COS 43-5 - Riparian denitrification prevents groundwater nitrate from reaching an intermittent blackwater stream

Wednesday, August 14, 2019: 9:20 AM
M111, Kentucky International Convention Center
C. Rhett Jackson, Warnell School of Forestry & Natural Resources, University of Georgia, Athens, GA, Johnson Jeffers, Warnell School of Forestry and Natural Resources, University of Georgia, Athens, GA, Natalie A. Griffiths, Environmental Sciences Division and Climate Change Science Institute, Oak Ridge National Laboratory, Oak Ridge, TN, Benjamin M. Rau, USDA Forest Service, Pisgah Forest, NC, Menberu Bitew, University of Georgia and Kellie B. Vache, Dept. of Biological and Ecological Engineering, Oregon State University, Corvallis, OR
Background/Question/Methods

Leaching of excess N fertilizer from managed lands and the resulting eutrophication of aquatic systems are among Earth’s major biogeochemical issues. Here we examined why groundwater contaminated with 0.5-1.5 mg/L nitrate has not affected nitrate concentrations in riparian groundwater or streamwater. Is the lack of stream response due to travel time delays or riparian denitrification? In 2010, our project team initiated an interdisciplinary evaluation of the water quality effects of short-rotation woody bioenergy crop plantations on one reference and two treatment watersheds at the Savannah River Site in South Carolina. After two years of pre-treatment monitoring, 50% of each treatment watershed was clearcut in 2012 using modern forestry BMPs including intact forested riparian buffers on all wetlands and streams. The harvest areas were then converted to intensive short-rotation pine production with planting in 2013. The watersheds are instrumented with in stream flumes, riparian piezometers, wells, interflow interception trenches, soil moisture sensors, and an eddy flux tower in a bioenergy plantation. Denitrification in SMZs was estimated using measurements of precursors and products (N2O and N2:Ar) in shallow groundwater.

Results/Conclusions

Groundwater nitrate concentrations jumped to 0.5-1.5 mg/L less than a year after the first fertilization, and these concentrations have since held steady in most wells, although plot studies suggest that leaching from the root zone slowed considerably after 2015. TN and nitrate concentrations in riparian piezometers and stream water have not responded to the elevated groundwater nitrate, although groundwater travel time modeling indicates N from near-stream plantation areas should have reached the stream. Two years of measurements of N gradients in a network of shallow wells from hillslope to stream and in-situ analysis of denitrification products indicate rapid denitrification in forested wetland valleys. Based on the reaction progress calculated from dissolved gas and initial nitrogen concentrations from data collected, net denitrification is reducing more than 80% of the nitrogen in the shallow groundwater system within the valley of the SMZ with upland locations reducing between 47-60%. However, N2O is the dominant product of incomplete denitrification in the upland environment. This case study suggests that intact forested riparian wetlands can partially or completely protect streams from nitrate-contaminated groundwater under certain conditions.