Treatment of Trace Organic Contaminants and Nutrients in Open-Water Unit Process Wetlands
- Author(s): Jasper, Justin Thomas
- Advisor(s): Sedlak, David L
- et al.
Treatment wetlands are becoming an increasing popular approach for nutrient removal from municipal wastewater effluent due to their low cost and energy requirements, as well as the ancillary benefits they provide. Recently, they have also been considered as a means of removing wastewater-derived trace organic contaminants. Initial studies of the removal of trace organic contaminants in treatment wetlands indicate that removal is often insignificant and is highly variable among systems. Efforts to improve wetland treatment efficiency have been hampered by a limited understanding of trace organic contaminant removal mechanisms. In this research, a novel type of wetland, consisting of a shallow cell lined with geotextile fabric to prevent the growth of emergent macrophytes, is considered for the removal of wastewater-derived trace organic contaminants and nutrients. When combined with other types of wetland cells in a unit process fashion (i.e., wetland cells designed to remove specific contaminants are arranged in series), open-water unit process wetlands may provide a basis for using treatment wetlands to remove a wide range of trace organic contaminants from municipal wastewater effluent.
To understand how open-water treatment wetlands could be integrated with other types of wetland cells in unit process wetlands, the treatment wetland literature was critically reviewed (Chapter 2). Removal mechanisms of trace organic contaminants and pathogens in treatment wetlands were considered, including sorption, biotransformation, photolysis, sedimentation, predation, and photoinactivation. Methods of enhancing these mechanisms in unit process wetland cells were also identified, both in commonly employed vegetated cells and through the development of novel wetland configurations, such as open-water cells. To further optimize unit process wetlands, the arrangement of wetland cells was evaluated. The application of the unit process concept to a wide range of wastewater contaminants has the potential to make treatment wetlands a more attractive component of urban water infrastructure.
To assess the ability of open-water cells to exploit sunlight photolysis to remove trace organic contaminants from municipal wastewater effluent, a photochemical model was calibrated using measured photolysis rates for atenolol, carbamazepine, propranolol, and sulfamethoxazole in wetland water under representative conditions (Chapter 3). Contaminant transformation by hydroxyl radical and carbonate radical were predicted from steady-state radical concentrations measured at pH values between 8 and 10. Direct photolysis rates and the effects of light screening by dissolved organic matter on photolysis rates were estimated using solar irradiance data, contaminant quantum yields, and light screening factors. The model was applied to predict the land area required to achieve 90% removal of a suite of wastewater-derived organic contaminants by sunlight-induced reactions under a variety of conditions. Results suggest that during summer, open-water cells that receive a million gallons per day of nitrified wastewater effluent can remove 90% of most compounds in an area comparable to existing full-scale wetland systems.
The bottoms of open-water wetland cells are rapidly colonized by a biomat consisting of an assemblage of photosynthetic and heterotrophic microorganisms. To assess the contribution of biotransformation in this system to the overall attenuation of trace organic contaminants, transformation rates of test compounds measured in microcosms were compared with attenuation rates measured in a pilot-scale system (Chapter 4). Biotransformation was the dominant removal mechanism in the pilot-scale system for atenolol, metoprolol, and trimethoprim, while sulfamethoxazole and propranolol were attenuated mainly via photolysis. In microcosm experiments, biotransformation rates increased for metoprolol and propranolol when algal photosynthesis was supported by irradiation with visible light. Biotransformation rates increased for trimethoprim and sulfamethoxazole in the dark, when microbial respiration depleted dissolved oxygen concentrations within the biomat. These observations are consistent with previous studies in wastewater treatment plants and wetlands at different dissolved oxygen concentrations. During summer, atenolol, metoprolol, and propranolol were rapidly attenuated in the pilot-scale system (t1/2 < 0.5 d), trimethoprim and sulfamethoxazole were transformed more slowly (t1/2 ≈ 1.5-2 d), and carbamazepine was recalcitrant (t1/2>30 d). The combination of biotransformation and photolysis resulted in overall transformation rates that were 10 to 100 times faster than those previously measured in vegetated wetlands.
In addition to removing trace organic contaminants, the diffuse biomat formed on the bottom of open-water wetland cells provides conditions conducive to NO3- removal via microbial denitrification, as well as anaerobic ammonium oxidation (anammox). To assess this process, nitrogen cycling was evaluated over a 3-year period in an open-water wetland cell (Chapter 5). Approximately two-thirds of the NO3- entering the cell was removed on an annual basis. Microcosm studies indicated that NO3- removal was mainly attributable to denitrification within the diffuse biomat (i.e., 80±20%), with accretion of assimilated nitrogen accounting for less than 3% of the NO3- removed. The importance of denitrification to NO3- removal was supported by the presence of denitrifying genes (nirS and nirK) within the biomat. While modest when compared to the presence of denitrifying genes, the anammox-specific gene hydrazine synthase (hzs) was detected at higher concentrations near the biomat bottom. This observation, along with the simultaneous presence of ammonium and nitrate in the biomat, suggested that anammox may have been responsible for some of the NO3- removal. The annual temperature-corrected areal first-order NO3- removal rate (k20=59.4±6.2 m yr 1) was higher than values reported for more than 75% of vegetated wetlands treating effluent where NO3- served as the main nitrogen species (e.g., nitrified secondary wastewater effluent and agricultural runoff). Inclusion of shallow, open-water cells in unit-process wetland systems has the potential to provide simultaneous removal of trace organic contaminants (Chapters 3 and 4) and pathogens, in addition to NO3-, in land areas similar to those occupied by existing full-scale treatment wetlands.