Groundwater Arsenic Contamination Inventories and Risk Assessment using Geographic Information System: Case Studies Kishoreganj and Netrokona Districts of Bangladesh

Groundwater in Bangladesh is contaminated with Arsenic, which occurs naturally in alluvial and deltaic sediments. The first official detection in 1993 and subsequent confirmation after 1995 of high levels of Arsenic in numerous shallow and deep wells in various parts of the country has raised serious health concerns. Recent investigations, though incomplete, confirm that the occurrence of Arsenic in groundwater is more widespread than assumed at first and that it already affects a large number of people. The latest statistics available on the arsenic contamination in groundwater indicates that 52 districts around 80% of the total area of Bangladesh and about 40 million people are at risk. It is estimated that at least 1.2 million people are exposed to Arsenic poisoning with tens of millions potentially exposed. The reported number of patients seriously affected by arsenic in drinking water has now risen to 60001 demands extensive research in this field. Emission inventories are essential tools for assessing releases to the environment. Analysis on emissions inventory data can be useful for environmental program planning and management purposes as well as identifying emissions that are potentially above the standard level. In identifying only the level of concentration is not enough; the concentrations resulting from the emissions are important to estimating exposure and risk. Geographic Information Systems (GIS) used in this study for visualizing water quality characteristics in Union census block, distribution of arsenic groundwater concentration, and exposure risk zones for two northeastern districts, Kishoreganj and Netrokona of Bangladesh. INTRODUCTION In recent years Geographic Information Systems (GIS), have become more available to various environmental agencies as these sophisticated computer information systems, once available only on large mainframe computers, have gradually migrated to workstations and personal computers. A GIS can store, manage, analyze, and display a wide range of geographic, demographic, and environmental quality data in a variety of formats. Various environmental agencies have begun to use GIS for numerous disaster planning tasks as well as the basis for a more general "environmental database" which can provide both local and international exposure with information about the depth of the disaster and also a further planning tool to make future studies. Arsenic groundwater contamination of Bangladesh has already got a wide range of international Paper Presented at the 92nd Annual Meeting of Air & Waste Management Association, St . Louis, Missouri, USA, 20 24 June 1999. 2 media coverage. So far multinational organizations have already promised in some capacity, help Government of Bangladesh promoting the deadly situation of the arsenic disaster. They are World Bank, UNDP, UNICEF, British DFID, and Switzerland Government. Field surveys on arsenic concentration and water quality measurements have already started using various methods by different organizations. These databases are scattered and no exchange of study findings is observed. It is really very necessary to have the results coordinated and shared with other organization. In this connection, Bangladesh Government in one of its recent movements set a National Arsenic Mitigation Information Center (NAMIC). This is a part of the World Bank financed Arsenic Mitigation Project, which will deal to collect, manage, interpret, and disseminate all relevant hydrogeological, water quality, health, socioeconomic, and technical information. Total project costs are estimated to be US $44.4 million. This includes an IDA credit of US$ 32.4 million and a US $ 3 million grant from the Swiss Agency for Development and Cooperation.2 It should be noted here that special care should be taken on laboratory methods and data accuracy, as some of the previous study results troubles to determine the nature and the extent of the arsenic contamination and its possible remedies due to the lack of reliable data.3 Groundwater contamination of arsenic has already affected 52 out of 64 districts of Bangladesh. It is estimated that around 0.7 million out of 2.5 million tubewells of the whole country are contaminated by arsenic. Based on this statistics it can be understood that working with such large set of environmental databases poses certain problems. As there are many sample sites, sampling time frames, and multiple parameters like tubewell depth, water qualities; not all of which are measured at each site and each time frame. A proper management of these databases is required so that it be handled on a geographic basis. The reason is to enable hydrologists, policy makers, and community leaders to discover the way, in which the arsenic flows through the aquifer and recommend sufferers to avoid drinking water, which contain high level of arsenic. It also helps taking measures not to sink new tubewells in the same geographical location and water aquifer where contaminated tubewell has been identified. Water quality refers to the characteristics of water that will influence its suitability for a specific use. Emphasis normally placed in the chemical and physical properties of water. Water quality measurement is important especially when certain chemical and physical quantity exceeds the normal levels. A formal risk assessment of arsenic exposure may be conducted under the standard framework used for chemical risk assessment. This consists of three stepsdose-response assessment, exposure assessment, and risk characterization. A retrospective case-control study showed a significant association between duration of consuming high-arsenic well water and cancers of the liver, lung and bladder.4 In this study, cancer deaths in the Blackfoot disease (BFD) endemic area of Taiwan between January 1980 and December 1982 were chosen for the case group. About 90% of the 86 lung cancers and 95 bladder cancers in the registry were histologically or cytologically confirmed and over 70% of the liver cancers were confirmed by biopsy. A control group of 400 persons living in the same area was frequency-matched with cases by age and sex. Standardized questionnaires of the cases (by proxy) and controls determined the history of artesian well water use, socioeconomic variables, disease history, dietary habits, and lifestyle. Similarly, in a 15-year study of a cohort of 789 patients of Blackfoot disease, an increased mortality from cancers of the liver, lung, bladder and kidney was seen among BFD patients when compared with the general population in the endemic area or when compared with the general population of Taiwan. Paper Presented at the 92nd Annual Meeting of Air & Waste Management Association, St . Louis, Missouri, USA, 20 24 June 1999. 