Ahmed, Kazi Matin Department
of Geology, University of Dhaka, Dhaka 1000, Bangladesh E-mail: email@example.com
Akhter Syed Humayun Department of Geology, University of Dhaka, Dhaka 1000, Bangladesh E-mail: firstname.lastname@example.org, email@example.com
Bhattacharya, Prosun Groundwater Arsenic Research Group, Division of Land and Water Resources, Royal Institute of Technology, SE-100 44 Stockholm, Sweden. E-mail: firstname.lastname@example.org
Chatterjee, Debashis Groundwater Arsenic Research Group, Department of Chemistry, University of Kalyani, Kalyani-741 235, West Bengal, India E-mail: email@example.com
Chowdhury, Dulali Environmental Health Programme, International Centre for Diarrhoeal Disease Research, Bangladesh, Mahakhali, Dhaka 1207, Bangladesh
Frisbie, Seth Environmental Health Programme, International Centre for Diarrhoeal Disease Research, Bangladesh, Mahakhali, Dhaka 1207, Bangladesh. E-mail: firstname.lastname@example.org
Gustafsson, Jon Petter Groundwater Arsenic Research Group, Division of Land and Water Resources, Royal Institute of Technology, SE-100 44 STOCKHOLM, Sweden E-mail: email@example.com
Hasan, M Aziz Department of Geology, University of Dhaka, Dhaka 1000, Bangladesh. E-mail: firstname.lastname@example.org
Hoque, Bilqis Amin Environmental Health Programme, International Centre for Diarrhoeal Disease Research, Bangladesh, Mahakhali, Dhaka 1207, Bangladesh. E-mail: email@example.com
Huq, S.M. Imamul Department of Soil Science, University of Dhaka, Dhaka 1000, Bangladesh. E-mail: firstname.lastname@example.org
Imam, M Badrul Geohazard Research Group, Department of Geology, University of Dhaka, Dhaka 1000, Bangladesh. Present address: Visiting Associate Professor, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia. E-mail: email@example.com
Jacks, Gunnar Groundwater Arsenic Research Group, Division of Land and Water Resources, Royal Institute of Technology, SE-100 44 Stockholm, Sweden E-mail: firstname.lastname@example.org
Khan, Aftab Alam Geohazard Research Group, Department of Geology, University of Dhaka, Dhaka 1000, Bangladesh. E-mail: email@example.com
Khan, Feroz, Environmental Health Programme, International Centre for Diarrhoeal Disease Research, Bangladesh, Mahakhali, Dhaka 1207, Bangladesh
Monsur, Md. Hussain Department of Geology, University of Dhaka, Dhaka 1000, Bangladesh. E-mail: firstname.lastname@example.org
Ravenscroft, Peter Mott MacDonald Ltd., 122 Gulshan Avenue, Dhaka 1212, Bangladesh E-mail: email@example.com
Sracek Andre Groundwater Arsenic Research Group, Division of Land and Water Resources, Royal Institute of Technology, SE-100 44 STOCKHOLM, Sweden. Recent address: Institute of Hydrogeology, Engineering Geology and Applied Geophysics, Faculty of Science, Charles University, Albertov 6, 128 43 Prague, Czech Republic. E-mail: firstname.lastname@example.org
09:30 : Guests take seats
09:35 : Recitation from the Holy Quran
09:40 : Welcome address by Dr. Aftab Alam Khan
09:55 : Introduction by Dr. Prosun Bhattacharya
10:10 : Address by the H. E. The Ambassador of Sweden
10:30 : Address by the Honourable Vice Chancellor, University of Dhaka
10:50 : Address by the Chairman, Department of Geology, University of Dhaka
11:00 : Tea
12:10-12:50 : Mobility of Arsenic and Geochemical Modelling Applications. Andre Sracek, Prosun Bhattacharaya, Gunnar Jacks and Jon Peter Gustafsson GARG, Division of Land and Water Resources, Royal Institute of Technology, Stockholm, Sweden.
12:50-13:30 : Geochemistry of the Holocene Alluvial sediments in the Bengal Delta Plains and their Implications in Groundwater Arsenic Contamination. Prosun Bhattacharaya, Gunnar Jacks, Andre Sracek, Jon Petter Gustafsson and Debashis Chatterjee1 GARG, Division of Land and Water Resources, Royal Institute of Technology, Stockholm, Sweden, 1Department of Chemistry, University of Kalyani, West Bengal, India.
13:30-14:10 : Lunch
14:50-15:30 : Etiology of Arsenic in the Groundwater of the Bengal Delta Plain - constraints from geological evidences. Aftab Alam Khan, Syed Humayun Akhter, M. Aziz Hasan, Kazi Matin Ahmed, Badrul Imam GRG, Department of Geology, University of Dhaka, Dhaka, Bangladesh.
15:30-16:10 : Artificial Recharge as a Remedial Measure. Gunnar Jacks, Prosun Bhattacharya and Debashis Chatterjee1 GARG, Division of Land and Water Resources, Royal Institute of Technology, Stockholm, Sweden, 1Department of Chemistry, University of Kalyani, West Bengal, India.
16:10 : Tea and closing of Day 1
Date : February 8, 1999
10:10-10:50 : Isotopic Applications in Contaminant Hydrogeology and Implications for Investigation of Arsenic Contamination. II. Radioactive isotopes. Prosun Bhattacharya, Andre Sracek and Gunnar Jacks GARG, Division of Land and Water Resources, Royal Institute of Technology, Stockholm, Sweden.
