Based on our results we recommend designing interspersed wetlands that offer more opportunities for a variety of biogeochemical processes to occur and using sediments with high carbon concentrations to promote denitrification.
2013 Urban Fellow
Research Topic: Water Quality and Hydrology
Faculty Advisor: Shimon Anisfeld
Factors Controlling Biogeochemical Removal of Nitrogen in Constructed Wetlands
The Long Island Sound (LIS) estuary is affected by summer hypoxia as a result of high nitrogen loads from New York and Connecticut watersheds. In order to mitigate hypoxia, managers have established a goal of reducing the nitrogen load from nonpoint sources by 10%. One strategy to reduce N loads from nonpoint sources is the use of constructed wetlands, which provide an ecosystem service by removing pollutants from stormwater runoff. This study examined the effectiveness of constructed wetlands in Hamden and Woodbridge, Connecticut in improving the quality of stormwater runoff. Our main objective was to determine the factors that contribute to N removal to provide design recommendations that optimize constructed wetlands performance. A total of 9 to 21 storms were monitored at four sites during the summer and fall of 2013. Weirs and water level loggers were installed at the inlet and outlet of the wetlands to measure water flow. Stormwater samples were collected using ISCO autosamplers at regular intervals over the duration of storm events. These were composited to obtain flow-weighted samples from the inlet and outlet of each wetland to determine nitrogen loads and mean concentrations per storm event. We also surveyed each site to determine plant diversity, sediment organic carbon concentration, and treatment ratios to determine their influence on N removal. Only two sites showed statistically significant biogeochemical removal of N. Our results indicate that wetland heterogeneity and interspersion between open water and vegetation, as well as high sediment carbon concentrations, promote N concentration reduction. Additionally, we examined the effects of input N concentrations, storm size and intensity, and water temperature using multiple linear regression. The models showed that only influent N concentration influences N concentration reduction. Based on our results we recommend designing interspersed wetlands that offer more opportunities for a variety of biogeochemical processes to occur and using sediments with high carbon concentrations to promote denitrification. Considering these variables might result in more effective N concentration reduction. This information contributes to the limited knowledge of constructed wetland design in Connecticut and can promote higher nitrogen removal rates from stormwater in the Long Island Sound watershed.