104 pp.

Shimizu, Y., Yamaguchi, H., and Itakura, T., 1991, Three-dimensional computation
of flow and bed deformation, J. Hydr. Eng., ASCE, 116(9), 1090-1108.

Smith, J.D., and McLean, S.R., 1984, A model for flow in meandering streams, Water
Resources Res., 20(9), 1301-1315.

Tennekes, H., and Lumley, J.L., A First Course in Turbulence, 1972, MIT Press, Cambridge, MA, 300pp.

6. CALCULATION CONDITIONS
A number of boundary conditions, initial conditions, numeric parameters, and other
variables must be specified before a simulation can be run using the iRIC application.
Typically these are specified once the grid has been generated and elevation has been
specified for all locations. To access the dialog to specify calculation conditions select
Calculation ConditionsSetting from the Menu Bar.

In this section all calculation conditions required by the Fastmech solver will be
described in the order they appear in the Setting dialog box in the iRIC interface (Fig a.
showing Calculation Condition Setting dialog). The left window of the dialog displays the
primary parameter groups that can be set using the FaSTMECH solver. Values for
specific parameters are entered in the fields on the right side of the window. For each
parameter group and each field we will describe what is required and provide guidelines
for selecting appropriate values.

Figure A. Example of the Calculation Condition dialog.

6.1. Discharge
Figure A shows the fields associated with specifying the discharge entering the upstream
boundary of the simulated reach. This discharge is applied to the areas of the grid that
are wet based on the initial water-surface conditions specified in the Initial Conditions
tab.

Discharge Type
Select either Constant or Variable from the drop down menu. Constant applies the same
discharge for each model iteration. Variable allows you to specify discharge through
time. When variable is selected, the Variable Discharge Edit button becomes enabled
allowing you to create a time-series or import a file.

Discharge
Enter the river discharge in cubic meters per second, m 3/s.

Depth Weighting Coefficient

The discharge is distributed by a power of the depth according to this coefficient. The
upstream boundary condition at each node is calculated as a function of the weighted
discharge.

Velocity Angle
The velocity angle specifies a uniform angle for velocity at each node on the upstream
boundary relative to the stream-wise direction defined by the grid orientation. The
default value is 0 which directs velocity parallel to the stream-wise direction of the grid.
Ideally the grid is constructed such that no adjustment to the velocity angle is
necessary, but this may not be possible in all cases. To adjust the angle, use the
following:

0 < Velocity Angle <= 90 directs velocity towards river left

-90 <= Velocity Angle <0 directs velocity towards river right.

Variable Discharge
When Discharge Type is set to variable the edit button to the right of this field becomes
enabled and opens a dialog box that allows you to create a discharge time-series, or
import a time-series file (Figure B). Time is specified in seconds and discharge is m 3/s.
Import files should contain the same information and be saved with the .txt or .csv
extension. The information here would be stored in the file as

0,100
30,500
45,500
62,500
 Figure B. Variable Discharge dialog.

Use Velocity Weighting Function

This option allows you to specify the magnitude of velocity by distance along the
upstream boundary. This data is normalized and velocity distribution is stretched to fit
the wetted length of the upstream boundary in the model.

This feature can be especially helpful in short reaches or in situations where the
upstream boundary is not sufficiently far enough upstream for the flow solution to be
resolved in the primary area of interest.

When “No” is selected, the model estimates velocity based on the depth weighting
coefficient.

When “Yes” is select the edit button next to Variable Discharge becomes enabled
allowing you to specify a time-series velocity file.

Variable Velocity Distribution

The edit button to the right of this field opens a dialog box that allows you to generate
or import a normalized distance and velocity profile to distribute the velocity across the
upstream boundary.

Normalized Distance is specified from the water’s edge at the right bank to the left
bank (as viewed when looking downstream), in meters. The coordinate system is
right-handed.

Normalized Velocity is the vertically averaged velocity magnitude in m/s.

Velocity Angle: this feature is not currently implemented.

 Figure C. Velocity weighting dialog.

6.2. Stage
Stage is the water-surface elevation at the downstream boundary for a given
discharge. Figure D. shows the fields associated with specifying stage at the
downstream boundary. All stage data should be entered as elevation in meters,
based on the same datum as the topography dataset.

