The use of sulfur isotopes to monitor the effectiveness of constructed wetlands in controlling acid mine drainage

Present address: S.C. Hsu, IT Corporation, 111 N. Canal St, Suite 941, Chicago IL 60606, USA Abstract Acid drainage containing high levels of iron, manganese, and sulfate is a pollution problem widely associated with abandoned coal mines. The remediation of these sites commonly falls to public agencies, so policy-makers need to allocate scarce resources over many sites and to find cost-effective remediation strategies. In recent years, constructed wetlands have become a popular alternative to more laborand capital-intensive treatment systems. Within these wetlands, bacterial sulfate reduction occurs and is often cited as an important remediation process, but few measurements of reduction rates have been made. We studied the Wills Creek site in NE Ohio, a series of staged constructed wetlands, to evaluate the use of stable isotopes of sulfur as a rapid technique to assess the effectiveness of wetland design. Sulfur isotopes are widely used in marine sediments to measure the rates of bacterial processes, but have not previously been applied to constructed wetlands. Water chemistry measurements indicated that the wetland system as a whole was effectively removing acidity and iron, but not sulfate. We found that sulfate reduction rates were slow relative to the sulfate loadings and that therefore sulfate reduction was ineffective as a remediation process for sulfate contamination in constructed wetlands designed like those at Wills Creek, in agreement with the five-year record of water chemistry measurements. Moreover, our results showed that cells especially constructed to enhance sulfate reduction in fact showed slower reduction rates than more typical compost wetland cells. We conclude that the concept of staged wetlands, although intuitively appealing, is no more effective than a series of ordinary wetlands of the same size. This study illustrates that stable isotopes of sulfur can be a quick and cost-effective way of assessing the functioning of bacterial remediation processes in constructed wetlands and could be used by regulatory agencies to evaluate various design strategies. 1 The problem of acid mine drainage Acid mine drainage is formed when sulfides (commonly pyrite) from coals and from shales associated with coal beds oxidize by coming in contact with oxygenated water, releasing acid, sulfate, and metals such as Fe, Mn, and Al into the water. The acidic conditions and elevated dissolved metals associated with mine drainage can be toxic to aquatic life and, if introduced into residential wells, potentially harmful to humans. Surface and groundwaters in areas with abandoned coal mines commonly exceed Secondary Maximum Contaminant Limits set by the USEPA for acidity, Fe, Mn, and sulfate. Not all coal-bearing sequences produce significant acid drainage, however. The amount and depositional nature of the pyrite present are important factors. If the coal was formed in a freshwater environment, as is common in the western United States, the sulfur content is generally low and significant acid mine drainage is not produced. However, coal beds formed within marine sequences, such as in the Ohio and West Virginia portion of the Appalachian region, contain several percent of pyrite sulfur and can produce significant acid mine drainage (Langmuir 1997). Also important is the size of pyrite crystals: the smaller the size, and thus the greater the surface area, the faster the pyrite oxidizes. Pyrites formed in marine environments are generally smaller than pyrites formed in freshwater environments (Caruccio and Ferm 1974) and so react faster. Other factors influencing the amount of acid mine drainage include the relationship of the coal bed to carbonate rocks, which can neutralize acid mine drainage, and the relationship of the coal bed to the water table: Coal beds above the water table can be oxidized faster than those below the water table. The processes of pyrite oxidation have been reviewed extensively by Stumm and Morgan (1981). The dominant reactions that take place are