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

-------