RiverFlow2D

Two-Dimensional
River Dynamics Model
Reference Manual
October 2015

Hydronia LLC

c
RiverFlow2D
model and documentation produced by Hydronia, LLC, Pembroke Pines, FL. USA.
Information in this document is subject to change without notice and does not represent a commitment on part of
Hydronia, LLC. The software described in this document is furnished under a license agreement.
RiverFlow2D, OilFlow2D, RiverFlow2D, and RiverFlow2D GPU are copyrighted by Hydronia, LLC. 2011-2015.
SMS is trademark of Aquaveo, LLC.
All other products or service names mentioned herein are trademarks of their respective owners.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in
any form or by any means electronic, mechanical, photocopying, recording or otherwise, without the prior written
permission of Hydronia, LLC.
Last document modification date: October, 2015.
Technical Support: [email protected]
Web site: www.hydronia.com

Contents

ii

List of Figures

iii

List of Tables

iv

Introduction
RiverFlow2D is a combined hydrologic and hydraulic, mobile bed and pollutant transport finitevolume model for rivers, estuaries and floodplains. It can route floods in rivers and simulate
inundation over floodplains and complex terrain at high resolution and with remarkable speed,
stability, and accuracy. The use of adaptive triangular-cell meshes enables the flow field to be
resolved around key features in any riverine environments.
RiverFlow2D user interface is based upon Surface-Water Modeling Solution (SMS) developed
by Aquaveo. This GIS-integrated software system provides interactive functions to generate and
refine the flexible mesh used by RiverFlow2D. SMS uses familiar GIS objects such as points,
arcs (polylines), and polygons to construct a high level representation of the model, facilitating
assigning boundary conditions and roughness values, and allowing the user to efficiently manage
the entire modeling process. SMS offers a comprehensive set of visualization tools including 3D
rendering, animations, and exporting graphs to Google Earth.
RiverFlow2D computation engine uses an accurate, and stable finite-volume solution method
that eliminates the boundary and hot start difficulties of some two-dimensional flexible mesh
models. The model can integrate hydraulic structures such as culverts, weirs, bridges, gates and
internal rating tables. The hydrologic capabilities include spatially distributed rainfall, evaporation
and infiltration. This version also accounts for distributed wind stress on the water surface.
This reference manual provides instructions to install the RiverFlow2D model, and explains the
fundamentals of the model and its components, as well as the numerical methods used to solve the
governing equations. It also presents detailed description of the input data files, and output files.
A separate document includes several tutorials that help you getting started using RiverFlow2D
and learning to apply several model components such as bridges, culverts, rainfall and infiltration,
weirs, etc.

1.1

Summary of RiverFlow2D Features and Capabilities

Figure 1.1 RiverFlow2D triangular-cell mesh.

1.1 Summary of RiverFlow2D Features and

Capabilities
1.1.1 Mesh Generator
Automatic generation of flexible triangular-cell mesh
Unlimited mesh refinement
Mesh quality report
Spatially varied interpolation methods including
Triangular (TIN)
Inverse Distance Weighting
Sampling, and many more

1.1.2 Numerical Engine

Spatial discretization using triangular elements
High performance Finite-Volume engine
Double-precision computations for higher accuracy
Fully parallelized with OpenMP for faster execution in Multiple-Core computers
GPU version for >150X faster simulations

1.1.3 Hydraulic Components

Culverts using the US Federal Highway Administration (FHWA) formulation
Bridge hydraulics in 2D including pressure flow and overtopping
Bridge pier drag forces
Weirs with variable crest elevations
Gates
Internal hydraulic structures
Sources and sinks
Spatially distributed rainfall and evaporation
Spatially distributed infiltration
Spatially distributed wind stress
Depth dependent Mannings n

1.1.4 Input Data Formats

Metric or English units
ASCII X, Y, Z
ESRI ASCII grid files
USGS DEM
ESRI shapefiles
Autodesk DXF
TIFF, GIF, JPG, etc. rasters

1.1.5 Initial Conditions

Dry-bed
User-defined water elevations

1.1.6 Boundary Conditions

Water discharge hydrograph
Water discharge and water elevation vs time
Water elevation vs time
Uniform flow
Rating tables

Introduction

1.1

Summary of RiverFlow2D Features and Capabilities

Free outflow
Inflow suspended sediment concentrations (ST Module)
Inflow pollutant concentrations (ST Module)

1.1.7 Output Options

Results at cross sections and profiles
Observation points: time series at user selected locations
Dynamic plots while the model runs
Velocity field, depth and water surface elevations
Bed shear stress
Erosion and deposition depths
Maximum velocity, water elevations, erosion or deposition depths, etc.
Bed elevation changes
Froude number
Sediment fluxes
Time to 0.3 m (1 ft), time to 1 m (3 ft) and time to peak depth
Inundation time during which depth is greater than 0.1 m or 4 in
Polygon, line and point ESRI shapefiles
Autodesk DXF mesh and velocity fields
GIS post processing plots including shapefiles and raster images
Paraview VTK

1.1.8 Sediment Transport Module (Optional)

Separate calculation of suspended and bed load sediment transport
Multiple size fractions
Bed-changes (erosion-deposition)
10 sediment-transport formulas
Sediment transport over rigid bed
Maximum erosion depth areas

Introduction

1.1.9 Pollutant Transport Module (Optional)

Advection-Dispersion-Reaction
Reaction rates between pollutants/solutes
Simultaneous computation of multiple solutes

1.1.10 Mud and Debris Flow Module (Optional)

Non-Newtoninan fluids
Seven rheological formulations
Granular flow

1.1.11 OilFlow2D: Overland Oil Spills

Model to simulate spills of crude oils and viscous fluids over complex terrain

Figure 1.2 SMS main window showing a velocity field computed with RiverFlow2D.

1.1

Summary of RiverFlow2D Features and Capabilities

Installing RiverFlow2D SMS Edition

RiverFlow2D requires SMS version 11.2 or later. This section will assist you to setup the RiverFlow2D
model to enable using the it from SMS.

2.1 Hardware Requirements

RiverFlow2D is supported on 64-bit computers running MS-Windows 7, 8.1, 10 platforms. It is
recommended to have a minimum of 4 GB of RAM and at least 10 GB of free hard disk space.
RiverFlow2D is capable of running in modern Intel single processor computers. If multiple-core
processors (Duo, Quad, etc.) are available, the model can execute in parallel processor mode,
thereby running much faster than in single processor computers. In addition, if you have a GPU
license, the model can take advantage of Graphic Processing Unit (GPU) cards to run up to 150
times faster than in single-processor computers.
It is recommended to uninstall all previous RiverFlow2D SMS setups before installing this version.

2.2 Installing RiverFlow2D and SMS

To install RiverFlow2D, please follow these steps.
1. When installing from a DVD, please insert the RiverFlow2D installation disk into your DVD
drive, double click on the RiverFlow2D SMS setup file and follow the on-screen instructions.
If you received a setup file link, please download the file and double-click on it.
2. If you are installing on a PC running Windows 7 or later, you must be logged on the PC as
an administrator before you begin the installation.
7

2.3

Activating SMS

2.3 Activating SMS

Depending on the SMS licensing options you have, you may need to use different procedures to activate the software. Please go to Aquaveo website at http://www.aquaveo.com/license-instructions
to find specific instructions to your particular SMS license.

2.4 Linking RiverFlow2D to SMS

In order to use RiverFlow2D from SMS GUI, it is necessary to link RiverFlow2D to SMS as follows:
1. Run the SMS program
2. On the Edit menu, click Preferences...
3. Click the File Locations Tab
4. Scroll down in the Model Executables frame till you see the Generic model.
5. Click the Browse button and find the folder C:\ProgramFiles(x86)\SMSRiverFlow2DPlusv4
and select the model executable file RiverFlow2Dm4.exe.
6. Click OK to exit.

Often you may find necessary to run SMS in administration mode. To do that, use the Windows
File Explorer and go to the C:\ProgramFiles\ folder. Right-click on SMS112.exe and select Run
this program as an administration as indicated in Figure ??

Figure 2.1 SMS executable Properties dialog with the Run this program as an administration
option set.

Installing RiverFlow2D SMS Edition

2.5 RiverFlow2D Documentation

Find the RiverFlow2D documentation including this manual in the following folder:
\RiverFlow2D_SMS\Documentation usually installed under ...\Documents or \MyDocuments
(see Figure ??).
Also under \RiverFlow2D_SMS, you will find example projects, videos, presentations, and other
useful resources.

Figure 2.2 Location of the RiverFlow2D Documentation folder.

2.6 RiverFlow2D Technical Support

If you have any questions or need assistance, please email our support team at: [email protected].
Please make sure you visit our web site www.hydronia.com regularly to find out about new products and updates.

2.6

RiverFlow2D Technical Support

10

Overview of SMS Capabilities

SMS is a general purpose Graphical User Interface (GUI) for free-surface flow models. SMS
represents the hydraulic or hydrologic problem based on a Conceptual Model that consists of
a simplified representation of the situation being simulated. The Conceptual Model includes the
geometric attributes of the area such as the extents, the boundary conditions acting on the domain
(such as inflow discharges or water levels) and the physical characteristics (such as bed roughness).
The Conceptual Model does not include numerical details such as mesh elements, etc. It is helpful
to construct the model using feature objects in the Map module and having an aerial image in the
background as a reference. After building the Conceptual Model, a mesh can be generated and
the data will be automatically transferred to the mesh cells and nodes. Figure ?? shows the main
parts of the SMS window.
A variety of tools in SMS allow the user to manage the data, set up the project and post-process
output results. The primary tools and windows can be described as follows:
Main Graphic Window: the data management is done in this window.
Toolbox:
1. Modules: these icons control what menu commands and tools are available at any
given time while operating in SMS. Each module corresponds to a specific type of data:
triangular-cell mesh, scattered data, etc. The commands that operate each module are
available once the module is active.
2. Static Tools: set of tools used to manipulate the display. They do not change for
different modules.
3. Dynamic Tools: tools used to create and edit entities specific to the module. They
change according to the selected module and the active model.
4. Macros: shortcuts for menu commands. They can be set in the Preferences dialog
accessible on the Edit menu and clicking Preferences...
Project Explorer or Tree Window: to view all data that makes up a part of a project. It is
used to switch modules, select coverage to work with, select a base set to be active, and set
display settings of the various entities in the active coverage. By right clicking on various
entities in the project explorer, the user may also transform, copy, or manipulate the entity.
11

3.1

Modules

12

Figure 3.1 SMS main window.

Time Steps Window: it is used to select results for specific time steps when a transient
dataset has been loaded into the project.
Edit Window: It is used to show and/or change the coordinates of selected entities.
Menu Bar: Allow managing data depending on the active module.
Status Bar: it shows the mouse coordinates when the model is in plan view and the information for selected entities.

3.1 Modules
Data is divided into modules based upon the data type they can accommodate.

There is always one active module.

The menus and toolbars change depending on the active module.
The current module may be selected in the module bar or by selecting an object in the
project explorer.

13

Overview of SMS Capabilities

3.1.1 Mesh Module

The mesh module is used to create, edit, and visualize mesh data. RiverFlow2D meshes are also
referred to as unstructured grids and are defined by nodes and elements also called cells.

Figure 3.2 Mesh module.

3.1.2 Scatter Module

The Scatter Module is used to create, edit, and visualize Triangulated Irregular Networks (TINs).
DEMs can be read in and converted to TINs.
Filter scatter sets to eliminate redundant data.
Datasets can be interpolated to other modules, e.g. a mesh.

3.1

Modules

14

Figure 3.3 Scatter module.

3.1.3 GIS Module

The GIS Module is used to import and visualize GIS data such as shapefiles.

Figure 3.4 GIS module.

Supports ESRI and MapInfo formats.

Uses ESRI files to use ArcGIS visualization options.
GIS data can be converted to feature data to the Map module.

15

Overview of SMS Capabilities

3.1.4 Map Module/Conceptual Model

The Map Module provides the GIS containers that allow creating the conceptual model

Figure 3.5 Map module.

Create and edit GIS data.

Used to create conceptual models as well as data for other purposes.
Conceptual model is a geometry, mesh independent representation of the numeric model
domain and its boundary conditions.
Features can be converted to model geometry and boundary conditions.
Conceptual model makes it easier to create, and edit model data.

3.1.5 Visualization Tools

Visualize scalar and vector datasets.
Linear, color filled or both.
Transparency.
Full control of ranges and colors.
Precision control for labels and legends.

3.1.6 Vectors
Visualize vector datasets as arrows.
Constant size or vary by magnitude.

3.1

Modules

16

Figure 3.6 Three-dimensional view of mesh terrain.

Show just a range of magnitudes.

Color by magnitude.

3.1.7 Plots
2D plots to visualize results and compare to measured values.
Profile plots view scalar data along polylines.
Time-series plots view scalar, vector, or flux (flowrate) data at a point or across a polyline.
Several kinds of plots can be used to compare model results with measured data.

3.1.8 Animations
Several types of AVI animations (film loops) can be generated.
Transient data animation shows model changes through time (contours, vectors, etc.).
Flow trace uses vector data to generate flow paths through the geometry.
Drogue plots use user specified starting locations and show how the particles would flow
through a vector field.
Multiple view animations show the data while transitioning between different views.

3.1.9 Export Options

Graphics window can be copied to the clipboard.
Current view can be exported in KML format for visualization in Google Earth.

17

Overview of SMS Capabilities

3.1.10 CAD Support

AutoCAD DXF and DWG files can be read into SMS.
CAD data is displayed in 3D.
CAD data can be converted to map or scatter data.

3.1.11 Import Wizard

Read columnar ASCII data into SMS.
Columns can be fixed width or delimited by specific characters.
Data can be read as mesh, scatter, or map data.

3.1.12 Image Support

Multiple images can be read/viewed at the same time.
Many image formats are supported including JPG, TIFF, PNG, MrSID, ECW, and more.
Images can be geo-referenced to view images along with other data.
Images can be draped over mesh or scatter data.
Image pyramids can be created which improves viewing at various zoom levels.

3.1

Modules

18

Overview of RiverFlow2D SMS edition

RiverFlow2D SMS Generic Model includes a set of standard options organized in groups that
facilitate the user interaction with the model. The standard window groups include: Check Mesh,
Define Model, Global Parameters, Assign BC, Material Properties and Run RiverFlow2D shown
in Table ??.
Table 4.1 SMS Menu for RiverFlow2D.
RiverFlow2D Menu
Check
Mesh
Define
Model
GLOBAL
PARAMETERS
Control
Data

Tab Name

Type

Information

Identifies potential issues in the

construction of the mesh or data.

Content

Template

Group locked. It contains the

structure of a RiverFlow2D Template.

TAB

WINDOW TO ENTER DATA

Data

Model run control parameters

such as simulation time, etc.

Graphic
Output
Options

Data

Parameters that can be set for the

graphical output plots that are
generated during runtime.

Sediment
Transport

Data

Parameters for the

Transport Model.

Data

Parameters for the Mud and Debris Flow Model.

Data

Parameters for the

Transport Model.

Data

Rainfall and Evaporation Data.

Mud and
Debris
Flow
Pollutant
Transport
Rainfall /
Evaporation

Continued on next page

19

Sediment

Pollutant

20
Table 4.1 continued from previous page
RiverFlow2D Menu
Tab Name
Type
Content
Component
Data
Turn on and off the different comControls
ponents as piers, weirs, culverts,
sources and sinks, bridges, gates
and internal rating tables.
ASSIGN
BOUNDARY
TAB
ASSIGN BCs
CONDITIONS
Exterior
Data
Specify boundary conditions on
exterior boundaries.
Interior
Data
Specify internal conditions on a
polyline.
Bridges
Data
Specify data for the Bridges component.
Gates
Data
Specify data for the Gates component.
Weir
Data
Specify data for the Weirs component.
MATERIAL
PROPERTAB
ASSIGN ROUGHNESS PATIES
RAMETERS
Materials
Data
Enter Mannings n that will be
assigned to polygons in the conceptual model.
Run
RiverFlow2D To execute RiverFlow2D
MESH
MODULE
CREATION
OF
NODES/ELEMENTS
2D unMesh
structured
Define the computational domain
grids
of the numerical model that consists of nodes group together to
form elements.
Terrain
Elevation
Bed elevations data file.
Data
Output
Nodal
Model output (.hdf5) will be read
Results
Values
as mesh datasets. Scalar datasets
are identified by the 123 icon, and
the vector dataset have the vector
icon.
SCATTER
MODULE
INTERPOLATE
SPATIAL
DATA VALUES
Nodal
Scatter
Interpolate values using one
values
of the available interpolation
schemes including linear, natural
neighbor and inverse distance
weighted. The module can be
also used to view and edit survey
data.
MAP
MODULE
USE
OF
GIS
OBJECTS
TO BUILD CONCEPTUAL
MODEL
Conceptual
GIS
GIS objects such as points, nodes,
Model
Objects
arcs and polygons are grouped
into layers or coverages and use to
create the hydraulic model. The
coverage type determines the attributes available for the objects
in the coverage.
Continued on next page

21

Overview of RiverFlow2D SMS edition

Table 4.1 continued from previous page

RiverFlow2D Menu
Tab Name
Type
Content
GIS
MODULE
TO LINK WITH GIS DATA
GIS
Data
Allow users to handle large files
more efficiently importing large
files and then select and convert
to feature objects only the portions that are needed.

The commands in SMS are divided based on the types of data they operate on. The dynamic
toolbar and available menu commands instantaneously change when the user switches from one
module to another. Only one module is active at any given time. The data associated with a
module (e.g. a 2D mesh) is preserved when the user switches to a different module. The different
SMS Modules used to set up a project in SMS are described below:
Table 4.2 SMS Modules on RiverFlow2D SMS edition.
SMS Module
Name
Check
Information
Mesh
MESH
Mesh

MODULE
2D unstructured
grids

Output
Results

Terrain
Data
Nodal
Values

SCATTER

MODULE

Scatter

Nodal
values

MAP

MODULE

Conceptual
Model

GIS
Objects

GIS
GIS

MODULE
Data

Elevation

Type

Content

Identifies potential issues in the construction of the mesh or data.

CREATION OF NODES/ELEMENTS
Define the computational domain of the
numerical model that consists of nodes
group together to form elements.
Bed elevations data file.
Model output (.hdf5) will be read as
mesh datasets. Scalar datasets are
identified by the 123 icon, and the vector dataset have the vector icon.
INTERPOLATE SPATIAL DATA
VALUES
Interpolate values using one of the
available interpolation schemes including linear, natural neighbor and inverse
distance weighted. The module can be
also used to view and edit survey data.
USE OF GIS OBJECTS TO BUILD
CONCEPTUAL MODEL
GIS objects such as points, nodes, arcs
and polygons are grouped into layers
or coverages and use to create the hydraulic model. The coverage type determines the attributes available for the
objects in the coverage.
TO LINK WITH GIS DATA
Allow users to handle large files more
efficiently importing large files and then
select and convert to feature objects
only the portions that are needed.

22

RiverFlow2D Model Equations

5.1 Hydrodynamic Model
One-dimensional hydraulic models are not adequate to simulate flooding when flows are unconfined
or velocities change direction during the course of the hydrograph. The cost of non-simplified threedimensional numerical models can be avoided using depth averaged two dimensional (2D) shallow
water equations [Toro 2001].
When dealing with the shallow water equations, realistic applications always include source
terms describing bed level variation and bed friction that, if not properly discretized, can lead
to numerical instabilities. In the last decade, the main effort has been put on keeping a discrete
balance between flux and source terms in cases of still water, leading to the notion of well-balanced
schemes or C property [V
azquez-Cendon (1999), Toro (2001), Rogers et al. (2003), Murillo et al.
(2007)]. Recently, in order to include properly the effect of source terms in the weak solution,
augmented approximate Riemann solvers have been presented [Rosatti et al. (2003), Murillo and
Garca-Navarro (2010)]. In this way, accurate solutions can be computed avoiding the need of
imposing case dependent tuning parameters which are used frequently to avoid negative values of
water depth and other numerical instabilities that appear when including source terms.
This section presents the system of equations, the formulation of the boundary conditions, and
the finite-volume scheme used in RiverFlow2D.

5.2 Assumptions of the Hydrodynamic Model

1. RiverFlow2D uses the Shallow Water Equations resulting from the vertical integration of
the Navier-Stokes equation. Therefore, the model does not calculate vertical accelerations,
vertical velocities and consequently cannot resolve secondary flows.
2. The bed shear stress is assumed to follow the depth-average velocity directions.
3. The model does not include dispersion nor turbulence terms. Turbulence dissipation and energy loses are accounted for only through the Mannings n term in the momentum equations.
23

5.3

Unsteady Flow Model

24

5.3 Unsteady Flow Model

Shallow water flows can be described mathematically by depth averaged mass and momentum
conservation equations with all the associated assumptions (?). That system of partial differential
equations will be formulated here in a conservative form as follows:
U F(U) G(U)
+
+
= S(U, x, y)
t
x
y

(5.1)

where U = (h, qx , qy ) is the vector of conserved variables with h representing the water depth,
qx = uh and qy = vh the unit discharges, with (u, v) the depth averaged components of the velocity
vector u along the (x, y) coordinates respectively. The flux vectors are given by:
F=

qy2
1
qx qy
+ gh2 ,
qx ,
h
2
h

!T
,

G=

1
qx qy qy2
,
+ gh2
qy ,
h
h
2

!T
(5.2)

where g is the acceleration of the gravity. The terms 21 gh2 in the fluxes have been obtained after
assuming a hydrostatic pressure distribution in every water column, as usually accepted in shallow
water models. The source term vector incorporates the effect of pressure force over the bed and
the tangential forces generated by the bed stress
S = (0, gh(S0x Sf x ), gh(S0y Sf y ))

(5.3)

where the bed slopes of the bottom level zb are

S0x =

zb
,
x

S0y =

zb
y

and the bed stress contribution is modeled using the Manning friction law so that:

n2 u u2 + v 2
n2 v u2 + v 2
Sf x =
,
Sf y =
h4/3
h4/3
with n the roughness coefficient.

(5.4)

(5.5)

5.4 Finite-Volume Numerical Solution

To introduce the finite-volume scheme, (??) is integrated in a volume or grid cell using Gauss
theorem:
Z
I
Z

Ud +
Endl =
Sd
(5.6)
t

where E = (F, G) and n = (nx , ny ) is the outward unit normal vector to the volume . In order
to obtain a numerical solution of system (??) the domain is divided into computational cells, i ,
using a fixed mesh. Assuming a piecewise representation of the conserved variables (Figure ??)
and an upwind and unified formulation of fluxes and source terms (?)

Z
Ud +
i

NE
X

k lk = 0
(En S)

k=1

(5.7)

5.4

Finite-Volume Numerical Solution

Figure 5.1 Piecewise uniform representation of the flow variables.

Figure 5.2 Cell parameters.

Figure 5.3 Cell parameters.

25

5.4

Finite-Volume Numerical Solution

26

en,k (?) of the

The approximate solution can be defined using an approximate Jacobian matrix J
1
2
3
e = (e
e 1 , built using
non-linear normal flux En and two approximate matrices P
e ,e
e ,e
e ), and P
en,k diagonal
the eigenvectors of the Jacobian, that make J
e 1 J
en,k P
ek =
ek
P
k

(5.8)

em in the main diagonal

e k is a diagonal matrix with eigenvalues
with
k

e1 0

0
ek =
e2 0

0
e3
0
0
k

(5.9)

Both the difference in vector U across the grid edge and the the source term are projected onto
the matrix eigenvectors basis
e k Ak
Uk = P
1

e kB
k=P
(S)

(5.10)

, 3 )Tk

, 3 )Tk

where Ak = ( ,
contains the set of wave strengths and B = ( ,
contains the
source strengths. Details are given in (?). The complete linearization of all terms in combination
k as
with the upwind technique allows to define the numerical flux function (En S)
3 
X

k = Ei nk +
(En S)

m
e e


e = 1 (
e ||)
e and m = 1
with
k
2

m

(5.11)

m=1

that when inserted in (??) gives an explicit first order

Godunov method (?)

Un+1
= Uni
i

NE
X

"

3 
X

Ei nk +

m
e e

e
k

m=1

k=1

lk
t
Ai

(5.12)

As the quantity Ei is uniform per cell i and the following geometrical property is given at any
cell
NE
X

nk lk = 0

(5.13)

k=1

(??) can be rewritten as

Un+1
i

Uni

" 3
NE X

X
m=1

k=1

#
m l t
k
e
e
Ai
k
e

(5.14)

The finite-volume method can be written using a compact wave splitting formulation as follows:
Un+1
= Uni
i

NE 
X
k=1

M
i,k

n l
k
t
Ai

(5.15)

with
M
i,k

3 
X
m=1

m
e e

e
k

(5.16)

27

RiverFlow2D Model Equations

The use of (??) is efficient when dealing with boundary conditions. At the same time it ensures
conservation. In (?) it was demonstrated how for a numerical scheme written in splitting form,
the total amount of contributions computed inside the domain at each cell edge, is equal to the
balance of fluxes that cross the boundary of the domain, proving exact conservation.

5.4.1 Numerical Optimizations

Once wave propagations in M
i in (??) are computed, the first order method can be applied
averaging the contributions of the local Riemann Problems (RPs) shaping the contour cell.
The approximate solution is always constructed as a sum of jumps or shocks, even in cases
involving rarefactions. One widely reported problem of linearized solvers is the entropy violation
in sonic rarefactions (?, ?), that produces negative values of depth in the shallow water equations,
even in absence of source term. The solution is restored by means of a suitable redefinition of the
approximate solution by means of entropy fixes.
The time and space linearization of the source terms in (??) can also have negative consequences, as numerical instabilities may arise when approximating their value. Their influence over
the approximate RP solutions is the key to construct appropriate fixes that avoid unphysical results. In (?) it was shown how errors in the integral approaches done over the source terms can
be avoided if imposing physically based restrictions over the approximate solution. By simply
modifying the source strength coefficients correct solutions are restored when necessary.

5.4.2 Stability Region

Once numerical fixes are applied the stability region for the homogeneous case can be used to
compute the size of the time step. In the 2D framework, considering unstructured meshes, the
relevant distance, that will be referred to as i in each cell i must consider the volume of the cell
and the length of the shared k edges.
i =

Ai
maxk=1,N E lk

(5.17)

Considering that each k RP is used to deliver information to a pair of neighboring cells of

different size, the distance min(Ai , Aj )/lk is relevant. The time step is limited by
t CF L t
e

t =
e

min(i , j )
em |
max |

(5.18)

with CF L=1/2, as the construction of finite-volume schemes from direct application of onedimensional fluxes leads to reduced stability ranges (?).
RiverFlow2D solution method uses variable time steps. The maximum allowed time-step is
controlled by the user-set Courant-Friederich-Lewy (CFL) number that is proportional to the
local element size, but also inversely proportional to velocity and depth. Smaller elements lead to
smaller time-steps. The maximum theoretical CFL value is 1, but in some runs it may be necessary
to reduce this number to lower values.

5.5 Open Boundary Conditions

There are two main boundary condition types that can be used in RiverFlow2D: Open boundaries
where flow can enter of leave the modeling area and closed boundaries that are solid no-flow walls
(see following figure). There is no restriction on the number of inlet or outlet boundaries. This
section describes the open boundary conditions.

5.5

Open Boundary Conditions

28

Figure 5.4 Open and closed boundary conditions.

RiverFlow2D allows having any number of inflow and outflow boundaries with various combinations of imposed conditions. Proper use of these conditions is a critical component of a successful
RiverFlow2D simulation. Shallow water equation theory indicates that for two-dimensional subcritical flow it is required to provide at least one condition at inflow boundaries and one for outflow
boundaries. For supercritical flow all conditions must be imposed on the inflow boundaries and no
boundary condition should be imposed at outflow boundaries. The table below helps determining
which conditions to use for most applications.
Table 5.1 Required boundary conditions.
FLOW
REGIME
AT BOUNDARY
Subcritical
Supercritical

INFLOW
BOUNDARY
CONDITION
Q or Velocity
Q and Water Surface Elevation

OUTFLOW
BOUNDARY CONDITION
Water Surface Elevation

It is recommended to have at least one boundary where water surface or stage-discharge is

prescribed. Having only discharge and no water surface may result in instabilities due to violation
of the theoretical boundary condition requirements of the shallow water equations.

The open boundary condition options are described in the table below.

29

RiverFlow2D Model Equations

Table 5.2 Open Boundary condition types for hydrodynamics.

1
5
6
9
10
11
12

BC TYPE
DESCRIPTION
Imposes Water Surface Elevation. An associated boundary condition file
must be provided.
Imposes water discharge and water surface elevation.
Converts water discharge in velocities perpendicular to boundary line.
Imposes single-valued stage-discharge rating table.
Free inflow or outflow condition. Velocities and water surface elevations are calculated by the model.
Free outflow condition. Velocities and water surface elevations are
calculated by the model, but only outward flow is allowed.
Uniform flow outflow condition.

5.5.1 Single Variable Boundary Condition Types (BCTYPE 1 and

6)
When imposing a single variable (water surface elevation, or Q), the user must provide a time series
for the corresponding variable. To model steady state the time series should contain constant values
for all times. There is no restriction on the time interval used for the time series. When imposing
water surface elevation it is important to check that the imposed value is higher than the bed
elevation.

5.5.1.1 Water Discharge Converted in Velocities (BCTYPE 6)

In this inflow condition the program calculates the flow area and the average water velocity corresponding to the imposed discharge that can be variable in time. Then, velocity is assigned to each
cell assuming perpendicular direction to the boundary line as shown:

Figure 5.5 Inflow water discharge imposed as velocities (BCTYPE 6).

5.5

Open Boundary Conditions

30

5.5.2 Discharge Rating Table (BCTYPE 9)

When using a single valued stage-discharge condition the model first computes the discharge on
the boundary then interpolates the corresponding water surface elevation from the rating table
and imposes that value for the next time step. If the boundary is dry, it functions as a free
condition boundary. Water surface elevations are imposed only on wet nodes. This condition
requires providing an ASCII file with the table values entries. See section ?? for details on the file
format.
Since these condition may generate wave reflection that can propagate upstream, it is important
to locate the downstream boundary on a reach sufficiently far from the area of interest, therefore
minimizing artificial backwater effects. Unfortunately, there is no general way to select such place,
but numerical experimenting with the actual model will be necessary to achieve a reasonable
location.
In most small slope rivers, the stage-discharge relationship is affected by hysteresis. In other words,
the stage-discharge curve is looped with higher discharges occurring on the rising limb than on the
rescission limb of the hydrograph. This is mainly caused by the depth gradient in the flow direction
that changes in sign throughout the hydrograph. In practice, this implies that there can be two
possible stages for the same discharge. Loop stage-discharge relationships are not considered in
this RiverFlow2D version.

5.5.3 Free Open Boundaries (BCTYPE 10, 11)

On free condition boundary nodes, the model calculates velocities and water surface elevations
applying the full equations from the internal elements. In practice this should be equivalent to
assume that derivatives of water surface elevations and velocities are 0. It is advisable to use
these conditions when there is at least another open boundary where water surface elevation or
stage-discharge is imposed. BCTYPE 10 allows water outflow and inflow, while BCTYPE 11 will
only allow flow out of the mesh.

5.5.4 Uniform Flow Boundary Condition (BCTYPE 12)

To apply this boundary condition the user provides only the bed slope S0 . The model will use
S0 , Mannings n, and discharge to create a rating table. Then for each time-interval, the program
will impose the water surface elevation corresponding to the boundary discharge interpolating on
the rating table. The rating table is calculated every 0.05 m (0.16 ft.) starting from the lowest
bed elevation in the outflow cross section up to 50 m (164 ft.) above the highest bed elevation
in the section. If S0 = 999, the model will calculate the average bed slope perpendicular to the
boundary line.

5.5.5 Numerical Implementation of Open Boundaries

Many simulation models are based on reliable and conservative numerical schemes. When trying
to extend their application to realistic problems involving irregular geometries at boundaries a
special care has to be put in preserving the properties of the original scheme. Conservation, in
particular, is damaged if boundaries are careless discretized.

31

RiverFlow2D Model Equations

In the cells forming the inlet discharge region the flow is characterized by the negative sign of
the following scalar product in the k boundary edges
qi ni,k = (hu)i ni,k < 0

(5.19)

and by the state of the flow, defined commonly through the Froude number
F ri =

ui ni,k
ci

(5.20)

with ci = ghi . When the Froude number defined as in (??) is greater than one, the flow is
supercritical and all the following eigenvalues are negative:
1 = ui ni,k + ci < 0

2 = ui ni,k < 0

3 = ui ni,k ci < 0

(5.21)

therefore the values of h, u, v, and must be imposed. The water solute concentration is independent of the eigenvalues, and therefore has to be provided at the inlet region for all flow regimes.
The cells in the outlet discharge region are defined by
qi ni,k = (hu)i ni,k > 0

(5.22)

for supercritical flow, all the following eigenvalues are positive:

1 = ui ni,k + ci < 0

2 = ui ni,k < 0

3 = ui ni,k ci < 0

(5.23)

in consequence, no extra information is required.

When in both inlet and outlet discharge region, the flow state is subcritical, the updating
information is not complete. The same happens at the cell edges acting like solid walls, that cannot
be crossed by the flow. Commonly the extra information provided upstream and downstream are
discharge functions. And, on solid boundaries, a zero normal discharge function is defined.
To decide whether we are dealing with a supercritical or a subcritical inlet or outlet is not easy
in a 2D mesh. A cell based characterization of the boundary flow regime at the boundaries leads
to complicated situations both from the physical and from the numerical point of view. On the
other hand, physical or external boundary conditions usually refer to average quantities such as
water surface level or total discharge that have to be translated into water depth or velocity at
each cell, depending on the practitioner criterion. To handle these situations, a suitable connection
between the two-dimensional and the one-dimensional models is required at the open boundaries.
The section Froude number is defined once the boundary section has a uniform water level as:
w
F rs = p
g(ST /bT )

(5.24)

being the cross sectional velocity w = Q/ST and defining the total wet cross section ST and total
breath as:

ST =

NB
X
j=1

Sj =

NB
X
j=1

hj lj

, bT =

NB
X

lj

(5.25)

j=1

where N B is the number of wet boundary cells, lj is the length of each edge conforming the wet
boundary and hj is the water depth at each boundary cell.

5.5

Open Boundary Conditions

32

5.5.5.1 Inlet discharge boundary

This is one of the boundary conditions that poses most difficulties because a correct and conservative representation of the steady or unsteady incoming flow must be defined and there is not
one obvious form to implement it. The total inflow discharge hydrograph Q = Q(t) is the usual
function given in flooding simulation, and it is important to analyse the best way to impose it
since it involves the full inlet cross section and we are dealing with a 2D discrete representation in
computational cells. Different cases may be found.

5.5.5.2 Simple cases

When the inlet cross section is of rectangular shape (Figure ??), that is, of flat bottom and limited
by vertical walls, the inlet wet cross section is just rectangular.

Figure 5.6 Rectangular inlet cross section.

The total inlet discharge at time t, QI (t), can be distributed along the inlet cross section using
a constant discharge per unit width, qI (m2 s1 ), that can be calculated as
qj = qI =

QI (t)
bT

(5.26)

In this simple case, qI is uniform along the inlet boundary and so is the resulting modulus of
the velocity, w = qI /h, with w = (u2 +v 2 )1/2 . It should be noted, that the direction of the entering
discharge is not necessarily the same as the direction normal to the inlet boundary. However, this
direction is usually chosen as the default information.

5.5.5.3 Complex cases

In real problems of general geometry the inlet cross section may change shape as water level changes
(drying/wetting boundary), and so does the number of boundary cells involved (Figure ??).

33

RiverFlow2D Model Equations

Figure 5.7 Irregular inlet cross section.

When dealing with inlet sections like that in Figure ??, a uniform value of qI as in (??) leads
to a completely unrealistic state of faster water at the section borders and slower water at the
middle of the cross section. Since the resulting velocities depend on the value of water depth h,
higher values will appear in those cells where water depth is smaller.
In order to seek a more appropriate distribution, a uniform modulus of the water velocity w
is enforced in the whole inlet boundary cross section. In this case, the unit discharge at each
boundary cell j is variable and defined depending both on the total cross section area, ST , and on
the individual cell transverse area, Sj as follows:
qj = QI

Sj
ST lj

(5.27)

On the other hand, the updating of the water depth values at the inlet cells provided by the
numerical scheme leads in the general case to a set of new water depths hn+1
(Figure ??) associated,
j
in general, to different water surface levels dj dj = hj + zj .

Figure 5.8 Evaluation of dmin .

For our purposes a horizontal water surface level is required in that region, in order to help
in the translation between the 2D and the 1D points of view at the open boundary. The value
of that uniform cross sectional water level is fixed taking into account mass conservation, that is,
conservative redistribution of water volume. The minimum value of the water levels among all the
wet cells in the inlet boundary, dmin , is found and the water volume VS stored in the inlet section
above dmin is evaluated as
VS =

NB
X
j=1

(dj dmin )Aj |dj >dmin

(5.28)

5.5

Open Boundary Conditions

34

and the wet surface above that level, Aw , is defined:

Aw =

NB
X

Aj |dj >dmin

(5.29)

j=1

They are used to redistribute the volume over the inlet section, keeping constant the wet section
breadth bT . As Figure 3 shows a new uniform water level at the section, dS , is given by:
ds = dmin +

VS
Aw

(5.30)

Apart from helping to decide the flow regime at the boundary, the modifications described
above make easier the treatment of supercritical inflow conditions. When modeling unsteady river
flow, high peaks in the hydrograph can be encountered. If those peaks are not correctly handled
from the numerical point of view, they can lead to local and unrealistic supercritical states in the
inlet boundary.
In that case of supercritical inlet flow, the specification of all the variables at the inlet boundary
cells is required. However, in many practical problems only the discharge hydrograph is available
as a function of time, with no data, in general, on the water level distribution or discharge direction
at the inlet boundary.
The alternative proposed is, when the inlet Froude number is bigger than 1
F rs = p

w
g(ST /bT )

>1

(5.31)

to enforce a maximum Froude number, F rs,max , to the inlet flow. For that purpose, keeping
the section breadth bT , a new inlet wet cross section area, ST , is computed from the F rs,max
imposed:
ST = (

Q2I
2
gF rs,max
/bT )

)1/3

(5.32)

If ST is greater than ST , it provides a new water surface level for the inlet section, d , also
greater than ds (Figure ??). The associated increment in water volume is balanced by means of a
reduction in the imposed discharge QI (t) in that time step.
Occasionally, both conditions, QI (t) and d(t) are known at supercritical inlets. For those cases,
imposing both data at the inlet boundary is enough. However, due to the discrete time integration
method used, this procedure does not follow the mass conservation criterion. To guarantee that
the mass balance is preserved, one of the conditions is imposed, the other must be modified, so
that the fluxes calculated in the following step lead to mass conservation. The best solution is to
impose directly the global surface water level at the inlet boundary section,d(t), and to adapt the
discrete inlet discharge to ensure that the final volume is conserved. The imposed value of d sets
an input volume that can be transformed into discharge by means of dividing it by the time step.
This value is added to the discharge leading to a correct mass balance.

