EPA/600/R-01/073
September 2001
EFDC1D - A One Dimensional Hydrodynamic
and Sediment Transport Model for
River and Stream Networks:
Model Theory and Users Guide
By
John M. Hamrick
Tetra Tech, Inc.
10306 Eaton Place, Suite 340
Fairfax, VA 22030
Project Officer:
Earl J. Hayter
Ecosystems Research Division
U.S. Environmental Protection Agency
Athens, GA
National Exposure Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
-------
Notice
The U.S. Environmental Protection Agency through its Office of Research and Development
funded and managed the research described here under contract (68-C-98-010; Work
Assignment 2-08) to Aqua Terra Consultants, Inc., Earl J. Hayter, EPA Project Officer. It has
been subjected to the Agency's peer and administrative review and has been approved for
publication as an EPA document. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
-------
Abstract
This technical report describes the new one-dimensional (ID) hydrodynamic and sediment
transport model EFDC1D. This model can be applied to stream networks. The model code and
two sample data sets are included on the distribution CD. EFDC1D can simulate bi-directional
unsteady flows and has the ability to accommodate unsteady inflows and outflows associated
with upstream inflows, lateral inflows and withdrawals, groundwater-surface water interaction,
evaporation and direct rainfall. The model also includes representation of hydraulic structures
such as dams and culverts. For sediment transport, the model includes settling, deposition and
resuspension of multiple size classes of cohesive and noncohesive sediments. The bed is
represented by multiple layers of mixed sediment classes. A bed consolidation model is
implemented to predict time variations of bed depth, void ratio, bulk density and shear strength.
The sediment bed representation is dynamically coupled to the cross-sectional area
representation to account for area changes due to deposition and resuspension.
-------
Table of Contents
Contents Page
Notice ii
Abstract iii
1. Introduction 1
2. Hydrodynamic Equations and Solution Procedures 3
3. Transport Equations and Solution Procedures 8
4. Heat Transport Formulation 10
5. Noncohesive Sediment Settling, Deposition and Resuspension 12
6. Cohesive Sediment Settling, Deposition and Resuspension 24
7. Sediment Bed Geomechanical Processes 31
8. Sorptive Contaminant Transport 42
9. Model Configuration and Input Files 43
10. Model Output Files and Post-Processing 84
11. References 85
Appendix A: Compilation Considerations 90
Appendix B: Sample Applications 92
-------
1. Introduction
This document outlines the theoretical and computational aspects of, EFDC1D
(Environmental Fluid Dynamics Code - One-Dimensional), a box or control volume
based one-dimensional hydrodynamic and sediment-contaminant transport model. The
model is implemented in a stand-alone version and also incorporated into HSPF
(Hydrologic Simulation Package: Fortran).
Model Features include:
A. Box or reach based spatial data structure, compatible with existing HSPF data
structure, for representing one-dimensional channel networks.
B. Utilization of water surface elevation dependent descriptions of channel cross-
section area, surface width, wetted perimeter and buoyancy centroid,
including representation of overbank regions.
C. Bi-directional unsteady flow and the ability to accommodate unsteady inflows
and outflows associated with upstream inflows, lateral inflows and
withdrawals, groundwater-surf ace water interaction, evaporation and direct
rainfall. The model includes representation of hydraulic structures such as
dams and culverts. Downstream boundary conditions include rating curves
and time varying water surface elevation.
D. The model includes a generic one-dimensional transport solver for salinity,
temperature and multiple sediment and contaminant classes. Longitudinal
dispersive transport is represented. Sources and sinks will be represented
consistent with continuity constraints.
E. Buoyancy effects due to salinity and temperature are dynamically coupled
with the hydrodynamic component using an equation of state. Temperature
transport includes a predictive surface heat exchange formulation representing
the effects of solar radiation, long wave back radiation, and latent and sensible
heat transfer.
F. For sediment transport, the model includes settling, deposition and
resuspension of multiple size classes of cohesive and noncohesive sediments.
The bed is represented by multiple layers of mixed sediment classes. A bed
consolidation model is implemented to predict time variations of bed depth,
void ratio, bulk density and shear strength. The sediment bed representation is
dynamically coupled to the cross-sectional area representation to account for
area changes due to deposition and resuspension.
The overall approach taken in developing the flow and sediment transport model was to
minimize code development by the utilization of existing process subroutines from the
multi-dimensional EFDC (Environmental Fluid Dynamic Code) model (Hamrick 1992;
1
-------
Hamrick 1996; Hamrick and Wu 1997). Process routines from the EFDC model are
utilized to satisfy the following requirements:
A. A fully dynamic one-dimensional solver for the momentum and continuity
equations with channel cross-section area, surface width, bottom width, wetted
perimeter and buoyancy centroid as functions of the water surface elevation.
B. Time varying upstream inflows, and lateral inflows and withdrawals including
corresponding sediment loads.
C. Hydraulic control structures and rating curve boundary conditions.
D. Time varying downstream boundary conditions for water surface elevation,
salinity, temperature and sediment concentration.
E. A generic one-dimensional transport solver utilizing a monotone, positive definite
scheme which minimizes numerical diffusion
F. A fully predictive surface heat exchange formulation which includes evaporation
G. An equation of state relating density to salinity and temperature.
H. A multiple class sediment processes module that incorporates a wide variety of
parameterization for settling, deposition and resuspension of cohesive and
noncohesive sediments.
I. A multiple layer bed module that includes a bed consolidation solver and
parameterizations relating void ratio, bulk and dry density, and shear strength.
Utilizing the above existing routines, code development focused on the main driver
program, input and output routines for the stand alone version of the model, interface
routines for the HSPF embedded version of the model, and a new hydrodynamic solver
optimized for one-dimensional channel network applications. The following sections of
this document summarize theoretical and computational formulations of the governing
hydrodynamic and transport equations, and model input and output files.
-------
2. Hydrodynamic Equations and Solutions Procedures
The one-dimensional momentum and continuity equations are
fr
(2.1)
(2.2)
where
A = cross-section area
b = dimensionless buoyancy = (p-p0)/p0
Cbp = dimensionless bottom resistance coefficient over the wetted
perimeter
g = acceleration of gravity
dc = distance down from water surface to channel area centroid
Q = discharge
RH = hydraulic radius
qi = lateral runoff per unit channel length
qs= rate of change of cross-sectional area due to sediment bed
resuspension and deposition
qr = tributary inflow per unit channel length
Ws = surface width of the channel
£= water surface elevation
The centroid depth is determined by
(2.3)
with the integrals denoting integration across the surface width and over the depth of the
channel. For a rectangular cross-section, dc is equal to one half of the depth. The cross-
section area, hydraulic radius, and water surface width are defined as functions of the
water surface elevation by tabular functions.
Equations (2.1) and (2.2) are solved using a box or control volume spatial representation.
e!l fen _
(2.4)
A :m A'.m-
-------
-a,=fe+a+Q-X (2-5)
Cross-sectional area, A, water surface elevation, £ and buoyancy, 6, are defined in
continuity control volumes. Discharge, Q, is defined in momentum control volumes
which are staggered relative to continuity control volumes, with momentum control
volume m being centered at the upstream face of continuity control volume m, as shown
in Figure 1. The continuity and momentum control volume lengths are denoted by /U.m
and AQ:m, respectively. Quantities defined at control volume centers, and determined by
interpolation, are denoted by the subscripts A:m and Q:m. Quantities at control volumes
upstream and downstream are defined by the subscripts m- and m+. The source terms in
the continuity equation are defined in continuity control volumes by
QL = lateral runoff discharge into volume m
Qs= rate of change of volume m due to bed resuspension and
deposition and bed consolidation.
QT = tributary discharge into volume m
The rate of change of the continuity control volume due to sediment deposition and
resuspension and bed consolidation is given by
Q^=-WBA.^A.Jfln (2.6)
where WB is the channel width over which the deposited sediment bed extends, and Bm is
the total thickness of the deposited bed.
-------
a) Continuity control volumes or cells
A(m-l)
C(m-l)
b(m-l)
q(m-l)
C(m-l)
A(m)
C(m)
b(m)
q(m)
C(m)
b) Momentum control volumes or cells
Q(m)
Figure 1. Continuity and momentum control volumes or cells. Dashed vertical lines
indicate positions of momentum volume boundaries in (a) and continuity
volume boundaries in (b).
-------
Equations (2.4) and (2.5) are integrated in time, between time levels n and n+1, using a
two-time level, semi-implicit scheme adopted from the EFDC model. The momentum
and continuity equations become
(2.8)
where 6 is the time-step. The implicit portion of the solution is represented by rewriting
(2.7) and (2.8) in the forms
n+1
and combining to give a system of linear equations:
These equations are solved for the new time level water surface elevation. New time
level discharges are then determined from (2.9). To guarantee volume continuity, the
new time level cross-sectional area is determined from (2.8).
Advective momentum fluxes in (2.7) and (2.9) are determined using simple upwind
differences:
-------
(212)
The data structure for the hydrodynamic and transport equation solutions is based on a
lookup table structure, with the upstream and downstream control volumes, m- and m+
being defined by m-= MM(m) and m+ = MP(m). An additional set of lookup tables are
used to define tributary inflow connections in the continuity equation.
At the upstream end of the main channel and each tributary channel, a no flow boundary
condition is implemented in the momentum equation and upstream inflow is represented
by a lateral source in continuity equation. This type of upstream boundary condition is
referred to as a waterfall boundary condition.
Downstream boundary conditions for the solution of the hydrodynamic equations
include:
Specified water surface elevation
C = CB® (2.13)
where £B is a specified time series or harmonic function.
Specified incoming wave characteristic
(— S— =AW ^
w
L—
\gws
where
-------
3. Transport Equations and Solution Procedures
The transport equation for a dissolved or suspended material, represented by the
concentration, C, is
dt (AC} + dx (QC} = dx(ADcdxC)+ qLCL + qsCs + qTCT + WBJCB - WSJCS (3.1)
where
C = concentration (salinity, temperature, or suspended sediment)
CL = concentration of lateral runoff inflow
Cs = concentration associated with pore water component of water column-sediment bed
exchange
CT = concentration of tributary inflow
DC = longitudinal dispersion coefficient
Ws= surface width of channel
WB = bed deposit! onal width of channel
Jcs= surface outflux of C
JCB = bed influx of C
When C represents temperature, the last two terms on the right of (3.1) are retained and
account for net water surface heat flux and convective heat flux from the bed,
respectively. When C represents suspended sediment, the next to the last term on the right
of (3.1) is retained and represents sediment resuspension and deposition, with the term
qsCs set to zero.
Equation (3.1) is solved by a fractional step method at the same time-step as the
hydrodynamic equations. With C defined in the continuity control volumes, the first step
is:
(3-2)
where
Qn+l/2=-(Qn+l
-------
It is noted that (3.2) is consistent with the continuity equation (2.6) that it reduces to
when all concentrations are set to unity. The second step is given by
Y+1 (3.4)
Jm ^ '
The surface and bottom fluxes are defined at the new time level to maintain a positive
concentration when Jcs represents the settling or deposition of suspended sediment. The
advective fluxes in (3.2) are given by a weighted upwind-central difference form
/•VJ+l/2/rn 7^ / />»«+l/2 ,~.\ ." ,n \ , r^ ( • //-^n+1/2 r\\ /-»n \
Qm CQnt=Fu(max(Qm ,0)Cm_)+Fu(mm(Qm ,0)Cm)
'
where the flux weights, Fu and Fc, are determined by a flux limiter that insures a
monotone and positive definite solution while also reducing numerical diffusion
associated with upwind flux.
-------
4. Heat Transport Formulation
For thermal transport and temperature simulations, the water surface heat flux for the
transport equation (3.1), when C represents heat (C=pwcpwT) is:
JTS = eaT* (0.39 - O.OS^2 )(l - BCCC) + 4arf? (T.-Ta) (4.1)
- Ta)+ cePaLjui + VI (ess - Rhesa)(0.622P~1)- Is
where cpw is the specific heat of water. The heat fluxes on the right hand side of (4.1) are
based on the NOAA Geophysical Fluid Dynamic Laboratory's atmospheric heat
exchange formulation (Rosati and Miyakoda 1988). The first two terms represent net
longwave back radiation where
Ts = water surface temperature
Ta = atmospheric temperature
e= emissivity
a= Stefan-Boltzman constant
ea = atmospheric vapor pressure in millibars
Cc = fractional cloud cover
Bc = empirical constant equal to 0.8.
The third term is the convective or sensible heat flux where
Ch = dimensionless transfer coefficient on the order of 10"3 in magnitude,
pa = atmospheric density,
cpa = specific heat of air.
The last term represents latent heat transfer where
ce = dimensionless transfer coefficient on the order of 10"3 in magnitude,
L = latent heat of evaporation
ess = saturation vapor pressures in millibars corresponding to the water surface
temperatures
esa = saturation vapor pressures in millibars corresponding to the atmospheric
temperature
Rh = fractional relative humidity
pa = atmospheric pressure in millibars
The incident shortwave solar radiation, /„ at the water surface (watts/m2) is given by
/, = 0.5/0 (l - Aa + rsec!")(l - <*)(! - 0.62Q + 0.0019$ (4.2)
where
10
-------
70 = shortwave solar radiation at the top of the atmosphere
Aa= water vapor plus ozone adsorption coefficient (0.09)
T= atmospheric attenuation coefficient (0.7)
\|/ = zenith angle
a = surface albedo
J> = solar noon angle in degrees.
The bottom heat flux is given by:
where
TB = bed temperature
pb = bed bulk density
cpb = specific heat of the water-solid bed mixture
CM = dimensionless convective heat exchange coefficient on the order of 10"3.
The remaining irradiance at the sediment bed-water interface being adsorbed into the
sediment bed is
4 = rls exp(-/?//)+ (1 - r)Is exp(-£#) (4.4)
where
fif= fast scale attenuation coefficient (I/meters)
fis = slow scale attenuation coefficient (I/meters)
r = a distribution fraction between zero and one
For shallow water environments, r is set to one and fif generally falls within the range of
0.2 to 4 per meter. The thermal balance for the bed is given by
(4-5)
where HB is the thermal thickness of the bed. Equation (4.5) serves to couple the bed
with the water column.
11
-------
5.
Noncohesive Sediment Settling, Deposition and Resuspension
Noncohesive inorganic sediments settle as discrete particles, with hindered settling and
multiphase interactions becoming important in regions of high sediment concentration
near the bed. At low concentrations, the settling velocity for the y'th noncohesive
sediment class corresponds to the settling velocity of a discrete particle:
>%•=••%. (5.1)
Useful expressions for the discrete particle settling velocity that depend on the sediment
density, effective grain diameter, and fluid kinematic viscosity are provided by van Rijn
(1984b):
r>
*
18
4'
1.1
d.
1 00/jm
is the reduced gravitational acceleration and
(5.4)
v
is the sediment grain densimetric Reynolds number.
At higher concentrations under hindered settling conditions, the settling velocity is less
than the discrete velocity and can be expressed in the form
( 1 VY (5-5)
where ps is the sediment particle density with values of n ranging from 2 (Cao et al.
1996) to 4 (Van Rijn 1984). The expression (5.5) is approximated to within 5 per cent by
(5'6)
.
i Psi
12
-------
for total sediment concentrations up to 200,000 mg/liter. For total sediment
concentrations less than 25,000 mg/liter, neglect of the hindered settling correction
results in less than a 5 per cent error in the settling velocity, which is well within the
range of uncertainty in parameters used to estimate the discrete particle settling velocity.
Noncohesive sediment is transported as bed load and suspended load. The initiation of
both modes of transport begins with erosion or resuspension of sediment from the bed
when the bed stress, Tb, exceeds a critical stress referred to as the Shield's stress, Tcs. The
Shield's stress depends upon the density and diameter of the sediment particles and the
kinematic viscosity of the fluid, and can be expressed in empirical dimensionless
relationships of the form:
o _
CSJ
gd,
(5.7)
Useful numerical expressions of the relationship (5.7), provided by van Rijn (1984b), are:
-0.64
-0.1
10
(5.8)
0.055 :
A number of approaches have been used to distinguish when a particular sediment size
class is transported as bed load or suspended load under specific local flow conditions
characterized by the bed stress or bed shear velocity:
(5.9)
where Tb is the kinematic bed stress (dynamic stress divided by water density). The
approach proposed by van Rijn (1984a) is adopted in the EFDC model and is as follows.
When the bed velocity is less than the critical shear velocity
(5.10)
no erosion or resuspension takes place and there is no bed load transport. Sediment in
suspension under this condition will deposit to the bed as will be subsequently discussed.
When the bed shear velocity exceeds the critical shear velocity but remains less than the
settling velocity,
w
so]
(5.11)
13
-------
sediment will be eroded from the bed and transported as bed load. Sediment in
suspension under this condition will deposit to the bed. When the bed shear velocity
exceeds both the critical shear velocity and the settling velocity, bed load transport ceases
and the eroded or resuspended sediment will be transported as suspended load. These
various transport modes are further illustrated by reference to Figure 2, which shows
dimensional forms of the settling velocity relationship (5.2) and the critical Shield's shear
velocity (5.10), determined using (5.8) for sediment with a specific gravity of 2.65. For
grain diameters less than approximately 0.130 mm (130 |im) the settling velocity is less
than the critical shear velocity and sediment resuspends from the bed when the bed shear
velocity exceeds the critical shear velocity will be transported entirely as suspended load.
For grain diameters greater than 0.130 mm, eroded sediment is transported as bed load in
the region corresponding to (5.11), and then as suspended load when the bed shear
velocity exceeds the settling velocity.
14
-------
1 G-TU | —
1 o+nn
to I
E
>s
tj 1 a I~H
o
"S -
>
O)
c
-1— 1
-l-l
m IP n?
D I
03
Oi
to
:
-
-
I c U't
C
s
/
/
t
1
t
t
t
t
t
t
t
t
1 1 1 1 1 1 II
ritical shields shear velocity
ettlina velocity
,
/'
t
,'
/ +
i
\ i 1 1 1 1 M
,.-'''
x^
s^
./
\ 1 1 1 1 III
^,-
^'-''
/
/^
\ 1 1 1 1 III
,,-'''
//
'
\ 1 1 1 1 1 II
1e-05 1e-04 1e-03 1e-02 1e-01
grain diameter (meters)
1e+00
Figure 2. Critical Shield's shear velocity and settling velocity as a function of sediment
grain size.