10:50-11:10 : Tea break
11:10-11:50 : Geophysical Signature vis-a-vis Arsenic Contaminated Aquifers - Case Studies. Aftab Alam Khan and Syed Humayun Akhter GRG, Department of Geology, University of Dhaka, Dhaka, Bangladesh.
11:50-12:30 : A Review of Available Geochemical Data and their Implications to origin of arsenic in Bangladesh groundwater. Kazi Matin Ahmed Department of Geology, University of Dhaka, Dhaka, Bangladesh.
12:30-13:10 : Quaternary Sedimentation Pattern in the Bengal Delta vis-a-vis Holocene Sea-level Changes. Md. Hussain Monsur Department of Geology, University of Dhaka, Dhaka, Bangladesh.
13:10-13:40 : Lunch
14:20-15:00 : Measurements of Arsenic in Drinking Water at the Field Level: A Mitigation Challenge. Bilqis Amin Hoque, Dulali Chowdhury, Feroz Khan and Seth Frisbie ICDDR-B, Dhaka, Bangladesh.
15:00-15:20 : Tea break
15:20 -16:00 : Sedimentology and Mineralogy of the Arsenic Contaminated Aquifers in the Bengal Delta of Bangladesh. Kazi Matin Ahmed, Badrul Imam, M. Aziz Hasan, Syed Humayun Akhter and Aftab Alam Khan GRG, Department of Geology, University of Dhaka, Dhaka, Bangladesh.
16:00 -17:00 : Discussions
Sida-SAREC Project Proposal
The regional problem of arsenic contamination in groundwater from the vast tract of alluvial aquifers in Bengal Delta Plains (BDP) is known to have affected a population of about 50 million in different districts of Bangladesh and approximately 38 million in the state of West Bengal in India. Consumption of groundwater with elevated arsenic levels (up to 3700 ug/L in certain wells) over a prolonged period of time have resulted in serious health hazards especially among the rural and semi-urban population in the region. Symptoms of arsenic toxicity are manifested as skin lesions, hyperkeratosis, melanosis, cancer in different organs and several other health disorders, which in some cases have proved to be lethal. Need of water for domestic as well as irrigation purposes had triggered rapid development of groundwater resources in the region during the last two decades. Such overdraft of groundwater in an indiscriminate manner is one of the key factors responsible for the spreading of arsenic epidemic in this region (Fig. 1).
Occurrence of arsenic was first noticed in groundwater from southwestern Bangladesh and the adjoining districts in West Bengal, India during the early 1980’s. Since then it was during 1996, only seven districts were declared affected by arsenic contamination in groundwater. Extensive survey of groundwater from domestic water wells during 1997 have indicated the presence of arsenic in about 48 out of the total 64 districts in Bangladesh. According to the reports available so far, it is estimated that arsenic occurs at concentrations above 0.05 mg/L in about 25 districts and nearly one-third population is feared to be at risk from chronic arsenic exposure in Bangladesh.
The mostly affected districts in Bangladesh include Chapai Nawabganj (1 mg/L, 70% affected tubewells), Kushtia (1.51 mg/L, 56% affected tubewells), Pabna (1.31 mg/L, 48% affected tubewells), Faridpur (1.53 mg/L, 75% affected tubewells) and Lakshmipur (1.11 mg/L, 90% affected tubewells). Apart from them nearly 79% tubewells in the district of Barisal, 87% in Gopalganj, 78% in Jessore, 74% in Rajbari, 72% in Bagerhat and 70% in Satkhira have reported arsenic concentrations above the WHO limits of safe drinking water standard.
In a joint declaration termed as "The Dhaka Declaration, 1998" at the concluding session of the International Conference on "Arsenic Contamination of Groundwater in Bangladesh: Causes, effects and remedies" held at Dhaka, Bangladesh during February 1998, it was unanimously pledged that the issue of groundwater arsenic contamination in the country needs to be addressed on an emergency basis in a concerted manner by the scientific community and all assistance should be directed to establish institutional research collaboration in order to safeguard the health of the people in this region.
The vast tract of alluvial aquifers within the Bengal Delta Plains (BDP) host an enormous reserve of groundwater in the region. Groundwater is contained in the Quaternary sedimentary successions of the BDP where the sediment column comprises sequences of complete or truncated fining upward meandering river deposits. The natural incidences of high-As in groundwater in the vast tract of alluvial aquifers within the BDP has resulted primarily due to large scale exploitation of groundwater resources to meet the rising demand of water for agricultural purposes as well as drinking water supply to the rural and semi-urban population. The occurrence of arsenic was earlier believed to be restricted to the meander belts in the Upper Delta Plain (UDP), but in later years arsenic flow had also been noted in groundwater extracted from the Lower Delta Plains (LDP). Arsenic is believed to have sources in crustal rocks of the Rajmahal Hills, and the parts of upland brought there by ancient river systems of the river Ganga or by the reworking of the basic volcanic suite of rocks that form part of the Indus Suture Zone by the river Brahmaputra in the east.
