throughout the

year and (3) Group-V, those having As concentrations

\50 lg/L during some parts of the year and C50 lg/L

Table 1 Statistical summary of the chemistry of groundwater (major and trace elements; n = 35) from Nabipur block, Murshidabad, West

Bengal

Parameter Pre-monsoon Monsoon Post-monsoon

Mean Median Min. Max. Standard

deviation

Mean Median Min. Max. Standard

deviation

Mean Median Min. Max. Standard

deviation

pH 7.4 7.4 7.2 7.7 0.1 7.2 7.2 6.8 7.5 0.2 7.2 7.2 6.9 7.6 0.2

HCO3- (mg/L) 569 559 260 958 151 492 488 312 847 112 492 470 231 694 104

Cl- (mg/L) 54.0 39.4 7.9 157 43.4 42 27.3 7.8 113 33.8 40.5 25.5 7.2 137 37.0

SO42- (mg/L) 38.1 32.6 10.6 129 25.7 36.2 27.7 19 91.0 20.1 34.7 25.6 15.9 83.4 19.1

Na? (mg/L) 27.2 21.5 2.7 89.8 19.3 27.7 19.0 6.9 108 22.5 25.3 17.6 5.1 97.0 19.8

K? (mg/L) 9.2 7.0 2.6 46.8 7.7 8.3 6.1 3.8 50.6 8.0 5.8 5.1 2.5 19.2 3.0

Mg2? (mg/L) 45.5 40.6 15.4 123 23.4 34.2 32.8 13.6 58.3 11.1 31.6 28.7 8.7 63.7 12.3

Ca2? (mg/L) 162 162 56 276 55.2 128 123 82.6 230 30.8 121 116 55.3 207 32.9

As (lg/L) 63.2 44.7 0.4 301 70.4 59.2 24.3 1.0 341 75.7 54.9 26.4 0.9 325 71.9

Fe (mg/L) 36.0 21.4 0.1 151 33.9 23.8 10.9 1.9 150 32.4 23.8 11.9 3.8 119 27.0

Mn (mg/L) 1.1 0.9 0.0 3.0 0.7 1.0 0.8 0.0 2.9 0.6 0.8 0.7 0.0 2.5 0.5

Cu (lg/L) 22.9 18.4 1.8 86.9 17.4 14.3 6.1 2.7 142 30.0 8.5 5.6 3.3 48 8.2

U (lg/L) 2.8 1.1 0.0 27.6 5.2 3.3 1.0 0.0 29.8 5.6 3.5 1.2 0.0 28.3 5.6

Fig. 2 Plot of the major ion composition of the groundwater samples

in a piper diagram

Environ Earth Sci

123

during other parts of the year (Table 2). The separation of

these groups is based on the limit of 50 lg/L set by the

Government of India to classify water as polluted or

unpolluted with regard to dissolved As concentrations. In

addition to low As values (mean 12.2 lg/L), Group-L is

characterized by lower concentrations of redox-sensitive

elements such as Fe (mean 15.3 mg/L) and Mn (0.9 mg/L),

compared to Group-H (As (mean) 128 lg/L, Fe (mean)

47.5 mg/L, Mn (mean) 1.1 mg/L; Table 2). It is worth

mentioning that the water of Group-L wells might be safe

for drinking with regard to As concentrations but in many

wells the concentration of other elements (e.g. dissolved

Mn etc.) exceeded the WHO limits, which makes the water

from these wells unacceptable for drinking purposes.

It is noteworthy to observe the difference in uranium (U)

concentrations between the two groups (Group L and H)

because the redox behavior of this element is generally

opposite to that of Fe and Mn. This is reflected by the

roughly four times higher mean U concentrations in Group-

L (5.2 lg/L) as compared to Group-H (Table 2). Group-L

has also roughly 10% lower HCO3- concentrations

(507 mg/L as compared to 547 mg/L in Group-H). All the

other measured elements have more or less similar mean

values. These hydrochemical characteristics indicate that

the higher As concentrations appear preferentially in the

context of a low redox hydrogeochemical environment.

Consequently, it is reasonable to assume that high pro-

portions of dissolved As and Fe results from the reductive

dissolution of Fe-oxyhydroxides. It is widely accepted that

Fe-oxyhydroxides in the Bengal delta plain sediments

serve as the source of As, which is released under reducing

conditions (Bagla and Kaiser 1996; Bhattacharya et al.

