Untitled DocumentQuantifying terrestrial-aquatic linkages and biogeochemical fluxes in semi-arid watersheds
Macro Theme Area:
River Systems [Project ID: R12]
PI:
James Hogan
CO-PI(s):
Tom Meixner, Paul Brooks
Basin focus:
Rio Grande, San Pedro
Specific area in
basin /
field sites:
San Pedro (Mexican border to Benson), Middle Rio Grande, Colorado
Summary/Goals: We are quantifying relationships between land use, vegetation characteristics, riparian structure and both water quantity and quality. Focus is regional with present work along the Rio Grande and San Pedro and proposed leveraged funding to extend this work to other regions within the Colorado River Basin.
Activities and outcomes during past year:
Research on the San Pedro (3 MS and one Ph.D students) and the Rio Grande (5 years of synoptic data) are ongoing. Below we detail recent results.
For the Middle Rio Grande (MRG) in central New Mexico, a persistent monsoon regime brought heavy rains during July and August making 2006 the 8th wettest year on record for Albuquerque, NM. This summer's heavy rains increased river discharge and inputs from tributaries and ephemeral streams which served to reconnect the river to its floodplain and riparian area and transport solutes to the river from normally disconnected sources. Discharge in the Middle Rio Grande was 39 cms in 2006 which is 250% higher than the discharge during the years 2001 and 2005. Isotopic data indicate that local monsoon precipitation is the primary source for this increase discharge. Under non-flood conditions, wastewater treatment plants (WWTP) are the primary source of nitrogen to the river. However, N loading from the Albuquerque WWTP was 40% lower in 2006 than previous years due to a decrease in effluent TDN concentrations. Tributaries had similar TDN concentrations in all years, but due to increased discharge, TDN loads from tributaries were an order of magnitude higher in 2006 and exceeded the TDN input from WWTPs. In all years, agricultural drains had lower TDN concentrations than the river water originally diverted for irrigation, suggesting that the agricultural areas function as a sink for nitrogen under a wide range of hydrologic conditions. Increased TDN concentrations and discharge rates resulted in a five-fold increase of N loading to Elephant Butte Reservoir from 1000 kgN/day to more than 5000 kgN/day. Analysis of DOC and stable isotopes of water should help to determine how hydrologic variability changes the relative role of different land uses and land covers as nutrient sources to the river.
Finally we used two different and simple models to evaluate the abiotic and biotic processes that generate the observed large-scale patterns in the Rio Grande. First, we used a Cl mixing model, validated with Br to quantify the effects of evapoconcentration, tributaries, and point sources on dissolved nitrogen and carbon concentrations. Ratios of observed to predicted concentrations close to 1 suggest that abiotic hydrologic processes are the dominant controls on concentrations. Alternatively, ratios lower or higher than 1 indicate that biological processes provide important controls on nutrient concentrations. Our conservative mixing model generally captured patterns in DOC concentrations, suggesting minimal, net biological processing. In contrast, both nitrate and TDN concentrations were altered biogeochemically in all reaches. In areas where observed and predicted values differed, the spatial variability of river characteristics was more strongly correlated to relative nutrient retention than seasonal or inter-annual discharge variability. Second, we used an integrated surface water - groundwater dynamic simulation model to evaluate the agricultural conveyance and riparian systems as potential locations for nitrogen removal. Under conservative behavior, modeled nitrate concentrations were higher than observed in the groundwater, river, and conveyance channels. We calibrated the model using denitrification in the groundwater and uptake by riparian vegetation and crops so that modeled concentrations matched observed concentrations. Neither uptake of nitrate by riparian vegetation and crops or denitrification alone reduced nitrate sufficiently. However, a combination of 10% denitrification in the groundwater, riparian uptake equal to 90% of ET and crop uptake equal to 50% of ET resulted in nitrate concentrations that generally matched the magnitude and seasonal variations of observed nitrate concentrations.
Plans for the upcoming year:
Future plans include:
1) Collaborate with other PI's working in the San Pedro to integrate results of the numerous individual projects performed over the last few years
2) Current San Pedro results, combined with the high spatial resolution sampling, existing data on potential rates of N transformation, and a dynamic flow model are being used to parameterize the DAYCENT biogeochemical for the stream-riparian-hyporheic corridor. If successful, this effort will provide a mechanism for scaling hotspots and hot moments in these dynamic environments.
3) Expand work along Rio Grande riparian systems in collaboration with Sandia National Lab (modeling with Vince Tidewell) and UNM (process measurements with Cliff Dahm)
4) Continue dynamic simulation modeling of water quantity and quality in collaboration with Kevin Lansey (funded by TRIF)
5) Leveraged funding pending to extend this research through integrated modeling, which combines surface water, groundwater and vegetation modeling, and extends this research to other riparian areas in Arizona.
Organization Involvement:
Dyanamic simulation modeling and process measurements along Rio Grande riparian sytems
Shared Resources / Joint Activities:
We obtained a fellowship funded by Sandia National Lab for PhD student Gretchen Oeslner. Gretchen Oelsner, working with research from Sandia and Cliff Dahm's group will focus on improving the riparian water balance (riparian ET) and linking water and nutrients at the scale of a river reach (10 to 30 km).