Gaps in our present understanding
There are several gaps in our present day knowledge about the genesis of arsenic contamination in groundwater from the alluvial aquifers. Although the source of arsenic is geological and located within the alluvial sediments, there has been a consistent lack of sedimentological studies and especially studies on lithofacies variation and provenance of the sediments comprising the BDP.
In order to understand the mobilization of arsenic from the alluvial aquifers it is necessary to delineate the vertical and lateral lithofacies variations. The information on the subsurface lthofacies is primarily available from the sediments recovered from the sites of installed tubewells. However these informations are sporadic and far from being accurate and thus unsuitable for the construction of aquifer panel diagrams. Some preliminary geophysical investigations carried out at Chapai Nawabganj using DC resistivity methods have indicated that the arsenic affected aquifers are mostly located above 100 m. However, further study is needed in this direction to characterize the lithofacies variation in the BDP in order to understand the aquifer geometry. Geophysical investigations have been carried out on alluvial aquifers using continuous vertical eletrical sounding (CVES) technique to characterize the geometry of the aquifer horizons in Zimbabwe. The method has remained successful for sub-surface geological mapping where the depth of the alluvial aquifers ranged only upto a maximum of 30-40 m.
Sedimentary petrographic studies on the alluvial aquifer sediments have revealed that the sands comprise nearly 55-65% quartz, 5-10% feldspar and 8-10% lithic fragments along with 4-22% mica mostly biotite. Conspicuous development of Fe-oxide surface coatings have been also found on the detrital framework grains of quartz, feldspar as well as ferruginous alteration products were abundant around the detrital biotite grains. It is therefore important to characterize the provenance of the sedimentary aquifers through petrological inves-tigations.
Chemical analyses of the aquifer sediments indicate significant enrichment in total arsenic. The presence of HFO, HAO and HMnO fractions have been revealed by the selective dissolution studies give a preliminary clue that they are the principal source of arsenic but this needs more detailed investigation. It is necessary to determine if primary sulfide minerals (pyrite, arsenopyrite) are present and the relation of Fe(OH)3 to these minerals. If HFO was formed as a result of sulfide oxidation, then it generally forms coatings on pyrite grains that can be observed under SEM. The precipitation of secondary HFO precipitates on the other hand should be independent of the oxidation of the primary sulfides.
The composition of water in the unsaturated zone collected by suction lysimeters has pivotal importance for determination of the arsenic release mechanism. There also should be several piezometers in groundwater fluctuations zone, to verify the role of redox conditions changes in release of arsenic. Site(s) with relatively shallow water table, about 20-25 m deep (or less), should be investigated because it is increasingly more difficult to recover a suction lysimeter sample from deeper unsaturated zone. There should be two different sites at least, one influenced by return flow from rice field and second under natural conditions. Also the monitoring and sampling should last one year at least, to determine the effect of monsoon recharge on arsenic loading.
The objectives of this project are:
The major work components in the proposed research can be grouped under 4 major headings. Geophysical investigation is the first task of the project, which would be followed by sedimentological and gechemical investigations on the sediments. Hydrochemical investigations would be carried out on the water samples from the saturated as well as unsaturated zone using suction lysimeter in order to avoid contamination risks from the adjacent aquifer units.
1. Characterization of the subsurface aquifer geometry using geophysical investigations along selected transects of the affected regions of the BDP. CVES (Continuous Vertical Electrical depth Sounding) and georadar investigations would be undertaken for the sub-surface geological mapping in the Chapai Nawabganj-Rajshahi, Ishurdi-Bheramara, Meherpur-Chuadanga-Jhenida transects for detailed geophysical investigations. Lateral and verical lithofacies variations as well as characterization of sediment/porewater chemistry would be established on the basis of the resistivity data. The resistivity data can be used to interpret the aquifer geometry as well as the groundwater characteristics. It is also pertinent to mention that these type of geophysical investigations is one of the powerful tool to unravel the subsurface aquifer geometry and so far not been employed in the region.
2. Sedimentary petrological investigation for the characterization of the source of aquifer sediments including petrography, X-ray diffractometry and SEM.
3. Geochemical investigations on the aquifer sediments (access to the sediment samples would be provided by BWDB. The geochemical studies include a) analyses of the sediments by total extraction ; b) selective extraction methods, with an emphasis on the quantification of arsenic released from sulfide minerals (arsenopyrite) or from secondary minerals (HFO, HAO and HMnO). Many sophisticated procedures for sequential extractions proposed in recent years and these need to be tested to characterize the occurrence of arsenic with different phases in the sediments as well as in the soils c). sequential leaching batch experiments under variable pH and Eh conditions on the aquifer sediments would simulate the field conditions and would help to quantify the amount of labile arsenic in the aquifer sediments. Thus the simulation of field conditions in the laboratory column experiments using drilling cores material would help to determine the rate of arsenic leaching.