15
-------
In the EFDC model, the preceding set of rules is used to determine the mode of transport
of multiple size classes of noncohesive sediment. Bed load transport is determined using
a general bed load transport rate formula:
where qB is the bed load transport rate (mass per unit time per unit width) in the direction
of the near bottom horizontal flow velocity vector. The function 0 depends on the
Shield's parameter
and the critical Shield's parameter defined by (5.7) and (5.8). A number of bed load
transport formulas explicitly incorporate the settling velocity. However, since both the
critical Shield's parameter and the settling velocity are unique functions of the sediment
grain densimetric Reynolds number, the settling velocity can also be expressed as a
function of the critical Shield's parameter with (5.12) remaining an appropriate
representation.
A number of bed load formulations developed for riverine prediction (Ackers and White
1973; Laursen 1958; Yang 1973; Yang and Molinas 1982) do not readily conform to (1)
and were not incorporated as options in the EFDC model. Two widely used bed load
formulations that do conform to (5.12) are the Meyer-Peter and Muller (1948) and
Bagnold (1956) formulas and their derivatives (Raudkivi 1967; Neilson 1992; Reid and
Frostick 1994) that have the general form
where
0 = 0(O or 0fa) (5.15)
The Meyer-Peter and Muller formulations are typified by
while Bagnold formulations are typified by
(5
with Bagnold's original formula having y equal to zero. The Meyer-Peter and Muller
formulation has been extended to heterogeneous beds by Suzuki et al. (1998), while
16
-------
Bagnold's formula has been similarly extended by van Niekerk et al. (1992). The bed
load formulation by van Rijn (1984a) has the form
0.053 <518>
and has been incorporated into the CH3D-SED model and modified for heterogeneous
beds by Spasojevic and Holly (1994). Equation (5.18) can be implemented in the EFDC
model with an appropriately specified 0. A modified formulation of the Einstein bed load
function (Einstein 1950) that conforms to (5.12) and (5.14) has been presented by
Rahmeyer (1999) and will be later incorporated into the EFDC model.
The procedure for coupling bed load transport with the sediment bed in the EFDC model
is as follows. First, the magnitude of the bed load mass flux per unit width is calculated
according to (5.12) at horizontal model cell centers, denoted by the subscript c. The cell
center flux is then transformed into cell center vector components using
(5.19)
where u and v are the cell center horizontal velocities near the bed. Cell face mass fluxes
are determined by downwind projection of the cell center fluxes
Ibfa = \
-------
Under conditions when the bed shear velocity exceeds the settling velocity and critical
Shield's shear velocity, noncohesive sediment will be resuspended and transported as
suspended load. When the bed shear velocity falls below both the settling velocity and
the critical Shield's shear velocity, suspended sediment will deposit to the bed. A
consistent formulation of these processes can be developed using the concept of a near
bed equilibrium sediment concentration. Under steady, uniform flow and sediment
loading conditions, an equilibrium distribution of sediment in the water column tends to
be established, with the resuspension and deposition fluxes canceling each other. Using a
number of simplifying assumptions, the equilibrium sediment concentration distribution
in the water column can be expressed analytically in terms of the near bed reference or
equilibrium concentration, the settling velocity and the vertical turbulent diffusivity. For
unsteady or spatially varying flow conditions, the water column sediment concentration
distribution varies in space and time in response to sediment load variations, changes in
hydrodynamic transport, and associated nonzero fluxes across the water column-sediment
bed interface. An increase or decrease in the bed stress and the intensity of vertical
turbulent mixing will result in net erosion or deposition, respectively, at a particular
location or time.
To illustrate how an appropriate suspended noncohesive sediment bed flux boundary
condition can be established, consider the two-dimensional, in the vertical plane, form of
the sediment transport equation,
dz(wS) = dz-
where u and w are the along channel and vertical velocities respectively, Kv is the vertical
diffusion coefficient, H is the water depth, and z is the dimensionless vertical coordinate
varying from 0 at the bed to 1 at the water surface. For nearly uniform horizontal
conditions, (5.22) can be approximated by:
Integrating (5.23) over the depth gives
dt(HS)=J0 (5.24)
where the over bar denotes the mean over the depth. Subtracting (5.24) from (5.23) gives
-J0
Assuming that the rate of change of the deviation of the sediment concentration from the
mean is small
dt(HS')« dt(HS) (5.26)
18
-------
allows (5.26) to be approximated by
(5.27)
Integrating (5.27) once gives
P_ ^i ^ (5-28)
Very near the bed, (5.28) can be approximated by
=- <5'29)
Neglecting stratification effects the near bed diffusivity is approximately
(5.30)
Introducing (5.30) into (5.29) gives
le__^A (5.31)
z zws
where
R = ^ (5.32)
is the Rouse parameter. The solution of (5.31) is
c__A,_£ (5.33)
o + R
The constant of integration is evaluated using
S=Seq : z = zeq and Jo = 0 (5.34)
which sets the near bed sediment concentration to an equilibrium value, defined just
above the bed under no net flux condition. Using (5.34), equation (5.33) becomes
eq
z J w.
19
-------
For nonequilibrium conditions, the net flux is given by evaluating (5.35) at the
equilibrium level
where Sne is the actual concentration at the reference equilibrium level. Equation (5.36)
clearly indicates that when the near bed sediment concentration is less than the
equilibrium value a net flux from the bed into the water column occurs. Likewise when
the concentration exceeds equilibrium, a net flux to the bed occurs.
For the relationship (5.36) to be useful in a numerical model, the bed flux must be
expressed in terms of the transport model's representation of sediment mean
concentration. For application in an one-dimensional or two-dimensional model, (5.35)
can be integrated over the depth to give
where
lnfc)
(5.38)
\
-1.
:R*l
defines an equivalent layer mean equilibrium concentration in terms of the near bed
equilibrium concentration. The corresponding quantity in the sediment transport
equation (3.1) is
(5-39)
An alternate formulation for one-dimensional and two-dimensional, depth averaged,
model applications retain the complete form of (5.28):
A more general form for the vertical diffusivity is:
Kv / . yi (5.41)
-f=Koq—=u,K2(l-z) ^ }
n n
Introducing (5.41) into (5.40) gives
20
-------
no, o *r*J0 (5-42)
z /-I \^
z(l-z) z wx
A close form solution of (5.42) is possible for A, equal to zero. Although the resulting
diffusivity is not as reasonable as the choice of k equal to one, the resulting vertical
distribution of sediment is much more sensitive to the near bed diffusivity distribution
than the distribution in the upper portions of the water column. For A, equal to zero, the
solution of (5.42) is
Evaluating the constant of integration using (5.43) gives
- (5'44)
z ' eq
For nonequilibrium conditions, the net flux is given by evaluating (5.44) at the
equilibrium level
(5-45)
where Sne is the actual concentration at the reference equilibrium level. Since zeq is on the
order of the sediment grain diameter divided by the depth of the water column, (5.45) is
essentially equivalent (5.35). To obtain an expression for the bed flux in terms of the
depth average sediment concentration, (5.44) is integrated over the depth to give
(5-46)
where
l
'-
(5.47)
The corresponding quantity in the sediment transport equation (3.1) is
21
-------
(5.48)
When multiple sediment size classes are simulated, the equilibrium concentrations given
by (5.38) and (5.47) are adjusted by multiplying by their respective sediment volume
fractions in the surface layer of the bed.
The specification of the water column-bed flux of noncohesive sediment has been
reduced to specification of the near bed equilibrium concentration and its corresponding
reference distance above the bed. Garcia and Parker (1991) evaluated seven
relationships, derived by combinations of analysis and experiment correlation, for
determining the near bed equilibrium concentration as well as proposing a new
relationship. All of the relationships essentially specify the equilibrium concentration in
terms of hydrodynamic and sediment physical parameters
Seq = Seq(d,ps,pw,ws,u^ v) (5.49)
including the sediment particle diameter, the sediment and water densities, the sediment
settling velocity, the bed shear velocity, and the kinematic molecular viscosity of water.
Garcia and Parker concluded that the representations of Smith and McLean (1977) and
Van Rijn (1984b) as well as their own proposed representation perform acceptably well
when tested against experimental and field observations.
Smith and McLean's formula for the equilibrium concentration is
(5.50)
where y0 is a constant equal to 2.4E-3 and Tis given by
T- T»~T™ _u^~u^
where Tb is the bed stress and Tcs is the critical Shields stress. The use of Smith and
McLean's formulation requires that the critical Shields stress be specified for each
sediment size class. Van Rijn's formula is
Seq= 0.015 Ps4T3l2R-d1'5 (5'52)
where zeq* ( = Hzeq ) is the dimensional reference height and Rd is a sediment grain
Reynolds number. When Van Rijn's formula is selected for use in EFDC1D, the critical
Shields stress in internally calculated using relationships from Van Rijn (1984b). Van
Rijn suggested setting the dimensional reference height to three-grain diameters. In the
22
-------
EFDC1D model, the user specifies the reference height as a multiple of the largest
noncohesive sediment size class diameter.
Garcia and Parker's general formula for multiple sediment size classes is
(5.54)
(5.55)
where A is a constant equal to 1.3E-7, dso is the median grain diameter based on all
sediment classes, A, is a straining factor, FH is a hiding factor and a^ is the standard
deviation of the sedimentological phi scale of sediment size distribution. Garcia and
Parker's formulation is unique in that it can account for armoring effects when multiple
sediment classes are simulated. For simulation of a single noncohesive size class, the
straining factor and the hiding factor are set to one. The EFDC1D model has the option
to simulate armoring with Garcia and Parker's formulation. For armoring simulation, the
current surface layer of the sediment bed is restricted to a thickness equal to the
dimensional reference height.
23
-------
6. Cohesive Sediment Settling, Deposition and Resuspension
The settling of cohesive inorganic sediment and organic particulate material is an
extremely complex process. Inherent in the process of gravitational settling is the process
of flocculation, where individual cohesive sediment particles and particulate organic
particles aggregate to form larger groupings or floes having settling characteristics
significantly different from those of the component particles (Burban et al. 1989, 1990;
Gibbs 1985; Mehta et al. 1989). Floe formation is dependent upon the type and
concentration of the suspended material, the ionic characteristics of the environment, and
the fluid shear and turbulence intensity of the flow environment. Progress has been made
in first principles mathematical modeling of floe formation or aggregation, and
disaggregation by intense flow shear (Lick and Lick 1988; Tsai et al. 1987). However,
the computational intensity of such approaches precludes direct simulation of flocculation
in operational cohesive sediment transport models for the immediate future.
An alternative approach, which has met with reasonable success, is the parameterization
of the settling velocity of floes in terms of cohesive and organic material fundamental
particle size, d\ concentration, S; and flow characteristics such as vertical shear of the
horizontal velocity, du/dz, shear stress, Avdu/dz, or turbulence intensity in the water
column or near the sediment bed, q. This has allowed semi-empirical expressions having
the functional form
to be developed to represent the effective settling velocity. A widely used empirical
expression, first incorporated into a numerical model by Ariathurai and Krone (1976),
relates the effective settling velocity to the sediment concentration:
(6.2)
i "-' i
w = w
where a is an empirical constant and the o subscript denotes a reference value.
Depending upon the reference concentration and the value of a, this equation predicts
either increasing or decreasing settling velocity as the sediment concentration increases.
Equation (6.2) with user-defined base settling velocity, concentration and exponent is an
option in the EFDC1D model. Hwang and Mehta (1989) proposed
aST (6.3)
based on observations of settling at six sites in Lake Okeechobee. This equation has a
general parabolic shape with the settling velocity decreasing with decreasing
concentration at low concentrations, and decreasing with increasing concentration at high
concentrations. A least squares regression analysis for the parameters a, m, and n in (6.3)
24
-------
was shown to agree well with observational data. Equation (6.3) does not have a
dependence on flow characteristics, but is based on data from an energetic field condition
having both currents and high frequency surface waves. A generalized form of (6.3) can
be selected as an option in the EFDC1D model.
Ziegler and Nisbet (1994, 1995) proposed the following formulation to express the
effective settling as a function of the floe diameter, df.
ws = ad] (6.4)
with the floe diameter given by:
( \l/2
I a, I
(6.5)
where S is the sediment concentration, et^is an experimentally determined constant and
Txz and Tyz are the x and y components of the turbulent shear stresses at a given position
in the water column. Other quantities in (6.4) have been experimentally determined to fit
the relationships:
(6.6)
b = -0.8 - 0.51o$ + - B (6-7)
where B} and B2 are experimental constants. This formulation is also an option in the
EFDC ID model.
A final settling option in EFDC ID is based on that proposed by Shrestha and Orlob
(1996). The formulation in EFDC1D has the form
(6.8)
0.039G
where
is the magnitude of the vertical shear of the horizontal velocity. It is noted that all of
these formulations are based on specific dimensional units for input parameters and
predicted settling velocities and that appropriate unit conversions are made internally in
their implementation in the EFDC1D model.
25
-------
Water column-sediment bed exchange of cohesive sediments and organic solids is
controlled by the near bed flow environment and the geomechanics of the deposited bed.
Net deposition to the bed occurs as the flow-induced bed surface stress decreases. The
most widely used expression for the depositional flux is:
r=\
b \ —
(6.10)
0 :
where Tb is the stress exerted by the flow on the bed, Tcd is a critical stress for deposition
that depends on sediment material and floe physiochemical properties (Mehta etal. 1989),
and Sd is the near bed depositing sediment concentration. The critical deposition stress is
generally determined from laboratory or in situ field observations and values ranging
from 0.06 to 1.1 N/m2 have been reported in the literature. Given this wide range of
reported values, in the absence of site specific data the depositional stress is generally
treated as a calibration parameter. The critical depositional stress is an input parameter in
the EFDC ID model.
Since the near bed depositing sediment concentration in (6.10) is not directly calculated,
the procedures of Chapter 5 can be applied to relate the near bed depositional
concentration to the bottom layer or depth average concentration. Using (5.33) the near
bed concentration during times of deposition can be determined in terms of the bottom
layer concentration for two-dimensional, in the vertical plane, and three-dimensional
model applications. Inserting (6.10) into (5.33) and evaluating the constant at a near bed
depositional level gives
Td+(l-Td)-£
sd
(6.11)
For use in a two-dimensional, depth averaged, or one-dimensional application, (6.11) is
integrated over the total depth to give
£,=
L+-
S:R=l
(6.12)
J
The corresponding quantities in the one-dimensional sediment transport equation (3.1)
are
26
-------
SB
l-R
(6.13)
An alternate formulation for depth averaged and one-dimensional model applications is
obtained by combining (6.10) with (5.43). The constant of integration is evaluated at a
near bed depositional level to give
(6.14)
Integrating (6.14) over the depth gives
(6.15)
\(2+R(l-Zd)
d ^ 2(1 + R)
The corresponding quantities in the one-dimensional sediment transport equation (3.1)
are
SB
(6.16)
It is noted that the assumptions used to arrive at the relationships (6.12) and (6.15) are
more tenuous for cohesive sediment than the similar relationships for noncohesive
sediment. The settling velocity for cohesive sediment is highly concentration dependent
and the use of a constant settling velocity to arrive at (6.12) and (6.15) is questionable.
The specification of an appropriate reference level for cohesive sediment is difficult. One
possibility is to relate the reference level to the floe diameter using (6.5). An alternative
is to set the reference level to a laminar sub-layer thickness
v(S)
Hu*
(6.17)
27
-------
where v(S) is a sediment concentration dependent kinematic viscosity, and the water
depth is included to nondimensionalize the reference level. A number of investigators,
including Mehta and Jiang (1990) have presented experimental results indicating that at
high sediment concentrations, cohesive sediment-water mixtures behave as high viscosity
fluids. Mehta and Jaing's results indicate that a sediment concentration of 10,000 mg/L
results in a viscosity ten times that of pure water, and that the viscosity increases
logarithmically with increasing mixture density. Use of the relationships (6.12) and
(6.16) is optional in the EFDC1D model. When they are used, the reference height is set
using (6.17) with the viscosity determined using Mehta and Jaing's experimental
relationship between viscosity and sediment concentration.
Cohesive bed erosion occurs in two distinct modes, mass erosion and surface erosion.
Mass erosion occurs rapidly when the bed stress exerted by the flow exceeds the depth
varying shear strength, Ts of the bed at a depth, Hme below the bed surface. Surface
erosion occurs gradually when the flow-exerted bed stress is less than the bed shear
strength near the surface but greater than a critical erosion or resuspension stress, Tce
which is dependent on the shear strength and density of the bed. A typical scenario under
conditions of accelerating flow and increasing bed stress would involve first the
occurrence of gradual surface erosion, followed by a rapid interval of mass erosion,
followed by another interval of surface erosion. Alternately, if the bed is well
consolidated with a sufficiently high shear strength profile, only gradual surface erosion
would occur. Transport into the water column by mass or bulk erosion can be expressed
in the form
m
jr _ j _ me
o ~ J SB ~
(T
-------
(6.20)
or
Tb>Tce (6.21)
where dme/dt is the surface erosion rate per unit surface area of the bed and Tce is the
critical stress for surface erosion or resuspension. The critical erosion rate and stress and
the parameters a, ft, and y are generally determined from laboratory or in situ field
experimental observations. Equation (6.20) is more appropriate for consolidated beds,
while (6.21) is appropriate for soft partially consolidated beds. The base erosion rate and
the critical stress for erosion depend upon the type of sediment, the bed water content,
total salt content, ionic species in the water, pH and temperature (Mehta et al. 1989) and
can be measured in laboratory and sea bed flumes.
The critical erosion stress is related to but generally less than the shear strength of the
bed, which in turn depends upon the sediment type and the state of consolidation of the
bed. Experimentally determined relationships between the critical surface erosion stress
and the dry density of the bed of the form
(6.22)
have been presented (Mehta et al. 1989). Hwang and Mehta (1989) proposed the
relationship
between the critical surface erosion stress and the bed bulk density with a, b, c, and pi
equal to 0.883, 0.2, 0.05, and 1.065, respectively for the stress in N/m2 and the bulk
density in gm/cm3. Considering the relationship between dry and bulk density
(A-/Q (6.24)
equations (6.22) and (6.23) are consistent. The EFDC1D model allows for a user defined
constant critical stress for surface erosion or the use of (6.23). Alternate predictive
expressions can be readily incorporated into the model.
Surface erosion rates ranging from 0.005 to 0.1 gm/m2-s have been reported in the
literature, and it is generally accepted that the surface erosion rate decreases with
increasing bulk density. Based on experimental observations, Hwang and Mehta (1989)
proposed the relationship
29
-------
dm\ ^ f 0.198 "| (6.25)
—- = 0.23 exp
dt J 17-1.0023 J
Iog10[-^]=0.23exp|
for the erosion rate in mg/cm2-hour and the bulk density in gm/cm3. The EFDC1D model
allows for a user defined constant surface erosion rate or predicts the rate using (6.25).