35

RiverFlow2D Model Equations

Figure 5.9 New water level for the inlet section.

When the boundary cell belongs to an open boundary where the inlet flow discharge is the
condition imposed and the flow is subcritical, the discharge is computed using (??) and imposed
in the boundary cell. Moreover, the water level is computed as a results of the contributions from
that other cell edges in (??) when updating the conserved values in the boundary cell at time level
n + 1 and is carefully redistributed as explained before.

5.5.5.4 Outlet boundaries

The analysis of the flow at the outlet boundary is simpler. For supercritical outflow no external
conditions have to be imposed. In RiverFlow2D, a preliminary sweep is performed over the wet
outlet boundary cells in order to evaluate the cell Froude number. If a supercritical cell is found,
the whole flow at the outflow boundary section is considered supercritical and no external condition
has to be enforced. Otherwise, all the cells are in a subcritical state, and receive an analogous
treatment to that of the inlet boundary described above. As before, a uniform cross sectional
water level is generated and a velocity distribution is set in cases in which a discharge rating curve
is the boundary condition to impose.

5.5.6 Closed Boundaries

Closed boundaries are rigid or solid walls that completely block the flow such as river banks or
islands. They constitute vertical walls that the flow can never overtop. A very thin viscous sublayer
occurs near these boundaries that would require extremely small elements to be appropriately
resolved. RiverFlow2D uses slip condition on closed boundaries and the model will set zero normal
flow across the boundary, but tangential velocities are allowed. The RiverFlow2D model detects
closed boundaries automatically.
This kind of boundary condition does not require any special treatment. As no flow must cross
the boundary, the physical condition u n = 0 is imposed on the cell velocity u after adding all the
wave contributions from the rest of the cell edges, where n is the solid wall normal (Figure ??).
In other words, if the boundary is closed, the associated boundary edge k is a solid wall, with a
zero normal velocity component. As there are no contributions from that edge, M
i,k = 0 is set
in (??) when updating the conserved values in the boundary cell at time level n + 1.

5.6

Dry/Wet Cell Modeling

36

Figure 5.10 Solid wall condition.

5.6 Dry/Wet Cell Modeling

RiverFlow2D is able to simulate the drying and wetting of the bed. This model capability is
important when simulating flood wave progression down an initially dry channel. In this case both
the channel bed and floodplain will get inundated. The channel bed can also dry again as the flood
wave recedes.
In RiverFlow2D the triangular-cell mesh can cover both dry and wet areas and the model will
handle these conditions using two distinct algorithms and depending on the following element
classification.

5.6.1 Cell definitions Based on Dry and Wet Conditions

A cell is considered dry if its water depth is less than a fraction of a millimeter. There is not a
partially dry cell situation. A cell edge is considered inactive if it separates two dry cells and is
excluded from the computation. Otherwise, the cell edge always contributes to the updating of
the variables on both sides. The so called wet/dry situation takes place at a cell edge when all the
following conditions hold:
One of the neighbor cells is wet and the other is dry.
The water level a the wet cell is below the bed level at the dry cell.
Flow is subcritical.
In that case, the procedure to follow is well described in (?).
RiverFlow2D drying and wetting algorithm is an adaptation of the the one originally proposed
by (?) and later improved by (?) and (?) in the finite-volume context and works as follows:
1. At the beginning of each time-step all cells are classified as wet or dry according to the
definition.
2. If a cell is dry and completely surrounded by dry cells, it is removed from the computations
and velocity components are set to zero for the ongoing time step.
3. All the internal cell edges are classified as active or inactive according to the definition.

37

RiverFlow2D Model Equations

4. Wet/dry cell edge contributions are computed assuming the edge is a solid boundary and
the velocities on both sides are set to zero.
5. The rest of the cell edge contributions are computed according to the numerical scheme as
described above.
6. Wet cells and dry cells surrounded by at least one wet cell are retained in the computation
and solved with the updating scheme using the contributions from the cell edges.
This method generates stable numerical solutions without spurious velocities over dry areas and
offers machine accuracy mass conservation errors allowing the use of the classical CFL condition.

5.7 Volume Conservation

The volume conservation or volume balance in the simulation domain can be defined through a
discharge contour integral:
t+t

(QI nI QO nO )dt

M (t) =

(5.33)

where QI and QO are the total discharge functions at the inlet and at the outlet boundaries
respectively, and nI and nO are the normal vectors to the boundaries. The normal discharge at
solid walls is zero. This balance is actually evaluated integrating at the contour cell by cell as
follows
M (t) =

N
BI
X

qI,j lj (nI nj )t

N
BO
X

qO,m lm (nO nm )t

(5.34)

m=1

j=1

where nj and nm are the directions of the flow in the inlet and in the outlet cells respectively.
The volume variation in the domain of calculation can be only due to
M (t) 6= 0

(5.35)

Therefore, the mass error of the numerical solution is measured by comparing the total amount
of water calculated at time t + t
V ol(t + t) =

N CELLS
X

hn+1
Si
i

(5.36)

i=1

with the total amount of water existing at time t

V ol(t) =

N CELLS
X

hni Si

(5.37)

i=1

as follows
Error = [V ol(t + t) V ol(t)] M (t)

(5.38)

This is usually expressed in relative terms as follows:

Relerror =

[V ol(t + t) V ol(t)] M (t)

V ol(t) + M (t)

(5.39)

5.8

Mannings n roughness Coefficients

38

5.8 Mannings n roughness Coefficients

The Mannings n usually estimated to determine head losses in channel and river flow is a global
measure that accounts not only for the effects of bed roughness, but also for internal friction
and variations in shape and size of the channel cross section, obstructions, river meandering (Ven
Te Chow, 1959). Therefore, estimations of Mannings n applicable for 1D models should be
adjusted, because 2D model equations consider two-dimensional momentum exchange within the
cross section that is only lumped in the 1D simplification. Several researchers have found in
practical applications of 2D models that the n values required can be 30% lower than those normally
used for 1D models on the same river reach (Belleudy, 2000). However, 2D models do not account
for lateral friction, therefore the final selection of Mannings n coefficients should be the outcome
of a calibration process where the model results are adjusted to measured data.

5.9 Sediment Transport Model: RiverFlow2D ST

Module
The science of sediment transport deals with the interrelationship between flowing water and
sediment particles. Despite of having been studied since the 1950s and being widely employed
in engineering practice, (?), (?), the sediment transport modeling remains at present one of the
most active topic in the field of hydraulic research. Although numerical modeling of free surface
flows with bed load transport over erodible bed in realistic situations involves transient flow and
movable flow boundaries, the conventional and pioneer methods for performing morphodynamic
simulations in coastal areas and rivers decouple the hydrodynamic and the erosion and deposition
components (?, ?, ?). Ignoring unsteady hydrodynamic effects means that the time scales of the
morphodynamics changes are smaller in comparison with the hydrodynamic ones. Assuming this
hypothesis, only a quasi-steady process of slowly varying bed-load can be reasonably modeled, so
that rapidly varying flows containing shocks or discontinuities remain excluded. For this reason
RiverFlow2D allows accurate simulation ranging from slow-evolving events to abrupt river bed
changes.

5.9.1 Model Equations

The relevant formulation of the model derives from the depth averaged equations expressing water
volume conservation, solute volume conservation and water momentum conservation. That system
of partial differential equations is formulated here in coupled form as follows
U F(U) G(U)
+
+
= S(U) + R(U)
t
x
y

(5.40)

where
U=

h,

qx ,

qy ,

h1 ,

...,

hNp ,

PNp

p=1 [zp (1 pp )]

T

(5.41)

are the conserved variables with h representing the water depth, qx = hu and qy = hv the
unit discharges, with (u, v) the depth averaged components of the velocity vector u along the
x and y coordinates respectively and p , with p = 1, ..., Np representing the scalar depth-averaged
concentration of the Np different sediments transported in suspension. The term z is the bed level
PNp
and pp is the bed porosity of each p bed sediment. The sum of p=1
[zp (1 pp )] is motivated for
the possibility of a heterogeneous soil, where different fractions of material may coexist.

39

RiverFlow2D Model Equations

On the other hand the fluxes are given by

F=

G=

2
qx
h

qx ,
qy ,

+ 12 gh2

qx qy /h,

qy2
h

qx qy /h,

qx 1 ,

...

qx Np ,

qsx

+ 12 gh2

qy 1 ,

...,

qy Np ,

qsy

T
T

(5.42)

where g is the acceleration of the gravity and qsx and qsy are the bed load transport terms computed
by means of an empirical law. The source terms of the system are split in three kind of terms.
The term S is defined as

S=

T
b,x pb,y
b,y
pb,x

, 0, . . . , 0, 0
0,
w
w w
w

(5.43)

with pb,x , pb,y and b,x , b,y are the pressure force along the bottom and the bed shear stress in
the x and y direction respectively, with w the density of water. The former can be formulated in
terms of the bed slopes of the bottom level z
pb,x
z
= ghS0x ,
= gh
w
x

pb,y
z
= ghS0y
= gh
w
y

(5.44)

and the friction losses are written in terms of the Mannings roughness coefficient n

b,x
= ghSf x
w

Sf x

n2 u u2 + v 2
=
,
h4/3

b,y
= ghSf y
w

Sf y

n2 v u2 + v 2
=
h4/3

(5.45)

The reaction source terms R, having a volumetric character, is defined as

R = 0, 0, 0, R1 , . . . , RNp ,

Np
X

T
Rp

(5.46)

p=1

and is evaluated as
Rp = sp (p p )

(5.47)

where the term p is employed for defining the equilibrium concentration, which is obtained
through a solid transport discharge law. The term p contains the information about the suspension
sediment quantity which is transported. Both concentrations takes in consideration the presence
of the settling velocity, sp .

5.9.2 Sediment Transport Laws

Two different ways of sediment transport govern the dynamics of the mobile bed considered in
RiverFlow2D: the suspended load and the bed load. Both of them may coexist or one may be
dominant.
As each sediment transport law is derived from different laboratory and field data sets, a calibration parameter in the form of a correction factor is considered in order to adjust the numerical
results.

5.9

Sediment Transport Model: RiverFlow2D ST Module

40

5.9.2.1 Bed Load

When bed load is the dominant sediment transport mechanism and the influence of the suspended
load is negligible, system (??) reduces to
Mass conservation

(h) (hu) (hv)

+
+
=0
t
x
y

(5.48)

Momentum conservation in x direction

(hu) [hu2 + (1/2)gh2 ] (huv)
pbx
bx
+
+
=

t
x
y
w
w

(5.49)

Momentum conservation in y direction

(hu) (huv) [hv 2 + (1/2)gh2 ]
pby
by
+
+
=

t
x
y
w
w

(5.50)

Bed elevation changes

PNp

p=1

[zp (1 pp )] qsx
qsy
+
+
=0
t
x
y

(5.51)

Where Np m and the terms of qsx and qsy are the volumetric sediment fluxes per unit width
can be determined through several deterministic laws or sediment transport formulas and include
the total transport from all sediment fractions.
The modulus of the sediment transport rate, qs , is defined as:
qs =

p
qsx + qsy

(5.52)

The bed load transport is often represented by the following dimensionless parameter,
|qs |
= p
g(s 1)d250

(5.53)

where s is the solid material vs water density ratio and d50 is the median diameter.
Table ?? collects the formulas that are implemented in RiverFlow2D, where d90 , d50 and d30
are the grain diameter for which 90%, 50% and 30% of the weight of a non-uniform sample is finer
respectively, s is the sediment density, c is the critical Shields parameter and cs is the critical
Shield parameter as expressed by Smart (1984).
Table 5.3 Summary of bed load transport formulas.
FORMULA
Meyer-Peter & Mueller
(1948)
Ashida Michiue (1972)
Engelund and Fredse
(1976)
Fernandez-Luque and
van Beek (1976)
Parker fit to Einstein
(1979)
Smart (1984)
Nielsen (1992)
Wong
Wong
Camenen-Larson

INPUT PARAMETERS
8 ( c )3/2
d50 ,c ,s



c


18.74 ( c )
0.7 c

d50 , c , s

5.7 ( c )3/2

d50 , c , s

11.2 (1 /c )9/2

d50 ,c ,s

4 (d90 /d30 )0.2 S00.6 C1/2 ( cs )

121/2 ( c )
4.93 ( c )1.6
3.97 ( c )3/2
123/2 exp (4.5c /)

d30 ,d50 , d90 , c , cS , s

d50 , c ,s
d50 , c , s
d50 , c , s
d50 , c , s

17 ( c )

d50 , c ,s

41

RiverFlow2D Model Equations

PNp
The sum of p=1
[zp (1 pp )] in (??) can consider heterogeneous soil, where different fractions
of material may coexist. In order to take into account this heterogeneity, every sediment transport
discharge (qSx , qSy ) associated to each sediment size is multiplied by its corresponding soil fraction.
For every sediment particle size it is necessary to include the fraction of the material.

5.9.3 Boundary Conditions for the Bed Load Sediment Model

When using the bed load sediment model, the user may impose inflow boundary condition BCType
= 6 and provide only the water discharge. In that case the model will determine the inflow sediment
discharge based on the selected sediment transport formula assuming equilibrium conditions.
Alternatively you may provide the sediment discharge for each sediment fraction at all inflow
boundaries using BCType = 26. This data should be included in the same file that sets the
discharge time series at each inlet. For instance, if you select two sediment fractions BCType =
26 where the inflow is set as discharge a typical file would be like this one:
3
0.0
2.0
100.0

1000.
7000.
7000.

0.16
1.16
1.16

0.23
2.24
2.24

In file there are 3 times: 0, 2 and 100 hours. The first column represents the time in hours.
The second column indicates the water discharge in m3 /s or f t3 /s. The last two columns have
the sediment discharge for each given fraction in in m3 /s or f t3 /s.

5.9.4 Suspended Load

When the suspension load plays the key role in the mobilization of the bed, the term of bed load
can be omitted and system (??) is written as
Mass conservation

(h) (hu) (hv)

+
+
=0
t
x
y

(5.54)

Momentum conservation in x direction

(hu) [hu2 + (1/2)gh2 ] (huv)
pbx
bx
+
+
=

t
x
y
w
w

(5.55)

Momentum conservation in y direction

(hu) (huv) [hv 2 + (1/2)gh2 ]
pby
by
+
+
=

t
x
y
w
w

(5.56)

Sediment mass of the fluid layer for the suspended sediment p

(hp ) (hup ) (hvp )
+
+
= sp (p p )
t
x
x

(5.57)

The term p defines the equilibrium volume concentration, which is obtained through a solid
transport discharge law. The term p contains the information about the suspension sediment
quantity which is transported. Both concentrations take in consideration the presence of the

5.9

Sediment Transport Model: RiverFlow2D ST Module

42

settling velocity, sp . The settling velocity can be computed through several laws as a function of
the sediment density and the grain diameter.
Table ?? collects the formulas considered in RiverFlow2D. It is worth noticing that these laws
have been derived for one single particle and under steady situations. Hence, in case of considering
a transient situation with a sediment concentration in the water column, the influence of the
nearby particles must be taken into consideration, which can strongly reduce the fall velocity in
comparison with clear water. This effect, known as hindered settling (?, ?), can be determined by
(?), computing a new settling velocity sm
n


s
(5.58)
sm = 1
2 d50
where s is the settling velocity of a single particle in clear water, is the volumetric concentration,
d50 is the mean diameter and n is an empirical exponent with a usual value equal to 3.
Table 5.4 Formulas to calculate sediment settling velocity s .
FORMULAr s
OBSERVATIONS


s
s = F
w 1 gds

Rubey
(1933)

If ds > 1 mm, F
= 0.79, otherwise F =

0.5
2
36 2

3 + gd3 (s /w 1)
s

0.5
36 2
gd3
s (s /w 1)

Zhang
(1961)
Zanke
(1977)

s =

r

13.95 ds

s = 10 ds

2

+ 1.09


s =

s =

Julien
(1995)

s =


1 gds 13.95 ds

s
w




s 1
1 + 0.01 w

Van
Rijn
(1984)
Raudkivi
(1990)

1 s w
18
w

d2
s

gd3
s
2

ds < 0.1 mm


32

(2/3)(s /w 1)gd2
s


24

p

1.2ds

1.5
25 + 1.2D2 5

s =

Jim
enezMadsen
(2003)

p
s = (C1 + C2 S ) (s 1)gds

r
s =

M
N ds

1
4

4N
3M 2

D3

1/n

ds > 1 mm

1.5ds

Cheng(1997)

WuWang(2006)

Valid for a wide range of

ds


1

(2/3)(s /w 1)gd2
s

ds

0.5

D
=

1/3
ds (s /w 1) g/ 2
C1 , C2 and S as in
Jim
enez-Madsen
!n

0.5

M,
N
and n as
in Wu and D =

1/3
ds (s /w 1) g/ 2

5.9.4.1 Suspended Load Formulas

As it has been mentioned before, the term p which appears in (??) is employed for defining the
equilibrium concentration, which is obtained through a solid transport discharge law as follows,
p =

qsuspp
hu

(5.59)

where the subscript p makes reference to the sediment class p. The formulas employed for computing the solid transport discharge are depicted below.
Bagnold (?) considered that the shear stress is proportional to the flow velocity and it was

43

RiverFlow2D Model Equations

s b u2
s w s

(5.60)

established the following formula,

qsuspp = 0.01

being b the shear stress generated at the bottom by the bed roughness which is taken into
account through the Mannings coefficient, n
gn2 u2
h1/3

b =

(5.61)

Van Rijn (?) proposed calculating the suspended load integrating the sediment flux within
the layer where the suspension plays a key role, i.e. between the term hs and h,
Z

qsuspp =

m udh

(5.62)

hs

where, qsusp is expressed by volume per unit time, and the terms of m and hs were defined
as follows

m = s

h
h0
h
hs

!s /(kU )
(5.63)

with

s =

1/3 


0.117

2
1
ds
(s 1)g
c
"

hs = 0.3ds ds

(s 1)g
2

1/3 #0.7 r

1
c

(5.64)

(5.65)

where is the non-dimensional shear stress, calculated using (??) and (??) and related to n,
the Manning roughness coefficient, c is the Shieldss parameter, ds is the sediment diameter,
s is the rate between densities, s = ws and is the kinematic viscosity.
=

b
w (s 1)gds

(5.66)

gn2 u2
h1/3

(5.67)

with
b =

5.9.5 Boundary Conditions for the Suspended Sediment Model

When using the suspended sediment model, the user must impose volume concentration for each
sediment fraction at all inflow boundaries. This data should be included in the same file that
sets the discharge time series at each inlet. For instance, if you select two sediment fractions and
BCType = 6 where the inflow is set as discharge a typical file would be like this one:
There are 3 times: 0, 2 and 100 hours. The first column is the time in hours. The second
column is the water discharge in m3 /s or f t3 /s. Then there are two columns that should have the
sediment volume concentration for each given fraction.

5.9
3
0
2
100

Sediment Transport Model: RiverFlow2D ST Module

1000.
67000.
67000.

0.001
0.001
0.001

44

0.002
0.002
0.002

5.9.6 Geomorphological Collapse

When managing transient geomorphological flows in realistic cases, the geotechnical equilibrium
bank characteristics can be ruined, leading to dramatic channel metamorphosis. This effect needs
to be modeled to reproduce correctly bed geometry evolution in combination with flow action.
In RiverFlow2D, the effect of the geomorphological collapse is introduced in the simulation by a
simple mass conservative mechanism of slope sliding failure, assuming that the angle of repose of
submerged material of the bed can be approximated by the friction angle. The failure mechanism is
applied by comparison between the bed slope in each cell edge and the angle of repose of saturated
bed material, (?).

5.9.7 Sediment Transport Finite-Volume Numerical Solution

The system of equations (??) is integrated in a volume or grid cell using Gauss theorem:

Ud +

Endl =

Sd +

Rd

(5.68)

where n = (nx , ny ) is the outward unit normal vector to the volume .

In order to obtain a numerical solution of system (??) we divide the domain in computational
cells, i , using a mesh fixed in time, and (??) is applied to each cell

Z
Ud +
i

NE Z
X
k=1

ek+1

ek

(En)k dlk

Z
=

Z
Sd +

Rd

(5.69)

with (En)k the value of the interface flux function through the edge k to be defined, nk = (nx , ny )
is the outward unit normal vector to the cell edge k, and N E is the number of edges in cell i. A
sketch of the fluxes is showed in Figure ??.

Figure 5.11 Cell parameters.

Assuming a piecewise representation of the conserved variables

Uni =

1
Ai

Z
i

U(x, y, tn )d

(5.70)

45

RiverFlow2D Model Equations

Equation (??) is written as

Z
Ud +
i

NE
X

(En)k lk =

k=1

Z
Sd +

Rd

(5.71)

where lk is the corresponding edge length. System (??) is solved following the theory of Roes
Riemann solver and using the upwind discretization [(?, ?)].

5.9.8 Entering Data for the Sediment Transport Model

To enter data for a sediment transport simulation use the Sediment Transport pannel in the
RiverFlow2D Data Input Program.

Figure 5.12 RiverFlow2D Sediment Transport panel.

5.9.9 Assumptions of the Sediment Transport Model

The main assumptions involved in the present version of RiverFlow2D model are:
When using the bed load option, the inflow sediment transport rate is equal to the transport
capacity under equilibrium conditions.
Each Sediment particle size is considered using a single characteristic diameter (D50 ) for all
formulas except for that of Smart (1984) that considers three sizes (D30 , D50 and D90 ).
The model does not account for bed armoring.
When assigning maximum erosion bed elevation, the model blocks further erosion when the
bed elevation reaches the given limit.

5.10

Mud and Debris Flow Model: RiverFlow2D MD module

46

Since the formulations for the bed load discharge, the suspended load discharge and the settling
velocity are based on deterministic laws supported by experimentation, tuning parameter factors
have been considered for each one. Hence, the model makes possible the calibration of the computed
sediment transport for each particular situation.

5.10 Mud and Debris Flow Model: RiverFlow2D MD

module
Mud/debris floods are unsteady flow phenomena in which the flow changes rapidly, and the properties of the fluids, typically non-Newtonian, are those of a hyperconcentrated mixture water and
sediment and include stop and go mechanisms. The global resistance behavior of the mud/debris
flow depends on the relative importance of the shear stresses arising from different sources that,
apart from turbulent shear stress at the river boundary, include viscous stress, yield stress, dispersive stress and inelastic collisions of solid particles within the fluid mixture Naef et al. 2006.
The mathematical model adopted in RiverFlow2D is based on that of Murillo & Garcia-Navarro
(2012), and regards bed and internal friction for free-surface flows ranging from clear water to
hyperconcentrated mixtures of sediments.

5.10.1 Assumptions of the Mud and Debris Flow Model

The model involves the following assumptions:
The flow is confined to a layer which is thin compared to the horizontal scale of interest;
The flow is governed by equation (??) with the friction terms evaluated as explained below;
The mixture of water and sediments is described by using the continuum approach and
assuming the same velocity for the liquid and for the solid phase;
The river bed does not erode;
The fluid is assumed to be an homogeneous single-phase mix of water and sediment and has
constant properties: e.g. density, yield stress, etc.

5.10.2 Rheological Formulation

The single-phase rheological formulation in RiverFlow2D MD accounts for different friction terms
that represent a variety of hyperconcentrated non-Newtonian fluids (see Figure ??). The friction
term in equation (??) is depth-averaged. Hence, the equations that describe the tangential forces
generated by the stresses can be lumped into the same mathematical formula despite of having a
different nature. In this way, several shear stresses can be considered to cover a wide range of mud
and debris flow conditions, including:
Turbulent stress t
Dispersive stress d
Coulomb-type frictional stress y
Yield stress y and

47

RiverFlow2D Model Equations

Figure 5.13 Rheological diagram of some common non-Newtonian formulations.

Viscous stress
Table ?? shows the rheological friction laws available in RiverFlow2D MD.
Table 5.5 Rheological flow resistance formulations used in RiverFlow2D MD.
FORMULATION
Turbulent
Full Bingham

FLOW RESISTANCE TERM

b = t
b = 0 with f1 (0 , y ) = 0

Simplified Bingham

b = 1.5y + 3

Turbulent
and
Coulomb
Turbulent and Yield
Turbulent,
Coulomb
and Yield
Quadratic
Granular

b = t + f
b = t + y


b = t + min y , f
b = t + y +
b = f

INPUT PARAMETERS
Mannings n
Bingham viscosity, Yield Stress,
Density
Bingham viscosity, Yield Stress,
Density
Mannings n, Friction angle, Density
Mannings n, Yield Stress
Mannings n, Yield Stress, Friction angle, Density
Mannings n, Yield Stress, Friction angle, Density
Mannings n = 0, Friction angle

5.11

Pollutant Transport Model: RiverFlow2D PL Module

48

5.10.3 Entering Data for the Mud and Debris Flow Model
To enter data for a mud and debris flow simulation use the Mud/Debris Flow panel in the
RiverFlow2D Data Input Program. Also make sure that the Mud/Debris Flow check box is active
in the Control Data panel.

Figure 5.14 RiverFlow2D Mud/Debris Flow panel.

Depending on the selected flow resistance relation, the user will need to enter different parameters. For example the Full Bingham formulation requires yield stress, Bingham viscosity and
Material density, etc..

5.11 Pollutant Transport Model: RiverFlow2D PL

Module
The study of solute transport phenomena and river mixing has become a great concern in hydraulic
and environmental problems. The RiverFlow2D Pollutant Transport Model provides a tool to
calculate concentrations of multiple pollutants in a variety of riverine and estuarine situations.
A solute or pollutant is defined as any substance that is advected by water and well mixed
in the vertical direction. The interest is usually focused around the time evolution of a solute
concentration within a complex hydrodynamic system, that is, given the solution concentration at
a specific time and space, the model determines the spatial distribution of the solute concentrations
at for future times. This physical process is accounted for the advection-dispersion equation and
can incorporate the effect of reaction with the water and with other solutes.

5.11.1 Model Equations

Although RiverFlow2D PL can handle multiple pollutants simultaneously, for the sake of clarity
in this section the transport of only one solute is presented coupled to the 2D model. The solute

49

RiverFlow2D Model Equations

transport equations will be expressed in a conservative form, assuming that the velocities and the
water depth may not vary smoothly in space and time.
Correspondingly, the 2D shallow water model with solute transport can be written in unique
coupled system:
U F(U) G(U)
+
+
= H(U) + R(U) + D(U)
t
x
y

(5.72)

where
T

U = (h, qx , qy , h)
F = qx , qxh2 + 12 gh2,

T
qx qy
h , hu


G = qy ,

qx qy
qy 2
h , h

H = (0, gh(S0x Sf x ), gh(S0y Sf y ), 0)

+ 21 gh2, hv

T

(5.73)

and is the depth-averaged solute concentration. The sources terms associated to the solute
transport equation are expressed as follows:
T

R = (0, 0, 0, Kh)

T
D = 0, 0, 0, (Dh )

(5.74)

where K is the uptake constant and D is an empirical diffusion matrix.

5.11.2 Pollutant Transport Finite-Volume Numerical Solution

In RiverFlow2D, the solute transport has been considered letting aside the consideration concerning
diffusion terms. However many strategies such as splitting and computing separately the advection
and the diffusion terms or solving the diffusion implicitly (?, ?), have been developed to avoid small
values in the time step size due to the combination of the CFL and Peclet number.
The numerical resolution of the solute transport equation under an explicit finite-volume
method is frequently performed by solving the depth-averaged concentration apart from the shallow
water equations, that is, using a simpler decoupled algorithm. Once the hydrodynamic equations
have been solved, the corresponding substances or solutes are advected with these flow field previously computed.
In order to get a fully conservative method, RiverFlow2D considers the complete system including the hydrodynamic and the transport equations. Mathematically, the complete system
conserves the hiperbolicity property, implying the existence of a 4 4 Jacobian matrix for the
2D model. On this basis we can apply the straightforward procedure described above, allowing a
Roes local linearization and expressing the contributions that arrive to the cell as a sum of waves.
To ensure conservation and bounded values in the final solute concentration even in extreme cases,
a conservative redistribution of the solute maximum fluxes as proposed in (?) was implemented
in RiverFlow2D.
According to (?), once the hydrodynamic part is properly formulated, a simple numerical flux
q , directly related to the Roes linearization, which is able to completely decouple the solute
transport from the hydrodynamic system in a conservative way is used. Therefore,
qk

= qi +

3 
X
m=1

ee
e1

m
k

(5.75)

5.11

Pollutant Transport Model: RiverFlow2D PL Module

50

where qi = (hun)i and the decoupled numerical scheme for the solute transport equation is
written as:
N

(h)n+1
= (h)ni
i

E
t X
(q)k lk
Ai

(5.76)

k=1

where
k


=

i
j

if
if

qk > 0
qk < 0

(5.77)

in cell i. A sketch of the fluxes is showed in Figure ??.

From a physical point of view, the new solute mass at a fixed cell can be seen as exchanging
water volumes with certain concentration through the neighboring walls and mixing them (finitevolume Godunovs type method) with the former mass existing in the previous time (Figure ??).
According to this philosophy, the outlet boundary cells will require a special treatment when
applying this technique in order to extract the corresponding solute mass through the boundary
walls. For this reason, it is necessary to define q = (hu n)BC and = BC at the boundary
wall and to include this contribution for the updating of the boundary cell BC (see Figure ??).

Figure 5.15 Physical representation of solute mass exchange between cells with qi1/2
, qi+1/2
> 0.

51

RiverFlow2D Model Equations

Figure 5.16 Extraction of mass solute in an outlet boundary cell.

As shown, the formulation reduces to compute a class of numerical flux q using the already
computed averaged values at each edge for the hydrodynamic paPollutant Transport tabrt. Apart
from ensuring a perfect conservation and bounded free-oscillatory solutions (Murillo et al, 2012),
this simply discretization decreases substantially the number of computations that would be necessary for the complete coupled system.

5.11.3 Entering Data for the Pollutant Transport Model

To enter data for a pollutant transport simulation use the Pollutant Transport tab in the Global
Parameters dialog accessible from the RiverFlow2D menu. Also make sure that the Pollutant
Transport check box is active in thee Control Data tab.

Figure 5.17 RiverFlow2D Pollutant Transport tab.

5.11

Pollutant Transport Model: RiverFlow2D PL Module

52

5.11.4 Assumptions of the Pollutant Transport Model

The main assumptions involved in the present version of RiverFlow2D model are:
1. There is no predetermined limit to the number of pollutants.
2. Interaction between solutes and between each solute and water are assumed to be first order
reactions.
3. All inflow boundaries where either discharge or water elevation is imposed must provide a
concentration time series for each pollutant.

Code Parallelization
6.1 RiverFlow2D CPU
RiverFlow2D code has been parallelized using OpenMP directives available in the Intel C++
compiler version 13.0. OpenMP Application Program Interface (API) supports multi-platform
shared-memory parallel programming in C/C++ and Fortran on architectures, including MAC
OS, Unix and Windows platforms (OpenMP, 2009). OpenMP provides instructions to parallelize
existing serial codes to run in shared-memory platforms ranging from affordable and widely available multiple-core computers to supercomputers. Using this parallelization approach RiverFlow2D
dynamically distributes the computational workload between as many processors or cores as are
available. In this way the model optimizes its computations to the particular architecture of each
computer.
Figure ?? shows the speedup of the model with respect to the number of processors/cores on a
DELL Precision 7400 computer with 2 Intel Xeon CPU X5472 @3.00GHz and 16GB of RAM. With
8 cores, the model runs more than 4 times faster than with the non-parallelized model. One hour
simulation takes approximately 6 minutes using the parallelized model in this particular computer
platform.

6.2 RiverFlow2D GPU

The GPU version of the RiverFlow2D model offers amazing speedups that considerably reduce
run times. RiverFlow2D GPU implements two strategies: OpenMP parallelization and GPU techniques. Since dealing with transient inundation flows the number of wet elements changes during
the simulation, a dynamic task assignment to the processors that ensures a balanced work load
has been included in the Open MP implementation. RiverFlow2D strict method to control volume conservation (errors of Order 1014 %) in the numerical modeling of the wetting/drying fronts
involves a correction step that is not fully local which requires special handling to avoid degrading the model performance. The GPU version reduces the computational time by factors of up
to 150X when compared with non-parallelized CPU (1-core) runs. Figure ?? shows performance
tests using recent GPU hardware technology, that demonstrate that the parallelization techniques
53

6.2

RiverFlow2D GPU

54

Figure 6.1 Speed up using RiverFlow2D parallelized code as a function of number of processor
cores.

Figure 6.2 Speed up using RiverFlow2D parallelized code as a function of number of processor
cores.

implemented in RiverFlow2D GPU can significantly reduce the computational cost.

Hydraulic and Hydrologic Components

RiverFlow2D components are internal boundary conditions that can be used to complement calculations that may not be directly handled using the 2D flow equations. Components can be specified
on polygons, polylines or points, depending on the required data.
The following hydrologic components are set over polygons:
Rainfall and Evaporation: accounts for spatially distributed rainfall and evaporation.
Infiltration: accounts for infiltration losses.
Wind: allows incorporating the effect of spatially distributed wind stress on the water
surface.
The following hydraulic components are set over polylines (feature arcs):
Bridges: account for general geometry bridges including pressure flow and overtopping.
Internal Rating Tables: provide an internal relationship of water elevation and discharge..
Gates: used to represent sluice gate structures.
Weirs: represent crested structures such as weirs, levees, sound walls, etc., where there is a
unique relationship between discharge and depth.
Hydraulic components that are entered on points are:
Bridge Piers: account for pier drag forces in a simplified formulation.
Culverts: one dimensional conveyance conduits where discharge can be calculated using
equations for circular or box structures, and rating tables.
Sources and Sinks: provide a mean to enter point inflows or outflows that may vary in
time.
55

7.1

Bridges Component

56

7.1 Bridges Component

RiverFlow2D provides several options to integrate bridge hydraulics into the 2D mesh calculations.
The most common option is to create the pier plan geometry generating a 2D triangular-cell mesh
that represents each pier as a solid obstacle. In that case, the model will compute the flow around
the pier and account for the pier drag. This would be the preferred approach when the user needs
to know the detailed flow around the piers, but it does not account for pressure flow or overtopping
conditions. In this option, the resulting mesh around piers has commonly very small cells which
can lead to increased computer times.
The Bridges component is a comprehensive bridge hydraulics computation tool that does not
require capturing bridge pier plan geometry in detail, therefore allowing longer time steps, while
allowing calculating the bridge hydraulics accounting for arbitrary plan alignment, complex bridge
geometry, free surface flow, pressure flow, overtopping, combined pressure flow and overtopping,
and submergence all in 2D.
This component requires defining the bridge alignment in plan and the bridge geometry cross
section. The bridge alignment is given in the .bridges data file which is generated by RiverFlow2D
model based on the user defined data in the RiverFlow2D Data Input Program. To run a simulation
with the bridges component, you need to select the option in the Control Data panel as shown in
Figure ??.

Figure 7.1 RiverFlow2D Control Data panel with Bridges component selected.

The bridge plan data is entered in the SMS conceptual model coverage. To create a bridge,
first make sure that the coverage where you defined the domain outline is active. Then, create an
arc that will define the bridge geometry in plan.
Now, draw a polyline clicking on each vertex until you reach the desired end when you can
double click to complete the polyline.

57

Hydraulic and Hydrologic Components

Figure 7.2 Bridge polyline.

Then select the polyline using the Select Feature Arch icon, and right click to redistribute the
vertices on the polyline. While the polyline is selected, right click on the arc and select Attributes...
and on the Feature arc Attributes dialog, select Boundary conditions and click Options...

Figure 7.3 Feature arc attributes.

On the Boundary Conditions Dialog, select the Bridges component and enter the desired
BridgeID (any text), and the bridge geometry file name that contains the bridge geometry data
and is explained in the next section.

Figure 7.4 Bridge dialog.

Since the bridge polyline must pass through nodes, it is essential that mesh generation engine
creates nodes along the bridge polyline. This is easily done recreating the mesh using the Feature
Objects/Map 2D Mesh command.

7.1

Bridges Component

58

Figure 7.5 Bridge polyline on mesh.

Note: There is no limit to the number of bridges that can be used.

7.1.1 Bridge Geometry Data File

The bridge geometry cross section file is necessary to define the bridge cross section. It is defined
by four polylines and the fined in five columns as follows:
Line 1: Number of points defining polylines.
NP
NP lines with these entries:
STATION(1) BEDELEV(1) ZLOWER(1) LOWCHORD(1) DECKELEV(1)
...
STATION(NP) BEDELEV(NP) ZLOWER(NP) LOWCHORD(NP) DECKELEV(NP)
.
The relationship between the four polylines must be as follows:
For all stations, STATION(I) STATION(I+1)
BEDELEV ZLOWER LOWCHORD DECKELEV
In a given line all elevations correspond to the same station.
The space between BEDELEV and ZLOWER is blocked to the flow.
The space between ZLOWER and LOWCHORD is open to the flow.
The space between LOWCHORD and DECKELEV is blocked to the flow.

7.1.1.1 Example of the Bridge Cross Section File

The following table is an example one of the geometry file that schematically represents the bridge
in ??.

7.1

Bridges Component

59

Figure 7.6 Front view of a bridge cross section.

7.1

Bridges Component

60

Table 7.1 Variable Descriptions for the bridge cross section geometry file.
VARIABLE

TYPE

RANGE

BEDELEV

m or ft

DECKELEV

m or ft

LOWCHORD

m or ft

NP

>1

STATION

m or ft

ZLOWER

m or ft

Note: R = Real variable. I = Integer

UNITS
DESCRIPTION
Bed elevation. Must be the lowest elevation for all polylines at a given point.
Elevation of the bridge deck. Must be
the highest elevation for all polylines at
a given point.
Elevation of the lower bridge deck.
LOWCHORD must be larger or equal
to ZLOWER and smaller or equal to
DECKELEV for a particular point.
The space between LOWCHORD and
DECELEV is a blocked area to the
flow.
Number of points defining cross section
polylines.
Distance from leftmost point defining
cross section polyline. All polylines
points must have a common station.
Elevation of lower polyline. ZLOWER
must be larger or equal to BEDELEV
and smaller or equal to LOWCHORD
for a given point. The space between
BEDELEV and ZLOWER is a blocked
area to the flow. The space between
ZLOWER and LOWCHORD is open
space. If the bridge has no holes,
ZLOWER must be identical to BEDELEV.
variable. S = Text variable.

7.1.2 Bridge Calculations

To model bridges, the source term in the dynamic equation is split in three terms S = Sz +Sf +Sb .
The term Sz defined as
Sz =


T
z
z
0, gh , gh
x
y

(7.1)

expresses the variation of the pressure force along the bottom in the x and y direction respectively,
formulated in terms of the bed slopes of the bottom level z. The term Sf
Sf =


T
f,x
f,y
0,
,

(7.2)

involves the the bed shear stresses f,x , f,y in the x and y direction respectively, with the density
of the fluid. The last term, Sb stands for local energy losses due to other processes
T

Sb = (0, ghSb,x , ghSb,y )

(7.3)

and will be used to represent bridges.