4. Sampling of water in the unsaturated zone using suction lysimeters. Special type of suction lysimeter available from Soil Moisture Company, Santa Barbara, California, would be used for sampling in very deep unsaturated zone which can be used to study the complete speciation of arsenic in pore water (arsenate, arsenite, organic complexes). These type of suction lysimeter can be allegedly used down to the depth about 100 m, while the normal type of suction lysimeter can only be used down to depth of about 18 m. Monitoring of seasonal changes in pore water composition within the unsaturated zone because Eh may be influenced very much by recharge pulses and there can be re-dissolution and re-precipitation of Fe(OH)3 depending on Eh variations related to changes in water saturation.. Redox transformation processes need to be examined in the field environment through lysimeter sampling would be of pivotal importance to determine the role of As-speciation during groundwater development. Adsorption affinity is higher for As(V) as compared to As(III). The optimum adsorption is at lower pH region for arsenate (about pH 4.0) than for arsenite (close to neutral pH). Behavior of arsenic in saturated zone is closely linked to the behavior of iron. As a general rule, an increase in iron concentration (Fe2+ concentration) is related to complimentary increase in the concentration of As. Surface complexation modeling can be used to predict the quantity of arsenic adsorbed on Fe3+ minerals. It is recommended to distinguish between filtered and unfiltered samples because a significant portion of arsenic may be adsorbed on colloidal iron. Specific groundwater sampling strategies need to be adopted using filtration and acidification techniques.
Thus, better understanding of chemical factors influencing the mechanism of retention and release of arsenic from the aquifer sediments in reduced and oxidized systems as well as in systems undergoing redox changes is essential for the evaluation of the risk of remobilization of arsenic in both unsaturated and saturated zones.
5. Development of low-cost geochemical techniques for the removal of arsenic suitable for application in the developing countries is another important objective of the present research proposal. Laterite, a local raw material ubiquitously found in the Indian penimnsula has been tested as an adsorbent for arsenic using the groundwater samples collected from the village Ghetugachi in Nadia district, West Bengal, India. Through a series of laboratory investigations, laterite was found to have an efficiency to adsorb 50-90 percent of arsenic depending mainly on the natural variability of the laterite and the variations in groundwater chemistry. Laterite can be used as a small-scale alternative for the people drinking water from arsenic affected aquifers. Long term groundwater management strategies must be adopted to prevent mobilization of arsenic to the groundwater by regulating the land use pattern in the region.
Groundwater sampling, installation of suction lysimeters and montoring
investigations and petrological
2001 Reports x x x
Project seminars x x x
Steady-state flow pattern of a hydrogeological system can be characterized using values of hydraulic head, hydraulic conductivity and boundary conditions based on geometry of the system and distribution of recharge. Hydraulic parameters like hydraulic conductivity are defined for representative elementary volume (REV), for which we can determine average values of hydraulic parameters. When appropriate data are available, a suitable groundwater modeling code based on particle tracking approach can be used to determine flow pattern and residence time of water in a hydrogeological system.
Arsenic, and many other metals and metalloids are redox sensitive species. This means that they have different forms with different behavior and toxicity depending on redox conditions in a hydrogeological system. There is generally vertical (and sometimes lateral) redox zonality in an aquifer. Only discrete depth sampling from piezometers opened at different depth can reveal this zonality. On the other hand, sampling from pumping wells with large screens gives only mixed samples and correct redox zonality and speciation of arsenic can not be determined correctly. It is also necessary to distinguish between filtered and unfiltered samples and properly preserve samples to maintain oxidation number of dissolved arsenic species until analysis. Speciation of arsenic can be done using the disposable cartridges for selective adsorption of As5+ in the field.
Arsenic is present in aqueous environment in +III and +V oxidation states. In an oxidizing environment, the principle attenuation mechanism of arsenic migration is its adsorption on amorphous Fe(OH)3. Adsorption affinity is higher for As5+ than for As3+ and both are pH-dependent. Maximum adsorption for As5+ occurs at pH about 4.0 and for As3+ about 7,0. However, Fe(OH)3 is unstable mineral phase which dissolves during changes of Eh and pH and releases adsorbed arsenic. The only important arsenic mineral in oxidizing environment is scorodite, but this mineral is quite soluble and, thus, less important for dissolved arsenic concentration control than adsorption on Fe(OH)3. In very reducing environment, where sulfate reduction takes place, secondary sulfide minerals like As-pyrite and orpiment, As2S3, can incorporate arsenic.
Geochemical modeling can be divided into a) speciation modeling, b) inverse modeling and
c) forward (reaction path) modeling. Speciation modeling is used to determine distribution of arsenic and other dissolved species between different complexes and also to evaluate stability of arsenic adsorbents like Fe(OH)3. Inverse modeling is used to suggest which processes (including, for example, sulfate reduction) take place in a hydrogeological system. Forward modeling is used for prediction of water chemistry downgradient, after completion of pre-determined chemical reactions. Especially useful is forward modeling of arsenic mobility using codes for surface complexation modeling. A common approach is based on diffuse double layer (DDL) model, which can incorporate the effect of different adsorbent/arsenic ratio, adsorption sites density, area available for adsorption, pH changes and competition of arsenic for adsorption sites with other dissolved species like phosphate. Application of all geochemical models requires information about solid phase composition. A case study about adsorption/desorption modeling of arsenic species and phosphate for samples from West Bengal is presented.
The problem of arsenic contaminated groundwater from the vast tract of alluvial aquifers in the Bengal Delta Plains, is spread over a large geographical area and known to have affected a population of nearly 90-100 million in West Bengal and Bangladesh. The sedimentary fills in the BDP are characterized by a thick succession of fluviatile sediments of Quaternary age. The Holocene sediments of the Upper Delta Plains (UDP) are mostly characterized by complete or truncated cycles of fining upward sequences typical for the deposits of the meandering channels and dominated by course to medium sand, fine sand, silt and clay sediments. The knowledge about the sedimentary architecture of the BDP is still inadequate and the lateral and vertical variability of the individual sedimentary facies is unclear and needs more detailed investigations.