1997; Nickson et al. 1998; Ravenscroft et al. 2001).

However, the absence of a strong correlation between

dissolved As and Fe in the water samples collected for this

study suggest that processes other than reductive dissolu-

tion are also involved in the release of As to groundwater.

86% of wells do not change their category and stay in

either of the two groups (Group-L; safe or Group-H;

unsafe) throughout the year. In the remaining 14% of wells

(Group-III), As concentrations vary throughout the year to

the extent that they change their association from one

group to the other. Thus, these wells can be labeled as

uncontaminated in some parts of the year whereas in the

other parts of the year they can be labeled as contaminated.

In both groups (Group-L and Group-H) some wells show

considerably large seasonal variations with regard to As

concentration. The standard deviation for the wells within

these groups vary between 0.3–25 and 2.1–36.7, respec-

tively, indicating that the seasonal variability of the As

concentration is considerably higher for Group-H as com-

pared to Group-L. The standard deviation for As concen-

trations in Group-V ranges from 22.6 to 99.5. These

variances indicate that the extent of the change in As

concentration is always relatively high when a well shifts

between the L and H groups.

In order to get a better understanding of the seasonal

variations, samples were further grouped into (1) Group-A

and (2) Group-B, based on the standard deviation of their

As concentrations in different seasons. The samples that

show coefficients of variation (CV) in As concentrations

\20% were allocated to Group-A while the samples that

show higher CV in As concentrations ([20%) were allo-

cated to Group-B (Table 3). 17 samples (49%) belong to

Group-A which include 7 samples with less than 50 lg/L

of As and 10 samples with more than 50 lg/L of As. The

Table 2 Statistical summary of the chemistry of groundwater (major and trace elements; n = 35) after dividing them into different groups

Parameter Group-L (Unpolluted, n = 18) Group-H (Polluted, n = 12) Group-V (n = 5)

Mean Median Min. Max. Standard

deviation

Mean Median Min. Max. Standard

deviation

Mean Median Min. Max. Standard

deviation

pH 7.3 7.3 7.0 7.7 0.2 7.2 7.2 6.8 7.5 0.2 7.3 7.3 7.0 7.6 0.2

HCO3- (mg/L) 507 491 231 958 128 547 527 359 847 134 486 459 368 790 105

Cl- (mg/L) 49.4 35.7 7.2 129 35.7 51.8 27.2 10.1 157 43.9 16.4 10.3 7.5 65.3 14.5

SO42- (mg/L) 36.9 31.1 12.3 84.1 19.2 41.1 30.8 11.0 129 26.5 23.0 22.3 10.6 49.8 9.1

Na (mg/L) 28.0 21.7 2.7 108 22.2 28.8 18.4 8.8 83.3 20.9 17.2 17.0 11.2 32.1 4.6

K (mg/L) 9.0 6.6 2.6 50.6 8.9 6.8 6.0 3.6 17.3 2.8 5.4 5.4 2.5 10.0 1.9

Mg (mg/L) 37.8 35.7 8.7 123 19.9 39.0 35.2 19.6 78.2 15.3 29.8 27.8 17.6 61.6 10.6

Ca (mg/L) 135 122 55.3 276 45.0 147 138 81.6 253 45.1 118 112 74.1 227 36.2

As (lg/L) 12.2 4.9 0.4 49.9 14.2 128 111 51.3 341 73.5 61.8 56.6 1.4 180 59.2

Fe (mg/L) 15.3 10.7 2.2 82.4 15.2 47.5 32.2 6.0 151 41.5 25.9 18.3 0.1 69.1 23.6

Mn (mg/L) 0.9 0.7 0.0 3.0 0.8 1.1 1.1 0.4 2.0 0.5 0.8 0.8 0.3 1.2 0.3

Cu (lg/L) 13.6 7.9 3.3 86.9 15.2 19.9 8.6 2.7 142 30.5 10.0 7.4 1.8 29.0 7.6

U (lg/L) 5.2 1.9 0.0 29.8 6.8 1.14 0.4 0.0 8.2 1.73 1.0 0.3 0.0 5.7 1.7

Environ Earth Sci

123

remaining 18 samples are in Group-B. The samples in

Group-B include 11 samples with less than 50 lg/L of As,

2 samples with more than 50 lg/L of As and 5 samples that

have\50 lg/L of As in some parts of the year and[50 lg/L

of As in the remaining part of the year. In addition to As,

samples of Group-B show greater variances in redox sen-

sitive elements (Table 3). A higher degree of variance in

Fe along with Cu and U in Group-B, suggests that a wide

fluctuation in the redox conditions in the aquifer is

responsible for the greater seasonal variation in As con-

centrations. The redox potential data from many previous

studies demonstrate that mild oxidizing to moderately and/

or strongly reducing conditions exists in the Bengal delta

plain (BGS and DPHE 2001; Bhattacharya et al. 2002).