The description of energy losses for the friction term associated to the bed stress in equation
(??) is commonly formulated as an extension of a 1D formulation. The same approach is applied
in RiverFlow2D deriving 1D closure relations for the definition of the bridge source term. Note
that although the terms Sb,x and Sb,y represent energy losses in the presence of bridges, they are
actually acting as a momentum sink. Empirical models for the energy loss caused by the bridge
are described next.

61

Hydraulic and Hydrologic Components

Figure 7.7 Top view of a bridge showing the cross sections of interest. Only two piers are depicted
for simplicity.

7.1.2.1 Energy dissipation in bridges

The formulation of Borda-Carnot for energy loss in sudden contractions or expansions in pipes
can also be used for channels (?). This in turn can model bridges with free water surface. The
energy loss will be expressed in terms of the total available head HBC , and represents the total
mechanical energy of the flow. In a 1D framework the head loss HBC is expressed as follows
HBC = (Hc + He )

(7.4)

where Hc and He are the contraction and expansion losses respectively

Hc

v
12
2g

He

v
42
2g

i  2

2
A1
1 + 19 A
2


2
A4
+ 91
A3 1
h

1
m

(7.5)

where m is a typical value for the contraction coefficient, m = 0.62 (?) and the areas A1 to A4
refer to effective cross sectional flow area. The numbering of areas is shown in Figures ??. Area
1 is a section upstream of the bridge while area 4 is a downstream section. Areas 2 and 3 are
sections inside the bridge, near the entrance and exit respectively.

Figure 7.8 Simple example of A1 ,A2 , A3 and A4 used to calculate head loss in free surface bridges.

Figure 7.9 Simple example of A1 ,A2 , A3 and A4 used to calculate head loss in a partially submerged
bridges.

The values v1 and v4 are the cross sectional averaged velocities

Q1
Q4
,
v4 =
(7.6)
A1 (d1 )
A4 (d4 )
with Q1 and Q4 the total discharges in areas A1 and A4 , expressed as a function of the water
surface elevation, d = h + z. Different regimes can be described. Figure ?? shows a skecth of
the areas considered in the free surface case, Figure ?? shows the equivalent areas for partially
submerged bridges and Figure ?? for fully submerged bridges.
v1 =

7.1

Bridges Component

62

Figure 7.10 Simple example of A1 ,A2 , A3 and A4 used to calculate head loss in fully submerged
bridges.

7.1.2.2 Integration of the energy losses associated to bridges

The unified formulation of the source terms accounting for energy losses associated to bridges also
ensures the well balanced property in steady cases with velocity. In order to do that it is necessary
to define Snb at the edge of the RP where the bridge exists. The source term Snb is formulated as

0
h Hnx
(Snb )k = ge
(7.7)
ge
h Hny k
with
H = H

en
u
|e
u n|

(7.8)

where H is the singular loss term used to represent bridges.

Computation of H in a real mesh is done as follows. The bridge is defined on cell edges
(bold line in Figure ??), and the cells on both sides of these edges are considered to form two
cross sections L and R (hatched cells in Figure ??). Note that it is possible to define bridges in
arbitrary orientations and in structured/unstructured meshes.

Figure 7.11 Application of the scheme in triangular structured meshes. Normal bridge (left) and
oblique bridge (right).

In each time step, the necessary variables for the calculation of the global bridge head loss are
averaged from the cells in both upstream and downstream sections as illustrated in Figure ??. The
discharge is computed as
X
X
QL =
(qn)k lk
QR =
(qn)k lk
(7.9)
kL

kR

and the cross sectional average water level surface is estimated as

P
P
kL dk lk
kR dk lk
dL = P
dR = P
kL lk
kR lk

(7.10)

63

Hydraulic and Hydrologic Components

involving cells with values of h > 0. The signs of QL and QR are used to determine which section
is upstream and which downstream. If QL 0, the discharge across the bridge is computed as
Q = QL and the areas are computed using d1 = dL and d4 = dR . In case that QL < 0,
the discharge across the bridge is computed as Q = QR and the sections are reversed setting
d1 = dR and d4 = dL . Next, the different areas and the cross-sectional top width are calculated
as a function of the average water level surface. From these values the total head loss H can be
evaluated.

7.1.2.3 Influence of the Bridge Width

The computation algorithm used in the Bridges component neglects the effect of the structure width
(distance perpendicular to the bridge alignment) on the head loss. According to (?) and (?), the
bridge width has a small influence on the flow variables such as water surface elevation and energy
loss. Yarnell performed experiments in a laboratory flume with bridges having rectangular piers
with width-to-length ratios (w :l ) of 1:4, 1:7 and 1:13, where w is the pier dimension perpendicular
to the flow direction and l the pier length parallel to the flow. Yarnell noted that the energy
loss increased less than 10% for the configuration with longest piers. (?) performed numerical
simulations to confirm Yarnells experiments using piers with the same with-to-length ratios and
a wide range of approach discharges (see Figure ??).

Figure 7.12 Bridge pier proportions used to asses the influence of the structure width.

Numerical results indicate that the changes in total head loss across the structure are very
similar for the three configurations (see Figure ??).

Figure 7.13 Influence of the structure width on the total head change (H) across the bridge as
a function of the Froud number downstream.

7.2 Bridge Piers

The Bridge Piers component allows accounting for the losses caused by piers in the flow field in a
simplified way, without requiring a refined mesh around the actual pier plan geometry.

7.2

Bridge Piers

64

To run a simulation with the Bridge Piers Component, you need to select the option in the
Control Data panel of the RiverFlow2D Data Input Program as shown in Figure ??.

Figure 7.14 RiverFlow2D Control Panel dialog with Bridge Piers component selected.

Bridge pier data is entered in the SMS conceptual model coverage. To create a bridge pier,
first make sure that the Pier Component is active and then follow these steps:
Click on the Create Feature Point icon
the source

and click on the location where you want to enter

Now click on the Select Feature Point icon

step

and click on the point entered in the previous

On the Feature Objects menu, click Attributes...

Select Elem BC and click Options... button:

Select Pier, enter the Pier Name and other data as shown:

Note: There is no limit to the number of Bridge Piers that can be used.

65

Hydraulic and Hydrologic Components

Figure 7.15 Pier Data dialog.

7.2.1 Bridge Pier Calculation

The Bridge Pier component can be used when the pier plan area is small compared to the element
area and there is no need to determine the details of the flow field around the structure. In this
component the model computes the drag force on each pier as a function of the drag coefficient,
water density, flow velocity and wetted pier projected area as shown in Eq. ??:

FD =

1
CD U 2 AP
2

(7.11)

Where CD is the pier drag coefficient, is the water density, U is the water velocity, and AP
is the pier wetted area projected normal to the flow direction. Piers are assumed to be located on
elements that not necessarily conform to the pier geometry as shown on the following figure.

Figure 7.16 Piers inside cells.

Piers can be circular or rectangular in plan. Rectangular piers are located on elements based
on the pier center coordinates and the angle between the axis along the largest dimension and the
X-axis as shown in the following figure.
Velocity vector magnitude and approach angle usually varies in time during unsteady flow
computations and is used to calculate the projected area. To account for the resistance force that
the pier exerts on the flow, RiverFlow2D converts it to the distributed shear stress on the element
where the pier centroid coordinate is located. The resulting pier shear stress expressions in x and

7.3

Culverts Component

66

Figure 7.17 Schematic view of a rectangular pier.

y directions are as follows:

px =

p
1
AP
CD U U 2 + V 2
2
Ae

(7.12)

py =

p
AP
1
CD V U 2 + V 2
2
Ae

(7.13)

where Ae is the element area.

7.3 Culverts Component

The culvert component in RiverFlow2D allows incorporating 1D hydraulic structures that convey
water between two locations on the mesh, or between a point on the mesh an another outside.
To run a simulation with the Culverts Component, you need to select the option in the Control
Data panel of the RiverFlow2D Data Input Program dialog as shown in Figure ??.

Figure 7.18 RiverFlow2D Global Parameters dialog with Culverts Component selected.

The culvert data is entered in the SMS conceptual model coverage:

67

7
1. Click on the Create Feature Point icon
the inlet point.

Hydraulic and Hydrologic Components

and click on the location where you want to enter

2. Now click on the Select Feature Point icon

previous step.

and right click on the point entered in the

3. On the Feature Objects menu, click Attributes...

4. Select Elem BC and click Options... button:

5. Select Culvert, enter the Culvert ID name (e.g. Culvert1) and enter additional data depending on the culvert formulation desired.

6. Then select the Culvert Outlet Element and enter the Culvert Inlet Name:

7. Before running the model, make sure that Culverts it is selected in the Global Parameters
dialog.
There are two options to compute culvert discharge in RiverFlow2D. When the user selects
Rating Table calculation and provides a rating table on the associated file, the model determines
the discharge by interpolation as a function of the depth upstream. If the user enters Culvert
calculation using culvert characteristics, the model will calculate the discharge based on the culvert geometric characteristics given in the file. Both procedures are described in more detail below.

Note: There is no limit to the number of culverts that can be used.

7.3

Culverts Component

68

7.3.1 Culvert Calculation using a Rating Table (CulvertType = 0)

When the user provides a rating table, the culvert calculation algorithm is as follows:
1. If at least one of the culvert ends is wet, determine the flow direction based on the water
surface elevations at each culvert end,
2. Interpolate flow discharge from the rating table using the depth at the culvert inlet,
3. If depth at the culvert inlet is lower than minimum value in the rating table, then the
discharge is assumed to be zero.
4. If depth at entrance is higher than maximum value in the rating table, then the discharge is
assumed to be equal to that of the maximum depth.
5. The computed discharge is subtracted from the inlet element and added to the outlet element
assuming instantaneous water volume transmission.

7.3.2 Culvert Calculation using a Culvert Characteristics

(CulvertType = 1,2)
For CulvertTypes 1 and 2, the model will calculate culvert discharge for inlet and outlet control
using the FHWA procedure (Norman et al. 1985). Later Froehlich (2003) restated the algorithm
in dimensionless form. The resulting formula is expressed as follows:
Q = Nb Cc Ac

p
2gHc

(7.14)

where Nb is the number of identical barrels, Cc is a discharge coefficient that depend on the
flow control and culvert geometric characteristics, Ac is the culvert area at full section, g is the
gravitational acceleration, Hc = W SELh Zbi for inlet control and Hc = W SELh W SEtw
for outlet control, W SEh is the water surface elevation at the culvert inlet, Zbi is the inlet invert
elevation, W SEtw is the water elevation downstream (tailwater). For inlet control calculation,

r
Dc

1 H
(Y +mS0 )

h
f or Hh > 1.2Dc
2c0
Cc =
(7.15)
 ( M1 0.5)

Hh
0(1
f
or
H

1.2D
h
c
2K 1/M ) Dc
where Hh = W SELh Zbi is the headwater depth. Dc is the culvert diameter for circular culverts
and the base dimension for box culverts, m = 0.7 for mitered inlets and m = 0.5 for all other
inlets. For outlet control, the following formula is used to determine Cc :

Cc =

1 + Ke +

2gn2c Lc
4/3

0.5
(7.16)

Rc

where Rc is the culvert hydraulic radius, Ke is the entrance loss coefficient that can be obtained
from Table ??, nc is the Mannings n obtained from Table ??, Lc is the culvert length, and Y , K 0 ,
M 0 , C 0 are inlet control coefficients (see Table ??).

69

Hydraulic and Hydrologic Components

Table 7.2 Mannings n roughness coefficients for various culvert materials. Adapted from Froehlich
(2003).
CULVERT
BARREL MATERIAL
Concrete

Corrugated metal

ENTRANCE DESCRIPTION

MANNINGS N Nc

Good joints, smooth walls

Projecting from fill, square-cut end
Poor joints, rough walls
2-2/3 inch 1/2 inch corrugations
6 inch 1 inch corrugations
5 inch 1 inch corrugations
3 inch 1 inch corrugations
6 inch 2 inch corrugations
9 inch 2-1/2 inch corrugations

0.012
0.015
0.017
0.025
0.024
0.026
0.028
0.034
0.035

Table 7.3 Entrance loss coefficients Ke . Adapted from Froehlich (2003).

TYPE
VERT

OF

Concrete pipe

CUL-

ENTRANCE DESCRIPTION*

Projecting from fill, grooved end

Projecting from fill, square-cut end
Headwall or headwall with wingwalls (concrete or cement sandbags)
Grooved pipe end
Square-cut pipe end
Rounded pipe end
Mitered end that conforms to embankment
slope
Manufactured end section of metal or concrete that conforms to embankment slope
Without grate
With grate
Corrugated metal pipe
Projecting from embankment (no heador pipe-arch
wall)
Headwall with or without wingwalls (concrete or cement sandbags)
Mitered end that conforms to embankment
slope
Manufactured end section of metal or concrete that conforms to embankment slope
Without grate
With grate
Reinforced
concrete
Headwall parallel to embankment (no
box
wingwalls)
Square-edged on three sides
Rounded on three sides to radius of
1/12 of barrel dimension
Wingwalls at 30 to 75 to barrel
Square-edged at crown
Crown edge rounded to radius of 1/12
of barrel dimension
Wingwalls at 10 to 30 to barrel
Square-edged at crown
Wingwalls parallel to embankment
Square-edged at crown
*See Table ?? for inlet configurations.

ENTRANCE
LOSS COEFFICIENT Ke
0.2
0.5

0.2
0.5
0.1
0.7

0.5
0.7
0.9
0.5
0.7

0.5
0.7

0.5
0.2

0.4
0.2

0.5
0.7

7.3

Culverts Component

70

Table 7.4 Culvert inlet control formula coefficients. Adapted from Froehlich (2003).
BARREL BARREL
MATESHAPE
RIAL
Concrete
Circular
Concrete
Circular
Concrete

Circular

Cor. metal
Cor. metal
Cor. metal
Concrete
Concrete

Circular
Circular
Circular
Circular
Circular

Concrete

Rectangular

Concrete

Rectangular

Concrete

Rectangular

Concrete

Rectangular

Concrete

Rectangular

Concrete

Rectangular

Concrete
Concrete
Concrete

Rectangular
Rectangular
Rectangular

Concrete

Rectangular

Concrete

Rectangular

Concrete

Rectangular

Concrete

Rectangular

Concrete

Rectangular

Concrete

Rectangular

Concrete

Rectangular

Concrete

Rectangular

Concrete

Rectangular

Cor. metal
Cor. metal
Cor. metal
Concrete
Cor. metal
Concrete
Concrete

Rectangular
Rectangular
Rectangular
Circular
Circular
Rectangular
Circular

INLET DESCRIPTION*
Headwall; square edge
Headwall;
grooved
edge
Projecting;
grooved
edge
Headwall
Mitered to slope
Projecting
Beveled ring; 45 bevels
Beveled
ring;
33.7 bevels
Wingwalls;
30 to
75 flares; square edge
Wingwalls;
90 and
15 flares; square edge
Wingwalls;
0 flares;
square edge
Wingwalls;
45 flare;
beveled edge
Wingwalls;
18 to
33.7 flare;
beveled
edge
Headwall; 3/4 inch
chamfers
Headwall; 45 bevels
Headwall; 33.7 bevels
Headwall;
45 skew;
3/4 in chamfers
Headwall;
30 skew;
3/4 in chamfers
Headwall;
15 skew;
3/4 in chamfers
Headwall; 10-45 skew;
45 bevels
Wingwalls; non-offset
45 /flares;
Wingwalls; non-offset
18.4 /flares; 3/4 in
chamfers
Wingwalls;
nonoffset
18.4 /flares;
30 /skewed barrel
Wingwalls;
offset
45 /flares; beveled top
edge
Wingwalls;
offset
33.7 /flares;
beveled
top edge
Wingwalls;
offset
18.4 /flares; top edge
bevel
Headwall
Projecting; thick wall
Projecting; thin wall
Tapered throat
Tapered throat
Tapered throat
Headwall; square edge

0.3153
0.2509

2.0000
2.0000

1.2804
0.9394

0.6700
0.7400

0.1448

2.0000

1.0198

0.6900

0.2509
0.2112
0.4593
0.1379
0.1379

2.0000
1.3300
1.5000
2.5000
2.5000

1.2192
1.4895
1.7790
0.9651
0.7817

0.6900
0.7500
0.5400
0.7400
0.8300

0.1475

1.0000

1.2385

0.8100

0.2242

0.7500

1.2868

0.8000

0.2242

0.7500

1.3608

0.8200

1.6230

0.6670

0.9941

0.8000

1.5466

0.6670

0.8010

0.8300

1.6389

0.6670

1.2064

0.7900

1.5752
1.5466
1.6611

0.6670
0.6670
0.6670

1.0101
0.8107
1.2932

0.8200
0.8650
0.7300

1.6961

0.6670

1.3672

0.7050

.7343

0.6670

1.4493

0.6800

1.5848

0.6670

1.0520

0.7500

1.5816

0.6670

1.0906

0.8030

1.5689

0.6670

1.1613

0.8060

1.5752

0.6670

1.2418

0.7100

1.5816

0.6670

0.9715

0.8350

1.5752

0.6670

0.8107

0.8810

1.5689

0.6670

0.7303

0.8870

0.2670
0.3023
0.4593
1.3991
1.5760
1.5116
0.3153

2.0000 1.2192 0.6900

1.7500 1.3479 0.6400
1.5000 1.5956 0.5700
0.5550 0.6305 0.8900
0.6400 0.9297 0.9000
0.6670 0.5758 0.9700
2.0000 1.2804 0.6700
Continued on next page

71

7
Table 7.4 continued from previous page
BARREL
BARREL
INLET DESCRIPMATESHAPE
TION*
RIAL
Concrete
Circular
Headwall;
grooved
edge
Concrete
Circular
Projecting;
grooved
edge
Cor. metal
Circular
Headwall
Cor. metal
Circular
Mitered to slope
Cor. metal
Circular
Projecting
Concrete
Circular
Beveled ring; 45 bevels
Concrete
Circular
Beveled
ring;
33.7 bevels
Concrete
Rectangular Wingwalls;
30 to75 flares; square
edge
Concrete
Rectangular Wingwalls;
90 and
15 flares; square edge
Concrete
Rectangular Wingwalls;
0 flares;
square edge
Concrete
Rectangular Wingwalls;
45 flare;
beveled edge
Concrete
Rectangular Wingwalls;
18 to
33.7 flare;
beveled
edge 1.5466
Concrete
Rectangular Headwall; 3/4 inch
chamfers
Concrete
Rectangular Headwall; 45 bevels
*See Table ?? for inlet configurations.

Hydraulic and Hydrologic Components

0.2509

2.0000

0.9394

0.7400

0.1448

2.0000

1.0198

0.6900

0.2509
0.2112
0.4593
0.1379
0.1379

2.0000
1.3300
1.5000
2.5000
2.5000

1.2192
1.4895
1.7790
0.9651
0.7817

0.6900
0.7500
0.5400
0.7400
0.8300

0.1475

1.0000

1.2385

0.8100

0.2242

0.7500

1.2868

0.8000

0.2242

0.7500

1.3608

0.8200

1.6230

0.6670

0.9941

0.8000

0.6670

0.8010

0.8300

1.6389

0.6670

1.2064

0.7900

1.5752

0.6670

1.0101

0.8200

Table 7.5 Culvert inlet configurations. Adapted from www.xmswiki.com/xms/.

INLET CONFIGURATION

DESCRIPTION

End of the culvert barrel projects out of the embankment.

Grooved pipe for concrete culverts decreases energy

losses through the culvert entrance.

This option is for concrete pipe culverts.

Continued on next page

7.3

Culverts Component

72

Table 7.5 continued from previous page

INLET CONFIGURATION
DESCRIPTION

Square edge with headwall is an entrance condition

where the culvert entrance is flush with the headwall.

Beveled edges is a tapered inlet edge that decreases

head loss as flow enters the culvert barrel.

Mitered entrance is when the culvert barrel is cut so

it is flush with the embankment slope.

Wingwalls are used when the culvert is shorter than

the embankment and prevents embankment material
from falling into the culvert

The culvert computation algorithm works as follows:

1. If at least one of the culvert ends is wet, Determine the flow direction based on the water
surface elevations at each culvert end.
2. Compute the culvert discharge using inlet control formulas.
3. Compute the culvert discharge using outlet control formulas.
4. Select the minimum discharge from the inlet and outlet control discharges.
5. If depth at the culvert inlet is lower than minimum value in the rating table, then the
discharge is assumed to be zero.
6. The computed discharge is subtracted from the inlet element and added to the outlet element
assuming instantaneous water volume transmission.

When using CulvertType 1 or 2, both ends of the culvert must be inside the mesh.

73

Hydraulic and Hydrologic Components

7.3.3 Assumptions of Culvert Calculations

1. The same rating table will be used to interpolate discharge regardless of the flow direction. In
other words, if the flow is from element A to element B at some point during the simulation,
depth at A will be used to interpolate discharge from A to B, but if at some other time flow
changes from B to A, discharge will be interpolated using depth at B.
2. There is no outlet control on the rating table discharge calculation.
3. When using CulvertTypes 1 and 2, both ends of the culvert must be inside the mesh. It is
not allowed to extract flow from the modeling domain when using these options.
4. Discharge calculation with CulverTypes 1 and 2 is only available for circular or box (rectangular) cross section culverts.
5. The entrance to a culvert is regarded as submerged when the head water depth, H, 1.2D,
where D is the diameter of the circular culvert or the height of box culverts.

7.4 Gates Component

The GATES components allows integrating gates inside the modeling region. Each gate needs to
be defined in terms of its plan alignment, crest elevation (Zc), gate height (Hgate) and the time
history of apertures (Ha) given as a table in a file associated to each structure (see Figure ??).
Figure ?? shows the flow modes that can be calculated through gates that include submergence
and overtopping.

Figure 7.19 Schematic cut view perpendicular to a gate structure.

7.4

Gates Component

74

Figure 7.20 Flow modes across gates.

To run a simulation with the gates component, you need to select the option in the Control
Data panel of the RiverFlow2D Data Input Program as shown in Figure ??.

Figure 7.21 RiverFlow2D Control Data panel with Gates Component selected.

The gate plan data is entered in the SMS conceptual model coverage. First create an arc that
will define the gate geometry in plan. Draw a polyline clicking on each vertex until you reach the
desired end when you can double click to complete the polyline.

75

Hydraulic and Hydrologic Components

Figure 7.22 Gate polyline.

Then select the polyline using the Select Feature Arch icon, and right click to redistribute the
vertices on the polyline.
While the polyline is selected, right-click on the arc and select Attributes... and on the Feature
arc Attributes dialog, select Boundary conditions and click Options...

Figure 7.23 Feature arc attributes.

On the Nodestring Boundary Conditions dialog, select the Gates component and enter the
desired Gate ID (any text), and gate file name that contains the aperture time series, the crest
elevation, crest height and gate Cd.

Figure 7.24 Gates dialog.

Since the gate polyline must pass through nodes, it is essential that the mesh generation engine creates nodes along the polyline. This is easily done recreating the mesh using the Feature
Objects/Map 2D Mesh command.

Note: There is no limit to the number of gates that can be used.

7.5

Rainfall, Evaporation and Infiltration

76

7.4.1 Gate Calculations

The gate is simulated by assuming that the discharge per unit breadth q crossing the gate is
governed by the difference between the water surface level (d = h + z) on both sides of the gate,
referred to as dl upstream of the gate and dr downstream of the gate, and by the allowable gate
opening, Go . Several situations are envisaged. In the case that Go = 0 the gate behaves as a solid
wall and no flow crosses the gate. When the gate opening is larger than the surface water level on
both sides, it no longer influences the flow. In any other case, assuming that dl < dr , without loss
of generality, two different flow situations can occur depending on the relative values of Go , zl , zr ,
dl and dr . When Go + max(zl , zr ) < min(dl , dr ), Figure ??, the discharge is given by
q = Go K1 (dr dl )1/2

(7.17)

with K1 an energy loss coefficient. In RiverFlow2D K1 =3.33 (?).

When Go + max(zl , zr ) > min(dl , dr ), Figure ??, the discharge is given by
1/2

q = Go K2 [dr max(zl , zr )]

(7.18)

with K2 another energy loss coefficient. In RiverFlow2D K2 =2.25 (?).

Figure 7.25 Water levels for discharge under a gate in submerged conditions formulated as in (G1).

Figure 7.26 Water levels for discharge under a gate in submerged conditions formulated as in (G2).

7.5 Rainfall, Evaporation and Infiltration

This section describes the Hydrologic computations implemented in the RiverFlow2D model. The
component includes capabilities to integrate rainfall, evaporation and infiltration in the model
simulations.

77

Hydraulic and Hydrologic Components

7.5.1 Rainfall
Rainfall is treated as a simple source term. It represents an additional input to the cell water depth
in the previous step of flow calculation. You can set up local rainfall events for several regions of
the watershed. This allows to simulate more realistic cases, in which data from several rain gauges
are available.
Notice that the rainfall is set before the infiltration calculation. This is an important detail because
the infiltration capacity strongly depends on the rainfall intensity, as we will see in the next section.

7.5.2 Infiltration
Infiltration represents another component of the hydrological budget and it can be defined as the
process by which surface water enters the soil. In the RiverFlow2D model, infiltration is treated as
a loss. This process is mainly governed by two forces: gravity and capillarity action. The model
includes three methods to compute the infiltration losses: Horton, Green & Ampt and SCS-Curve
Number.

7.5.2.1 Horton Infiltration Model

Hortons infiltration model (?) suggests an exponential equation (??) for modelling the soil infiltration capacity fp :
fp = fc + (f0 fc ) ekt

(7.19)

where f0 and fc are the initial and final infiltration capacities, both measured in m/s or in/s and
k represents the rate of decrease in the capacity (1/s).
The parameters f0 and k have no physical basis, so they must be determinated from experimental data. A good source for experimental values of these parameters for different types of
soils can be found in (?) and summarized in (?). Table ?? and Table ?? show the parameters
for some general types of soil, as presented in (?). Note that no k values are shown. A value of
k = 4.14 hr1 is recommended in the absence of any field data (?).
Table 7.6 Horton initial infiltration for different soils. Source: Akan(1993).
SOIL TYPE
f0 (mm/hr)
Dry sandy soils with few to no vegetation
Dry loam soils with few to no vegetation
Dry clay soils with few to no vegetation
Dry sandy soils with dense vegetation
Dry loam soils with dense vegetation
Dry clay soils with dense vegetation
Moist sandy soils with few to no vegetation
Moist loam soils with few to no vegetation
Moist clay soils with few to no vegetation
Moist sandy soils with dense vegetation
Moist loam soils with dense vegetation
Moist clay soils with dense vegetation

127
76.2
25.4
254
152.4
50.8
43.18
25.4
7.62
83.82
50.8
17.78

7.5

Rainfall, Evaporation and Infiltration

78

Table 7.7 Horton final infiltration for different soils. Source: Akan(1993).
SOIL TYPE
fc (mm/hr)
Clay loam, silty clay loams
127
Sandy clay loam
1.3 - 3.8
Silt loam, loam
3.8 - 7.6
Sand, loamy sand, sandy loams
7.6 - 11.4

The equation (??) has to be applied after the surface ponding. In other words, we are assuming
conditions of unlimited water supply at the surface. Under this consideration, the cumulative
infiltration up to time t can be calculated by integrating the infiltration capacity:
Z t


f0 fc
fp (t)dt = fc t +
F =
1 ekt
(7.20)
k
0
It is important to highlight the difference between the infiltration capacity fp and the infiltration
rate f. If we consider a rain event starting with a weak rainfall intensity (R fp ), then all the rain
will be infiltrated into the soil. On the other hand, if the rain exceeds the soil infiltration capacity
or if the surface becomes ponded, this magnitude will determine the infiltration rate:
R fp f = R

R > fp , t > t p f = f p

(7.21)

where tp represents the ponding time.

Following (?), for the two first rain intervals, the rainfall intensity is less than the infiltration
capacity, so the real infiltration rate is equal to the rainfall rate. Because of this fact, the actual
infiltration capacity does not decay as given by Hortons equation. The reason, as indicated above,
is the Hortons model assumption of water supply always exceeding the infiltration capacity from
the beginning. Hence, the soil has more infiltration capacity and we have to compute the real
infiltration at t=20 min, so we need to determine the ponding time tp by solving (??):
Z tp


f0 f c
F =
R(t)dt = fc tp +
1 ektp
(7.22)
k
0
where F stands for the cumulative infiltration (that is equal to the rainfall volume) until this
ponding time.
The above equation needs to be solved by an iterative procedure, for instance the NewtonRaphson method. Thus, the infiltration capacity is now a function of the actual infiltrated water,
not just a function of time. Finally, the real infiltration capacity at t=20 min is calculated by
evaluating (??) at tp :
fp = fc + (f0 fc ) ektp

(7.23)

When rainfall intensity exceeds the soil infiltration capacity, the real infiltration rate is equal
to this capacity and decays following Hortons equation by replacing f c = f p and t = t t0 , being
t0 at which the rainfall intensity exceeds the soil infiltration capacity:
0

f = fc + (fp fc ) ek(tt )

(7.24)

An additional consideration must be taken into account. It is possible that the recalculated
infiltration capacity will be greater than the rainfall intensity. This implies a non-physical situation
with negative storage or run-off. The reason for this behaviour is that the soil cannot infiltrate
more than the rainfall rate, so a limit in the recalculated infiltration capacity must be imposed:
fp R

(7.25)

79

Hydraulic and Hydrologic Components

7.5.2.2 Green-Ampt Infiltration Model

The infiltration Green-Ampt model is a simple model with a theoretical base on Darcys law, so it is
not strictly empirical. Moreover, its parameters have physical meaning and they can be computed
from soil properties. The most common soil parameters are shown in Table ??, as presented in
(?).
Table 7.8 Mean values and standard deviation for Green-Ampt model parameters. Source: Rawls
et al. (1983).
SOIL TYPE
s
Sf (mm/hr)
Ks (mm/hr)
Sand
0.437(0.374-0.500)
4.95(0.97-25.36)
Loamy sand
0.437(0.363-0.506)
6.13(1.35-27.94)
Sandy loam
0.453(0.351-0.565)
11.01(2.67-45.47)
Loam
0.463(0.375-0.551)
8.89(1.33-59.38)
Silt loam
0.501(0.420-0.582)
16.68(2.92-95.39)
Sandy clay loam
0.398(0.332-0.464)
21.85(4.42-108.0)
Clay loam
0.464(0.409-0.519)
20.88(4.79-91.10)
Silty clay loam
0.471(0.418-0.524)
27.30(5.67-131.50)
Sandy clay
0.430(0.370-0.490)
23.90(4.08-140.2)
Silty clay
0.479(0.425-0.533)
29.22(6.13-139.4)

11.78
2.99
1.09
0.66
0.34
0.15
0.10
0.10
0.06
0.05

The original Green-Ampt model starts from the assumption that a ponding depth h is maintained over the surface. The Green-Ampt method approximates the soil infiltration capacity as
follows:
Ks (s i ) Sf
(7.26)
F
being Ks the effective hydraulic conductivity, Sf the suction head at the wetting front, i the
initial uniform water content and s the porosity. The integration of (??) provides the cumulative
infiltration:


dF
F
fp =
= Ks t = F (s i ) Sf ln 1 +
(7.27)
dt
(s i ) Sf
Solving for the cumulative infiltration F in equation (??) requires an iteration procedure (e.g.
Picard iterations or Newton-Rhapson method). The effective suction head can be replaced by the
average value (?).
Equations (??) and (??) assume that the soil is ponded from the beginning. Additional considerations should be taken into account in order to model an unsteady storm pattern (?). Three
posibilities can occur in every timestep: 1) ponding occurs at the beginning of the interval; 2) there
is no ponding within the interval; 3) ponding occurs within the interval. The first step consists of
computing the actual infiltration capacity fp from the known value of the cumulative infiltration
F at time t. From (??):



+1
(7.28)
fp = K s
F
The result from eq. (??) is compared with the rainfall intensity i. If fp i, case 1 occurs and
the cumulative infiltration at the end of the interval is given by (??). Moreover, the real infiltration
f rate will be equal to the potential one fp i:


Ft+t +
Ft+t F ln
= K
(7.29)
F +
fp = Ks +

7.5

Rainfall, Evaporation and Infiltration

80

If fp > i, there is no ponding at the beginning of the interval. We assume that there is no
ponding during the entire interval, so the real infiltration rate is equal to the rain rate and a
tentative value for the cumulative infiltration at the end of the period can be computed as:
0
Ft+t
= F + it.

(7.30)
0
fp,t+t

From equations (??) and (??) a tentative infiltration capacity

can be calculated. If
0
fp,t+t
> i, there is no ponding during the interval, the assumption is correct and the problem
0
0
corresponds to situation number 2, so Ft+t
= Ft+t . If fp,t+t
i, there are ponding condition
within the interval (case 3). The cumulative infiltration at ponding time Fp is found by taking
fp = i and F = Fp at (??):
Fp =

Ks
i Ks

(7.31)

Then, the ponding time is computed as t + t0 , where:

Fp F
(7.32)
i
Finally, the cumulative infiltration can be found by replacing F = Fp and t = t t0 in
equation (??).
t0 =

7.5.2.3 SCS-CN Runoff Model

The Soil Conservation Service-Curve Number (SCS-CN) runoff model was originally developed by
the USDA Natural Resources Conservation Service for estimating runoff from rainfall events on
agricultural watersheds (?). Nowadays it is also used for urban hydrology. The main parameter
of the method is the Curve Number (CN) which is essentially a coefficient for reducing the total
precipitation to runoff or surface water potential, by taking into account the losses (evaporation,
absorption, transpiration and surface storage). In general terms, the higher the CN value the
higher the runoff potential. The main equation of the SCS-CN method is:
(
(RV Ia )2
(RV > Ia )
RO = RV Ia +S
(7.33)
0
(RV Ia )
where RV =Runoff of effective precipitation, RV =Rain Volume, Ia =Initial abstraction which infiltrates before runoff begins, and S =Potential maximum retention, estimated by means of the
Curve Number:
25400
254
(7.34)
CN
As seen on Eq. ??, runoff cannot begin until initial abstraction has been met. Initial abstraction
is written in terms of its ratio and S :
S=

Ia = S

(7.35)

The initial abstraction ratio was traditionally taken as = 0.2 for every watershed (?) but
recent studies suggest that there is a wide range of values that work better than this value,
depending on the soil properties. The influence of this parameter was widely studied in Caviedes
et al. (?).
It is important to remark that SCS-CN method was not designed to consider time. Following
(?), when the method is implemented in a complex simulator, a time-advancing methodology is
used. The method is not applied to the entire catchment. Runoff is calculated for every cell in
every time step, using the cumulative rainfall since the beginning of the storm.

81

Hydraulic and Hydrologic Components

For the determination of the Curve Number the guidelines provided in (?) should be followed.
Another important fact that needs to be considered is the Antecedent Moisture Content (AMC) or
Antecedent Runoff Condition (ARC), which represents the preceding relative moisture of the soil
prior to the storm event (?). This parameters result useful in order to attempt to account for the
CN variation among storm events. Three posible assumptions can be considered: dry conditions
(AMC I), average conditions (AMC II) or wet conditions (AMC III) (see Table ??).
Table 7.9 Antecedent Moisture Content groups (adaptation from Mishra
). et al. (2003)
SOIL AMC

I
II
III

Total
5-day
rainfall
(dormant season)
Less than 13 mm
13 mm to 28 mm
More than 28 mm

Total
5-day
rainfall (growing
season)
Less than 36 mm
36 mm to 53 mm
More than 53 mm

Traditionally (?), the Curve Number for dry or wet conditions has been recalculated in terms
of the standard conditions according to Eqs. ?? and ??:
CN (I) =

4.2CN (II)
10 0.058CN (II)

(7.36)

23CN (II)
10 0.13CN (II)

(7.37)

CN (III) =

On the other hand, some newer references (?, ?) recommend to use an empirical data table to
compute both values.

7.5.3 Entering Rainfall, Evaporation and Infiltration Data

To enter data for a simulation considering rainfall, evaporation and infiltration simulation use the
Rainfall/Infiltration panel in the RiverFlow2D Data Input Program.
You may use this component to make hydrologic simulations with the following options:
Rainfall and evaporation and impermeable soil (no infiltration)
Infiltration (no rainfall)
Rainfall, evaporation and infiltration.

In the infiltration without rainfall option only the Horton or the Green and Amp methods can be
used since they take into account the water depth to determine the infiltration rates. The SCS-CN
method calculates infiltration as a function of the given rainfall and does not consider the pounded
water.
Please follow the Rainfall, Evaporation and Infiltration Tutorial that explains how to setup a
hydrologic simulation from start to finish.

7.6

Internal Rating Tables

82

Figure 7.27 RiverFlow2D Rainfall/Infiltration panel.

7.6 Internal Rating Tables

Internal Rating Tables is an internal condition along a polyline where the model imposes the
interpolated water elevation from the calculated discharge from a user provided rating table.
If the rating table is not fully compatible with the computed 2D flow, results can be erroneous. It
is suggested to use this condition with care to avoid inconsistencies.
To run a simulation with Internal Rating Tables, you need to select the option in the Control
Data panel of the RiverFlow2D Data Input Programas shown in Figure ??.
Internal Rating Table (IRT) plan data is entered in the SMS conceptual model coverage. To
create a IRT polyline, draw a polyline clicking on each vertex until you reach the desired end when
you can double click to complete the polyline.
Then select the polyline using the Select Feature Arch icon, and right click to redistribute the
vertices on the polyline. While the polyline is selected, right click on the arc and select Attributes...
and on the Feature arc Attributes dialog, select Boundary conditions and click Options...
On the Nodestring Boundary Conditions dialog, select the Interior Boundary Condition component and enter the desired IRT ID (any text), and IRT file name that contains of water elevation
vs discharge table.

7.6

Internal Rating Tables

83

Figure 7.28 RiverFlow2D Control Data panel with IRT component selected.

Figure 7.29 Internal Rating Table polyline.

Figure 7.30 Feature arc attributes.

84

Hydraulic and Hydrologic Components

Figure 7.31 Internal Rating Table dialog.

85

Hydraulic and Hydrologic Components

Enter the desired IRT (any text), and click Define... to enter the rating table. The following
dialog pops up and the Internal Rating Data (Discharge vs. Water Surface Elevation) can be
entered.

Figure 7.32 Window to enter data pairs (discharge-WSE) for an Internal Rating Table.

Since the IRT must pass through nodes, it is essential that mesh generation engine creates nodes
along the IRT nodestring. This is easily done recreating the mesh using the Feature Objects/Map
2D Mesh command. The resulting mesh nodes line along the IRT polyline as shown in Figure
??.

Figure 7.33 Internal Rating Table dialog.

Note: There is no limit to the number of Internal Rating Tables that can be used.

