vvEPA
                United States
                Environmental Protection
                Agency
                Region V
                 Great Lakes National
                 Program Office
                 536 South Clark Street, Room 932
                 Chicago, IL 60605
EPA-9O5/9-82-O01
ANSWERS  -
USERS MANUAL
I

-------
              ANSWERS
(Areal Nonpoint Source Watershed Environment Response Simulation)
                   User's Manual
          David B. Beasley, PhD, PE and Larry F. Huggins, PhD, PE
 Agricultural Engineering Department, Purdue University, West Lafayette, Indiana


                         Assisted by:
                U.S. Environmental Protection Agency, Region V
                 230 South Dearborn St., Chicago, IL 60604

               Agricultural Experiment Station, Purdue University

              U.S. Department of Agriculture (ARS, SCS, and ASCS)

          Allen County Soil and Water Conservation District, Ft. Wayne, Indiana

          Indiana Heartland Coordinating Commission, Indianapolis, Indiana
                         U S  Environmental Protection Agency
                         Region 5, Library (PL-12J)
                         77 West Jackson Boulevard, 12th Floor
                         Chicago, IL  60604-3590

-------
Contents
  I. Introduction  1
      Acknowledgment  1
      Disclaimer  1
 II. Concepts  3
      Philosophy  3
      Model Structure  4
      Hydrplogic Considerations  5
      Applicability  6
 III. Component Relationships  7
      Flow Characterization  7
      Rainfall Rate  9
      Infiltration  10
      Sediment Detachment and Movement  12
      Best Management Practices  15
      Programming Overview   15
 IV. Data Preparation for Answers  17
      General  17
      Data File Construction  17
       Simulation Requirements  17
       Rainfall Information  19
       Soils Information  19
       Land Use and Surface Information  21
       Channel Descriptions  21
       Individual Element Information  22
 V. Answers Output  24
      General 24
       Input Information "Echo"  24
       Watershed Characteristics  24
       Flow and Sediment Information and BMP Effectiveness 24
       Net Transported Sediment Yield or Deposition 26
       Channel Deposition  27
      Graphical Output  27
List of References   30

-------
Appendices

 A. Soil Parameters  33
      Section A.1  33
        Total Porosity (TP)  33
        Field Capacity (FP)  33
        Infiltration Control Zone Depth (DF)  34
        Antecedent Soil Moisture (ASM)  34
      Section A.2  35
        Infiltration Rate Descriptors (FC and A)  35
        Infiltration Exponent (P) 37
      Section A.3  37
        Soil Erodibility-USLE "K" (K)  37
        Subsurface Drainage Characteristics 38
      Section A.4  38
 B. Land Use and Surface Parameters  39
      Section B.1  39
        Interception Parameters (PIT and PER)  39
        Manning's n (N)  39
        Relative Erosiveness (C) 39
      Section B.2  40
        Surface Storage Descriptors (HU and RC)  40
 C. Individual Element Information   41
      Section C.1  41
        Simplified Data File Construction  41
        Data File Construction Using Topographic Maps 42
      Section C.2  43
 D. Enhanced Sediment Model   45
      General   45
      Model Development  45
      Input and Output Format Changes   51

-------
                       CHAPTER I: INTRODUCTION

This manual is intended to present the theory behind the ANSWERS simulation model, a broad
overview of the computational algorithms employed in the current implementation of the model and,
most importantly, a practical guide for persons who want to use the model as a tool in developing
nonpoint source pollution control measures. Depending upon the intended application, only certain
sections of the manual may be relevant.

Chapter II gives a broad overview of the model and should generally be read by anyone considering
its use. Chapter III goes into computational details of the component relationships and is necessary
reading only if a thorough understanding of the internal workings of the model is desired. Two
different sediment detachment/transport sub-models  are presented in this manual. Chapter III
details the current, "production" version of the model,  while Appendix D describes the new, more
detailed model capable of providing information on particle-size distributions. Chapters IV and V,
together with Appendices A, B, and C, detail procedures for simulating watershed behavior and
interpreting the results.

Finally, it must be noted that developmental work on improving the utility of ANSWERS is still
on-going.  A listing of the ANSWERS computer program was deliberately excluded from this
publication in order that interested users would obtain the latest version  of the model rather than
attempt to manually reproduce a  possibly obsolete  listing. In  particular, developmental work is
nearing completion on a version which includes direct simulation of nutrient losses in addition to the
sediment yield discussed herein. A supplement to this manual will be distributed with all releases of
the expanded version of the model.


                               Acknowledgments
The ANSWERS simulation model  development was financed  with Federal funds from the U.S.
Environmental Protection Agency under Sections 108a and 208 of PL 92-500 and by the Purdue
Agricultural Experiment Station. The EPA grants were administered by the Soil and Water Cpnserva-
tion District of Allen County, Indiana and by the Indiana Heartland Coordinating  Commission in
Indianapolis. Special recognition is due the U.S. Dept. of Agriculture, Agriculture Research Service,
for  technical assistance with  field  rainujator experiments conducted  in  cooperation with the
Department of Agronomy, Purdue University. These field tests, together with other plot data and
professional consultation made available by ARS personnel, provided the basic information for
development of the erosion and sediment transport components of the ANSWERS model. Earlier
research supported in part by the  Dept. of Interior, Office of Water Resources Research provide a
foundation for the hydrologic portion of the mode!.

Numerous graduate students of the Department of Agricultural Engineering and professional col-
leagues have contributed greatly to various components and programming algorithms. Individuals
deserving special mention include: J.R. Burney, T.A. Dillaha, III, G.R. Foster, H.M. Galloway, J.V.
Mannering, T.D. McCain, E.J. Monke,  D.W. Nelson, and W.H. Wischmeier.

                                    Disclaimer
The contents of this  publication  do not necessarily reflect the views  and policies of the U.S.
Environmental Protection Agency,  nor does mention of trade names or commerical products consti-
tute endorsement or recommendation for use. While every reasonable effort has been made in the
development of the ANSWERS model to provide a computer program free of logical errors, no
absolute assurance can be given that this effort has been entirely successful. Furthermore, neither
the U.S. Environmental Protection  Agency, the Allen County Soil and Water  Conservation District,
the Indiana Heartland Coordinating Commission or Purdue University assumes any responsibility
for liability, either direct or indirect, as a result of actions predicated on model simulation results.

-------
In order to assure that modeling results are as reliable as possible, always use the latest release
of ANSWERS. Contact Dr. David B. Beasley in order to determine the most recent version of the
ANSWERS model and to obtain information concerning acquisition of ANSWERS and related pro-
grams.

-------
                          CHAPTER II: CONCEPTS


                                     Philosophy
 ANSWERS is a model intended to simulate the behavior of watersheds having agriculture as their
 primary land use, during and immediately following a rainfall event. Its primary application was
 envisioned to be planning and evaluating various strategies for controlling nonpoint source pollu-
 tion from intensively cropped areas. For such situations, a watershed's hydrologic response to a
 storm event is the controlling mechanism for transporting pollutants to the drainage network.

 A fundamental characteristic of  the ANSWERS model is its  distributed parameter approach as
 contrasted to the more common lumped parameter modeling efforts. While this approach is gener-
 ally more computationally intensive, its inherent advantages were deemed to more than offset this
 factor.

 A distributed parameter watershed model incorporates the  influences of the spatially  variable,
 controlling parameters, e.g., topography, soils, land use, etc., in a manner internal to its computa-
 tional algorithms. In contrast, lumped parameter models incorporate, to whatever degree they do
 not ignore, these effects by an a priori analysis of a watershed's spatially variable characteristics.
 In other words,  the lumped approach uses some type of averaging technique to generate an
 "effective" coefficient(s) for characterizing the influence of specific, non-uniform distributions of
 each parameter. The influence of this distribution is then represented by the "lumped" coefficients,
 and the resulting model is treated as a  mathematical transformation of input into output, i.e., a
 "black box", for the subsequent simulation.

 A primary advantage of a distributed parameter analysis is its potential for providing a more
 accurate simulation of natural catchment behavior. The term potential is used because increased
 accuracy is by no means a direct consequence of using a distributed analysis; rather, it is realized
 only if the model is designed to avoid the constraints imposed by lumped parameters.

 Lumped models almost  invariably employ some weighting function to account for the spatial
 variability of watershed parameters such as soil type, cover and slope steepness. Such weighting
 functions, regardless of how elaborate, are applied to the catchment prior to modeling runoff. This
 constrains the parameter values to be independent of the magnitude and temporal distribution of the
 storm event. Such a constraint is  valid only for linear systems. Thus, the assumptions and limita-
 tions of behaving, at least to some degree, as a linear system are subtly imposed on lumped models.
 Another linear system assumption implicit to almost all weighting functions used with lumped
 models  results from ignoring the  influence of geographic placement  of spatially varying factors
 within the watershed boundaries. The magnitude of error associated with such approximations has
 been demonstrated by Muggins, et al. (1973).

 A second major advantage of a distributed model is its inherent ability to simultaneously  simulate
 conditions at all points within the watershed. This ability to depict what is happening throughout the
 watershed greatly increases the amount  and utility of information provided. In addition, it permits
 simulation of processes that change both spatially and temporally throughout the watershed. The
 accuracy with which interacting processes can be modeled is thereby increased.

 Finally,  distributed models greatly facijitate incorporation of relationships developed from small
 scale "plot-size" studies to yield predictions on a watershed scale. It is much easier to formulate the
 individual  processes  being modeled  as  independent equations applicable at a  point, letting the
 subsequent integration  process incorporate  effects of spatial and temporal variability, than to
 develop an elaborate weighting function for each process. This approach also directly accounts for
 process interactions that would otherwise be ignored or require complex modifications of weighting
 functions.