Our study suggests a slight but continuous decrease of

the As concentrations from the pre-monsoon to post-

monsoon season due to dilution with percolating rain water

(Table 1). During the period preceding the monsoon sea-

son, this decrease is possibly offset by processes that lead

to a new episodic increase of the As concentrations.

Among the possible mechanisms one may consider are (1)

the mobilization of As under reducing conditions (Smedley

and Kinniburgh 2002; Nickson et al. 2000) generated by

the microbially mediated decay of natural organic matter

(NOM) and/or dissolved organic carbon (DOC) (Harvey

et al. 2002) and (2) the mobilization of As by organic acids

(a fraction of DOC) that may play an important role in

mineral degradation and metal mobilization (Baker 1973;

Kalbitz and Wennrich 1998; Bauer and Blodau 2006). In

West Bengal high quantities of organic matter are left over

in the paddy fields due to traditional paddy cultivation. In

paddy cultivation, harvested crop is cut from the middle of

stem and the remaining half of the stem and roots are

ploughed back for the next cultivation. During the mon-

soon, paddy fields are flooded with rain water and the

decomposition of such plant remains increases the avail-

ability of DOC in general and of organic acids in particular.

The availability of DOC and production of organic acids

during the monsoon season is further enhanced by the local

jute industry (biggest jute producer in the world), which

follows the traditional practice of decomposing the jute

crop in ponds in order to get jute fibres. In this way sub-

stantial quantities of DOC and organic acids are produced

(as shown in supplementary data sheet 2), which can per-

colate down with rain water, and, on their way to

groundwater react with mineral surfaces or act as electron

donor for microbes. As a consequence, the reductive dis-

solution of Fe-oxyhydroxides would be enhanced. This

might be one cause of the extremely high dissolved

Fe-concentrations (\151 mg/L) in some wells. Further-

more, the presence of DOC modifies the solubility and

mobility of many contaminants (Sensei et al. 1994). The

negatively charged DOC, for example, increases the

desorption of As from the binding sites through electro-

static effects (Bauer and Blodau 2006). A slight but quasi

constant decrease of the pH values from pre-monsoon

(mean 7.4) to monsoon/post-monsoon (mean monsoon 7.2,

post-monsoon 7.2) further suggests percolation and mixing

of organic acids, though some other factors may also play a

considerable role in the fluctuation of pH values (Fig. 3).

Higher pH values during pre-monsoon coupled with

relatively minor seasonal fluctuations but with a clear

Table 3 Grouping of samples based on the variance in seasonal As concentration

Parameter Group-A (Coefficient of variance in As concentrations\20%)

(n = 17)

Group-B (Coefficient of variance in As concentrations [20%)

(n = 18)

Mean Median Min. Max. Standard

deviation range

CV Mean Median Min. Max. Standard

deviation range

CV

pH 7.2 7.2 6.8 7.5 0.1–0.2 0.8–3.4 7.3 7.3 7.0 7.7 0.1–0.3 1.3–3.5

HCO3- (mg/L) 532 503 348 858 12–209 3.0–34.0 504 479 231 958 32–265 5.0–41.0

Cl- (mg/L) 49.6 29.4 10.1 157 0.3–29.5 2.9–62.8 41.6 29.9 7.2 145 0.2–27 1.7–74.7

SO42- (mg/L) 40.5 30.9 11 129 1.2–28.4 3.8–64.4 32.4 25.6 10.6 84.1 0.6–14.8 2.4–51.5

Na (mg/L) 29.3 19.1 8.8 89.8 0.9–33.6 2.5–65.5 24.4 18.3 2.7 108 0.5–28.8 2.9–56.3