7.6.1 Internal Rating Table Calculations

An internal rating table is implemented as a set of values of total discharge in terms of the water
surface level Q = Q(h + z). This table is defined along a polyline in the mesh. First, a common
average water surface level is computed considering all the upstream cells along the polyline. Then,
the discharge is imposed at the cells sharing the edges on both sides in the polyline according to
the common upstream water surface level and following the internal rating table.
The IRT calculation algorithm works as follows:
1. For each calculation time interval, estimate an average water surface level at each side of the
IRT polyline.
2. Compute the discharge passing through the IRT polyline from the average water levels in 1
using the rating table.
3. Define an average velocity from the discharge and the cross sectional wetted area.

7.7

Sources and Sinks

86

4. Assign a common unit discharge to every pair of cells sharing a polyline segment.

Some inappropriate IRT polyline configurations or very long polylines can over-constrain the model
and should be avoided.

7.6.2 Assumptions of Internal Rating Table Calculations

The rating table does not account for outlet control.

7.7 Sources and Sinks

Sources and Sinks component allows accounting point inflows (source) or outflows (sink) of water
on the mesh. This allows simulating for example water intakes at any location on the mesh.
To run a simulation with Sources or Sinks, you need to select the option in the Control Data
panel of the RiverFlow2D Data Input Program as shown in Figure ??.

Figure 7.34 RiverFlow2D Control Data panel with the Sources and Sinks component selected.

Sources and Sinks data is entered in the SMS conceptual model coverage. To create a Source/Sink
follow these steps:
Then, click on the Create Feature Point icon
enter the source
Now click on the Select Feature Point icon
step

and click on the location where you want to

and click on the point entered in the previous

On the Feature Objects menu, click Attributes...

87

Hydraulic and Hydrologic Components

Select Elem BC and click Options... button:

Select Source/Sink, enter the Source/Sink ID name (e.g. SOURCE1) and click Define...

Figure 7.35 Source/Sink dialog.

Enter the discharge time series (unsteady) or a single discharge (steady flow).

Figure 7.36 Time series editor to enter discharge for a Source/Sink.

Note: There is no limit to the number of sources and sinks that can be used.

7.8 Weirs
RiverFlow2D Weirs component may be convenient when trying to simulate levee or road overtopping. The tool allows defining a polyline representing the structure alignment and assigning crest

7.8

Weirs

88

elevations that can vary along the polyline.

To run a simulation with weirs, you need to select the option in the Control Data panel of the
RiverFlow2D Data Input Program as shown in Figure ??.

Figure 7.37 RiverFlow2D Control Data panel with Weirs component selected.

Weir plan data is entered in the SMS conceptual model coverage. To create a weir, first make
sure that the conceptual model coverage where you defined the domain outline is active. Then,
create an arc that will define the weir geometry in plan. Now, draw a polyline clicking on each
vertex until you reach the desired end when you can double click to complete the polyline.

Figure 7.38 Weir polyline.

Then select the polyline using the Select Feature Arch icon, and right click to redistribute the
vertices on the polyline. While the polyline is selected, right click on the arc and select Attributes...
and on the Feature arc Attributes dialog, select Boundary conditions and click Options...

89

Hydraulic and Hydrologic Components

Figure 7.39 Feature arc attributes.

On the Nodestring Boundary Conditions dialog, select the Interior Boundary Condition component and enter the desired Weir name, crest elevation and weir coefficient Cf . This version of
the software considers variable crest elevation along the weir.

Figure 7.40 Weirs data dialog.

Since RiverFlow2D requires that the weir passes through nodes, it is essential that mesh generation engine creates nodes along the weir polyline. To achieve this it is necessary that each weir
polyline coincides exactly with corresponding arcs on SMS Conceptual Model.

Note: There is no limit to the number of weirs that can be used.

7.8.1 Weir Calculations

The weir calculation algorithm works as follows:
1. For each calculation time interval, the model checks for each segment defined by two pair of
opposing cells (L, R) along the weir that at least one of the opposite cells is wet and that its
water surface elevation is above the crest elevation.
2. Then the model calculates the water elevation at each weir segment as:
dw = hcrest + M AX (zL , zR )
where hcrest is the crest elevation and dw the segment water elevation.

(7.38)

7.8

Weirs

90

3. When the water surface levels on both sides is below the weir level, M AX (dL , dR ) dw ,
the velocity component normal to the weir segment direction is set to zero.
4. Otherwise the model calculates the normal discharge for the segment according to the water
levels on both sides.
5. The discharge is imposed on both the and cells.
The weir is modelled by assuming that the discharge per unit breadth q crossing the weir is
governed by the difference between the water surface level (d = h + z) on both sides of the weir,
referred to as dl upstream and dr downstream of the weir, and by the weir crest elevation, Hw .
Several situations are accounted for. In the case that both water elevations are below the weir
crest elevation the weir behaves as a solid wall and no flow crosses it. When dl < dr , without loss
of generality, two different flow situations can occur depending on the relative values of Hw , zl , zr ,
dl and dr . When Hw + max (zl , zr ) < min (dl , dr ), the discharge is given by
3/2

q = Cf (dr dl )

(7.39)

with Cf the discharge coefficient.

When Hw + max (zl , zr ) > min (dl , dr ), the discharge is given by
q = Cf (dr Hw )

3/2

(7.40)

7.8.2 Assumptions of Weir Calculations

1. The weir crest elevation may vary along the weir but must be higher than both cells opposing
each weir segment.

RiverFlow2D Data Input Program (DIP)

Data Input Program (DIP) user interface provides the tools to enter non-spatial data, and run
RiverFlow2D model. All changes introduced on the DIP will be saved to the native RiverFlow2D
data files not to the .2dm file.
The RiverFlow2D DIP appears when you use the Run RiverFlow2D command from SMS RiverFlow2D menu. You can also access the DIP double-clicking on the RiverFlow2D icon
on the
desktop. In that case, the program will give you a list of previous projects and let you open any
one of them:

Figure 8.1 RiverFlow2D DIP Open Project Dialog.

RiverFlow2D DIP provides a data input environment with panels that include all the nonspatial options to run the RiverFlow2D model. The left column on the main window allows you
select modules, components, output options, etc. When you click on one of the elements, the
appropriate right side panel is activated. Each panel contains the data corresponding to each of
the RiverFlow2D data files. For example, the Control Data Panel has all the data of the .DAT file.

91

8.1

Control Data Panel (.DAT files)

92

Figure 8.2 Main RiverFlow2D Data Input Program window.

RiverFlow2D DIP lets you select different model engines. Clicking on Model Selection opens the
dialog where you can select between RiverFlow2D Plus GPU, RiverFlow2D Plus or RiverFlow2D
FE. Note that to run the GPU version you need a special license as well as the appropriate GPU
hardware.

Figure 8.3 RiverFlow2D Model Selection Dialog.

The following sections describe the panel dialogs of RiverFlow2D DIP.

8.1 Control Data Panel (.DAT files)

This panel determines the general run options like time step control parameters, metric or English
units, physical process (components), graphical outputs, and initial conditions. It also provides
buttons to open and saving files, and running the RiverFlow2D model. The program will launch
with the Control Data panel visible.

93

RiverFlow2D Data Input Program (DIP)

Figure 8.4 Control Data panel.

Table 8.1 Buttons in the Control Data Panel.
BUTTON NAME
DESCRIPTION
Open .DAT
Opens an existing .DAT file.
Save .DAT
Saves a .DAT file with the data shown on the Panel.
Buttons on Bottom Row Duplicated on Every Panel.
Open All Files
Saves data from all enabled Tabs. Note: This operation
does not alter the node coordinates and elevations, triangular mesh topology, Manning roughness coefficients, and
other mesh related parameters in the .FED file.
Run RiverFlow2D
Runs the RiverFlow2D model.
Exit
Closes the RiverFlow2D Data Input Program.
About Hydronia RiverFlow2D Shows a concise description of RiverFlow2D.
?
Opens the RiverFlow2D Users Guide.
www.hydronia.com
Opens browser to Hydronia home page.

8.1

Control Data Panel (.DAT files)

94

Table 8.2 Time Control Data Frame on the Control Data Panel.
CONTROL NAME
Simulation time (hrs.)
Output Interval (hrs.)
CFL (RiverFlow2D Plus and
RiverFlow2D Plus GPU only)

Time
step
(sec.)
(RiverFlow2D FE only)

DESCRIPTION
Total simulation time in hours.
Time interval for output reporting.
Courant-Friederich-Lewy condition (CFL). Set this number to a value in the (0,1] interval. By default CFL is set
to 1.0 which is the recommended value for maximum performance. A few rare applications may require reducing
CFL to 0.5 or so avoid model oscillations in the model
results.
Constant time-step for model computation (DT) available only in RiverFlow2D FE model. Set this time-step
to run the model on constant time-step mode. Setting
DT is a critical issue for adequate stability and conservation. RiverFlow2D explicit time scheme is conditionally
stable, meaning that there is a maximum DT above which
the simulations will become unstable. This threshold can
be approximated by the Courant-Frederick-Lewy condition defined as follows:
 1/2
4
3
t

(1 )1/2
x
3 gh
where t =DT is the time-step, x is a measure of the
triangle element size, g is the acceleration of gravity, h is
the flow depth and  is the selective lumping parameter
(see description below). Initially the time-step may be
adequate, but during the simulation velocities and depth
of flow may increase causing the stability condition to be
exceeded. In this case the model will stop and it will
be necessary to run the model with a smaller time-step.
Alternatively, the variable time-step option may be used.
Closes the RiverFlow2D data input interface.
=

Time
step
factor
(RiverFlow2D FE only)
About Hydronia RiverFlow2D

Variable
time
(RiverFlow2D FE only)

Hot start

step

Time-step multiplier for variable time-step option. For

variable time-step simulations, RiverFlow2D FE computes
an estimate of the maximum DT using the theoretical
Courant-Frederick-Lewy (CFL) condition. This time-step
estimation may be high and the Time-step factor variable
can be used to adjust it. The Time-step factor will be multiplied by the estimated CFL time step to get the actual
DT used in computations. Typical project values range:
0.3-0.7.
Option to select variable time-step modeling. Otherwise
a constant time-step (DT) is applied. For variable timesteps, an estimate of the maximum DT using the theoretical Courant (CFL) condition is computed that is modified
by the time-step factor DTMULT. The time-step decreases
with high velocity flows and increases when the flow slows
down.
Use this option to restart the model from a previously
simulation.

Table 8.3 Units Frame on the Control Data Panel.

CONTROL NAME
Metric

DESCRIPTION
Select this option to work in metric units. Coordinates
are given in meters, velocities in m/s, discharge in m3 /s,
etc. Text output is provided in metric units.
Continued on next page

95

RiverFlow2D Data Input Program (DIP)

Table 8.3 continued from previous page

DESCRIPTION
Select this option to work in English units. Coordinates
are given in feet, velocities in f t/s, discharge in f t3 /s, etc.
Text output is provided in English units.
Note: Changing the units on this panel does not change the contents of the data files.
CONTROL NAME
English

Table 8.4 Computation Control Data Frame on the Control Data Panel.
CONTROL NAME
Selective lumping parameter
(RiverFlow2D FE only)

Mannings n factor

Mannings n variable with

depth

Minimum depth for dry bed

Wetting-Dry
method
(RiverFlow2D FE only)
Wetting-Dry
method
(RiverFlow2D FE only)

Wetting-Dry
method
(RiverFlow2D FE only)

DESCRIPTION
The selective lumping parameter,  (EPSILON) is used to
improve the model capabilities to simulate steep gradient
water surface changes and rapid flows. Lower EPSILON
values tend to smooth out depth and velocity gradients
while higher values would steepen frontal waves, but tend
to make solutions prone to instability. A recommended
starting EPSILON = 0.9.
Use this factor (XNMAN) to test the sensitivity of results
to the Mannings n. Each element Mannings n-value will
be multiplied by XNMAN. Over several simulations, using
this factor may speed up the model calibration. Default
is XNMAN = 1.
Select this option to set Mannings n as a function of
depth. The user must enter polygons over the mesh and
each polygon should have an associated file containing the
depth vs Mannings n table.
In the RiverFlow2D FE model this parameter sets the
depth limit (HMIN) for dry-wet calculation. If depth
is less than HMIN, node will be considered dry. In the
RiverFlow2D Plus and GPU models, this parameter indicates the depth below which cell velocity will be assumed
0. By default it is set to -1 which will allow the model to
dynamically set the dry cell conditions.
Water surface elevation on partially wet element is not
modified.
Water surface elevation on partially wet element is enforced to be horizontal and velocities are set to 0 on all
nodes of the element.
Not implemented.

Table 8.5 Closed Boundary Conditions Data Frame on the Control Data Panel (RiverFlow2D FE
only).
CONTROL NAME
Slip (RiverFlow2D FE only)

Non-Slip (RiverFlow2D FE
only)
Semi-Slip (RiverFlow2D FE
only)
Semi-Slip
factor
(RiverFlow2D FE only)

DESCRIPTION
Sets to 0 the velocity normal to the boundary at each node
allowing tangential flow. Node normal is approximated
using the perpendicular to the line joining the previous
and next node on the boundary.
Sets to 0 both components of the velocity at all nodes on
closed boundaries.
The model will impose a Semi-Slip condition where the
tangential velocity will be multiplied by the Semi-Slip factor.
Factor to multiply the tangential velocity when using
Semi-Slip condition.

8.1

Control Data Panel (.DAT files)

96

Table 8.6 Output Options Data Frame on the Control Data Panel.
CONTROL NAME
Text file output
Graphic output
Output results for cross sections
Output results for profiles

Output maximum values

Output results at observation

points

DESCRIPTION
Reports results to ASCII text output files.
Activates plot simulation results while program runs and
reporting to graphic files.
Use this option to generate results for user defined cross
sections. The cross section can be edited in the Cross
Section Output Panel. This data goes in a .XSECS file.
Use this option to generate results along a user defined
polyline. The polyline data can be edited in the Profile
Cut Output Panel. This data goes in a .PROFILES file.
Switch to allow reporting maximum values throughout the
simulation to .MAXI, .MAXE, and maximum value .EXP output files. In the RiverFlow2D Plus model maximum values will only be generated using the post processing Plot
RiverFlow2D results button on the Graphic Output Options panel.
Switch to allow reporting time series of results at specified
locations defined by coordinates in the Observation Points
Panel.

Table 8.7 Components Data Frame on the Control Data Panel.

CONTROL NAME
Rainfall/Evaporation

Bridge piers

Bridges

Culverts

Gates
Infiltration

Internal rating tables

Mud/Debris Flow

Oil Spill on Land

Pollutant Transport

Sources/Sinks

DESCRIPTION
Option to activate rainfall and/or evaporation. The required data has to be entered in the Rainfall /Evaporation
Panel. This data is written to .LRAIN file.
Switch to allow accounting for pier drag force. The Bridge
piers data can be edited in the Bridge Piers Panel. The
data is written to .PIERS file.
Switch to model Bridges using the bridge cross section
geometry and accounting for energy losses. The data can
be edited in the Bridges Panel. The data is written to
.BRIDGES file.
Switch indicating if one dimensional culverts will be used.
The Culverts data can be edited in the Culverts Panel.
The data is written to .CULVERTS file.
Switch to model gates. The data can be edited in the
Gates Panel. The data is written to .GATES file.
Option to activate Infiltration loss calculations. The required data has to be entered in the Infiltration Panel.
This data is written to .LINF file.
Switch to allow using internal rating tables. The data can
be edited in the Internal Rating Tables Panel. The data
is written to .IRT file.
Option to activate the Mud and Debris Flow modeling.
The data can be edited in the Mud/Debris Flow Panel.
The data is written to the .MUD file.
Option to activate the Overland Oil Spill modeling. The
data can be edited in the Oil Spill on Land Panel. The
data is written to the .OILP file.
Option to activate pollutant transport modeling. The pollutant transport data can be edited in the Pollutant Transport Panel. The data is written to the .SOLUTES file.
Switch to indicate existence of sources of sinks. The
sources/sinks data can be edited in the Sources/Sinks
Panel. The data is written to .SOURCES file.
Continued on next page

97

CONTROL NAME
Weirs

Wind

Sediment transport

RiverFlow2D Data Input Program (DIP)

Table 8.7 continued from previous page

DESCRIPTION
Switch to indicate the existence of Weir computation. The
Weir data can be edited in the Weirs Panel. The data is
written to .WEIRS file.
This option activates the calculation of wind stress on the
water surface. The data can be edited in the Wind Panel.
The data is written to .WIND file.
Option to activate sediment transport modeling with erosion and deposition for a mobile bed. The sediment transport data can be edited in the Sediment Transport Panel.
In the RiverFlow2D FE model this data is written to .SED
file and in the RiverFlow2D Plus model to the .SEDS and
.SEDB files.

Table 8.8 Components Data Frame on the Control Data Panel.

CONTROL NAME
Dry bed

Read initial water elevations

from .FED file
Horizontal water surface elevation
Initial water elevation

DESCRIPTION
The simulation will start with a fully dry bed. For discharge boundary conditions, an arbitrary depth (> 0.0)
is assigned to calculate the inflow for the first time-step.
Subsequently the flow depth at the boundary will be determined by the model.
Initial water surface elevations will be read from the .FED
file. It is possible to assign a spatially variable initial water
surface elevation in the Initial Conditions Layer.
Use this option to start a simulation with a user provided
initial horizontal water surface elevation.
Initial water surface elevation on the whole mesh. If initial
water elevation is set to -9999, the program will assign a
constant water elevation equal to the highest bed elevation
on the mesh.

8.2 Sediment Transport Data Panel for the

RiverFlow2D Plus model (.SEDS and .SEDB Files)
This panel allows entering sediment transport data for the RiverFlow2D FE model. To activate
this panel, first select Sediment Transport on the Components Frame of the Control Data Panel.
Table 8.9 Parameters on the Sediment transport Mode frame and buttons of the Sediment Transport
Panel (RiverFlow2D Plus model).
CONTROL NAME
Suspended sediment

Bed load Sediment

Buttons
Open .SED*
Save .SED*

DESCRIPTION
When this check box is selected, the model will compute sediment concentrations using the suspended sediment transport component. See comment 1.
Selecting this check box will activate the bed load sediment
transport component. See comment 1.
Opens an existing .SEDS or .SEDB files.
Saves the sediment data to .SEDS and .SEDB files.

8.2

Sediment Transport Data Panel for the RiverFlow2D Plus model (.SEDS and .SEDB Files)
98

Figure 8.5 Sediment Transport Panel (RiverFlow2D Plus model).

Table 8.10 Parameters on the Suspended Sediment transport frame of the Sediment Transport
Panel (RiverFlow2D Plus model).
CONTROL NAME
Equilibrium Concentration
formula

DESCRIPTION
When this check box is selected, the model will compute
sediment concentrations using one of the following suspended sediment transport formulas:
1. Bagnold (1966)lb
2. Van Rijn (1984a)

Settling Velocity Formula

It is a unique formula that applies for all fractions. This

drop-down list includes the following formulas:
1. Rubey (1933)
2. Zhang (1961)
3. Zanke (1977)
4. Van Rijn (1984a)
5. Raudkivi (1990)
6. Julien (1998)
7. Cheng (1997)
8. Jimenez-Madsen (2003)
9. Wu-Wong (2006)

Factor
Table
Density
Initial Concentration
Diameter
Porosity
Shields Stress
Friction Angle

This factor multiplies the settling velocity calculated by

the selected formula.
Suspended sediment density (kg/m3 or lb/f t3 ).
Initial volumetric sediment concentration. See comment
2.
Characteristic sediment size for this fraction (m or f t).
Bed porosity.
Critical Shield stress.
Sediment friction angle (degrees).
Continued on next page

99

CONTROL NAME
Factor

RiverFlow2D Data Input Program (DIP)

Table 8.10 continued from previous page

DESCRIPTION
Equilibrium concentration formula factor for each fraction.
This factor multiplies the equilibrium concentration calculated by the selected formula.

Buttons
Add Suspended Sediment
Fraction
Remove Selected Fraction

Used to add a new fraction. Up to 10 fractions may be

used.
Deletes the selected fraction.

Table 8.11 Parameters on the Bed Load Sediment transport frame of the Sediment Transport Panel
(RiverFlow2D Plus model).
CONTROL NAME
Sediment load formula

DESCRIPTION
Allows selection of one of the following sediment transport
formulas:
1. Meyer-Peter & Muller (1948)
2. Ashida (1972)
3. Engelund (1976)
4. Fernandez (1976)
5. Parker fit to Einstein (1979)
6. Smart (1984)
7. Nielsen (1992)
8. Wong 1 (2003)
9. Wong 2 (2003)
10. Camenen-Larson (1966)

Table
Density
Diameter D30
Diameter
Diameter D90
Porosity
Shields Stress
Friction Angle
Fraction
Factor
Buttons
Add Bed Load Fraction
Remove Selected Fraction

Sediment density (lb/f t3 or kg/m3 ).

Sediment D30 size (m). 30% of the sediment is finer than
D30. Only used for Smart Formula.
Characteristic sediment size for this fraction (m).
Sediment D90 size (m). 90% of the sediment is finer than
D90. Only used for Smart Formula.
Sediment porosity.
Critical Shield stress.
Sediment friction angle (degrees).
Fraction of material in bed. All fractions must add up to
1.
Transport formula factor for each fraction. This factor
multiplies the result of the transport formula selected.
Used to add a new fraction. Up to 10 fractions may be
used.
Deletes the selected fraction.

8.2.0.1 Comments for the .SEDS and .SEDB Files

1. You can select either one or both options. When the suspended sediment transport options
all inflow data files should contain time series of volumetric concentrations for each fraction
entered.
2. Volumetric concentration should be provided as a fraction of 1. Note that the typically total
maximum suspended load concentration do not exceed 0.08. Concentrations greater than
0.08 is generally considered hyperconcentrated flow which falls beyond the validity of the
sediment transport algorithms. Therefore, the sum of all initial concentrations should also
not exceed 0.08.

8.3

Mud and Debris Flow Data Panel for the RiverFlow2D Plus model (.MUD File)

100

8.3 Mud and Debris Flow Data Panel for the

RiverFlow2D Plus model (.MUD File)
This panel allows entering mud and debris flow data for the RiverFlow2D model. To activate this
panel, first select Mud on the Components Frame of the Control Data Panel.

Figure 8.6 Mud/Debris Flow Panel (RiverFlow2D Plus model).

Table 8.12 Parameters on the Mud/Debris Flow Panel (RiverFlow2D Plus model).
CONTROL NAME
Flow resistance relation

DESCRIPTION
1. Turbulent flow
2. Full Bingham
3. Simplified Bingham
4. Turbulent and Coulomb
5. Turbulent and Yield
6. Turbulent, Coulomb and Yield
7. Quadratic
8. Granular flow

Yield stress
Bingham viscosity
Internal friction angle
Material density

Yield stress (P a or lb/in2 ).

Bingham viscosity (Poise or lb/in2 ).
Internal friction angle (degrees).
Material density (kg/m3 or lb/f t3 ).

101

RiverFlow2D Data Input Program (DIP)

8.4 Oil Spill on Land for the OilFlow2D model (.OILP

File)
This panel allows entering viscous fluid properties for the OilFlow2D model. To activate this panel,
first select Oil Spill on Land on the Components Frame of the Control Data Panel.

Figure 8.7 Oil Spill on Land Panel (OilFlow2D model).

Table 8.13 Parameters on the Oil Spill on Land Panel (OilFlow2D Plus model).
CONTROL NAME
Fluid viscosity
Fluid density

DESCRIPTION
Viscosity (Pascal s or lbs/in2 ).
Density (kg/m3 or lb/f t3 ).

8.5 Graphic Output Options Tab (.PLT File)

This panel allows entering options to control RiverFlow2D output. To activate this panel, first
select Graphic Output Options from the Output group on the left panel of the RiverFlow2D DIP.

8.5

Graphic Output Options Tab (.PLT File)

102

Figure 8.8 Graphic Output Panel.

Figure 8.9 Color Palettes.

Table 8.14 Parameters on the Graphic Output Option Panel.

CONTROL NAME
Buttons
Open .PLT
Save .PLT
Plot RiverFlow2D Results

DESCRIPTION
Opens an existing .PLT output data file.
Saves only the graphic output data to a .PLT file.
Use this button to create output files and visualize
RiverFlow2D results from existing simulations. The output files for the simulation must be available.

Plot Frame
Continued on next page

103

CONTROL NAME
Plot

RiverFlow2D Data Input Program (DIP)

Table 8.14 continued from previous page

DESCRIPTION
Chose the desired plot from the list:
Velocity field using black arrows.
Velocity field using colored arrows based on velocity
magnitude.
Velocities in black over colored depths.
Velocities in black over colored bed elevations.
Flow depth.
Bed elevation.
Water elevations.
Velocities in black over colored water elevations.
Erosion and deposition.
Concentration

Color palette
Plot axis
Create DXF vector field files

Create graphic output files

Output for creating shapefiles
Background Image Frame
Background image
Background file
Transparency

List of variables Frame

Min
Max
Max velocity vector
Velocity vector factor
Plot Window Frame
X1
Y1
X2
Y2

Choose between 6 color schemes (see Figure ??).

Display axes on plot.
Generate velocity vector DXF (CAD) files. This option
will also export the mesh in DXF format to the file:
<ProjectName>_MESH.DXF.
Generates a raster image in GIF format for each report
time step.
Using this option RiverFlow2D will generate support files
that can be used to create ESRI Shapefiles for velocity
fields, depth and water surface elevations.
Check this if you want to import a background aerial to
for dynamic plots.
Selects images for background. Supported formats include: .BMP, .GIF, .PNG and .TIF.
Transparency of plots to allow visualizing the background
image under dynamic plots. 100=full transparency, 0=full
opacity.
Minimum value to plot of the current variable chosen in
Plot option.
Maximum value to plot of the current variable chosen in
Plot option.
Maximum velocity vector to plot.
Multiplication factor to amplify arrows in vector plots.
Use to define zoom or area to visualize. If all values are 0,
then maximum extent is plotted.
X coordinates of lower left corner of rectangle.
Y coordinates of lower left corner of rectangle.
X coordinate of upper right corner of rectangle.
Y coordinate of upper right corner of rectangle.

8.6 Profile Output Panel (.PROFILES File)

Use this panel to enter polyline coordinates where the model results are to be generated. RiverFlow2D
will generate output .PRFI and .PRFE files. To activate this panel, first select the Profile Output
in Output from the Output group on the left panel of the RiverFlow2D DIP.
See output file section (??) for output file content description.

8.8

Inflow Boundary Data Panel

104

Figure 8.10 Profile Output File.

Table 8.15 Parameters on the Profile Output Panel.
COLUMN NAME
Profile data table
Profile name
N Vertices
N Intervals
X, Y
Buttons
Open .PROFILES
Save .PROFILES

DESCRIPTION
Profile name. Should not contain spaces and must have
less than 26 characters.
Number of vertices in each profile.
Intervals to divide each profile sub-segment between vertices. Results will be reported at each interval.
Coordinates for each vertex in polyline.
Opens an existing .PROFILES file.
Saves only the profile data to a .PROFILES file.

8.7 Cross Section Output Panel (.XSECS File)

Use this panel to enter coordinates for cross sections that intersect the triangular element mesh
where you want to output model results. RiverFlow2D will generate output .XSECI and .XSECE
files. To activate this panel, first select the Cross Section Output from the Output group on the
left panel of the RiverFlow2D DIP.
See output file section (4.4.1) for output file content description.
Table 8.16 Parameters on the Inflow Boundary Data Panel.
COLUMN NAME
XS data table
XS Name
N Intervals
X1 Y1 X2 Y2
Buttons
Open .XSECS
Save .XSECS

DESCRIPTION
Cross section name. Should not contain spaces and must
have less than 26 characters.
Intervals to divide each section. Results will be extracted
and reported at each interval.
Each row corresponds to the coordinates of the initial
(X1,Y1) and ending (X2,Y2) of one cross section.
Opens an existing .XSECS file.
Saves only the cross section data to a .XSECS file.

8.8

Inflow Boundary Data Panel

Figure 8.11 Cross Section Output Panel.

Figure 8.12 Inflow Boundary Data Panel.

105

8.9

Outflow Boundary Data Panel

106

Use the Plot button to generate a graph of the time series as shown in Figure ??.

Figure 8.13 Inflow hydrograph plot.

Table 8.17 Parameters on the Inflow Boundary Data Panel.

COLUMN NAME
Select boundary condition
Discharge vs. Time
Water Surface Elevation
(WSE) vs. Time
Data table
Time (hrs.)
Q; WSEL*
Buttons
Open Inflow File
Save Inflow File
Plot
*Varies according to boundary

DESCRIPTION
type
To select a discharge vs. time hydrograph.
To select a water surface elevation vs. time data series.

Time in hours.
Water discharge (m3 /s or f t3 /s); Water surface elevation
(m or f t).
Opens an existing file with inflow data.
Saves only the inflow table to a file.
Generates a graph of the time series.
condition type.

8.9 Outflow Boundary Data Panel

Use this panel to create, edit, and plot hydrographs (Discharge vs. Time, Water Surface Elevation
vs. Time, or Discharge Rating Curve tables) that can be assigned as outflow boundary conditions.
To activate this panel, first select the Outflow Boundary Data Output from the Components group
on the left panel of the RiverFlow2D DIP.

107

RiverFlow2D Data Input Program (DIP)

Figure 8.14 Outflow Boundary Data Panel.

Use the Plot button to generate a graph of the time series.

Table 8.18 Parameters on the Outflow Boundary Data Panel.
COLUMN NAME
Select boundary condition
Discharge vs. Time
Water Surface Elevation
(WSE) vs. Time
Discharge Rating Curve
Data table
Time; Q*
Q; WSE*
Buttons
Open Outflow File
Save Outflow File
Plot
*Varies according to boundary

DESCRIPTION
type
To select a discharge vs. time hydrograph.
To select a water surface elevation vs. time data series.
To select a discharge rating curve.
Time in hours; Q (m3 /s or f t3 /s).
Water discharge (m3 /s or f t3 /s); Water surface elevation
(m or f t).
Opens an existing file with outflow data.
Saves only the outflow table to a file.
Generates a graph of the time series.
condition type.

8.10

Culverts Panel (.CULVERTS File)

108

8.10 Culverts Panel (.CULVERTS File)

This panel is used to display the content of the .CULVERTS file and enter data for culverts. Figure
?? shows the Culvert panel with a three culverts. Selecting Culvert1 on the first row shows the
associated rating table. To activate this panel, first select the Culverts from the Components group
on the left panel of the RiverFlow2D DIP.

Figure 8.15 Culverts Panel showing data in rating curve.

Figure ?? shows the corresponding data entry controls that appear when selecting the first row
for Culvert1 that is a circular culvert.
Table 8.19 Parameters on the Culverts Panel.
COLUMN NAME
Culvert data
Culvert Name
Type

Culvert File

X1, Y1, X2, Y2

Mannings roughness coefficient
Entrance loss coefficient
Culvert inlet control formula
coefficients
m

DESCRIPTION
Culvert name. Should not contain spaces and must have
less than 26 characters.
Type of culvert. For Type = 0, culvert discharge is computed from a user given rating table on the Culvert File.
For Types = 1 and 2, discharge is computed using culvert
equations based on culvert characteristics provided in the
Culvert File.
Culvert rating table file name or culvert characteristic
data. Name Should not contain spaces and must have
less than 26 characters.
Coordinates of vertices defining each culvert line.
Culvert Mannings n Coefficient given by Table qweqweqwe.
Culvert entrance loss coefficient given by Table qweqweqwe.
Culvert inlet control formula coefficients given by Table
qweqweqwe.
Inlet form coefficient. m=0.7 for mitered inlets, m=-0.5
for all other inlets.
Continued on next page

109

RiverFlow2D Data Input Program (DIP)

Table 8.19 continued from previous page

DESCRIPTION
Barrel height for box culverts (ft or m). Only for box
culverts: CulvertType = 1.
Barrel width (Base)
Barrel width for box culverts (m or ft). Only for box
culverts: CulvertType = 1.
Diameter (Dc)
Barrel diameter for circular culverts (m or ft). Only for
circular culverts: CulvertType = 2.
Number of barrels
Number of identical barrels.
Use element elevations
When this check box is selected the model will extract the
inlet and outlet invert elevations from the element or cell
elevations of the culvert ending points. If the check box is
not selected, the user can enter the inlet invert elevation
(Z1) and outlet invert elevations (Z2) that may be different
from the element/cell elevations.
Buttoms
Open .CULVERTS
Opens an existing .CULVERTS file.
Save .CULVERTS
Saves only the culvert data to a .CULVERTS file.
COLUMN NAME
Barrel height (Hb)

8.11

Internal Rating Tables Panel (.IRT File)

110

8.11 Internal Rating Tables Panel (.IRT File)

This panel is used to display the content of the .IRT file and enter data for culverts. In this Panel
can also edit Internal Rating Table polylines, type, and data file name. To activate this panel, first
select the Internal Rating Table from the Components group on the left panel of the RiverFlow2D
DIP.

Figure 8.16 Internal Rating Tables Panel.

Table 8.20 Parameters on the Internal Rating Tables Panel.

COLUMN NAME
Data table
IRT Name
Type

File Name
X, Y
Buttoms
Open .IRT
Save .IRT

DESCRIPTION
Name of internal rating table. Should not contain spaces
and must have less than 26 characters.
Boundary condition is always equal to 19 in this version,
corresponding to discharge vs. water surface elevation tables.
Name of file containing internal rating table data in the
format described as a stage-discharge data file.
Coordinates of vertices defining each IRT polyline.
Opens an existing .IRT file.
Saves only the internal rating table data to a .IRT file.

111

RiverFlow2D Data Input Program (DIP)

8.12 Weirs Panel (.WEIRS File)

This panel is used to display the content of the .WEIRS file. In this Panel can also create weir
polyline data. To activate this panel, first select the Weirs from the Components group on the left
panel of the RiverFlow2D DIP.

Figure 8.17 Weirs Panel.

Table 8.21 Parameters on the Weirs Panel.

COLUMN NAME
Data table
Weir Name
N Vertices
Cf
X, Y
Buttoms
Open .WEIRS
Save .WEIRS

DESCRIPTION
Name of weir. Should not contain spaces and must have
less than 26 characters.
Number of points defining each weir polyline.
Weir coefficient.
Coordinates of vertices defining each weir polyline (m or
ft).
Opens an existing .WEIRS file.
Saves only the weir data to a .WEIRS file.

8.13

Sources/Sinks Panel (.SOURCES File)

112

8.13 Sources/Sinks Panel (.SOURCES File)

This panel is used to display the content of the .SOURCES file. Use this Panel to also create sources
and sinks location data, type, and sources/sink data file. To activate this panel, first select the
Sources and Sinks from the Components group on the left panel of the RiverFlow2D DIP.

Figure 8.18 Sources/Sinks Panel.

Table 8.22 Parameters on the Sources/Sinks Panel.

COLUMN NAME
Data table
Source/Sink Name
File Name
X, Y
Buttoms
Open .SOURCES
Save .SOURCES

DESCRIPTION
Name of point source or sink. Should not contain spaces
and must have less than 26 characters.
Name of file containing the time series of each point source
or sink.
Coordinates of point.
Opens an existing .SOURCES file.
Saves only the sources and sinks data to a .SOURCES file.

113

RiverFlow2D Data Input Program (DIP)

8.14 Open Boundary Conditions Panel (.IFL File)

This panel is used to display the content of the .IFL file. Since this file is automatically generated
by the RiverFlow2D model based on the user input in the SMS interfaces, any change made will be
disregarded when the model runs. To activate this panel, first select the Open Boundary Conditions
from the Components group on the left panel of the RiverFlow2D DIP.

Figure 8.19 Open Boundary Conditions Panel.

Table 8.23 Parameters on the Open Boundary Condition Panel.

COLUMN NAME
Data table
Node
BC Type
File
Buttoms
Open .IFL
Save .IFL

DESCRIPTION
Node number.
Code to indicate type of open boundary. See boundary
condition options in the .IFL file description section.
Boundary condition file name. Should not contain spaces
and must have less than 26 characters.
Opens an existing .IFL file.
Saves only the open boundary conditions data to a .IFL
file.

8.16

Observation Points Panel (.OBS File)

114

8.15 Bridge Piers Panel (.PIERS File)

This panel is used to display the content of the .PIERS file. In this Panel can also create bridge
pier location and pier characteristic data. To activate this panel, first select the Bridges Piers
from the Components group on the left panel of the RiverFlow2D DIP.

Figure 8.20 Bridge Piers Panel.

Table 8.24 Parameters on the Bridge Piers Panel.

COLUMN NAME
Data table
X,Y
Angle X
Length
Width
Drag Coeff.
Pier Name
Buttoms
Open .PIERS
Save .PIERS

DESCRIPTION
Coordinates of pier centroid.
Pier angle with respect to X axis.
Pier length (m or f t).
Pier width (m or f t).
Drag coefficient of the pier.
Name of pier. Should not contain spaces and must have
less than 26 characters.
Opens an existing .PIERS file.
Saves only the bridge piers data to a .PIERS file.

8.16 Observation Points Panel (.OBS File)

Use this panel to create, edit and display the content of the .OBS file. To activate this panel, first
select the Observation Points from the Output group on the left panel of the RiverFlow2D DIP.

115

RiverFlow2D Data Input Program (DIP)

Figure 8.21 Observation Points Panel.

Table 8.25 Parameters on the Observation Points Panel.
COLUMN NAME
Data table
Observation Point Name
X,Y
Buttoms
Open .OBS
Save .OBS

DESCRIPTION
Name of observation point. Should not contain spaces and
must have less than 26 characters.
Coordinates of point.
Opens an existing .OBS file.
Saves only the observation point data to a .OBS file.

8.17

Tools Panel

116

8.17 Tools Panel

This section describes various utilities that are available through the RiverFlow2D Data Input
Program.

8.17.1 Two-Dimensional Model Result Interpreter: 2dMRI

2dMRI is a RiverFlow2D utility developed by Northwest Hydraulic Consultants (http://www.
nhc-sea.com). The program gathers the data from RiverFlow2D output files and generates plots
and animations for cross sections, profiles and observation points.
To access the program use first select Tools from the Output group on the left panel of the
RiverFlow2D DIP.

Figure 8.22 Two-dimensional Model Result Interpreter: 2dMRI.

Explanation and guidelines on how to use the 2dMRI utility is described in the 2dMRI manual
included in the documentation folder:
...\Documents\RiverFlow2Dv4\Documentation.

8.17.2 HEC-RAS Data Extraction Tool

The purpose of this tool is to facilitate migrating existing HEC-RAS projects to RiverFlow2D.
The program allows extraction of point elevation data from geo-referenced cross-section from the
HEC-RAS one-dimensional model developed by the USACE. The tool reads HEC-RAS geometry
files with extension .g01, .g02, etc., and creates X Y Z files that can be readily imported in SMS.
The utility discriminates the elevations in the channel between the left and right bank on each
cross section and exports the files as detailed in the following table.

117

RiverFlow2D Data Input Program (DIP)

Table 8.26 Files generated by the HEC-RAS Data Extraction Tool.

