Impacts of Channel Morphodynamics on Fish Habitat Utilization

Abstract

It is reasonable to expect that hydro-morphodynamic processes in fluvial systems can affect fish habitat availability, but the impacts of morphological changes in fluvial systems on fish habitat are not well studied. Herein we investigate the impact of morphological development of a cohesive meandering stream on the quality of fish habitat available for juvenile yellow perch (Perca flavescens) and white sucker (Catostomus commersonii). A three-dimensional (3D) morphodynamic model was first developed to simulate the hydro-morphodynamics of the study creek. The results of the morphodynamic model were then incorporated into a fish habitat availability assessment. The 3D hydro-morphodynamic model was successfully calibrated using an intensive acoustic Doppler current profiler (ADCP) spatial survey of the entire 3D velocity field and total station surveys of topographic changes in a meander bend in the study creek. Two fish sampling surveys were carried out at the beginning and the end of the study period to determine presence–absence of fish as an indicator of the habitat utilization of each fish species in the study reach. It was shown that morphological development of the stream was a significant factor for the observed changes in the habitat utilization of juvenile yellow perch. It is shown that juvenile yellow perch mostly utilized habitat where deposition occurred whereas they avoided areas of erosion. The results of this study and the proposed methodology could provide some insights into the potential impact of sediment transport processes on the fish occurrence, and distribution and has implications for management of small fluvial systems.

Appendix

The Delft3D hydrodynamic model solves 3D Navier–Stokes equations for incompressible flow under Boussinesq assumptions. The partial differential equations include the following flow and momentum continuity equations:

$$\frac{{\partial \eta }}{{\partial t}} + \frac{{\partial (hU)}}{{\partial x}} + \frac{{\partial (hV)}}{{\partial y}} = 0$$

(2)

$$\frac{{\partial u}}{{\partial t}} + u\frac{{\partial u}}{{\partial x}} + v\frac{{\partial u}}{{\partial y}} + w\frac{{\partial u}}{{\partial z}} = - g\frac{{\partial \eta }}{{\partial x}} + \nu _{\mathrm {h}}\left( {\frac{{\partial ^2u}}{{\partial x^2}} + \frac{{\partial ^2u}}{{\partial y^2}}} \right) + \frac{\partial }{{\partial z}}\left( {\nu _{\mathrm {v}}\frac{{\partial u}}{{\partial z}}} \right)$$

(3)

$$\frac{{\partial v}}{{\partial t}} + u\frac{{\partial v}}{{\partial x}} + v\frac{{\partial v}}{{\partial y}} + w\frac{{\partial v}}{{\partial z}} = - g\frac{{\partial \eta }}{{\partial y}} + \nu _{\mathrm {h}}\left( {\frac{{\partial ^2v}}{{\partial x^2}} + \frac{{\partial ^2v}}{{\partial y^2}}} \right) + \frac{\partial }{{\partial z}}\left( {\nu _{\mathrm {v}}\frac{{\partial v}}{{\partial z}}} \right)$$

(4)

In shallow water applications, the vertical momentum equation is reduced to the hydrostatic pressure assumption:

$$\frac{{\partial p}}{{\partial z}} = - \rho g$$

(5)

where h is the water depth, η is the water surface elevation, U and V are the depth-averaged velocities in x and y directions, respectively, and u, v, and w denote velocity components; g is the gravitational acceleration; ρt is the time; υ_h and υ_v are, respectively, horizontal and vertical kinematic eddy viscosity coefficients.

After applying the approach of Reynold’s averaging, turbulence closure models are employed to solve the Reynolds-averaged Navier–Stokes (RANS) equations. Delft3D-Flow code is numerically solved based on the finite difference method. We employed σ coordinate system in which the vertical layers are bounded by the planes which follow the free surface and the bottom topography. The k–ε turbulence closure model, based on eddy viscosity theory of Kolmogorov and Prandtl (Deltares 2014), was used to calculate the 3D turbulence. The morphodynamic module of Delft3D is capable of simulating the sediment transport of suspended load and bedload for non-cohesive sediments and suspended load for cohesive sediments. As mentioned in Section 2, the study creek has cohesive bed and bank materials. For suspended sediments, Delft3D solves the 3D advection–diffusion equation:

$$\frac{{\partial c}}{{\partial t}} + \frac{{\partial uc}}{{\partial x}} + \frac{{\partial vc}}{{\partial y}} + \frac{{\partial \left( {w - w_{\mathrm {s}}} \right)c}}{{\partial z}} = \frac{\partial }{{\partial x}}\left( {D_x\frac{{\partial c}}{{\partial x}}} \right) + \frac{\partial }{{\partial y}}\left( {D_y\frac{{\partial c}}{{\partial y}}} \right) + \frac{\partial }{{\partial z}}\left( {D_z\frac{{\partial c}}{{\partial z}}} \right)$$

(6)

where c is mass concentration of the sediment (kg/m³), D_x, D_y, and D_z are sediment eddy diffisivities (m²/s), and w_s is sediment settling velocity (m/s). Eddy diffisivities and local flow velocities are calculated according to hydrodynamic model results. Delft3D calculates the sedimentation and erosion of the cohesive sediment employing the Partheniades–Krone formulations (Partheniades 1965):

$$E = {MS}(\tau _{{\mathrm {cw}}},\tau _{{\mathrm {cr,e}}})$$

(7)

$$D = w_{\mathrm {s}}c_{\mathrm {b}}S(\tau _{{\mathrm {cw}}},\tau _{{\mathrm {cr,d}}})$$

(8)

$$c_{\mathrm {b}} = c\left( {z = \frac{{\Delta z_{\mathrm {b}}}}{2},t} \right)$$

(9)

where E is erosion flux, M is a user-defined erosion parameter, D is deposition flux, c_b is the average sediment concentration in the near bottom computational layer, $S(\tau _{{\mathrm {cw}}},\tau _{{\mathrm {cr,e}}})$ is an erosion step function:

$$S\left( {\tau _{{\mathrm {cw}}},\tau _{{\mathrm {cr,e}}}} \right) = \left\{ {\begin{array}{*{20}{c}} {\left( {\frac{{\tau _{{\mathrm {cw}}}}}{{\tau _{{\mathrm {cr,e}}}}} - 1} \right),} & {{\mathrm {when}}\;\tau _{{\mathrm {cw}}} > \tau _{{\mathrm {cr,e}}}} \\ {0,} & {{\mathrm {when}}\;\tau _{{\mathrm {cw}}} \le \tau _{{\mathrm {cr,e}}}} \end{array}} \right.$$

(10)

$S\left( {\tau _{{\mathrm {cw}}},\tau _{{\mathrm {cr,d}}}} \right)$ is a deposition step function:

$$S\left( {\tau _{{\mathrm {cw}}},\tau _{{\mathrm {cr,d}}}} \right) = \left\{ {\begin{array}{*{20}{c}} {\left( {1 - \frac{{\tau _{{\mathrm {cw}}}}}{{\tau _{{\mathrm {cr,d}}}}}} \right),} & {{\mathrm {when}}\;\tau _{{\mathrm {cw}}} < \tau _{{\mathrm {cr,d}}}} \\ {0,} & {{\mathrm {when}}\;\tau _{{\mathrm {cw}}} \ge \tau _{{\mathrm {cr,d}}}} \end{array}} \right.$$

(11)

τ_cw is maximum bed shear stress due to waves and current calculated through the wave–current interaction, τ_cr,_e is the user-defined critical shear stress for erosion, and τ_cr,_d is the user-defined critical shear stress for deposition.

The Delft3D morphodynamic module also includes bed level update as well as the bank erosion. Bank erosion is a function of erosion flux in the adjacent dry cell. In the developed model, 50% of the erosion in the wet cell was redistributed to the neighboring dry cells. Wet cells were defined to have at least 10 cm of water depth.