Dissolved chromium pollution in rainfall and surface waters in mid-Wales during the mid-1980s

Atmospheric pollution by dissolved chromium occurred in the mid-1980s in a rural part of upland Wales with no local industrial sources and at a distance of similar to 100-200 km from the major industrial centres of South and North Wales and the English Midlands. The pollution peaked in rainfall, throughfall and stemflow during 1986-1987 when average dissolved chromium concentrations tripled from a background of 2 mu g/l. Correspondingly, there were large increases in the concentrations in runoff from streams in the area; annual mean values quadrupling during 1986-1987 from 1983-1985 values of similar to 2 mu g/l. Subsequently, the stream concentrations declined approximately exponentially to <0.2 mu g/l. Thus, the peak stream concentrations have to be set against a declining long term background trend presumably related to general improvements in emission controls for UK industry and changing industrial production patterns. The annual average dissolved chromium concentrations in the stream during the peak years approached and in some cases surpassed values which are thought damaging to sensitive aquatic organisms. For the period prior to 1992, the dissolved chromium concentrations in the streams were higher than in the rainfall, throughfall and stemflow by similar to 30-40% and background weathering sources are probably small. Since evapotranspiration of rainfall should increase dissolved chromium concentrations in the stream by similar to 25-30%, it seems that most of the atmospheric chromium supplies are not being significantly inhibited on transport from the atmosphere to the stream other than by physical storage: differences between input and output are probably related to errors influx estimation. This means that the dissolved chromium is relatively inert within the catchment, implying that it exists either in the hexavalent anionic state or in a highly complexed trivalent form in both the atmospheric input and the stream output. If it is the hexavalent form which predominates, which seems most likely, then this is particularly disturbing as it is in this state that it is at its most environmentally harmful. Subsequent to 1991, the rainfall input of chromium is >4 times higher than the stream outflow and in this case the catchment is retaining much of the chromium input. This implies that the chromium input differs for this later period: the predominance of trivalent chromium is thus inferred. The reason for the pollution incidents throughout the study period is unclear as dissolved chromium concentrations are not correlated with determinands linked either to general air circulation patterns or to the usual sources of chromium such as the steel industry and cement manufacture.