Alternate predictive expressions can be readily incorporated into the model. The use of
bulk density functions to predict bed strength and erosion rates in turn requires the
prediction of time and depth in bed variations in bulk density which is related to the water
and sediment density and the bed void ratio, e, by
(6.26)
Selection of the bulk density dependent formulations in the EFDC1D model requires
implementation of a bed consolidation simulation to predict the bed void ratio as
discussed in the following chapter.
30
-------
7. Sediment Bed Geomechanical Processes
This chapter describes the representation of the sediment bed in the EFDC1D model. To
make the information presented self-contained, the derivation of mass balance equations
and comparison with formulations used in other models are also presented.
Consider a sediment bed represented by discrete layers of thickness Bk, which may be
time varying. The conservation of sediment and water mass per unit horizontal area in
layer k are given by:
, 0))
where e is the void ratio, ps and pw are the sediment and water density and Js and Jw are
the sediment and water mass fluxes with k- and k+ defining the bottom and top
boundaries, respectively, of layer k. The mass fluxes are defined as positive in the
vertical direction and exclude fluxes associated with sediment deposition and erosion.
The last term in equation (7.1) represents erosion and deposition of sediment at the top of
the upper most bed layer, k=kb, where
Consistent with this partitioning of flux,
Jsk+ = Q;k*kb (7.4)
The last term in (7.2) represents the corresponding entrainment of bed pore water into the
water column during sediment erosion and entrainment of water column water into the
bed during deposition. The water flux, Jw±+, at the top of the upper most layer, kb, is not
necessarily zero, since it can include ambient seepage and pore water expulsion due to
bed consolidation.
Assuming sediment and water to be incompressible, (7. 1) and (7.2) can be written as:
A ' ' A
(7.6)
-------
where the water specific discharges
Jw±- = Pvftv*- (7.7)
^w.k+ ~ Pvflw.k-
have been introduced into (7.6). Four approaches for the solution of the mass
conservation equations (7.5) and (7.6) have been previously utilized. The solution
approaches, hereafter referred to as solution levels, increase in complexity and physical
realism and will be briefly summarized below.
The first level or simplest approach assumes specified time-constant layer thicknesses
and void ratios with the left sides of (7.5) and (7.6) being identically zero. Sediment
mass flux at all layer interfaces are then identical to the net flux from the bed to the water
column.
/s±- = /* :"- = l>"i
b (7-8)
=
s±+
Bed representations at this level, as exemplified by the RECOVERY model (Boyer et al.
1994), typically omit the water mass conservation equations. However, it is noted that
the water mass conservation is ill posed unless either q^. the specific discharge at the
bottom of the deepest layer or qkb+, the specific discharge at the top of the water column
adjacent layer, is specified. If q^. is set to zero, qkb+ is then required to exactly cancel the
entrainment terms is (7.6).
The second level of bed mass conservation representation assumes specified time
invariant layer thicknesses. The mass conservation equations (7.5) and (7.6) become
A P,
(7.10)
This system of 2 x kb equations includes fa unknown void ratios, fa unknown internal
sediment fluxes, and fa+i unknown specific discharges and is under determined unless
additional information is specified. The constant bed layer thickness option in the
WASPS model (Ambrose et al. 1993) uses specified burial velocities to define the
following internal sediment fluxes:
32
-------
J,:k-=~'Wb:k-Sk
wkk+=wkk+l_
s A (7.12)
*
where wb is the burial velocity and S is the sediment concentration (mass per unit total
volume). Use of the burial velocity eliminates the indeterminacy in (7.9) and allows its
solution for the void ratio. In the event that the sediment concentration in the upper most
layer becomes negative, the layer is eliminated and the underlying layer becomes water
column adjacent. The left side of the water mass conservation equations (7.10) is now
known, and the equation is more appropriately written as
ekm
The determination of the specific discharges using (7.13) can be viewed as either under
determined or physically inconsistent. As shown for the first level approach, the solution
of (7.13) is ill posed unless either q^_, the specific discharge at the bottom of the deepest
layer or qkb+, the specific discharge at the top of the upper most layer is independently
specified. If q^. is specified and the internal specific discharges are determined from
(7.13), qkb+ is then required to partially cancel the entrainment terms in (7.13). As will be
subsequently shown, the specific discharges can be dynamically determined using
Darcy's law. However, the specific discharges determined using Darcy's law and the
known void ratios are not guaranteed to satisfy (7.13); thus, the level two formulation is
dynamically inconsistent with respect to water mass conservation in the sediment bed.
The constant bed layer thickness option in WASPS ignores this problem entirely by not
considering the water mass balance and hence neglecting pore water advection of
dissolved contaminants.
The third level of bed mass conservation representation assumes specified time invariant
layer void ratios. The mass conservation equations (7.5) and (7.6) become
' 1 ^ _ 1 \ ( \J* (?'14)
Ps s*~ s''k+ ' * Ps
ek )„_ „,. . ^ i j,. . \ . (j^ .Yi C7-15)
This system of equations exhibits the same under determined nature as (7.9) and (7.10).
Specification of internal sediment fluxes or burial velocities allows (7.14) to be solved for
the layer thicknesses. Solution of (7.15) for the specific discharges then requires the
specification of either q^., the specific discharge at the bottom of the deepest layer or ^+,
33
-------
the specific discharge at the top of the uppermost layer. The variable bed layer thickness
option in the WASPS model (Ambrose et al. 1993) exemplifies the third level of bed
representation. Specifically, the thickness of the water column adjacent layer is allowed
to vary in time, while the thicknesses of the underlying layers remain constant. A
periodic time variation is specified for the bottom sediment flux in the uppermost layer
(7.16)
where At is the standard water time-step and NAt is the sediment compaction time. This
results in the thickness of the uppermost layer periodically returning to its initial value at
time intervals of NAt unless the thickness becomes negative due to net resuspension. In
that event, the underlying layer becomes the water column adjacent layer. The water
mass conservation (7.15) for all but the uppermost layer becomes
qk+=qk-=ql- • k*kb (7.1?)
which indicates that all internal specific discharges are equal to a specified specific
discharge at the bottom of layer 1. Given the solution for the time variation of the water
column adjacent thickness and bottom specific discharge, (7.15) can be solved for the
specific discharge at the top of the layer. The constant porosity bed option in EFDC1D is
also a level three approach. In EFDC1D, the internal sediment fluxes are set to zero and
the change in thickness of the water column adjacent layer is determined directly using
(7.14) while the underlying layers have time invariant thicknesses. As a result, the
internal water specific discharges are set to zero and the water entrainment and expulsion
in the water column adjacent layer are determined directly from (7.15). As a result, the
internal water specific discharges are set to zero and the water entrainment and expulsion
in the water column adjacent layer are determined directly from (7.15). The EFDC1D
model is configured to have a user specified maximum number of sediment bed layers.
At the start of a simulation, the number of layers containing sediment at a specific
horizontal location is specified. Under continued deposition, a new water column layer is
created when the thickness of the current layer exceeds a user specified value. If the
current water column adjacent layer's index is equal to the maximum number of layers,
the bottom two layers are combined and the remaining layers renumbered before addition
of the new layer. Under continued resuspension, the layer underlying the current water
column adjacent layer becomes the new adjacent layer when all sediment is resuspended
from the current layer.
The fourth level of bed representation accounts for bed consolidation by allowing the
layer void ratios and thicknesses to vary in time. The simplest and most elegant
formulations at this level utilize a Lagrangian approach for sediment mass conservation.
The Lagrangian approach requires that the sediment mass per unit horizontal area in all
layers, except the upper most, be time invariant and without loss of generality, the
34
-------
internal sediment fluxes can be set to zero. Consistent with these requirements (7.5)
becomes
P,
Expanding the left side of the water conservation equation (7.6), and using (7. 18) gives
The Lagrangian approach for sediment mass conservation also requires that the number
of bed layers vary in time. Under conditions of continued deposition, a new water
column adjacent layer would be added when either the thickness, void ratio or mass per
unit area of the current water column adjacent layer reaches a predefined value. Under
conditions of continued resuspension, the bed layer immediately under the current water
column adjacent layer would become the new water column adjacent layer when the
entire sediment mass of the current layer has been resuspended.
At the fourth and most realistic level of bed representation, three approaches can be used
to represent bed consolidation. Two of the approaches are semi-empirical with the first
assuming that the void ratio of a layer decreases with time. A typical relationship that is
used for the simple consolidation option in the EFDC model is
e=em + (eo- em )exp(-or(r - 10 )) (7.20)
where e0 is the void ratio at the mean time of deposition, t0, em is the ultimate minimum
void ratio corresponding to complete consolidation, and ens an empirical or experimental
constant. Use of (7.20) in the EFDC ID model involves specifying the deposit! onal void
ratio, the ultimate void ratios and the rate constants. The ultimate void ratio can be
specified as a function depth below the water column-bed interface. The actual
calculation involves using the initial void ratios to determine the deposition time t0, after
which (7.20) is used to update the void ratios as the simulation progresses. After equation
(7.20) is used to calculate the new time level void ratios, equation (7.18) provides the
new layer thicknesses. The water conservation equation (7.19) can then be solved using
: k+ = ~ " -
to determine the water specific discharges, provided that the specific discharge q^., at the
bottom of layer 1 is specified. When this option is specified in the EFDC1D model, the
specific discharge at the bottom of the bottom sediment layer is set to zero. Layers are
added and deleted in the manner previously described for EFDClD's constant porosity
option. The SED2D-WES model (Letter et al. 1998) utilizes a similar approach based on
a specified time variation of bulk density
35
-------
A = T = A, + (A0 - A Jexp(-^ - 0)
which in turn defines the variation in void ratio.
The second semi-empirical approach assumes that the vertical distribution of the bed bulk
density or equivalently the void ratio at any time is given by a self-similar function of
vertical position, bed thickness and fixed surface and bottom bulk densities or void ratios.
Functionally this is equivalent to
e=V(z,BT,ea,,e^ (7.23)
where V represents the function, z is a vertical coordinate measured upward from the
bottom of the lowest layer, and BT is the total thickness of the bed. This approach is used
in the original HSCTM model (Hayter and Mehta 1983), the new HSCTM model (Hayter
et al. 1999) and is an option in the CE-QUAL-ICM/TOXI model (Dortch et al. 1998).
The determination of the new time level layer thickness and void ratios requires an
iterative solution of equations (7.18) and (7.23). The solution is completed using (7.21)
to determine the water specific discharges.
The third and most realistic approach is to dynamically simulate the consolidation of the
bed. In the Lagrangian formulation, (7.18) is directly solved for the equivalent sediment
thickness
and the water conservation equation (7.19) is integrated to determine the void ratio.
The specific discharges in (7.25) are determined using the Darcy equation
where K is the hydraulic conductivity and u is the excess pore pressure defined as the
difference between the total pore pressure ut, and the hydrostatic pressure u^.
u = ut-uh (7.27)
The total pore pressure is defined as the difference between the total stress crand
effective stress oe.
36
-------
ut = a-ae (7.28)
The total stress and hydrostatic pressure are given by
(7.30)
wherepb is the water column pressure at the bed z\,. Solving for the excess pore pressure
using (7.27) through (7.30) gives
(7.31)
The specific discharge (7.26), can alternately be expressed in terms of the effective stress
07.32)
or the void ratio
where de/dac is a coefficient of compressibility.
For consistency with the Lagrangian representation of sediment mass conservation, a new
vertical coordinate £ defined by
d£ = 1 (7.34)
dz l + e
is introduced. The discrete form of (7.34) is
r _ **+-**- __ ^_ (7.35)
'- - -
where Ak is the equivalent sediment thickness previously defined by (7.24). Introducing
(7.34) into (7.26), (7.32), and (7.33) gives
K (7.36)
q = -- - - -du ^ }
37
-------
K (737)
where
^=_L£S (7.39)
is a compressibility length.
Three formulations for the solution the consolidation problem can be utilized. The void
ratio-excess pore pressure formulation, used in the EFDC1D model, evaluates the
specific discharges at the current time level «, using (7.36) and explicitly integrates (7.25)
where #is the time-step, to give the new time level void ratios. The layer thicknesses are
then determined by explicit integration of (7.18).
D \"+l / D \"
=
M +
Constitutive equations required for consolidation prediction generally express the
effective stress and hydraulic conductivity as functions of the void ratio. Thus the new
time level void ratio is used to determine new time level values of the effective stress and
hydraulic conductivity. The new time level excess pore pressure is then given by
the transformed equivalent of (7.31). The primary advantage of the void ratio-excess
pore pressure formulation is the simplicity of its boundary conditions
u=«b:£=Cb (7.43)
38
-------
(7-44)
The water column-sediment bed interface boundary condition generally sets Ub to zero if
the surface water flow is hydrostatic but can incorporate wave induced pore pressures.
The bottom boundary conditions allows either the specification of pressure or specific
discharge. The primary disadvantage of this formulation is the stability or positivity
criterion imposed on the time-step
A^ _ _ (7.45)
(7.46)
In practice, these criteria are readily satisfied if the consolidation time-step is identical to
the time-step of the hydrodynamic model. In the event that these criteria are not met
using the hydrodynamic time-step, the bed consolidation is sub-cycled using an integer
number of time sub-steps, satisfying (7.45) and (7.46), per each hydrodynamic time-step.
Alternately, the consolidation problem can be directly formulated in terms of the
effective stress or void ratio. Combining (7.25) and (7.37) using (7.39) gives the
effective stress formulation
(7.47)
The continuum equivalent is
(7.48)
which is parabolic since A, is negative. Combining (7.25) and (7.38) using (7.39) gives
the void ratio formulation
39
-------
(7.49)
The continuum equivalent is
Equation (7.50) is the discrete form of the finite strain consolidation equation first
derived by Gibson et al. (1967). Equation (7.50) was used by Cargill (1985) in the
formulation of a model for dredge material consolidation and by Le Normant et al.
(1998) to represent bed consolidation in a three-dimensional cohesive sediment transport
model.
The classic linear consolidation equation (Middleton and Wilcock 1994) omits the second
term associated with self weight in (7.50) and introduces a constant consolidation
coefficient
Cc = _(1+,)^JL (7.5D
* gpw
reducing (7.50) to
dte=Ccdzze (7.52)
Equation (7.52) has separable solutions of the form
-------
Bear (1979) notes that curve fitting of experimental data typically results in relationships
of the form
e-£o = -av(ae-aeo} (7.56)
for noncohesive and cohesive soils respectively, where av is the coefficient of
compressibility and Cc is the compression index. Graphical presentation of experimental
forms of (7.54) and (7.55) are presented in Cargill (1985) and Palermo et al. (1998)
which are generally consistent with (7.57) and suggest
(7.58)
as a candidate relationship between the void ratio and hydraulic conductivity for cohesive
sediment beds. Similarly, a linear relationship
e-eoocK-Ko (7.59)
would likely suffice for noncohesive sediment beds.
41
-------
8. Sorptive Contaminant Transport
The EFDC1D model currently includes the capability for simulating sorptive
contaminants including, heavy metals and hydrophobic organic compounds. Complete
documentation of this capability will be included in subsequent versions of this
document.
42
-------
9. Model Configuration and Input Files
Configuration of the EFDC1D model requires the development of model input files
describing the geometry of the channel network, initial conditions, boundary conditions,
solution options and output options. An alphabetical listing of input files is given in
Table 1. The second and third columns indicate the method by which the files are created
for the standalone version (SA) and the HSPF embedded version (HS) of EFDC1D. User
created files are denoted by "u", files automatically created by the model are denoted by
"a", files created by the HSPF/EFDC1D interface are denoted by "i", and provided
default data files are denoted by "d". Certain files can be user or interface created. Files
having standard EFDC model formats are denoted under the file description column.
The recommended sequence for the construction of the input files for configuration of the
model and set up for a simulation generally begins with the cellnet.inp file defining the
channel network. Following the creation of the cellnet.inp file, the channel section
properties file, chansec.inp, the master input file, efdcldinp, and inflow time series file,
qser.inp, should be created. If active screen display is selected in the efdcldinp, the
screen display control file show.inp should be properly set to the desired display location.
Default, zero concentration, time series files for dye, dser.inp, salinity, sser.inp,
temperature, tser.inp, cohesive sediment, sdser.inp, and noncohesive sediment, snser.inp
are also required. To complete the configuration for hydrodynamic simulations, the
qctl.inp file, defining flow control structures or rating curve properties, and thepser.inp
file, defining water surface elevation open boundary conditions must be created as
required for a specific application. Temperature or thermal simulations requires
atmospheric conditions from the aser.inp file and wind conditions from the wser.inp file.
If it is desired to include direct rainfall and evaporation in the simulation, the aser.inp can
be used even if thermal simulation is not selected. All of the EFDC ID input files
conform to standard templates which the model is designed to read, however user data
input into the templates is treated as free form across an input line. All of the input files
are internally documented, and the sample input file sets distributed with the model can
be used as starting templates for setting up a new simulation. The following subsections
of this chapter provide additional details regarding the user created input files.
43
-------
Table 1. Input files for the EFDC1D model.
File Name
SA
HS
Description
Aser.inp
u
u
File containing time series for atmospheric
conditions including air temperature, relative
humidity, rainfall, shortwave solar radiation and
cloud cover. Standard EFDC format
Cell.inp
Horizontal cell type identifier for the
computational domain
Standard EFDC format
Cellllt.inp
Horizontal cell type identifier for saving mass
transport to drive external water quality model
Standard EFDC format
Cellnet.inp
u
u
File specifying the channel network, including
certain boundary conditions
Chanjunij.inp
File defining junctions in the channel network
Standard EFDC format
Chansec.inp
u
u
File defining channel cross-section properties as a
function of depth or water surface elevation
Chansecij.inp
A reformated version of chansec.inp
Standard EFDC format
Dser.inp
u
u
File containing time series for dye tracer
concentration for use a boundary conditions or
loadings
Standard EFDC format
44
-------
Table 1 (continued). Input files for the EFDC1D model.
File Name
SA
HS
Description
Dxdy.inp
File containing channel cell control volume
information
Standard EFDC format
Efdcdat.inp
Aversion of multidimensional efdc.inp file
containing default data and solution options.
Should be modified only by experienced EFDC
users. Standard EFDC format
Efdcld.inp
u
u
Master input file providing run time control,
certain boundary conditions, and output control
Flowld.inp
File contains initial flow and discharge data from
the cellnet.inp file which has been reordered to the
computational grid. This information is used for
cold starts.
Flwctl.inp
File containing data on inflow and control
structure locations
Lxly.inp
File containing geographic orientation information
for channel cells
Standard EFDC format
Mapld2ij.inp
File containing network to computational grid
mapping data
Mappgns.inp
File containing additional network to
computational grid mapping data
Standard EFDC format
45
-------
Table 1 (continued). Input files for the EFDC1D model.
File Name
SA
HS
Description
Pser.inp
u
u
File containing time series of water surface
elevation for tidal open boundary conditions
Standard EFDC format
Qctl.inp
u
u
File specifying properties of hydraulic control
structures
Standard EFDC format
Qser.inp
u
u/i
File containing time series for dye tracer
concentration for use a boundary conditions or
loadings
Standard EFDC format
Restart.inp
File for restarting a simulation
Standard EFDC format
Sdser.inp
u
u/i
File containing time series for cohesive sediment
concentration for use a boundary conditions or
loadings. Standard EFDC format
Show.inp
u
u
File controlling screen output during model
execution. Standard EFDC format
Snser.inp
u
u/i
File containing time series for noncohesive
sediment concentration for use a boundary
conditions or loadings Standard EFDC format
Sser.inp
u
u
File containing time series for salinity
concentration for use a boundary conditions or
loadings. Standard EFDC format
46
-------
Table 1 (continued). Input files for the EFDC1D model.
File Name
Tmserld.inp
Tser.inp
Txser.inp
Wser.inp
SA
a
u
u
u
HS
a
u/i
u/i
u
Description
File contains time series output save locations
referenced to computational grid
File containing time series for temperature for use
a boundary conditions or loadings
Standard EFDC format
File containing time series for sorptive toxic
contaminant concentration for use a boundary
conditions or loadings
Standard EFDC format
File containing time series wind speed and
direction
Standard EFDC format
47
-------
9.1 Unit Conventions Used in EFDC1D
The EFDC1D model uses an arbitrary time convection. Simulation start time is specified
by TBEGIN in data block 5 of the efdcldinp file. This starting time can be viewed as the
elapsed time in days, hours, etc. from the 0.0 time origin. The time origin is arbitrary, but
is usually taken as midnight January 1 of a user defined reference year. The reference
year is not input as data into any of the input files, but a good practice is noting the
reference year in comment areas of all input files involving temporal data. The reference
time year should also be noted in the simulation title in data block 1 of efdcJdinp. For
example if the reference year were 1999, midnight March 1, 1999 would be 90.0 days.
Likewise, midnight March 1, 2000 would be 456.0 days, noting that 2000 is a leap year.
All files having time data include time adjustment factors and unit conversions. The
convention used is
Model Internal Time = TCxxx * (Input Time + TAxxx)
where the internal time is always seconds, and Taxxx and TCxxx are adjustment and
conversion factors. TCxxx converts from various input time units to seconds, (TCxxx =
86400, for input time units of days). TAxxx must always have the same units as input
time data.
The internal unit system in EFDC1D is meters-kilograms-seconds, with the exception of
concentration variables. Unit conversion factors are included for flow, elevation, and
concentration in all appropriate input files allowing, for example, a flow time series to be
input in CFS and converted to CMS. Salinity and temperature variables must the PSU
(practical salinity units) and Celsius since the internal equation of state requires these
units. The dye concentration state variable can have arbitrary input units, which are
carried through to output. Suspended sediment concentrations must be in mg/liter,
(which is equivalent to gm/m3) which is the conventional unit and is required for
sediment concentration influences in the equation of state. For the sediment bed, the
internal model variable is sediment mass per unit area, in gm/m2, for each sediment bed
layer. For correct coupling with water column sediment processes, settling velocities and
resuspension rates must the specified in m/s and gm/m2-sec, respectively. Toxic
contaminant concentrations can be in mg/liter or jig/liter at the user's discretion. For
toxic contaminant simulation, the equilibrium partition coefficients must be input in
liter/mg, the inverse of the sediment concentration units, as opposed to the traditional unit
of liter/Kg.
9.2 Aser.inp Input File
The atmospheric conditions file, aser.inp, specifies information for necessary for
temperature simulation including atmospheric pressure, air temperature, wet bulb air
temperature or relative humidity, rainfall rate, evaporation rate, water surface incident
solar short wave radiation, and fractional cloud cover. If wet bulb temperature is
specified instead of relative humidity, the relative humidity is internally calculated. Input
48
-------
relative humidity must be a decimal fraction rather than percent. The evaporation rate
can be either specified or internally calculated if the temperature simulation option is
active. When temperature simulation is not active, rainfall and evaporation rates from the
aser.inp file are still used as source and sink terms in the continuity equation. Input solar
short wave radiation should be converted using the SOLRCVT conversion factor to
watts/m2, as appropriated. An example of the aser.inp file is shown in Figure 3, and can
be used as a template for file construction.
9.3 Cellnet.inp Input File
The cellnet.inp file specifies the connectivity and individual cell properties of the channel
network. If also defines the location of control structures or rating curve boundaries and
volumetric inflows and associated concentrations. Control table identifier and time series
identifier for flow and concentration are also defined. It is noted that the use of a zero
concentration time series identifier with a nonzero flow identifier results in zero
concentrations being associated with that inflow.
Figure 4 shows an example channel network having the following properties:
Channel segment 5 (cells 32-36) flows freely into cell 28 of channel segment 4
(cells 22-31)
Channel segment 4 (cells 22-31) flows freely into cell 17 of channel segment 3
(cells 15-21)
Channel segment 3 (cells 15-21) flows by control structure into cell 6 of channel
segment 1 (cells 1-6)
Channel segment 2 (cells 7-14) flows freely into cell 4 of channel segment 1 (cells
1-6)
Channel segment 1 (cells 1-6) flows out of the domain by control rating curve,
discharge or water surface elevation time series, or tidal open boundary condition
(obc). The cell table above uses tidal obc.
or — denotes a cell having volume, cross-section area, surface width, etc,
determined by water surface elevation
+ denotes flow face where discharge is defined
F is free flowing connection
C is connection by control structure or rating curve
O is tidal open boundary
49
-------
The cellnet.inp file for this network is shown in Figure 5. The main feature notation
requiring mastery in construction of the cellnet.inp file is the concept of segment order.
The main stem channel segment is defined as order 1. There can be only one main stem
segment. Channel segments flowing into the main stem or 1st order segment are defined
as 2nd order segments. Channel segments flowing into the 2nd order segments are defined
as 3rd order segments and so on. Stream segments defined in this ordering fashion cannot
have internal control structures. Thus a continuing reach of a stream channel above a
control structure must be defined as a new segment and be considered as a tributary of the
stream segment below the control structure. Data columns in the cellnet.inp file are
internally explained in the file header lines.
C aser.inp
C
file, in free format across line, repeats naser=l times
C ATMOSPHERIC FORCING FILE, USE WITH 7 APRIL 97 AND LATER VERSIONS OF EFDC
C
C MASER
C TCASER
C TAASER
C IRELH
C RAINCVT
C EVAPCVT
C SOLRCVT
C CLDCVT
C IASWRAD
C REVC
C RCHC
C SWRATNF
C SWRATNS
C FSWRATF
C DABEDT
C TBEDIT
C HTBED1
C HTBED2
C PATH
C TDRY/TEQ
C TWET/RELH
C RAIN
C EVAP
C SOLSWR
C CLOUD
C
C MASER
C
=NUMBER OF TIME DATA POINTS
=DATA TIME UNIT CONVERSION TO SECONDS
=ADDITIVE ADJUSTMENT OF TIME VALUES SAME UNITS AS INPUT TIMES
=0 VALUE TWET COLUMN VALUE IS TWET, =1 VALUE IS RELATIVE HUMIDITY
=CONVERTS RAIN TO UNITS OF M/SEC
=CVRT EVAP TO M/SEC, IF EVAPCVT<0 EVAP IS INTERNALLY COMPUTED
=CONVERTS SOLAR SW RADIATION TO JOULES/SQ METER
=MULTIPLIER FOR ADJUSTING CLOUD COVER
=0 DIST SW SOL RAD OVER WATER COL AND BED, =1 ALL TO SURF LAYER
=1000*EVAP TRANSFER COEF, REVC<0 USE WIND SPD DEPD DRAG COEF
=1000*CONV HEAT TRAN COEF, REVC<0 USE WIND SPD DEPD DRAG COEF
=FAST SCALE SOLAR SW RADIATION ATTENUATION COEFFCIENT 1 . /METERS
=SLOW SCALE SOLAR SW RADIATION ATTENUATION COEFFCIENT 1 . /METERS
=FRACTION OF SOLSR SW RADIATION ATTENUATED FAST 00 .
=SOLAR SHORT WAVE RAD AT WATER SURFACE ENERGY FLUX/UNIT AREA
=FRATIONAL CLOUD COVER
TCASER TAASER IRELH RAINCVT EVAPCVT SOLRCVT CLDCVT
C IASWRAD REVC RCHC SWRATNF SWRATNS FSWRATF DABEDT TBEDIT HTBED1 HTBED2
C
C TASER(M)
C
3447
0 1.
0. 0354
120. 0354
120.0771
120. 1188
PATM(M) TDRY(M) TWET(M) RAIN(M) EVAP (M) SOLSWR (M) CLOUD (M)
/TEQ /RELH /HTCOEF
86400. 0. 1 1.0 -1. 0.95 1.00
5 1.5 2.5 0. 1. 1. 14. 0 0.001 0.
1005.47 11.72 0.74 O.OOOE+00 0.00 0.00 0.95
1005.47 11.72 0.74 O.OOOE+00 0.00 0.00 0.95
1004.46 11.72 0.77 O.OOOE+00 0.00 0.00 0.95
1003.44 12.22 0.75 O.OOOE+00 0.00 0.00 0.95
Figure 3. Example of the aser.inp file.
50
-------
36
21
20
19
35
+
34
seg 5 +
order 4 33
32
seg 3 18 | F
order 2 +
17 | F H 1 1 1 1 1 1 1 1 1
+ 22 23 24 25 26 27 28 29 30 31
16 |
+ seg 4
15 I order 3
seg 1
order 1
5 |
+
4 I F
3 I
7 8 9 10 11 12 13 14
seg 2
order 2
Figure 4. Example of a cell network.
51
-------
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
file cellnet.inp - cell network file
channel network data 1
nsegs - number of stream segments
nords - highest order segment
ncells - total number of cells
ncells - total number of cells
ncells - total number of cells
nsegs nords ncells nctl nqts nets nobc
5 4 36 0 0 0 0
channel network data 2
iseg - stream segment id number (main stem must be 1)
isord - order of this segment (main stem is 1)
isgdn - number of segment downstream from this segment
iscdn - cell number of most downstream cell in this segment
iscup - cell number of most upstream cell in this segment
iseg isord isgdn iscdn iscup
11016
2 2 1 7 14
3 2 1 15 21
4 3 3 22 31
5 4 4 32 36
Figure 5. Example of the cellnet.inp file.
52
-------
c
c cell data
c
c icell - cell id number
c iseg - segment id number
c itype - cell type id number
c 1 interior
c 2 downstream end of channel segment with free flowing
W connection to downstream segment
c 3 downstream end of channel segment with control
W structure or rating curve connection to downstream segment
c 4 downstream end of channel segment with time series outflow
c 5 downstream end of channel segment with tidal open
W boundary condition
c 6 head of channel segment
c iupc - number of upstream cell (use zero at head of channel segment)
c idnc - number of downstream cell (use zero at end of channel segment)
c ifru - number of upstream cell interacting by free flowing
W connection (use zero for null)
c ifrd - number of downstream cell interacting by free flowing
W connection (use zero for null)
c iqcu - number of upstream cell interacting by control structure
W or rating curve (use zero for null)
c iqcd - number of downstream cell interacting by control structure
W or rating curve (use zero for null)
c nqctu - id number of control table or rating curve for upstream to
W downstream flow (use zero for null)
c nqctd - id number of control table or rating curve for downstream
W upstream flow (use zero for null)
c nqts - id number of time series inflow or withdrawal (use zero for null)
c nets - id number of concentration inflow time series (use zero for null)
c nobs - id number of tidal open boundary condition (use zero for null)
c nsec - id number of cross-section property table for this cell
c rlen - length of cell
c xcrd - x coordinate of cell center (any consistent system,
W used only for graphics)
c ycrd - y coordinate of cell center (any consistent system,
W used only for graphics)
c angc - cell orientation angle (bearing from downstream cell
W boundary to upstream boundary)
c qint - initial discharge across downstream cell face
c hint - initial depth in cell
c bint - initial thalweg bed elevation of cell
c idum - dummy data can follow in columns after the last column
Figure 5 (continued). Example of the cellnet.inp file.
53
-------
c
c icell
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
26
27
28
29
30
31
32
33
34
35
36
iseg
1
1
1
1
1
1
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
itype
5
1
1
1
1
6
2
1
1
1
1
1
1
6
3
1
1
1
1
1
6
2
1
1
1
1
1
1
1
1
6
2
1
1
1
6
iupc
2
3
4
5
6
0
8
9
10
11
12
13
14
0
16
17
18
19
20
21
0
23
24
25
26
27
28
29
30
31
0
33
34
35
36
0
idnc
0
1
2
3
4
5
0
7
8
9
10
11
12
13
0
15
16
17
18
19
20
0
22
0
24
25
26
27
25
26
30
0
31
32
33
34
if ru
0
0
0
7
0
0
0
0
0
0
0
0
0
0
0
0
22
0
0
0
0
0
0
0
0
0
0
32
0
0
0
0
0
0
0
0
ifrd
0
0
0
0
0
0
4
0
0
0
0
0
0
9
0
0
0
0
0
0
0
17
0
0
0
0
0
0
0
0
0
28
0
0
0
0
iqcu
0
0
0
0
0
15
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
iqcd
0
0
0
0
0
0
0
0
0
0
0
0
0
0
6
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
nqctu
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
nqctd
0
0
0
0
0
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
0
0
0
0
Figure 5 (continued). Example of the cellnet.inp file.
54
-------
nqts nets
1 1
1 1
1 1
1 1
1 1
1 1
2 2
2 2
2 2
2 2
2 2
2 2
2 2
6 6
3 3
3 3
3 3
3 3
3 3
3 3
6 6
4 4
4 4
4 4
4 4
4 4
4 4
4 4
4 4
4 4
8 8
5 5
5 5
5 5
5 5
9 9
nobc
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
nsec
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
26
27
28
29
30
31
32
33
34
35
36
rlen
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
xcrd
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
ycrd
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
angc
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
qint
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
hint
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
bint
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
licell
! 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
! 26
! 27
! 28
! 29
! 30
! 31
! 32
! 33
! 34
! 35
! 36
Figure 5 (continued). Example of the cellnet.inp file. Note that the above lines actually
appear inline with those of the previous page in the actual input
file.
55
-------
9.4 Chansec.inp Input File
The chansec.inp file is used to specify channel cross-sectional area, wetted perimeter, and
surface width as a function of water depth (ISECDAT = 1) or water surface elevation
(ISECDAT = 0). If the table is defined for depth, the bottom elevation of the lowest
point in the cross-section must be specified. A separate table must be entered for each
cell. Properties are presumed to be defined at the center of the cell, corresponding to the
continuity control volume. A unity conversion factor is included for each section. A safe
practice is to extend the property table well beyond the raw data range, since a simulation
will stop if model computed water surface elevation or depth falls outside of the table
range. Figure 6 shows an example of the chansec.inp file that can be used as a template
for file construction.
9.5 Dser.inp, Sser.inp, and Tser.inp Input Files
Figure 7 shows an example of the dye concentration file dser.inp. The salinity and
temperature time series files follow identical formats. The example file in Figure 7 can
be used as a template for file construction. All time series for a given concentration
variable are loaded sequentially into this file. Each time series has either one or two
header lines. If ISTYP is equal to 1, a single concentration is entered and a second
header lines of weights are used. If ISTPY is equal to zero, the number of concentration
columns must equal to number of layers. For EFDC1D, the number of layers is by
default equal to one. Both adjustment and conversion factors can be specified for time
and concentration units.
56
-------
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
chansec
. inp file, in free format across line, repeats lc-2 times
EFDC 1-D CHANNEL CROSS-SECTION DATA FILE
NSEC
ISECDAT
NXYSDAT
BELVB
RMULADJ
EHXYS
AREA
WPER
SURF
NSEC
EHXYS (I
1
0.00
0.47
0. 95
1.42
2
0.00
0.47
0. 95
1.42
1.90
= section ID number
= 0 for elevation table, 1 for depth table
= number of elevation (or depth) vs area and wetted perimeter
data points for section
= channel bot elev corresponding to depth of zero (isecdat=l)
= multiplicative unit conversion, multiplies BELVB, EHXYS, WPER
AREA is converted by RMULADJ** 2
= elevation or depth relative to BELVB
= corresponding cross- sectional area
= corresponding wetted perimeter
= corresponding surface width
ISECDAT NXYSDAT BELVB RMULADJ
, J,NXY) AREA(I, J,NXY) WPER ( I , J, NXY) SURF ( I , J, NXY)
1 4 253.68 1.00 notes = 75 76 2
0.00 0.00 0.00
13.83 35.55 35.43
32.58 41.64 41.35
52.90 44.17 43.67
1 5 255.18 1.00 notes = 74 75 2
0.00 0.00 0.00
21.73 47.22 47.01
44.50 49.50 49.07
68.25 51.77 51.13
92.98 54.04 53.19
Figure 6. Example of the chansec.inp file with first two property tables.
57
-------
C dser.inp file, dye, in free format across line
C repeats ncser(3) times, test case
C
C ISTYP MCSER(NS,3) TCCSER(NS,3) TACSER(NS,3) RMULADJ(NS,3) ADDADJ(NS,3)
C
C if istyp.eq.l then read depth weights and single value of CSER
C
C (WKQ(K),K=1,KC)
C
C TCSER(M,NS,3) CSER(M,NS,3) !(mcser(ns,3) pairs for ns=3,ncser(3) series)
C
C else read a value of dser for each layer
C
C TCSER(M,NS,3) (CSER(M,K,NS,3),K=l,KG) !(mcser(ns,3) pairs)
C
1 7 86400.0 0.0 1.0 0.0 !01 tracer test
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
-1000. 0.0
0 . 0.0
121.99 0.0
122.01 1.0
122.99 1.0
123.01 0.0
1000. 0.0
0 7 86400.0 0.0 1.0 0.0 !02 tracer test
-1000. 0.0
0. 0.0
121.99 0.0
122.01 1.0
122.99 1.0
123.01 0.0
1000. 0.0
Figure 7. Example of the dser.inp, sser.inp, and tser.inp files. File shown with two
concentration time series.
58
-------
9.6 Efdcld.inp Input File
An example of the master input file, efdcJdinp is shown in Figure 8, and can be used as a
template for file construction. The file is organized into 34 data blocks. The definitions
of all variables in each data block are included in the block headers. The user cannot
insert any extract lines in the block headers, but can make notes and comments beyond
the end of the header and data lines. The remainder of this section highlights some
important features of selected data blocks.
Data block 2 controls restart and diagnostic options. For ISRESTI = 0 the model cold
starts using initial depths and flows from the cell.net files. The ISLOG option controls
the generation of a log file containing diagnostic information. The ISNEGH switch,
when activated, causes a model simulation to stop when a negative depth is computed,
since the one-dimensional solution procedure does not allow drying and wetting. If a
negative depth is encountered, information on its location will be printed in the efdc.log
output file, and the user should verified the channel section properties for that cell as well
as consider using a smaller time-step for the time integration. Data block 3 contains
additional solution options. The water surface elevation solution tolerance should be a
very small number below machine precision. If the solution does not achieve this
precision after ITERM iterations, the simulation will stop. The actual number of
iterations is recorded in the efdc.log file, and should be between 10 and 100. If the
maximum number of iterations is exceeded, the time-step should be reduced before
lowering the value of RSQM by a few orders of magnitude. For serious start up
problems, such as negative depth or NaN (not a number) output, the ISDSOLV option
can be activated. However, familiarity with the numerical solution scheme and
subroutine CALPUV1D is necessary to interpret the diagnostic output files. Potential
problems with control structures, such as control structure out of bounds messages, and
inflows can be investigated using the ISDIQ option. Use of this option requires
familiarity with subroutine CALQVS. Data block 4 is used to activate transport of
various concentration variables. It is recommended that the solution options ISADAC
and ISFCT remain 1. For thermal transport, ISTOPT on the temperature line should be
set to 1. Reading and writing concentration variables from and to the restart.inp file is
controlled by ISCI and ISCO.
Temporal run time control is specified in data block 5. The beginning time and its unit
conversion to seconds is specified by TEBEGIN and TCON. The duration of the
simulation is defined as
Simulation Duration = NTC * TREF
The example in Figure 8 sets a 152-day simulation time, since 152 time cycles, each
having a duration of 1 day (86400 seconds) is specified. The model time set is specified
by
Time Step = TREF/NTSPTC
59
-------
where NTSPTC is the number of time-steps per reference time cycle. For the example,
the time-step is 86400/1440 = 60 seconds. The value of NTSPTC should be chosen to
result in even whole number time-step sizes to avoid round off errors in time
accumulation in the model code. The EFDC1D model employs explicit momentum and
mass advection which limits the time-step by the Courant criteria
Al (9.1)
Q
where 6 is the time-step, and A, A, Q, and u are the cross-sectional area, flow direction
length, discharge, and cross-sectional average velocity in a cell, respectively, and a is a
safety factor less than unity. Experience has shown that the safety factor should be less
than 0.5, with 0.25 being a reasonable value for rapidly varying flow. Note that the time-
step is constrained by the minimum value of the right side of (9.1) over all cells. The
most common manifestation of instability due to using too large a time-step is the
occurrence of negative depths, divide by zeros, or NaN outputs. Generally when a
negative depth is detected in particular cell, the show.inp file should be set to display that
cell, with the user observing the velocity at which the instability occurs and then resetting
the time-step accordingly using (9.1).
Data block 6 allows adjustment or over riding of the Manning roughness values specified
in the cellnet.inp file. The convention is
Actual Roughness = ZBRADJ + ZBRCVT * (cellnet.inp Roughness)
Data block 7 specifies the use of various forcing files. The MTIDE variable defines the
number of tidal constituents to be specified in data block 8 if a tidal harmonic boundary
condition is used or an in place least squares harmonic analysis of water surface
elevations in selected cells is specified in data blocks 32 and 33. If MTIDE is set to 0, no
data are read from data block 8. If a tidal open boundary is to be specified by a water
surface elevation time series or the combination of a water surface elevation time series
and tidal harmonics, NPSER should be set to one. NPFOR should be set to 1 if harmonic
forcings are used. If harmonics are used, the mean sea level water surface elevation
should be specified by pser.inp if a zero mean sea level is not used. The parameters
NASER and NWSER specify the number of atmospheric and wind time series, which
should be either 0 or 1. Specifying 0 eliminates the requirement for the aser.inp and
wser.inp files and sets all variables normally entered in these files to zero. Both aser.inp
and wser.inp files are required for thermal simulation. Data block 8 specifies harmonic
water surface elevations amplitudes and phases for NPFOR equal to 1 and MTIDE
greater than zero. Harmonic constituent amplitudes should be in meters and phases
should be in seconds relative to the zero time origin of the reference year.
Data block 10 specifies the number of toxic state variables, and the number of classes of
cohesive and noncohesive sediment. The distribution executable is limited to one toxic,
one cohesive sediment class and two noncohesive sediment classes. The model can be
extended to more variables by modifying the efdc.par file following instructions in that
file and recompiling the code to create a new executable. The remaining number of time
60
-------
series for each concentration variable is not active. Instead, if a time series for inflows or
open boundaries is specified in the cellnet.inp file, each active concentration variable
must have a corresponding time series with the total number of time series for each
concentration variable specified in cellnet.inp.
Data blocks 11 through 18 define sediment transport process options, consistent with
descriptions provided in Chapters 6, 7 and 8 of this document. Model users are also
referred to subroutine SSEDTOX of the model for details of the computational
implementation of sediment transport and sorptive contaminant fate and transport.
Data block 22 provides additional information for constituent transport. The parameter
BSC controls density or buoyancy coupling of salinity, temperature and suspended
sediment. The parameter TEMO is an initial temperature and also the constant
equilibrium temperature if the simple equilibrium temperature option, ISTOPT(2) = 1, is
used. In this case, HEQT is the bulk heat transfer coefficient. If a tidal open boundary is
specified at the downstream end of segment 1 in the cellnet.inp file, NOBC should be set
to 1 and inflowing concentrations should be specified in data blocks 24 through 27. It is
noted that these specified values are used if no concentration time series is specified for
the open boundary cell in the cellnet.inp file. If a concentration time series is specified,
the values specified in data blocks 24 through 27 are additive with the time series values.
Note that surface values in data blocks 26 and 27 should be set to the corresponding
bottom values in data blocks 24 and 25. Only the first parameter in data block 23,
NTSCRS needs to be specified when NOBC is set to 1. This parameter specifies the
transition period as NTSCRS time-steps from the last outflow concentration to the
specified inflow concentration. A value of 0 immediately changes the concentration to
the inflow concentration at the start of incoming tidal flow.
Data block 28 controls writing to the primary output file, dump Id.out, of the EFDC1D
model. To activate output to this file, ISDUMP should be set to 3. The write frequency
is specified as the number of time-steps by NSDUMP, while the beginning and ending
times are specified by TSDUMP and TEDMP, which must be in days relative to the
reference year time origin. The last five parameters are not currently used. Data blocks
29 through 31 allow output of model results using formats originally developed for the
multi-dimensional EFDC model. Users familiar with EFDC may elect to use these
options. Data blocks 32 and 33 control in place least squares harmonic analysis of
selected state variables. This option is useful in analyzing model predictions in tidal
systems by users who are familiar with harmonic descriptions of tidal varying quantities
and least squares harmonic analysis. Data block 33 activates time series output at
locations specified in the cellnet.inp file. Starting and stopping time-step numbers and
the write frequency, in number of time-steps are specified by NBTMSR, NSTMSR, and
NWTMSR, respectively. Output file names follow the following conventions
dyetsNN.out - dye
q3dtsNN.out - net volume flow into cell
saltsNN.out - salinity
sedtsNN.out - cohesive sediment
seltsNN.out - water surface elevation and depth
sndtsNN.out - noncohesive sediment
61
-------
temtsNN.out - temperature
toxtsNN.out - toxics concentration
uvetsNN.out - velocity
uvttsNN.out - discharge
where NN denotes the output time series identifier assigned to particular cells in the
cellnet.inp file.
WELCOME TO THE ENVIRONMENTAL FLUID DYNAMICS COMPUTER CODE SERIES
DEVELOPED BY JOHN M. HAMRICK.
THIS IS THE MASTER INPUT FILE efdcld.inp, AND SHOULD BE USED WITH THE
10 JANUARY 2001 OR LATER VERSION OF efdcld.f
THIS FILE IS SELF DOCUMENTED WITH DEFINITIONS AND GUIDENCE FOR EACH
INPUT VARIABLE CONTAINED IN ITS CARD IMAGE SECTION. REFER TO USERS MAN
AVAILABLE FROM DEVELOPER AT ham@visi.net FOR ADDTIONAL DOCUMENTATION
Cl TITLE FOR RUN
C
TITLE OR IDENTIFIER FOR THIS INPUT FILE AND RUN
C
Cl (LIMIT TO 80 CHARACTERS LENGTH)
'brandywine creek, Christina river basin'
C2 RESTART, GENERAL CONTROL AND DIAGNOSTIC SWITCHES
C
ISRESTI: 1 FOR READING INITIAL CONDITIONS FROM FILE restart.inp
-1 AS ABOVE BUT ADJUST FOR CHANGING BOTTOM ELEVATION
2 INTIALIZES A KG LAYER RUN FROM A KC/2 LAYER RUN FOR KC.GE.4
10 FOR READING IC'S FROM restart.inp WRITTEN BEFORE 8 SEPT 92
ISRESTO:-! FOR WRITING RESTART FILE restart.out AT END OF RUN
N INTEGER.GE.0 FOR WRITING restart.out EVERY N REF TIME PERIODS
ISLOG: 1 FOR WRITING LOG FILE efdc.log
2 FOR WRITING LOG FILE efdc.log WITH OVER WRITE
EVERY REFERENCE TIME PERIOD
ISNEGH: 1 FOR SEARCHING FOR NEGATIVE DEPTHS AND WRITING TO SCREEN
2 FOR STOPPING SIMULATION WHEN NEGATIVE DEPTH OCCURS
ISHOW: 1 TO SHOW PUV&S ON SCREEN, SEE INSTRUCTIONS FOR FILE show.inp
C
C2 ISRESTI ISRESTO ISLOG ISNEGH ISHOW
01221
Figure 8. Example of the efdc.inp file.
62
-------
C3
C
C
C3
SOLUTION OPTION PARAMETERS AND SWITCHES
RSQM: TRAGET SQUARE RESIDUAL OF ITERATIVE SOLUTION SCHEME
ITERM: MAXIMUN NUMBER OF ITERARTIONS
ISDSOLV: 1 TO WRITE DIAGNOSTICS FILES FOR EXTERNAL MODE SOLVER
ISDIQ: 1 TO WRITE FLOW AND CONTROL DIAGNOSTIC FILE, diaq.out
ISSSMMT: 0 WRITES MEAN MASS TRANSPORT TO restran.out AFTER EACH
AVERAGING PERIOD (FOR RESEARCH PURPOSES)
2 DOES NOT WRITE
ISHDMF: 1 TO ACTIVE HORIZONTAL MOMENTUM AND MASS DIFFUSION
RSQM ITERM
l.E-20 300
ISDSOLV
0
ISDIQ
0
ISSSMMT
2
ISHDMF
0
C4
C4
C4
C
C
C4
DISSOLVED AND SUSPENDED CONSTITUENT TRANSPORT SWITCHES
SAL=1,TEM=2,DYE=3,TOX=5, (cohesive sediment) SED=6,
(noncohesive sediment) SND=7
ISTRAN: 1 OR GREATER TO ACTIVATE TRANSPORT
ISTOPT: NONZERO FOR TRANSPORT OPTIONS, SEE USERS MANUAL
ISCDCA: 0 FOR STANDARD DONOR CELL UPWIND DIFFERENCE ADVECTION
1 FOR CENTRAL DIFFERENCE ADVECTION FOR THREE TIME LEVEL STEPS
2 FOR EXPERIMENTAL UPWIND DIFFERENCE ADVECTION (FOR RESEARCH)
ISADAC: 1 TO ACTIVATE ANTI-NUMERICAL DIFFUSION CORRECTION TO
STANDARD DONOR CELL SCHEME
ISFCT: 1 TO ADD FLUX LIMITING TO ANTI-NUMERICAL DIFFUSION CORRECTION
ISPLIT: 1 TO OPERATOR SPLIT HORIZONTAL AND VERTICAL ADVECTION
(FOR RESEARCH PURPOSES)
ISADAH: 1 TO ACTIVATE ANTI-NUM DIFFUSION
SPLIT ADVECTION STANDARD DONOR
ISADAV: 1 TO ACTIVATE ANTI-NUM DIFFUSION
SPLIT ADVECTION STANDARD DONOR
ISCI: 1 TO READ CONCENTRATION FROM FILE restart.inp
ISCO: 1 TO WRITE CONCENTRATION TO FILE restart.out
CORRECTION TO HORIZONTAL
CELL SCHEME (FOR RESEARCH)
CORRECTION TO VERTICAL
CELL SCHEME (FOR RESEARCH)
ISTRAN
0
0
0
0
1
1
ISTOPT ISCDCA
ISADAC
1
1
1
1
1
1
ISFCT
1
1
1
1
1
1
ISPLIT
0
0
0
0
0
0
ISADAH
0
0
0
0
0
0
ISADAV
0
0
0
0
0
0
ISCI
0
0
0
0
0
0
ISCO
0
0
0
0
0
0
!sal
Item
!dye
!tox
! sed
! snd
Figure 8 (continued). Example of the efdc.inp file.
63
-------
C5 TIME-RELATED INTEGER PARAMETERS
C
NTC: NUMBER OF REFERENCE TIME PERIODS IN RUN
NTSPTC: NUMBER OF TIME-STEPS PER REFERENCE TIME PERIOD
NTCMMT: NUMBER OF NUMBER OF REF TIME TO AVERAGE OVER TO OBTAIN
RESIDUAL OR MEAN MASS TRANSPORT VARIABLES
TCON: CONVERSION MULTIPLIER TO CHANGE TBEGIN TO SECONDS
TBEGIN: TIME ORIGIN OF RUN
TREF: REFERENCE TIME PERIOD IN SEC (i.e. 44714.16s or 86400s)
C
C5
NTC NTSPTC NTSMMT TCON TBEGIN TREF
152 1440 720 86400. 121.0 86400.
C6 GRID, ROUGHNESS AND DEPTH PARAMETERS
C
ZBRADJ: LOG BDRY LAYER CONST OR VARIABLE ROUGH HEIGHT ADJ IN METERS
ZBRCVRT: LOG BDRY LAYER VARIABLE ROUGHNESS HEIGHT CONVERT TO METERS
C 0.0 1.00
C6 ZBRADJ ZBRCVT
0.0 1.00
C7
C
C
C7
TIDAL & ATMOSPHERIC FORCING
MTIDE: NUMBER OF PERIOD (TIDAL) FORCING CONSTITUENTS
NPSER: NUMBER OF TIME SERIES FORCINGS
NPFOR: NUMBER OF HARMONIC FORCINGS
NWSER: NUMBER OF WIND TIME SERIES (0 SETS WIND TO ZERO)
NASER: NUMBER OF ATMOSPHERIC CONDITION TIME SERIES (0 SETS ALL
ZERO)
MTIDE
0
NPSER NPFOR
0 0
NWSER
1
NASER
1
C8 PERIODIC FORCING (TIDAL) CONSTITUENT SYMBOLS AND PERIODS
C
SYMBOL: FORCING SYMBOL (CHARACTER VARIABLE) FOR TIDES, THE NOS SYMBOL
PERIOD: FORCING PERIOD IN SECONDS
C8 SYMBOL
PERIOD
Figure 8 (continued). Example of the efdc.inp file.
64
-------
C9
C
C
C9
PERIODIC FORCING (TIDAL) SURF ELEV OR PRESSURE BOUNDARY COND. FORCINGS
NPFOR: FORCING NUMBER
SYMBOL: FORCING SYMBOL (FOR REFERENCE HERE ONLY)
AMPLITUDE: AMPLITUDE IN M (PRESSURE DIVIDED BY RHO*G)
PHASE: FORCING PHASE RELATIVE TO TBEGIN IN SECONDS
NPFOR
SYMBOL
AMPLITUDE
PHASE
CIO
C
C
CIO
SPECIFY NUM OF SEDIMENT AMD TOXICS AND NUM OF CONCENTRATION TIME SERIES
NTOX: NUMBER OF TOXIC CONTAMINANTS (DEFAULT = 1)
NSED: NUMBER OF COHESIVE SEDIMENT SIZE CLASSES (DEFAULT = 1)
NSND: NUMBER OF NON-COHESIVE SEDIMENT SIZE CLASSES (DEFAULT = 1)
NSSER: NUMBER OF SALINITY TIME SERIES
NTSER: NUMBER OF TEMPERATURE TIME SERIES
NDSER: NUMBER OF DYE CONCENTRATION TIME SERIES
NTXSER: NUMBER OF TOXIC CONTAMINANT CONCENTRATION TIME SERIES
EACH TIME SERIES MUST HAVE DATA FOR NTOX TOXICICANTS
NSDSER: NUMBER OF COHESIVE SEDIMENT CONCENTRATION TIME SERIES
EACH TIME SERIES MUST HAVE DATA FOR NSED COHESIVE SEDIMENTS
NSNSER: NUMBER OF NONCOHESIVE SEDIMENT CONCENTRATION TIME SERIES
EACH TIME SERIES MUST HAVE DATA FOR NSND NONCOHESIVE SEDIMENTS
NTOX
1
NSED
1
NSND
2
NSSER
0
NTSER NDSER
0 0
NTXSER NSDSER NSNSER
000
Figure 8 (continued). Example of the efdc.inp file.
