Developing temperature-dependent habitat suitability curves for instream flow modeling

Abstract Objective Habitat simulation models used to define instream flow needs for fish typically assume that flow–habitat relationships are constant across temperature, which is at odds with empirical observation, where velocity and depth microhabitat selection appears temperature dependent. To address this issue, an approach and workflow is presented for generating temperature-dependent habitat suitability curves (HSCs)—the biological component of habitat simulation models. Methods The approach builds on the development of bioenergetic net rate of energy intake models that describe the energy balance of drift-feeding fish as a function of hydraulics (depth and velocity) and prey availability. I describe a process to express net rate of energy intake model predictions on a daily time step that incorporates temperature-dependent energy costs and consumption. Results A simple worked example for juvenile Coho Salmon Oncorhynchus kisutch is provided that illustrates temperature dependence of velocity and depth habitat HSCs. The general model behavior broadly aligns with ecological theory and observations of salmonid microhabitat selection and behavior across temperature gradients. In particular, models predict a shift toward higher optimal velocity as temperature warms, which is a function of increasing metabolic demands. Conclusions Temperature-dependent HSCs allow temperature variation to be directly considered within instream flow modeling. Thus, they may have high utility for management applications concerned with interactive changes in flow and temperature. However, there are key uncertainties and limitations with both model structure and ecological processes that should be considered when these models are applied. Given the certainty of warming rivers with climate change and land use activity, there is a need for a better empirical understanding of warming effects on environmental flow needs, as well as applied approaches for modeling them.