Geochemical studies on the alluvial sediments from Hanspukuria, Mahishbathan and Ghetugachi villages in Nadia District in West Bengal have revealed significant enrichment of arsenic. Total extraction of the sediments by 7M HNO3 treatment reveals arsenic concentrations in the range 20-133 mg/kg in the aquifers at different depths. Total concentrations of Fe, Al, Mn and PO4 in the sediments show significant positive correlation with arsenic. Estimation of leachable As from these sediments has been carried out through sequential leaching batch experiments using deionized water (pH 6.95) and with 0.01M NaHCO3 (pH 8.65). The total concentrations of arsenic leached by sequential extraction was found to be in the range of 116-383 ?g/l, strikingly similar to the As concentrations in groundwater. Selective extraction of the sediments has been carried out in oxalate and pyrophosphate solutions to understand relationship between As and secondary amorphous Fe, Al and Mn phases, and also organically bound As. Oxalate extraction of the sediments revealed that ferric oxyhydroxides dominate (Feox=264-1238 mg/kg) above aluminum hydroxides (Alox=27-294 mg/kg). However, the clayey sediments at depth exhibit the presence of both Feox (983 mg/kg) and Alox (294 mg/kg) fractions, combined with high Siox (229 mg/kg), thus indicating formation of secondary aluminosilicates as a potential adsorbent for arsenic. The amounts of pyrophosphate extractable Fe, Al, Mn and As are low and suggest that bulk of the arsenic is bound to the inorganic fractions of Fe, Al and Mn in the sediments.
Ferric hydroxides have variable pH-dependent surface charge. At low pH values the surface charge is positive, where the adsorbed arsenic is predominantly in As(V) anionic form and to less extent in As(III) in neutral form. With an increase in pH, the surface charge of ferric hydroxide becomes more negative and arsenic is desorbed even under oxidizing conditions. Change from oxidizing to reducing conditions in the sedimentary environment results in the dissolution of ferric hydroxide and release bulk of the arsenic due to reductive dissolution of the ferric oxyhydroxides as an impact of groundwater development. Transformation of freshly precipitated amorphous ferric hydroxide to goethite may result in the decrease of the surface area available for adsorption, which may cause release of arsenic with time.
Groundwater with arsenic concentrations
greater than those recommended for drinking occurs in most part of Bangladesh,
probably affecting more than twenty million people. Over the country, it
is estimated that about 26% of wells are contaminated. Arsenic is irregularly
distributed on both a regional and local scales. The occurrence of arsenic
is systematically related to the Quaternary geological history of the basin.
The paper describes the procedures followed in compiling the available
field and laboratory data to determine the geographical and vertical distribution
of arsenic in groundwater across the country. The scales of spatial variation
will be explored and the alternative methods of presentation of data discussed.
The distribution of arsenic has been correlated with surface geological,
geomorphological and hydrogeological parameters. Of the extensive geological
units, the most contaminated groundwater is found beneath the Chandina
Alluvium, Deltaic Silt and Deltaic Sand. Of the extensive geomorphological
units, the most contaminated groundwater is found beneath the Meghna River
Floodplain and Old Meghna Estuarine Floodplain. In the shallowest part
of the aquifer system, arsenic concentrations tend to increase with depth
but then decrease at depths below 100 m. Not more than 3% of wells deeper
than 300 m are contaminated, and even these have much lower total arsenic
concentrations than shallow aquifers. In spite of a shortage of monitoring
data, there is strong indirect evidence that arsenic concentrations increase
with time. The paper explores the correlation of arsenic distribution with
groundwater levels, abstraction, transmissivity and recharge parameters.
Finally, a conceptual geological model to explain the distribution of arsenic
The major source of arsenic in the sediments of the Bengal Delta is inferred to derive from the Rajmahal volcanics and the coal basins situated in the Rajmahal Hills. Denudation and erosion of rock material have resulted in transport of arsenic downstream along with the sediments in suspension and deposition of the sediments of the Bengal Delta Plain (BDP) during the Quaternary period. The arsenic bearing solid phases in these Quaternary sediments have been probably formed due to the precipitation of oxyhydroxides, but they were probably transformed to sulfidic mineral phases at depths. The precipitations of carbonates and phosphates have also been envisaged. The positive correlation of As, Fe and S in the sediments at depth below 40m suggest formation of secondary sulfides with arsenic. However, there is an excess of iron compared to sulfate in groundwater and precipitation of secondary sulfides like pyrite could not play a significant role in control of dissolved arsenic concentration. Reductive dissolution of abundant ferric hydroxide is believed to be responsible for the release of arsenic from adsorbed and co-precipitated state.
Various sedimentary conditions have been inferred for the arsenic contaminated aquifers viz., aquifers in the meandering belt of incised channels, occurrence of peaty clay and calcareous clay below and intercalated with aquifer zones, unconfined nature of aquifers and the residence time of groundwater in the aquifers. The concentration of arsenic in groundwater seems to be directly proportional to its residence time and inversely proportional to the rate of flushing.
Downstream in the Delta Plain, iron from acid mine drainage in the region around the Rajmahal Hills has also been suggested to contribute new suspended sediments enriched in arsenic. However, this contribution cannot be very significant regarding relatively limited period of mining activities compared to the sedimentary history of the BDP.
Artificial recharge has been used to augment the groundwater availability. In a limited extent it has also been used to improve the groundwater quality. In Finland and in Sweden it has been used to remove iron from the groundwater. Tests have been done in Denmark to remove nitrate from groundwater by recharge through straw beds supplying organic matter for denitrification. In India groundwater recharge has been used to decrease the fluoride content of groundwater. If the high arsenic concentrations in groundwater met with in the Bengal Delta Plain are mobilised from an adsorbed pool through reduction of ferric compounds to dissolved ferrous iron, then this would give the possibility of using groundwater recharge to remove at least a substantial part of the arsenic. The aim must then be to lift the redox-state from the ferric/ferrous level to a higher level by introducing another electron-acceptor than ferric iron. The are essentially two choices, oxygen or nitrate.
Oxygen has a limited solubility at high ambient temperature met with in the area. Furthermore if pond recharge is used, algae growth and subsequent degradation may consume the oxygen rather fast. In well recharge, clogging by ferric precipitates may pose a problem. Denitrification requires three conditions, the presence of nitrate, anaerobic conditions and a degradable organic matter. Nitrate can be added and anaerobic conditions are already prevailing in the aquifers, manifested in the almost ubiquitous presence of dissolved iron in the groundwater. The key factor is the degree of degradability of the organic matter. The organic matter must be "chewable" for the denitrifiers. It may not be a disadvantage if the organic matter is a bit refractory as this would imply that the effect of recharge would be rather soft along the water pathway, avoiding local clogging with ferric precipitates. In Denmark it is also considered that ferrous iron can chemically reduce nitrate. Blue clay layers sandwiched in the postglacial aquifers are considered as reducing agents in the Danish groundwater environment.
An indication whether nitrate could be used as an oxidant could be obtained by analysis of the 15N/14N ratio in the traces of nitrate occurring in the groundwater. Denitrification discriminates against 15N and causes accumulation of the isotope in residual nitrate. The applicability of nitrate as oxidant should be tested in batch tests with sediment samples, preferably non-oxidised, cored material. In the field the recharge proposal could first be tested in areas where the upper aquifer is unconfined so that recharge ditches or ponds could be used. Well recharge requires far more knowledge about the hydraulics the groundwater flow and is expensive. Advanced systems like the VYREDOX technology can be used where larger investments are possible such as water supply to larger urban agglomerations. Rather simple and reliable recharge wells have been designed by Vivekanand Research and Training Institute in Gujarat in India. They are based on sand-filtration before the entry into the recharge well and have been in use for 8 years up till now.
If groundwater recharge is a viable
solution, the in situ oxidation of ferrous iron offers the advantage to
avoid large amounts of ferric sludge. As there is a considerable amount
of arsenite present in the groundwater a subsequent oxidation by Mn(IV)-oxides
or application of some other oxidant followed by filtering may be required.
Isotopes can be divided into stable isotopes (they are stable in time) and radioactive isotopes (they decay in time). Stable isotopes are used to determine which physical processes (for example, evaporation: enrichment in both isotopes deuterium and oxygen-18) and geochemical processes (for example, carbonate dissolution: enrichment of dissolved inorganic carbon (DIC) in carbon-13, and sulfate reduction: enrichment of residual sulfate in sulfur-34 etc.) take place. Radioactive isotopes (tritium, carbon-14, chlorine-36) are used for dating of groundwater. Only deuterium, oxygen-18, and tritium are a part of water molecule. Other isotopes are a part of solids dissolved in water and this fact complicates their interpretation.
Potential isotopic applications in investigation of arsenic contamination include: determination of the origin of sulfate and evaluation of the possibility of sulfate reduction (34S(SO4) and 13C(DIC)), determination of the origin of groundwater (D, 18O), evaluation of the possibility of nitrate reduction (15N(NO3)), and determination of the age of dissolved organic carbon (14C(DOC)).
Dating of groundwater using tritium
and 14C(DIC) can be used to determine the age of groundwater
in deeper aquifers. Tritium and 14C(DIC) interpretation can
be combined with determination of groundwater age based on chlorofluorocarbons
Vertical Electrical Sounding (VES) of geophysical technique is generally applied to detect vertical discontinuities resulting from variable physical properties in the subsurface geological environment. A substantially high anomalous VES signatures have been observed at two different regions of severe arsenic contamination at Chapai of Nawabganj district and at Bera of Pabna district while searching for alternate aquifer zones for groundwater development. The occurrence of abnormally high resistive zones have been envisaged at two different physical conditions viz., within the aquifer zone and underlying the aquifer zone.
Groundwater from the aquifers located above the resistive zones and sandwiched within the resistive zones shows high arsenic concentration. It is further observed that the sediments from the high resistive zones give a positive correlation with the concentration of arsenic, iron, aluminum, silica, phosphorous and sulphur. The occurrence of high resistive zones have been inferred to be due to the presence of oxides, silicates, phosphates, and carbonates as precipitates in the sediments with high arsenic content.
A thick clay layer of about 100m thickness, which corresponds to high resistivity value has been encountered in the drilled holes. The presence of organic matter in this clay layer has been suggested because the presence of organic matter in clay generally increases the resistivity of sedimentary layers.
An interesting relation has also been observed between low resistivity and limited concentrations of sulphur, phosphorous, silica and arsenic in the sediments.
Geochemical analyses of groundwater, pore water and aquifer sediments are essential for understanding the origin and release mechanism of dissolved arsenic in the aquifer. This paper attempts to critically review all available geochemical analyses results of groundwater, pore water and aquifer sediments. Results will be interpreted in relation to origin and release mechanism of dissolved arsenic in the Bangladesh aquifers.
Although about twenty thousand water samples have so far been analysed using laboratory techniques, only a small proportion of those samples have been analysed for full range of geochemical parameters. At the same time a very limited samples have been analysed for isotopic constituents. The hydrogeochemical investigations by all the different laboratories agree to the fact that groundwater in Bangladesh aquifers exists under strongly reducing environment. Contaminated groundwater is in most cases found to be Ca-HCO3 -type with lesser amount of Na-Ca-Mg-HCO3 -type and very few NaCl-type water. Very small amount of sulphate and high amount of phosphate and manganese are other characteristic features. In many cases high arsenic is found to be associated with high iron, but not all the high iron groundwater contains arsenic. As(III) and As(V) species distribution shows a very wide variation from study to study and this might have been resulted from improper sampling and preservation techniques.
A very limited number of pore water profiles are available which demonstrate a similar geochemical facies as shown by groundwater. However, pore water from the finer fraction of the subsurface section contains high amount of total arsenic then the pore water from the coarser sediments.
The d 18O and d 2H data show that most of the water samples are close to the World Meteoric Water Line (WMWL) and hence indicate local recharge and no evidence of evaporation. The d 13C data show that most samples are depleted in this isotope and these values are associated with involvement of carbon in biogenic processes. This data also suggest that the dissolved inorganic carbon in the groundwater was derived from oxidation of organic matter in the aquifers.
Sediment analyses from different aquifers
show that none of the samples contain abnormally high concentration of
arsenic and in all cases within the range of concentration reported from
different parts of the world. Finer sediments always contain high concentrations
of arsenic. In all cases, sedimentary arsenic correlates well with iron.
Presence of high amount of organic matter and very low amount of sulphur
is characteristic of all samples. Presence of framboidal pyrite, corroded
iron oxyhydroxides and precursor minerals have particular relevance to
arsenic release and mobilisation. Most of the arsenic present in the sediments
are leachable. Vertical profiles of arsenic show a characteristic pattern,
there is one particular horizon with anomalously high arsenic and apart
from that arsenic concentrations decrease sharply with depth. The high
concentration zone is not related to the position of the water table.
The available geochemical data support the hypothesis that arsenic in Bangladesh groundwater is released by the reductive dissolution of arsenic adsorbed on sediment grains as surface coatings. At the same time the data do not support the pyrite oxidation hypothesis widely believed to be the source of arsenic contamination in Bangladesh.
It is now well established that mid latitude glaciation was accompanied by dry climatic condition in tropical regions. On the other hand, during the interglacials, wet and humid climate caused heavy rainfall in the tropics. Deep reddish brown Early Pleistocene deposits are called Madhupur Formation in Bangladesh, Ilumbazar Formation in West Bengal and Lukundal Formation in Kathmandu Valley. Paleoclimatic signature is the common characteristic of these Formations. Upper parts of these Formations are highly oxidized and deeply weathered. A wet and humid climate was prevailing in the northeastern part of Indian subcontinent, caused this intense weathering and oxidation. Sona Tila Gravel Beds of Lower Pleistocene time, exposed as capping rocks of most of the hillocks of Jaintiapur area, are also strongly oxidized and highly weathered. An equivalent gravel deposits are called Kamalpur Gravel Beds, can be found in Damadar-Ajay interfluve close to Durgapur city of West Bengal. Similarly Talchir Gravels are the equivalent gravel beds, exposed in near the Talchir township of Orissa State. All these beds have deep weathering profiles indicating a humid-wet palaeoclimatic condition.
Late Quaternary sediments are quite fresh in Bangladesh, India and Nepal. In Bangladesh, upper Pleistocene sediments are represented by the alternation of coarse sand and well-rounded smooth gravel deposits in the piedmont area of north Bengal. These are called Panchagarh Gravel Beds, composed of granitic, quartzitic, gneissic and chistose gravels. This gravel may equivalent to the Thimi Formation of Kathmandu valley.
Late Quaternary maximum glaciation
was about 18,000 yrs BP. By that time, sea level was about 100 to 130 m
lower than the present sea level. During the upper Pleistocene time, a
major part of Himalayan mountains was glaciated and the Bengal plain was
acting like an outwash plain. Melt was flowing through the deeply incised
narrow river valleys, discharged into the Bay of Bengal at some hundred
of kilometer southward from the present coast line. These deeply incised
river valleys were filled up with Holocene sediments during the marine
transgression of about 5,500 years BP. It was the Holocene high stand of
sea level caused a wave cut shoreline from Cox’s Bazar to Teknaf and Garakghata
to Uttar Nalbila in Maiskhali Island. After attaining the maximum height
at about 5,500 yr. BP, sea level drop about 1 to 2m, resulted a raised
beach or benchmark (Supra tidal flat) in the eastern coast of the Bay of
Bengal. Lowering of Holocene sea-level after the highstand, submerged islands
of the Bay of Bengal became into aerial exposition. Kutubdia, Materbari,
Sandip, Hatia (in the eastern coast of the Bay of Bengal) and some islands
of Chilka lake (in the western coast of the Bay of Bengal) are the best
Intake of arsenic polluted ground water is supposed to be causing the problem of arsenic contamination to a vast majority of the country’s population. While this phenomenon is common, yet in many instances it has been reported and observed that some people are not affected even if they are drinking water from the same source as affected people. This divergence in the manifestation of arsenic contamination led the author to the thought that nutritional matters might be linked to the problem, which is ultimately related to the food habit and intake of food material of the subject. The possibility of biomagnification of arsenic in the body needs to be given attention. As such, about a dozen of vegetable samples, collected from the villages "Chandigram" in Feni, where arsenic contamination is severe, were analyzed for water-soluble (inorganic) arsenic.
The concentrations of arsenic ranged from 107 to about 2000 ppb. Edible vegetables "arum", "gourd", and "kalmi" were found to contain very high amount of water soluble arsenic. The possibility of arsenic contamination through food chain is discussed.
The problem of arsenic-affected drinking water in Bangladesh is vast and complex. The accuracy, precision, timeliness and cost of analyzing arsenic in this water should vary according to the objective of this measurement.
The proper measurement of arsenic concentration within appropriate limits of accuracy and precision is essential for effective mitigation. It is relatively easy to obtain quality results under laboratory conditions; however, obtaining quality results under field condition where various types of arsenic measurement kits are used is more difficult. Here we review local reports, which compare result from field kits and/or from laboratory, in addition to sharing our experience in this regard.
None of the locally available field kits are developed to detect arsenic at 0.01mg/L concentration (WHO recommended standard). These kits allow qualitative as well as semi-quantitative measurement of the arsenic content in water. All kits, except the E-Merck kit, detect 0.05 mg/l arsenic. E-Merck kit detects arsenic from 0.1 mg/L but a locally modified method is often used to measure arsenic below 0.1 mg/L. " Groundwater studies for arsenic contamination in Bangladesh" showed a positive correlation between Field and Laboratory test results. The occupational hazard in using some of these kits has been also noted.
We have used E-Merck kits (field kit) and silver diethyl dithiocarbomate (laboratory method) methods in our mitigation research project. These test kits yielded results agreed almost 100% perfectly with our laboratory method when the water had no detectable levels of arsenic. But the values of correlation coefficient between results obtained by these two methods under different project activities as well as under grouped data analysis varied widely.
A proper mix of field and laboratory based analysis is essential to get reliable data at an affordable cost. Field kits are very useful for practical purposes, but immediate attention is needed to improve its performance.
Sedimentological and mineralogical investigations were conducted on the samples collected by the Ground Water Circle of Bangladesh Water Development Board from 18 exploratory boreholes. Grain size distribution, vertical lithofacies sequence analysis, thin section microscopy and X-ray diffraction techniques were used for the study. Texturally the samples are composed of clay, silty clay, silty sand and sand. The median grain diameter (MZ) show that most of the samples are fine grained with few medium sands. The degree of sorting (sI) indicates that samples moderate to well sorted which is characteristic of moderate energy condition in the depositional environment. Skewness (SkI) values are mostly positive this normally is an indication of river environment. Kurtosis (KG) value ranges indicate very platykurtic to leptokurtic in nature of the distribution. The bi-variate plot of skewness versus standard deviation shows that all the samples fall in the river domain.
Four distinct lithofacies have been identified from the vertical lithofacies sequence. Texturally these are clay, silty clay, sandy silt, and sand (very fine-, fine- and medium-grained). From the vertical lithofacies profiles, it is evident that the facies occur as two distinct facies associations – a dominantly sandy channel-fill association and a fine-grained overbank association. These associations are typical for meandering river deposits.
Thin sections were prepared for microscopic identification of minerals. Quartz (50-65%) feldspar (7-15%) and lithic grains (7-20%) are the major components in most of samples. Mica (biotite and muscovite) occurs as an important mineral (5-15%) in all the samples and forms major component in some samples. Calcite is also present as a minor constituent in some samples. The heavy mineral content is generally low (<1%) to moderate (<5%), whereas opaque heavies including pyrite and arsenopyrite are insignicant. Non-opaque heavies include mostly hornblende, garnet, tourmaline, epidote and rutile. XRD identification of the whole rock samples did not find any peak characteristic of pyrite/arsenopyrite.
A characteristic feature, in most of the samples, is the pronounced development of surface coating on detrital quartz, feldspar, mica and calcite grains. The coatings are deep brown to black in color, generally thin but often thick to very thick. They are mostly partial and discontinuous. The coatings are considered to be mainly ferrugineous in composition. However, in some samples coatings are indistinct and sporadic.
From the thin section microscopy study of the sand samples, it does not seem likely that the pyrite/arsenopyrite could be a source of arsenic because of low to very low content of opaque heavy minerals. On the other hand the presence of grain coatings points towards the possibility of presence of arsenic in adsorbed state. Such adsorption takes place under oxidising conditions during the sediment-water interaction at the time of transport. This adsorbed arsenic can be released into groundwater under reducing conditions. These coating need to be chemically characterized by SEM and EPMA techniques to ascertain their exact role in the arsenic release and mobilization process.