Figure D. Stage dialog.

Stage Type
Allows you to select three different stage conditions; constant, time-series, and rating-
curve from the pull down menu. Constant applies the same stage to the downstream
 boundary for the entire simulation. Time-series enables the Time-series button which
allows you to enter or import a file that defines how stage changes through time.
Rating-curve enables the Rating-curve button which opens a dialog that allows you to
enter or import a file that defines how stage and discharge are related.

Constant Stage
Enter the water-surface elevation at the downstream boundary in meters.

Stage Time-Series
The edit button to the right of this field opens a dialog box that allows you to enter a
discharge time-series, or import a time-series file for stage (Figure E). Time is specified
in seconds and stage is meters. Import files should contain the same information and
be saved with the .txt or .csv extension. The format of the file is time comma stage
similar to that of discharge in Section 6.1.

Figure E Time-series stage dialog.

Stage Rating-Curve
The edit button to the right of this field opens a dialog box that allows you to generate
stage-rating curve, or import a rating-curve file. Discharge is specified in m3/s and stage
is in meters. Import files should contain the same information and be saved with the .txt
or .csv extension.

*Note, if you use the extend grid feature, you will need to modify the stage to account
for the extended grid.

6.3. Roughness
A hydraulic roughness parameter combines many potential sources of roughness into
one value. Different sources of roughness include grain roughness, bed forms such as
dunes and ripples, bank alignment, and vegetation, etc that may be difficult to measure
in the field. Roughness is often determined through a calibration process that evaluates
 which roughness value produces the best fit with measured data such as water-surface
elevations and velocity fields.

Roughness can be specified based on either drag coefficients or z 0 parameters (Figure

F). Roughness can be either a constant throughout the model reach or vary spatially. To
use a spatially varying value you will need to create polygons with different roughness
values that are then interpolated to the grid using tools in the Pre-Processor window.

Roughness Type
Select either drag coefficient or znaught (z0) from the pull down menu.

Drag coefficient
Specify the drag coefficient. Typically the drag coefficient is determined by
calibrating the value to measured water surface elevations through the reach. In the
absence of measured water surface elevations you can estimate the drag coefficient
base on the mean depth through the reach and Mannings roughness coefficient (n)
based on the following equation, where h is the mean flow depth and g is gravity.
𝑛2 𝑔
𝐶𝑑 = 1
ℎ ⁄3
Z0
Z0 is defined as the height above the bed where flow velocity equals zero. It can be
related to bed grain size using Z0=C * grain size, where C is a coefficient typically set to
0.1. The Z0 roughness parameter can be a good choice when roughness is likely to be
spatially variable due to depth dependence.

The FastMech Solver converts Z0 to Cd from an estimate of depth based on the initial
water-surface elevation condition. This conversion can produce unrealistically low or high
drag coefficient estimates. Any values that are lower than a user entered minimum
value is reset to the miniumum value and any values higher than the user entered
maximum value is reset to the maximum drag coefficient.
 Figure F Roughness dialog.

6.4. Lateral Eddy Viscosity

The lateral eddy viscosity is a correction to the eddy viscosity used in the vertically
averaged equations to treat lateral separation eddies. It can be estimated using:

LEV = (0.01 – 0.001)x average depth x average velocity

In some cases the model may be unstable for reasonable estimates of LEV. Typically this
occurs when the grid centerline is perpendicular to the channel such that the initial
velocities are directed towards one or the other bank. In that case it may be necessary
to start the solution at a higher LEV value and then decreases LEV to a reasonable
estimate once the solution is partly converged. See Figure G for the dialog that controls
LEV.
 Figure G. Lateral Eddy Viscosity in the Calculation Condition dialog.

Lateral Eddy Viscosity Type

Select Constant or Variable from the pull down menu. Constant uses the same LEV for
each iteration. The variable option linearly reduces the LEV value between a specified
beginning and ending value over a specified iteration interval. The fields to control the
variable LEV features become enabled when the Variable option is selected.

Starting Iteration
Enter the iteration at which the LEV will begin to decrease.

