Ion input/output budgets for five wetlands constructed for acid coal mine drainage treatment

Abstract

Five wetlands, each 6 m wide and 30 m long and containing 30 cm of an organic substrate (Sphagnum peat to which limestone and fertilizer were surface-applied on a quarterly basis, Sphagnum peat, sawdust, straw/manure, spent mushroom compost), were exposed to controlled inputs of acid coal mine drainage (AMD; pH 2.89, soluble Fe, Mn, and SO4 2− concentrations of 119, 19, and 3132 mg L−1, respectively) at a mean flow rate of 8513 L da−1 for 111 weeks, beginning in July of 1989. All wetlands were net sources, rather than sinks, for base cations (Ca2+, Mg2+, Na+, K+). The Sphagnum peat wetland was the least effective in treating the AMD, retaining 35% of the soluble Fe influx, but not retaining substantial H+, soluble Mn, soluble Al, SO4 2−, or acidity. The straw/manure and mushroom compost wetlands were the most effective in treating the AMD, retaining 53 and 67% of the H+ influx, 80 and 78% of the soluble Fe influx, 7 and 20% of the soluble Mn influx, 54 and 53% of the soluble Al influx, 15 and 11% of the SO4 2− influx, and 57 and 63% of the acidity influx. For these two wetlands especially, treatment effectiveness was substantially diminished during the cold winter months of January through March. Moreover, from March through July of the final year of the study, treatment effectiveness was minimal with outflow pH and concentrations of soluble Fe, Mn, Al, SO4 2− and acidity that were similar to inflow values. Decreases in treatment effectiveness over time appeared to be related to a decrease in the ability to counter the substantial acid load entering the wetlands in the AMD. Lime or limestone dissolution and bacterial dissimilatory sulfate reduction may have contributed substantially to pH improvement and acidity consumption in the straw/manure and mushroom compost wetlands, but after 2 years the cumulative input of acidity apparently had overwhelmed biotic and abiotic alkalinity generating mechanisms, as reflected in a progressive decrease in both substrate pH and abiotic acid neutralization capacity (ANC) over time, especially in the surface substrates. Also over time, effluent H+ and acidity concentrations became more like influent and H+ and acidity concentrations. Although samples of wetland interstitial water were not collected for chemical analysis, as substrate pH and ANC decreased and as influent and effluent water chemistry became more similar, it is likely that wetland interstitial water became progressively more acidic, potentially inhibiting bacterial processes that could contribute to effective treatment, favoring dissolution rather than formation of insoluble metal precipitates, and thereby contributing to the eventual failure of the wetlands to effectively treat the AMD. In general, when constructed wetlands are used to treat particularly acidic (pH<4) AMD, if abiotic and biotic alkalinity generation cannot balance the influent acid load, long-term effective treatment will not be achieved.