A complete discussion of the  concepts and chronologic development of the ANSWERS model can be
found in Muggins and  Monke (1966), Beasley (1977), Beasley, et al. (1980), and Dillaha (1981).

-------
                                 Model Structure
ANSWERS is a deterministic model based upon the fundamental hypothesis that:
    "At every point within a watershed, functional relationships exist between water flow
    rates and those hydmlogic parameters which govern them, e.g., rainfall intensity, infiltra-
    tion, topography, soil type, etc. Furthermore, these flow rates can be utilized in conjunc-
    tion with appropriate component relationships as the basis for modeling other transport-
    related phenomenon such as soil erosion and chemical movement within that watershed."
An important feature of the above hypothesis is its applicability on a "point" basis.

In order to apply this approach on a practical scale, the point concept is relaxed to refer instead to a
watershed "element". An element is defined to be an area within which all hydrologically significant
parameters are uniform. Of course, this process of going from a point to an elemental area could be
extended indefinitely until one assumed the entire watershed was composed of a single element with
"averaged" parameter values, i.e., a lumped model. The actual geometric size of an element is not
critical because there is no finite-sized area within which some degree of variation in one or more
parameters does not exist. The crucial concept is that an element must be sufficiently small that
arbitrary changes pf parameter values for a single element have a negligible influence upon the
response of the entire watershed.

A watershed to be modeled is assumed to be composed of  "elements" as shown in Figure 2-1. A
square element shape was chosen to ease the task of data file preparation and to facilitate
computational  convenience. It is this user supplied elemental data file, listing the physical charac-
teristics of each watershed element, which permits ANSWERS to simulate the unique behavior of any
particular watershed.
         Figure 2-1. Watershed Divided into Elements with Channel Elements Shaded.
                                           4

-------
Component relationships characterizing water and  pollutant production need only simulate the
behavior of a small, uniform elemental area. Parameter values are allowed to vary in an unrestricted
manner between elements; thus, any degree of spatial variability within a watershed is easily
represented. Individual elements collectively act as a composite system because of supplied
topographic data for each element delineating flow directions in a manner consistent with the
topography of the watershed being modeled. Element interaction occurs because surface flow
(overland and channel), flow in tile lines and groundwater flow from each element becomes inflow to
its adjacent elements. Pollutants are generated and transported by these flows and by raindrop
impact.

It is the distributed model structure, which inherently provides the ability to simulate the fate of any
type of pollutant and to integrate the response of individual elements to yield a composite watershed
simulation, that is the foundation of ANSWERS. The component relationships included in a specific
release of the model will determine which pollutant processes are described and to what extent.
Specific relationships included have no effect whatsoever on the integration algorithm or distributed
model  concepts. Of course, relationships chosen have a marked  impact upon the accuracy with
which the model can characterize real watershed behavior. The significant consequence is that to
substitute one component relationship for another when subsequent research develops improved
relationships is a relatively trivial task.

                           Hytlroiogie Considerations
As indicated earlier, hydrologic processes are the driving force within the model. Consequently, a
conceptual understanding of  those processes, as they apply to each independent watershed ele-
ment, is a prerequisite to studying component algorithms.

Hydrologic processes, for which component relationships have been incorporated within ANSWERS,
are shown qualitatively in Figure2-2. After rainfall begins, some is intercepted by the vegetal canopy
until such time as the interception  storage potential is met. When the rainfall rate exceeds the
interception rate, infiltration into the soil begins. Since the infiltration rate decreases in an exponen-
tial manner as the soil water storage increases, a point may be reached when the rainfall rate
exceeds the combined infiltration and interception rates. When this occurs, water begins to accumu-
late on the surface in micro-depressions.
                                                             INTERCEPTION



                                                             INFILTRATION
     LU
     Ct
h-

2
ID

CC
LJ
a.

LJ
h-

QC
                                                        SUBSURFACE  DRAINAGE
          Figure 2-2. Water Movement Relationships for Small Watershed Elements.

-------
Once surface retention exceeds the capacity of the micro-depressions, runoff begins. The accumu-
lated water, when in excess of surface retention capacity, produces surface runoff and is termed
surface detention. Subsurface drainage begins when the pressure potential of the groundwater
surrounding a tile drain exceeds atmospheric potential. A steady-state infiltration rate may be
reached if the duration and intensity of the rainfall event are sufficiently large.

When rainfall ceases, the surface detention storage begins to dissipate until surface runoff ceases
altogether.  However, infiltration continues until depressional water is no longer available. Subsur-
face drainage continues as long as there is excess soil water surrounding the drains. The long
recession curve on the outflow hydrograph, typical of tile drained areas, is then produced. Slowly
falling recession limbs are also produced by interflow, the emergence of groundwater into the
surface drainage network.

Soil  detachment, transport, and deposition are very closely related to concurrent  hydrologic
processes in a watershed. Detachment and transport can both be accomplished by either raindrop
impact or overland flow. Detachment by rainfall occurs throughout a storm even though overland
flow may not occur. Thus, most of the soil particles detached prior to flow initiation are deposited
and to some extent, reattached. Detachment of soil particles by overland flow occurs when the shear
stress at the surface is sufficient to overcome the gravitational and cohesive forces of the particles.
Whether or  not a detached soil particle moves, however, depends upon the sediment load in the flow
and its capacity for sediment transport.

The transport of chemical pollutants from a land area is also highly related to the  hydrologic
behavior of a catchment and, for certain chemicals, e.g., phosphorus, to the soil erosion that occurs.
These processes can be readily incorporated into a distributed model by developing  component
relationships that characterize what happens within an individual watershed element. These pro-
cesses may be much simpler than those that must work on the watershed scale required by a lumped
approach.

Natural rainfall events do  not exhibit the steady appearance shown in Figure 2-2.  Furthermore,
uniformity of coverage over a watershed will usually vary during an event. In addition, hydrologic
responses of various areas within a watershed may vary greatly. Hence, the resultant hydrograph
for the entire watershed will contain at least some of the effects of all of these highly complex,
unsteady, non-uniform interactions. The distributed  approach provides a straightforward and
accurate method of simulating such a complex situation.

All materials leaving an element are assumed  to be transported with one  of the various flow
components. Overland flow either moves into an element's shadow channel element, if a channel is
specified for that element, or onto its adjacent elements. Tile and groundwater flows are assumed to
outlet directly into the specified channel network. It is this network of interacting flows which causes
the "independent" watershed elements to act together as a composite system.

                                    Applicability
The distributed nature of the ANSWERS model is both its strength and ultimate limiting factor. As will
be shown with subsequent examples, the degree of information provided to decision making bodies
and program  planners is extraordinary. However, costs for preparing an elemental data file and for
computer time increase with the number of elements required. Thus, economics and  the size  of
computer available limit the feasible number of elements and, consequently, the size of watershed
which can be simulated. Of course, selection of a larger element size permits simulation of a larger
area  for a given number  of elements. However, this comes at the expense of inaccuracy  in
representing the spatial diversity of a watershed and of component relationships within the model.
As a practical matter, element sizes normally range from 1 to 4 ha. with watershed sizes commonly
less than 10,000 ha. Additionally, the utilization of the expanded capability sediment transport
model (which allows particle-size calculations) requires additional computer resources and time.
Hence, all other things being equal, the expanded model will not simulate as large an area and it will
cost more (as much as two times) on a  unit area basis.


-------
           CHAPTER III: COMPONENT RELATIONSHIPS

This chapter outlines the specific mathematical relationships currently used to quantify the various
model component processes and also gives an overview of the programming algorithms employed.
jt is intended primarily for the person who  is doing research in model development or who is
interested in modifying some features of the current release of the model. Parameter names which
are given in all capital letters correspond to the programming names used for those variables.

The specific component relationships selected for the current version of ANSWERS are separately
discussed below. All except the time integration of the continuity equation are incorporated on a
modular basis. Modification or replacement of component relations such as infiltration or sediment
production does not affect  the algorithms for other components. In other words, the component
relationships are sufficiently independent from each other that user-supplied subroutines may be
substituted for those supplied with the "official" release of the model. This framework also permits
users to append  additional component relationships to simulate other processes important to
specific applications.

                            Flow Characterization
Mathematically, each element's hydraulic response is computed, as a function of time, by  an
explicit, backward difference solution  of the continuity equation:

                                        z  _ Q  = dS                               (3-1)
         where:   I  =  inflow rate to an element  from rainfall  and adjacent elements,
                 Q  =  outflow rate,
                 S  =  volume of water stored  in  an element,
                 t  =  time.


This equation may be solved when it is combined with a stage-discharge relationship. Manning's
equation, with appropriately different coefficients, is used as the stage-discharge equation for both
overland and channel flow routing.

Within its topographic boundary, a catchment is divided into an irregular matrix of square elements,
as shown in Figure 2-1.  Every element acts as an overland flow plane having a user specified slope
and direction of steepest descent. Channel flow is analyzed by a separate pattern of channel
elements (referred to hereafter as channel segments) which underlie i.e., are in the shadow of, the
grid of overland flow elements.

Elements designated  to have channel flow may be viewed as dual elements. These elements act as
ordinary overland flow elements, with the exception that all overland flow out of that element goes
into its "shadow" channel segment. Flow from a channel segment goes into the next downstream
channel segment. This downstream channel segment will also receive flow from any other channel
segments which are directed toward it and from its own overland flow element.

