Measuring and Modeling the Hydraulic Environment for Assessing Instream Flows

Abstract

Detailed measurements of water depth and velocity in natural channels, though rare, show that the velocity fields are complex and irregular even in streams with moderate gradients and gravel substrates. This complexity poses a challenge for instream flow studies, most of which use the physical habitat simulation (PHABSIM) model, a set of computer models that combine the results of hydraulic modeling with estimates of channel substrate or cover and habitat suitability criteria to compute weighted usable area (WUA), an index of habitat. Some recent studies have replaced the transect-based one-dimensional hydraulic modeling in PHABSIM with two-dimensional models that allow better definition of the depth and velocity fields in the modeled stream reach. The accuracy of the estimates as a function of channel geometry and data collection effort remains unclear, however, as does the utility of the estimates for evaluating instream flow needs. Here we review the assumptions, accuracy, and precision of hydraulic modeling and the measurements that provide input data for the models; we also consider some implications of the limitations of hydraulic modeling for describing fish habitat and assessing instream flows. Highly accurate hydraulic modeling seems infeasible for streams with complex channel geometry, and in any event practical hydraulic modeling cannot resolve flow patterns at the short length scales at which fish often respond to the hydraulic environment. Information on depth, velocity, and substrate is important for assessing instream flows, but information from hydraulic models should be treated with great caution and is not a substitute for biological understanding.