TEMPORALLY-VARIABLE HYPORHEIC DYNAMICS IN AN ANTARCTIC STREAM: ANALYSIS AND IMPLICATIONS

A stream tracer injection conducted in the Dry Valleys of Antarctica yielded oscillations and spikes in conservative and reactive chemical constituents. Some of this variability can be directly explained by daily floods during the 33 hour injection. Anabranches formed at high flows, resulting in storage and exchange of water and solutes on a scale orders of magnitude larger than expected given results from previous injections. Variable flows may also affect near-stream hyporheic storage area and exchange rates, but the subsequent variability of solute breakthrough curves limit our ability to quantify these processes via standard methods.

Interpretation of this tracer data was accomplished through coupling of a groundwater flow model with principal component analysis (PCA) and solute breakthrough curves. PCA was run for both conservative and reactive, injected and background solutes. Conservative chemistry PCA correctly quantified stream flow and the timing of solute leaching from stream sediments. Reactive chemistry PCA identified a decreasing trend in nitrate, consistent with hyporheic denitrification. The groundwater flow model was used to simulate storage and exchange of water in the anabranching reach, and successfully predicted temporally variable exchange between the stream and subsurface. These results explain much of the solute breakthrough oscillations, and lay the foundation for interpretation of reactive stream chemistry resulting from denitrification.

The results of this tracer highlight the potential for temporal and spatial variability in hyporheic dynamics, and the utility of modeling hyporheic interactions within a groundwater framework. While transient storage models are typically used to characterize hyporheic zones, this work suggests a groundwater flow model may be better able simulate dynamics that result from hydraulic head gradients. However, groundwater models also require significantly more effort to correctly parameterize. When is such a model worth the effort? In light of our results we propose a new method to quantify stream head variability that may enable faster characterization of exchange rates. When coupled with a groundwater model, these techniques may aid in quantifying hyporheic dynamics and scaling from the reach to the watershed.