K (mg/L) 8.3 6.1 3.1 50.6 0.3–24.9 4.0–72.5 7.2 6.2 2.5 22.5 0.8–8.0 11.7–86.9

Mg (mg/L) 39.2 37.7 19.6 123 1.9–51.6 5.7–81.3 35.1 30.5 8.7 118 2.1–48.9 7.4–79.1

Ca (mg/L) 141 124 82 270 8.7–93.6 7.6–57.8 133 120 55 276 3.1–93.2 2.7–55.3

As (lg/L) 84.4 70.8 1.2 341 0.3–28.8 1.0–19.7 35.2 20.1 0.4 180 0.3–99.5 23.5–157.5

Fe (mg/L) 31.4 16 3 151 1.4–18.5 7.7–106 24.6 13 0.1 93.8 1.1–41.4 5.6–136.2

Mn (mg/L) 937 793 83 1,990 47.1–412 4.9–87.1 961 862 23 3,000 7.8–785 5.0–48.8

Cu (lg/L) 17.8 7.9 3.3 143 1.4–65.5 10.5–132 12.8 7.8 1.8 86.9 2.9–47.3 35.9–147.1

U (lg/L) 3.2 1.0 0.0 27.6 0–11.7 0.9–82.1 3.2 1.2 0.0 29.8 0–13.6 2.5–94.4

CV = SD 9 100/mean

Environ Earth Sci

123

decreasing trend in mean sulphate concentration (pre-

monsoon 38.1 mg/L, monsoon 36.2 mg/L and post-mon-

soon 34.7) argue against the formation of sulphate by pyrite

oxidation, because pyrite oxidation would imply the low-

ering of pH (Das et al. 1996).

The samples that remain in Group-L (As \ 50 lg/L)

throughout the year show only a week correlation between

Fe and As (r2 = 0.34) during monsoon; however, the wells

from Group-H (r2 = 0.42) and specially Group-V show a

very strong correlation (r2 = 0.97) between As and Fe

concentrations during monsoon period (Fig. 4). This sug-

gests that the release of As and its high seasonal variability

is controlled not only by rain water dilution, but also to

some extent by changes in the redox environment.

Although dilution effects alone could cause a covariance

between As and Fe, the simultaneous increase of their

concentrations during the pre-monsoonal season, certainly

should have a different explanation. Furthermore, the wells

that fall in Group-V shows a higher correlation (r2 = 0.70)

between As and HCO3- during monsoon, which suggests

that the release of As is associated with the degradation of

organic material. The microbial degradation of organic

matter leads to the formation of HCO3-. The displacing

effect of HCO3- for As sorbed onto Fe oxyhydroxides is

well documented (Appelo et al. 2002). Sometimes higher

concentrations of HCO3- fosters the release of As into the

groundwater (Kim et al. 2000) and, consequently, be a

further source for dissolved As.

Apart from various other factors, distance from pools

rich in organic matter and organic acids (paddy fields,

ponds and waste dumping sites) may also play an important

role in the seasonal variation of As concentrations in

groundwater. Open (dug) wells, which are often located

away from organic matter pools (agricultural fields/jute

decomposing ponds), were found to have relatively low As

concentrations, often well below (1.7–3.2 lg/L) than the

WHO prescribed limit. These wells are just 3–5 m deep, a

depth range in which direct recharge through vertical flow

dominates, while mixing and recharge through horizontal

flow is assumed to be negligible. The rain water/run-off

which percolates from the surface into the groundwater,

feeding these wells, has very low organic matter content

and As concentrations.

Differences among wells with respect to the seasonal

fluctuation of As concentrations may additionally be con-

trolled by various local factors, such as the screening depth

of tube wells, the presence of clay intercalations/pockets in

the cap-rocks of the aquifer, the proximity of organic acid

production sites and of waste dumping sites, etc. There is

no clear relationship between well depth and As concen-

tration. Pockets and intermittent layers of clayey sedi-

ments, which are typical for deltaic sequences, prolongs the

percolation time of surface water. Thus provides longer

interaction time with the sediments that have higher spe-

cific surface area and abundant reactive sites for adsorption

or desorption of As. There is no information about the

hydrogeology of each well in order to further establish a

possible relationship between well depth and hydrogeo-

logical units.