Ending Iteration
Enter the iteration at which the LEV will stop decreasing

Starting LEV
Enter the beginning LEV value

Ending LEV
Enter the ending LEV value

Constant LEV
Enter the LEV value
6.5. Grid Extension
It is always best to choose a downstream boundary where the flow is uniform. In practice
this is often not the case. The model does not allow for upstream flow at the downstream
boundary condition, which can occur when there is a recirculating eddy at the boundary.
There are two ways to handle issues at the downstream boundary by either 1) extending
the boundary with uniform channel geometry to try to prevent recirculating flow or 2)
forcing downstream flow at the boundary.

Grid extension works by duplicating the topography in the last row of nodes by a specified
number of streamwise nodes while reducing the elevation of each node based on a grid
extension slope. For example, if the grid cell size in the stream-wise direction is 10 m, and
you select 10 grid extension nodes, the total grid extension will be 100 m. It is important to
note that the downstream boundary condition in the input file must account for this change
in the model grid. In other words, you must decrease the water-surface elevation at the
downstream boundary so that it corresponds with the extended topography.

Forcing the downstream velocity boundary to prevent recirculating flow often resolves
issues related to recirculating flow. However it will affect the solution near the downstream
boundary.

Figure H. Grid Extension in the Calculation Condition dialog.

Grid Extension Nodes

Enter the number of nodes to extend the grid in the streamwise direction

Extension Slope
Enter the slope to decrement the elevation of the extended grid.
View Extension
Select yes or no from the pull down menu to display the extended topography in the Post-
processing window.

Downstream Boundary Velocity

Select No or Force no-recirculation from the pull down menu.

6.6. Initial Conditions

The model requires an initial estimate or guess of the water-surface elevation for each node
in the grid. This provides both a starting point in the numerical model and determines which
nodes are wet and dry in the calculation. Better initial water-surface elevations lead to
faster model convergence. There are a number of ways to specify the initial water-surface
elevation; however the 1D step-backwater solution is recommended in most situations.

This input dialog also allows you to select a hot-start file to initialize the model. A hot-start
file is simply the output from a previous model solution. This feature allows you to start the
model using a converged solution, which can be particularly useful in simulations that
simulate bed evolution. If you want to use the hot start option there are two requirments.
1. The solution to be used to hot start the current solution must have been saved
using the File->Save Project command in iRIC.
2. The 2D Solution Output must have “Initial Velocity for Hot Start” selected.

The initially wet or active nodes in FaSTMECH are determined by the water-surface
elevation initial guess. The initial water surface elevation is specified along the grid
centerline. Wet and dry nodes are determined by extending the 1-dimensional water surface
elevation along the grid centerline to each row or cross-section of nodes. If the water
surface elevation is above the bed elevation the node is wet, otherwise it is dry. By default
nodes will dry during the solution iteration process if the water surface elevation solution
becomes less than the bed elevation.
 Figure I. Initial Conditions in the Calculation Condition dialog.

Initial Water-Surface Elevation

Select Upstream Stage, Uniform Slope, 1D Step-backwater, or Hot Start from the pull down
menu. The necessary fields are then enabled below.

Upstream Stage
This option is enabled when Upstream Stage is selected from the Initial Water-Surface
Elevation pull down menu. Enter the stage at the upstream boundary. The water-surface
elevation is then linearly interpolated between the boundaries for each node in the grid.

Uniform Slope
This option is enabled when Uniform Slope is selected from the Initial Water-Surface
Elevation pull down menu. Enter a slope. The water-surface elevation is then projected
upstream starting at the downstream boundary along the grid centerline.

1D Discharge
This option is enabled when 1D Step-backwater is selected from the Initial Water-Surface
Elevation pull down menu. Enter the discharge you are interested in modeling.

1D Stage
This option is enabled when 1D Step-backwater is selected from the Initial Water-Surface
Elevation pull down menu. Enter the stage at the downstream boundary.
1D Drag Coefficient
This option is enabled when 1D Step-backwater is selected from the Initial Water-Surface
Elevation pull down menu. Select a roughness value. You may wish to select a higher
roughness value to force the initial water-surface elevation to be somewhat higher.

Hot Start File

Select the solution file you wish to use from the dialog window. The solution used for the
hot start must have the requirements as specified in the introduction to this section.

Hot Start Time-step

Select the iteration or time-step from the solution file to initialize the model.

