The underlying goal of most environmental flow assessments is to ensure viability of target instream populations or ecosystems in the face of flow alterations. Rivers are characterized by dynamic feedbacks among system components, a high degree of spatial and temporal variability, and connectivity between habitats. Accordingly, recent reviews have advocated for environmental flow assessment methods that create explicit links between changes in physical habitat and other abiotic features, the supply of macroinvertebrates that comprise fish diets, and long-term viability of target fish species. An ongoing challenge is to identify ways of recognizing these characteristics in practical methodology. Here, we focus on tools for linking physical habitat variation and changes in flow to variation in benthic macroinvertebrate densities that can be incorporated into indices of fish habitat quality. We begin by modeling macroinvertebrate transport using a field validated two dimensional flow model of the Robinson Reach in the Merced River, California, U.S.A., where macroinvertebrates are the food source for threatened Chinook salmon. We demonstrate how results from the computationally intensive two dimensional flow model can be used to construct simpler one dimensional representations, which we use in turn to explore macroinvertebrate redistribution across river microhabitats under a range of discharge conditions.
Results/Conclusions
We found that the trajectories of simulated invertebrates were dominated by a high velocity core under baseflow and 75% bankfull discharge conditions. Resulting macroinvertebrate redistributions were well-described by a one-dimensional dispersal function that included flow dependence. We collapsed both the two dimensional flow field and resulting macroinvertebrate transport into one dimensional representations that allowed their use in population dynamic models for the invertebrates. This in effect extrapolated the dynamics observed in the two dimensional flow model to broader spatial and temporal scales with minimum computational effort. Simulations incorporating the one dimensional flow representation and resulting dispersal dynamics yielded distributions of invertebrates that showed a strong inverse relationship with flow velocity, especially between pools and riffles. The strength of the relationship was influenced by other parameters, namely the rate at which drift dispersal is initiated and the rate at which dispersers settle to the benthos. We will discuss ongoing parameterization efforts and how variation in modeled macroinvertebrate densities can be used as inputs to salmon bioenergetic and population dynamic models.