Can Weighted Useable Area Predict Flow Requirements of Drift‐Feeding Salmonids? Comparison with a Net Rate of Energy Intake Model Incorporating Drift–Flow Processes

Abstract

We compared a process‐based invertebrate drift and drift‐feeding net rate of energy intake (NREI) model and a traditional hydraulic‐habitat model (using the RHYHABSIM [River Hydraulics and Habitat Simulation] software program) for predicting the flow requirements of 52‐cm Brown Trout Salmo trutta in a New Zealand river. Brown Trout abundance predicted by the NREI model for the constant drift concentration–flow scenarios were asymptotic or linear, depending on drift concentration, increasing through the mean annual low flow (MALF; 17 m3/s). However, drift concentration increased with flow, consistent with passive entrainment. The predicted fish abundance–flow relationship based on flow‐varying drift concentration increased logistically, and more steeply, with flow through the MALF and beyond. Predictions for the relationship between weighted useable area (WUA) and flow were made for three sets of drift‐feeding habitat suitability criteria (HSC) developed on three midsized and one large New Zealand river (flow at sampling was 2.8–4.6 m3/s and ~100 m3/s, respectively) and the South Platte River, Colorado (flow at sampling, 7–18 m3/s). The midsized‐river HSC ascribe lower suitability to water velocities > 0.6 m/s. They predicted WUA peaking at 10–11 m3/s, well below the MALF. The WUA–flow relationships for the two large‐river HSC were asymptotic at about 22 m3/s. Overall, WUA appears to underestimate the flow needs of drift‐feeding salmonids. The NREI model showed that assessing flow needs of drift‐feeding fish is more complex than interpreting a WUA–flow relationship based only on physical habitat suitability. The relationship between predicted fish abundance and flow is an emergent property of flow‐dependent drift‐foraging dynamics interacting with flow‐dependent drift concentration and drift flux, local depletion of drift by feeding fish, and flow‐related replenishment of drift from the bed and dispersion. It is time that the principles and predictions of drift–NREI models influence assessments of habitat capacity and instream flow needs of drift‐feeding fish. Received April 10, 2015; accepted November 10, 2015 Published online April 27, 2016

Species traits and channel architecture mediate flow disturbance impacts on invertebrate drift

“…Because concentration is often the common standard for measuring drift (references in Dewson et al . 2007), another implication from our study is the need to carefully consider dilution effects when interpreting drift‐flow relationships (Hayes et al ., ; Wooster, Miller & DeBano, ).…”

Developing flow–ecology relationships: Implications of nonlinear biological responses for water management

“…Effects of discharge on a) drift concentration (circles; concentration = 3.53 · discharge −0.288 ; R 2 = .33, F 1,41 = 20, p < .0001) and drift flux (diamonds; flux = 3.66 · discharge 0.69 ; R 2 = .76, F 1,45 = 144, p < .0001) in the Umatilla River, Oregon (Wooster et al., , see Supporting Information). Note that total flux is positively related to discharge, while drift concentration decreases with flow; (b) contrasting flow–ecology relationships for drift concentration (circles) and flux (diamonds) in the Mataura River, New Zealand, where both drift flux and concentration increase with discharge (filled symbols represent means; Hayes et al., , see Supporting Information); (c) flow‐related changes in available energy in the drift (solid line) and weighted useable area ( WUA ; broken line) for a single stream pool in Husdon Creek, British Columbia; both drift and WUA are standardised to a maximum of 1 (from Rosenfeld & Ptolemy, ); (d) flow‐related changes in WUA (dotted line; based on standard habitat simulation modelling) or trout abundance based on bioenergetic modelling (from Hayes et al., ). The broken line represents standardised abundance of fish on a maintenance ration versus predicted abundance of fish experiencing positive net energy intake (solid line).…”

“…The broken line represents standardised abundance of fish on a maintenance ration versus predicted abundance of fish experiencing positive net energy intake (solid line). Note the strongly contrasting thresholds and intercepts between changes in WUA versus predicted trout abundance (Hayes et al., ) [Colour figure can be viewed at wileyonlinelibrary.com]…”

“…This is illustrated with modelling by Hayes et al. (), who compared changes in available habitat estimated with conventional habitat suitability curves to predicted changes in trout abundance using a bioenergetic‐drift foraging model. Although WUA modelled using PHABSIM showed minimal overall trend with declining flow (i.e.…”

“…a positive y ‐intercept at zero flow), trout abundance predicted using energy flux showed strong declines for fish on a maintenance ration, which became increasingly nonlinear (negative y ‐intercept at zero flow) if fish were assumed to have positive growth (Figure d). Consequently, the implicit assumption in standard habitat simulation modelling that prey concentration is flow invariant needs to be carefully assessed, since this may exert strong leverage on the shape of the habitat capacity versus discharge curve (Hayes et al., ; Rosenfeld et al., ). Divergent predictions from habitat simulation versus bioenergetic‐drift foraging models further emphasise the importance of quantitatively understanding how flow affects prey abundance in general (Naman et al., ), and of validating predicted population changes from habitat modelling against measured changes in fish abundance (e.g.…”