FILE NAME
<HEC-RASfilename>_ALL_STATIONS.EXP

<HEC-RASfilename>_ALL_POLY.EXP
<HEC-RASfilename>h0_CHANNEL_STATIONS.EXP

<HEC-RASfilename>_CHANNEL_POLY.EXP

DESCRIPTION
Contains all elevation points in
all cross sections in the for all
reaches and cross sections in the
<HEC-RASfilename>.g0? file.
Contains polygons that include all elevation points in each reach.
Contains only the elevation points between the left and right banks in all
cross sections in the for all reaches in
the <HEC-RASfilename>.g0? file.
Contains polygons that include only
the elevation in the main channel for
each reach.

To activate this panel, first select the Tools from the Output group on the left panel of the
RiverFlow2D DIP.

Figure 8.23 Two-dimensional Model Result Interpreter: 2dMRI.

The tutorial: Using HEC-RAS elevation data in the RiverFlow2D Users Guide, describes in
detail the procedure to use this tool.

8.17

Tools Panel

118

Input Data File Reference

Data files for non-spatial information required to run the RiverFlow2D model are created by the
Data Input Program and the spatial data is generated based on the .2dm file. All RiverFlow2D
input data files are in ASCII free-form format, which can be opened using any text editor or
spreadsheet program. In some instances it may be convenient to directly edit the data. However,
it is recommended to edit files with extreme caution, and only after having gained a thorough
understanding of RiverFlow2D file formats. This section explains the input data files, and the
parameters included in each file.
The RiverFlow2D installation program creates a folder with several example projects that can
be consulted to review the model data files. Depending on your operating system and settings,
this folder can be found in:
...\Documents\RiverFlow2D SMS\ExampleProjects.
RiverFlow2D data files will share the same file name and use the file extensions listed in the
table below. For example a run named Run1 will have files as follows: Run1.DAT, Run1.FED, etc.
The following table summarizes the data files used by RiverFlow2D model.
Table 9.1 List of Input Data Files.
NAME
SMS
SMS project file

FILE
EXTENSION

DEPENDENCIES
CONTENT

This is the project file where

SMS stores the triangular element mesh, the data and projection.
2DM file
.2DM
Required when
This is the file where SMS stores
using SMS GUI all the data for a simulation using
the generic mesh.
INDIVIDUAL DATA FILES CREATED FROM THE *2DM
SPATIAL DATA FILES
Elevation data
any
Required
Scattered elevation data points.
Continued on next page
.SMS

Required when
using SMS GUI

119

120
Table 9.1 continued from previous page
FILE
EX- DEPENDENCIES
CONTENT
TENSION
Triangle-element
.FED
Required
Node coordinates and elevamesh data
tions, triangular mesh topology,
boundary condition type and file
names, initial water elevations,
and Mannings n coefficients.
Mesh boundary
.TBA
Internal file
List of external and island boundnodes
ary nodes. This file is automatically generated by RiverFlow2D.
I/O
boundary
.IFL
Internal file
List of external boundary nodes,
conditions
inflow and outflow conditions
This file is automatically generated by RiverFlow2D.
CONTROL FILES
Run control data
.DAT
Required
General run control options, including time step, simulation
time, metric or English units,
graphical output options, initial
conditions, components, etc.
Plot results op.PLT
Optional
Graphical output options.
tions
River Animator
.PLOT
Optional
Options for River Animator to
options
plot results in Google Earth, create animations and Windows Media Video files (.WMV).
Observation
.OBS
Optional
Location of observation points
points data
where the model will report time
series of results.
Cross section out- .XSECS
Optional
List of cross sections where the
put
model will output results. Each
cross section is defined by coordinates of its two ending points.
Profile output
.PROFILES
Optional
Mesh profile cut where results are
desired.
BOUNDARY CONDITIONS
user defined
Required
Hydrograph, water surface eleTime series or
vations vs.
time, etc.
The
rating table files
model requires one file for each
for
inflow
or
open boundary condition, except
outflow boundary
free boundary condition types.
conditions
Initial concentra.CINITIAL
The Pollutant Created by SMS using an .h5 file.
tion of each polluTransport comtant
ponent is selected
COMPONENT FILES
Bed load sedi- .SEDB
Sediment
Bed load sediment transport data
ment transport
transport
for the RiverFlow2D Plus model.
model
Bridge piers
.PIERS
The pier comBridge pier data used to calculate
ponent is se- pier drag forces.
lected
Culverts
.CULVERTS
The
culvert
Culvert location and associated
component is
culvert data files.
selected
Internal
rating .IRT
The
Internal
Data to impose discharge rating
tables
Rating Table
tables along internal boundaries.
component is
selected
Mud
.MUD
The
Provides the parameters necesMud/Debris
sary to model mud and debris
Flow compo- flow using the RiverFlow2D Plus
nent is selected
model.
Continued on next page
NAME

121

Input Data File Reference

Table 9.1 continued from previous page

FILE
EX- DEPENDENCIES
CONTENT
TENSION
Pollutant trans- .SOLUTE
The Pollutant Data for passive or reactive polport
Transport com- lutants.
ponent is selected
Rainfall/Evaporation.RET
The
Rainfall Time series for rainfall and evapcomponent is
oration.
selected
Sediment trans- .SED
Sediment
Sediment transport data for the
port
transport
RiverFlow2D FE model.
model
Suspended sedi- .SEDS
Sediment
Suspended sediment transport
ment transport
transport
data for the RiverFlow2D Plus
model
model.
Sources and sinks
.SOURCES
The
Sources This file contains location of incomponent is
put discharge sources or output
selected
discharge sinks and associated
time series of discharge data files.
Weirs
.WEIRS
The Weir com- This file contains weirs polylines
ponent is se- and associated weir data.
lected
Wind
.WIND
The Wind com- This file contains wind specific
ponent is se- density and velocity data. The
lected
wind stress component is only
available for the RiverFlow2D FE
in this release.
NAME

9.1 Run Control Data

9.1.1 Run Control Data File: .DAT
This file contains parameters to control the model run including time step, simulation time, metric or English units, physical processes or component switches, and graphical output and initial
conditions options.
Line 1: Internal program version number.
RELEASE
Line 2: Model selector switch.
IMS
Line 3: Physical processes or component switches.
IRAIN ISED IPIERS IWEIRS ICULVERTS ISOURCES IINTRC
Line 4: Wet-dry bed method switch.
IWETDRY
Line 5: Output control switches.
IEXTREMES IXSEC IPROFILE DUMMY IOBS
Line 6: Time control data.
DUMMY CFL DUMMY TOUT TLIMT

9.1

Run Control Data

122

Figure 9.1 Example of a RiverFlow2D Mesh.

Line 7: Initial conditions and hot start control switches.

IINITIAL IHOTSTART
Line 8: Manning n variable with depth switch.
IMANN
Line 9: Mannings n value global multiplication factor.
XNMAN
Line 10: Dummy parameter.
DUMMY
Line 11: Unit system definition switch.
NUNITS
Line 12: Surface detention or minimum value of flow depth for dry areas.
HMIN
Line 13: Initial water surface elevation.
INITIAL WSE
Line 14: Pollutant transport model switch.
IADDISP
Line 15: Wind stress switch.
IWIND

123

Input Data File Reference

Line 16: Mud/Debris flow model switch.

IMUD
Line 17: Number of cores or GPU ID.
IDGPU

9.1.1.1 Example of the .DAT file

201301
1
0000110
2
00000
0 0.5 0.25 0.25 8
10
1
1
0.9
1
-1
0
0
0
0
4

Table 9.2 Variable Descriptions for the .DAT File.

VARIABLE
CFL

TYPE
R

RANGE
(0, 1]

UNITS
-

DUMMY

HMIN

1 or > 0

m/f t

IADDISP

0,1

DESCRIPTION
Applies to RiverFlow2D Plus and
RiverFlow2D Plus GPU models.
Courant number. Default value
is set to 0.7. CFL may need to
be set to lower values if results
show signs of unexpected oscillations or instability. CFL may often be increased to 1 with stable
results.
Dummy parameter for future use.
Ignored in RiverFlow2D Plus.
In RiverFlow2D Plus HMIN is the
depth limit for dry-wet calculation. If depth is less than HMIN,
cell velocity will be set to 0. If
HMIN = -1, dry RiverFlow2D FE
HMIN defines the depth above
which the node or element will be
considered wet.
Switch to activate the pollutant
transport model.
1. Turn off pollutant transport computations.
2. Calculate pollutant transport.
Continued on next page

9.1

Run Control Data

VARIABLE
ICULVERTS

124
Table 9.2 continued from previous page
TYPE
RANGE
UNITS
DESCRIPTION
I
0,1
Switch
indicating
if
onedimensional culverts will be
used.
1. No culverts will be used.
2. Use culverts.

IDGPU

IEXTREMES

0,1

Requires .CULVERTS file. See details on the Culverts Section of

this manual.
RiverFlow2D Plus: This parameter indicates how many processors or cores will be used in
the parallel computation. The
maximum number will depend
on the processor capabilities.
RiverFlow2D Plus GPU: If your
computer has multiple GPU
cards, this parameter allows selecting which card will be used
for the run. Since the model allows only one concurrent run per
cards, this option allows running
simultaneous simulations in different cards.
Switch to reporting maximum
values throughout the simulation.
1. Do not report maximum
values.
2. Report maximum values.

IHOTSTART

0,1

Switch to start run from scratch

or continue a previous simulation.
1. Start simulation from initial time.
2. Start simulation from previous run.

IINTRC

0,1

Switch for internal rating tables.

1. Do not use internal rating
table component.
2. Use internal rating tables.
See details on Internal Rating
Tables Section of this manual.
Continued on next page

125

VARIABLE
IINITIAL

Input Data File Reference

Table 9.2 continued from previous page

TYPE
RANGE
UNITS
DESCRIPTION
I
0,1,2,Initial condition switch for water
9999
surface elevations.
1. Prescribed horizontal water surface elevation
2. Initial dry bed on whole
mesh.
3. Initial water surface elevations read from .FED file
-9999: Assigns a horizontal water elevation equal
to the maximum bed elevation plus 0.5 m. (1.64
ft.). See comment 3.

INITIAL WSE

m/f t

IMANN

1,2

Initial water surface elevation on

the whole meshes. This will be
the initial water surface if IINITIAL is 0. See comment 3.
Variable Manning n with depth
switch.
1. Mannings n is constant
for all depths.
2. Mannings n may vary
with depths as defined in
the .MANNN file.

IMS

0,1,2

Model switch used to select the

hydrodynamic model engine.
1. RiverFlow2D FE.
2. RiverFlow2D Plus.
3. RiverFlow2D Plus GPU.

IMUD

0,1

Switch to select mud/debris flow

model.
1. Do not run mud/debris
flow model.
2. Run mud/debris
model.

IOBS

0,1

flow

Requires .MUD file. See details on

the Mud/Debris Flow Model section of this manual.
Switch to report time series of
results at specified locations defined by coordinates.
1. Do not report on observation points.
2. Report
points.

on

observation

Requires .OBS file. See details on

the Observation Points section of
this manual.
Continued on next page

9.1

Run Control Data

VARIABLE
IPIERS

126
Table 9.2 continued from previous page
TYPE
RANGE
UNITS
DESCRIPTION
I
0,1
Switch to allow accounting for
pier drag force.
1. Do not use pier drag force
option.
2. Use pier drag force option.

IPOLLUTANT

0,1

Requires .PIERS file. This option may be used if the mesh does
not account for the pier geometry. See details on Bridge Piers
Section of this manual.
Switch to select pollutant transport model.
1. Do not run pollutant
transport model.
2. Run pollutant transport
model. Requires .AD file.

IPROFILE

0,1

See details on the Pollutant

Transport Model section of this
manual.
Switch to control profile output.
1. No profile results output.
2. Results will be output
along a prescribed profile.

IRAIN

0-4

Requires .PROFILES file.

See
comment 4.
Switch for rainfall and evaporation input.
1. No rainfall modeling.
2. Not used.
3. Rainfall/evaporation.
4. Infiltration.
5. Rainfall/evaporation and
Infiltration.

ISED

0,1

Sediment transport switch.

1. No sediment
modeling.

transport

2. Sediment transport, mobile bed erosion, and deposition will be simulated.

Requires .SED file.

ISOURCES

0,1

See details on the Sediment

Transport section of this manual.
Switch for sources and sinks.
1. No sources or sinks are
present.
2. Sources or sinks are
present.
Requires
.SOURCES file.
See details on the Sources section
of this manual.
Continued on next page

127

VARIABLE
IWEIRS

Input Data File Reference

Table 9.2 continued from previous page

TYPE
RANGE
UNITS
DESCRIPTION
I
0,1
Weir computation on internal
boundary switch.
1. Do not use weir computation on internal boundaries.
2. Use weir computation on
internal boundaries.

IWIND

0,1

See details on the Weirs section

of this manual.
This component is only available
for the RiverFlow2D FE in this release. Switch to account for wind
stress on the water surface.
1. Do not
stress.

consider

wind

2. Account for wind stress.

Requires .WIND file.

IXSEC

0,1

See details on the Wind Stress

section of this manual.
Cross section output switch.
1. No cross section result
output.
2. Cross section results will
be output to file. Requires
.XSECS file. See comment
5.

NUNITS

0,1

Variable to indicate unit system:

1. Metric units.
2. English units.

RELEASE

TLIMT
TOUT

R
R

hrs.
hrs.

XNMAN

>0

T LIM T
[0.1-2]

Release number ID used internally for reference. Should not be

modified.
Total simulation time.
Output time interval for reporting results.
Mannings n coefficient multiplier. See comment 6.

Note: I = Integer variable. R = Real variable.

9.1.1.2 Comments for the .DAT file

1. Setting the CFL (Courant Friederich-Lewy) or Courant number is critical for adequate stability and ensure mass conservation. RiverFlow2D explicit time scheme is conditionally
stable, meaning that there is a maximum time step above which the simulations will become
unstable. This threshold can be theoretically approximated by a Courant-Frederick-Lewy
condition defined as follows:

t gh
1
CF L =
x

(9.1)

where t = DT is the time-step, x is a measure of the minimum triangular element or cell

size, g is the acceleration of gravity, and h is the flow depth. It may occur that during the
initial stages of a hydrograph, velocities are small and the selected time step is adequate.

9.2

Mesh Data

128

During the simulation, however, velocities and flow depth may increase causing the stability
condition to be exceeded. In those cases it will be necessary to rerun the model with a smaller
CFL. Alternatively, the variable time step option may be used.
2. For variable time step simulations, RiverFlow2D estimates the maximum DT using the theoretical Courant-Frederick-Lewy (CFL) condition. Sometimes, the estimated DT may be too
high, leading to instabilities, and it may be necessary to reduce CFL to with a value less than
one to adjust it. Typical CFL values range from 0.3 to 1, but may vary project to project.
3. There are three initial conditions options. If IINITIAL = 0, the initial water elevation will
be a constant horizontal surface at the elevation given as INITIAL WSE. If INITIAL WSE
is = -9999 then the program will assign a constant water elevation equal to the highest bed
elevation on the mesh. If IINITIAL = 1, the whole computational mesh will be initially dry,
except at open boundaries where discharge is prescribed and depth > 0 is assumed for the
first time step. If IINITIAL = 2, initial water surface elevations are read from the .FED data
file for each node in the mesh. These initial elevations can be set creating a dataset file (.h5)
on SMS layer, and can be arbitrary. This last option allows modeling cases where part of the
domain is flooded (e.g. a reservoir) and other parts of the mesh are dry. The initial velocities
are always assumed to be equal to 0.
4. Use the IPROFILE option to allow RiverFlow2D to generate results along a polyline. The
polyline and other required data should be given in the Profiles file (.PROFILES), which is
defined later in this document.
5. Use this option to allow RiverFlow2D to generate results along prescribed cross sections. The
cross sections and other required data should be given in Cross Section file (.XSECS) which
is defined later in this document.
6. Use the XNMAN option to test the Mannings n value sensitivity on the results. The
prescribed Mannings coefficient assigned to each element will be multiplied by XNMAN.
This option is useful to test model sensitivity to Mannings n during model calibration.
7. The model will create output files with maximum values of each output variable.
8. The user can specify an initial water surface elevation setting IINITIAL = 0 and entering
INITIAL WSE.

9.2 Mesh Data

9.2.1 Mesh Data File: .FED
This file has the data that defines the triangular element mesh, and includes node coordinates,
connectivity for each triangular element, node elevations, Mannings n coefficients and other parameters. This file is created with SMS GUI. The templates for RiverFlow2D provided with SMS
assures that the .FED file will be created error free and consistent with the boundary conditions
and other mesh parameters. Manually editing this file may introduce unexpected errors.
Line 1: Number of elements and nodes.
NELEM NNODES DUMMY DUMMY
NNODES lines containing node coordinates and node parameters.
IN X(IN) Y(IN) ZB(IN) INITWSE(IN) MINERODELEV(IN) BCTYPE BCFILENAME

129

Input Data File Reference

NELEM lines containing mesh connectivity and element/cell parameters.

IE NODE(IE,1) NODE(IE,2) NODE(IE,3) MANNINGN(IE) ELZB(IE) ELINITWSE(IE)
ELMINERODELEV(IE)

9.2.1.1 Example of a .FED file

1773 954 5 5
1 798427.180 305984.903 160.800 0.000 -9999.000 0 0
2 798375.129 305901.823 160.400 0.000 -9999.000 11 0
3 798456.481 305851.866 160.600 0.000 -9999.000 11 0
4 800145.687 304890.733 166.800 0.000 -9999.000 11 0
5 800164.855 304802.764 162.400 0.000 -9999.000 11 0
6 800219.908 304878.337 171.200 0.000 -9999.000 0 0
7 800083.504 304852.721 161.800 0.000 -9999.000 11 0
8 798619.614 309561.972 161.834 0.000 -9999.000 0 0
9 798519.268 309583.373 161.736 0.000 -9999.000 6 qin UN.dat
10 798566.714 309501.694 162.334 0.000 -9999.000 0 0
11 798579.326 309637.425 160.892 0.000 -9999.000 6 qin UN.dat
1 1 2 3 0.040 160.400 0.000 -9999.000 0.000
2 4 5 6 0.040 163.200 0.000 -9999.000 0.000
3 4 7 5 0.040 161.350 0.000 -9999.000 0.000
4 8 9 10 0.040 162.322 0.000 -9999.000 0.000
5 8 11 9 0.040 161.238 0.000 -9999.000 0.000
6 12 11 8 0.040 160.944 0.000 -9999.000 0.000
7 12 13 11 0.040 160.634 0.000 -9999.000 0.000
8 14 13 12 0.040 159.446 0.000 -9999.000 0.000
9 14 15 13 0.040 159.544 0.000 -9999.000 0.000
This mesh has 1773 elements, 954 nodes.
Table 9.3 Variable Descriptions for the .FED File.
VARIABLE
BCTYPE

TYPE
I

RANGE
-

UNITS
-

BCFILENAME

< 26

m or f t

DUMMY
I
ELINITWSE(IE) R

DESCRIPTION
Code to indicate type of open
boundary. See further details
about boundary conditions on
the .IFL file description below.
Boundary condition file name.
Should not contain spaces and
must have less than 26 characters. See further details on the
.IFL file description below.
Always equal to 2.
Initial water surface elevation
for
element/cell
EL.
Used in RiverFlow2D Plus
and RiverFlow2D Plus GPU.
Continued on next page

9.2

Mesh Data

130

Table 9.3 continued from previous page

VARIABLE
TYPE
RANGE
UNITS
DESCRIPTION
ELMINERODELEV
R
0
Minimum
erosion
elevation
(IE)
allowed at each cell.
Used
in
RiverFlow2D
Plus
and
RiverFlow2D Plus GPU.
ELZB (IE)
R
m or f t
Initial
bed
elevation
for
element/Cell
EL.
Used
in
RiverFlow2D
Plus
and
RiverFlow2D Plus GPU.
INITWSE(IN)
R
m or f t
Initial water surface elevation for
node IN.
IE
I
>0
Element index. Consecutive from
1 to NELEM.
IN
I
>0
Node number. Consecutive from
1 to NNODES.
MANNINGN(IE) R
>0
Mannings n value for element
IE.
MINERODELEV R
0
m or f t
Minimum erosion elevation al(IN)
lowed at each node.
NELEM
I
1-5
Number of triangular elements.
NNODES
I
>0
Number of triangular element
nodes.
NODE(IE,1),
I
>0
Node numbers for element IE
NODE(IE,2),
given in counter clockwise direcNODE(IE,3)
tion.
X(IN)
R
m or f t
X coordinate for node IN.
Y(IN)
R
m or f t
Y coordinate for node IN.
ZB (IN)
R
m or f t
Initial bed elevation for node IN.
Note: I = Integer variable. R = Real variable. S = Text variable..
All variables are separated by at least one blank space.

9.2.2 Open Boundary Conditions Data File: .IFL

This file contains boundary condition data used only internally by the model. This file is automatically generated by RiverFlow2D.
Line 1: Number of nodes on external boundary.
NNODESBOUNDARY
NNODESBOUNDARY lines containing the external boundary conditions data.
NODE BCTYPE BCFILENAME

9.2.2.1 Example of a .IFL file

1165
365
367
431
...

1
1
1

WSE97out.txt
WSE97out.txt
WSE97out.txt

This .IFL file has 1165 nodes on the boundary. Node 365 has BCTYPE=1 (Water Surface Elevation) and the time series of water surface elevations vs. time is in file WSE97out.txt.

131

Input Data File Reference

Table 9.4 Variable Descriptions for the .IFL File.

VARIABLE
BCTYPE

TYPE
I

RANGE
-

UNITS
-

BCFILENAME

< 26

NODE
I
NNODESBOUNDARY
I

DESCRIPTION
Code to indicate type of open
boundary. See Table ?? and comment 1.
Boundary condition file name.
Should not contain spaces and
must have less than 26 characters. See comments 2 and 3.
Node number.
Total number of nodes on boundary.

Note: I = Integer variable. S = Text variable.

Table 9.5 Boundary Condition Types available for the RiverFlow2D FE model.
BCTYPE
0
1
2

5
6
7

9
10
11
12
13-18
19
20-25
26
61

DESCRIPTION
Closed impermeable boundary. Slip boundary condition (no normal flow) is
imposed. See comment 5.
Imposes Water Surface Elevation. An associated boundary condition file
must be provided. See comments 2 and 4.
This code is only applicable for RiverFlow2D FE. Imposes U velocity component. An associated boundary condition file must be provided. See comment
2.
This code is only applicable for RiverFlow2D FE. Imposes V velocity component. An associated boundary condition file must be provided. See comment
2.
This code is only applicable for RiverFlow2D FE. Imposes U and V velocity
components. A two-variable boundary condition file must be provided. See
comment 3.
Imposes water discharge and water surface elevation. A two-variable boundary condition file must be provided. See comment 3.
Imposes water discharge. An associated boundary condition file must be
provided. See comment 2.
This code is only applicable for RiverFlow2D FE. Imposes U velocity and
water surface elevation. A two-variable boundary condition file must be
provided. See comment 3.
This code is only applicable for RiverFlow2D FE. Imposes V velocity and
water surface elevation. An associated boundary condition file must be provided. See comment 3.
Imposes single-valued stage-discharge rating table. An associated boundary
condition file must be provided. See comment 6.
Free inflow or outflow condition. Velocities and water surface elevations
are calculated by the model. See comment 7.
Free outflow condition. Velocities and water surface elevations are calculated by the model. Only outward flow is allowed. See comment 7.
Uniform flow outflow condition. See comment 10.
For future use.
Imposes single-valued stage-discharge rating table along an internal polyline.
An associated boundary condition file must be provided. See comment 8.
For future use.
Imposes water discharge and sediment discharge time series. An associated
boundary condition file must be provided. See comment 9.
This code is only applicable for RiverFlow2D FE. Imposes water discharge
creating sources on all inflow elements. An associated boundary condition
file must be provided. See comment 11.

9.2

Mesh Data

132

9.2.2.2 Comments for the .IFL file

1. RiverFlow2D allows having any number of inflow and outflow boundaries with various combinations of imposed conditions. Proper use of these conditions is a critical component of a
successful RiverFlow2D simulation. Theoretically, for subcritical flow it is required to provide
at least one condition at inflow boundaries and one for outflow boundaries. For supercritical flow all conditions must be imposed on the inflow boundaries and none on outflow
boundaries. Table ?? helps determining which conditions to use for most applications.
Table 9.6 Supercritical Flow Regime Boundary Conditions.
FLOW REGIME
BOUNDARY
Subcritical
Supercritical

AT

INFLOW BOUNDARY
CONDITION
Q or Velocity
Q and WSE

OUTFLOW BOUNDARY CONDITION

Water Surface Elevation
Free

NOTE: It is recommended to have at least one boundary where WSE or stage-discharge is

prescribed. Having only discharge and no WSE may result in instabilities due to violation of
the theoretical boundary condition requirements of the shallow water equations.
2. When imposing a single variable (water surface elevation, Q, U or V), the user must provide
an ASCII file with the time series for the corresponding variable. See section Boundary
Conditions Data Files for details on the format for one-variable boundary condition files.
3. When imposing two variables (water surface elevation and discharge, etc.), it is required to
provide an ASCII file with the time series for the variables. See section Boundary Conditions
Data Files for details on the format for two-variable boundary condition files.
4. When imposing water surface elevation it is important to check that the imposed value is
higher than the bed elevation. Even though RiverFlow2D can run with that condition, it
could lead to volume conservation errors.
5. A closed boundary condition is imposed by default on all boundary nodes. In this case,
the model calculates velocities and water surface elevations for all nodes on the boundary
depending on the value of the ISPLIPBC parameter. For example ISLIPBC = 1 will impose
slip conditions setting zero-flow across the boundary. Tangential flow is free corresponding
to a slip condition. In RiverFlow2D FE this condition can be overridden using ISLIPBC 1
or any of the other open inflow or outflow conditions described herein.
6. When using a single valued stage-discharge condition the model first computes the discharge
on the boundary then interpolates the corresponding water surface elevation from the rating
table and imposes that value for the next time step. In case the boundary is dry, it functions
as a free condition boundary (see comment 7). Water surface elevations are imposed only
on wet nodes. This condition requires providing an ASCII file with the table values entries.
See section Boundary Conditions Data Files for details on the file format. In general it is
preferable to use stage hydrograph rather than stage-discharge condition. In most small
slope rivers, the stage-discharge relationship is affected by hysteresis. In other words, the
stage-discharge curve is looped with higher discharges occurring on the rising limb than
on the rescission limb of the hydrograph. This is mainly caused by the depth gradient
in the flow direction that changes in sign throughout the hydrograph. In practice, this

133

Input Data File Reference

implies that there can be two possible stages for the same discharge. If the stage-discharge
relationship is not well known or if it just computed assuming steady state uniform flow,
it may lead to considerable errors when used as downstream boundary condition. That it
is why it is often preferred to use the stage hydrograph for that purpose. However, such
hydrograph may not be available to study changes in the river and evaluating proposed
conditions. For those cases, it is useful to use a stage-discharge relationship, preferably
measured over an extensive range of discharges. When this relationship is not available, one
option would be to assume steady state flow to determine a single-value rating curve. Since
this condition may generate wave reflection that can propagate upstream, it is important to
locate the downstream boundary on a reach sufficiently far from the area of interest, therefore
minimizing artificial backwater effects. Unfortunately, there is no general way to select such
place, but numerical experimenting with the actual model will be necessary to achieve a
reasonable location.
NOTE: loop stage-discharge relationships are not implemented in this RiverFlow2D version.

7. On free condition boundary nodes, the model calculates velocities and water surface elevations applying the full equations from the internal elements. No condition is imposed per se
on these nodes, which on the finite element context, is equivalent to apply natural boundary
conditions. In practice this should be equivalent to assume that derivatives of water surface
elevations and velocities are 0. It is advisable to use this condition when there is at least
another open boundary where WSE or stage-discharge is imposed.
8. When using a single valued stage-discharge condition on internal sections, the model first
computes the discharge across the boundary then interpolates the corresponding water surface elevation from the rating table, imposing that value for the next time step for all nodes
on the internal boundary. This condition requires providing an ASCII file with the table
values entries. See section Boundary Conditions Data Files for details on the file format.
9. When imposing a water and sediment discharge, it is required to provide an ASCII file with
the time series for water discharge and volumetric sediment discharge for each of the fractions. Note that sediment discharge is always expected in volume per unit time. See section
Boundary Conditions Data Files for details on the format for multiple-variable boundary
condition files.
10. The user must provide a file with the energy slope S0 for the corresponding boundary. This
file will only contain a single value S0 . The model will use S0 , Mannings n, and discharge to
create a rating table from which water surface elevations will be imposed as a function of the
computed outflow discharge. The rating table is calculated every 0.05 m (0.16 ft.) starting
from the lowest bed elevation in the outflow cross section up to 50 m (164 ft.) above the
highest bed elevation in the section. If S0 = 999, the model will calculate the average bed
slope perpendicular to the boundary line. Please, note than when letting the model calculate
the average bed slope, it uses the elevations on the elements adjacent to the boundary line,
which may result in adverse slopes or slopes that do not capture the general trend the reach.
11. This boundary condition is similar to the BCTYPE = 6 for inflow water discharge. However,
in this case, instead of converting the discharge into velocities that are imposed on all the
inflow nodes; the model creates sources on all the elements adjacent to the boundary line.
The condition then can be visualized as if the given discharge enters over the inflow elements.
For each time, the model evenly divides the discharge between all the inflow elements. For
example if there are Ne inflow elements and the imposed discharge is Qin, each element will

9.2

Mesh Data

134

receive a discharge equal to Qin/Ne. The water volume will naturally flow away from the
inflow depending on the bed slopes, etc. Care must be taken when the inflow boundary
elements have lower bed elevations than the surrounding elements. When imposing this
condition the user must provide an ASCII file with the discharge time series. See section
Boundary Conditions Data Files for details on the format for one-variable boundary condition
files.

9.2.3 Mesh Boundary Data File: .TBA

.TBA file is for internal use by the model and contains the list of boundary nodes in counterclockwise order for the external boundary polygon and in clockwise order for the internal boundaries.
This file is automatically generated by RiverFlow2D.
Line 1: Start of boundary indicator.
IBOUNDARYID
Line 2: Number of nodes in external boundary of mesh.
NNODESBOUNDARY
NNODESBOUNDARY lines containing the list of boundary nodes in counter clockwise direction.
BOUNDARYNODE (1:NNODESBOUNDARY)
The next lines are only used if there are islands in the mesh.
For each island:
Start of boundary parameter indicator for each island or internal closed contour.
IBOUNDARYID
Number of nodes in island boundary.
NNODESISLANDBOUNDARY
NNODESISLANDBOUNDARY lines containing the list of boundary nodes in clockwise direction.
ISLANDBOUNDARYNODE (1:NNODESISLANDBOUNDARY)

9.2.3.1 Example of a .TBA file

-9999
132
1
2
3
173
...
224
175
1
-9999
34
5
...

135

Input Data File Reference

5
In this example the external boundary has 132 nodes and there is one island with 34 nodes.
Table 9.7 Variable Descriptions for the .DAT File.
VARIABLE
IBOUNDARYID

TYPE
I

NNODESBOUNDARY

BOUNDARYNODE

NNODESISLANDBOUNDARY I
ISLANDBOUNDARYNODE

RANGE UNITS DESCRIPTION

-9999
Always = -9999. This value
is used to indicate the start
of a new boundary.
>0
Number of nodes on the
mesh external boundary.
>0
Node number on external
boundary. See comments 1
and 2.
>0
Number of nodes on island
boundary.
>0
Node number on island
boundary.

Note: I = Integer variable.

9.2.3.2 Comments for the .TBA file

1. There should be a single external boundary polygon and any number of internal islands or
closed contours.
2. The external boundary should also be the first on the file. The first boundary must always
be the external one. The internal boundaries as islands, piers, etc. should follow the external
domain polygon.

9.3 Component data

RiverFlow2D provides components to model specific hydraulic problems, including bridge piers,
culverts, internal rating tables, weirs, and sediment transport and pollutant transport. This section
explains the data file formats used in these components.

9.3.1 Bridges
RiverFlow2D provides four options to account for bridge piers. The most common option is to
create the pier plan geometry generating a 2D triangular-cell mesh that represents each pier as a
solid obstacle. In that case, the model will compute the flow around the pier and account for the
pier drag. This would be the preferred approach when the user needs to know the detailed flow
around the piers and the flow does not overtop the bridge deck. However, the resulting mesh may
have very small elements, leading to increasing computer times.
The second option (Bridge Piers) is a simplified formulation that does not require defining the
mesh around the piers, but will compute the pier drag force based on geometric data. This would
be the preferred approach when the flow does not overtop the bridge deck and the user does not
need to have detailed depiction of the flow around the piers but needs to account for the general
effect that the pier would have on the flow.
The third option represented in the Bridges component is a comprehensive bridge hydraulics
computation tool that does not require capturing bridge pier plan geometry in detail, therefore

9.3

Component data

136

allowing longer time steps, while allowing calculating the bridge hydraulics accounting for arbitrary
plan alignment, complex bridge geometry, free surface flow, pressure flow, overtopping, combined
pressure flow and overtopping, and submergence all in 2D. This is the recommended option for
most bridges.
There is a fourth option using the Internal Rating Table component, but for most applications
it is recommended to use one of the above since they better represent the bridge hydraulics.

9.3.2 Bridges Data File: .BRIDGES

This component requires the .BRIDGES data file that is automatically generated by the model based
on the geometrical representation entered in the SMS GUI. The .BRIDGES file has the following
format:
Line 1: Number of bridges.
NUMBEROFBRIDGES
NUMBEROFBRIDGES lines containing the data for each bridge.
Bridge Id.
BRIDGE ID
Bridge Cross Section Geometry file name.
BRIDGE GEOMETRY FILE
Number of cells pairs along bridge alignment.
NC
NUMBEROFCELLS lines containing pairs of cells numbers along bridge alignment.
CELL A[1] CELL B[1]
...
CELL A[NC] CELL B[NC]

9.3.2.1 Example of a .BRIDGES file

1
BRIDGE1
1894878.176 586966.254 1895274.636 586613.844
BRIDGEGEOM.DAT
9
133
1294
131
1296
129
1298
127
1300
125
1302
123
1304
121
1306
119
1308
94
1310

137

Input Data File Reference

Table 9.8 Variable Descriptions for the .BRIDGES File.

VARIABLE
BRIDGE GEOMETRY FILE

TYPE
S

RANGE UNITS DESCRIPTION

< 26
Contains the geometry of the
bridge cross section as explained below.
BRIDGE ID
S
< 26
Bridge ID.
CELL A[i] CELL B[i]
I
Cell pair along bridge alignment.
NC
I
>0
Number of cell pairs along
the bridge alignment.
NUMBEROFBRIDGES
I
>0
Number of bridges.
Note: I = Integer variable. S = Text variable.

9.3.3 Bridge Cross Section Geometry Data File

The bridge geometry cross section file is necessary to define the bridge cross section and is defined
by four polylines and the fined in five columns as follows:
Line 1: Number of points defining polylines.
NP
NP lines with these entries:
STATION(1) BEDELEV(1) ZLOWER(1) LOWCHORD(1) DECKELEV(1)
...
STATION(NP) BEDELEV(NP) ZLOWER(NP) LOWCHORD(NP) DECKELEV(NP)
The relationship between the four polylines must be as follows:
For all stations, STATION(I)STATION(I+1).
BEDELEVZLOWERLOWCHORDDECKELEV .
In a given line all elevations correspond to the same station.
The space between BEDELEV and ZLOWER is blocked to the flow.
The space between ZLOWER and LOWCHORD is open to the flow.
The space between LOWCHORD and DECKELEV is blocked to the flow.

9.3.3.1 Example of the Cross Section Geometry Data File

The following table is an example one of the geometry file that schematically represents the bridge
in Figure ??.

138

Input Data File Reference

Figure 9.2 Front view of a bridge cross section.

139

Input Data File Reference

Table 9.9 Variable Descriptions for the bridge cross section geometry file.
VARIABLE
BEDELEV

TYPE
R

RANGE
-

UNITS
m or f t

DECKELEV

m or f t

NP

>1

STATION

m or f t

ZLOWER

m or f t

LOWCHORD

m or f t

Note: R = Real variable. I = Integer variable.

DESCRIPTION
Bed elevation. Must be the lowest elevation for all polylines at a
given point.
Elevation of the bridge deck.
Must be the highest elevation for
all polylines at a given point.
Number of points defining cross
section polylines.
Distance from leftmost point
defining cross section polyline.
All polylines points must have a
common station.
Elevation of lower polyline.
ZLOWER must be larger or
equal to BEDELEV and smaller
or equal to LOWCHORD for a
given point. The space between
BEDELEV and ZLOWER is a
blocked area to the flow. The
space between ZLOWER and
LOWCHORD is open space.
If the bridge has no holes,
ZLOWER must be identical to
BEDELEV.
Elevation of the lower bridge
deck.
LOWCHORD must be
larger or equal to ZLOWER and
smaller or equal to DECKELEV
for a particular point. The space
between LOWCHORD and DECELEV is a blocked area to the
flow.

9.3

Component data

140

9.3.4 Bridge Piers Drag Forces File: .PIERS

This option requires the .PIERS data file that is automatically generated by the model based on the
geometrical representation entered in the SMS GUI. The .PIERS data file has the following format:
Line 1: Number of piers.
NUMBEROFPIERS
NUMBEROFPIERS lines containing the data for each pier.
X Y ANGLEX LENGTH WIDTH CD PIERID

9.3.4.1 Example of a .PIERS file

124
2042658.82
2042690.52
2042786.89
2042818.95
2042851.05
2042883.13
2042915.40
2040307.66
2040328.07
2040351.38
2040375.99

14214769.48
14214739.87
14214649.58
14214619.53
14214589.46
14214559.39
14214529.15
14214876.12
14214790.33
14214705.48
14214622.12

47.33
46.66
44.88
45.17
45.72
45.10
43.01
0.00
0.00
0.00
0.00

19.00
19.00
19.00
19.00
19.00
19.00
19.00
70.00
70.00
70.00
70.00

4.00
4.00
4.00
4.00
4.00
4.00
4.00
1.00
1.00
1.00
1.00

0.64
0.64
0.64
0.64
0.64
0.64
0.64
0.90
0.90
0.90
0.90

P1
P2
P3
P4
P6
P7
P8
P9
P10
P11
P12

Table 9.10 Variable Descriptions for the .PIERS File.

VARIABLE
ANGLEX

TYPE
R

RANGE
0 180

UNITS
Deg.

CD

0.5 2.5

LENGTH
PIERID

R
S

< 260

m or f t
-

WIDTH
R
X
R
Y
R
Note: R = Real variable. S =

m or f t
m or f t
m or f t
Text variable.

DESCRIPTION
Pier angle with respect to X axis.
See comment 1.
Non-dimensional drag coefficient
of the pier. See comment 2 and
Error! Reference source not
found.
Pier length.
Name of pier. Should not contain
spaces and must have less than 26
characters.
Pier width.
X coordinate of pier centroid.
Y coordinate of pier centroid.

