NATURAL ARSENIC IN GROUNDWATER: OCCURRENCE, REMEDIATION AND MANAGEMENT - CHAPTER 11 pot

Section 2: Environmental health
assessment-arsenic in the food chain
Copyright © 2005 Taylor & Francis Group plc, London, UK
95
Natural Arsenic in Groundwater: Occurrence, Remediation and Management –
Bundschuh, Bhattacharya and Chandrasekharam (eds)
© 2005, Taylor & Francis Group, London, ISBN 04 1536 700 X
Arsenic in groundwater and contamination of the food chain:
Bangladesh scenario
S.M. Imamul Huq
Department of Soil, Water & Environment, University of Dhaka, Dhaka, Bangladesh
Ravi Naidu
Centre for Environmental Risk Assessment and Remediation (CERAR), University of South Australia,
Mawson Lake Campus, Adelaide, Australia
ABSTRACT: Ingestion of arsenic (As) contaminated groundwater is the major cause of As poi-
soning in Bangladesh. However, poisoning among the population is not consistent with the level
of water intake. Moreover, there is also a spatial variation of the manifestation of arsenicosis in the
country. This has raised the question about the role of the food habit, nature and amount of food
intake in the As dilemma. Even if an As-safe drinking water supply is ensured yet, the same As
contaminated groundwater will continue to be the main source of irrigation for about 40% of the
net cultivable area as more than 60% of irrigation water comes from groundwater. This leaves a
risk of soil accumulation of the toxic element and eventual exposure of the food chain to As con-
tamination through plant uptake and animal consumption. Water (hand tube wells and irrigation
pumps), soil and vegetables/crop samples collected from as many as 150 different locations cover-
ing 15 districts of the country have revealed that the average background concentration of
As in Bangladesh is much below 10mg/kg soil. However, in some areas where soils receive As-
contaminated groundwater irrigation, the concentration has been found to be as high as 80 mg/kg
soil. The soil As varies both spatially and vertically. The soil formation and the aquifer characters
control the spatial variation, while the vertical distribution is controlled by the clay contents. The
maximum As concentration in irrigation water was found to be 0.55 mg/L; irrigating a rice field
with this water when the requirement is 1000 mm of water, it has been calculated that the As load

will come to 5 kg As/ha/yr. Many crops receiving As contaminated water as irrigation have been
found to accumulate As at levels that exceed the maximum allowable daily limit (MADL) of 0.2 mg
per kg dry weight (dw). Some vegetables crops like Arum (Colocassia antiquorum), Kalmi (Ipomea
aquatica), Amaranthus (Amaranthus spp.) etc. were found to be As accumulators. The transfer fac-
tor for As has been found to exceed the value of 0.1 in a number of plants indicating their affinity
towards this element. In Arum, the concentration of As have been found to be as high as more than
150 mg/kg dw. Rice and wheat receiving As-contaminated irrigation water have been found to
sequester the toxic metalloid into roots and stems. However, the quantities of rice consumed per
person per day with the content of As in the grain may in many instances, surpass the MADL.
1 INTRODUCTION
Contamination of groundwater by arsenic (As) in the deltaic region, particularly in the Gangetic
Alluvium of Bangladesh has become one of the world’s most important natural calamities. In
many areas of the country, water containing more than 0.05 mg/L As, a limit value set for drinking
water has been reported (DPHE-BGS 1999). Efforts are being directed towards ensuring safe
drinking water either through mitigation technique or through finding alternative sources. Even if an
As-safe drinking water supply is ensured the same groundwater will continue to be used for irrigation
Copyright © 2005 Taylor & Francis Group plc, London, UK
purpose, leaving a risk of soil accumulation of this toxic element and eventual exposure to the
food chain through plant uptake and animal consumption. Between 30 to 40% of net cultivable
land is under irrigation and more than 60% of this irrigation is met from groundwater.
Studies in the last few years have focused on ingestion of As through intake of groundwater
containing As. The observation that As poisoning among the population is not consistent with the
level of water intake has raised question on potential pathways of As ingestion. In a preliminary
study from a village of Laksimpur, Huq et al. (2001) observed that certain vegetables growing on
soil supposed to be affected by As contamination, accumulated substantial amount of As in them.
This observation prompted the authors to undertake further study to see the loading of arsenic in
soils from irrigation water and subsequently to different crops growing on those soils.
2MATERIALS AND METHODS
For the study, information on As contamination in Bangladesh was obtained from secondary sources
(DCH 1997, DPHE/BGS 1999). On the basis of the information, some areas were identified as con-