Overland and tile outflow from an element flows into neighboring elements according to the direction
of the element's slope. The slope direction is designated on input as the angle, in degrees, counter-
clockwise from the positive horizontal (row) axis. For the example shown in Figure 3-1 below, slope
direction is in the fourth quadrant. The slope direction angle equals 270 degrees plus angle ANG.

-------
                                                       Direction of
                                                       steepest slope
                              Al  +  A2
                          Figure 3-1. Partitioning of Overland Flow.


The fraction of outflow going into the adjacent row element, RFL, is:
                             RFL =
    tan (ANG)
        2
 if ANG < 45 deg.
                                                                                  (3-2a)
                     RFL = 1 -
tan (90-ANG)
     2
if 45 deg. < ANG <90 deg.
                                                                                  (3-2b)
with the remaining outflow going into the adjacent column element. Since everything, including
surface slope, within an element is assumed to be constant, this method of partitioning overland
flow seems intuitively obvious.

The most appropriate manner to partition tile flow is not as obvious as is overland flow. In general,
records are seldom available which delineate the layout of tile systems. However, limitations on
feasible installation depths mean that tile slopes must, with only temporary deviations, follow the
general topography. Therefore, the use  of an element's slope properties was chosen as a close
approximation which has the secondary benefit of eliminating the need for additional input informa-
tion.

Baseflow, the emergence of groundwater into the channel system, is simulated only crudely in the
current release of ANSWERS. All infiltrated water which moves past the  zone of tile drainage is
assumed to enter a single groundwater  storage reservoir. Water  is then released evenly into all
channel segments at a rate proportional  to the volume of accumulated storage.

The flow relationship utilized in conjunction with the continuity equation to perform the overland
flow routing is Manning's equation. The hydraulic radius is assumed equal to the average detention
depth in an element. The flow width is assumed to be the maximum width for that element, i.e., the
length measured  in a direction perpendicular to the overland flow  direction.

Surface detention is the water volume which must build up to sustain overland flow. Detention depth
is calculated as the total volume of surface water in an element, minus the retention volume (which
can only infiltrate), divided by the area of the element. This implies that the entire specified retention
volume of an element be f il led before any water becomes available for surface detention and runoff.
                                            8

-------
Surface detention is a component that can have a pronounced effect on surface runoff and drainage
characteristics of a watershed. Rough ground can store large amounts of water. Muggins and Monke
(1966), using several field surfaces, developed a relationship describing the surface storage poten-
tial of a surface as a function of the water depth in the zone of micro-relief. The form of that equation
used by ANSWERS is:


                                                   H  1./ROUGH                      (3_3)
                               DEP  =  HU  *  ROUGH * (
         where:    DEP = volume of stored water, in depth units,
                     H = height above datum,
                    HU = height of maximum micro-relief,
                 ROUGH = a surface characteristic parameter.


The specified surface roughness determines the volume of surface storage and also influences
infiltration rates during the recession limb of a hydrograph, as explained later.

Although the channel flow system is unrestricted in direction and branching, it is necessary that it be
continuous and that each element contain only one channel segment. To achieve greater definition it
is necessary to assume a smaller element size for all elements, with a consequent increase in core
storage and computer execution time.

As all overland flow from a dual element is constrained to enter its shadow channel segment, the
surface slope direction of a dual element is irrelevant for partitioning overland flow to adjacent
elements. Therefore, instead of specifying the direction of surface slppe for a dual element, this
parameter position  in the data file is used to specify the f^w direction for the shadow  channel
segment. This slope direction is of vital importance in establishing channel continuity. All outflow
from a channel segment  must enter a single adjacent channel segment located in one of the eight
directions of the cardinal axes or the diagonals. Thus, only 45 degree increments in slope direction
should be specified for dual  elements.

It is permissible to specify the slope, width and Manning's roughness coefficient for each channel
segment independent of the corresponding values for its overland flow element. Typically, rather
than having a unique set of  values for each channel segment, they are grouped into reaches with
similar coefficients. Manning's  equation is again used as the flow relationship necessary, in
conjunction with the continuity equation, to perform the routing calculations.

Much programming effort has been expended to develop computational algorithms which solve the
flow routing equations very efficiently. For example, a piece-wise linear segmented curve is used to
approximate Manning's equation and thereby eliminate the iteration process that would otherwise
be necessary when solving the continuity equation. Unfortunately, this computational optimization
has resulted in some very complex programming  logic. Therefore, it is recommended that only
someone intimately familiar with all aspects of the ANSWERS mode! attempt to modify this portion
of the computer program.

                                    Rainfall  Rate
The net rainfall rate, that which reaches the ground surface,  is dependent on the user specified
pluviograph(s) and on the rate of interception by vegetation. The net rainfajl rate for each rain gauge
and  crop is  calculated  by FUNCTION RAIN. Since a rain  gauge identifier is specified for each
watershed element,  it is theoretically possible to have each element subjected to a different storm
pattern. As a practical matter, the standard release of the model  is dimensioned to permit only 4
gauges.

-------
Interception is that water extracted from the incoming rainfall upon contact with and retention by the
vegetal canopy. Water retained by the vegetation, i.e., interception storage, is held primarily by
surface tension forces. This initial interception volume is quickly satisfied, particularly in more
intense storms. Since a dense vegetal cover can expose an immense surface area to rainfall, the
amount of moisture evaporating during a long duration storm can be appreciable. Intercepted water
which evaporates will be replenished during the storm. This creates a low level of interception
demand throughout a storm. However, for the high intensities of primary interest from the stand-
point of nonpoint source pollution from cropland, interception is a relatively minor hydrologic
component. In order to reduce simulation costs, interception was assumed to be uniform in rate and
total volume over each type of vegetation. Evaporation losses during a storm were assumed to be
negligible.

Horton  (1919) did a great deal of work in the area of estimating the amount and mechanisms
controlling interception. He studied the water intercepted by several species of trees as well as some
economically important crops. Values from 0.5 millimeter to 1.8 millimeters of interception storage
volume were found  to exist for trees and nearly as much for well developed crops. Values for
potential interception storage, PIT, are based primarily on his recommended  relationship.

The maximum potential interception (PIT), supplied as an input value, represents the available leaf
moisture storage in depth units (volume per unit land area). In each time increment  in which
interception storage remains unsatisfied, rainfall supplied to interception storage is calculated as
incremental intercepton (RIT), i.e., the product of the rainfall amount (RATE) and the portion of the
element covered by foliage (PER). The value of potential interception (PIT) and the net rainfall (RAIN)
are correspondingly decreased until all interception storage is satisfied. At this stage,  PIT is set
equal to zero and the net rainfall rate is subsequently equal to the  gauge rainfall rate for the
remainder of the simulation.

                                     Infiltration
Infiltration is one of the components to which ANSWERS is most sensitive, especially during low to
medium runoff  storms. Although  many years of research have been conducted on infiltration
phenomena, there is still no universally accepted method for describing infiltration on a watershed
scale. The  widely used, time-dependent equations 9? Horton (1939) and  Philip (1957) presume a
continuous water supply rate adequate to meet all infiltration demands.  This approach  does not
readily allow for the observed recovery in soil infiltration capacity during periods of light or zero
rainfall. To avoid this difficulty, the infiltration relation chosen for ANSWERS was the one developed
by Holtan (1961) and Overton (1965). In a dimensionally homogeneous form it can be expressed as:


                                   FMAX = FC  + A * (P™/                          (3-4)


         where:  FMAX = infiltration capacity with surface inundated,
                   FC = final  or steady state infiltration capacity,
                    A = maximum infiltration  capacity in excess  of  FC,
                   TP = total  volume of pore  space within the control  depth,
                  PIV = volume of water that  can be stored within the control
                          volume prior  to its becoming saturated,
                    P = dimensionless  coefficient relating the rate of decrease
                          in  infiltration rate with increasing soil moisture
                          content.


This form uses the soil water content, rather than time, as the independent variable.
                                           10

-------
During periods of zero rainfall rate, any infiltration which occurs must be supplied by water stored
as either retention or detention. Since the surface of an element is seldom entirely inundated, the
computed infiltration capacity is reduced in direct proportion to the percent of the soil surface not
submerged. Thus, the specified surface micro-relief parameter  also can reduce the infiltration
capacity during these recession periods.

According to Holtan's conceptualization of the infiltration process, a "control zone" depth of soil
determines the infiltration rate at the  surface.  He defined the depth of this control zone as the
shallower of the depth to an impeding soil  layer or that required for the hydraulic gradient to reach
unity. Extending this same concept somewhat,  the ANSWERS model maintains an accounting of
water that leaves this control zone.

Holtan's equation requires six infiltration parameters to be specified fpr a given soil type: total
porosity, field  capacity, depth of the control zone, steady-state infiltration rate (FC), and the two
unsteady-state coefficients (A and P). Data from both  large, plot-sized simulated rainfall  tests
(Skaggs, et al., 1969) and field rainulator tests conducted in cooperation with USDA-ARS as a part of
the Black Creek Project have been used to estimate parameter values. All of these tests indicated
that surface crusting conditions have a major impact on the  observed infiltration relationship.
Experience with Holtan's equation has indicated the influence  of  crusting can be modeled by
adjustment of the specified depth of the control zone. Crusting requires the use of a much shallower
control zone.

The rate of water movement from the control zone is a function of the moisture content of that zone.
The two conditions which can exist are handled according the following rules:
     1. when the moisture content of the control zone is less than field  capacity, no water
       moves from this zone,
     2. when the control zone moisture exceeds field capacity, the water moves  from this
       zone according to the equation:


                                                    PIV 3                             (3-5)
                                    np  = pr  * M  _  r •*•'
                                    UK  hi,   U
         where:   DR =  drainage  rate  of  water from control  zone,
                 GWC =  gravitational  water capacity of the  control  zone
                          (total  porosity minus field capacity).