141

Input Data File Reference

Table 9.11 Drag Coefficients for Bridge Piers. Adapted from Froehlich (2003).
PIER PLAN SHAPE AND APPROACH VELOCITY

DRAG COEFFICIENT CD

Round cylinder

1.0

Square cylinder

2.2

Square cylinder
Square cylinder
with
rounded corners

1.6
R/B
0
0.02
0.17
0.33

Hexagonal cylinder

CD
2.2
2.0
1.2
1.0

1.0

Hexagonal cylinder

0.7
L/B
1
2
4
6
L/B
1
2
4
6

Round-nosed section

Square-nosed section

CD
1.0
0.7
0.68
0.64
CD
2.2
1.8
1.3
0.9

9.3.4.2 Comments for the .PIERS File

1. Angle ANGLEX applies only to piers that are rectangular in plan. For example ANGLEX
= 90 corresponds to a pier whose longest axis is perpendicular to the X-axis.
2. The drag coefficient CD is related to the drag force though the following formula:
FD =

1
CD U 2 AP
2

(9.2)

where CD is the pier drag coefficient, is the water density, U is the water velocity, and AP
is the pier wetted area projected normal to the flow direction.
To account for the drag force that the pier exerts on the flow, RiverFlow2D converts it to the
distributed shear stress on the element where the pier centroid coordinate is located. The
resulting pier shear stress expressions in x and y directions are as follows:
px =

p
1
CD U U 2 + V 2
2

AP
Ae


(9.3)

9.3

Component data

142

py

p
1
= CD V U 2 + V 2
2

AP
Ae


(9.4)

where Ae is the element area.

9.3.5 Culverts Data File: .CULVERTS

The culvert component allows accounting for hydraulic structures that convey flow between two
locations. The discharge between the structure inflow and outflow ends will be computed based
on a user provided hydraulic structure rating table. The model will determine the flow direction
based on the hydraulic conditions on the structure ends.
Line 1: Number of culverts.
NCULVERTS
NCULVERTS groups containing
CulvertID
CulvertType
CulvertFile
X1 Y1 X2 Y2
...

9.3.5.1 Example of a .CULVERTS file

2
CulvertA
2
CulvertA.txt
799550.846 309455.307 799363.544 309031.842
CulvertB
1
CulvertB.txt
798858.644 309313.609 799153.441 309004.154

Table 9.12 Variable Descriptions for the .CULVERTS File.

VARIABLE
CulvertFile

TYPE
S

RANGE
< 26

UNITS
-

CulvertID

< 26

CulvertType

0, 1, 2

NCULVERTS
X1 Y1 X2 Y2

I
R

>0
-

Note: I = Integer variable. R = Real variable.

DESCRIPTION
Culvert rating table or culvert
characteristic file name. See next
section for details about the culvert characteristic file. Should
not contain spaces and must have
less than 26 characters.
Culvert name. Should not contain spaces and must have less
than 26 characters.
Type of culvert. See comments 1
and 2.
Number of culverts.
or f t
Coordinates of vertices defining
each culvert line.
S = Text variable.

143

Input Data File Reference

9.3.5.2 Culvert Depth-Discharge Rating table Data Files for CulvertType=0

This format applies to the culvert depth vs. discharge rating table.
Line 1: Number points in data series
NDATA
NDATA lines containing depth and discharge.
DEPTH(I) Q(I)
Where DEPTH(I) is depth corresponding to discharge Q(I).
INVERT Z1
INVERT Z2
Where INVERT Z1 and INVERT Z2 are the invert elevations for the inlet and outlet respectively.

9.3.5.3 Example of the Culvert Depth-Discharge Rating Table File

The following example shows a depth-discharge rating table for a culvert. NDATA is 7 and there
are 7 lines with pairs of depth and corresponding discharge:
7
0 0.20
0.1 1.00
1.00 36.09
2.00 60.00
3.00 84.78
4.00 110.01
100.00 110.02
5.0
1.0
Table 9.13 Variable Descriptions of Culvert Depth-Discharge Data files.
VARIABLE
NDATA
INVERT Z1

TYPE
I
R

RANGE
>0
>0

UNITS
m or f t

INVERT Z2

>0

m or f t

DEPTH
Q

R
R

>0
>0

m or f t
m3 /s or
f t3 /s
Note: I = Integer variable. R = Real variable.

DESCRIPTION
Number of lines in data file.
Inlet invert elevation.
If INVERT Z1 = -9999, the model
makes INVERT Z1 equal to the
average bed elevation of the inlet
element.
Outlet invert elevation. If INVERT Z2 = -9999, the model
makes INVERT Z2 equal to the
average bed elevation of the inlet
element.
Water depth.
Water discharge.

9.3

Component data

144

9.3.5.4 Culvert Characteristic Data Files for CulvertType = 1, 2

The culvert characteristic data has the following structure:
Nb
Ke
nc
Kp
M
Cp
Y
m
If CulvertType=1
Hb
Base
Else if CulvertType=2
Dc
INVERT Z1
INVERT Z2

9.3.5.5 Example of the culvert characteristic data file

1
0.5
0.012
1
1
1.1
0.6
-0.5
0.10
5.0
1.0
This example culvert characteristics data file indicates that the culvert one barrel (Nb =1),
Ke=0.4, nc=0.012, Kp=1, cp =1, M =1.1, Y=0.6, m=-0.5, and Dc=0.10, INVERT Z1=5.0 and
INVERT Z2 = 1.0.
Table 9.14 Variable Descriptions for the Culvert Characteristic file.
VARIABLE
Nb

TYPE
I

RANGE
-

UNITS
-

Ke

0-1

nc

0.01-0.1

0.1-2.0

0.6-2.0

DESCRIPTION
Number of identical barrels. The
computed discharge for a culvert
is multiplied by Nb to obtain the
total culvert discharge.
Entrance Loss Coefficient given
by Table ??.
Culvert Mannings n Coefficient
given by Table ??.
Inlet Control Coefficient given by
Table ??.
Inlet Control Coefficient given by
Table ??.
Continued on next page

145

Input Data File Reference

Table 9.14 continued from previous page

TYPE
RANGE
UNITS
DESCRIPTION
R
0.6-2.0
Inlet Control Coefficient given by
Table ??.
Y
R
0.5-1.0
Inlet Control Coefficient given by
Table ??.
m
R
0.7,-0.5
Inlet form coefficient.
m=0.7
for mitered inlets, m=-0.5 for all
other inlets.
Hb
R
>0
m or f t
Barrel Height for box culverts.
Only for CulvertType = 1.
Base
R
>0
m or f t
Barrel Width for box culverts.
Only for CulvertType = 1.
Dc
R
>0
m or f t
Diameter for circular culverts.
Only for CulvertType = 2.
INVERT Z1
R
>0
m or f t
Inlet invert elevation.
If INVERT Z1 = -9999, the model
makes INVERT Z1 equal to the
average bed elevation of the inlet.
INVERT Z2
R
>0
m or f t
Outlet invert elevation. If INVERT Z2 = -9999, the model
makes INVERT Z1 equal to the
average bed elevation of the inlet
element.
Note: I = Integer variable. R = Real variable.
VARIABLE
c

Table 9.15 Mannings n roughness coefficients for various culvert materials. Adapted from Froehlich
(2003).
Culvert barrel material

Concrete

Corrugated metal

Entrance description
Good joints, smooth walls
Projecting from fill, square-cut end
Poor joints, rough walls
2-2/3 inch 1/2 inch corrugations
6 inch 1 inch corrugations
5 inch 1 inch corrugations
3 inch 1 inch corrugations
6 inch 2 inch corrugations
9 inch 2 1/2 inch corrugations

Mannings
nc
0.012
0.015
0.017
0.025
0.024
0.026
0.028
0.034
0.035

Table 9.16 Entrance loss coefficients Ke . Adapted from Froehlich (2003).

Type of culvert

Concrete pipe

Entrance description*

Entrance loss
coefficient Ke
0.2
0.5

Projecting from fill, grooved end

Projecting from fill, square-cut end
Headwall or headwall with wingwalls
(concrete or cement sandbags)
Grooved pipe end
0.2
Square-cut pipe end
0.1
Rounded pipe end
0.7
Mitered end that conforms to embankment slope
Without grate
0.5
With grate
0.7
Continued on next page

9.3

Component data

Type of culvert

146
Table 9.16 continued from previous page
Entrance description*

Corrugated metal pipe

Projecting from embankment (no headwall)

Headwall with or without wingwalls
(concrete or cement sandbags)
or pipe-arch
Mitered end that conforms to embankment slope
Manufactured end section of metal or
concrete that conforms to embankment
slope
Without grate
With grate
Headwall parallel to embankment (no
wingwalls)
Square-edged on three sides
Rounded on three sides to radius of
1/12 of barrel dimension
Reinforced concrete box
Wingwalls at 30 to 75 to barrel
Square-edged at crown
Crown edge rounded to radius of
1/12 of barrel dimension
Wingwalls at 10 to 30 to barrel
Square-edged at crown
Wingwalls parallel to embankment
Square-edged at crown
See Table ?? for inlet configurations.

Entrance loss
coefficient Ke
0.9
0.5
0.7

0.5
0.7

0.5
0.2

0.4
0.2

0.5
0.7

Table 9.17 Culvert inlet control formula coefficients. Adapted from Froehlich (2003).
Barrel
material
Concrete
Concrete
Concrete
Cor.
metal
Cor.
metal
Cor.
metal
Concrete
Concrete
Concrete
Concrete
Concrete
Concrete
Concrete
Concrete
Concrete
Concrete
Concrete

Barrel
shape
Circular
Circular
Circular
Circular

Inlet description*

Headwall; square edge

Headwall; grooved edge
Projecting; grooved edge
Headwall

0.3153
0.2509
0.1448
0.2509

2.0000
2.0000
2.0000
2.0000

1.2804
0.9394
1.0198
1.2192

0.6700
0.7400
0.6900
0.6900

Circular

Mitered to slope

0.2112

1.3300

1.4895

0.7500

Circular

Projecting

0.4593

1.5000

1.7790

0.5400

0.1379
0.1379
0.1475

2.5000
2.5000
1.0000

0.9651
0.7817
1.2385

0.7400
0.8300
0.8100

0.2242

0.7500

1.2868

0.8000

0.2242

0.7500

1.3608

0.8200

1.6230

0.6670

0.9941

0.8000

1.5466

0.6670

0.8010

0.8300

1.6389

0.6670

1.2064

0.7900

1.5752
1.5466
1.6611

0.6670
0.6670
0.6670

1.0101
0.8107
1.2932

0.8200
0.8650
0.7300

Circular
Beveled ring; 45 bevels
Circular
Beveled ring; 33.7 bevels
Rectangular Wingwalls; 30 to 75 flares;
square edge
Rectangular Wingwalls; 90 and 15 flares;
square edge
Rectangular Wingwalls; 0 flares ;square
edge
Rectangular Wingwalls; 45 flare; beveled
edge
Rectangular Wingwalls; 18 to 33.7 flare;
beveled edge
Rectangular Headwall; 3/4 inch chamfers
Rectangular Headwall; 45 bevels
Rectangular Headwall; 33.7 bevels
Rectangular Headwall; 45 skew; 3/4 in
chamfers

Continued on next page

147

Barrel
material
Concrete
Concrete
Concrete
Concrete
Concrete
Concrete

Concrete
Concrete

Concrete

Cor.
metal
Cor.
metal
Cor.
metal
Concrete
Cor.
metal
Concrete
Concrete
Concrete
Concrete
Cor.
metal
Cor.
metal
Cor.
metal
Concrete
Concrete
Concrete
Concrete
Concrete
Concrete
Concrete
Concrete
Concrete
See Table

Input Data File Reference

Table 9.17 continued from previous page

Barrel
Inlet description*
K
M
shape
Rectangular Headwall; 30 skew; 3/4 in
1.6961 0.6670
chamfers
Rectangular Headwall; 15 skew; 3/4 in
1.7343 0.6670
chamfers
Rectangular Headwall;10-45 skew; 45 1.5848 0.6670
bevels
Rectangular Wingwalls;
non-offset 1.5816 0.6670
45/flares
Rectangular Wingwalls;
non-offset 1.5689 0.6670
18.4/flares; 3/4 in chamfers
Rectangular Wingwalls;
non-offset 1.5752 1.2418
18.4/flares;
30/skewed
0.6670
barrel
Rectangular Wingwalls; offset 45/flares; 1.5816 0.6670
beveled top edge
Rectangular Wingwalls;
offset 1.5752 0.6670
33.7/flares;
beveled top
edge
Rectangular Wingwalls;
offset 1.5689 0.6670
18.4/flares;
top
edge
bevel
Rectangular Headwall
0.2670 2.0000

1.3672

0.7050

1.4493

0.6800

1.0520

0.7500

1.0906

0.8030

1.1613

0.8060

0.7100

0.9715

0.8350

0.8107

0.8810

0.7303

0.8870

1.2192

0.6900

Rectangular Projecting; thick wall

0.3023

1.7500

1.3479

0.6400

Rectangular Projecting; thin wall

0.4593

1.5000

1.5956

0.5700

Circular
Circular

Tapered throat
Tapered throat

1.3991
1.5760

0.5550
0.6400

0.6305
0.9297

0.8900
0.9000

Rectangular
Circular
Circular
Circular
Circular

Tapered throat
Headwall; square edge
Headwall; grooved edge
Projecting; grooved edge
Headwall

1.5116
0.3153
0.2509
0.1448
0.2509

0.6670
2.0000
2.0000
2.0000
2.0000

0.5758
1.2804
0.9394
1.0198
1.2192

0.9700
0.6700
0.7400
0.6900
0.6900

Circular

Mitered to slope

0.2112

1.3300

1.4895

0.7500

Circular

Projecting

0.4593

1.5000

1.7790

0.5400

Circular
Beveled ring; 45 bevels
Circular
Beveled ring; 33.7 bevels
Rectangular Wingwalls; 30 to75 flares;
square edge
Rectangular Wingwalls; 90 and 15 flares;
square edge
Rectangular Wingwalls; 0 flares; square
edge
Rectangular Wingwalls; 45 flare; beveled
edge
Rectangular Wingwalls; 18 to 33.7 flare;
beveled edge
Rectangular Headwall; inch chamfers
Rectangular Headwall; 45 bevels
?? for inlet configurations.

0.1379
0.1379
0.1475

2.5000
2.5000
1.0000

0.9651
0.7817
1.2385

0.7400
0.8300
0.8100

0.2242

0.7500

1.2868

0.8000

0.2242

0.7500

1.3608

0.8200

1.6230

0.6670

0.9941

0.8000

1.5466

0.6670

0.8010

0.8300

1.6389
1.5752

0.6670
0.6670

1.2064
1.0101

0.7900
0.8200

9.3

Component data

148

Table 9.18 Culvert inlet configurations. (Adapted from www.xmswiki.com/xms/).

Inlet configuration

Description

End of the culvert barrel projects out of the embankment.

Grooved pipe for concrete culverts decreases energy

losses through the culvert entrance.

This option is for concrete pipe culverts.

Square edge with headwall is an entrance condition

where the culvert entrance is flush with the headwall.

Beveled edges is a tapered inlet edge that decreases

head loss as flow enters the culvert barrel.

Mitered entrance is when the culvert barrel is cut so

it is flush with the embankment slope.

Wingwalls are used when the culvert is shorter than

the embankment and prevents embankment material
from falling into the culvert.

9.3.5.6 Comments for the .CULVERTS and culvert characteristics files

1. For CulvertType 0, culvert discharge is computed using a given rating table on the CulvertFile
file.
2. For CulvertTypes 1 and 2, the model will calculate culvert discharge for inlet and outlet

149

Input Data File Reference

control using the FHWA procedures (Norman et al.,1985) that were later restated in dimensionless form by Froehlich (2003). See Users Guide for more details.

9.3.6 GATES Data Files: .GATES

This component requires the .GATES data file that is automatically generated by the model based
on the geometrical representation entered in the SMS GUI. The .GATES file has the following format:
Line 1: Number of gates.
NUMBEROFGATES
NUMBEROFGATES lines containing the data for each gate.
Gate Id
GATES ID
Crest elevation height Cd
CRESTELEV GATEHEIGHT CD
Time series of gate aperture
GATE APERTURES FILE
Number of cells pairs along gates alignment
NC
NUMBEROFCELLS lines containing pairs of cells numbers along gate alignment
CELL A[1] CELL B[1]
...
CELL A[NC] CELL B[NC]

9.3.6.1 Example of the .GATES File

2
Gate2
102.00
2.00
Gate2.dat
5
3105
29
3103
79
3101
87
3099
137
3097
141
GATE1
111.00
11.00
GATE1.dat
8
4099
285
4097
283
4033
281

1.720

1.710

9.3

Component data

4031
4029
4027
4026
4024

150

279
277
156
82
16
Table 9.19 Variable Descriptions for the .GATES File.
VARIABLE
CD

TYPE
R

CRESTELEV
R
GATE APERTURES FILE S
GATEHEIGHT
GATE ID
CELL A[i] CELL B[i]

R
S
I

NC

NUMBEROFGATES

Note: I = Integer variable. R

RANGE
>0

UNITS

DESCRIPTION
Gate discharge coefficient.
>0
Gate crest elevation.
< 26
Gate aperture time series.
>0
Gate height.
< 26
Gate ID.
Cell numbers of cell pairs
along gate alignment.
>0
Name of pier. Should not
contain spaces and must
have less than 26 characters.
>0
Number of cells along the
gate alignment.
= Real variable, S = Text variable.
-

9.3.7 Gate Aperture Time Series File

Line 1: Number of points in time series of gate aperture data.
NPOINTS
NPOINTS lines containing:
Time Aperture.
TIME APERTURE

9.3.7.1 Example of a Gates Aperture Data File

3
0
2
4

0.0
0.5
1.0
Table 9.20 Variable Descriptions for the .GATES File.
VARIABLE
NPOINTS

TYPE
I

RANGE
>1

UNITS
-

TIME
H(I)

R
R

>0
-

Hrs.
m-f t

Note: I = Integer variable. R = Real variable.

DESCRIPTION
Number of data points in the gate
aperture time series.
Time.
Gate aperture for the corresponding time.

151

Input Data File Reference

9.3.8 Internal Rating Table Data File: .IRT

This data file allows modeling complex hydraulic structures inside the modeling domain. The user
would enter polylines coincident with mesh nodes and assign a rating table of discharge vs. water
surface elevation to the polyline. In other words, the IRT polylines must connect nodes of the
triangular element mesh. For each time step, the model will compute the discharge crossing the
polyline and find by interpolation the corresponding water surface elevation from the provided rating table. The model will then impose that water surface elevation to all nodes along the polyline.
Velocities will be calculated using the standard 2D equations. Therefore, in internal rating table
polylines, computed velocities may not necessarily be perpendicular to the IRT polyline.
The file structure is as follows:
Line 1: Number of internal rating table (IRT) polylines.
IRT NPL
IRT NPL line groups containing the IRT polyline ID, the number of vertices defining each polyline,
the IRT boundary condition type (always equal to 19 in this version), the Rating Table file name,
followed by the list of polyline coordinate vertices as shown:
IRT ID
IRT NV IRT BCTYPE IRT FILENAME
X IRT(1) Y IRT(1)
X IRT(2) Y IRT(2)
...
X IRT(IRT NV) Y IRT(IRT NV)

9.3.8.1 Example of the .IRT file

2
IRT A
4 19 IRT A.DAT
799429.362 308905.287
799833.895 308354.857
799986.424 307738.111
799847.158 307141.259
IRT B
4 19 IRT A.DAT
799482.440 309453.678
799135.525 309118.164
798914.020 309269.634
798787.701 309467.583
This file has 2 internal rating table polylines, the ID of the first one is IRT A, which has 4
vertices, BCTYPE 19 and file name IRT A.DAT.
Table 9.21 Variable Descriptions for the .IRT File.
VARIABLE
IRT NPL
IRT NV

TYPE
I
I

RANGE
>0
2

UNITS
-

DESCRIPTION
Number of IRT polylines.
Number of points defining each
IRT polyline.
Continued on next page

9.3

Component data

152

Table 9.21 continued from previous page

TYPE
RANGE
UNITS DESCRIPTION
S
< 26
Name of IRT. Should not contain
spaces and must have less than 26
characters.
IRT BCTYPE
I
19
Boundary
condition
always
equals to 19 in this version
corresponding to discharge vs.
water surface elevation tables.
Future versions will include
further options.
X IRT Y IRT
R
m or f t
Coordinates of vertices defining
each IRT polyline. See comment
1.
IRT FILENAME
S
< 26
File name containing internal rating table in the format described
as a stage-discharge data file.
Should not contain spaces and
must have less than 26 characters.
Note: I = Integer variable. R = Real variable, S = Text variable.
VARIABLE
IRT ID

9.3.8.2 Comments for the .IRT file

1. IRT polylines should be defined avoiding abrupt direction changes (e.g. 90 degree turns).
Polyline alignments as such may create errors in the model algorithm that identifies the nodes
that lie over the polyline. Therefore, it is recommended that the IRT follow a more or less
smooth path.

9.3.9 Mud and Debris Flow Data File: .MUD

This file only applies to the RiverFlow2D Plus model. This file provides the parameters necessary
to model mud and debris flow using the RiverFlow2D Plus model.
Line 1: Flow resistance relation.
MF FRR
Line 2: Yield stress.
MF YS
Line 3: Bingham viscosity.
MF BVIS
Line 4: Internal friction angle.
MF THETA
Line 5: Material density.
MF DENS

9.3.9.1 Example of the .MUD file

1
300.

153

Input Data File Reference

0.00899
3.5
2200.

Table 9.22 Variable Descriptions for the .MUD File.

VARIABLE
MF BVIS
MF DENS
MF FRR

TYPE
I
R
R

RANGE
>0
>0
1-7

UNITS
Poise or lb/in2
kg/m3 or lb/f t3
-

MF THETA
R
>0
Degrees
MF YS
R
>0
N/m2 or lb/in2
Note: I = Integer variable. R = Real variable.

DESCRIPTION
Bingham viscosity.
Material density.
Flow resistance relation
(see Table ??).
Internal friction angle.
Yield stress.

9.3.9.2 Comments for the .MUD file

1. The equations that describe the tangential forces generated by the stresses can be lumped into
the same mathematical formula despite of having a different nature. In this way several shear
stresses formulations can be considered: turbulent stress t , dispersive stress d , Coulombtype frictional stress f , yield stress y and viscous stress . Table ?? includes turbulentcollisional, yield stress and laminar flow resistance terms in a quadratic rheological law, where
a standard value of = 24 is assumed. In this table is also included the input parameters
which are required by its computation.
Table 9.23 Flow resistance relation.
MF FFR Flow Resistance
Relation
1
Turbulent (default)
2
Full Bingham

Flow resistance term

Input parameters

b = t
b = 0 with f1 (b , 0 ) = 0

b = 1.5y + 3

Mannings n, density
Bingham viscosity, yield
stress, density
Bingham viscosity, yield
stress, density
Mannings n, density,
Friction angle
Mannings n, density,
yield stress
Mannings n, density,
yield stress, friction angle

Simplified
ham
Turbulent
Coulomb
Turbulent
Yield
Turbulent,
Coulomb
Yield
Quadratic

Granular flow

4
5
6

Bingand

b = t + f

and

b = t + y
b = t + min y , f

and
b = t + y +
b = f

Mannings n, density,
yield stress, friction angle
Friction angle, density.
For this formula, the
model automatically sets
Mannings n = 0 for all
cells.

9.3

Component data

154

9.3.10 Oil Spill on Land Data File: .OILP

This file only applies to the OilFlow2D Plus model. This file provides the parameters necessary to
model flow of viscous fluids including oil over complex terrain using the OilFlow2D model.
Line 1: Flow resistance relation.
OL FRR
Line 2: Yield stress. Not used in this release.
OL YS
Line 3: Fluid viscosity.
OL BVIS
Line 4: Internal friction angle. Not used in this release.
OL THETA
Line 5: Oil density.
OL DENS
Line 6: Temperature time series file.
OL TEMPTSERIES
Line 7: Temperature - Viscosity - Density table file.
OL TEMPVISCDENS

9.3.10.1 Example of the .OILP file

3
1.
0.00899
1
2200.
Temptseries.txt
Tempviscdensetable.txt

Table 9.24 Variable Descriptions for the .OILP File.

VARIABLE
OL BVIS

TYPE
I

RANGE UNITS
>0
P a s or
lb s/in2

OL DENS

>0

kg/m3 or
lb/f t3

OL FRR

OL TEMPTSERIES

26

DESCRIPTION
Fluid viscosity that will be
used if the lines corresponding to the Temperature-viscositydensity-table file and temperature data file are not provided.
Fluid density that will be used
if the lines corresponding to the
Temperature-viscosity-densitytable file and temperature data
file are not provided.
Flow resistance relation. Always
should be equal to 3 in this release.
Temperature time series file.
Continued on next page

155

Input Data File Reference

Table 9.24 continued from previous page

VARIABLE
TYPE
RANGE UNITS
DESCRIPTION
OL TEMPVISCDENS S
26
Temperature-viscosity-density
table file. The model will use
the temperature for a given time
to interpolate in the table the
viscosity and density.
OL THETA
R
>0
Degrees
Not used in this release.
OL YS
R
>0
N/m2 or Yield stress. Not used in this relb/in2
lease.
Note: I = Integer variable. R = Real variable. S = Text variable.

9.3.10.2 Example of the Temperature Time Series file

7
0.0
2.0
4.0
6.0
8.0
10.0
12.0

60
65
70
75
77
79
65

In this example there are 7 entries in the Temperature Time series. For the initial time 0.0
hours the temperature is 60 F , and so on.

9.3.10.3 Example of the Temperature-Viscosity-Density Table file

2
50 1.38E-06 58.00
120 6.94E-07 56.00
In this example there are 2 lines in the Temperature-Viscosity-Density table file. For 50 F the
viscosity is 1.38E-06 lb s/in2 and the density 50 lb/f t3 .

9.3.11 Pollutant Transport Module Data File: .SOLUTES

The pollutant transport module calculates concentration of passive or reactive pollutants (solutes)
based on advection and dispersion. The required date is included in the .SOLUTES data file that
has the following format:
Line 1: Number of pollutants (solutes)
NPOLLUTANTS MAX
Line 2: Number of pollutants used in this run
NPOLLUTANTS USED
Line 3: List of solutes used (all in one line)
I 1 I 2 ... I NPOLLUTANTS USED

9.3

Component data

156

Line 4: Longitudinal and transversal dispersion coefficients

DISPL DISPT
Line 5...: List of names of pollutants (one in each line)
NAME I1
NAME I1
...
NAME NPOLLUTANT MAX
Line 6 to NPOLLUTANT MAX+6: First order reaction rate coefficient matrix K(I,J)
K(1,1) K(1,2) ... K(1,NPOLLUTANT MAX)
K(2,1) K(2,2) ... K(2,NPOLLUTANT MAX)
...
K(NPOLLUTANT MAX,1) K(NPOLLUTANT MAX,2) ... K(NPOLLUTANT MAX,NPOLLUTANT MAX)

9.3.11.1 Example of the .SOLUTES file

5
5
1 2 3 4 5
0.1 0.02
NITRATE
AMMONIUM
TEMPERATURE
CHLORINE
PHOSPHATE
0.2 0.0 0.0 0.0
0.0 0.1 0.0 0.0
0.0 0.0 1.1 0.0
0.0 0.0 0.0 0.2
0.0 0.0 0.0 0.0

0.0
0.0
0.0
0.0
0.5
Table 9.25 Explanation of the example .SOLUTES file.

DATA
5
5
1 2 3 4 5
0.1 0.02
D
AMMONIUM
TEMPERATURE
CHLORINE
PHOSPHATE
0.2 0.0 0.0 0.0
0.0 0.1 0.0 0.0
0.0 0.0 1.1 0.0
0.0 0.0 0.0 0.2
0.0 0.0 0.0 0.0

0.0
0.0
0.0
0.0
0.5

DESCRIPTION
There are five pollutants
Five pollutants will be used in this run
The active pollutants will be 1 through 4
Longitudinal dispersion coefficient is 0.1 and
0.02.
Pollutant No. 1 name is: NITRATE
Pollutant No. 2 name is: AMMONIUM
Pollutant No. 3 name is: TEMPERATURE
Pollutant No. 4 name is: CHLORINE
Pollutant No. 5 name is: PHOSPHATE
First order reaction constant for pollutant No.
First order reaction constant for pollutant No.
First order reaction constant for pollutant No.
First order reaction constant for pollutant No.
First order reaction constant for pollutant No.

transversal =

1
2
3
4
5

is
is
is
is
is

0.2
0.1
1.1
0.2
0.2

157

Input Data File Reference

Table 9.26 Variable Descriptions for the .SOLUTES File.

VARIABLE
DISPL

TYPE
R

DISPT

K(I,J)

RANGE UNITS
>0
m2 /s or
f t2 /s
>0
m2 /s or
f t2 /s
s1

NPOLLUTANTS MAX

0, 1, 2

NPOLLUTANTS USED

>0

DESCRIPTION
Longitudinal dispersion coefficient.
Transversal dispersion coefficient.
First-order reaction constants, where K(I,I) is the
reaction constant for pollutant I and K(I,J) the reaction constant of pollutant I
with pollutant J.
Maximum number of pollutants.
Number of pollutants used
in the present run. Should
be equal to NPOLLUTANTS MAX

Note: I = Integer variable. R = Real variable.

9.3.12 Pollutant Transport Module Initial Concentration Data File:

.CINITIAL
The initial concentration data file is created by when the dataset with the initial concentrations is
exported from SMS based on the data entered as Initial Concentrations.
Each column in this file corresponds to each of the pollutants indicated in the corresponding
.SOLUTES file. Each line corresponds to the initial concentrations for each cell. Therefore the
.CINITIAL file must have NPOLLUTANTS MAX columns and NELEM lines.

9.3.13 Manning n Variable with Depth Data File: .MANNN

Use this file to enter spatially distributed Mannings n variable with depth data. The model assumes that the rainfall and evaporation can vary over the modeling area.
Line 1: Number of zones defined by polygons where Mannings n variable with depths are defined.
NNZONES
NRZONES group of lines containing Mannings n variable with depth data file for each zone
MANNNFILE
Number of vertices of polygon i
NPZONE[i]
List of NPZONE[i] vertex coordinates
X(1) Y(1)
...
X(NPZONE[i]) Y(NPZONE[i])

9.3.13.1 Example of the .MANNN file

2
Manning1.txt

9.3

Component data

158

4
25.0 25.0
25.0 75.0
75.0 75.0
75.0 25.0
Manning2.txt
4
25.0 125.0
25.0 175.0
75.0 175.0
75.0 125.0
In this example, there are two polygons. The Manning n data file for the first polygon is
Manning1.txt and the polygon is defined by four vertices.
Table 9.27 Variable Descriptions for the .MANNN File.
VARIABLE
NNZONE[i]
NNZONES
MANNNFILE
X(i) Y(i)

TYPE
I
I
S
R

RANGE
1
26
>0

UNITS
m or f t

DESCRIPTION
Number of vertices defining zone i.
Number of zones.
Mannings n file. See comment 1.
Vertex coordinates of the polygon
defining Zone i.
Note: I = Integer variable. R = Real variable. S = Text variable.

9.3.13.2 Comments for the .MANNN file

1. The spatial distribution of Mannings n variable with depth is given as a number of nonoverlapping polygons that would cover or not the mesh area. Zones not covered by any
polygon (complementary area) would be assigned the default Mannings n file.

9.3.13.3 Mannings n variable with depth data file

Line 1: Number of points in Mannings n file.
NPRE
ND lines containing:
DEPTH MANNINGS N

9.3.13.4 Example of a Mannings variable with depth data file

3
0. 0.1
0.3 0.1
1.0 0.03

159

Input Data File Reference

Table 9.28 Variable Descriptions for the Mannings n Cariable with Depth Data File.
VARIABLE
DEPTH
MANNINGS N

TYPE
R
R

RANGE UNITS
0
m or f t
0
-

ND
I
Note: I = Integer variable. R = Real variable.

DESCRIPTION
Flow depth. See comment 1.
Mannings
n
corresponding
DEPTH. See comment 1.
Number values in file.

to

9.3.13.5 Comments for the Mannigns n variable with depth data file
1. To calculate the Mannings n over the mesh, the model will first identify the polygon over
each cell and then will use the interpolated n value for cell depth from the table corresponding
to the polygon. In the example above, for all depth between 0.3 and 1, Mannings n will be
obtained by linear interpolation between 0.1 and 0.03.
2. The user should provide a DefaultManningsn.dat file in the project folder and the program
will apply the data contained in that file to the complementary area to the polygons provided.
If the DefaultManningsn.dat does not exist, the model will apply a default value of 0.035
to the areas not covered by Mannings n polygons.

9.3.14 Rainfall And Evaporation Data File: .LRAIN

Use this file to enter spatially distributed and time varying rainfall and evaporation data. The
model assumes that the rainfall and evaporation can vary over the modeling area.
Line 1: Number of zones defined by polygons where rainfall time series are defined.
NRZONES
NRZONES group of lines containing hyetograph and evaporation data file for each zone
RAINEVFILE[i]
Number of vertices of polygon i
NPZONE[i]
List of NPZONE[i] vertex coordinates
X(1) Y(1)
...
X(NPZONE[i]) Y(NPZONE[i])

9.3.14.1 Example of the .LRAIN file

2
hyeto1.txt
4
25.0 25.0
25.0 75.0
75.0 75.0
75.0 25.0
hyeto2.txt
4

9.3
25.0
25.0
75.0
75.0

Component data

160

125.0
175.0
175.0
125.0

In this example, there are two polygons. The rainfall and evaporation data file for the first
polygon is hyeto1.txt and the polygon is defined by four vertices.
Table 9.29 Variable Descriptions for the .LRAIN File.
VARIABLE
NPZONE[i]
NRZONES
RAINEVFILE
X(i) Y(i)

TYPE
I
I
S
R

RANGE
1
26
>0

UNITS
m or f t

DESCRIPTION
Number of vertices defining zone i.
Number of zones.
Rainfall intensity. See comment 1.
Vertex coordinates of the polygon
defining Zone i.
Note: I = Integer variable. R = Real variable. S = Text variable.

9.3.14.2 Comments for the .LRAIN file

1. The spatial distribution of rainfall and evaporation is given as a number of non-overlapping
polygons that would cover or not the mesh area. Zones not covered by any polygons would
have no rainfall or evaporation imposed onto the mesh.

9.3.14.3 Hyetograph and Evaporation data file

Line 1: Number of points in time series of rainfall and evaporation.
NPRE
NPRE lines containing:
Time Daily Rainfall, Daily Evaporation.
TIME RAINFALL EVAPORATION

9.3.14.4 Example of a Hyetograph and Evaporation data file

3
0.
24
48

0.0 1.0
4.0 2.0
12.0 3.0
Table 9.30 Variable Descriptions for the Hyetograph and Evaporation Data File.
VARIABLE
TYPE
EVAPORATION R
NPRE

RANGE UNITS
0
mm/hr
or in/hr
-

RAINFALL

mm/hr
or in/hr
TIME
R
>0
hours
Note: I = Integer variable. R = Real variable.

DESCRIPTION
Evaporation intensity. See comment
1.
Number of times in rainfall and evaporation time series.
Rainfall intensity. See comment 1.
Time interval

161

Input Data File Reference

9.3.14.5 Comments for the Hyetograph and Evaporation data file

1. To calculate the rainfall/evaporation over the mesh, the model will use the interpolated
rainfall and evaporation intensities between each time interval. For instance in the example above, for all times between 0 and 24 hours the rainfall will be determined by linear
interpolation between 0 mm/hr and 4 mm/hr, and so on for evaporation.
2. If the user has a DefaultRainEvap.dat file in the project folder, the program will apply the
data contained in that file to the complementary area to the polygons provided.

9.3.15 Infiltration Data File: .LINF

Use this file to enter spatially distributed infiltration parameters.
Line 1: Number of zones defined by polygons where infiltration parameters are defined.
NIZONES
NIZONES group of lines containing:
Infiltration data file for each zone
INFILFILE
Number of vertices of polygon i
NPZONE[i]
List of NPZONE[i] vertex coordinates
X(1) Y(1)
...
X(NPZONE[i]) Y(NPZONE[i])

9.3.15.1 Example of the .LINF file

2
inf1.inf
4
0.0 0.0
0.0 200.0
200.0 200.0
200.0 0.0
Inf2.inf
4
200.0 200.0
400.0 200.0
400.0 0.0
200.0 0.0
In this example, there are two polygons. The infiltration data file for the first polygon is
inf1.inf and the polygon is defined by four vertices.

9.3

Component data

162
Table 9.31 Variable Descriptions for the .LINF File.

VARIABLE
NPZONE[i]
NIZONES

TYPE
I
I

RANGE UNITS
1
-

DESCRIPTION
Number of vertices defining zone i.
Number of zones. See Comments 1 and
2.
INFILFILE
S
26
Infiltration parameter file.
X(i) Y(i)
R
>0
m or f t
Vertex coordinates of the polygon defining Zone i.
Note: I = Integer variable. R = Real variable. S = Text variable.

9.3.15.2 Comments for the .LINF file

1. The spatial distribution of infiltration parameters is given as a number of non-overlapping
polygons that would cover or not the mesh area. Zones not covered by any polygons would
have no infiltration loss calculated.
2. Each polygon can have a different infiltration method assigned.
3. If the user has a DefaultInfiltration.dat file in the project folder, the program will apply
the data contained in that file to the complementary area to the polygons provided.

9.3.15.3 Infiltration parameters data file

Line 1: Model to calculate infiltration.
INFILMODEL
Line 2: Number of infiltration parameters.
NIPARAM
If INFILMODEL = 1: Horton method then:
Line 3: K fc f0
If INFILMODEL = 2: Green and Ampt method then:
Line 3: KH PSI DELTATHETA
If INFILMODEL = 3: SCS-CN method then:
Line 3: CN POTRETCONST FUTUREUSE FUTUREUSE

9.3.15.4 Example of a Infiltration parameter data file

1
3
8.3E-04

3.47E-06

2.22E-5

In this example the infiltration loss method is set to 1 corresponding to the Horton model.
There are 3 parameters as follows: K = 8.3E-04, fc = 3.47E-06 and f0 = 2.22E-5.

163

Input Data File Reference

Table 9.32 Variable Descriptions for the Infiltration Parameter File.

VARIABLE
CN

TYPE
R

RANGE UNITS
>0
-

DeltaTheta

>0

fc

[0,5E-4]

m/s or f t/s

f0

[0,5E-4]

m/s or f t/s

INFILMODEL

1,2,3

[0,30]

1/s

Kh

m/s or f t/s

NIPARAM

[3-4]

POTRETCONST R

[0-1]

PSI

[0-1]

m or in

DESCRIPTION
Curve Number.
See USDA
(1986) to determine adequate values depending on land cover.
Typical values range from 10 for
highly permeable soils to 99 for
paved impermeable covers.
Difference between saturated and
initial volumetric moisture content. Default value = 3E-5.
Final infiltration rate. Default =
2E-5.
Initial infiltration rate. Default
= 7E-5.
Infiltration method. 1: Horton,
2: Green and Ampt, 3: SCS-CN.
Decay coefficient used in Horton
method. Default = 1.
Hydraulic conductivity used in
Green and Ampt method. Default = 0.00001.
Number of data parameters depending on the infiltration model
selected. Should be set as follows:
3 for Horton of Green and Ampt
methods, 4 for SCS-CN method.
Potential maximum retention
constant. Typically = 0.2.
Wetting front soil suction head.
Default = 0.05.