trol (where groundwater As contamination was not reported), less affected, moderately affected,
severely affected (according to the incidence of As patients). Thus water, soil, and vegetables/crop
samples were collected from 143 locations, covering 13 districts of Bangladesh (Huq & Naidu 2003).
Water samples from hand tube-wells or irrigation pumps were collected. Soon after collection of water
samples 2–3 drops of concentrated HCl was added to the vials containing water and transported to
the laboratory for further analysis.
Surface (0–15 cm) and subsurface (15–30 cm) soil samples were collected in replicate from
locations covering the alluvial flood plains of Ganges, Teesta and Meghna-Brahmaputra rivers as
well as from the Pleistocene terraces (Huq et al. 2003). Arsenic poisoning was reported from all
these areas except the Pleistocene terraces, which was taken as control for comparison. In order to
monitor the As load on soils from water, samples were collected from regions subjected to hand
tube-well, shallow tube-well, and surface water irrigation. Random grid sampling was adopted
and from each site the number of samples collected ranged from 25 to 40 per acre. After collec-
tion, samples were air dried, ground and screened to pass through 0.5 mm sieve and stored in plas-
tic vials and set for further analysis. Replicate samples of consumable parts of the vegetables/crops
commonly grown in the sampling area were collected.
All plant samples were dusted free of adhering soil particles, washed with deionized water and
0.05 M HCl and then washed with deionized water 3 times to ensure dislodging of adhering dust
particles. Samples were then dried in oven at 60 Ϯ 5°C for 48 h, ground, screened to pass through
0.2 mm sieve, stored in plastic vials, and kept for further analysis. The As in soil was extracted by
digesting with aqua regia while As in plant samples was extracted with HNO
3
digestion (Portman &
Riley 1964). Arsenic in water, soil extract, and plant extract was estimated by HG-AAS technique and
certified reference materials were used to ensure QA/QC.
3 RESULTS AND DISCUSSION
The As concentration in water used for irrigation was found to vary between 0.14 to 0.55mg/L.
So, for a Boro rice requiring 1000 mm of irrigation water per season, the load of As comes to
between 1.36 to 5.5 kg/ha/yr. Similarly, with winter wheat requiring 150 mm of irrigation water
per season, the load of As is estimated to be between 0.12 to 0.82kg/ha/yr. The results on soil As

indicated that the collected soils were not contaminated with As, and contained less than 10 mg/kg
As. Expressed in terms of kg per hectare; the values did not exceed 20kg/ha (Fig. 1). Moreover,
the As concentration was higher in the surface layer than the sub-surface layer with a few excep-
tions in soils collected from Meghna Alluvium. On the other hand, soils from arsenic contam-
inated area show higher values ranging from around 2 to Ͼ80 mg/kg. It needs to be mentioned
96
Copyright © 2005 Taylor & Francis Group plc, London, UK
here that soils belonging to the Gangetic Alluvium contained higher amount of As than the soils
belonging to the Teesta Floodplain. Usually in soils contaminated through anthropogenic activity
the arsenic contents may rise up to 50 mg/kg. In the present study, although the source of As is
geogenic (Nickson 1998) yet, in some cases the values were equal to that of anthropogenic activ-
ity. This is an indication of As accumulation in the soil due to irrigation.
Results on As contents of the analyzed vegetables/crops showed that some of the vegetables/crops
accumulated As in the plant tissues (Table 1). It is also apparently clear that plants of the same type
growing on uncontaminated soil had much less As content in their tissues.
For rice and wheat, most of the As taken up by the plants were sequestered in the roots and
stems, an insignificant amount was found to have accumulated in the grains (Fig. 2). But some
leafy vegetables, particularly arum (Colocassia antiquorum) appeared to be an As accumulator.
This was true for all the areas studied (Fig. 3), indicating that As from groundwater was entering
into the food chain through soil to crops transfer.
The levels of As in plants seldom exceed 1mg/kg (Markert 1992). In the present study, quite a
few plant samples had values much higher than this level. Farago & Mehra (1992) have considered
that when the plant/soil ratios for any particular element are 0.1, then the plant can be considered
as excluding the element from its tissues. In our case, many plants have shown this phenomenon while
some like arum and a number of leafy vegetables had shown the reverse phenomenon indicating
97
0
5
10
15