Out of the five wells which shift their category in

response to seasonal changes, well numbers 10 and 15

show a decrease in As concentration during the monsoon

season probably due to dilution by rainwater. In both of

these wells As concentrations increase during post-mon-

soon season (Fig. 5a, b), which might reflect a delay in the

development of sufficiently reducing conditions. The

development of reducing conditions in these wells is

indicated by a significant increase of the Fe and HCO3-

concentrations during the post-monsoon season (as shown

in supplementary data sheet 1).

Wells 25 and 34 show a prominent ‘‘dilution effect’’

during post-monsoon rather than the monsoon season

(Fig. 5c, d). Both of these wells are screened at a greater

depth (46 and 98 m, respectively) as compared to the other

sampled wells. The water table is at a very shallow depth in

the study area (3 m) and the mixing of percolating rain

6.6

6.8

7.0

7.2

7.4

7.6

7.8

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35

Well No.

pH

Pre-monsoonMonsoonPost-monsoon

Fig. 3 Variations in pH values in response to seasonal changes

R2(H)= 0.42

R2(V)= 0.97

R2(L)= 0.340

50

100

150

200

250

300

350

400

0 20 40 60 80 100 120 140 160

Fe (mg/L)

As

(ug

/L)

Group-L

Group-H

Group-V

Fig. 4 Graph showing Fe and As correlation during monsoon. The

samples which are allocated to Group-L show only a weak correlation

between Fe and As during monsoon (r2 = 0.34), while samples from

Group-H and Group-V show a moderate (r2 = 0.42, Group-II) to very

strong (r2 = 0.97; Group-V) correlation between Fe and As in the

same season

Environ Earth Sci

123

water with deeper groundwater is assumed to take longer

time. Clay lenses and clay patches that are common in

deltaic deposits exhibit very limited permeability. The slow

infiltration through such lithologies can effectively delay

the dilution of groundwater by precipitation derived water.

A significant increase in the As concentration during

monsoon in well 16 (Fig. 5e) could be due to some par-

ticular local conditions, which may affect As mobility.

Proximity of waste dumping site, heterogeneity of aquifer

characteristics, anthropogenic activity, etc. are potential

causes for such special conditions.

It is critical to monitor whether the inter-seasonal vari-

ability or total As concentrations increase with time;

therefore, it is necessary that monitoring should be carried

out over longer periods during all the three seasons. Even if

only 14% of the wells are misclassified by disregarding the

seasonal variability of As concentrations, the huge number

of wells in the Bengal delta means that such a wrong

labeling would affect the health of millions of people.

Savarimuthu et al. (2006) also point out seasonal variations

in As concentrations, but opposite to our results, the

highest As concentrations were found during monsoon and

the lowest in the pre-monsoon time. A monsoon season

with less intensive rainfall is likely to trigger greater sea-

sonal variations in As concentrations. A drop in average

rainfall will not only result in lesser dilution of As, but will

also lead to relatively high DOC and organic acids con-

centrations, thus inducing stronger reducing conditions.

The results of this study are only partly in line with the

findings of Dhar et al. (2008) conducted in Araihazar area

in Bangladesh, where over a monitoring period of

2–3 years the As concentrations in some wells were

increasing, while in others a decreasing trend was

observed, and some wells remained practically unchanged.

Conclusions

This study clearly shows that seasonal variability of dis-

solved As occurs in several wells but also suggests that a big

majority (86%) of wells did not show such a prominent

change in their As concentrations that they can be labeled as

contaminated in one season and uncontaminated in the

another. Nevertheless, a very high standard deviation in the

seasonal As concentrations of these wells highlight a dire

need for long term seasonal monitoring. Considering the

high population density in West Bengal and the huge number

of wells in the Bengal delta Plain, a wrong labeling of only a

low percentage of the wells can pose a severe health threat to

a considerably large human population. In general, this study

suggests a kind of cyclical seasonal pattern, characterized by

a decreasing trend of the As concentrations from the pre-

Fig. 5 a–e Pattern of As concentrations in wells that exhibit a change from one group to another in response to seasonal variations

Environ Earth Sci

123

monsoon to post-monsoon period, followed by an increase

during the shorter pre-monsoon time. The decreasing branch

of the cycle is dominated by dilution through run-off and by a

redox shift towards more oxic conditions, while the shorter

increasing branch is characterized by a more intensive water/

sediment interaction and the development of more reducing

conditions in the groundwater. There are also indications that

the well depth and lithology may have an influence on the

seasonal variability along with local peculiarities (dump

sites etc.). Arsenic release may also be heavily influenced by

the infiltration of organic carbon from agricultural processes;

however, its influence on the variability of dissolved As

concentrations could not be established from this study.

