Predicting The Global Extent of Arsenic Pollution of Groundwater and Its Potential Impact on Human Health

Natural arsenic pollution of groundwater and surface water affects more than 140 million people in at least 70 countries worldwide. In half the countries where arsenic pollution is now known, it was discovered within the last ten years, and it is almost certain that it will be found in many more. The objective of this study is to predict the other countries and regions in which there is a significant risk of finding arsenic pollution.
 
By examining known arsenic occurences, their geochemistry, and their climatic and geological associations, it has proved possible to develop a set of predictive rules that have been embodied in a GIS model. This model not only predicts the locations where there is a risk that groundwater is polluted by arsenic, it also estimates the population of the ’at risk’ regions, and hence provides a basis for predicting the size of the population which may be exposed to drinking arsenic above the WHO guideline value and/or local standards. The GIS model does not predict all known forms of natural arsenic pollution, but based on known occurrences, is expected to predict more than 90% of them.
 
The GIS model has been applied to the whole world, and this report provides maps of the at risk areas and listings of the at-risk population in each country. The model successfully predicts the vast majority of the major known cases of pollution. The model predicts a risk of arsenic pollution in 54 countries where pollution is known to occur, and also in a further 53 countries where it is not known. Nine countries, with large at-risk populations but no reported cases of pollution, have been selected for more detailed mapping. These countries were chosen to represent a range of conditions and include four in Asia (Indonesia, the Philippines, Iraq, and Uzbekistan), two in South America (Colombia and Venezuela) and three in Africa (Ethiopia, Sudan and Marocco).
 
Where a risk of arsenic pollution is predicted, relevant courses of action are recommended. In countries where arsenic is a familiar problem, it may be sufficient to provide guidance for additional surveys. Where arsenic pollution is predicted but not known, a second tier of desk study is recommended to evaluate the model predictions in the light of national data sets and, provisionally, to identify areas, personnel and testing protocols for field surveys.

Contents:

1. Introduction
1.1 Background
1.2 Scope and Purpose
1.3 Acknowledgments

2. Scientific Basis of the Model
2.1 Geochemical Mobilisation Mechanisms
2.2 Geological and Climatic Setting
2.3 Population: Exposed and At-Risk
2.4 Prediction Principles

3. Formulation of Predictive Models in Geographical Information Systems
3.1 Arsenic Risk in Alluvial Aquifers
3.2 Sulphide-Oxidation in Bedrock Aquifers
3.3 Glacial Aquifers

4. Discussion of Model Results
4.1 Preamble
4.2 Europe, North America and Australasia
4.3 Asia
4.4 South and Central America
4.5 Africa
4.6 The Sulphide-Oxidation Model
4.7 Secondary Data Processing
4.8 Summary and Screening by Country

5. How to Interpret and Use Models Results
5.1 Initial Enquiries
5.2 If Reductive Dissolution is the Predicted Mechanism
5.3 If Alkali-Desorption is the Predicted Mechanism
5.4 Surveys of Arsenic in Well Waters

6. Predictions of Arsenic Risk by Region
6.1 Europe, North America and Australasia
6.2 Asia
6.3 South and Central America
6.4 Africa

7. Conclusions and Recommendations