This relationship satisfies the continuity requirement that at saturation, when PIV=0, the drainage
rate from the control zone equals the steady state infiltration  rate, FC. In addition, it exhibits
intuitively desirable  properties of rapidly decreasing moisture movement as the soil dries from
saturation and of asymptotically approaching a zero drainage rate.

Water leaving the control zone contributes to tile drainage, if the element is tiled, or to baseflow. In
both cases the water is assumed to re-emerge into the channel segments. Water moves from the
infiltration control zone into the "pools" available for tile and/or baseflow at a rate equal to DRA.

Individual elements may selectively be designated as being tile drained. In addition to water coming
from the control zone, tile inflow may be occurring from adjacent tiled elements. The sum of these
two rates constitutes the rate of subsurface inflow into an element. Subsurface water moves out the
element's tile at this inflow rate up to a maximum outflow rate equal to the tile drainage coefficient.
Whenever the rate of subsurface inflow to an element exceeds its drainage coefficient, that excess
water is diverted to baseflow storage. Elements which  are not tiled have a drainage coefficient of
zero.


                                            11

-------
Subsurface water entering an element at rates in excess 9f its drainage coefficient is stored in a
single "pool." To simulate baseflow it is released directly into channel segments at a rate propor-
tional to the volume of water in storage. Water is released at an equal rate to each channel segment.
For small catchments having no defined channels, only overland flow will appear at the outlet.


                    Sediment Detachment and Movement
Soil erosion, as it relates to nonpoint source pollution, can be viewed as two separate processes,
detachment of particles from the soil mass and transport of these particles into the streams and
lakes. Detachment of either primary soil particles or aggregates can result from either rainfall or
flowing water. These same factors can cause detached particles to be transported to the water
supply network. Thus, there are four processes for which quantifying relationships must be deve-
loped, as shown in Figure 3-2. Two different transport models are available for use in ANSWERS. The
simpler, less descriptive transport relationship is presented in this chapter. A more detailed trans-
port model, which predicts particle size distributions and  differential deposition, is presented in
Appendix D.
                                                 TRANSPORT  BY  FLOW
                                                          DETACHMENT  BY
                                                          RAINDROP  IMPACT
                                 DETACHMENT
                                 BY  FLOW
                                       6
                      Figure 3-2. Sediment Detachment and Transport.


The detachment of soil particles by water is accomplished by two processes. The first involves
dislodging soil as a result of the kinetic energy of rainfall. Rainfall is the major detachment process
on relatively flat watersheds. The second process involves the separation of particles from the soil
mass by shear and lift forces generated by overland flow.

Detachment of soil particles by raindrop impact is calculated using the relationship described by
Meyer and Wischmeier (1969):
                             DETR  =  .108 * CDR * SKDR * A. *
                                                                                  (3-6)
         where:  DETR = rainfall  detachment  rate, kg/min,
                  CDR = cropping  and  management  factor, C (from Universal Soil
                          Loss Equation  -  Wischmeier and Smith, 1978),
                 SKDR = soil  erosivity factor, K  (from Universal Soil Loss
                          Equation),     r.
                   A. = area  increment,  m  ,
                    R = rainfall  intensity during a time interval, mm/min.
                                           12

-------
The detachment of soil particles by overland flow was described by Meyer and Wischmeier (1969)
and modified by Foster (1976) as follows:
                           DETF =  .90 * CDR * SKDR * A. * SL * Q
                                                                                   (3-7)
         where:  DETF = overland flow detachment rate,  kg/min,
                   SL = slope steepness,
                    Q = flow rate per unit width, m/min.
Once a soil particle has been detached, sufficient energy must be available to transport it or the
particle will be deposited. The transport of sediment by overland flow is self-regulating, i.e., soil
particle detachment by overland flow does not occur unless there is excess energy available in
addition to the amount required to transport suspended sediments. However, detachment by rainfall
impact often occurs when there is little or no flow available for transport.

After a literature study which included Yalin (1963), Meyer and Wischmeier (1969), Foster and Meyer
(1972), and Curtis (1976), as well as an inspection of soils data, a relationship for particle transport in
overland flow was chosen as shown in Figure 3-3. The two portions of the curve generally represent
the laminar and turbulent flow regions. Obviously, the transport capacity does  not continue to
increase as the square of f jow forever. However, within the range of flows generally encountered in
ANSWERS simulations, this generalized relationship (based in part on Yalin's work and in part on
observed data) has produced reasonable results. Equations and their  region of application are:


                         TF = 161 * SL * Q'5     if Q <_ .046 m2/min               (3-8a)



                         TF = 16,320 * SL * Q2   if Q > .046 m2/min               (3-8b)
         where:  TF = potential transport rate of sediment, kg/min-m.
The erosion portion of the ANSWERS model was simplified further by the following assumptions:
    1.  Subsurface or tile drainage produces no sediment. (Data indicate around two percent
       of the average annual loading  originated from  subsurface systems on the Black
       Creek Watershed).
    2.  Sediment detached at one point and deposited at another is reattached to the soil
       surface.
    3.  Re-detachment of sediment requires the same amount of energy as required for the
       original detachment.
    4.  For channel segments rainfall detachment is assumed to be zero and only deposited
       sediment is made available for flow detachment, i.e., original channel linings are not
       erodible.
Although these assumptions were made primarily to reduce the computational cost of using the
model, some were also required because little or no data were available in the literature to quantify
the particular process. Also, after consideration of the relative magnitude of the four detachment/-
transport processes, the transport of soil particles by rainfall  was assumed  negligible.
                                           13

-------
                                        TURBULENT FLOW
                                                       2
10
                      q = m /min
  Figure 3-3. Transport Relationship Used in the ANSWERS Model.
                         14

-------
Combining the above equations and assumptions gives a composite soil movement model wherein
soil particles are dislodged from the soil mass by both rainfall and flowing water. Detached solids
then become available for transport by overland flow. Within an element the material available for
transport is the combination of that detached within the element and that which enters with inflow
from adjacent elements.

Once the available detached  sediment within an element is known, the transport  capacity is
computed. If it is insufficient to carry the available material, the excess is deposited in the element.
The overall accounting relationship for this pmcess is the differential form of the continuity equation
as applied ab9ve to water flow. Sediment carried out of an element is apportioned between adjacent
elements in direct relation to overland flow.

                         Best  Management Practices
Land use changes, tillage techniques and management procedures which qualify as Best Manage-
ment Practices (BMPs) for controlling nonpoint source pollution are simulated with ANSWERS by
using appropriate parameter values for the component relationships discussed above. For example,
conservation tillage generally results in a rougher surface, reduced C-factor and increased infiltra-
tion. Gully stabilization structures such a drop spillways or chutes may be simulated by reducing the
slope steepness of the associated channel segments. Certain structural BMPs cannot be adequately
accommodated with these component relationships. Currently, four specific BMPs which require
special computational  provision have been included:  ponds, parallel tile-outlet terraces, grass
waterways and field borders.

The four special structural BMPs are dealt with by a subroutine named STRUCT. The programming
philosophy for STRUCT was to have it adjust "standard" parameter values for those elements which
contain a structural BMP. These modifications  are made when subroutine DATA is called. This
approach permits simulating the  important impacts of the BMP without major modification to the
fundamental computational logic of the entire computer program.

Both ponds and PTOs are handled in a similar manner using a trap efficiency concept.  Sediment
trapped from the water flowing into a pond or PTO is diverted into a special, psuedp element which
provides a means of tabulating the combined effectiveness of all such BMPs. Water is 3)59 assumed
to be diverted, in the same ratio as sediment is trapped, into the tile drainage system. In this manner,
effects of both reduced  sediment loads and downstream overland flow rates are simulated.

Grass waterways and field border strips are also treated similarly to one another. It is assumed that
the  vegetated area within the affected element is no longer subject to any sediment detachment.
Computationally, this is accomplished by adjusting the specified  slope steepness  by an amount
which produces the desired change in sediment detachment rate for the element. Deposition within
the vegetation of a grass waterway is deliberately prohibited, since any waterway that effectively
traps sediment would soon fill and  become ineffective. Specifying that an element  has a grassed
waterway forces the presence of a shadow channel element if none was already present.

                            Programming Overview
ANSWERS is programmed with a  MAIN section which includes all the logic to simulate the various
component processes each one  of which is  structured as an independent subroutine. Prior to
beginning any simulation calculations, which follow the descriptions given above, MAIN calls a
large subroutine named DATA.

It is the responsibility of DATA to read the user supplied data file and convert that parametric and
geographic information into a structure which maximizes the computational efficiency of the
lengthy  simulation calculations. Specifically, it  converts the gridded  watershed into a  one-
dimensional array structure.  This  is done to avoid wasting machine memory with incomplete
two-dimensional arrays that  would result from irregularly shaped watersheds. Computational
                                         15

-------
speed  is also significantly improved  by this process.  The process of transforming the two-
dimensional layout of a watershed elemental map into one-dimensional computational arrays is
primarily accomplished by subroutine RELEM.

Subroutine RELEM is designed to utilize three adjacent rows of elements in a watershed file to
determine position sequence numbers for all elements. The three-row system is utilized in order to be
able to account for discontinuous rows of elements. The subroutine operates in such a manner that it
can detect whenever the specified stope direction of an element would direct any portion of the
runoff from that element into a non-existent element, i.e., out of the catchment. A warning message
is printed for each offending element.