A more detailed study on temporal variations of As is

needed as it has important implications on the exposure and

precision of the risk assessments. In addition to the various

natural controls on As release from sediments to ground-

water, it is also important to consider the role that organic

matter plays in As mobility.

Acknowledgments The authors gratefully acknowledge support from

German Academic Exchange Service (DAAD) through research fel-

lowship. Our thanks are also due to Mrs. Claudia Mossner of the Institute

of Mineralogy and Geochemistry (IMG) for the chemical analyses.

References

Acharyya SK, Shah BA, Ashyiya ID, Panday Y (2005) Arsenic

contamination ion groundwater from parts of Ambagarh chowki

block, Chattisgarh, India: source and release mechanism.

Environ Geol 49:148–158

Ahamed S, Sengupta MK, Mukharjee A, Hossain AM, Das B, Nayak

B, Pal A et al (2006) Arsenic groundwater contamination and its

health effects in the state of Uttar Pradesh (UP) in upper and

middle Ganga plain, India: a severe danger. Sci Total Environ

370:310–322

Ahmed KM, Bhattacharya P, Hasan MA, Akhter SH, Alam SMM,

Bhuyian MAH, Imam MB, Khan AA, Sracek O (2004) Arsenic

enrichment in groundwater of the alluvial aquifers in Bangla-

desh: an overview. Appl Geochem 19:181–200

Ahmed S, Sayed MHS, Khan MH, Karim MN, Haque MA, Bhuiyan

MSA, Rahman MS, Faruquee MH (2007) Sociocultural aspects

of arsenicosis in Bangladesh: community perspective. J Environ

Sci Health A 42:1945–1958

Appelo CAJ, van der Weiden MH, Tournassat C, Charlet L (2002)

Surface complexation of ferrous ion and bicarbonate on ferrihy-

drite and the mobilization of arsenic. Environ Sci Technol

36:3096–3103

Arnold HL, Odam RB, James WD (1990) Disease of the skin: clinical

dermatology, 8th edn. WB Saunders, Philadelphia, p 121

Bagla P, Kaiser J (1996) India’s spreading health crisis draws global

arsenic experts. Science 274(5285):174–175

Baker WE (1973) The role of humic acids from Tasmanian podzolic

soils in mineral degradation and metal mobilization. Geochim

Cosmochim Acta 37:269–281

Bauer M, Blodau C (2006) Mobilization of arsenic by dissolved

organic matter from iron oxides, soils, and sediments. Sci Total

Environ 354:179–190

Berg M, Tran HC, Nguyen TC, Pham HV, Schertenleib R, Giger W

(2001) Arsenic contamination of groundwater and drinking

water in Vietnam: a human health threat. Environ Sci Technol

35:2621–2626

BGS and DPHE (2001) Arsenic contamination of groundwater in

Bangladesh, vol 2. Final Report, BGS Technical Report WC/00/

19

Bhattacharjee S, Chakravarty S, Maity S, Dureja V, Gupta KK (2005)

Metal contents in the groundwater of Sahebganj district,

Jharkhand, India, with special reference to arsenic. Chemosphere

58:1203–1217

Bhattacharya P, Chatteriee D, Jacks G (1997) Occurrence of arsenic

contaminated groundwater in alluvial aquifers from the Delta

Plains, eastern India: options for safe drinking water supply. Int J

Water Resour Manag 13:79–92

Bhattacharya P, Jacks G, Ahmed KM, Khan AA, Routh J (2002)

Arsenic in groundwater of the Bengal Delta Plain aquifers in

Bangladesh. Bull Environ Contam Toxicol 69:538–545

Chakraborti D, Das D, Samanta BK, Mandal BK, Chowdhury RT,

Chanda CR, Chowdhury PP, Basu GK (1996) Arsenic in

groundwater in six districts of West Bengal, India. Environ

Geochem Health 18:5–15

Chakraborti D, Biswas BK, Basu GK, Mandal BK, Chowdhury UK,

Mukherjee S, Gupta SD, Chaudhury SR, Rathore KC (1999)