65
-------
Cll SEDIMENT INITIALIZATION AND WATER COLUMN/BED REPRESENTATION OPTIONS
DATA REQUIRED EVEN IF ISTRAN(6) AND ISTRAN(7) ARE 0
ISEDINT: 0 FOR CONSTANT INITIAL CONDITIONS
1 FOR SPATIALLY VARIABLE WATER COLUMN INITIAL CONDITIONS
FROM sedw.inp AND sndw.inp
2 FOR SPATIALLY VARIABLE BED INITIAL CONDITIONS
FROM sedb.inp AND sndb.inp
3 FOR SPATIALLY VARIABLE WATER COL AND BED INITIAL CONDITIONS
ISEDBINT: 0 FOR SPATIALLY VARYING BED INITIAL CONDITIONS IN MASS/AREA
1 FOR SPATIALLY VARYING BED INITIAL CONDITIONS IN MASS FRACTION
OF TOTAL SEDIMENT MASS (REQUIRES BED LAYER THICKNESS
FILE bedlay.inp)
ISEDWC: 0 COHESIVE SED WC/BED EXCHANGE BASED ON BOTTOM LAYER CONDITIONS
1 COHESIVE SED WC/BED EXCHANGE BASED ON WAVE/CURRENT/SEDIMENT
BOUNDARY LAYERS EMBEDDED IN BOTTOM LAYER
ISMUD: 1 INCLUDE COHESIVE FLUID MUD VISCOUS EFFECTS USING EFDC
FUNCTION CSEDVIS(SEDT)
ISNDWC: 0 NONCOH SED WC/BED EXCHANGE BASED ON BOTTOM LAYER CONDITIONS
1 NONCOH SED WC/BED EXCHANGE BASED ON WAVE/CURRENT/SEDIMENT
BOUNDARY LAYERS EMBEDDED IN BOTTOM LAYER
ISEDVW: 0 FOR CONSTANT OR SIMPLE CONCENTRATION DEPENDENT
COHESIVE SEDIMENT SETTLING VELOCITY
>1 CONCENTRATION AND/OR SHEAR/TURBULENCE DEPENDENT COHESIVE
SEDIMENT SETTLING VELOCITY. VALUE INDICATES OPTION TO BE USED
IN EFDC FUNCTION CSEDSET(SED,SHEAR,ISEDVWC)
1 HUANG AND METHA - LAKE OKEECHOBEE
2 SHRESTA AND ORLOB - FOR KRONES SAN FRANCISCO BAY DATA
3 ZIEGLER AND NESBIT - FRESH WATER
ISNDVW: 0 USE CONSTANT SPECIFIED NONCHOESIVE SED SETTLING VELOCITIES
OR CALCULATE FOR CLASS DIAMETER IS SPECIFIED VALUE IS NEG
>1 FOLLOW OPTION 0 PROCEDURE BUT APPLY HINDERED SETTLING
CORRECTION. VALUE INDICATES OPTION TO BE USED WITH EFDC
FUNCTION CSNDSET(SND,SDEN,ISNDVW) VALUE OF ISNDVW INDICATES
EXPONENTIAL IN CORRECT (1-SDEN(NS)*SND(NS)**ISNDVW
KB: MAXIMUM NUMBER OF BED LAYERS (EXCLUDING ACTIVE LAYER)
ISEDAL: 1 TO ACTIVATE STATIONARY COHESIVE MUD ACTIVE LAYER
ISNDAL: 1 TO ACTIVATE NONCOHESIVE ARMORING LAYER ACTIVE LAYER
C
Cll
ISEDINT
0
ISEDBINT
0
ISEDWC
0
ISMUD
0
ISNDWC
0
ISEDVW
0
ISNDVW
0
KB
1
ISEDAL
0
ISNDAL
1
Figure 8 (continued). Example of the efdc.inp file.
66
-------
C12 BED MECHANICAL PROPERTIES PARAMETER SET 1
DATA REQUIRED EVEN IF ISTRAN(6) AND ISTRAN(7) ARE 0
IBMECH:
TIME INVARIANT CONSTANT BED MECHANICAL PROPERITES
1 SIMPLE CONSOLIDATION CALCULATION WITH CONSTANT COEFFICIENTS
2 SIMPLE CONSOLIDATION WITH VARIABLE COEFFICIENTS DETERMINED
EFDC FUNCTIONS CSEDCON1,2,3(IBMECH)
3 COMPLEX CONSOLIDATION WITH VARIABLE COEFFICIENTS DETERMINED
EFDC FUNCTIONS CSEDCON1,2,3(IBMECH). IBMECH > 0 SETS THE
C38 PARAMETER ISEDBINT=1 AND REQUIRES INITIAL CONDITIONS
FILES bedlay.inp, bedbdn.inp and bedddn.in
IMORPH: 0 CONSTANT BED MORPHOLOGY (IBMECH=0, ONLY)
HBEDMAX
BEDPORC
SEDMDMX
SEDMDMN
SEDVDRD
SEDVDRM
SEDVDRT
1 ACTIVE BED MORPHOLOGY: NO WATER ENTRAIN/EXPULSION EFFECTS
2 ACTIVE BED MORPHOLOGY: WITH WATER ENTRAIN/EXPULSION EFFECTS
TOP BED LAYER THICKNESS (M) AT WHICH NEW LAYER IS ADDED OR IF
KBT(I,J)=KB, NEW LAYER ADDED AND LOWEST TWO LAYERS COMBINED
CONSTANT BED POROSITY (IBMECH=0, OR NSED=0)
ALSO USED AS POROSITY OF DEPOSITIN NONCOHESIVE SEDIMENT
MAXIMUM FLUID MUD COHESIVE SEDIMENT CONCENTRATION (mg/1)
MINIMUM FLUID MUD COHESIVE SEDIMENT CONCENTRATION (mg/1)
VOID RATIO OF DEPOSITING COHESIVE SEDIMENT
MINIMUM COHESIVE SEDIMENT BED VOID RATIO (IBMECH > 0)
BED CONSOLODATION RATED CONSTANT (I/SEC) (IBMECH = 1,2)
C
C12
IBMECH IMORPH HBEDMAX BEDPORC SEDMDMX SEDMDMN SEDVDRD SEDVDRM SEDVRDT
0 0 2.0 0.5 1.1E5 4000. 20. 4. l.E-5
C13 BED MECHANICAL PROPERTIES PARAMETER SET 2
C DATA REQUIRED EVEN IF ISTRAN(6) AND ISTRAN(7) ARE
C
BMECH1 BED MECHANICS FUNCTION COEFFICIENT
BMECH2 BED MECHANICS FUNCTION COEFFICIENT
BMECH3 BED MECHANICS FUNCTION COEFFICIENT
BMECH4 BED MECHANICS FUNCTION COEFFICIENT
BMECH5 BED MECHANICS FUNCTION COEFFICIENT
BMECH6 BED MECHANICS FUNCTION COEFFICIENT
C
C13
BMECH1 BMECH2
0. 0
0 .
BMECH3
0.
BMECH4
0 .
BMECH5 BMECH6
0 . 0 .
Figure 8 (continued). Example of the efdc.inp file.
67
-------
C14 COHESIVE SEDIMENT PARAMETER SET 1 REPEAT DATA LINE NSED TIMES
DATA REQUIRED EVEN IF ISTRAN(6) AND ISTRAN(7) ARE 0
SEDO:
SEDBO:
SDEN:
SSG:
WSEDO:
SEDSN:
SEXP:
TAUD:
ISEDSCOR:
C
C14 SEDO
0.0
PER UNIT AREA
SSG=2.5 AND
CONSTANT INITIAL COHESIVE SEDIMENT CONG IN WATER COLUMN
(mg/liter=gm/m**3)
CONSTANT INITIAL COHESIVE SEDIMENT IN BED
(gm/sq meter) IE 1CM THICKNESS BED WITH
N=.6,.5 GIVES SEDBO 1.E4, 1.25E4
SEDIMENT SPEC VOLUME (IE 1/2.25E6 m**3/gm)
SEDIMENT SPECIFIC GRAVITY
CONSTANT OR REFERENCE SEDIMENT SETTLING VELOCITY
IN FORMULA WSED=WSEDO*( (SED/SEDSN)**SEXP )
NORMALIZING SEDIMENT CONG (COHESIVE SED TRANSPORT)
EXPONENTIAL (COHESIVE SED TRANSPORT)
BOUNDARY STRESS BELOW WHICH DEPOSITION TAKES PLACE ACCORDING
TO (TAUD - TAU) / TAUD
TO CORRECT BOTTOM LAYER CONCENTRATION TO NEAR BED CONG
(gm/m**3)
SEDBO
1.0E5
SDEN
0.4E-6
SSG
2 .5
WSEDO
l.OE-4
SEDSN
1.0
SEXP
0.
TAUD
2.0E-4
ISEDSCOR
0
C15 COHESIVE SEDIMENT PARAMETER SET 2 REPEAT DATA LINE NSED TIMES
C DATA REQUIRED EVEN IT ISTRAN(6) AND ISTRAN(7) ARE 0
C
IWRSP: 0 USE RESUSPENSION RATE AND CRITICAL STRESS BASED ON PARAMETERS
ON THIS DATA LINE
>1 USE BED PROPERTIES DEPENDEDNT RESUSPENSION RATE AND CRITICAL
STRESS GIVEN BY EFDC FUNCTIONS CSEDRESS,CSEDTAUS,CSEDTAUB
FUNCTION ARGUMENSTS ARE (BDENBED,IWRSP)
1 HWANG AND METHA - LAKE OKEECHOBEE
WRSPO: REF SURFACE EROSION RATE IN FORMULA
WRSP=WRSPO*( ((TAU-TAUR)/TAUN)**TEX ) (gm/m**2-sec)
TAUR: BOUNDARY STRESS ABOVE WHICH SURFACE EROSION OCCURS (m/s)**2
TAUN: NORMALIZING STRESS (EQUAL TO TAUR FOR COHESIVE SED TRANS)
TEXP: EXPONENTIAL (COH SED)
C 0 0.005 4.E-4 4.E-4 1.
C15 IWRSP WRSPO TAUR TAUN TEXP
0 0.005 4.E-4 4.E-4 1.
Figure 8 (continued). Example of the efdc.inp file.
68
-------
C16 NONCOHESIVE SEDIMENT PARAMETER SET 1 REPEAT DATA LINE NSND TIMES
C DATA REQUIRED EVEN IT ISTRAN(6) AND ISTRAN(7) ARE 0
C
SNDO: CONSTANT INITIAL NONCOHESIVE SEDIMENT CONG IN WATER COLUMN
(mg/liter=gm/m**3)
SNDBO: CONSTANT INITIAL NONCOHESIVE SEDIMENT IN BED PER UNIT AREA
(gm/sq meter) IE 1CM THICKNESS BED WITH SSG=2.5 AND
N=.6,.5 GIVES SNDBO 1.E4, 1.25E4
SDEN: SEDIMENT SPEC VOLUME (IE 1/2.65E6 m**3/gm)
SSG: SEDIMENT SPECIFIC GRAVITY
SNDDIA: REPRESENTATIVE DIAMETER OF SEDIMENT CLASS
WSNDO: CONSTANT OR REFERENCE SEDIMENT SETTLING VELOCITY
IF WSNDO < 0, SETTLING VELOCITY INTERNALLY COMPUTED
SNDN: MAX MASS/TOT VOLUME IN BED (NONCOHESIVE SED TRANS) (gm/m**3)
SEXP: DIMENSIONLESS RESUSPENSION PARAMETER GAMMA ZERO
TAUD: DUNE BREAK POINT STRESS (m/s)**2
ISNDSCOR: 1 TO CORRECT BOTTOM LAYER CONCENTRATION TO NEAR BED CONG
C 3.77E-5 2.65
C16 SNDO SNDBO SDEN SSG SNDDIA WSNDO SNDN SEXP TAUD ISNDSCOR
0.0 1.E5 3.77E-5 2.65 8.0E-5 -1.00 1.E6 l.E-3 7.E-5 0
0.0 1.E5 3.77E-5 2.65 3.0E-4 -1.00 1.E6 l.E-3 7.E-5 0
C17 NONCOHESIVE SEDIMENT PARAMETER SET 2 REPEAT DATA LINE NSND TIMES
C DATA REQUIRED EVEN IT ISTRAN(6) AND ISTRAN(7) ARE 0
C
ISNDEQ: >1 CALCULATE ABOVE BED REFERENCE NONCHOHESIVE SEDIMENT
EQUILIBRIUM CONCENTRATION USING EFDC FUNCTION
CSNDEQC(SNDDIA,SSG,WS,TAUR,TAUB,SIGPHI,SNDDMX,IOTP)
WHICH IMPLEMENT FORMULATIONS OF
1 GRACIA AND PARKER
2 SMITH AND MCLEAN
3 VAN RUN
ISBDLD: 0 BED LOAD PHI FUNCTION IS CONSTANT, SBDLDP
1 VAN RIJN PHI FUNCTION
2 OHTER PHI FUNCTION
3 OHTER PHI FUNCTION
ISBDLD: 0 BED LOAD PHI FUNCTION IS CONSTANT
TAUR: CRITICAL SHIELDS STRESS IN (m/s)**2 (ISNDEQ=2)
NOTE: IF TAUR < 0, THEN TAUR, TAUN, AND TEXP ARE INTERNALLY
COMPUTED USING VAN RIJN'S FORMULAS
TAUN: EQUAL TO TAUR FOR NONCHOESIVE SED TRANS (ISNDEQ=2)
TEXP: CRITICAL SHIELDS PARAMETER (ISNDEQ=2)
C
C17
ISNDEQ
3
3
ISBDLD
0
0
TAUR
-1 .E + 6
-1 .E + 6
TAUN
-l.E+6
-l.E+6
TEXP
-0 .2
-0 .2
Figure 8 (continued). Example of the efdc.inp file.
69
-------
CIS
c
c
c
CIS
NONCOHESIVE SEDIMENT PARAMETER SET 3 (BED LOAD FORMULA
PARAMETERS)
DATA REQUIRED EVEN IT ISTRAN(6) AND ISTRAN(7) ARE 0
SBDLDA:
SBDLDB :
SBDLDG:
SBDLDP:
SBDLDA
2 .1
ALPHA EXPONENTIAL FOR BED LOAD FORMULA
BETA EXPONENTIAL FOR BED LOAD FORMULA
GAMMA CONSTANT FOR BED LOAD FORMULA
CONSTANT PHI FOR BED LOAD FORMULA
SBDLDB SBDLDG SBDLDP
0. 0 . 0.
C19 TOXIC CONTAMINANT INITIAL CONDITIONS AND PARAMETERS
C USER MAY CHANGE UNITS OF WATER AND SED PHASE TOX CONCENTRATION
C AND PARTIATION COEFFICIENT ON C44 - C46 BUT CONSISTENT UNITS MUST
C MUST BE USED FOR MEANINGFUL RESULTS
C DATA REQUIRED EVEN IT ISTRAN(5) IS 0
C
NTOXN: TOXIC CONTAMINANT NUMBER ID (1 LINE OF DATA BY DEFAULT)
ITXINT: 0 FOR SPATIALLY CONSTANT WATER COL AND BED INITIAL CONDITIONS
1 FOR SPATIALLY VARIABLE WATER COLUMN INITIAL CONDITIONS
2 FOR SPATIALLY VARIABLE BED INITIAL CONDITIONS
3 FOR SPATIALLY VARIABLE WATER COL AND BED INITIAL CONDITION
ITXBDUT: SET TO 0 FOR CONST INITIAL BED GIVEN BY TOTAL TOX (ugm/litr)
SET TO 1 FOR CONST INITIAL BED GIVEN BY
SORBED MASS TOX/MASS SED(mg/kg)
TOXINTW: INIT WATER COLUNM TOT TOXIC VARIABLE CONCENTRATION (ugm/litr)
TOXINTB: INIT SED BED TOXIC CONG SEE ITXBDUT
RKTOXW: FIRST ORDER WATER COL DECAY RATE FOR TOX VARIABLE IN I/SEC
TKTOXW: REF TEMP FOR 1ST ORDER WATER COL DECAY DEG C
RKTOXB: FIRST ORDER SED BED DECAY RATE FOR TOX VARIABLE IN I/SEC
TKTOXB: REF TEMP FOR 1ST ORDER SED BED DECAY DEG C
C ck blw kevin uses 6.0
C19 NTOXN ITXINT ITXBDUT TOXINTW TOXINTB RKTOXW TKTOXW RKTOXB TRTOXB COMMENTS
1 0 0 1 . 1. 0. 0. 0. 0. DUMMY
Figure 8 (continued). Example of the efdc.inp file.
70
-------
C20
C
C
ADDITIONAL TOXIC CONTAMINANT PARAMETERS
DATA REQUIRED EVEN IT ISTRAN(5) IS 0
NTOXN: TOXIC CONTAMINANT NUMBER ID (1 LINE OF DATA BY DEFAULT)
VOLTOX: WATER SURFACE VOLITIALIZATION RATE MULTIPLIER (0. OR 1.)
RMOLTX: MOLECULAR WEIGHT FOR DETERMINING VOLATILIZATION RATE
RKTOXP: REFERENCE PHOTOLOYSIS DECAY RATE I/SEC
SKTOXP: REFERENCE SOLAR RADIATION FOR PHOTOLOYSIS (WATTS/M**2)
DIFTOX: DIFFUSION COEFF FOR TOXICANT IN SED BED PORE WATER (M**2/S)
C
C20
NTOXN VOLTOX
1 0.
RMOLTX
0.
RKTOXP SKTOXP DIFTOX
0. 0. l.E-9
COMMENTS
DUMMY
C21 TOXIC CONTAMINANT SEDIMENT INTERACTION PARAMETERS
C 2 LINES OF DATA REQUIRED EVEN IT ISTRAN(5) IS 0
C
NTOXC: TOXIC CONTAMINANT NUMBER ID. NSEDC+NSEDN LINES OF DATA
FOR EACH TOXIC CONTAMINANT (DEFAULT = 2)
NSEDN/NSNDN: FIRST NSED LINES COHESIVE, NEXT NSND LINES NON-COHESIVE.
REPEATED FOR EACH CONTAMINANT
ITXPARW: EQUAL 1 FOR SOLIDS DEPENDENT PARTITIONING (WC) GIVEN BY
TOXPAR=PARO*(CSED**CONPAR)
TOXPARW: WATER COLUMN PARO (ITXPARW=1) OR EQUIL TOX CON PART COEFF BETWEEN
EACH TOXIC IN WATER AND ASSOCIATED SEDIMENT PHASES (liters/mg)
CONPARW: EXPONENT IN TOXPAR=PARO*(CSED**CONPARW) IF ITXPARW=1
ITXPARB: EQUAL 1 FOR SOLIDS DEPENDENT PARTITIONING (BED)
TOXPARB: SEDIMENT BED PARO (ITXPARB=1) OR EQUIL TOX CON PART COEFF BETWEEN
EACH TOXIC IN WATER AND ASSOCIATED SEDIMENT PHASES (liters/mg)
CONPARB: EXPONENT IN TOXPAR=PARO*(CSED**CONPARB) IF ITXPARB=1
0.8770 -0.943
0.025
C21 NTOXN NSEDN ITXPARW TOXPARW CONPARW ITXPARB TOXPARB CONPARB
1.
1.
1.
0.
0.
0.
1.
1.
1.
0 .
0.
0.
COMMENTS
FOR NSED=1
FOR NSND=1
FOR NSND=2
Figure 8 (continued). Example of the efdc.inp file.
