Mountain Streams—Modeling Hydraulics and Substance Transport

Abstract

One-dimensional models are useful tools for predicting hydraulics, substance transport, temperature, and water quality in rivers for flow distances that are long compared to the lateral mixing distance. However, the application of such models to mountain streams is difficult because, in addition to small scale bed roughness, the effects of large-scale roughness, irregularity in bed geometry, and pool and riffle sequences have to be taken into account at a spatially averaged scale. Representation of pools, stagnant zones due to bed irregularity, and interstitial zones in the gravel bed with a dead zone model leads to a description of substance transport that is strongly simplified but correct in the representation of basic mechanisms. However, this approach leads to more severe problems regarding parameter identifiability compared to inclusion of these effects in an enlarged friction coefficient. To overcome these problems, we applied a friction estimator developed for river sections without pools and were then able to identify the dead zone volume and exchange parameters. In a case study in the southern Swiss Alps, the dead zone volume fraction and the residence time in the dead zone did not strongly depend on stream discharge, but these parameters varied strongly from one stream section to another. This indicates that the evaluation of a tracer test for one single flow condition significantly decreases prediction uncertainty of substance transport for a specific stream reach.