Note: I = Integer variable. R = Real variable.

9.3.16 RiverFlow2D FE Sediment Transport Data File: .SED

This file only applies to the RiverFlow2D FE model. Sediment transport files for the RiverFlow2D
Plus model are described in the next section.
The .SED file provides the parameters necessary to model sediment transport using the RiverFlow2D
FE model.
Line 1: Sediment discharge formula, sediment by fraction switch and number of sediment steps.
ST FORMULA STFRAC NSEDSTEP
Line 2: Sediment size in mm, sediment specific gravity, porosity and rating curve coefficients.
D50 D90 SG POROS A B
Line 3: Critical Shield Stress.
THETAC

9.3.16.1 Example of the .SED file

101
0.01 0.1 2.65 0.4 0 0
0.06

9.3

Component data

164
Table 9.33 Variable Descriptions for the .SED File.

VARIABLE
A,B

TYPE
R

RANGE
>0

UNITS
-

D50

>0

mm

D90

>0

mm

SG
POROS
STFRAC
NSEDSTEP

R
R
I
I

2-2.7
0.3-0.6
0,1
>0

ST FORMULA

0-9

DESCRIPTION
Coefficients in user defined rating formula for sediment discharge qs =
Aq B , where qs is the volumetric sediment discharge per unit width and q
is the volumetric discharge per unit
width. See comment 1.
Sediment median size. 50% of the sediment is finer than D50.
Sediment median size. 90% of the sediment is finer than D90. Only used for
Van Rijn Formula.
Sediment particle specific gravity.
Sediment porosity.
For future use.
The sediment transport will be calculated every NSEDSTEP of the hydrodynamic model. See comment 2.
Code indicating the sediment transport rate to use.
1. Rating Formula qs = Aq B
2. Meyer-Peter & Muller (1948)
3. Karim-Kennedy (1998)
4. Ackers-White (1975)
5. Yang (Sand)
6. Yang (Gravel)
7. Parker-Klingeman-Mclean
(1982)
8. Van Rijn (1984a-c)
9. Engelund Hansen (1967)

THETAC
R
0.03-0.06
Note: I = Integer variable. R = Real variable.

Critical Shield Stress

9.3.16.2 Comments for the .SED file

1. Sediment rating formula may be approximate to the total sediment discharge, and should be
used with caution. B-coefficient varies typically between 1.90 and 2.20 (see Table ?? below
from Garde & Ranga Raju, 1985).
Table 9.34 B-coefficient in formula qs = Aq B for various rivers (from Garde & Ranga Raju,
1985).
RIVER
Missouri river at Kansas (USA)
Red river (USA)
Colorado river (USA)
Hii river (Japan)
Toni river (Japan)
Kimotsuki river (Japan)
Kanzoko river (Japan)
Shinano river (Japan)
Chenab river at Khanki (Pakistan)

B
2.16
2.07
2.00
1.92
2.00
2.00
2.20
1.92
2.00

165

Input Data File Reference

A-coefficient should be ideally be calibrated, comparing model results with measurements.

Using untested coefficients may lead to out of range qs values and to model instability.
2. By default the model will compute sediment transport for every time step, couple with the
hydrodynamic computations. Must of the time, bed elevation changes occur at a much slower
rate than the hydrodynamic changes, and it makes sense to decouple the sediment transport
calculations. Use this parameter to model sediment transport every NSEDSTEP time steps.
For example, if NSEDSTEP is equal to 10, sediment transport will be calculated every 10
time steps of the hydrodynamic model.

9.3.17 RiverFlow2D Plus Sediment Transport Data Files: .SEDS

and .SEDB
This file only applies to the RiverFlow2D Plus model. Sediment transport files for the RiverFlow2D
FE model are described in the previous section.
This file provides the parameters necessary to model sediment transport using the RiverFlow2D
Plus model.

9.3.17.1 .SEDS file for suspended sediment data

The .SEDS file provides the data for the suspended sediment transport model.
Line 1: Suspended sediment activation switch.
ISSACT
Line 2: Number of suspended sediment fractions.
NSSNFRAC
Line 3: Equilibrium concentration formula.
ISSTF
Line 4: Sediment density for each fraction.
SSDEN(1) ... SSDEN(NSSNFRAC)
Line 5: Initial suspended sediment concentration for each fraction.
INICON(1) ... INICON(NSSNFRAC)
Line 6: For future use.
DUMMY1(1) ... DUMMY1(NSSNFRAC)
Line 7: Suspended sediment D50 size for each fraction.
D50(1) ... D50(NSSNFRAC)
Line 8: For future use.
DUMMY2(1) ... DUMMY2(NSSNFRAC)
Line 9: Porosity for each fraction.
SSPOR(1) ... SSPOR(NSSNFRAC)

9.3

Component data

166

Line 10: Critical Shield Stress for each fraction.

THETAC(1) ... THETAC(NSSNFRAC)
Line 11: Friction angle for each fraction.
FRICANG(1) ... FRICANG(NSSNFRAC)
Line 12: Equilibrium concentration formula factor for each fraction.
SSTFACT(1) ... SSTFACT(NSSNFRAC)
Line 13: Settling velocity formula.
ISETFOR
Line 14: Settling velocity formula factor.
SETFORFACT

9.3.17.2 Example of the .SEDS file

1
2
1
165.00 165.00
0.01 0.01
0.003937008 0.003937008
0.0039370 0.0039370
0.003937008 0.003937008
0.40 0.40
0.047 0.047
35.00 35.00
1.00 1.00
1
1
Table 9.35 Variable Descriptions for the .SEDS File.
VARIABLE
INICON

TYPE
R

RANGE UNITS
0,0.3
-

ISSACT

0,1

NSSNFRAC

1-10

ISSTF

1,2

DESCRIPTION
Initial volumetric sediment concentration. See comment 1.
Suspended sediment activation switch.
If ISSACT = 1the model will compute
suspended sediment transport. If ISSACT = 0 suspended sediment transport
will not be calculated.
Number of suspended sediment fractions.
Equilibrium concentration formula.
This code indicates the formula
according to this list:
1. Bagnold (1966)
2. Van Rijn (1984a)

DUMMY1

>0

Not used in this release, but must be

present.
Continued on next page

167

VARIABLE
D50

DUMMY2
SSPOR
THETAC
FRICANG
ISETFOR

Input Data File Reference

Table 9.35 continued from previous page

TYPE
RANGE UNITS
DESCRIPTION
R
>0
mm or in
Sediment median size. 50% of the sediment is finer than D50. See comment
1.
R
>0
Not used in this release, but must be
present.
R
0.3-0.6
Porosity. See comment 1.
R
0.03Critical Shield Stress. See comment 1.
0.06
R
5-45
Friction angle. See comment 1.
I
0-9
Settling velocity formula. It is a unique
formula for all fractions. This code indicates the formula according to this list:
1. Rubey (1983)
2. Zhang (1961)
3. Zanke (1977)
4. Van Rijn (1984a)
5. Raudkivi (1990)
6. Julien (1998)
7. Cheng (1997)
8. Jimenez-Madsen (2003)
9. Wu-Wong (2006)

SETFORFACT R

SSDEN

SSTFACT

kg/m3 or
lb/f t3
-

Settling velocity formula factor. This

factor multiplies the settling velocity
calculated by the formula selected in
ISETFOR. Its a factor that may be
used for calibrating the model.
Suspended sediment density. See comment 1.
Equilibrium concentration formula factor for each fraction. This factor multiplies the equilibrium concentration formula ISSTF. Its a factor that may be
used for calibrating the model. See
comment 1.

Note: I = Integer variable. R = Real variable.

9.3.17.3 Comments for the .SEDS file

1. There should be one value for each sediment fraction up to NSSNFRAC.

9.3.17.4 .SEDB file for bed load transport data

The .SEDS file provides the data for the bed load transport model.
Line 1: Bed load sediment transport activation switch.
IBLACT
Line 2: Number of bed load sediment fractions.
NBLNFRAC

9.3

Component data

168

Line 3: Sediment transport formula.

IBLTF
Line 4: Sediment density for each fraction.
BLDEN(1) ... BLDEN (NBLNFRAC)
Line 5: Sediment D30 size for each fraction.
D30(1) ... D30(NBLNFRAC)
Line 6: Sediment D50 size for each fraction.
D50(1) ... D50(NBLNFRAC)
Line 7: Sediment D90 size for each fraction.
D90(1) ... D90(NBLNFRAC)
Line 8: Porosity for each fraction.
BLPOR(1) ... BLPOR(NSSNFRAC)
Line 9: Critical Shield Stress for each fraction.
THETAC (1) ... THETAC (NBLNFRAC)
Line 10: Friction angle for each fraction.
FRICANG (1) ... FRICANG (NBLNFRAC)
Line 11: Fractions on the bed.
BEDFRACT (1) ... BEDFRACT (NBLNFRAC)
Line 12: Transport formula factor for each fraction.
BLFORFACT(1) ... BLFORFACT (NBLNFRAC)

9.3.17.5 Example of the .SEDB file

1
2
2
165.0 165.0
0.0039370 0.0039370
0.0039000 0.0039370
0.0039370 0.0039370
0.40 0.40
0.047 0.047
35.00 35.00
0.50 0.50
1.00 1.00
Table 9.36 Variable Descriptions for the .SEDS File.
VARIABLE
IBLACT

TYPE
I

RANGE UNITS
0,1
-

DESCRIPTION
Bed load transport activation switch. If
IBLACT = 1the model will compute bed
load sediment transport. If IBLACT = 0
bed load transport will not be calculated.
Continued on next page

169

VARIABLE
NBLNFRAC
IBLTF

TYPE
I
I

Table 9.36 continued

RANGE UNITS
1-10
1,10
-

Input Data File Reference

from previous page

DESCRIPTION
Number of sediment fractions.
Sediment transport formula. This code indicates the formula according to this list:
1. Meyer-Peter & Muller (1948)
2. Ashida (1972)
3. Engelund (1976)
4. Fernandez (1976)
5. Parker fit to Einstein (1979)
6. Smart (1984)
7. Nielsen (1992)
8. Wong 1 (2003)
9. Wong 2 (2003)
10. Camenen-Larson (2005)

D30

>0

m or in

D50

>0

m or in

D90

>0

m or in

BLPOR
THETAC

R
R

FRICANG
BEDFRACT

R
R

0.3-0.6
0.030.06
5-45
-

BLDEN

BLFORFACT R

kg/m3
or
lb/f t3
-

Sediment D30 size. 30% of the sediment

is finer than D30. Only used for Smart
Formula. See comment 1.
Sediment median size. 50% of the sediment is finer than D50. See comment 2.
Sediment D90 size. 90% of the sediment
is finer than D90. Only used for Smart
Formula. See comment 1.
Porosity. See comment 1.
Critical Shields Stress. See comment 3.
Friction angle. See comment 1.
Sediment fraction. The sum of all fractions
should add to 1. See comment 1.
Sediment density. See comment 1.

Transport formula factor for each fraction.

This factor multiplies the result of the
transport formula selected (IBLTF). Its a
factor that may be used for calibrating the
model. See comment 1.

Note: I = Integer variable. R = Real variable.

9.3.17.6 Comments for the .SEDB file

1. There should be one value for each sediment fraction up to NBLNFRAC.
2. Characteristic diameter for all sediment transport formulas.
3. The default critical Shields stress is 0.047.

9.3.18 Sources and Sinks Data File: .SOURCES

Use this file to enter data to simulate point inflows or outflows at any location. This feature is
typically used when modeling intakes (outflow) or point inflows. The user may provide time varying hydrographs that will be applied to each point.
Line 1: Number of source and sink points.
NSOURCES

9.3

Component data

170

NSOURCES groups of lines containing source/sink point identification text, name of the file containing the discharge time series, and the coordinates of the point as follows:
SOURCEID
ISFILENAME
X S(I) Y S(I)
...

9.3.18.1 Example of the .SOURCES file

2
IntakeA
Intake.txt
799019.633 309402.572
DischargeIn
Discharge.txt
799222.740 309048.493
This file has 2 sources/sinks. The first one is named IntakeA located at coordinate: X = 799019.633
and Y = 309402.572 and discharge data is on file: Intake.txt.
Table 9.37 Variable Descriptions for the .SOURCES File.
VARIABLE TYPE
NSOURCES
I
ISFILENAME S

RANGE UNITS
>0
-

DESCRIPTION
Number of source or sink points.
Name of file containing the time series of
each point source or sink. Must not contain blank spaces. See comments 1 and 2.
SOURCEID
S
< 26
Name of point source or sink. Should have
less than 26 characters and must not contain blank spaces.
XSYS
R
m or f t
Coordinates of source/sink.
Note: I = Integer variable. R = Real variable. S = Text variable.

9.3.18.2 Comments for the .SOURCES File

1. The file name is arbitrary but must not contain blank spaces. The file format is the same as
the one variable boundary condition file described in Section ??.
2. To model inflows use positive discharge values and to model outflows use negative values.

9.3.19 Weirs Data File: .WEIRS

This data file allows using weir calculations along user defined polylines representing road or weir
overtopping. The user selects the weir coefficient associated with each weir and the model will
determine the nodes on each polyline and the discharge across each pair of nodes based on the weir
formulae and methods described on Hydraulics of Bridge Waterways FHWA, 1978 (see comment
1). The present version allows defining a variable crest elevation along the weir polyline.
Line 1: Number of weir polylines.
NWEIRS

171

Input Data File Reference

NWEIRS group of lines including weir ID, number of vertices defining each weir polyline, the weir
coefficient followed by the coordinates each vertex as shown:
WEIR ID
NV CF WRCRESTELEV
X W(1) Y W(1) WRCREST(1)
X W(2) Y W(2) WRCREST(2)
...
X W(NV) Y W(NV) WRCREST(NV)

9.3.19.1 Example of the .WEIRS file

2
WEIR A
4 3.0 -999
799429.362 308905.287
799833.895 308354.857
799986.424 307738.111
799847.158 307141.259
WEIR B
4 3.04 -999
799482.440 309453.678
799135.525 309118.164
798914.020 309269.634
798787.701 309467.583

200.
201.
202.
203.

203.5
204.0
204.9
205.0

This file has 2 weirs. The first one is named WEIR A and is defined by a polyline with 4
vertices. Weir discharge coefficient is equal to 3.0.
Table 9.38 Variable Descriptions for the .WEIRS File.
VARIABLE
CF
NWEIRS
NV

TYPE
R
I
I

RANGE
>0
>0
2

UNITS
-

WEIR ID

< 26

m of f t

WRCRESTELEVR

WRCREST(I)
R
m of f t
X W(I)
R
m of f t
Y W(I)
Note: I = Integer variable. R = Real variable.

DESCRIPTION
Weir coefficient. See comment 1.
Number of weir polylines.
Number of points defining each weir
polyline.
Name of weir. Should have less than 26
characters and must not contain blank
spaces.
Weir crest elevation for all the weir. If
WRCRESTELEV = -999 a weir elevation is provided for each weir polyline
vertex.
Weir crest elevation for vertex I.
Coordinates of vertices defining each
weir polyline. See comment 2.
S = Text variable.

9.3

Component data

172

9.3.19.2 Comments for the .WEIRS File

1. Weir discharge is computed between pairs of nodes along the polyline based on the following
formula:
Q = Cf LH 3/2

(9.5)

where L is the distance between nodes, H is the total head upstream of the polyline segment
and Cf is the discharge coefficient. The model checks for submergence and it occurs Cf will
be corrected according to the correction factor defined by (FHWA, 2001).
2. Weir polylines should be defined avoiding abrupt direction changes (e.g. 90 degree turns),
because such angles may create errors in the algorithm that identifies the nodes that lie over
the polyline.

9.3.20 Wind Data File: .WIND

Use this file to enter spatially distributed and time varying wind velocity data. The model assumes
that the wind velocity can vary over the modeling area. The user should provide a set of polygons
and a time series of velocities for each polygon.
Line 1: Number of zones defined by polygons where wind velocity time series are defined.
NWZONES
Line 2: Wind stress coefficient.
CD
Line 3: Air density.
AIRDENSITY
NWZONES group of lines containing hyetograph and evaporation data file for each zone.
WINDFILE
Number of vertices of polygon i.
NPZONE[i]
List of NPZONE[i] vertex coordinates.
X(1) Y(1)
...
X(NPZONE[i]) Y(NPZONE[i])

9.3.20.1 Example of the .WIND file

2
0.009
1.225
Wind1.txt
4
25.0 25.0
25.0 75.0
75.0 75.0

173

Input Data File Reference

75.0 25.0
Wind2.txt
4
25.0 125.0
25.0 175.0
75.0 175.0
75.0 125.0
In this example, there are two polygons. The Cd coefficient is set to 0.009 and the wind density
to 1.225 kg/m3 . The wind velocity file for the first polygon is wind1.txt and the polygon is
defined by four vertices.
Table 9.39 Variable Descriptions for the .WIND File.
VARIABLE
AIRDENSITY

TYPE
R

RANGE UNITS
0
-

DESCRIPTION
Air density. Always given in metric
units. Default = 1.225.
CD
R
0
Wind stress coefficient. Always given in
metric units. Default = 0.009.
NPZONE[i]
I
1
Number of vertices defining zone i.
NWZONES
I
Number of zones.
WINDFILE
S
26
Wind velocity vector time series file.
See Comment 1.
X(I) Y(I)
R
>0
m or f t
Vertex coordinates of the polygon defining Zone i.
Note: I = Integer variable. R = Real variable. S = Text variable.

9.3.20.2 Comments for the .WIND File

1. The spatial distribution of wind is given as a number of non-overlapping polygons that would
cover or not the mesh area. Zones not covered by any polygons will be considered as having
no wind stress.
2. If the user has a DefaultWind.dat file in the project folder, the program will apply the
data contained in that file to the complementary area to the polygons provided.

9.3.20.3 Wind Velocity Data File

Line 1: Number of points in time series of wind velocity data.
NPOINTS
NPOINTS lines containing:
Time Wind velocity component in X and Y directions.
TIME UX UY

9.3.20.4 Example of a Wind Velocity and Data File

3
0.
24
48

0.0
4.0
4.0

0.0
-3.0
-3.0

9.4

Output control data

174

Table 9.40 Variable Descriptions for the Wind Velocity File

VARIABLE
NPOINTS

TYPE
I

RANGE UNITS
>1
-

TIME
UX(I) UY(I)

R
R

>0
-

Hrs
m/s or f t/s

DESCRIPTION
Number of data points in the wind
velocity time series.
Time.
Wind velocity components in x and
y directions.

Note: I = Integer variable. R = Real variable.

9.4 Output control data

9.4.1 Observation Points Data File: .OBS
This file contains data to allow the model reporting time series of results at user specified locations
based on point coordinates. The user would indicate the number of observation points and then
the list of point coordinates.
Line 1: Number of observation points.
NOBSPOINTS
NOBSPOINTS groups of lines containing the observation point ID, and coordinate of each point:
ObsID
X OP(I) Y OP(I)
...

9.4.1.1 Example of the .OBS file

3
PointA
798798.380 309627.950
PointB
799146.926 309430.876
PointC
799721.8608 309041.615
This .OBS file has three points. The first point is named PointA and has coordinates: X=798798.380
Y=309627.950.
Table 9.41 Variable Descriptions for the .OBS File
VARIABLE
ObsID

TYPE
S

RANGE UNITS
< 26
-

DESCRIPTION
Name of observation point. Should
have less than 26 characters and
must not contain blank spaces.
NOBSPOINTS I
>0
Number of observation points.
X OP Y OP
R
m or f t
Coordinates of point.
Note: I = Integer variable. R = Real variable. S = Text variable.

175

Input Data File Reference

9.4.2 Graphical Output Control Data File: .PLT

This file provides parameters to control graphic output options for plots generated while the model
is running.
Line 1: Plot control variables.
IGRAPHCODE COLORSCHEME IAXES IDXF IGRAPHFILES IVSF
Line 2: Velocity vector scale multiplier.
SF MULT
Line 3: Coordinates for plot window.
XMING XMAXG YMING YMAXG
Line 4: Limits of plotted variable.
MINVARG MAXVARG
Line 5: Maximum velocity to plot.
MAXVELOC
Line 6: Transparency.
USEBACKIMAGE
Line 7: Transparency.
TRANSP
Line 8: Background aerial image.
IMAGEFILE
Line 9: Background aerial image world file.
IMAGEWF

9.4.2.1 Example of the .PLT file

100 5 1 0 1 0
5
0 0 0 0
0 0
7
1
0.6
C:\Projects\Example\Aerial.gif
C:\Projects\Example\Aerial.gwf
Table 9.42 Variable Descriptions for the .PLT File
VARIABLE
TYPE RANGE UNITS
COLORSCHEMEI
1-6
-

DESCRIPTION
Variable to select plot option. See comment
1.
Continued on next page

9.4

Output control data

VARIABLE
IAXES

176

Table 9.42 continued from previous page

TYPE RANGE UNITS DESCRIPTION
I
0,1
Switch to control weather to plot axes.
1. Do not plot X and Y axes.
2. Plot X and Y axes.

IDXF

0,1

Switch to control velocity field output in

DXF CAD format.
1. Do not output DXF mesh and velocity
field.
2. Create mesh and velocity field DXF
files for each output time.

IGRAPHCODE I

100,
101,
102,
103,
110,
201,
202,
203,
204,
600-610

Parameter to indicate the plot type to display

while the program is running.
100: Plot velocity field using vectors
in black.
101: Plot velocity field using vectors
in black over depths.
102: Plot velocity field over bed elevations.
103: Plot water elevations.
110: Plot velocity field using vectors
in color.
201: Plot depths.
202: Plot bed elevations.
203: Plot velocity field over water elevations.
204: Plot bed elevation changes.
600-610: Plot suspended sediment or
pollutant concentrations.

IGRAPHFILES I

0,1

Variable to control whether to output

graphic files.
1. Do not output graphic files.
2. Output graphic files.

IMAGEFILE

IMAGEWF

IVSF

0,1

Name of aerial image file including path and

extension. Supported formats include .BMP,
.GIF, .PNG and .TIFF. Other graphic file formats will be included in forthcoming releases.
Name of aerial image world file including
path and extension. World file format should
follow ESRI specifications. See comment 2.
Switch to control velocity field output in a
file that allows creating shapefiles.
1. Do not output SF velocity field.
2. Create velocity field SF files for each
output time. See comment 3.
Continued on next page

177

VARIABLE
MAXVELOC

Input Data File Reference

Table 9.42 continued from previous page

TYPE RANGE UNITS DESCRIPTION
R
m/s or Use this variable to control the maximum vef t/s
locity displayed in vector plots.
If MAXVELOC = 0, the whole velocity range will be plotted.
If MAXVELOC > 0, it will define the
maximum velocity to be displayed.

MINVARG,
MAXVARG

SF MULT

>1

TRANSP

[0,1]

USEBACKIMAGE
I

[0,1]

R
XMING,
XMAXG,
YMING,
YMAXG
Note: I = Integer variable. R

These variables define the minimum and

maximum values to be displayed of the selected variable. If equal to 0, the maximum
range will be displayed.
Variable to control velocity vector scale. Use
this variable to adjust velocity vectors. Velocities will be scaled according to SF MULT.
Variable to control plot transparency when
using a background image. TRANSP should
be in the range from 0.0 to 1.0, where
0.0 means a fully transparent color and 1.0
means a fully opaque color.
Variable to controls whether to use a background image for dynamic plots during
model run. If value is = 1, the plot will include as background the image provided in
IMAGEFILE and IMAGEWF.
m or f t
These variables indicate the coordinates of a
rectangle that define the plot window. If all
values are 0, the full extent of the modeling
area will be displayed.
= Real variable. S = Text variable.

9.4.2.2 Comments for the .PLT file

1. COLORSCHEME defines the color palette that will be used for all plots. The available
palettes are shown in this figure:

Figure 9.3 Color palettes.

2. The generic meaning of world file parameters are:

9.4

Output control data

178

Line 1: A: pixel size in the x-direction in map units/pixel

Line 2: D: rotation about y-axis (ignored in this version)
Line 3: B: rotation about x-axis (ignored in this version)
Line 4: E: pixel size in the y-direction in map units, almost always negative
Line 5: C: x-coordinate of the center of the upper left pixel
Line 6: F: y-coordinate of the center of the upper left pixel.

Example:
2.05
0.00
0.00
-2.05
795944.99
310049.73
In this example, 2.05 is the pixel size in x-direction, rotation in x and y axes is 0.00, pixel
size in y direction is 2.05 (shown in negative), x-coordinate of upper left pixel is 795944.99
and y-coordinate of upper left pixel is 310049.73.
The following table indicates the supported image formats and their corresponding world file
extensions.
Table 9.43 Supported image formats and their corresponding world file extensions.
IMAGE FILE
MAT
.BMP
.GIF
.PNG
.TIF, .TIFF

FOR-

WORLD FILE EXTENSION

.BMPW, .BPW
.GFW, .GIFW, .WLD
.PGW, .PNGW, .WLD
.TFW, .WLD

3. Vector field, depth and water surface elevation text files are used for creating vector field
shapefiles. Using the IVSF = 1 option, will allow the model to generate ESRI Shapefile
support files for each report interval. This option is not used in the RiverFlow2D SMS.

179

Input Data File Reference

9.4.3 Data for Profile Result Output: .PROFILES

Use this file to provide profiles (polylines) along which results will be generated.
Line 1: Number of profiles.
NPROFILES
NPROFILES group of files including: Profile ID, number of vertices in profile I, the number of
intervals to divide each profile, and coordinates for each vertex in polyline.
PROFILEID
NVERTICES PR(I) ND PR
X PRF(I), Y PRF(I)
...

9.4.3.1 Example of the .PROFILES file

2
ProfileA
2 10
800500.45 306895.63
799095.07 307457.34
ProfileB
3 10
800503.45 306896.63
799500.00 306900.00
799095.07 307457.34
This file indicates there are 2 profiles. First profile ID is: ProfileA which is defined with a
2-vertex polyline and will be divided in 10 segments.
Table 9.44 Variable Descriptions for the .PROFILES File
VARIABLE
ND PR

TYPE RANGE UNITS

I
>2
-

DESCRIPTION
Intervals to divide each profile subsegment between vertices. Results will
be reported at each interval.
NPROFILES
I
>0
Number of profiles.
NVERTICES PR(I)
I
>1
Number of vertices in each profile.
PROFILEID
S
< 26
Profile name. Should have less than 26
characters and must not contain blank
spaces.
X PRF(I,J),
R
m or f t
Coordinates of each vertex J in profile
Y PRF(I,J)
I.
Note: I = Integer variable. R = Real variable. S = Text variable.

9.4.4 Cross Section Data for Result Output File: .XSECS

Cross sections are used to output numeric results at user defined lines on the mesh.
Line 1: Number of cross sections.
NCROSS SECTIONS

9.4

Output control data

180

NCROSS SECTIONS groups of lines containing the cross section ID, the number of vertices defining the cross section (always equal to 2), the number of intervals to divide the cross section and
the list of coordinates of initial and final point in cross section:
XSECID
NPXSEC ND CS
X1 CS(I) Y1 CS(I)
X2 CS(I) Y2 CS(I)

9.4.4.1 Example of the .XSECS file

3
CrossSectionA
2 40
800500.45 306895.63
799095.07 307457.34
CrossSectionB
2 40
800492.17 307163.36
799171.99 307594.56
CrossSectionC
2 40
800449.99 307404.31
799223.97 307690.20
This .XSECS file indicates there are 3 cross sections. The first one has ID = CrossSectionA and
will be divided in 40 segments.
Table 9.45 Variable Descriptions for the .XSECS File
VARIABLE
TYPE RANGE UNITS
NCROSS SECTIONS I
>0
ND CS
I
>2
-

DESCRIPTION
Number of cross sections.
Cross section will be divided in ND CS
segments. Results will be reported at
each segment. See comment 1.
NPXSEC
I
2
Number of points defining cross section. In the present version only the
two extreme points are allowed to define the cross section, therefore this
value should always be 2.
X1 CS,
Y1 CS, R
m or f t
Coordinates of initial and ending point
X2 CS, Y2 CS
of each cross section.
XSECID
S
< 26
Cross section name. Should have less
than 26 characters and must not contain blank spaces.
Note: I = Integer variable. R = Real variable. S = Text variable.

9.4.4.2 Comments for the .XSECS File

1. The model will cut the mesh using the cross section line and extract results at the division points. If ND CS is too small, the program may not capture anything in between the
divisions, and the computed cross section discharges may have big errors.

181

Input Data File Reference

9.5 Elevation data

9.5.1 X Y Z data with header
These files contain scattered data in the format suitable to import it in the SMS data layer, for
example the BedElevations data layer. It usually has .EXP extension, but can have any other file
extension provided that the format is as described herein. Each point is identified by its X and Y
coordinates and the elevation value for that coordinate.
Line 1: Number of points and number of parameters (header)
NUMBER OF DATA POINTS NUMBER OF PARAMETERS
NUMBER OF DATA POINTS lines with X, Y and parameters data.
X(POINT) Y(POINT) P1(POINT) P2(POINT) ... PN(POINT)

9.5.1.1 Example of an .EXP File

11086 1
798439.73
798477.04
798489.45
798498.09
798504.45
798511.71
798516.09
798517.37
...

306063.87
309506.95
309522.30
306222.29
305915.63
306075.55
309412.73
309592.42

160.00
201.10
200.93
162.00
160.00
161.00
201.74
163.14

In this example .EXP file, there are 11086 elevation data points, one parameter per point (the
elevation for each point).
Table 9.46 Variable Descriptions for the .EXP File.
VARIABLE
NUMBER
OF
DATA POINTS
NUMBER OF PARAMETERS

TYPE RANGE UNITS

I
>0
-

DESCRIPTION
Number of data points in the file.

>0

m or f t

m or f t

Number of parameters for each point.

In the case of the elevation data file
this value is equal to 1.
X Coordinate of each elevation point.
See comment 1.
Y Coordinate of each elevation point.
See comment 1.
Parameter value. See comment 2.

P
R
m or f t
Note: I = Integer variable. R = Real variable.

9.5.1.2 Comments for the .EXP Data File

1. X and Y coordinates may be given in either meters or feet, depending on the units being used in the project. Coordinate system should always correspond to plane projection.
RiverFlow2D does not support geographical coordinates in Latitude/Longitude format. If

9.6

Boundary conditions data

182

the available data is in Lat/Lon format, it will need to be converted to UTM or plane Cartesian coordinates before importing coordinates into SMS. Presently RiverFlow2D does not
provide tools to make this conversion.
2. Elevation values should be given in the same units as the corresponding coordinates.

9.5.2 General X Y Z data

General XYZ data can be imported using PANEL, space or comma delimited ASCII files or ASCII
Grid Files using the PIE/Import.../Import Data command.

9.6 Boundary conditions data

9.6.1 One Variable Boundary Condition Files
This format applies to the following data files:
Time vs. Water Surface Elevation (BCTYPE = 1)
Time vs. Discharge (BCTYPE = 6)
Time vs. U velocity component in x direction (BCTYPE = 2)
Time vs. V velocity component in y direction (BCTYPE = 3)
Note: BCTYPE parameter is described in Table 7.
Line 1: Number points in data series.
NDATA
NDATA lines containing
TIME(I) VARIABLE(I)
Where VARIABLE(I) is WSE, Q, U or V, depending on the boundary condition BCTYPE.

9.6.1.1 Example of the Boundary Condition File for One Variable Time Series
The following example shows an inflow hydrograph where NDATA is 7 and there are 7 lines with
pairs of time and discharge:
7
0.
1.
1.3
2.
4.
5.
7.

20.
30.
50.
90.
120.
200.
250.

183

Input Data File Reference

Table 9.47 Variable Descriptions of Boundary Condition Files.

VARIABLE
NDATA
TIME

TYPE RANGE UNITS

I
>0
R
>0
Hrs

VARIABLE

DESCRIPTION
Number of points in data series.
Time in hours. The time interval is
arbitrary.
Represents Water Surface Elevation,
Discharge, U or V velocity components
depending on the boundary condition.

Note: I = Integer variable. R = Real variable.

9.6.2 Two Variables Boundary Condition Files

This format applies to the following data files:
Time vs. U and V velocity components (BCTYPE = 4)
Time vs. Discharge and Water Surface Elevation (BCTYPE = 5)
Time vs. U velocity component and Water Surface Elevation (BCTYPE = 7)
Time vs. V velocity component and Water Surface Elevation (BCTYPE = 8)
Line 1: Number points in data series.
NDATA
NDATA lines containing time and two values.
TIME(I) VARIABLE1(I) VARIABLE2(I)
Where VARIABLE1(I) and VARIABLE2(I) depend on the boundary condition type as follows:

BCTYPE
4
5
7
8
26

VARIABLE1
U
Q
U
V
Q

VARIABLE2
V
WSE
WSE
WSE
Qs

9.6.2.1 Example of the Two-Variable Boundary Condition File

The following example shows a file for BCTYPE=5 where discharge and WSE are given, NDATA
is 10 and there are 10 lines with pairs of time, discharge and WSE:
10
0. 20. 1420.
1. 30. 1421.5
1.3. 50. 1423.
...
7. 250. 1420.
8.1 110. 1426.

9.6
10.
20.

Boundary conditions data

60.
20.

184

1423.5
1421.
Table 9.49 Variable Descriptions of Two-Variable Boundary Condition Files.
VARIABLE
NDATA
TIME

TYPE RANGE UNITS

I
>0
R
>0
Hrs

VARIABLE1

VARIABLE2

DESCRIPTION
Number of points in data series.
Time in hours. The time interval is
arbitrary.
Represents Water Surface Elevation,
Discharge, U or V velocity components
depending on the boundary condition.
Represents Water Surface Elevation,
Discharge, U or V velocity components
depending on the boundary condition.

Note: I = Integer variable. R = Real variable.

9.6.3 Multiple-Variable Boundary Condition Files

This format applies to the following data file:
Time vs. Q water discharge and Qs sediment discharge (BCTYPE = 26)
Line 1: Number points in data series.
NDATA
NDATA lines containing time and two values.
TIME(I) VARIABLE1(I) VARIABLE2(I) ... VARIABLEN(I)
Where VARIABLE1(I) ... VARIABLEN(I) depend on the boundary condition type as follows:

BCTYPE
26

VARIABLE1
Q

VARIABLE2
Qs

9.6.3.1 Example of the Multiple-Variable Boundary Condition File

The following example shows a file for BCTYPE=26 where water discharge and sediment discharge
for two fractions are given, NDATA is 10 and there are 10 lines with pairs of time, discharge and
WSE:
10
0. 20.
1. 30.
1.3. 50.
...
7. 250.
8.1 110.
10. 60.

0.001 0.002
0.002 0.005
0.003 0.010
0.01 0.015
0.005 0.009
0.004 0.007

185
20.

9
20.

0.003

Input Data File Reference

0.005.

Table 9.51 Variable Descriptions of Multiple-Variable Boundary Condition Files.

VARIABLE
NDATA
TIME

TYPE RANGE UNITS

I
>0
R
>0
Hrs

VARIABLE1
VARIABLE2..N

R
R

DESCRIPTION
Number of points in data series.
Time in hours. The time interval is
arbitrary.
Represents Water Discharge.
Represents Sediment Discharge for the
given fraction.

Note: I = Integer variable. R = Real variable.

9.6.4 Stage-Discharge Data Files

This format applies to the stage (water surface elevation) vs. discharge table used for BCTYPE
= 9 and 19.
Line 1: Number points in data series.
NDATA
NDATA lines containing stage and discharge.
STAGE(I) Q(I)
Where STAGE(I) is water surface elevation and Q(I) is the corresponding discharge.

9.6.4.1 Example of the Stage-Discharge Boundary Condition File

The following example shows a stage-discharge rating table where NDATA is 21 and there are 21
lines with pairs of stage and corresponding discharge:
21
-1.00 0.00
-0.75 1.79
-0.50 5.20
-0.25 9.45
0.00 14.23
0.25 19.37
0.50 24.76
0.75 30.36
1.00 36.09
1.25 41.95
1.50 47.89
1.75 53.92
2.00 60.00
2.25 66.14
2.50 72.31
2.75 78.53
3.00 84.78
3.25 91.05
3.50 97.35

9.6
3.75
4.00

Boundary conditions data

186

103.67
110.01
Table 9.52 Variable Descriptions of Two-Variable Boundary Condition Files.
VARIABLE TYPE RANGE UNITS
NDATA
I
>0
STAGE
R
>0
m or f t
Q
R
>0
m3 /s or f t3 /s
Note: I = Integer variable. R = Real variable.

DESCRIPTION
Number of lines in data file.
Water surface elevation.
Water discharge.

9.6.5 Culvert Depth-Discharge Data Files

This format applies to the culvert depth vs. discharge table.
Line 1: Number points in data series.
NDATA
NDATA lines containing depth and discharge.
DEPTH(I) Q(I)
Where DEPTH(I) is depth corresponding to discharge Q(I).

9.6.5.1 Example of the Culvert Depth-Discharge File

The following example shows a depth-discharge rating table for a culvert. NDATA is 7 and there
are 7 lines with pairs of depth and corresponding discharge:
7
0 0.20
0.1 1.00
1.00 36.09
2.00 60.00
3.00 84.78
4.00 110.01
100.00 110.02
Table 9.53 Variable Descriptions of Culvert Depth-Discharge Data Files.
VARIABLE TYPE RANGE UNITS
NDATA
I
>0
DEPTH
R
>0
m or f t
Q
R
>0
m3 /s or f t3 /s
Note: I = Integer variable. R = Real variable.

DESCRIPTION
Number of lines in data file.
Water depth.
Water discharge.

10

Output File Reference

RiverFlow2D generates results in many output ASCII text files. These files can be easily accessed
with any text editor and they can be imported into SMS program and other GIS software for
visualization and analysis. RiverFlow2D always creates also output ASCII files in both English
and metric units depending on the units provided in the data files.

10.1 Output File Overview

The following table summarizes the output files generated by RiverFlow2D:
Table 10.1 List of Output Data Files.
RESULTS

DATA
FILE
EXTENSION
GENERAL OUTPUT
Run control param.OUTI, .OUTE
eters,
components
used, etc..

