The last 10 years has seen a considerable improvement in our ability to
measure and to model the complex flow fields in a range of river
environments, even to the point that we have numerical schemes and data
acquisition strategies that allow us to model the interaction between
the detailed geometry of individual clasts and clast clusters and the
associated three-dimensional flow and sediment transfer fields. At the
same time, habitat modeling has developed significantly, as we have
developed methods for coupling hydrodynamic models to measured habitat
preferences. Here, we show how we have developed this linkage through
coupling of a depth-averaged hydrodynamic model to habitat preferences,
using a fuzzy rule-based approach. We demonstrate how this approach can
be used to determine habitat suitability for two fish species (Atlantic
salmon and Brown trout), in relation to spawning, nursery and rearing
habitat, and related to proposed changes in compensation flow releases
downstream from a dam, and possible future climate changes to the
2050s. The results demonstrate the crucial need to explore flow-biology
interactions at the within-reach scale, especially in relation to low
flows. In reflecting on these findings, we show that whilst
three-dimensional modeling may reduce some of the uncertainties that
follow from a depth-averaged approach, its applicability is limited by
both the small scale of predictions generated and the lack of
ecological knowledge of how to interpret those scales of prediction.
The scales of habitat that matter in ecological terms are larger and
more continuous, questioning approaches based on modeling
`representative reaches' and necessitating exploration of new ways of
coupling biological and hydrodynamic knowledge.