Riparian Buffer Effectiveness at Removal of NO3-N from Groundwater in the Middle Coastal Plain of North Carolina

Abstract

Non-point source pollution from agriculture is one of the causes of surface water quality degradation in the Coastal Plain of North Carolina. Riparian buffers are an important best management practice for reducing NO3 concentrations in natural waters, predominantly by vegetation uptake and denitrification. However, there continues to be debate over the optimal design of buffers, specifically buffer width, and vegetation type. This project was designed to investigate the effects of vegetation type, groundwater depth, and buffer width on NO3 removal from groundwater. Four buffers have been established at a research farm in the Middle Coastal Plain of North Carolina to investigate these factors; individual buffers are comprised of five vegetation types, two buffer widths, and two well depths. The influence of vegetation type on NO3-N groundwater decreases were as follows: revegetation had a decrease of 14% (5.75 mg N/L to 4.97 mg N/L); switchgrass had a decrease of 40% (9.19 mg N/L to 5.48 mg N/L); trees had a decrease of 32% (9.18 mg N/L to 6.20 mg N/L); native vegetation had a decrease of 35% (8.36 mg N/L to 5.41 mg N/L); fescue had a decrease of 23% (7.34 mg N/L to 5.67 mg N/L); the control had a decrease of 0% (5.85 mg N/L to 5.86 mg N/L).

Influence of width and depth on NO3-N decreases were as follows: deep wells in 15 m buffers had a NO3-N decrease of 77% (5.76 mg N/L to 1.34 mg N/L), deep wells in 8 m buffers had a decrease of 53% (4.55 mg N/L to 2.13 mg N/L), intermediate wells in 15 m buffers had a decrease of 47% (7.51 mg N/L to 4.00 mg N/L), and intermediate wells in 8 m buffers had a decrease of 14% (8.38 mg N/L to 7.19 mg N/L) There was a significant three-way interaction (p = 0.001) between vegetation type, buffer width, and well depth. This interaction was desegregated by depth: at the deep depth, the effect of switchgrass was significant (p=0.0120) in removal of NO3-N in both the narrow and wide buffer widths. The effect of the revegetation treatment was significant (p=0.0093) at removal of NO3-N in the narrow width. The ratio of NO3-N/Cl was evaluated to determine if dilution of groundwater was responsible for observed NO3-N concentration decreases. Dilution was slight and did not significantly account for any observed NO3-N decreases. Reduction potential (Eh) values indicated reducing conditions at the deep well depth in three of the four buffers, suggesting denitrification was most likely responsible for observed NO3-N decreases in groundwater. Inhibition of denitrification rates could be occurring in buffers due to low levels of organic C (≈3.4 ± 0.6 mg C/L). To test this hypothesis, a laboratory study was designed to complement the field study. Flow-thru soil columns were constructed to determine the effect of dissolved organic carbon (DOC) concentration on denitrification rates and products in buffer soils. Three DOC concentrations (2.0 mg DOC/L, 4.0 mg DOC/L, 8.0 mg DOC/L, and 16.0 mg DOC/L) and a control (0.0 mg DOC/L) were utilized to study this relationship between DOC and denitrification. There was no trend between DOC concentration and rate of NO3-N loss. DOC concentrations > 4.0 mg DOC/L increased up until 12.0 mg DOC/L, after which rates leveled off. There was a linear relationship between DOC concentration and rate of N2O-N production with the exception of 12.0 mg DOC/L, with the rate of N2O-N production increased with increasing concentrations of DOC.