71
-------
C22 BUOYANCY, TEMPERATURE, DYE DATA AND CONCENTRATION BC DATA
C
BSC: BUOYANCY INFLUENCE COEFFICIENT 0 TO 1, BSC=1. FOR REAL PHYSICS
TEMO: REFERENCE, INITIAL, EQUILIBRUM AND/OR ISOTHERMAL TEMP IN DEG C
HEQT: EQUILIBRUM TEMPERTURE TRANSFER COEFFICIENT M/SEC
RKDYE: FIRST ORDER DECAY RATE FOR DYE VARIABLE IN I/SEC
NCOBC: 1 TO ACTIVATE CONCENTRATION BC ON TIDAL OPEN BOUNDARIES
C
C22 BSC TEMO HEQT RKDYE NCOBC
1.0 20.0 l.E-6 0. 0
C23 LOCATION OF CONG EC'S ON TIDAL OPEN BOUNDARY
C
NTSCRS: NUMBER OF TIME-STEPS TO RECOVER SPECIFIED VALUES ON CHANGE
TO INFLOW FROM OUTFLOW
NSSERS: BOUNDARY CELL SALINITY TIME SERIES ID NUMBER
NTSERS: BOUNDARY CELL TEMPERATURE TIME SERIES ID NUMBER
NDSERS: BOUNDARY CELL DYE CONG TIME SERIES ID NUMBER
NTXSERS: BOUNDARY CELL TOXIC CONTAMINANT CONG TIME SERIES ID NUM.
NSDSERS: BOUNDARY CELL COHESIVE SED CONG TIME SERIES ID NUMBER
NSNSERS: BOUNDARY CELL NONCOHESIVE SED CONG TIME SERIES ID NUMBER
C
C23 NTSCRS NSSERS NTSERS NDSERS NTXSERS NSDSERS NSNSERS
C24 TIME CONSTANT BOTTOM CONG ON OPEN BOUNDARY
C
SAL: ULTIMATE INFLOWING BOTTOM LAYER SALINITY
TEM: ULTIMATE INFLOWING BOTTOM LAYER TEMPERATURE
DYE: ULTIMATE INFLOWING BOTTOM LAYER DYE CONCENTRATION
TOX: NTOX ULTIMATE INFLOWING BOTTOM LAYER TOXIC CONTAMINANT
CONCENTRATIONS NTOX VALUES TOX(N), N=1,NTOX
C
C24 SAL TEM DYE TOX1-TOXN
C25 TIME CONSTANT BOTTOM CONG ON OPEN BOUNDARY
C
SED: NSED ULTIMATE INFLOWING BOTTOM LAYER COHESIVE SEDIMENT
CONCENTRAIONS FIRST NSED VALUES SED(N), N=1,NSND
SND: NSND ULTIMATE INFLOWING BOTTOM LAYER NONCOHESIVE SEDIMENT
CONCENTRATIONS LAST NSND VALUES SND(N), N=1,NSND
C
C25 SED1 SND1 SND2
Figure 8 (continued). Example of the efdc.inp file.
72
-------
C26 TIME CONSTANT SURFACE CONG ON OPEN BOUNDARY
C
SAL: ULTIMATE INFLOWING SURFAC LAYER SALINITY
TEM: ULTIMATE INFLOWING SURFAC LAYER TEMPERATURE
DYE: ULTIMATE INFLOWING SURFAC LAYER DYE CONCENTRATION
TOX: NTOX ULTIMATE INFLOWING SURFAC LAYER TOXIC CONTAMINANT
CONCENTRATIONS NTOX VALUES TOX(N), N=1,NTOX
C
C26 SAL TEM DYE TOX1-TOXN
C27 TIME CONSTANT SURFACE CONG ON OPEN BOUNDARY
C
SED: NSED ULTIMATE INFLOWING SURFAC LAYER COHESIVE SEDIMENT
CONCENTRAIONS FIRST NSED VALUES SED(N), N=1,NSND
SND: NSND ULTIMATE INFLOWING SURFAC LAYER NONCOHESIVE SEDIMENT
CONCENTRATIONS LAST NSND VALUES SND(N), N=1,NSND
C
C27 SED1 SND1 SND2
C28 CONTROLS FOR WRITING ASCII OR BINARY DUMP FILES
C
ISDUMP: GREATER THAN 0 TO ACTIVATE
1 SCALED ASCII INTERGER (0
-------
C29 CONTROLS FOR HORIZONTAL PLANE SCALAR FIELD CONTOURING
C
ISSPH: 1 TO WRITE FILE FOR SCALAR FIELD CONTOURING IN HORIZONTAL PLANE
NPSPH: NUMBER OF WRITES PER REFERENCE TIME PERIOD
ISRSPH: 1 TO WRITE FILE FOR RESIDUAL SALINITY PLOTTING IN
HORIZONTAL
ISPHXY: 0 DOES NOT WRITE I,J,X,Y IN ***cnh.out and r***cnh.out FILES
1 WRITES I,J ONLY IN ***cnh.out and r***cnh.out FILES
2 WRITES I,J,X,Y IN ***cnh.out and r***cnh.out FILES
C
DATA LINE REPEATS 7 TIMES FOR SAL,TEM,DYE,SFL,TOX,SED,SND
C29
ISSPH
0
0
0
0
0
0
0
NPSPH
1
1
1
1
1
1
1
ISRSPH
0
0
0
0
0
0
0
ISPHXY
1
1
1
1
1
1
1
!SAL
ITEM
!DYE
!SFL
!TOX
!SED
!SND
C30 CONTROLS FOR HORIZONTAL SURFACE ELEVATION OR PRESSURE CONTOURING
C
ISPPH: 1 TO WRITE FILE FOR SURFACE ELEVATION OR PRESSURE CONTOURING
IN HORIZONTAL PLANE
NPPPH: NUMBER OF WRITES PER REFERENCE TIME PERIOD
ISRPPH: 1 TO WRITE FILE FOR RESIDUAL SURFACE ELEVATION CONTOURNG IN
HORIZONTAL PLANE
C
C30
ISPPH
1
NPPPH
1
ISRPPH
0
C31 CONTROLS FOR HORIZONTAL PLANE VELOCITY VECTOR PLOTTING
C
ISVPH: 1 TO WRITE FILE FOR VELOCITY PLOTTING IN HORIZONTAL PLANE
NPVPH: NUMBER OF WRITES PER REFERENCE TIME PERIOD
ISRVPH: 1 TO WRITE FILE FOR RESIDUAL VELOCITY PLOTTIN IN
HORIZONTAL PLANE
C31 ISVPH
0
NPVPH
1
ISRVPH
0
Figure 8 (continued). Example of the efdc.inp file.
74
-------
C32
C
C
C32
INPLACE HARMONIC ANALYSIS PARAMETERS
ISLSHA: 1 FOR IN PLACE LEAST SQUARES HARMONIC ANALYSIS
MLLSHA: NUMBER OF LOCATIONS FOR LSHA
NTCLSHA: LENGTH OF LSHA IN INTEGER NUMBER OF REFERENCE TIME PERIODS
ISLSTR: 1 FOR TREND REMOVAL
ISHTA : 1 FOR SINGLE TREF PERIOD SURFACE ELEV ANALYSIS
90
ISLSHA MLLSHA NTCLSHA ISLSTR ISHTA
00 29 0 0
C33 HARMONIC ANALYSIS LOCATIONS AND SWITCHES
LLLSHA: CELL INDEX
LSHAP: 1 FOR ANALYSIS OF SURFACE ELEVATION
LSHAB: 1 FOR ANALYSIS OF SALINITY
LSHAUE: 1 FOR ANALYSIS OF EXTERNAL MODE HORIZONTAL VELOCITY
LSHAU: 1 FOR ANALYSIS OF HORIZONTAL VELOCITY IN EVERY LAYER
CLSL:
LOCATION AS A CHARACTER VARIALBLE
C33 LLLSHA
LSHAP LSHAB LSHAUE LSHAU CLSL
C34
C
C
C34
CONTROLS FOR WRITING TO TIME SERIES FILES
ISTMSR: 1 OR 2 TO WRITE TIME SERIES OF SURF ELEV, VELOCITY, NET
INTERNAL AND EXTERNAL MODE VOLUME SOURCE-SINKS, AND
CONCENTRATION VARIABLES, 2 APPENDS EXISTING TIME SERIES FILES
NBTMSR: TIME-STEP TO BEGIN WRITING TO TIME SERIES FILES
NSTMSR: TIME-STEP TO STOP WRITING TO TIME SERIES FILES
NWTSER: WRITE INTERVAL FOR WRITING TO TIME SERIES FILES
TCTMSR: UNIT CONVERSION FOR TIME SERIES TIME. FOR SECONDS, MINUTES,
HOURS,DAYS USE 1.0, 60.0, 3600.0, 86400.0 RESPECTIVELY
ISTMSR
1
NBTMSR
1
NSTMSR
2000000
NWTMSR
180
TCTMSR
86400.
Figure 8 (continued). Example of the efdc.inp file.
75
-------
9.7 Pser.inp Input File
The water surface elevation open boundary condition file, pser.inp, is shown in Figure 9.
The file in this figure can be used as a template for file construction.
9.8 Qctl.inp Input File
The input file qctl.inp specifies data to implement flow between pairs of cells controlled
by hydraulic structures or rating curves. The flow is unidirectional between an upstream
and downstream cell. Bi-directional flow is implemented by a control structure for each
direction. An example of the file, which can contain data sequences for an arbitrary
number of structures is shown in Figure 10, which can also be used as a template for file
construction. The header line contains a number of adjustment and multiplier factors for
the upstream and downstream water surface elevations and the flow value. The data table
has a head difference in the first column and one or more flow data columns. The flows
in the flow data column can be adjusted both by addition and multiplication. For
example, if the flow in the flow data column represents flow per unit width over a
spillway in CFS, RMULADJ would include both the width of the spillway and the unit
conversion from CFS to the required units of CMS. When the flow is calculated in the
EFDC1D model in subroutine CALQVS, the value of the head difference is calculated
by:
hdifctl = hctlum ® (selu + hctlua) - hctldm ® (seld + hctldd) (9.2)
where selu and seld are the water surface elevations upstream and downstream of the
control structure. For example, a rating curve or flow over a spillway has only upstream
control and hctldm would be set to zero. For a culvert, both hctlum and hctldm would be
set to one or a unit conversion factor if required. For a rating curve in terms of the flow
depth, the hctlua adjustment would be set to the negative of the channel bottom elevation.
Likewise for a spillway rating curve based on the upstream head above the spillway crest,
hctlua would be set to the negative of the spillway crest elevation.
76
-------
C pser.inp file, in free format across line, repeats npser times
C
C MPSER(NS) TCPSER(NS) TAPSER(NS) RMULADJ(NS) ADDADJ(NS)
C
C TPSER(M,NS) PSER(M,NS) !(mpser(ns) pairs for ns=l,npser series)
C
5040 86400. 0. 1.0 0.0
59.042 0.157
60.042 0.157
60.083 0.163
60.125 0.161
60.167 0.165
60.208 0.169
Figure 9. Example of the pser.inp file.
qctl.inp file, repeats nqctl times, for Christina River
ISTYP MQCTL HCTLUA HCTLUM HCTLDA HCTLDM RMULADJ ADDADJ ACQCTL
hctlua is negative of dam crest elevation in meters
if istyp.eq.l then read depth weights and single value of QCTL
(WKQ(K),K=1,KC)
HDIFCTL(M,NS) QCTL(M,1,NS) !(mqctl(ns) pairs for ns=l,nqser series)
else read a value of qser for each layer
HDIFCTL(M,NS) (QCTL(M,K,NS),K=l,KG) !(mqctl(ns) pairs)
0 7 -3.5 1.0
0.0000 0.0000
0.0800 0.0386
0.1600 0.1091
0.2400 0.2005
0.3200 0.3086
0.4000 0.4313
12.0000 70.8712
0 5 -50.6 1.0
0.0000 0.0000
0.0800 0.0386
0.1600 0.1091
0.2400 0.2005
0.3200 0.3086
0.0 0.0 25.0
0.00 0.0
1 Brandywine Park Dam
0.0 0.0 16.0 0.00 0.0
2 Brandywine-E.Br. (Lenape)
Figure 10. Example of the qctl.inp file with two control tables.
77
-------
9.9 Qser.inp Input File
The qser.inp file, shown in Figure 11, is used to specify time varying inflow and
withdrawals. Its structure is similar to other EFDC1D time series input files.
Withdrawals from a model cell are specified by a negative flow multiplier or negative
values in the flow column.
9.10 Sdser.inp and Snser.inp Input File
The sdser.inp and snser.inp files are used to specify inflow and open boundary cohesive
and noncohesive sediment concentrations. Since the EFDC1D model can simulate
multiple classes of each sediment group, their time series files can be slightly different
than those for other concentration variables. Figure 12 shows an example of the sdser.inp
for a single class of cohesive sediment. The file is identical in form to the dser.inp file.
Figure 13 shows an example of the snser.inp for two classes of noncohesive sediment.
Note that the sediment concentration adjustment factors for the second sediment class are
included on a second header line before the time series data.
9.11 Show.inp Input File
The show.inp file, shown in Figure 14 is used to control screen display during run time.
Use of this option is set in data block 2. The show.inp file is reread after NSHOWR time-
steps, allowing the user to edit the file and change options between subsequent reads.
When active, time, water surface elevation, velocity, concentration variables, and vertical
diffusivity (for 3D applications only) in computational grid cell ISHOWC, JSHOWC, are
displayed numerically on the screen. A map relating the computational I,J locations to
the cellnet.inp defined cell indices is generated by the model and output in file
mapld2ij .out, at the start of a simulation. Options defined by NSTYPE include:
1 - strip chart (not recommended)
2 - time-step rather than time, concentration variable shown is salinity
3 - time in days and salinity
4 - time in days and temperature
5 - time in days and total cohesive and total noncohesive sediment
The total cohesive and noncohesive concentrations displayed for option 5 represent the
sum of all classes of each sediment type. The parameter ISHPRT defines the frequency
of screen output in number of time-steps.
9.12 Txser.inp Input File
Figure 15 shows an example of the toxic contaminant time series file for two
contaminants. The convention used follows that for multiple sediment classes.
78
-------
9.13 Wser.inp Input File
An example of the wind speed and direction time series file, wser.inp is shown in Figure
16. The file includes time adjustment and unit conversion factors and a speed conversion
factor for wind speed. When ISWDINT equal to zero or one, direction is bearing (angle
CW from north) to which the wind is going (0) or coming from (1). For ISWDINT equal
to 3, the second and third data columns represent the east and north vector components of
the wind velocity.
QSER.INP for Brandywine Creek, Christina River Basin
flow in cfs; harmonic mean flow for nonpoint sources
ISTYP MQSER(NS) TCQSER(NS) TAQSER(NS) RMULADJ(NS) ADDADJ(NS) ICHGQS
if istyp.eq.l then read depth weights and single value of QSER
(WKQ(K),K=1,KC)
TQSER(M,NS) QSER(M,1,NS) !(mqser(ns) pairs for ns=l,nqser series)
else read a value of qser for each layer
TQSER(M,NS) (QSER(M,K,NS),K=l,KG) !(mqser(ns) pairs)
0 7 86400.00 0.00 0.02832 0.00000 0 ns = 1
-10000.0000 2.6306
121.5000 2.6306
122.5000 2.4074
123.5000 2.6689
124.5000 3.3037
125.5000 3.4919
10000.0000 0.9863
0 7 86400.00 0.00 0.02832 0.00000 0 ns = 2
-10000.0000 4.3250
121.5000 4.3250
122.5000 3.9562
123.5000 4.3863
124.5000 5.4313
125.5000 5.7381
126.5000 4.3250
Figure 11. Example of qser.inp file with two flow time series.
79
-------
C sdser.inp file - cohesive sediment
C repeats nsdser times, test case
C
C ISTYP MCSER(NS,1) TCCSER(NS,1) TACSER(NS,1) RMULADJ(NS,1) ADDADJ(NS,1)
C
C if istyp.eq.l then read depth weights and single value of CSER
C
C (WKQ(K),K=1,KC)
C
C TCSER(M,NS,1) CSER(M,NS,1) !(mcser(ns,1) pairs for ns=l,ncser(1) series)
C
C else read a value of dser for each layer
C
C TCSER(M,NS,1) (CSER(M,K,NS,1),K=l,KG) !(mcser(ns,1) pairs)
C
0 4 86400.0 0. 1. 0.
-100.0 1.
100.0 1.
1000.0 1.
10000.0 1.
0 4 86400.0 0. 1. 0.
-100.0 1.
100.0 1.
1000.0 1.
10000.0 1.
Figure 12. Example of the sdser.inp file with two time series for a single cohesive
sediment class.
80
-------
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
snser.inp file - noncohesive sediment - 2 size classes
repeats nsnser times, test case
ISTYP MCSER
if istyp.eq.
(WKQ (K) ,K=1,
TCSER(M,NS,1
else read a
TCSER(M,NS,1
0 4
-100.0
100 .0
1000.0
10000. 0
0 4
-100 . 0
100.0
1000 .0
10000.0
(NS,1) TCCSER(NS
1 then read depth
KG)
) CSER(M,NS, 1)
value of dser for
) (CSER(M,K,NS,
86400.0 0.
1.
1.
1.
1.
1.
1.
1 .
1.
86400.0 0.
2 .
2 .
2.
2.
2 .
2 .
2 .
2.
,1) TACSER(NS,1) RMULADJ(NS, 1) ADDADJ(NS,1)
weights and single value of CSER
! (mcser (ns , 1) pairs for ns=l ,ncser (1) series)
each layer
1),K=1,KC) ! (mcser (ns, 1) pairs)
1. 0.
1. 0. ! adjustment factors for 1st class
Isilt (very fine sand)
! sand (fine to coarse sand)
1. 0.
1. 0. ! adjustment factors for 2nd class
Isilt (very fine sand)
! sand (fine to coarse sand)
Figure 13. Example of the snser.inp file with two time series for two classes of
noncohesive sediments.
C show.inp file, in free format across line
C
C NSTYPE NSHOWR ISHOWC JSHOWC ISHPRT
C
C ZSSMIN ZSSMAX SSALMAX
C
3 360 2 31
-500. 500. 35.
Figure 14. Example of the show.inp file.
81
-------
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
txser.inp file - sorptive toxic contaminant time series file
repeats ntxser times, test case
ISTYP MCSER(NS, 1) TCCSER(NS,1) TACSER(NS,1) RMULADJ (NS , 1 ) ADDADJ(NS,1)
if istyp.eq.l then read depth weights and single value of CSER
(WKQ (K) ,K=1,
TCSER(M,NS,1
else read a
TCSER(M,NS,1
0 4
-100.0
100 .0
1000.0
10000. 0
0 4
-100 .0
100.0
1000 .0
10000.0
KG)
) CSER (M,NS, 1) ! (mcser (ns, 1) pairs for ns=l ,ncser (1) series)
value of dser for each layer
) (CSER(M,K,NS,1) , K=1,KC) ! (mcser (ns , 1 ) pairs)
86400 .0 0 . 1. 0.
1. 0. ! adjustment factors for 2nd toxic
1. ! toxic 1 concentration
1. ! toxic 2 concentration
1.
1.
1.
1.
1 .
1.
86400.0 0. 1. 0.
1. 0. ! adjustment factors for 2nd toxic
1. ! toxic 1 concentration
1. ! toxic 2 concentration
1.
1.
1 .
1.
1.
1.
Figure 15. Example of the txser.inp file for two time series and two toxic contaminants.
82
-------
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
wser.inp file
WIND FORCING
MASER(NW)
TCASER (NW)
TAASER (NW)
WINDSCT(NW)
ISWDINT(NW)
MASER TCASER
, in free format across line, repeats nw=l,nwser times
FILE, USE WITH 7 APRIL 97 AND LATER VERSIONS OF EFDC
=NUMBER OF
=DATA TIME
TIME DATA POINTS
UNIT CONVERSION TO SECONDS
=ADDITIVE ADJUSTMENT OF TIME VALUES SAME UNITS AS INPUT TIMES
=WIND SPEED CONVERSION TO M/SEC
0
1
2
TASER(M) WINDS
3447 86400
0. 0354
120. 0354
120.0771
120. 1188
4
4
4
4
=DIRECTION
DIRECTION
DIRECTION
WINDS IS
TAASER
CONVENTION
TO
FROM
EAST VELOCITY, WINDD IS NORTH VELOCITY
WINDSCT ISWDINT
(M) WINDD(M)
0.
.16 310
.16 310
.16 330
.68 330
1. 0
.00
.00
.00
.00
Figure 16. Example of the wser.inp file.
83
-------
10. Model Output Files and Post-processing
The primary model output files are the dump Id.out file and the cell time series files.
Controls for writing the dumpld.out files are found in data block 28 of the efdcld.inp
file. Controls for writing the cell time series files are in the cellnet.inp file and data block
34 of the efdcld.inp files. At each write to the dumpld.out file a time header line
127.0000 TIME IN DAYS
is written followed by a variable definition line which in sequence defines the
corresponding values written in ASCII test format. Definition of the variable header is as
follows:
LID - Cell index number from cellnet.inp
L - EFDC1D single count computational index
I - EFDC1D I double count computational index
J - EFDC1D J double count computational index
X - cell center east coordinate from cellnet.inp
Y - cell center north coordinate from cellnet.inp
BEL - cell bottom elevation
DEP - cell depth relative to lowest point in cross-section
AREA - cell cross-section area
WPER - cell wetted perimeter
BSRF - cell water surface width normal to flow direction
VEL - velocity (negative indicates downstream direction)
Q - discharge (negative indicates downstream direction)
SAL - cell salinity
TEM - cell temperature
SED - concentration of cohesive sediment (1st class if
multiple classes)
SND1 - concentration of 1st class of noncohesive sediment
SND2 - concentration of 2nd class of noncohesive sediment
SEDB - bed mass per unit area of 1st class of cohesive sediment
SNDB1 - bed mass per unit area of 1st class of noncohesive
sediment
SNDB2 - bed mass per unit area of 2nd class of noncohesive
sediment
TAUBED - bed shear stress
The current output of only the first cohesive sediment class and the first two noncohesive
sediment classes was designed for consistency with HSPF embedded applications of
EFDC1D which have these class number limitations. The output format can be changed
by modification of subroutine DUMP ID.
84
-------
11. References
Ackers, P., and W. R. White, 1973: Sediment transport: New approaches and analysis. J.
Hyd. Div. ASCE, 99, 2041-2060.
Ariathurai, R., and R. B. Krone, 1976: Finite element model for cohesive sediment
transport. J. Hyd Div. ASCE, 102, 323-338.
Ambrose, R. B., T. A. Wool, and J. L. Martin, 1993: The water quality analysis and
simulation program, WASPS: Part A, model documentation version 5.1. U. S. EPA,
Athens Environmental Research Laboratory, 210 pp.
Bear, J., 1879: Hydraulics ofgroundwater, McGraw-Hill, New York.
Bagnold, R. A., 1956: The flow of cohesionless grains in fluids. Phil. Trans. Roy. Soc.
Lond., Series A, Vol 249, No. 964, 235-297.
Blumberg, A. F., B. Galperin, and D. J. O'Connor, 1992: Modeling vertical structure of
open-channel flow. J. Hydr. Engr., 118, 1119-1134.
Boyer, J. M., S. C. Chapra, C. E. Ruiz, and M. S. Dortch, 1994: RECOVERY, a
mathematical model to predict the temporal response of surface water to contaminated
sediment. Tech. Rpt. W-94-4, U. S. Army Engineer Waterways Experiment Station,
Vicksburg, MS, 61 pp.
Burban, P. Y., W. Lick, and J. Lick, 1989: The flocculation of fine-grained sediments in
estuarine waters. J. Geophys. Res., 94, 8323-8330.
Burban, P. Y., Y. J. Xu, J. McNeil, and W. Lick, 1990: Settling speeds of floes in fresh
and seawater. J. Geophys. Res., 95, 18,213-18,220.
Cargill, K. W., 1985: Mathematical model of the consolidation and desiccation processes
in dredge material. U.S. Army Corps of Engineers, Waterways Experiment Station,
Technical Report D-85-4.
Dortch, M., C. Ruiz, T. Gerald, and R. Hall, 1998: Three-dimensional contaminant
transport/fate model. Estuarine and Coastal Modeling, Proceedings of the 5th
International Conference, M. L. Spaulding and A. F. Blumberg, Eds., American Society
of Civil Engineers, New York, 75-89.
Einstein, H. A., 1950: The bed load function for sediment transport in open channel
flows. U.S. Dept. Agric. Tech. Bull., 1026.
Galperin, B., L. H. Kantha, S. Hassid, and A. Rosati, 1988: A quasi-equilibrium
turbulent energy model for geophysical flows. J. Atmos. Sci., 45, 55-62.
85
-------
Garcia, M., and G. Parker, 1991: Entrainment of bed sediment into suspension. J. Hyd.
Engrg., 117, 414-435.
Gibbs, R. J., 1985: Estuarine Floes: their size, settling velocity and density. J. Geophys.
Res., 90, 3249-3251.
Gibson, R. E., G. L. England, and M. J. L. Hussey, 1967: The theory of one-dimensional
consolidation of saturated clays. Geotechnique, 17, 261-273.
Hamrick, J. M., 1992: A three-dimensional environmental fluid dynamics computer code:
Theoretical and computational aspects. The College of William and Mary, Virginia
Institute of Marine Science, Special Report 317, 63 pp.
Hamrick, J. M., 1996: Users manual for the environmental fluid dynamic computer code.
The College of William and Mary, Virginia Institute of Marine Science, Special Report
328, 224 pp.
Hamrick, J. M., and T. S. Wu, 1997: Computational design and optimization of the
EFDC/HEM3D surface water hydrodynamic and eutrophication models. Next
Generation Environmental Models and Computational Methods. G. Delich and M. F.
Wheeler, Eds., Society of Industrial and Applied Mathematics, Philadelphia, 143-156.
Hayter, E. J., and A. J. Mehta, 1983: Modeling fine sediment transport in estuaries.
Report EPA-600/3-83-045, U.S. Environmental Protection Agency. Athens, GA.
Hayter, E.J., M. Bergs, R. Gu, S. McCutcheon, S. J. Smith, and H. J. Whiteley, 1999:
HSCTM-2D, a finite element model for depth-averaged hydrodynamics, sediment and
contaminant transport. Technical Report, U. S. EPA Environmental Research Laboratory,
Athens, GA.
Hwang, K.-N, and A. J. Mehta, 1989: Fine sediment erodibility in Lake Okeechobee.
Coastal and Oceanographic Engineering Dept, University of Florida, Report
UFL/COEL-89/019, Gainesville, FL.
Laursen, E., 1958: The total sediment load of streams J. Hyd. Div. ASCE, 84, 1-36.
Letter, J. V., L. C. Roig, B. P. Donnell, Wa. A. Thomas, W. H. McAnally, and S. A.
Adamec, 1998: A user's manual for SED2D-WES, a generalized computer program for
two-dimensional, vertically averaged sediment transport. Version 4.3 Beta Draft
Instructional Report, U. S. Army Corps of Engrs., Wtrwy. Exper. Sta., Vicksburg, MS.
Le Normant, C., E. Peltier, and C. Teisson, 1998: Three dimensional modelling of
cohesive sediment in estuaries, in Physics of Estuaries and Coastal Seas, (J. Dronkers and
M. Scheffers, Eds.), Balkema, Rotterdam, pp 65-71.
Lick, W., and J. Lick, 1988: Aggregation and disaggregation of fine-grained lake
sediments. J Great Lakes Res., 14, 514-523.
86
-------
Mehta, A. 1, E. J. Hayter, W. R. Parker, R. B. Krone, A. M. Teeter, 1989: Cohesive
sediment transport. I: Process description. J. Hyd. Engrg., 115, 1076-1093.
Mehta, A. J., T. M. Parchure, J. G. Dixit, and R. Ariathurai, 1982: Resuspension potential
of deposited cohesive sediment beds, in Estuarine Comparisons, V. S. Kennedy, Ed.,
Academic Press, New York, 348-362.
Mehta, A. J., and F. Jiang, 1990: Some field observations on bottom mud motion due to
waves. Coastal and Oceanographic Engineering Dept, University of Florida, Gainesville,
FL.
Mellor, G. L., and T. Yamada, 1982: Development of a turbulence closure model for
geophysical fluid problems. Rev. Geophys. Space Phys., 20, 851-875.
Meyer-Peter, E. and R. Muller, 1948: Formulas for bed-load transport. Proc. Int. Assoc.
Hydr. Struct. Res., Report of Second Meeting, Stockholm, 39-64.
Middleton, G. V., and P. R. Wilcock, 1994: Mechanics in the Earth and Environmental
Sciences. Cambridge University Press, Cambridge, UK.
Nielsen, P., 1992: Coastal bottom boundary layers and sediment transport, World
Scientific, Singapore.
Park, K., A. Y. Kuo, J. Shen, and J. M. Hamrick, 1995: A three-dimensional
hydrodynamic-eutrophication model (HEM3D): description of water quality and
sediment processes submodels. The College of William and Mary, Virginia Institute of
Marine Science. Special Report 327, 113 pp.
Rahmeyer, W. J., 1999: Lecture notes for CEE5560/6560: Sedimentation Engineering,
Dept. of Civil and Environmental Engineering, Utah State University, Logan, Utah.
Raukivi, A. J., 1990: Loose boundary hydraulics. 3rd Ed. Pergamon, New York, NY.
Ried, L, and L. E. Frostick, 1994: Fluvial sediment transport and deposition, in Sediment
Transport andDepositionalProcesses, K. Pye, ed., Blackwell, Oxford, UK, 89-155.
Rosati, A. K., and K. Miyakoda, 1988: A general circulation model for upper ocean
simulation. J. Phys. Ocean., 18, 1601-1626.
Shrestha, P. A., and G. T. Orlob, 1996: Multiphase distribution of cohesive sediments and
heavy metals in estuarine systems. J. Environ. Engrg., 122, 730-740.
Smagorinsky, J., 1963: General circulation experiments with the primitive equations, Part
I: the basic experiment. Mon. Wea. Rev., 91, 99-152.
Smith, J. D., and S. R. McLean , 1977: Spatially averaged flow over a wavy bed. J.
Geophys. Res., 82, 1735-1746.
87
-------
Smolarkiewicz, P. K., and T. L. Clark, 1986: The multidimensional positive definite
advection transport algorithm: further development and applications. J. Comp. Phys., 67,
396-438.
Smolarkiewicz, P. K., and W. W. Grabowski, 1990: The multidimensional positive
definite advection transport algorithm: non-oscillatory option. J. Comp. Phys., 86, 355-
375.
Spasojevic, M., and F. M. Holly, 1994: Three-dimensional numerical simulation of
mobile-bed hydrodynamics. Contract Report HL-94-2, US Army Engineer Waterways
Experiment Station, Vicksburg, MS.
Stark, T. D., 1996: Program documentation and users guide: PSDDF primary
consolidation, secondary compression, and desiccation of dredge fill. Instructional Report
EL-96-xx, US Army Engineer Waterways Experiment Station, Vicksburg, MS.
Suzuki, K., H. Yamamoto, and A. Kadota, 1998: Mechanism of bed load fluctuations of
sand-gravel mixture in a steep slope channel, Proc. of the 11th congress of APD IAHR,
Yogyakarta, pp.679-688.
Tsai, C. H., S. lacobellis, and W. Lick, 1987: Flocculation of fine-grained lake sediments
due to a uniform shear stress. J Great Lakes Res., 13, 135-146.
van Niekerk, A., K. R. Vogel, R. L Slingerland, and J. S. Bridge, 1992: Routing of
heterogeneous sediments over movable bed: Model development. J. Hyd. Engrg., 118,
246-262.
Van Rijin, L. C., 1984a: Sediment transport, Part I: Bed load transport. J. Hyd. Engrg.,
110, 1431-1455.
Van Rijin, L. C., 1984b: Sediment transport, Part II: Suspended load transport. J. Hyd.
Engrg., 110, 1613-1641.
Villaret, C., and M. Paulic, 1986: Experiments on the erosion of deposited and placed
cohesive sediments in an annular flume and a rocking flume. Coastal and Oceanographic
Engineering Dept, University of Florida, Report UFL/COEL-86/007, Gainesville, FL.
Ziegler, C. K., and B. Nesbitt, 1994: Fine-grained sediment transport in Pawtuxet River,
Rhode Island. J. Hyd. Engrg., 120, 561-576.
Ziegler, C. K., and B. Nesbitt, 1995: Long-term simulation of fine-grained sediment
transport in large reservoir. J. Hyd. Engrg., 121, 773-781.
Yang, C. T., 1973: Incipient motion and sediment transport. J. Hyd. Div. ASCE, 99, 1679-
1704.
Yang, C. T., 1984: Unit stream power equation for gravel. J. Hyd. Engrg., 110, 1783-
1797.
88
-------
Yang, C. T., and A. Molinas, 1982: Sediment transport and unit streams power function.
J. Hyd. Div. ASCE, 108, 774-793.
89
-------
Appendix A: Compilation Considerations
Compilation of the EFDC1D model requires the FORTRAN source code efdcld.f and
two include files, efdc.cmn and efdc.par. The efdc.cmn file is a global common block file
identical to that used for the multi-dimensional EFDC model. The efdc.par file contains
parameter statements defining the dimensions of global and local arrays. All parameters
in efdc.par are defined within the file. The primary reason for recompilation of EFDC ID
would be to enlarge array storage to accommodate a greater number of cells or
more/longer time series files, more inflow locations and more control structure locations
and/or control tables.
The number of cells and the domain size are specified by LCM, ICM, and JCM. The
allowed number of active cells is actually LCM-2, since the internal data structure utilizes
2 dummy cells. The parameters ICM and JCM define the dimensions of a computational
space. For EFDC ID, guidelines for setting these two parameters are:
ICM = (number of cells in the longest even order segment) + 6
JCM = (sum of number of cells in all odd order segments) + (number of odd order
segments) + 4
The junctions parameters, NJUNX and NJUNY, should be greater than the number of
segments plus two. Other parameters that might need to be changed include:
NCSERM = maximum number of concentration time series for any variable
NDCSER = maximum length of any concentration time series
NQSIJM = maximum number of inflow and flow withdrawal locations
NQSERM = maximum number of flow time series
NDQSER = maximum length of longest flow time series
NQCTLM = maximum number of flow control locations
NQCTTM = maximum number of flow control tables
NDQCLT= maximum length of longest control table
NDQCLT2 = NDQCLT
NDASER = maximum length of the atmospheric time series
NDWSER = maximum length of the wind time series
90
-------
NPSERM = maximum length of the water surface elevation time series
NSCM = maximum number of cohesive sediment classes
NSNM = maximum number of noncohesive sediment classes
NSTM = NSCM + NSTM
NTXM = maximum number of sorptive toxic contaminants
NSTVM = 5 + NSCM + NSTM + NTXM
The EFDC1D model source code and include files conform to standard FORTRAN 77.
The limited number of non-standard extensions in the code are supported by compilers
from major PC compiler vendors including Absoft, Compaq, and Lahey. The code has
been verified as compiling clean under FORTRAN 90/95 using the Wintel and Macintosh
versions of Absoft's Pro FORTRAN compilers after changing the file descriptor
ACCESS=APPEND' to 'POSITION=APPEND' throughout the source. For compilers
distinguishing between upper and lower case letters, a fold to upper case option (-N109
for the Absoft compilers) should be used. The multidimensional EFDC model source
code is compatible with SUN Unix FORTRAN, and EFDC ID should also be compatible
although it has not been tested.
91
-------
Appendix B: Sample Applications
Two sample application file sets are currently distributed with EFDC1D. The first file set
is for the Los Angles River, shown in Figure Bl. The cell network is relatively simple,
being composed of the main stem of the river and seven-second order tributaries. Due to
large bed elevation differences between the main stem and the tributaries, where the
tributaries join the main stem, all of the tributaries are connected to the main stem by
flow control structures representing free over falls.
The second file set is for Brandywine Creek in the Christina River Basin of Delaware and
Pennsylvania, shown in Figure B2. The main stem, segment 1, corresponds to the lower
portion of the creek extending upstream to a dam just below the confluence of the East
and West Branches. Segment 2, corresponding to the East Branch, connects to segment 1
by a control structure. The lower portion of the West branch, segment 3, joins the East
Branch segment 2, by a free flowing connection. Segment 3 extends upstream along the
West Branch to a dam just below the confluence with Buck Run. Segment 4 is along the
West Branch to a dam just beyond the confluence with Buck Run. Buck Run, segment 5,
joins segment 4 by a free flowing connection. Segments 6, 7 and 8 represent the
remaining portion of West Branch and are separated by dams.
92
-------
Western Burbank Chann
r~^
Verdugo Vtesh
A/M3(d) Listed Segments
[ j Cataloging UnitBountfaries
Figure B1. Los Angles River and Tributaries.
93
-------
•••NPOES Di
Figure B2. Brandywine Creek and Tributaries
94
------- | |