Generality of Nitrate Removal in Streambed Sediment on the Southern Delmarva Peninsula
Abstract
Nitrogen fertilizers have accumulated in the unconfined Columbia aquifer on the Eastern Shore of Virginia (ESVA) and increased nitrate concentrations, which could potentially affect water quality in down-gradient surface-water bodies. The streambeds and riparian zones of the small, low-gradient ESVA streams appear to be attenuating the nitrate load of discharging groundwater via denitrification, thereby reducing groundwater nitrate concentrations by an order of magnitude or more in some cases. We measured concentrations of nitrate, chloride, and dissolved oxygen in sediment pore water as well as vertical head gradients and hydraulic conductivity in the streambed of four streams on the ESVA (Coal Kiln, Machipongo, Phillips Creek, and Parker Creek). Unlike measurements made in some other streams on the ESVA, the data did not show a consistent decrease in nitrate concentrations as groundwater discharged upward through the streambed. For Coal Kiln, Machipongo, and Phillips Creek, dissolved oxygen concentrations were consistently low (generally <3 mg/L) throughout sediment pore water, down to at least 60 cm (maximum depth of measurements). In Parker Creek, dissolved oxygen concentrations were higher and there was a general proportionality between nitrate and oxygen concentrations, consistent with microbially mediated redox reactions gradually shifting the electron acceptor from oxygen to nitrate as pore-water moved upward through the sediments. Whereas previous measurements at another stream (Cobb Mill Creek) on the ESVA indicated that denitrification occurred primarily in the upper 30-60 cm of the streambed sediment, the data from Coal Kiln, Machipongo, and Phillips Creek suggest that denitrification may also be important elsewhere in the catchment, perhaps deeper in the sediment profile. In Parker Creek, the streambed appeared to be reducing nitrate concentrations as groundwater discharged to the stream, however, the magnitude of nitrate removal during vertical flow through the bed was insufficient to account for the low nitrate concentrations in the stream. Perhaps a key factor contributing to the larger groundwater nitrate concentrations entering Parker Creek's streambed was due to the discrepancy in pore-water velocities. The average pore-water velocity at maximum sample depth was a magnitude of order larger at Parker Creek than the other streams and could be representative of velocities earlier along flow paths. Calculating commonly used metrics, such as Damkohler numbers for oxygen and nitrate, for inferring the poise of systems for net nitrification or denitrification may not be meaningful when the zones of denitrification and dominant flow paths are complex and uncertain.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.H43A1420M
- Keywords:
-
- 1830 HYDROLOGY Groundwater/surface water interaction;
- 1832 HYDROLOGY Groundwater transport;
- 1813 HYDROLOGY Eco-hydrology