Arsenic groundwater contamination and sufferings of people in

Rajnandgaon district, Madhya Pradesh, India. Curr Sci 77:502–

504

Chakraborti D, Mukherjee SC, Pati S, Sengupta MK, Rahman MM,

Chaudhury UK et al (2003) Arsenic groundwater contamination

in Middle Ganga plain, Bihar, India: a future danger. Environ

Health Prospect 111:1194–1201

Chandrasekharam D, Karmakar J, Berner Z, Stuben D (2001) Arsenic

contamination in groundwater, Murshidabad district, West

Bengal. In Cidu A (ed) Proceedings of the 10th international

symposium on water-rock interaction, vol 2, pp 1051–1054

Chatterjee A, Das D, Mandal B (1995) Arsenic in groundwater of six

districts of West Bengal: the biggest arsenic calamity in the

world. Part 1. Arsenic species in drinking water and urine of

affected people. Analyst 120:640–650

Chen CJ, Wu MM, Lee SS, Wang JD, Cheng SH, Wu HY (1988)

Atherogenicity and carcinogenicity of high-arsenic artesian well

water: multiple risk factors and related malignant neoplasms of

blackfoot diseases. Arteriosclerosis 8:452–460

Chen CJ, Chiou HY, Chiang MH, Lin LJ, Tai TY (1996) Dose-

response relationship between ischemic heart disease mortality

and long-term arsenic exposure. Arterioscler Thromb Vasc Biol

16(4):504–510

Cheng Z, van Geen A, Seddique AA, Ahmed KM (2005) Limited

temporal variability of arsenic concentrations in 20 wells

monitored for 3 years in Araihazar, Bangladesh. Environ Sci

Technol 39:4759–4766

Das D, Chatterjee A, Samanta G, Mandal B, Chowdhury TR, Samanta

G, Chowdhury PP, Chanda C, Basu G, Lodh D, Nandi S,

Chakraborty T, Mandal S, Bhattacharya SM, Chakraborti D

(1994) Arsenic contamination in ground water in six districts of

West Bengal: the biggest arsenic calamity in the world. Analyst

119:168–170

Das D, Samanta G, Mandal BK, Chowdhury TR, Chanda CR,

Chowdhury PP, Basu GK, Chakraborti D (1996) Arsenic in

groundwater in six districts of West Bengal, India. Environ

Geochem Health 18:5–15

Dhar RK, Zheng Y, Stute M, van Geen A, Cheng Z, Shanewaz M,

Shamshudduha M, Haque MA, Rahman MW, Ahmed KM

(2008) Temporal variability of groundwater chemistry in shallow

and deep aquifers of Araihazar, Bangladesh. J Contam Hydrol

99:97–111

Environ Earth Sci

123

Eiche E, Neumann T, Berg M, Weinman B, van Geen A, Norra S,

Berner Z, Pham TKT, Pham HV, Stuben D (2008) Geochemical

processes underlying a sharp contrast in groundwater arsenic

concentrations in a village on Red River Delta, Vietnam. Appl

Geochem 23:3143–3154

Harvey CH, Swartz C, Badruzzaman ABM, Keon-Blute N, Yu W, Ali

MA, Jay J, Beckie R, Niedan V, Brabander D, Oates P, Ashfaque

KN, Islam S, Hemond HF, Ahmed MF (2002) Arsenic mobility

and groundwater extraction in Bangladesh. Science 298:1602–

1606

Kalbitz K, Wennrich R (1998) Mobilization of heavy metals and

arsenic in polluted wetland soils and its dependence on dissolved

organic matter. Sci Total Environ 209:27–39

Karim M (2000) Arsenic in groundwater and health problems in

Bangladesh. Water Resour 34:304–310

Kim MJ, Nriagu J, Haack S (2000) Carbonate ions and arsenic

dissolution by groundwater. Environ Sci Technol 34:3094–3100

Kinniburgh DG, Smedley PL (2001) Arsenic contamination of

groundwater in Bangladesh, vol 2. Final Report, British

Geological Survey, BGS Technical Report No WC/00/19

McArthur JM, Banerjee DM, Hudson-Edwards KA, Mishra R,

Purohit R, Ravenscroft P, Cronin A, Howarth RJ, Chatterjee

A, Talukder T, Lowry D, Houghton S, Chadha DK (2004)

Natural organic matter in sedimentary basins and its relation to

arsenic in anoxic ground water: the example of West Bengal and

its worldwide implications. Appl Geochem 19:1255–1293

National Research Council (1999) Arsenic in drinking water. National

Academy Press, Washington, DC

Nickson RT, McArthur JM, Ravenscroft P, Burgess WG, Ahmed KM,

Rahman M (1998) Arsenic poisoning of Bengal groundwater.