Specifically, subroutine RELEM returns to subroutine DATA a 3-dimensional  array, IEL. The first
subscript refers to the row's position in the 3-row array. Eventually, each row of watershed elements
"ripple through" all three row positions in IEL. The second subscript in IEL refers to the element's
watershed column number. The third subscript corresponds to the parameter number for that
watershed element according to:
    1.  — contains the element's watershed row number,
    2.  — contains its column, except for IEL(i,1,2) which contains the column no. of the last
         element in a given row,
    3.  — contains the position sequence number for the element,
    4.  — contains the element's slope steepness,
    5.  — contains the flow direction for the element,
    6.  — contains the channel designator and soil type,
    7.  — contains the crop type,
    8.  — contains the rain gauge name,
    9.  — contains the tile flow designator,
   10.  — contains the slope steepness of any shadow channel element,
   11.  — contains the identifier number for a structural BMP,
   12.  — contains size information about any structural BMP,
   13.  — contains size information about any structural BMP.

Array ITEMP is used  to temporarily hold parameter values for the  most recently read element.
Subroutine RELEM must be entered with array ITEMP loaded with parameter values for the very first
element of the catchment. The first task of RELEM is to transfer row-2 values for array IEL into row-1
and row-3 values into row-2. Once entered, execution continues within RELEM until an entire row of
watershed elemental parameters has been read and stored in row-3 of IEL.

Upon return to subroutine DATA from RELEM, DATA works with row-2 of IEL to transfer its parameter
data into the single dimension, sequential arrays that will ultimately be used for simulation
computations. After an entire watershed row has been so set up, it again returns to subroutine
RELEM which "ripples" the data in array IEL so it is ready to accept the next row of watershed
elemental parameter values.

After all watershed elements have been set up by RELEM and DATA, the shadow channel elements
that have  been detected are moved  to  occur sequentially at the end  of the regular elements.
Parameter values for each element are next combined into computationally efficient coefficients and
control is returned to MAIN for the actual simulation.
                                          16

-------
      CHAPTER IV: DATA PREPARATION FOR ANSWERS
                                      General
The data file used by the ANSWERS model provides a detailed description of the watershed
topography, drainage networks, soils, land uses, and BMPs. Most of the information can be readjly
gleaned from USDA-SCS Soil Surveys and land use and cropping surveys or summaries. Also, aerial
photographs of the area, USGS topographic maps, and BMP construction or implementation data is
quite useful in developing descriptions of actual watershed  areas. This chapter will detail the
techniques used in putting together the data files necessary for running ANSWERS. In addition,
information describing various surface conditions, soil responses, and BMP data is included in the
Appendices. The formats shown in this section are for the current release of the ANSWERS program.
The expanded sediment detachment/transport  model, detailed in Appendix D, utilizes additional
information which is described in that section.

Input information for the ANSWERS model contains six general types of data:
    1.  Simulation requirements (measurement units and output control),
    2.  Rainfall information (times and intensities),
    3.  Soils information (antecedent moisture, infiltration, drainage response and potential
       erodibility),
    4.  Land use and surface  information (crop  type, surface roughness and  storage
       characteristics),
    5.  Channel descriptions (width and roughness),
    6.  Individual element information (location, topography, drainage, soils, land use and
       BMPs).

The individual element information is the largest body of data and the  most time consuming to
collect. However, once the topography, soils, land use and drainage patterns have been determined
for all of the elements, changes in watershed management or BMPs can be added very easily without
having to totally reconstruct the input file.

Figure 4-1 shows the configuration of a typical ANSWERS data file. Each of the six data areas listed
above are noted and will be covered individually in succeeding sections. The ANSWERS data file was
designed to be self explanatory. The information contained in the  soils, land use, and individual
element information sections are physically measurable and can be checked for validity without
having to go through a complicated process of differentiating one or more lumped parameters.


                            Data File Construction
The configuration oi an ANSWERS input data file allows the file to be constructed in two parts. All
data except for the individual element data can be contained in a separate file. This first or "predata"
file contains all of the general information necessary to describe the various soils, Jand uses, and
management systems in a given county, planning region, state, etc. This "predata" file can be used
with numerous elemental data files which describe greatly differing watersheds. Therefore, once a
"predata" file has  been constructed for a given area, subsequent  simulations, even on  different
watersheds, may be possible with very little or no additional general information collection.

Simulation requirements. Figure 4-2 depicts the portion of the ANSWERS input file that supplies
data on simulation requirements. The first line of this two-line portion of the "predata" file contains
an alpha-numeric header used to describe which "predata" file is being used. All 80 columns of the
card are used, and this information will be reprinted at execution time. Metric or English units are
acceptable for input/output. The system of units used must be specified in columns 2-8 (left justified)
on the second card. Inches and feet are both used when ENGLISH units are specified. Millimeters and
                                          17

-------
STANDARD PREDATA FILE FOR
ALLEN CO.
, INDIANA— 800823 1 •!
ENGLISH UNITS ARE USED ON INPUT/OUTPUT
PRINT 1 '

RAINFALL DATA FOR 2 GAUGE(S)
GAUGE NUMBER Rl
0 0. 0.00
0 9. .52
0 15. 1.55
0 20. 2.40
0 30. 1.59
0 35. .85
0 45. .50
1 300. 0.00
GAUGE NUMBER R2
0 0. 0.00
0 7. .45
0 14. 1.25
0 18. 2.66
0 25. 1.65
0 33. .60
0 42. .35
1 300. 0.00
FOR EVENT


















OF: — TEST— -












































2































SOIL INFILTRATION, DRAINAGE
NUMBER OF SOILS = 8
S 1, TP =.46, FP =.75, PC =
S 2, TP =.46, FP =.65, PC =
S 3, TP =.46, FP =.70, FC =
S 4, TP =.42, FP =.70, FC =
S 5, TP =.44, FP =.80, FC =
S 6, TP =.46, FP =.75, FC =
S 7, TP =.46, FP =.75, PC =
S 8, TP =.35, FP =.65, PC =
AND GRDUNDWATER CONSTANTS FOLLOW

.40, A =
.40, A =
.40, A =
.60, A =
.60, A =
.60, A =
.40, A =
.90, A =
DRAINAGE COEFFICIENT FOR TILE DRAINS =
GRDUNDWATER RELEASE FRACTION
.005

.80, P
.80, P
.80, P
1.0, P
1.0, P
1.0, P
.80, P
1.6, P

=.65, DF = 4.0, ASM =.70
=.65, DF = 3.0, ASM =.70
=.75, DF = 3.0, ASM =.70
=.65, DF = 4.0, ASM =.70
=.65, DF = 5.0, ASM =.70
=.65, DF = 5.0, ASM =.70
=.65, DF = 3.0, ASM =.70
=.60, DF = 6.0, ASM =.70

K =.36
K =.32
K =.17
K =.36
K =.32
, K =.36
K =.38
K =.35
0.25 IN/24HR



SURFACE ROUGHNESS AND CROP CONSTANTS
NUMBER OF CROPS AND SURFACES
C 1, CRDP= SI CORN, PIT=.01,
C 2, CROP= CORN-NT, PIT=.06,
C 3, CHOP=BEANS TP, PIT=.01,
C 4, CROP=S. GRAINS, PIT=.04,
C 5, CROP= PASTURE, PIT=.03,
CIO, CROP= WOODS , PIT=.10,
= 6
PER=0.0,
PER=. 75,
PER=.45,
PER=.90,
PER=1.0,
PER=.90,
FOLLOW

RC=.47,
RC=.55,
RC=.47,
RC=.55,
RC=.40,
RO.55,


HU= 2.0, N=.075, C=.50
HU= 3.5, N=.150, C=.30
HU= 1.5, N=.070, C=.60
HU= 2.0, N=.150, C=.15
HU= 1.5, N=.200, C=.04
HU= 3.5, N=.250, C=.15




4



CHANNEL SPECIFICATIONS FOLLOW
NUMBER OF TYPES OF CHANNELS = 4,
CHANNEL 1 WIDTH =15.0 FT, ROUGHNESS COEFF. (N) = .035"
CHANNEL 2 WIDTH =10.0 FT, ROUGHNESS COEFF. (N) = .040
CHANNEL 3 WIDTH = 7.5 FT, ROUGHNESS COEFF. (N) = .045
CHANNEL 4 WIDTH = 5.0 FT, ROUGHNESS COEFF. (N) = .070
ELEMENT SPECIFICATIONS FOR MIDDLETOWN WATERSHED
EACH ELEMENT IS 528.0ft SQUARE
' OUTFLOW FROM ROW 21 COLUMN 3
1 15 4 270 1 1 Rl 0
1
7
7
7
8
8
8
8
20
20
21
21
22
22
16
20
22
23
11
12
13
14
19
20
3
4
13
14
5
9
2
3
20
17
16
12
8
8
9
28
2
9 5
259
186
135
180
270
180
180
148
180
180
270
180
0
90
3
2
1
3
201
302
303
1
1
"l
102
108
3
1
1
1
2
5
2
2
1
1
2
2
1
1
4
4
Rl
Rl
Rl
Rl
R2
R2
R2
R2
R2
R2
R2
R2
R2
R2

•

TILE
TILE
TILE
TILE
TILE
TILE
TILE
TILE
TILE


0
0
0
0
5
5
5
0
0
0
3
3
0
0

2


4 32
4 20
4 20
1
3 20

4 40
4 40


5
827.8
828.8
822.8
835.3
835.3
799.7
799.3
802.2
809.2
802.0
803.0
766.5
762.0
799.5
798.0
Figure 4-1. Typical ANSWERS Input Data File.
                   18

-------
meters are used when METRIC units are chosen. The presence of the word PRINT in the second line
of the "predata" file in columns 58-62 causes ANSWERS to produce a detailed listing of all predata
information. The absence of the word PRINT on line 2 suppresses this output.
                       STANDARD EKEDATA FILE FOR ALLEN CXI., INDIANA—800823
                     ENGLISH UNITS ARE USED ON INPUT/OUTPUT             PRINT
                           Figure 4-2. Simulation Requirements.