20
Old
Pleistocene
Alluvium soil
Teesta
Alluvium soil
Megna
Alluvium soil
Gangetic
Alluvium soil
Soil As (kg/ha)
0-15 cm
15-30 cm
Figure 1. Arsenic content in different soils of Bangladesh.
Table 1. Arsenic content (mg/kg dw) in common plants from uncontaminated and
contaminated areas of Bangladesh.
Uncontaminated Contaminated
Name of plants areas areas
Green papaya (Carica papaya) 0.46 2.22
Arum (Colocassia antiquorum) 0.39 153.2
Bean (Dolicos lablab) 0.092 1.16
Indian spinach (Brasilia alba) 0.15 1.00
Long bean (Vicia faba) 0.30 2.83
Potato (Solanum tuberosum) 0.62 2.43
Bitter gourd (Momordicum charantia) 1.56 2.12
Aubergine (Solanum melongena) 0.23 2.3
Chili (Capsicum spp) 0.41 1.52
Copyright © 2005 Taylor & Francis Group plc, London, UK
their affinity to As accumulation (Table 2). It was observed that arum (Colocassia antiquorum)
showed very high accumulation of As when grown with As contaminated water. This plant grows

in wet areas. As a result it is all the time taking up As from groundwater. Moreover, this plant is
consumed very widely in the rural areas of Bangladesh. It is a good source of vitamin A, C and Fe.
Every parts of the plant such as leaves, stems, rhizomes, and creepers are edible and are con-
sumed. For this reason, this plant was thoroughly analyzed. The average As content in arum plants
collected from different areas of Bangladesh is presented in Table 3.
The maximum allowable daily level of As in foodstuff is taken as 0.22mg per day (OEHHA
2003). On the basis of this level, calculations were made on the possibility of exceeding this
MADL for the various plants analyzed. For example, a person who daily consumes 100g of arum that
contains 2.2 mg/kg of As would have a MADL from arum alone. However, when the concentration
98
0
20
40
60
80
Grain Husk Leaf Stem Root
As content (mg/kg dw)
Average
Maximum
Minimum
Figure 2. Arsenic content in different parts of rice plant.
Plant of Gangtetic Alluvium soil
Plant of Meghna Alluvium soil
Plant of Teesta Alluvium soil
Plant of Pleistocene terrace soil
10
9
8
7
6

5
4
3
2
1
0
As content (mg/kg dw)
Arum
Brinjal
Amaranthus
Gourd
Onion
Papaya
Indian spinach
Radish
Figure 3. Arsenic content in different crops collected from soils of various origin.
Copyright © 2005 Taylor & Francis Group plc, London, UK
is as high as 22 mg/kg, only 10g would give the MADL. Similarly, 440 g of rice with 0.5 mg/kg
would also represent the MADL. Such inputs are comparable to drinking 4.4 L of water with
0.05 mg/L.
On the basis of the As content in rice, the amount of average daily consumption, the extent of
As contamination in the area, the incidence of arsenicosis patients and the number of population
at risk to exposure of arsenic ingestion, the dietary load estimation, i.e., the possibility of the per
cent of population risking the exposure to excess of MADL, has been calculated for Jessore (rep-
resenting Gangetic Alluvium) and Rangpur (representing Teesta Alluvium) areas. In Jessore area,
32% of the people are above the MADL, while in Rangpur area the value is only 2% (Huq et al.
2001). This again substantiates the fact that the groundwater in the Gangetic alluvial plain is more
contaminated than the other parts of Bangladesh.
Cooked food from the households of arsenic affected people were collected and analyzed.
Arsenic contaminated water has been used to cook these foods. The foods contained various

amount of As in them (Table 4). It is interesting to note that As could not be detected in cooked
lentil soup, locally called “Dal” and in eggs. The cooked rice contained different amounts of As;
the differences could be due either to the variety of rice and also due to variation in the As content
in the cooking water. It can be seen that in many of the cooked foods the values are well above the
99
Table 2. Arsenic transfer factors in different plants of Bangladesh.
Arsenic transfer Name of Arsenic transfer
Name of crops factor crops factor
Mustard 0.02 Bean 0.27
Patal 0.08 Karalla 0.32
Chicinga 0.11 Brinjal 0.35
Ladies finger 0.11 Cabbage 0.44
Coriander 0.12 Amaranthus 0.55
Kalmi 0.12 Tomato 0.55
Pui sak 0.12 Garlic 0.57
Cow-pea 0.13 Turmaric 0.68
Pumpkin 0.14 Gourd 0.69
Lentil 0.17 Jhinga 0.69
Radish 0.18 Pea 0.83
Chili 0.2 Cauliflower 0.84
Carrot 0.23 Wheat 1.46
Papaya 0.26 Arum 2.64
Table 3. Arsenic content in arum collected from different locations of Bangladesh.
As content (mg/kg dw)
Location Average Maximum Minimum
Brahmanbaria 23.55 138.33 0.83
Chuadanga 0.99 3.78 0.13
Comilla 2.03 4.66 0.03
Dhaka 0.40 0.96 0.05
Dinajpur 0.15 0.21 0.09

Jessore 5.27 11.37 0.92
Meherpur 0.71 1.52 0.15
Munshigonj 0.78 3.05 0.18
Narayangonj 3.08 20.50 0.02
Pabna 24.65 115.32 1.05
Rangpur 1.10 3.82 0.14
Copyright © 2005 Taylor & Francis Group plc, London, UK
MADL of 0.22 mg/kg. This is a matter of concern and it is an indication that As ingestion in
human beings is affected not only through water but also through foodstuffs. The food habit and
the nutritional status of a person thus could be related to the manifestation of arsenicosis. This,
however, is also related to the biomethylation activity of the individual (Alauddin et al. 2002).
The above information indicates that there are other pathways of As ingestion in human body
besides drinking water, and that is through food chain. Crops receiving As contaminated irrigation
water take up this toxic element and accumulate it in different degrees depending on the species and
variety as well as on the type of soils on which these plants are growing.
However, the portion of this As that goes directly to the different metabolic pathways and causes
the problem of arsenicosis needs to be assessed. The bioavailability of this arsenic in the different
food materials also needs to be assessed. In a preliminary study by the authors with swine feeding
trials, it has been observed that 27% of the total amount of As in silverbeet and 82% of As in rice
were bioavailable.
ACKNOWLEDGEMENTS
The present work is a part of a joint research financed by Australian Center for International
Agricultural Research (ACIAR) and the Ministry of Education, Government of Bangladesh.
REFERENCES
Alauddin, M., Chowdhury, D., Bhattacharya, M., Bibi, H., Begum, S., Islam, M.S. & Rabbani, G. 2002.
Speciation of arsenic metabolic intermediates in human urine by chromatography and flow injection
hydride generation atomic absorption spectrometry. Paper presented at the 4th International Conference on
“Arsenic Contamination of Ground Water in Bangladesh: Cause, Effect &Remedy” held at Dhaka,
Bangladesh during 12–13 January 2002.
DCH (Dhaka Community Hospital) 1997. Arsenic pollution in groundwater in Bangladesh, Dhaka, Bangladesh.

