Hydrology, Nutrient Removal, and Cost Effectiveness of Small, Edge-of-Field Tile Drainage Treatment Wetlands

Abstract

Constructing treatment wetlands is a recommended practice for mediating nutrient pollution from non-point sources in the Mississippi River Basin. This research investigated the nitrogen and phosphorus removal effectiveness of a small, edge-of-field, constructed treatment wetland using field, laboratory, and modeling data. In the field, the wetland removed 67% (48-100%) of nitrate discharging from tile drainage but released soil legacy phosphorus from 2013 through 2016. Denitrification in the shallow groundwater and vegetation harvest were the greatest sinks for nitrogen and phosphorus, respectively. In the laboratory, three plant communities from the wetland (a wet prairie forb-dominant mix, a switchgrass and prairie cordgrass-dominant community, and a reed canary grass monoculture) were compared for nitrate removal. The wet prairie mix removed the most nitrate, and it had the lowest dissolved oxygen concentration and greatest ratio of denitrifying bacteria to total bacteria (nosZ:16S rRNA genes) – measured using a quantitative polymerase chain reaction (qPCR) – in its root zone. For the modeling component, the ACPF toolbox, the SWAT model, and a spreadsheet model were used to estimate the mass of nitrate-N removed from tile drainage if more edge-of-field wetlands were constructed in the Elm Creek HUC12 watershed. These smaller wetlands removed more nitrate-N per wetland area than larger wetlands (watersheds > 60 ha) but cost the same per mass removed if the small wetlands were designed to have a high saturated hydraulic conductivity. Results from this study suggest that edge-of-field wetlands can be more effective with a dual treatment of surface flow and shallow groundwater flow for nitrate removal and vegetation harvest for phosphorus removal. However, reed canary grass invasion could potentially decrease the nitrate removal effectiveness. If the wetland soils have a high conductivity, the smaller, edge-of-field designs could be as cost effective as large treatment wetlands but remove less land from agricultural production. This dissertation is composed of three individual chapters that will be published in peer reviewed scientific journals. The first chapter pertains to a field study that observed a small, edge-of-field tile drainage treatment wetland. This chapter will be submitted to Ecological Engineering. In the second chapter, the nitrate removal in three plant communities from the wetland was compared using mesocosms. Total bacteria and denitrifying bacterial populations in the root zones of these communities were also compared using qPCR. The work from this chapter was submitted to the Journal of Environmental Quality and is currently under review. The final chapter will be submitted to Agriculture, Ecosystems & Environment. This chapter compared the effectiveness of small, edge-of-field treatment wetlands with watersheds less than 60 ha to large treatment wetlands with watersheds greater than 60 ha. Multiple models were used to determine the best locations for each wetland in the Elm Creek watershed in southern Minnesota. Conclusions were drawn that small, edge-of-field wetlands are effective nutrient removal practices and can be improved with high saturated hydraulic conductivity, harvested vegetation, and diverse plant communities.