Rainfall information. Figure 4-3 shows the portion of the "predata" file dedicated to the descrip-
tion of the precipitation event to be simulated. The first line must contain the word RAINFALL
beginning in column 2. The number of gauges used in the simulation are inserted in column 20.
ANSWERS permits up to four rain gauges to be used.  Beginning in column 47, an eight character
identifier of the particular storm being modeled is inserted. Each  new rain gauge file contains a two
character identifier for the gauge in columns 17-18.

The Thiessen polygonal method  is used to determine which areas of a watershed are affected by
specific rain gauges. For hypothetical situations, the various gauges can be used,to simulate the
movement of a storm across the watershed.

The three columns contain the following data:
    1.  Data column 1 - last entry? (indicated by 1),
    2.  Data columns 3-10 - time  (in minutes),
    3.  Data  columns 11-20  - precipitation intensity (in inches/hour or millimeters/hour,
       depending on units chosen) which ended at corresponding time (Column 2).
in this example, the intensity was 0.52 iph from 0 to 9 minutes and 1.55 iph from 9 to 15 minutes for
gauge R1. During a similar time interval, the rates were 0.45 iph from 0 to 7 minutes and 1.25 iph from
7 to 14 minutes  at gauge R2. The last entry for both gauges, at 300 minutes, indicates the end of
simulation us, although the precipitation ended at approximately 45 minutes, ANSWERS would
continue simulating the watershed response until the time was equal to or greater than 300 minutes.
RAINFALL DATA
GAUGE NUMBER
0 0.
0 9.
0 15.
0 20.
0 30.
0 35.
0 45.
1 300.
GAUGE NUMBER
0 0.
0 7.
0 14.
0 18.
0 25.
0 33.
0 42.
1 300.
FOR 2 GAUGE(S) FOR EVENT OF: — TEST—
Rl
0.00
.52
1.55
2.40
1.59
.85
.50
O.OO
R2
0.00
.45
1.25
2.66
1.65
.60
.35
0.00
                              Figure 4-3. Rainfall Information.

Soils information. Figure 4-4 lists the information necessary to describe the response of the
various soils to precipitation inputs. The word SOIL must be found in columns 3-6 of the first data
card. The number of identified soil groups or types is entered in columns 19-22 (right justified) of the
second card. ANSWERS is normally dimensioned to handle up to 20 different soils. If additional soils
are present, the model can be easily set up to accept as many soils as are needed.
                                           19

-------
SOIL INFILTRATION, DRAINAGE AND GROUNDWATER CONSTANTS FOLLOW
NUMBER OF SOILS = 8
S 1, TP
S 2, TP
S 3, TP
S 4, TP
S 5, TP
S 6, TP
S 7, TP
S 8, TP
=.46,
=.46,
=.46,
=.42,
=.44,
=.46,
=.46,
=.35,
FP =.75,
FP =.65,
FP =.70,
FP =.70,
FP =.80,
FP =.75,
FP =.75,
FP =.65,
FC = .
FC = .
FC = .
PC = .
FC = .
FC = .
FC = .
FC = .
DRAINAGE COEFFICIENT FOR TILE
GROUNDWATER RELEASE FRACTION =
40, A =
40, A =
40, A =
60, A =
60, A =
60, A =
40, A =
90, A =
DRAINS =
.005
.80,
.80,
.80,
1.0,
1.0,
1.0,
.80,
1.6,
0.25

P =.65,
P =.65,
P =.75,
P =.65,
P =.65,
P =.65,
P =.65,
P =.60,
IN/24HR

DF =
DF =
DF =
DF =
DF =
DF =
DF =
DF =


4.0,
3.0,
3.0,
4.0,
5.0,
5.0,
3.0,
6.0,


ASM
ASM
ASM
ASM
ASM
ASM
ASM
ASM


=.70,
=.70,
=.70,
=.70,
=.70,
=.70,
=.70,
=.70,


K =.36
K =.32
K =.17
K =.36
K =.32
K =.36
K =.38
K =.35


                                Figure 4-4. Soils Information.

The infiltration descriptors, antecedent moisture content, and the potential credibility of the various
soils are listed in this section. Table4-1 describes each parameter and gives its location on the data
card. Methods for determining the values of TP, FP, FC, A, P, DF, and ASM are presented in Appendix
A. The K parameter is identical to the K parameter in the Universal Soil Loss Equation. Because the
soil parameters may be used to describe the response of several similar soil types, the listed K value
is usually an effective or average value.

The next two user-selectable constants are used to  help describe the contribution of subsurface
drainage to the total water yield in the watershed. The tile drainage coefficient indicates the design
coefficient (inches/day or millimeters/day) of tile drains in those areas designated as having tile
drainage. Generally, this value is 0.25 to 0.5 inches/day (6.4 to 12.7 millimeters/day). The ground-
water release fraction is a measure of the contribution of lateral groundwater movement or interflow
to total  runoff. Appendix A contains information to be used in the selection of a release fraction.

The tile drainage coefficient is contained in columns 40-44 (right justified). The groundwater release
fraction is contained in columns 32-41  (right justified).

                        Table 4-1. Soils Information File Construction.
      Parameter
Definition
Column(s)   Appendix
TP
FP
FC
A
P
DF
ASM
K
Total porosity (percent volume)
Field capacity (percent saturation)
Steady state infiltration rate
Difference between steady state and
maximum infiltration rate
Exponent in infiltration equation
Infiltration control zone depth
Antecedent soil moisture (percent saturation)
USLE "K"
11-13
20-22
29-33
39-43
49-51
58-62
70-72
78-80
A.I
A.I
A. 2
A. 2
A. 2
A.I
A.I
A. 3
      All  entries  are right justified.
                                            20

-------
Land use and surface information. The layout of the land use and surface description portion of
the "predata" file is given in Figure 4-5. Line one should C9ntain the word SURFACE in columns 3-9.
The number of different land uses defined in this section is contained in columns 32-34 (right
justified) of the second card. ANSWERS is normally set up to handle up to 20 different combinations
of land use, management, and surface conditions. The model can be easily re-dimensioned to accept
more entries, if necessary.

The particular land use and, in some cases, the management system used are noted in the crop
section. The potential interception (PIT), percentage cover (PER), surface storage (RC and HU),
surface roughness (N), and relative erosiveness (C) parameters are listed and described in Table4-2.
Methods for determining the values of these descriptors are given in Appendix B. The C value used in
this section is a seasonally-adjusted combination of the C and P factors used in the Universal Soil
Loss Equation.
                     SURFACE ROUGHNESS AND CHOP CONSTANTS FOUCW
                    NIMBER OF CROPS AND SURFACES =  6
                    C 1, CBOP= SI CORN, PIT=.01, PER=0.0, RC=.47, HU= 2.0, N=.075, C=.50
                    C 2, CROP= CORN-NT, PIT=.06, PER=. 75, RO.55, HU= 3.5, N=.150, C=.30
                    C 3, CROP=*EANS TP, PIT=.01, PER=.45, RC=.47, HU= 1.5, N=.070, C=.60
                    C 4, CROP=S.GRAINS, PIT=.04, EER=.90, RC=.55, HU= 2.0, N=.150, C=.15
                    C 5, CROP= PASTURE, PIT=.03, PER=1.0, RC=.40, HU= 1.5, N=.200, O.04
                    CIO, CROP= WOODS , PIT=.10, PER=.90, RC=.55, HU= 3.5, N=.250, O.15
                        Figure 4-5. Land Use and Surface Information.


                 Table 4-2. Land Use and Surface Condition File Construction.



        Parameter                  Definition                  Column(s)   Appendix
CROP
PIT
PER
Specific land use and management
Potential interception volume
Percentage of surface covered by
12-19
26-28

B.I
B.I

                     specific land use (CROP)                    35-37        B.I

            RC       Roughness coefficient (a shape  factor)      43-45        B.2

            HU       Maximum roughness height                    51-54        B.2

            N        Manning's n                                 59-62        B.I

            C        Relative erosiveness of a particular
                     land  use (function of time and
                     USLE  "C" and "P")                           67-69        B.I


        All entries  are right justified.


Channel  descriptions. The part of the  "predata" file used to describe the channels in the
watershed is shown in Figure 4-6. The first line must contain the word CHANNEL in columns 3-9. The
number of different channel configurations (width or roughness) is listed in columns 31-33 (right
justified) on the second line. The  assumption is made that the  channel is rectangular in cross-
                                            21

-------
section. The width of the channel is noted in columns 19-22 and the value of Manning's n is listed in
columns 50-54 (both right justified). The channel is also assumed to be sufficiently deep to handle the
runoff. ANSWERS does not predict flooding, nor does it presently use variable geometry in the
channel cross-section.
                          CHANNEL SPECIFICATIONS FOLLOW
                         NUMBER OF,TYPES OF CHANNELS = 4,- "" '
                         CHANNEL 1 Wl*±H =15"6 FT, ROUSHNEltS*COEFF. (N) = .035
                         CHANNEL 2 WIOTH =10.0 FT, ROUGHNESS COEFF. (N) = .040
                         CHANNEL 3 WIDTH = 7.5 FT, TOUGHNESS COEFF. (N) = .045
                         CHANNEL 4 WIDTH = 5.0 FT, ROUGHNESS COEFF. (N) = .070
                              Figure 4-6. Channel Descriptions.

Individual element information. This file allows the user to describe, in detail, the particular
watershed and its special characteristics. Information on topography, soils, land use, drainage and
BMPs is included in this section. Figure 4-7 shows the configuration of the individual element
information file.
ELEMENT SPECIFICATIONS FOR MIDDLETOWN WATERSHED
EACH ELEMENT IS 528.0ft SQUARE
OUTFLOW FROM ROW 21 COLUMN 3
1
1
7
7
7
8
8
8
8
20
20
21
21
22
22
15
16
20
22
23
11
12
13
14
19
20
3
4
13
14
4
5
9
2
3
20
17
16
12
8
8
9
28
2
9 5
270
259
186
135
180
270
180
180
148
180
180
270
180
0
90
1
3
2
1
3
201
302
303
1
1
1
102
108
3
1
1
1
1
2
5
2
2
1
1
2
2
1
1
4
4
Rl
Rl
Rl
Rl
Rl
R2
R2
R2
R2
R2
R2
R2
R2
R2
B2




TILE
TILE
TILE
TILE
TILE
TILE
TILE
TILE
TILE
•

0
0
0
0
0
5
5
5
0
0
0
3
3
0
0


2


4 32
4 20
4 20
1
3 20

4 40
4 40


827
828
822
835
835
'799
799
802
809
802
803
766
762
799
798
.8
.8
.8
.3
.3
.7
.3
.2
.2
.0
.0
.5
.0
.5
.0
                         Figure 4-7. Individual Element Information.

The top three lines of this file are header informatipn. The first card contains the watershed name or
identifier in columns 29-72. The size of the element is shown in columns 17-22 on the second card. The
third card contains the row and column coordinates for the outlet element in columns 18-21 (row) and
30-33 (column). All information is right justified.

Each subsequent line of this file contains the information necessary to characterize the physical
configuration of each element. The position and definition pf all of the parameters used in describing
each element are shown in Table 4-3. Appendix C describes various  methods for setting up the
elemental data  file.
                                             22

-------
              Table 4-3. Element Information File Construction.
              Parameter Definition                 Column(s)    Appendix


Row number of element                                  1-3

Column number of element                               4-6

Last element flag, this field should be blank
except for the last watershed element;  that
element should have a value of 9 entered here          7-8

Slope steepness of the element, entered in
tenths of a percent, e.g., an element with a
slope of 2.9 percent would be entered as 29           9-11       C.I

Direction of steepest slope, entered in
degrees counterclockwise referenced to  a
horizontal line from the center of
the element and directed to the right                12-15       C.I

Channel size category, if this element  has a
well defined channel flowing through it, other-
wise these columns should be blank                   16-17

Soil type number                                     18-19        A

Crop/management type number                          20-23        B

Rain gauge designator (alpha-numeric)                27-28

Tile flag; presence of the letters "TI" indicate
tile drainage, anything else indicates  no tile       30-31

Channel slope steepness in tenths of a  percent;
when no value is present and a channel  exists,
the slope will be taken to be equal to  the over-
land slope                                           34-37

BMP identification number                            39-40       C.2

First BMP descriptor                                 41-44       C.2

Second BMP descriptor                                45-48       C.2

Mean elevation of element (optional)                 64-70       C.2


All entries are right justified.
                                  23

-------
                   CHAPTER V: ANSWERS OUTPUT
                                      General
By using a very descriptive data file and the distributed parameter concept, the ANSWERS model is
capable of producing a detailed accounting of the erosion and hydrologic response of a watershed
subjected to a precipitation event. Figure 5-1 displays a condensed version of the typical ANSWERS
output. The output listing consists of five basic sections:
    1.  An "echo" of the input data (can be suppressed by removing PRINT parameter in line
       2 of input data),
    2.  Watershed characteristics (calculated from elemental data),
    3.  Flow and sediment information at the watershed outlet and effectiveness of structu-
       ral BMPs,
    4.  Net transported sediment yield or deposition for each element,
    5.  Channel deposition.
The various tabular output sections of the ANSWERS  model can be used with  several plotting
programs to yield graphical depictions of the watershed's hydrologic and erosion response.

This chapter will describe the five sections of the ANSWERS tabular output as  listed above. In
addition, examples of graphical representation of the output (to lend spatial significance) will be
presented.

Input information "echo". This portion of the ANSWERS, when enabled with the PRINT parameter
in line 2 of the input file, essentially reprints the "predata" file. It can be quite useful in tracking down
problems with  input formats, since the output will be exactly what it read.

Watershed characteristics. Included in this section is the header information from the elemental
data file. The size of the elements is calculated from the side length and the area of the watershed is
displayed, based on the total number of elements. Also, a count of the number of channel elements is
made to give some feeling for drainage density. Information on the minimum, average and maximum
slopes for both overland flow and channels is collected. Information on subsurface drainage, mean
antecedent moisture and outlet element location is given.

The second section of this portion of the output contains a short synopsis of each of the land uses
and soils that are actually present in the watershed. Percentage of area occupied as well as several
pertinent descriptors for each soil or land use are displayed.

The final section of the watershed characteristics involves the number of elements containing those
BMPs which involve structures or land use changes (identified in elemental data file). If no BMPs of
this type are present, this section will not appear.

Flow  and  sediment information and BMP  effectiveness. This portion of  the output  most
nearly resembles the "normal" output from most watershed  models. A flow hydrograph, along with
associated sediment concentration and accumulative yield, is displayed. The time interval between
print points is determined by the number of hydrograph print points (set internally to 101). This
parameter can be changed easily.

An accounting of total rainfall, total flow and average sediment yield is presented at the end of the
hydrograph. Also, the amount of sediment trapped by each structural BMP (if any) is noted.

Although the only hydrograph printed is that of the outlet element, it is possible to access the output
 information from any element(s) within the watershed. A rather minor modification to the output
 section of the ANSWERS model is all that is required to produce a  hydrograph and associated
sediment information for any point in the basin.
                                          24

-------
DISTRIBUTED HYDROLOGIC AND WATER QUALITY SIMULATION
               BY ANSWERS UER 4.800326
  STANDARD PREDATA FILE FOR  ALLEN CO.,  INDIANA—800423
    RAINFALL HYETOGRAPH FOR EUENT OF  B-DIST
    GAGE NUMBER GA
    TIME - MIN.
          0.0
          9.0
         18.0
         23.0
         29.7
         37.6
         45.0
         54.0
         63.0
         72=0
         90.0
        350.0
              RAINFALL RATE - 1N./HR.
                         0.00
                         0.52
                         0.76
                         0.92
                         1.59
                         5.52
                          .63
                         1.15
                         0.83
                         0.72
                         0.60
                         0.00
SOIL PROPERTIES
SOIL  POROSITY  FIELD CAP.  INFILTRATION CONSTANTS  CONTROL  ANTECEDENT EROSION
      (PERCENT   (PERCENT
  1
  2
  3
  4
  5
  6
  7
  8
UOL.)
46.0
46.0
46.0
42.0
44.0
46.0
46.0
35.0
                   SAT.)
                   75.0
                   65.0
                   70.0
                   70.0
                   80.0
                   75.0
                   75.0
                   65.0
FC
IN./HR.
0.4
0.4
0.4
0.6
0.6
0.6
0.4
0.9
ft
IIS./HR
0.8
0.3
0.3
1.0
1.0
1.0
0.8
1.6
0.65
0.65
0.75
0.65
0.65
0.65
0.65
0.60
ZONE
 IN.
  4.0
  3.0
  3.0
  4.0
  5.0
  5.0
  3.0
  6.0
TILE DRAINAGE COEFF.  = 0.25 IN./24H
GROUNBWATER RELEASE FRACTION = 0.500E-03

COUER/MANAGEMENT PRACTICES
CROP MAX. POT.
INTERCEPTION

1 FALLOW
2 51 CORN
3 S2 CORN
4 CORN-NT
5 BEANS TP
6 BEANS FC
7 S. GRAINS
8 PASTURE
9 WOOES
11 HOMESITE
IN.
0.00
0.02
0.04
0.06
0.03
0.03
C . 04
o.os
0.10
0.02
PERCENT
CQUER

0.
20.
60.
85.
75,
75.
95.
100.
90.
100.
ROUGH
COEFF

0.47
0.47
0.55
0.55
0.47
0.53
0.55
g.CO
0.55
0.40
                                            .8
                                            ,4
                                 ROUGH.
                                 HEIGHT
                                   IN.
                                   2,5
                                   2.
                                   2.
                                   4.
                                   2.
                                   2.G
                                   3.0
                                   0.4
                                   5.1
                                   2.4
                                                 MANNINGS
                                                     N
                                                   0.100
                                                   0.075
                                                   0.070
                                                   0.150
                                                   0.060
                                                   O.OS5
                                                   0.090
                                                   0.100
                                                   0.200
                                                   0.090
  MOISTURE   CONST.
(PERCENT SAT)
    70.0     0.36
    70.0     0.32
    70.0     0.17
    70.0     0.36
    70.0     0.32
    70.0     0.36
    70,0     0.38
    70.0     0.35
               EROSION
                CONST.

                1.00
                0.50
                0.50
                0.30
                0.60
                0.55
                0.15
                                                               04
                                                               15
                                                             O.SO
CHANNEL PROPERTIES
TYPE   MIDTH   MANNINGS M
         FT.
  1    10.0      0.035
  2     8.0      0,040
  3     6.0      0.040
  4     5.0      0.050
  5     3.0      O.OSO

      BKUNSON ~ 1980 PRACTICES
    WftTbRSHED CHARACTERISTICS
NUMBER OF  6.40 AC.  OUEELAND FLOW ELEMENTS =
NUMBER OF CHANNEL SEGMENTS =  41
AREA OF CATCHMENT =  1644.8 AC.
CATCHMENT SLOPE:  MIN =   0.2"  AUE =   1.35
CHANNEL SLOPE:   MIN =   0.45  AUE =   1.63
PERCENT OF AREA TILED = 100.0 WITH A D.C. OF
MEAN ANTECEDENT SOIL MOISTURE = 70,, FIELD CAPACITY = 70. PERCENT SATURATION
GROUNDWATER RELEASE FRACTION = 0.0005
OUTLET IS ELEMENT  2b5 AT ROW  18 COL   2
                                              257
                                             MAX =
                                             MAX =
                                             0.25 IN./24H
                                             4.00 PERCENT
                                             2.50 PERCENT
                   Figure 5-1. Typical ANSWERS Output Listing.
                                      25

-------
      SURFACE COUER/NANAGEMENT CONDITIONS
        CROP  PERCENT PERCENT  N
                         SOIL ASSOCIATION PROPERTIES
              PRESENT
       S2 CORM  87,5
       PASTURE   0.4
        WOODS    3.3
      HOMESITE   2.7
COUER
 SO. 0.070
100. 0,100
 90. O.SOO
100. 0.090
                   MO.  PERCENT   FC
                           INITIAL  CONTROL  K
0.50
0,04
0,15
0.20
1
2
3
6
7
8
           PRESENT  IN./HR.  IN./HR.' DEPTH IN.
23.4
35.4
18,
 S.
10.
.7
.a
.5
                                               0.8
        STRUCTURAL MEASURES INCLUDED
               TYPE         NUMBER
           1  PTO TERRACES    17
           3  G. WATERUftYS    15
           4  FIELD BORDER     5
OUTLET HYDROGRftPHS — UER

TIME
S1IN.
0.0
3.5
7.0
10.5
14.0
17.5
21.0
24.5
28.0
31.5
35.0
38.5
42.0
45.5
43.0
52.5
56.0
53.5
63.0
66.5
70.0
73.5
77.0
80.5
84.0
87.5
31. 0
34.5
«
276.5
280.0
283.5
287.0
336.6
339.5
343.0
346.5
350.0

RAINFALL
IN./HR.
0.00
0.52
0.52
0.76
0.76
0.76
0.92
1,59
1.59
5.52
5.52
1.63
1.63
1.15
1.15
1.15
0.83
0.83
0.83
0.72
0.72
0.60
0.60
0.60
0.60
0.60
0,00
0,00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00

RUNOFF
IM./HR.
0.0000
0.0000
0.0000
0.0000
0.0000
0.0001
0.0001
0.0001
0.0002
0.0004
0.0019
0.0126
0.0554
0.1705
0.3071
0.3877
0.4237
0.4403
0.4479
0.4495
0.44C9
0.4411
0.4325
0.4217
0.4032
0.3357
0.3820
0 . 3663
0.0160
0.0159
0.0158
0.0157
V
0.0155
0.0155
0.0155
0.0155
0.0155
YIELD
SEDIMENT
LB.
0.
0.
0.
0.
0,
0.
0.
0.
3.
19.
139.
1247.
7200.
S6052.
63336.
114398.
172411.
233857.
295959.
360548.
423912.
486639.
548335.
608331.
667707.
724795.
780001.
833187.
1503945!
1504636.
1505257.
1505810.
1508849'
1508879.
1508901.
1508916.
15089H6.
4.300326
CONC.
SEDIMENT
PPM
0.
0.
0.
0.
0.
0.
3.
200.
1179.
3524.
6097.
8150.
8226.
7138.
6SS3.
6517.
6500.
6520.
6520.
6508.
6502.
6506.
65.15.
6527.
6538.
6547.
6551.
6560.
•
2105!
1913.
1725.
1544.
•
107!
78.
56.
39.
27.
      0.4
      0,4
0.8
0.8
4.0
3.0
                                        1.7
0.36
0.32
0.17
 .36
 .38
                                    6.0   0.35
         RUNOFF UOLUME PREDICTED  FROM    1.96 IN. OF RAINFALL =  0.625 IN.
                    AUERftGE SOIL  LOSS =   317.  LB./AC
      STRUCTURAL PRACTICE  1 REDUCED TOTAL SEDIMENT YIELD BY  56384.1. LB.

                                    Figure 5-1, cont.

 Net transported sediment yield or deposition. Statistics are kept, throughout the simulation,
on the actual amount of soil removed from or deposited on each element. Although this soil actually
leaves the element (in the case of a negative number), it may or may not reach the channel system.
This is due to the fact that deposition may occur further down slope.

Another problem with treating "hot spots" identified with this information is that the removal of one
"hot spot" may actually cause another one. If the management practice removes sediment, yet still
allows nearly the same amount of water to flow downslope, the detachment rate will increase in the
                                          26

-------
          ELEMENT SEDIMENT
            NO.
             1
             5
             3
           137
           141
           145
           149
           153
           £41
           245
           249
           253
           257
LB./AC
 -301.
 -839.
 -980.
 -109.
-2160.
 -5GB,
-1762.
-2255.
-1S58.
-1663.
-1993.
-1653.
-1884.
        MAX EROSION-RATE =
           NO.
            43
            72
            31
AMOUNT
    0.
  1S5.
  324.
INDIUIDUAL ELEMENT  SEDIMENTATION
ELEMENT SEDIMENT ELEMEI*
  NO.    LB./AC     NO.
   2        40.      3
   6      -909.      7
  10      -887.     11
 138     -1779.     139
 142     -1832.     143
 146      -183.     147
 150      2203.     151
 154        21.     155
 242      -839,     243
 246     -1S84,     247
 250     -1882,     251
 254     -2018.     255
         2948.  LB./AC
           STD.  BEU. =
IEDIMENT
LB./AC
-189.
-2214.
-1744.
e
-1072^
-2QB6.
-333.
56.
-1895..
a
-851 1
-1745.
-1137.
-1137.
ELEMENT
ISO.
4
8
12
0
140
144
148
152
156
244
248
252
256
SEDIMENT
LB./AC
-2201.
-1063.
151.
-isss!
-32S.
-1S25.
-615.
-1654.
0
-149]
-1410.
-1434.
-1333.
               MAX DEPOSITION RATE =
                893.  LB./AC
   CHANNEL DEPOSITION  —  LB.
 NO.     AMOUNT    (SO.     fi[10UNT
  57         0.     ISO         0.
  75         0.     73         0.
  32        79.     33       234.
                                                                 2203.  LE./fiC
                        NO.
                         64
                         83
                        108
                         ftMCUNT
                             0.
                           467.
                          ' 661.
           134
           206
           255
   37.
    0.
    0.
 193
 207
 Oo
27.
290
PPP
0.
0.
                                    Figure 5-1, cont.
201
E39
94.
 0.
downhill elements, since the load is less from above. In most cases, however, practices which
reduce the sediment load also reduce the runoff rate in the downslope area.

Channel deposition. This section of the output information details the amount of sediment
deposited  in each of the "shadow" or channel elements. ANSWERS does not presently predict
channel erosion. It does allow the channel flow to entrain deposited sediment if the flow regains
enough excess transport capacity to do so.

                                Graphical Output
Several plotting programs have been constructed to use the input to and the output from the
ANSWERS model  to provide visual enhancement and better understanding of the information
provided by this program. The QCKPLT program, mentioned in a previous section, uses the elemental
data portion of the input file to produce the map shown in Figure 5-2. The arrows indicate the flow
direction for each  element. The shaded areas indicate "dual" elements or overland elements with
companion channel elements. This map can be produced so that it will fit any scale. This feature
allows one to check input data on maps with different scales by producing overlays at the specific
scale needed. The  grid is also very useful in physically locating the predicted "hot spot" areas.

Figure 5-3 shows the graphic output from the HYPLT program. HYPLT uses standard CALCOMP-
compatible calls and plots the rainfall hyetograph, the runoff hydrograph and the sediment concen-
tration curve. The  program directly uses the hydrograph portion of the output listing.
                                          27

-------
          1  ?  5  4   5  6  7   P  S  12  1 1  \2 13  14  IS 18  17  18 1S2021 222524
   1 -1

   2 -7

   5 -IS

   4 -IS

   S -28

   6 -58

   7 -48

   8 -60

   S -75

   10  -S2

   11  -115

   12  -154

   15  -157

   14  -180

   IS  -205

   16  -226

   17  -2G0

   18  -275

   IS  -2S4

   20  -512

   21  -550

   22  -542
vF^
    Figure 5-2. Example Elemental Map Produced by QCKPLT Program.
 .OOO-i
4.000-
6.000 ->
        1.200-
        .800-
        .400-
                                Intensity
                                Runoff
                                               BRUNSON DITCH
                                            8 YR. - 1.5 HR. STORM
                                               .Sediment Concentration
                                                                         18.00
                                                                        - 12.00
                                                                        -  9.00
                                                                        - 3.00
        .000-4 - ' — -f - 1 - 1 - : - 1 —     r  "   T
             .0     40.0     80.0    120.0     160.0    200.0     240.0    280.0     320.0
                                      TIME - MINUTES
                                                                           .00
           Figure 5-3. