The acid mine drainage (AMD) situation at three opencast collieries in the Witbank Coalfield, South Africa, was evaluated from different perspectives. Twenty-seven test pits were dug into the spoils of these collieries to characterise the in situ conditions of spoils of varying age. Detailed static testing was done on surrounding drill holes and additional samples were taken from the test pits. Selected samples were tested using different techniques, namely ABA, mineralogy and kinetic tests. Groundwater monitoring boreholes from the areas surrounding these spoils were also used for characterisation. The ABA results compared favourably with the in situ determinations; acidic test pits proved to be potentially acid-generating from ABA, even when the samples had neutral paste pH. Kinetic test results showed these samples to yield low pH leachate over the test period. In several of these results, the theoretical depletion of neutralising potential and the onset of acidity were well matched. NAG pH-values and final observed water chemistry corresponded for the cells predicted by ABA to acidify, but compared poorly in cells that did not acidify. Results from the non-acidic spoils were consistent with the in situ observations. Alternative kinetic test methodologies were also employed. Duplicate samples were tested in humidity cells and in aerated trays that were manually wetted on a regular basis._x000D_
Results from these tests showed an excellent correlation. Detailed mineralogical determinations using XRD and XRF were used for comparison to the ABA and, in some cases, humidity cell results. These comparisons show that the ABA results are consistent with the results expected from the mineralogy. The conclusion from these assessments is that ABA, in most cases, provides information that can be correlated to the field, mineralogy and to humidity cells. The use of multiple techniques leads to a higher degree of confidence in the use of the data.