How sampling integration scale affects estimates of coupled groundwater and nitrogen fluxes into an agricultural stream
Abstract
Coupled fluxes of groundwater and non-point source contaminants from aquifers to streams may be estimated using streambed point (SP) measurements, seepage meter (SM) measurements, or a reach mass-balance approach (RMB), each with different spatial support scales and pros/cons. We have applied all three measurement schemes concurrently in the same stream to assess how sampling integration scale affects estimates of (1) coupled groundwater and dissolved nitrogen fluxes through a streambed and (2) the total amount of denitrification that has occurred along groundwater flowpaths. Our study site was a 2700m reach in West Bear Creek (WBC), a channelized and overall gaining stream in the agricultural Coastal Plain of North Carolina. In a July 2012 sampling campaign, groundwater fluxes through the WBC streambed were similar for the SP and RMB approaches (35 cm/day and 33 cm/day, respectively), despite very different measurement scales and different reach sizes (RMB groundwater flux is based on a 200m stream reach containing the smaller 58m SP reach). However, the RMB approach gave a lower calculated streambed nitrate flux (136 mmol m-2 d-1, versus 231 mmol m-2 d-1 for SP) for the 58m reach. The lower nitrate flux by the RMB approach is linked to a lower mean groundwater nitrate concentration estimated by RMB (361 μM, vs. 808 μM for SP). Unlike the SP approach, the RMB approach samples groundwater that has had significant interaction with the stream channel and thus, apparently, nitrate loss from uptake and/or denitrification. The SM approach used novel flexible streambed 'blankets' and gave lower fluxes: 10 cm/day for groundwater (due perhaps to incomplete sampling of streambed variability in this flux or other methodological issues) and 53 mmol m-2 d-1 for nitrate; it also gave an intermediate estimate of nitrate concentration in the groundwater discharge to the stream (527 μM), likely a reflection of the intermediate amount of channel interaction (collected after passing through the hyporheic zone, but before subsequent channel interaction) for the groundwater sampled by this approach. Noble gas concentrations (Xe, Ar, Ne, Kr) are being used to model the amount of N2 derived from denitrification (N2-den) in the groundwater feeding the stream. Preliminary results from a subset of SP samples (n=9) suggest significant amounts of N2-den because measured groundwater N2 concentration is up to 75% higher than modeled N2 concentration. The three approaches offer different strengths and weaknesses appropriate for answering different questions, and in concert may provide a fuller picture of N fluxes from groundwater to surface water in areas of non-point N pollution.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.H12E..03G
- Keywords:
-
- 1832 HYDROLOGY Groundwater transport;
- 1830 HYDROLOGY Groundwater/surface water interaction;
- 1803 HYDROLOGY Anthropogenic effects;
- 1895 HYDROLOGY Instruments and techniques: monitoring