Do transient storage parameters directly scale in longer, combined stream reaches? Reach length dependence of transient storage interpretations

SummaryLittle work has been done to assess parameterizations and related interpretations (i.e., metrics of exchange) of transient storage modeling (TSM) over multiple spatial scales in streams. In this paper, we simulate conservative solute transport in a small mountain stream over combinations of five consecutive sub-reaches (38 m, 105 m, 281 m, 433 m, and 619 m below injection point) to (1) determine how optimized parameter estimates vary with reach length and reach combination, and (2) evaluate whether equally well-optimized simulations of solute transport in the channel result in varying interpretations of tracer exchange with the storage zone. Each simulated stream solute concentration breakthrough curve (BTC) showed consistently accurate fits to observations. However, our results indicate approximate equifinality (similar fits from different parameter sets) in the simulations of concentrations of stream tracer across individual sub-reaches and combined reaches leading to varying interpretations of transient storage exchange parameterization (i.e., variable optimized parameter estimates) and concentration time series of tracer in the storage zone. These results suggest strong reach-length dependence in simulated exchange. Based on stream BTCs alone, the TSM is useful in characterizing the influence of transient storage on in-stream solute transport, though it does not consistently reproduce storage zone dynamics. Characterization of the solute exchange between the stream and the storage zones remains problematic, and the effects of transient storage cannot be directly compared within overlapping stream reaches, an important consideration in designing and interpreting stream solute transport experiments.

► We simulated solute transport over multiple reaches from the 1973 Uvas Cr data set.
► Combined reach simulations fit field data as well as simulations of each reach.
► Optimized parameters do not scale to longer reaches via length-weighted averages.
► Interpretations of transient storage processes are dependent upon reach length.