Estimated reactive water contributions to discharge during the July 17-18, 2001 flood.

M. Conklin, J. Villinski (UA-HWR)

Our overall goal is to understand the coupling between the hydrologic cycle and nutrient inputs into riparian systems. Our hypothesis is that during flood events different compartments of the riparian system are linked due to overland flow and subsurface water inputs.

Science Questions

Our overall question is: What role do flooding events play in contributing nutrients to semi-arid rivers, both temporally and spatially?

• What are the important hydrologic pathways for nutrient delivery to semi-arid rivers? In particular, what are the roles of tributaries, relic floodplains, and channels for nutrient delivery?
• What are the linkages between sediment scour and fill (due to flooding events) and nutrient uptake rates in the sediments?

Approach

• Two autosamplers have been placed at both ends of our Boquillas Ranch study reach. Samples are collected hourly during storm events, and analyzed for major anions, nitrate, and ammonium. Precipitation and soil water samples were also collected.
• Two transects of scour pans and chains have been established to measure sediment scour and fill in a point bar at the Boquillas Ranch site. Over the last six months, water quality sampling and monthly respiration measurements were taken along a flowpath. Tracer tests were performed to establish flowpaths on the point bar at Boquillas Ranch.

Results

Two storms flood events were sampled: a complete small flood event (July 17-18, 2001) and the receding limb of a larger flood (August 8-10, 2001).

Sampling Event of July 17-18, 2001. Using the nitrate, chloride, oxygen isotope and dissolved organic carbon (DOC) data, two scenarios can be proposed: a) there is a continuous release of water from either post-flood bank drainage or another source after peak flow or b) in-stream metabolic processes are at work during the receding limb. The second scenario is less likely, due to the short duration of the event and the condition that most organisms are rapidly transported downstream during the flood. These results have also been analyzed using a mixing model to determine the extent to which a particular hydrologic component has contributed to stream flow (Figure 3-4). Oxygen isotope and Cl- concentrations in precipitation, stream water and soil water were used as endmembers. Our mixing model results yielded that old water contribution was greatest just before and during peak flow, indicating a significant contribution from either groundwater or soil water flushing.

Sampling Event of August 8-10, 2001. August samples were taken during the receding limb of a larger flood (> 300 cfs). These data illustrate the tendency for the stream to return to its normal chemistry levels after a flood. However, all stream chemical signals exhibited a significant change ~15 hours before the smaller flood occurred. DOC data reflect this phenomenon, but not as dramatically. These results provide evidence that a) relic flood channels may be releasing water to the stream during the flood recession, b) flood flow is "pushing" water through these relic flood channels, or c) bank drainage is occurring.

Point bar work to date has established water quality and respiration conditions that occur before and after leafout:

• Tracer injections were performed in February (pre-leafout) and May (post-leafout) to establish the direction of subsurface flowpaths.
• Respiration rates are highest at the stream-parafluvial interface due to higher dissolved oxygen and nutrient concentrations flowing into the parafluvial zone from the stream.
• Respiration rates do not appear to be depth-dependent in the spring baseflow period.
• After leafout, aerobic microbial activity in parafluvial sediments, which are anoxic before disturbed and extracted, is positively correlated with dissolved oxygen concentration. This finding implies that despite an extended period without aerobic conditions, the parafluvial microorganisms are able to use oxygen as soon as it becomes available.

Plans

• Continue to monitor sediment and nutrient fluxes on a meander point bar. Complement these measurements with laboratory respiration measurements using groundwater from the site with and without nutrient additions.
• Repeat current water quality and sediment scour measurements in a reach that contains a tributary. Most river basins have key geomorphic characteristics. To scale up the nutrient work to a basin scale, we need to examine nutrient movement associated with geomorphic characteristics. Tributaries should provide an inflow of both nutrients and sediments during flood events.
• Focus on the role of flowpaths in floodplains and relic channels for nutrient contributions during and after flood events. As we are trying to identify the important geomorphic characteristics that facilitate nutrient movement, characterizing the role of these units will help to provide a basin-wide description of nutrient movement and will facilitate the development of an ecosystem model.
• Develop real-time instrumentation for a cross-section of the San Pedro to quantify water quality and sediment movement on different temporal scales. A proposal by Villinski and Conklin will be submitted to leverage the infrastructure in place through SAHRA.