6.7. Wetting and Drying

The wetting and drying dialog controls how dry nodes are treated in the calculation, and
provides settings to re-evaluate whether or not nodes are wet or dry during the simulation.
The wet or dry state of a node can be visualized in the IBC solution scalar. A value of 0 is
dry and a value of -1 is wet.

Figure J. Wetting and Drying in the Calculation Condition dialog.

Drying Type
Select from Set Node Off and Set Node to Drying Depth in the pull down menu. The result of
this option depends on the Drying Depth set below. Set Node Off turns the node off (or sets
it to a dry state) if the depth drops below the minimum depth set below. Set Node to Drying
Depth keeps the node in a wet state if it drops below the minimum depth but keeps the
depth at the specified value. This can be useful when you have instabilities in the solution
in very shallow locations. In general this latter option is rarely used..
Drying Depth
Set the minimum depth at which a node becomes dry.

Re-Wetting Option
Select from Off and On in the pull down menu. Off means that the wet status of a node is
set by the initial water-surface condition and is not re-evaluated during the simulation. On
allows the user to prescribe when the wet and dry status of a node is re-evaluated during
the simulation. Using this option can make a stable solution difficult to acquire. If the
channel has large areas of relatively shallow regions then it may be necessary to use this
option to get a solution that predicts well these shallow area’s.

Re-wetting Minimum Depth

Enter the minimum depth at which a dry node becomes wet. This value should be larger
than the Drying Depth set above. Generally good results are found by setting the drying
depth to 0, and the re-wetting depth to 0.015.

Iteration to check Re-wetting

Enter the iteration to check if the re-wetting condition has been met.

Iteration to stop check Re-wetting

Entering the iteration to stop checking if the re-wetting condition has been met.

6.8. Solution Parameters

The solution parameters dialog controls how many iterations are run using a constant
discharge simulation and provides controls for time-stepping features for quasi steady
discharge simulations.
 Figure K. Solution Parameters in the Calculation Condition dialog.

Solution Type
Select Constant Discharge or Quasi-steady Discharge from the pull down menu. Constant
Discharge is used when solving for a single discharge value. Quasi-Steady Discharge is
when simulating a hydrograph or morphologic change over a period of time. The discharge
may be constant or time-dependent.

Iterations
Enter the number of iterations used to generate a solution. The number required to reach a
converged solution varies. This field is set for both constant discharge and quasi-steady
discharge options. In the quasi-steady option it specifies the number of iteration at the
start of the calculation.

Output Iteration Solution

Select No or Yes from the pull down menu. No will output solution results for the final
iteration. Yes will output solution results for each iteration entered in the Iteration Plot
Interval field. Generally this option is not used.

Iteration Plot Interval

Enter the iteration interval to output solution results. For example, if your iteration plot
interval is set to 50 and the iteration is set to 200, you will get output data at iteration 1,
50, 100, 150, and 200? enter 50 and have 200 This fields is enabled when the Output
Iteration Solution is set to Yes.
Start Time
Enter the time in seconds to start the simulation. The field is enabled when Quasi-steady
Discharge is selected from the Solution Type pull down menu.

Time Step
Enter the time step in seconds to evaluate the time-stepping solution.

Num. Iterations Between Time-steps

This field specifies the number of iterations the solver completes before incrementing to the
next time-step.

Wetting and Drying iteration Multiplier

If the solution is solved for a variable discharge hydrograph then sometimes it is advantageous to
increase the number of iteration between each time step when the solution either increases or
decreases in the number of dry nodes. In the case where the number of wet or dry nodes changes
then the num. iteration between time-steps increases by a multiple of the number specified in this
field.

Plot Interval
Select the interval, in seconds, to record solution results for Quasi-steady Discharge
simulations.

Use Straight channel Wall Boundary

This is in development and should be specified as no.

Stop Solution for debugging

Select Yes or No from the pull down menu. Yes, allows you to ouput solution results at
specific time-step and iteration with in a time-step to better understand what causes a
model to crash before completing the solution.

Time-step to stop
Enter the time-step to stop the simulation and save a solution.

Iteration to stop
Enter the iteration within the time-step to stop the simulation and save a solution.