Nature 395:338

Nickson RT, McArthur JM, Ravenscroft P, Burgess WG, Ahmed KM

(2000) Mechanism of arsenic release to groundwater, Bangla-

desh and West Bengal. Appl Geochem 15:403–413

Norra S, Berner Z, Aggarwala P, Wagner F, Chandrasekharam D,

Stuben D (2005) Impact of irrigation with As-rich groundwater

on soil and crops: a geochemical case study in Maldah District,

West Bengal. Appl Geochem 20:1890–1906

Nriagu JO, Bhattacharya P, Mukherjee AB, Bundschuh J, Zevenho-

ven R, Loeppert RH (2008) Arsenic in soil and groundwater: an

overview. In: Bhattacharya P, Mukherjee AB, Bundschuh J,

Zevenhoven R, Loeppert R (eds) Trace metals and other

contaminants in the environment. Elsevier, The Netherlands,

pp 3–63

Paul AB, Kar D (2004) Ground water arsenic in Assam: a report from

Brahmaputra and Barak valley. Environ Ecol 22:588–589

Ravenscroft P, McArthur JM, Hoque BA (2001) Geochemical and

paleohydrological controls on pollution of groundwater by

arsenic. In: Chappell W, Abernathy CO, Calderon R (eds)

Arsenic exposure, heals effects (IV). Elsevier, Oxford, pp 78–83

Savarimuthu X, Hira-Smith MM, Yuan Y, von Ehrenstein OS, Das S,

Ghosh N, Guha DN, Smith AH (2006) Seasonal variation of

arsenic concentration in tube wells in West Bengal, India. J

Health Popul Nutr 24:1–5

Sensei N, Brunetti G, La CP, Miano TM (1994) Adsorption of

alachlor by humic acids from sewage sludge and amended and

non-amended soils. Soil Sci 157:176–184

Smedley PL, Kinniburgh DG (2002) A review of the source,

behaviour and distribution of arsenic in natural waters. Appl

Geochem 17:517–568

Stuben D, Berner Z, Chandrasekharam D, Karmakar J (2003) Arsenic

enrichment in groundwater of West Bengal, India: geochemical

evidence for mobilization of As under reducing conditions. Appl

Geochem 18:1417–1434

Sun G (2004) Arsenic contamination and arsenicosis in China.

Toxicol Appl Pharmacol 198:268–271

Sun G, Li X, Pi J, Sun Y, Li B, Jin Y, Xu Y (2006) Current research

problems of chronic arsenicosis in China. J Health Popul Nutr

24(2):176–181

van Geen A, Radloff K, Aziz Z, Cheng Z, Huq MR, Ahmed KM,

Weinman B, Goodbred S, Jung HB, Zheng Y, Berg M, Trang

PTK, Charlet L, Metral J, Tisserand D, Guillot S, Chakraborty S,

Gajurel AP, Upreti BN (2008) Comparison of arsenic concen-

trations in simultaneously-collected groundwater and aquifer

particles from Bangladesh, India, Vietnam, and Nepal. Appl

Geochem 23:3244–3251

Wagner F, Berner ZA, Stuben D (2005) Arsenic in groundwater in

Bengal Delta Plain: geochemical evidences for small scale redox

zonation in the aquifer. In: Bundschuh J, Bhattacharya P,

Chandrasekharam D (eds) Natural arsenic in groundwater:

occurrence, remediation and management. Taylor & Francis

Group, London, pp 3–15

World Health Organization (2006) Guidelines for drinking-water

quality. Recommendations, vol 1, 3rd edn. World Health Orga-

nization, Geneva

Yakota H, Tanabe K, Sezaki M, Akiyoshi Y, Miyata T, Kawahara K,

Tsushima S, Hironaka H, Takafuji H, Rahaman M, Ahmad SKA,

Sayed MHSU, Faruquee MH (2001) Arsenic contamination of

ground and pond water and water purification system using pond

water in Bangladesh. Eng Geol 60:323–331

Environ Earth Sci

123