Modelling SO2 emissions from Anglogold Ashanti's East Acid Plant: detemining impacts on ambient air quality

Abstract

Since the promulgation of the new air quality legislation in South Africa, sulphur dioxide (SO2) has been a pollutant of concern especially in the heavily industrial South African regions. AngloGold Ashanti’s sulphuric acid (H2SO4) plant located in Klerksdorp, North West province is an important local source of SO2. Other important sources in the North West province include the platinum mine smelters which are responsible for elevated SO2 concentrations in the Rustenburg area. The impacts of these emissions are exacerbated by the poor atmospheric dispersion potential for a substantial portion of the year. An air dispersion modelling study undertaken by Scorgie and Venter (2004a) indicated that the AngloGold Ashanti’s East Acid Plant was likely not to comply with its Air Pollution Prevention Act (APPA) registration certificate (RC) conditions and the proposed South African ambient air quality standards. AngloGold Ashanti subsequently implemented emission reduction measures to minimise elevated SO2 levels and for the first time initiated continuous emission monitoring in the stack and the nearby village. This study aimed at determining the impacts from implementing emission control measures in 2007 whilst establishing the relationship between quantified stack emissions, modelled and monitored ambient air quality data. Other AngloGold Ashanti SO2 sources i.e. South Uranium Plant and Great Noligwa No. 8 Gold Plant were included in the model runs to assess their contribution to the cumulative SO2 concentrations. AERMOD was applied to examine the dispersion potential of stack and fugitive emissions. Modelled SO2 stack concentrations were within the current South African ambient air quality standards for all averaging periods prior and post East Acid Plant shutdown1. However, exceedances were noted for 1-hour and 24-hour averaging periods for modelled stack and volume sources combined i.e. East Acid Plant, South Uranium Plant and Great Noligwa No. 8 Gold Plant. The stack emissions and ambient data compared well with an exception of the fugitive emissions. The model demonstrated a satisfactory performance to calculate stack emissions from the East Acid Plant. However, the model compared poorly with the monitored ambient air quality data partly due to the lack of comprehensive emission factors for fugitive sources. Based on these results it can be concluded that the Acid Plant stack concentrations solely, do not pose significant health risk to the nearby receptors and that the implemented air pollution abatement measures are mostly effective. However, it is important to note that the model may have underestimated fugitive emissions which contribute to low-lying emissions thus impacting on sensitive receptors.