6.9. Relaxation Coefficients

Fastmech uses relaxation to iterate on a solution. Relaxation is essentially a way to speed
up or slow down changes in a given parameter between one iteration and the next. For
example, the ERelax controls the relaxation of the water-surface elevation. A value of 0.4,
weights the current water-surface elevation solution at 0.4, and the previous solution at 0.6
to obtain the Water-surface used in the next iteration. In other words the current solution is
made up of 60 percent of the previous solution and 40 percent of the current solution.
It is often necessary to hunt for the best parameters. The default values work in most
cases but if the channel is complex, then the relaxation coefficient may need to be changed.
Generally it is the ratio between ERelax and URelax which is important. Generally ARelax is
set to the same value as URelax. In steep channels a values of 0.3, 0.6, 0.6 for each of the
three values respectively are found to work well.

Figure L. Solution Relaxation Coefficients in the Calculation Condition dialog.

ERelax
Enter the relaxation coefficient for the water-surface elevation

URelax
Enter the relaxation coefficient for the velocity.

ARelax
Adjusts the global slope of the water surface at each iteration.

6.10. 2D Solution Output

This dialog allows you to calculate several additional parameters as part of the solution.
These include:

Parameter Description
Velocity SN Outputs velocity vectors based on SN,
or stream-wise and stream-normal
coordinates.
Shear Stress XY Outputs shear stress based on XY
coordinates
 Shear Stress XN Outputs shear stress based on stream-
wise and stream normal coordinates.
Shear Stress Divergence Outputs shear stress divergence
Drag Coefficient Outputs the drag coefficient used by
the similar.
Initial Velocity for Hotstart Saves the initial velocity to use as a
Hot Start

Figure M. 2D Solution Output in the Calculation Condition dialog.

6.11. Quasi-3D Solution

The vertical structure of the flow can be developed using the Quasi-3D option.

Calculate Quasi-3D Solution

Select Yes or No from the pull down menu. Yes enables the Quasi-3D solution.

Number of Vertical Nodes

Enter the number of vertical nodes to include in the calculation

Use Streamline-curvature
Select Yes or No from the pull down menu. This option uses the streamline curvature in
addition to the grid curvature to develop the secondary flow.

Minimum Streamline Curvature

This value should not be set less than approximately 5 times the mean depth.
6.12. 3D Solution Output
Select whether or not to include aspects of 3D flow in the solution calculation. The 3D
Scalar and Vector option can make file sizes very large.

Figure N

Parameter Description
Helix Strength The difference in angle between the
surface and near bed velocities
3D Scalar/Vector Outputs the 3D Scalar and Vector
quantities

6.13. Sediment Transport

The sediment Transport dialog controls many functions of sediment transport and bed evolution
simulations.
 Figure O

Calculate Sediment Transport

Choose yes or no to calculate sediment transport.

Transport Equation
 Yalin (bedload)
 Engelund-Hansen (total load)
 Wilcock-Kenworthy (two-fraction model, uses fine fraction only)

Grainsize
Mean grainsize to use in sediment transport calculations.

Dune Height
If dune height and dune wavelength are set boundary shear stress will be reduced in the sediment
transport calculations. Set the average dune height in meters

Dune Wavelength
Set the average dune wavelength in meters.
Number of smoothing passes
Smooth the sediment transport rate using a box-car smoothing algorithm.

Smoothing Weight
Specify the weight of the adjacent nodes in the smoothing algorithm.

Automatic Time-Stepping
When this value is set to yes the time step for each new calculation is determined from the
previous time step based on the Max. Fraction of Depth per Time-step set below. Based on the
current sediment flux and depths, the time step is determined such that no where in the solution
will the change in elevation of the bed result in a change in depth greater than the value specified
below.

Max. Fraction of Depth per Time-step

The time step is constrained by this fraction. A value of 0.1 means that the time-step calculated
from the previous solution will be set such that the change in elevation will not exceed 10
percent of the depth anywhere in the solution domain.

Downstream Node Number to Begin Transport.

The model can become unstable with the transport is started near the upstream boundary. This
sets the node number in the stream-wise direction to begin the transport calculation.

Fraction of Equilibrium Load

Over or under supply he load at the upstream boundary by specifying this value to be greater
than or less than 1 respectively.

Gravitational Correction
Specify the gravitational correction to use.