Triangular
element
mesh information

.MESHOUTI,
.MESHOUTE

Run progress results

.ROUT

General results tabular output

.UVHI, .UVHE

DESCRIPTION

Echoes input data read from files including

modeling control parameters, mesh data,
boundary conditions, and for each report
time interval inflow and outflow discharges
and velocities .OUTI file is in metric units
and .OUTE file in English units.
These files provide comprehensive information about the triangular element mesh.
.MESHOUTI is in metric units and .MESHOUTE
file in English units.
This file report for each output interval the
computer time, average time step, inflow
and outflow water and sediment transport
discharge at open boundaries and volume
and mass conservation errors.
For each output interval nodal velocities,
depths, water surface and bed elevations,
bed elevation changes, wet-dry condition,
Froude number are written to file .UVHI is
in metric units and .UVHE in English units.
Continued on next page

187

10.1

Output File Overview

Table 10.1 continued from previous page
DATA
FILE
DESCRIPTION
EXTENSION
Maximum values tab- .MAXI, .MAXE
For each output interval maximum nodal
ular output
velocity modules, depths, and are written
to file .MAXI is in metric units and .MAXE in
English units.
Time series at obser- .OUTI, .OUTE
These files report time series of results at
vation points
observation points. The model finds the
nodes closest to the provided coordinates
and the files report time series of velocities, depths, water surface and bed elevations, bed elevation changes, wet-dry condition, Froude number and sediment transport discharge. File name format is as follows: RESvsT_NODENUMBER.OUTI for metric
units and RESvsT_NODENUMBER.OUTE for English units.
Mass balance
.MBAL
Generated only in the RiverFlow2D FE
model to report total inflow and outflow
discharges and volume change for each output interval.
Hot start
.VTPA
Generated only in the RiverFlow2D FE
model to restart a simulation from previously computed results. The file contains
the file name from which the model will
restart when using hot start option.
Inundation area
.WETAREAS
These files report the area of inundation
(wet area) calculated as the sum of areas
of all wet elements. See comment 1.
PROFILE AND CROSS SECTION OUTPUT
Cross section output
.XSECI, .XSECE
For all output intervals, these files provide
bed elevation, depth, water surface elevation, depth average velocity, and Froude
number, water and sediment discharge.
.XSEC is in metric and .XSECE in English
units.
Cross section hydro.XSECH,
Report a hydrograph table for each cross
graphs
.XSECSED
section. .XSECH is the water hydrograph
and .XSECSED the sediment flux hydrograph for each cross section. See Comment
1.
Profile output
.PRFI, .PRFE
For each output interval and for a number of points along user defined polylines,
these files provide bed elevation, depth, water surface elevation, depth average velocity, and Froude number are written to file
.PRFI is in metric and .PRFE in English
units.
COMPONENT OUTPUT
Culverts
.OUT
Output discharge at every culvert for each
report interval. File name format is as follows: CULVERT_culvertID.OUT where culvertID is the user provided name.
Internal Rating Ta- .OUT
Output discharge at every IRT for each rebles
port interval.File name format is as follows:
IRT_irtID.OUT, where irtID is the text provided by the user to identify the Internal
Rating Table.
Weirs
.WEIRI, .WEIRE
Report results for weirs. .WEIRI is in metric
units and WEIRE in English units.
GIS SUPPORT OUTPUT
Continued on next page
RESULTS

188

189

10

Output File Reference

Table 10.1 continued from previous page

DATA
FILE
DESCRIPTION
EXTENSION
Spatial distribution .EXP
For each output interval nodal velocities,
of results in GIS
depths, water surface and bed elevations,
format
bed elevation changes, wet-dry condition,
Froude number, and sediment transport
fluxes, are written to file. These files can
be directly imported into GIS programs.
The file names are TIME_METRIC_DDDD_HH_
MM_SS.EXP for metric units and TIME_ENG_
DDDD_HH_MM_SS.EXP for English units.
Maximum values in
.EXP
These ASCII files report maximum values
GIS format
of velocity module, depth and water surface elevations. The files named as follows:
For English units: <ProjectName>_ENG_
MAX.EXP. For metric units: <ProjectName>
_METRIC_MAX.EXP. See Comment 1.
Shapefile
support .EXP
Support files that allow generating ESRI
files
Shapefiles for velocity vectors, depths and
water surface elevations. See Comment 1.
GRAPHIC OUTPUT
Mesh
.DXF
When the export to DXF is activated the
program will export the mesh in DXF format to the file: <ProjectName>_MESH.DXF.
See comment 1.
Velocity vectors
.DXF
Velocity vectors in AutoCAD DXF format.
See comment 1.
Various
.GIF
Various outputs in graphic raster GIF format. See comment 1.
AQUAVEO SMS OUTPUT
General results
.HDF5
RiverFlow2D model will output .HDF5
binary files, that report velocities, depths,
water surface and bed elevations, bed
elevation changes, wet-dry condition,
Froude number and sediment transport
discharge for each output time interval.
.HDF5 files can be used by SMS and
other post-processing software to visualize
RiverFlow2D results. See comment 1.
Visualization Toolkit Files
General results
.vtk
When the Create Graphic Output Files
check box is selected in the Graphic Output Panel, RiverFlow2D Plus model and
RiverFlow2D Plus GPU will output .vtk
files, that report velocities, depths, water
surface and bed elevations, bed elevation
changes, wet-dry condition, Froude number and sediment transport discharge for
each output time interval. These files can
be used by third party software including
Paraview to generate high quality graphs
of RiverFlow2D results. ParaView (www.
paraview.org) is an open-source, multiplatform data analysis and visualization application. ParaView users can quickly build
visualizations to analyze their data using qualitative and quantitative techniques.
The data exploration can be done interactively in 3D or programmatically using ParaViews batch processing capabilities.
RESULTS

10.2

General Output Files

190

10.1.0.1 Comments for Output Files

1. In the RiverFlow2D Plus and Plus GPU models these files are only generated when post
processing results using the Plot RiverFlow2D results on the Data Input Program Graphic
Options panel. Please see the Users Guide for more detail.

10.2 General Output Files

This section describes content of each output file.

10.2.1 Run Options Summary

These files replicate the input data read from files including modeling control parameters, mesh
data, boundary conditions, and inflow and outflow discharges and velocities for each output interval. .OUTI is in metric units and .OUTE in English units. Part of a typical .OUTI/.OUTE file format
is shown below:

191

10

Output File Reference

10.2.2 Mesh Data and Mesh Metrics

Mesh data is written to files with extensions: .MESHOUTI (metric units) and .MESHOUTE (English
units). These files provide comprehensive information about the triangular element mesh. The
following table summarizes the available output.
Table 10.2 Variables reported on the .MESHOUTI and .MESHOUTE Files.
VARIABLE
Number of elements
Number of nodes
X
Y
BEDEL
INITIAL WSE
BC ID
BC File
Node1, Node2, Node3
Mannings n
Area
Angle
Total mesh area
Average element area
Average element size
Element with minimum area
Minimum element size
Minimum element area
Element with maximum area
Maximum element size
Maximum element area
Minimum element angle
Element with minimum angle

DESCRIPTION
Total number of elements in the mesh
Total number of nodes in the mesh
x-coordinate of node
y-coordinate of node
Node initial bed elevation
Node initial water surface elevation
Boundary condition code
Boundary condition file name
Nodes numbers of each element in counterclockwise order
Manning n roughness coefficient
Element/cell area
Minimum angle in element/cell
Sum of areas of all elements/cells on the mesh
Total mesh area divided by number of elements/cells
Average size of elements/cells on mesh
Smallest element
Approximate linear size of smallest element/cell
Area of smallest element/cell
Largest element/cell
Approximate linear size of largest element/cell
Area of largest element/cell
Smallest element/cell internal angle
Element/cell that has the smallest internal angle

This file also reports the list of acute cells that have an internal angle of less than 22.5 degrees.
If there are acute cells, the model will give an error message and will not be able to execute.
Part of a typical file format is shown below:

192

10

Output File Reference

10.2

General Output Files

193

194

10

Output File Reference

195

10

Output File Reference

10.2.3 Run Summary

Run summary report is written to file with extension: .ROUT. These files report for each output
interval the computer time, average time step, and for each open boundary inflow (positive) or
outflow (negative) discharge (m2 /s or f t3 /s), volume conservation error (%), volumetric sediment
discharge (m2 /s or f t3 /s) and sediment mass conservation error (%)

10.2.3.1 Example of the .ROUT file

10.2

General Output Files

196

10.2.4 Tabular Output Files

These files report nodal velocities, depths, water surface and bed elevations, bed elevation changes,
wet-dry condition, Froude number, and sediment transport fluxes for each output interval. .UVHI
is in metric units and .UVHE in English units.
A typical output .UVHE file is shown below:

197

10

Output File Reference

10.2.5 Area of Inundation Files

These files report the area of inundation (wet area) calculated as the sum of areas of all wet elements. The file name has extension .WETAREAS. RiverFlow2D Plus model maximum values will
only be generated using the post processing Plot RiverFlow2D results button on the Graphic Output Options panel.
A typical output is shown below:

10.2

General Output Files

198

10.2.6 Maximum Value Tabular Files

These files report maximum nodal values of velocity module, depth, water surface elevations, and
bed changes over the complete simulation. .MAXI is in metric units and .MAXE in English units.
The reported variables are described in the following table:
Table 10.3 Variables Reported on the Observation Point Files.
COLUMN VARIABLE

DESCRIPTION

1
2
3
4

Time in hours
Velocity component in x direction
Velocity component in y direction
Maximum
velocity magnitude

U2 + V 2
Maximum water depth
Maximum water elevation
Maximum bed elevation*
Maximum bed elevation*
Minimum erosion depth*

Node
X
Y
MAX VEL

5
6
7
8
9

MAX DEPTH
MAX WSEL
MAX BEDEL
MIN BEDEL
MIN
EROSION DEPTH
10
MAX
DE- Maximum deposition depth*
POSITION
DEPTH
*Output when using the sediment transport component.

A typical output .MAXE file follows:

ENGLISH
UNITS
ft
ft
f t/s

METRIC
UNITS
m
m
m/s

ft
ft
ft
ft
ft

m
m
m
m
m

ft

199

10

Output File Reference

10.2.7 Observation Point Output Files

These files report time series of results at observation points. The program finds the closest node
to the observation point coordinate and writes the result time series of the following variables:
Table 10.4 Variables Reported on the Observation Point Files.
COLUMN VARIABLE

DESCRIPTION

1
2
3
4

Time
U
V
VELOCITY

5
6
7
8
9
10
11

DEPTH
WSEL
BEDEL ORI
BEDEL
DELTA BED
Froude
QSX

12

QSY

13

QS

Time in hours
Velocity component in x direction
Velocity component in y direction
Maximum velocity magnitude

U2 + V 2
Maximum water depth
Maximum water elevation
Maximum bed elevation*
Maximum bed elevation*
Minimum erosion depth*
Maximum deposition depth*
Volumetric sediment discharge
per unit width in x direction
Volumetric sediment discharge
per unit width in y direction
Volumetric sediment
discharge
q
magnitude Qs =

ENGLISH
UNITS
ft
ft
f t/s

METRIC
UNITS
m
m
m/s

ft
ft
ft
ft
ft
ft
f t2 /s

m
m
m
m
m
m
m2 /s

f t2 /s

m2 /s

f t2 /s

m2 /s

Q2sx + Q2sy

The file name for each node is:

RESvsT_NODENUMBER.OUTI for metric units and
RESvsT_NODENUMBER.OUTE for English units.
For example: RESvsT_0000010.OUTI is the file name for time series results of node 10. An example
of this file is shown below.

10.2

General Output Files

200

10.2.8 Hot Start File

The hot start 2binitialized.HOTSTART file is used to restart a simulation from previously computed results and when hot start option is selected. By default the file contains the name of the
last time and state file computed. Those results will be used as initial condition to restart the
simulation if the hot start option is activated. For example, if the user stops the simulation at 5
hours to review results or for any other reason the, 2binitialized.HOTSTART file would have the
following text:
18000state5.out
Note that the stateN.out files are named sequentially. For instance, stateN.out corresponds
to the Nth report interval. RiverFlow2D can be restarted from the last report time interval by
reading the initial conditions from the stateN.out file pointed in the file. To restart from a time
different from the last one calculated, just edit the 2binitialized.HOTSTART file and enter the
desired time in seconds and corresponding stateN.out file name that is to be used as initial conditions. For example, to hot start from hour 3 and assuming the report interval is 0.5 hours, the
2binitialized.HOTSTART would contain:
18000state6.out

201

10

Output File Reference

10.3 Component Output Files

10.3.1 Culvert Output Files
For each culvert, RiverFlow2D creates an output file named: CULVERT_culvertID.OUT, where
culvertID is the text provided by the user to identify the culvert. Report includes discharge for
each report interval as shown:

10.3

Component Output Files

202

10.3.2 Internal Rating Table Output Files

For each Internal Rating Table, RiverFlow2D creates an output file named: IRT_irtID.OUT, where
irtID is the text provided by the user to identify the Internal Rating Table. Report includes
discharge for each report interval as shown:

203

10

Output File Reference

10.3.3 Weir Output Files

These files report results for each weir and for each output interval. File extension is .WEIRI for
metric units and .WEIRE for English units. Output includes the following information:
Table 10.5 Variables Reported on the Weir Point Files.
VARIABLE
EDGE
N1
N2
WSE1
WSE2
D1
D2
Distance
Q

DESCRIPTION
Edge number
Cell at side 1 of the edge
Cell at side 2 of the edge
Water surface elevation at cell N1
Water surface elevation at cell N2
Depth at cell N1
Depth at cell N2
Edge length
Edge discharge

A typical weir output file format is shown below:

10.4

Profile and Cross Section Output Files

204

10.4 Profile and Cross Section Output Files

10.4.1 General Cross Section Output Files
When using the Output results for cross sections option, the model will generate files with extensions .XSECI and .XSECE, that report results along user provided cross sections. For each output
interval and for each user defined cross sections the bed elevation, depth, water surface elevation,
depth average velocity, Froude number and volumetric sediment discharge per unit width is written
to file .XSECI is in metric and .XSECE in English units. A typical .XSECI file is as follows:

When running only hydrodynamics the .XSECE and .XSECI files will display the cross section
water discharge. When running sediment transport, in addition to the water discharge these files
will report the total sediment discharge in f t3 /s or m3 /s.

205

10

Output File Reference

10.4.2 Cross Section Hydrograph Output Files

These files will only be generated using the post processing Plot RiverFlow2D results button on the
Graphic Output Options panel. When using the Output results for cross sections option, the model
will generate files with extension .XSECH and .XSECSED (if using sediment transport component),
that report a hydrograph table for each cross section. A typical path.XSECH file is as follows:

10.4

Profile and Cross Section Output Files

206

10.4.3 Profile Output Files

When using the Output results for profiles option, the model will generate files with extensions
.PRFI and .PRFE, that report results along user provided polylines. For each output interval and
for the number of points along user defined polylines these files list bed elevation, depth, water
surface elevation, depth average velocity, and Froude number. .PRFI is in metric and .PRFE in
English units. An example output is shown below:

10.4

Profile and Cross Section Output Files

207

10.5

Output Files for GIS Post-processing

208

10.5 Output Files for GIS Post-processing

10.5.1 General Spatial Distribution Files
These ASCII files allow seamless transfer to Geographic Information System software. These files
use the .EXP extension and are named as follows:
For English units: TIME_ENG_DDDD_HH_MM_SS.EXP
For Metric units: TIME_METRIC_DDDD_HH_MM_SS.EXP
Where DDDD is days, HH is hours, MM is minutes and SS seconds. For example
TIME_ENG_0001_12_01_34.EXP
corresponds to a file in English units for time: 1 day, 12 hours, 1 minute and 34 seconds.
The format for these files is as follows. The first line indicates the number of node parameters
(18 by default). Then follows NNODES lines with results for each node in the triangular element
mesh as shown:
Table 10.6 Variables Reported on TIME .EXP Output Files.
COLUMN VALUE
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

Node number
Node x-coordinate
Node y-coordinate
Velocity component in x direction U
Velocity component in y direction V
~  2
Velocity magnitude U
= U +V2
Water surface elevation
Depth H
Initial bed elevation
Bed elevation
Bed elevation change since time = 0
Dry or wet
Froude number
Volumetric sediment discharge per
unit width in x direction: Qsx
Volumetric sediment discharge per
unit width in y direction: Qsy
Volumetric
q sediment discharge magniQ2sx

ENGLISH
UNITS
ft
ft
f t/s
f t/s

METRIC
UNITS
m
m
m/s
m/s

f t/s
ft
ft
ft
ft
ft
Dry=1 / Wet=0
f t2 /s

m/s
m
m
m
m
m
Dry=1 / Wet=0
m2 /s

f t2 /s

m2 /s

f t2 /s

m2 /s

Q2sy

+
tude: Qs
Bed shear stress*:
lb/f t2
= HSf = (U n/k)2 /H 1/3
18
For future use
19
For future use
*English units: = 62.4lb/f t3 , k = 1.49.
*Metric units: = 9810N/m3 ; k = 1, n = Mannings coefficient.
17

The following file is an example of a typical ASCII .EXP file:

Pa

209

10

Output File Reference

10.5.2 Pollutant Concentration Files

These ASCII files contains the pollutant. These files use the .EXP extension and are named as
follows:
CONC_DDDD_HH_MM_SS.EXP
Where DDDD is days, HH is hours, MM is minutes and SS seconds. For example
CONC_0001_12_01_34.EXP
corresponds to a file in English units for time: 1 day, 12 hours, 1 minute and 34 seconds.
The format for these files is as follows. The first line indicates the number of solutes used in
the PL run (NPOLLUTANTS MAX). Then follows NNODES lines with results for each node in
the triangular element mesh as shown:
Table 10.7 Variables Reported on CONC*.EXP Output Files.
COLUMN VALUE
1
Node number
2
Concentration for solute 1
3

Concentration for solute 2

...
5

...
Concentration for solute NPOLLUTANTS MAX

UNITS
Same as
BCs
Same as
BCs
...
Same as
BCs

in
in

in

10.5

Output Files for GIS Post-processing

210

The following file is an example of a typical CONC*.EXP file:

3
1
2
3
4
5
6
7
...

0.000000
0.000000
0.000000
0.202378
0.326602
0.291721
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000

In this example, the CONC*.EXP has 3 pollutants.

10.5.3 Suspended Sediment Concentration Files

These ASCII files contain suspended sediment concentrations. These files use the .EXP extension
and are named as follows:
ST_DDDD_HH_MM_SS.EXP
Where DDDD is days, HH is hours, MM is minutes and SS seconds. For example
ST_0001_12_01_34.EXP
corresponds to a file in English units for time: 1 day, 12 hours, 1 minute and 34 seconds.
The format for these files is as follows. The first line indicates the number of suspended sediment
fractions used in the ST run (NSSNFRAC). Then follows NNODES lines with results for each node
in the triangular element mesh as shown:
Table 10.8 Variables Reported on ST*.EXP Output Files.
COLUMN
1
2
3
...
5

VALUE
Node number
Concentration by volume for fraction 1
Concentration by volume for fraction 2
...
Concentration by volume for fraction NSSNFRAC

The following file is an example of a typical ST*.EXP file:

2
1
2
3
4
5
6
7

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000

0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000

UNITS
Fraction of 1
Fraction of 1
...
Fraction of 1

211
8
9
10
11
12
13
...

10
0.000000
0.000287
0.000456
0.000356
0.000334
0.000000

Output File Reference

0.000000
0.000287
0.000456
0.000356
0.000334
0.000000

In this example, the ST*.EXP has 2 suspended sediment fractions.

10.5.4 Maximum Value GIS Files

These ASCII files report maximum values of velocity module, depth and water surface elevations
and allow seamless transfer to Geographic Information System software. These files use the .EXP
extension and are named as follows:
For English units: <ProjectName>_ENG_MAX.EXP
For Metric units: <ProjectName>_METRIC_MAX.EXP
The format for these files is as follows. The first line indicates the number of node parameters (5
by default). There follows NNODES lines with velocity module, depth and water surface elevation
for each node as shown:
Table
10.9

Variables
Reported
<ProjectName> METRIC MAX.EXP Files.

on

the

COLUMN VALUE
1
2
3
4

Node number
Node x-coordinate
Node y-coordinate
Maximum
velocity
q

magnitude

Ux2 + Uy2
5
Maximum depth
6
Maximum water surface elevation
7
Maximum bed elevation*
8
Minimum bed elevation*
9
Maximum erosion depth*
10
Maximum deposition depth*
*Output when using the sediment transport component.

<ProjectName> ENG MAX.EXP

ENGLISH
UNITS
ft
ft
f t/s

METRIC
UNITS
m
m
m/s

ft
ft
ft
ft
ft
ft

m
m
m
m
m
m

or

The following file is an example of a maximum values <ProjectName>_ENG_MAX.EXP or <ProjectName>

_METRIC_MAX.EXP file:

10.5

Output Files for GIS Post-processing

212

10.5.5 Time to Depth GIS File

This ASCII file are generated when using the Plot RiverFlow2D Results button on the Graphic
Output Option panel of the DIP. The file reports the time at which certain depths are reached
during the simulation and allow seamless transfer to Geographic Information System software. In
metric units the file includes time to 0.30 m of depth, time to 1 m and time to maximum depth for
each node. In English units the file includes time to 1ft of depth, time to 3ft and time to maximum
depth for each node. Time is always given in hours. The time to depth file use the .EXP extension
and have this name:
<ProjectName>_time2depths.exp
The format for this file is as follows. The first line indicates the number of node parameters (3
by default). There follows NNODES lines with time to 0.30 m (1ft), time to 1m (3ft), and time
to maximum depth for each node as shown in Table ??. When the node remains dry or depth is
below 0.30 m (1ft) the reported value is -1.
Table 10.10 Variables Reported on the <ProjectName> time2depths.exp File.
COLUMN
1
2
3
4
*Time = -1

VALUE
UNITS
Node number
Time to 0.30 m (Metric) or 1 ft. (English)*
hrs.
Time to 1 m (Metric) or 3 ft. (English)*
hrs.
Time to maximum depth*
hrs.
when node is dry or depth does not reach 0.3 m (1ft).

The following file is an example of a time to depths file <ProjectName>_time2depths.exp file:

213
3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

10

0.110
0.080
0.800
-1.000
-1.000
-1.000
-1.000
0.370
0.180
1.010
0.700
0.660
0.660
0.710
0.390
0.390
0.380
0.390
0.040
0.040
0.050
0.040
0.780
0.740
0.380

0.280
0.150
-1.000
-1.000
-1.000
-1.000
-1.000
0.510
0.380
-1.000
-1.000
1.000
-1.000
-1.000
0.590
0.580
0.570
0.580
0.100
0.100
0.120
0.100
-1.000
-1.000
0.560

Output File Reference

1.960
1.960
1.960
-1.000
-1.000
-1.000
-1.000
0.640
1.930
1.960
1.970
1.100
1.100
1.930
1.980
1.980
1.980
1.980
1.980
1.980
1.980
1.980
1.970
1.100
1.980

10.5.6 Support Files to Generate ESRI Shapefiles

These ASCII files can be imported into Geographic Information System (GIS) programs.
Velocity field files use the .EXP extension and are named as follows:
VelField_SF_TIME_DDDD_HH_MM_SS.EXP
Water surface elevation files use the .EXP extension and are named as follows:
WSE_SF_TIME_TIME_DDDD_HH_MM_SS.EXP
Where DDDD is days, HH is hours, MM is minutes and SS seconds.

10.5.7 Support Files to Generate USBR Hazard Maps in ESRI

Shapefiles format
When you use the Output for creating shapefiles in the Graphic Output Options panel of the DIP,
the program creates the USBR_Hazard.EXP that can be imported into Geographic Information
systems (GIS) to create Hazard Map shapefiles based on the United States Bureau of Reclamation
classification of flood hazards (USBR, 1988). The USBR_Hazard.EXP file has as many polygons as
flooded elements containing five attributes corresponding of hazard level for houses, mobile homes,

10.6

Graphic Output Files

214

vehicles, adults and children. The attributes can get the values of 1, 2 or 3 depending on the
hazard level:
Table 10.11 USBR Hazard Classification in USBR Hazard.EXP file.
ATTRIBUTE
VALUE
1
2
3

HAZARD CLASSIFICATION
Low-danger zone
Judgment zone
High-danger zone

For further details about the USBR Hazard classification, consult the RiverFlow2D Users
Guide and USBR (1988).

10.6 Graphic Output Files

10.6.1 DXF Files
These files contain the velocity field for each output interval using colored vectors in AutoCAD
DXF format. The file name is of this form: VelField_TIME_DDDD_HH_MM_SS.DXF.

10.6.2 GIF Output Files

These are raster output files in .GIF format showing the plot for each output interval. They can
be imported in any word processor, spreadsheet or presentation programs.

10.6.3 VTK Output Files

RiverFlow2D Plus model and RiverFlow2D Plus GPU will output .vtk files, that report velocities, depths, water surface and bed elevations, bed elevation changes, wet-dry condition, Froude
number and sediment transport discharge for each output time interval. These files can be used by
third party software including Paraview to generate high quality graphs of RiverFlow2D results.
ParaView (www.paraview.org) is an open-source, multi-platform data analysis and visualization
application. ParaView users can quickly build visualizations to analyze their data using qualitative
and quantitative techniques. The data exploration can be done interactively in 3D or programmatically using ParaViews batch processing capabilities.

10.6

Graphic Output Files

Figure 10.1 Output color coded by depth and with velocity vectors

215

10.6

Graphic Output Files

216

11

Data Input Program Tools

This section describes various utilities that are available through the RiverFlow2D Data Input
Program.

11.1 HEC-RAS Data Extraction Tool

The purpose of this tool is to facilitate migrating existing HEC-RAS projects to RiverFlow2D.
The program allows extraction of point elevation data from geo-referenced cross-section from the
HEC-RAS one-dimensional model developed by the USACE. This elevation data can then be used
by RiverFlow2D pre-processors as the base bathymetric and topographic data or to complement
existing topography. The tool reads HEC-RAS geometry file names with extension .g01, .g02,
etc., and creates X Y Z files that can be readily imported in SMS. The utility discriminates the
elevations in the channel between the left and right bank on each cross section and exports the
files as detailed in the following table.
Table 11.1 HEC-RAS files with elevation data.
FILE NAME

CONTENT
Contains all elevation points in
all cross sections in the for all
<HEC-RAS file name> ALL STATIONS.EXP
reaches and cross sections in the
<HEC-RAS file name>.g0 file.
Contains polygons that include
<HEC-RAS file name> ALL POLY.EXP
all elevation points in each reach.
Contains only the elevation
points between the left and right
<HEC-RAS file name> CHANNEL STATIONS.EXPbanks in all cross sections in the
for all reaches in the <HEC-RAS
file name>.g0 file.
Contains polygons that include
only the elevation in the main
<HEC-RAS file name> CHANNEL POLY.EXP
channel for each reach.

217

11.1

HEC-RAS Data Extraction Tool

218

The HEC-RAS Data Extraction Tool is accessible from the Tools panel of the Data Input Program.

Figure 11.1 HEC-RAS command on tools panel.

12

References

Abderrezzak, K. & Paquier, A. (2011), Applicability of sediment transport capacity formulas to

dam-break flows over movable beds, Journal of Hydraulic Engineering, ASCE 137, 209221.
Ackers, P. & White, W. (1973), Sediment transport, new approach and analysis, Journal of
Hydraulic Div. ASCE 99(11), 20412060.
Akan, A. O. (1993), Urban Stormwater Hydrology: A Guide to Engineering Calculations, CRC
Press.
ASCE (1996), Hydrology Handbook, 2nd edn, American Society of Civil Engineers (ASCE).
Ashida, K. & Michiue, M. (1972), Study on hydraulic resistance and bed-load transport rate in
alluvial streams, Transactions Japan Society of Civil Engineering 206, 5969.
Bagnold, R. (1973), The nature of saltation of bed load transport in water, Proc. of Royal Society.
Ser. A 332, 473504.
Bai, L. & Jin, S. (2009), A conservative coupled flow/transport model with zero mass error, J.
of Hydrodynamics 21, 166 175.
Baldock, T., Tomkins, M., Nielsen, P. & Hughes, M. (2004), Settling velocity of sediments at high
concentrations, Coastal Engineering 51, 91100.
Belleundy, P. (2000), Restoring flow capacity in the loire river bed, Hydrological Processes
14, 23312344.
Brufau, P., Garca-Navarro, P. & Vazquez-Cendon, M. (2004), Zero mass error using unsteady
wetting-drying conditions in shallow flows over dry irregular topography, Int. Journal for
Numerical Methods in Fluids 45, 10471082.
Camenen, B. & Larson, M. (2005), A bedload sediment transport formula for the nearshore,
Estuarine, Coastal and Shelf Science 63, 249260.
Caviedes-Voulli`eme, D., Garca-Navarro, P. & Murillo, J. (2012), Influence of mesh structure on
2D full shallow water equations and SCS Curve Number simulation of rainfall/runoff events,
Journal of Hydrology 448-449(0), 39 59.
219

220
Cheng, N. (1997), Simplified settling velocity formula for sediment particle, J. Hydraulic Eng.,
ASCE. 123(2), 149152.
Cunge, J., Holly, F. & Vervey, A. (1980), Practical Aspects of Computational River Hydraulics,
Pitman: London.
De Vriend, H., Zyserman, J., Nicholson, J., Roelvink, J., Pechon, P. & Southgate, H. (1993),
Medium-term 2DH coastal area modelling, Coastal Engineering 21(1-3), 193224.
Engelund, F. & Fredsoe, J. (1976), A sediment transport model for straight alluvial channels.,
Nordic Hydrology 7, 293306.
Engelund, F. & Hensen, E. (1967), A monograph on sediment transport to alluvial streams, Report,
Copenhagen: Teknique Vorlag.
Fernandez-Luque, R. & van Beek, R. (1976), Erosion and transport of bed sediment, Journal of
Hydraulic Research, IAHR 14(2), 127144.
FHWA (1978), Hydraulics of bridge waterways, Report EPD-86-101, FHWA (Federal Highway
Administration).
Froehlich, D. (2003), Users Manual for FESWMS FST2DH Two-dimensional Depth-averaged Flow
and Sediment Transport Model, Report No. FHWA-RD-03-053, Federal Highway Administration, Washington, DC.
Garca, R., Espinoza, R., Valera, E. & Gonzalez, M. (2006), An explicit two-dimensional finite
element model to simulate short- and long-term bed evolution in alluvial rivers, Journal of
Hydraulic Research 44(6), 755766.
Garca, R., Gonz
alez, N. & OBrien, J. (2009), Dam-break flood routing. Chapter 4 in: Dam-Break
Problems, Solutions and Case Studies, WIT Press, Southampton-Boston.
Garde, R. & Ranga Raju, K. (1985), Mechanics of Sediment Transportation and Alluvial Stream
Problems, John Wiley & Sons, New York.
Godunov, S. (1959), A difference method for numerical calculation of discontinuous solutions of
the equations of hydrodynamics, Mat. Sb. 47, 271306.
Gupta, R. S. (1995), Hydrology and Hydraulic Systems, Waveland Press.
Henderson, F. (1966), Open Channel Flow, MacMillan series in civil Engineering.
Horton, R. (1933), The role of infiltration in the hydrologic cycle, Trans. Am. Geophys. Union
14, 446 460.
Jimenez, J. & O.S., M. (2003), A simple formula to estimate settling velocity of natural sediments,
J. Waterway, Port, Coastal, Ocean Eng 129(2), 7078.
Juez, C., Murillo, J. & P., G.-N. (2014), Numerical assessment of bed-load discharge formulations
for transient flow in 1d and 2d situations, Journal of Hydroinformatics 15(4), 12341257.
Julien, P. (1998), Erosion and Sedimentation, Cambridge University Press.
Jury, W. & Horton, R. (2004), Soil Physics, John Wiley and Sons.
Karim, F. (1988), Bed material discharge prediction for non-uniform bed sediments, Journal of
Hydraulic Engineering, ASCE 124(6), 597604.

221

12

References

LeVeque, R. (2002), Finite-Volume Methods for Hyperbolic problems, Cambridge University Press,
New York.
Mein, R. G. & Larson, C. L. (1973), Modeling infiltration during a steady rain, Water Resources
Research 9(2), 384394.
Meyer-Peter, E. & Muller, R. (1948), Formulas for bed-load transport, Proc. of the Second Meeting. IAHR, Stockholm, Sweden pp. 3964.
Mishra, S. & Singh, V. (2003), Soil Conservation Service Curve Number (SCS-CN) Methodology,
Kluwer Academic Publishers.
Morvan, H., Knight, D., Wright, N., Tang, X. & Crossley, A. (2008), The concept of roughness
in fluvial hydraulics and its formulation in 1-d, 2-d and 3-d numerical simulation models,
Journal of Hydraulic Research 46(2), 191208.
Murillo, J., Burguete, J., Brufau, P. & Garca-Navarro, P. (2005), Coupling between shallow water
and solute flow equations: analysis and management of source terms in 2D, International
Journal for Numerical Methods in Fluids 49(3), 267299.
Murillo, J., Burguete, J., Brufau, P. & Garca-Navarro, P. (2007), The influence of source terms
on stability, accuracy and conservation in two-dimensional shallow flow simulation using triangular finite volumes, International Journal of Numerical Methods in Fluids 54, 543590.
Murillo, J. & Garca-Navarro, P. (2010a), An exner-based coupled model for two-dimensional
transient flow over erodible bed, Journal of Computational Physics 229(23), 87048732.
Murillo, J. & Garca-Navarro, P. (2010b), Weak solutions for partial differential equations with
source terms: Application to the shallow water equations, Journal of Computational Physics
229(11), 43274368.
Murillo, J., Garca-Navarro, P. & Burguete, J. (2008), Analysis of a second-order upwind method
for the simulation of solute transport in 2d shallow water flow, International Journal for
Numerical Methods in Fluids 56(6), 661686.
Murillo, J., Garca-Navarro, P. & Burguete, J. (2012), Wave riemann description of friction terms
in unsteady shallow flows: Application to water and mud/debris floods, Journal of Computational Physics 231, 19632001.
Naef, D., Rickenmann, D., Rutschmann, P., & McArdell, B. (2006), Comparison of flow resistance relations for debris flows using a one-dimensional finite element simulation model, Nat.
Hazards Earth Syst. Sci. 6, 155156.
Nielsen, P. (1992), Coastal Bottom Boundary Layers and Sediment Transport, Advanced Series on
Ocean Engineering. World Scientific Publishing.
Norman, J. M., Houghtalen, R. J. & Johnston, W. J. (1985), Hydraulic design of highway culverts,
Hydraulic Design Series No. 5 FHWA-IP-85-15, Federal Highway Administration, Washington,
D.C.
Parker, G. (1979), Hydraulic geometry of active gravel rivers, Journal of Hydraulic Engineering,
ASCE 105(9), 11851201.
Parker, G., Klingeman, P. & McLean, D. (1982), Bed load and size distribution in paved gravel
bed streams, Journal of Hydraulic Engineering 108(4), 544571.

222
Ratia, H., Murillo, J. & Garca-Navarro, P. (2014), Numerical modelling of bridges in 2D shallow
water flow simulations, International Journal for Numerical Methods in Fluids 75(4), 250
272.
Raudkivi, A. (1990), Loose boundary hydraulics, Pergamon Press, Inc., Tarrytown, N.Y.
Rawls, W. & Brakensiek, D. (1983), A procedure to predict green and ampt infiltration parameters,
in Proceeding of ASAE Conferences on Advances in Infiltration, Chicago, Illinois, pp. 102
112.
Rawls, W., Yates, P. & Asmussen, L. (1976), Calibration of selected infiltration equation for the
georgia coastal plain, Report ARS-S-113, Agriculture Research Service.
Roe, P. (1981), Approximate riemann solvers, parameter vectors, and difference schemes, Journal
of Computational Physics 43(2), 357 372.
Rogers, B. & Borthwick, A.G.L. Taylor, P. (2003), Mathematical balancing of flux gradient and
source terms prior to using roes approximate riemann solver, J. Comput. Phys. 192(2), 422
451.
Rosatti, G., Murillo, J. & Fraccarollo, L. (2008), Generalized roe schemes for 1D two-phase,
free-surface flows over a mobile bed, J. Comput. Phys. 227(24), 1005810077.
Rubey, M. (1933), Settling velocities of gravel, sand and silt particles, Amer. J. Soc. 225, 325338.
Sha, Y. (1965), Introduction to Sediment Dynamics, Industry Press, 302-310.
Smart, G. (1984), Sediment transport formula for steep channels, J. Hydr. Eng., ASCE.
111(3), 267276.
Te Chow, V. (1959), Open Channel Hydraulics, McGraw-Hill, New York.
Te Chow, V., Maidment, D. & Mays, L. (1988), Applied Hydrology, McGraw-Hill Civil Engineering
Series, MCGRAW-HILL Higher Education.
Toro, E. (2001), Shock-Capturing Methods for Free-Surface Shallow Flows, Wiley, New York, p.
109.
USBR (1988), Downstream hazard classification guidelines, ACER Technical Memorandum 5,
United States Bureau of Reclamation (USBR).
USDA (1986), Urban hydrology for small watersheds, Environment agency report, United States
Department of Agriculture (USDA).
USDA (2004), National engineering handbook part 630 hydrology, Technical report, United States
Department of Agriculture (USDA).
Van Rijn, L. (1984a), Sediment pick-up functions, J. Hydr. Eng., ASCE. 110(10), 14941502.
Van Rijn, L. (1984b), Sediment transport, part i: bed load transport, J. Hydr. Eng., ASCE.
110(10), 14311456.
Van Rijn, L. (1984c), Sediment transport, part ii: suspended load transport, J. Hydr. Eng.,
ASCE. 110(11), 16131641.
V
azquez-Cend
on, M. (1999), Improved treatment of source terms in upwind schemes for the shallow water equations in channels with irregular geometry, Journal of Computational Physics
148(2), 497526.

223

12

References

Vreugdenhil, C. (1994), Numerical methods for shallow water flow, Water Science and Technology
Library. Kluwer Academic Publishers.
Wilcock, P. & Crowe, J. (2003), Surface-based transport model for mixed-sized sediment, Journal
of Hydraulic Engineering 129(2), 120128.
Wong, M. (2003), Does the bedload transport relation of meyer-peter and muller fits its own
data?, Proc. 30th IAHR-Congress, Thessaloniki, Greece, 8 pp. .
Wong, M. & Parker, G. (2006), Re-analysis and correction of bed load relation of meyer-peter and
mller using their own database, Journal of Hydraulic Engineering 132(11), 11591168.
Wu, W. (2008), Computational River Dynamics, Taylor and Francis.
Wu, W. & Wang, S. (2006), Formulas for sediment porosity and settling velocity, J. Hydraulic
Eng., ASCE. 132(8), 858862.
Yang, C. (1996), Sediment Transport Theory and Practice, McGraw Hill, New York.
Yarnell, D. (1934a), Bridge piers as channel obstructions, U.S. Department of Agriculture, Tech.
Bull. 442.
Yarnell, D. (1934b), Pile trestles as channel obstructions, U.S. Department of Agriculture, Tech.
Bull. 429.
Zanke, U. (n.d.), Berechnung der sinkgeschwindigkeiten von sedimenten, Technical report, Technischen Universitt Hannover.
Zhang, R. (1961), River Dynamics (in Chinese), Industry Press.