DPHE/BGS 1999. Groundwater studies for arsenic contamination in Bangladesh, Phase 1: Rapid Investigation
Phase, Final Report in 4 volumes prepared for the Government of Bangladesh and the Department for
International Development (UK).
Farago, M.E. & Mehra, A. 1992. Uptake of elements by the copper tolerant plant Armeria maritima. Metal
compounds in environment and life 4, (Interrelation between Chemistry and Biology), Science and
Technology Letters, Northwood.
Huq, I., Smith, E., Correll, R., Smith, L., Smith, J., Ahmed, M., Roy, S., Barnes, M. & Naidu, R. 2001. Arsenic
Transfer in Water-Soil-Crop Environments in Bangladesh I: assessing Potential Exposure Pathways in
100
Table 4. Arsenic content (mg/kg) in some cooked food collected
from various Arsenic contaminated areas.
Sample As Sample As
Rice 0.35 Egg Ͻb.d.l.
Rice 0.36 Egg Ͻb.d.l.
Rice 0.11 Eggplant 0.66
Rice 0.13 Eggplant 1.45
Vegetable curry 0.81 Lentil soup Ͻb.d.l.
Vegetable curry 0.95 Lentil soup Ͻb.d.l.
Spinach 0.13 Pumpkin 0.27
Spinach 0.12 Pumpkin 0.25
Spinach 0.34 Fish curry 0.34
Spinach 0.33 Fish curry 0.39
Ͻb.d.l. ϭ below detection level.
Copyright © 2005 Taylor & Francis Group plc, London, UK
Bangladesh. Book of abstracts, “Arsenic in the Asia-Pacific Region Workshop Adelaide 2001”, held dur-
ing 20–23 Nov 2001, Adelaide, South Australia.
Huq, S.M.I., Jahan Ara, Q.A., Islam, K., Zaher, A. & Naidu, R. 2001. The possible contamination from arsenic
through the food chain. In Groundwater Arsenic Contamination in the Bengal Delta Plains of Bangladesh
(Proceedings of the KTH-Dhaka University Seminar, University of Dhaka, Bangladesh), Eds. G Jacks,
P Bhattacharya, and AA Khan KTH Special Publication, TRITA-AMI Report 3084, ISSN 14001306,

ISRN KTH/AMI/REPORT 3084-SE, ISBN 91-7283-076-X, ©2001, KTH, pp. 91–96.
Huq, S.M.I. & Naidu, R. 2003. Arsenic in groundwater of Bangladesh: Contamination in the food chain. In
M.Feroze Ahmed (ed.), Arsenic Contamination: Bangladesh Perspective.ITN-Bangladesh, BUET, Dhaka,
Bangladesh, ISBN 984-32-0350-X, pp. 203–226.
Huq, S.M.I., Rahman, A., Sultana, N. & Naidu, R. 2003. Extent and severity of arsenic contamination in soils
of Bangladesh. In M.Feroze Ahmed, M.Ashraf Ali and Zafar Adeel (eds.), Fate of Arsenic in the Environment.
BUET, Dhaka , The United Nations University, Tokyo, ISBN 984-32-0507-3, pp. 69–84.
Merkert, B. 1992. Multi-element analysis in plant materials – Analytical tools and biological questions. In DC
Adriano (ed), Biogeochemistry of Trace Metals, Lewis Publishers, Boca Raton.
Nickson, R., Mcarthur, J.M., Burgess, W, Ahmed, K.M., Ravensroft, P. & Rahman, M. 1998. Arsenic poison-
ing of Bangladesh groundwater. Nature 395: 338.
OEHHA 2003. Maximum allowable dose level (MADL) for reproductive toxicity for arsenic (inorganic oxides)
for oral exposure. Proposition 65, Office of Environmental Health Hazard Assessment, Reproductive and
Cancer Hazard Assessment Section, May, 2003. 7p.
Portman, J.E. & Riley, J.P. 1964. Determination of arsenic in seawater, marine plants and silicate and carbonate
sediments. Anal. Chem. Acta. 31: 509–519.
101
Copyright © 2005 Taylor & Francis Group plc, London, UK