September 2001

1. Executive summary.

Acid rock drainage is common in both the gold and coal mining districts of South Africa. In general, acidic water is produced by the interaction of water with sulphide minerals e.g. iron pyrites in the ore bodies (conglomerate or coal, respectively) or in tailings dumps associated with mining activity in these environments. Our experience of acid rock drainage has provided examples where pH has been reduced from a normal value of ~6 to as low as 1.5. Sulphate contents frequently exceed 5 000 mg/l and we have measured point sources in excess of 40 000 mg/l. The severe reduction in water quality has major implications for industrial, urban and agricultural activity further down stream whilst also adversely affecting the long-term habitability of the earth in these areas. In order to ameliorate the effects of existing acid rock drainage and to plan for future mining and industrial activity, it is necessary to have a better quantitative understanding of the migration paths for ground water and the interaction between mine water and both surface and ground water. Quantification of water quality in terms of source, travel distance, route and time is thus very important for effective management of South Africa's scarce water resources.

Studies to distinguish between water derived from various sources are currently based on routine chemical analyses for elements and ions such as Ca, Mg, Na, Fe, Cl-, N032-, S042-, C032- and HC03-. One of the major disadvantages associated with such standard chemical investigations of water is that the different elements are variably prone to precipitation or adsorption onto clay particles and hydroxides. The chemical composition of water may thus vary non-linearly with distance from the source. This is particularly prevalent for acid rock drainage where pH typically varies from <2 in badly contaminated situations to ~6-8, resulting in enhanced diversity of chemical behaviour between the different elements.

This project was designed to investigate the feasibility of using various isotopes of elements common in ARD to provide supporting information for standard chemical and hydrogeological studies of water quality downstream of acid rock drainage sources. Previous studies have considered the isotopic composition of elements such as oxygen and deuterium but have not considered elements such as strontium (Sr) and lead (Pb) which also show significant variations dependant on the source of the water and water-rock exchange processes along the fluid pathway.

The aims of the project were initially defined as:

Following interaction with and advice from the Steering Committee, these major objectives were redefined as:

No artificial or radioactive isotope tracers have been considered in this study. The isotopes selected for use were deuterium, oxygen, strontium and lead. The first two elements are the major components of water and therefore provide a direct indication of the history of the water sampled. Strontium is a common trace constituent of the rocks and soil through which ground water passes and is always present at measurable levels in water. Lead is a trace component of sulphide minerals such as iron pyrite, the oxidation and dissolution of which is the prime cause of acid rock drainage. Lead therefore provides a direct link to the source of contamination in ARD situations. Lead is also present in certain silicate minerals, usually at lower concentrations than is strontium.

Deuterium and oxygen are classified as light stable isotopes whereas strontium and lead are termed radiogenic stable isotopes. All the isotopes concerned are readily measured in ground and surface water provided that certain precautions to avoid contamination are taken during field sampling and in the laboratory. These are particularly important for lead because of its low concentration in natural waters.

Chemical constituents which are to be utilised for monitoring purposes should meet the following criteria (Hutchison and Ellison, 1994):

Very few, if any macrochemical constituents ever meet all the requirements for selection of suitable constituents for monitoring purposes whereas the isotopes described in this report meet most and, in some cases, all of these criteria in many situations. Significant variations in isotope composition occur in South African waters and these can be used for site specific investigations as well as for more regional ones. The advantages of unique, very precise isotope ratios relative to less precise and potentially variable chemical element variations warrant the use of isotopes in many more studies than have been performed to date.

Particular advantages of the isotopes over chemical constituents alone include the following:

This study has demonstrated that radiogenic isotopes in water vary according to a structured, easily determined manner and are sensitive indicators of both mixing of water with different isotopic compositions and of water-rock exchange processes. Together with light stable isotopes and macrochemical constituents, they are ideally suited for precise, often quantitative studies of ARD pollution. They may be used to identify different, isotopically distinct sources of pollution and, in certain situations, may also quantify relative proportions contributed from the different sources.

Isotope investigations should not be performed in isolation from standard chemical and hydrogeological studies but should rather be considered another 'tool' in the kit. Isotope studies should normally be applied on a selected sub-set of the samples considered during a full investigation and will be of particular use where the chemical constituents do not distinguish sufficiently between end-member contaminant sources or where it is necessary to obtain a more quantitative assessment of the relative contributions from such sources.

This study has demonstrated that the use of radiogenic stable isotopes, especially in conjunction with light stable isotopes and macrochemistry, is a very sensitive tool for the study of acid rock drainage type water pollution. Surface and ground water resources in South Africa are under threat from a number of other sources of pollution, in particular from pollutants generated by the manufacturing industry and from waste disposal activities in general. It is recommended that the extent to which radiogenic stable isotopes may assist in the identification and quantification of these other types of pollution also be determined.

The present study has used computational techniques for calculating source apportionment which, whilst well documented in the literature, are not readily available for use by regulatory authorities and environmental consultants in general. It is recommended that a Windows '951'98/2000/NT version of this software be developed and made available in South Africa.

It would also be of use to develop software which can take into account the measurement uncertainties of the different parameters e.g. chemical element concentrations and isotope ratios, when performing source apportionment calculations. These techniques should also be distributed for general use by consultants and regulatory authorities.