3 A significant dose-response relationship was found between arsenic levels in artesian well water in 42 villages in the southwestern Taiwan and age-adjusted mortality rates from cancers at all sites, cancers of the bladder, kidney, skin, lung, liver and prostate.5 An ecological study of cancer mortality rates and arsenic levels in drinking water in 314 townships in Taiwan also corroborated the association between arsenic levels and mortality from the internal cancers.6 Chen and others7 conducted an analysis of cancer mortality data from the arsenic-exposed population to compare risk of various internal cancers and compare risk between males and females. The study area and population have been described by Wu and others.5 It is limited to 42 southwestern coastal villages where residents have used water high in arsenic from deep artesian wells for more than 70 years. Arsenic levels in drinking water ranged from 0.010 to 1.752 ppm. The study population had 898,806 person-years of observation and 202 liver cancer, 304 lung cancer, 202 bladder cancer and 64 kidney cancer deaths. The study population was stratified into four groups according to median arsenic level in well water (< 0.10 ppm, 0.100.29 ppm, 0.30-0.59 ppm and 0.60+ ppm), and also stratified into four age groups (< 30 years, 30-49 years, 50-69 years and 70+ years). Mortality rates were found to increase significantly with age for all cancers and significant doseresponse relationships were observed between arsenic level and mortality from cancer of the liver, lung, bladder and kidney in most age groups of both males and females. The data generated by Chen and others7 provide evidence for an association of the levels of arsenic in drinking water and duration of exposure with the rate of mortality from cancers of the liver, lung, bladder, and kidney. This study has been conducted on the database of North East Minor Irrigation Project, Ministry of Agriculture, Bangladesh Government. Under this project water samples from tubewells were collected from six northeastern districts of Bangladesh. Water samples were analyzed to measure arsenic concentration and various water quality characteristics. This paper will deal with databases of two northeastern districts: Kishoreganj and Netrokona. These basic databases are utilized for visualizing geographically located arsenic contamination level, water quality characteristics, and a general outline for arsenic risk assessment. DATA AND METHOD Districts Kishoreganj and Netrokona are two affected northeastern districts where arsenic groundwater contamination exceeds Bangladesh permissible limit of 0.05 ppm. These two districts having population of 1.73 and 2.3 millions and total areas of 2689 and 2810 km2, respectively. 375 water samples from Kishoreganj and 333 samples from Netrokona district are analyzed for arsenic concentration using Silverdiethyl Dithocarbonate (SDDC) method (Standard Method #307B). On the other hand, tubewell water quality measurements for 30 samples of Kishoreganj district and 30 from Netrokona district using standard methods. The following parameters of water were analyzed: pH, electrical conductivity (ECw), Sodium ion (Na +), Potassium ion (K+), Calcium ion (Ca++), Magnesium ion (Mg++), Ferrous ion (Fe++), Nitrate ion (NO3 -), Chlorine ion (Cl-), Boron (B), and Carbonate/Bicarbonate (CO3/HCO3), and on the basis of experimental data Sodium Adsorption Ratio (SAR) was estimated. The population data for the smaller geographic area Unions are collected from the 1991 census data adjusted for 1995 of Bangladesh Bureau of Statistics.8 These databases are composed of Union level data of male and female groups, age wise population distribution in each geographic location. Paper Presented at the 92nd Annual Meeting of Air & Waste Management Association, St . Louis, Missouri, USA, 20 24 June 1999. 4 pH was measured with a HANNA pH meter. Electrical conductivity was measured with a HANNA condectometer and the results are expressed in Deci Siemens per meter (dS/m). A Jenway PFP7 flame analyzer was used for the determination of Na+ and K+ against the usual practice of calibration with standard solutions of Na+ and K+. Calcium and Magnesium were determined by complexometric titration. For greater accuracy preconcentration of water samples were done to an exact volume when necessary. Both Ca++ and Mg++ were obtained by using Erichrome Black T (EBT) indicator and back titration with standard Mg++ solution at pH 8-10. For Ca++ only, calcon indicator was used after removing Mg++ at about pH 12. Alternatively Ca++ was titrated using Murexide indicator. Iron was reduced to the Fe (II) State by NH2OH and pH was adjusted to 3.23.3. 1, 10 Phenanthroline reagent was added in presence of acetate buffer and the resulting orangered color was measured in a spectrophotometer at 510 nm against a reagent blank as usual. Nitrate was determined by using the reagent phenol disulphonic acid and measured the absorbence in a spectrophotometer at 420 nm against the standard calibration as usual. Chloride content was determined by using Mohr's method of titration using AgNO3 as the standard titrating reagent in presence of K2CrO4 indicator or for back titration with standard NH4CNS solution using ferricalum indicator. Boron was determined by the carmine method in concentrated H2SO 4 and the absorbence was measured at 585 nm in the spectrophotometer Photic-100. Carbonate and Bicarbonate was determined by titrimetric method using mixed indicator (universal indicator) with the help of standard acid/alkali solutions. The sodium adsorption ratio (SAR) is the standard measure of the sodicity of soil/water. The sodium adsorption ratio (SAR) is calculated from the concentrations in milliequivalents per liter (me/l) of sodium, calcium, and magnesium in the saturation extract: SAR = Na+ + Ca++ + Mg++ A dose response relationship has been developed to estimate the health impacts of arsenic contamination in two districts. The health impact for relative risk measurement can be estimated using the following relationship: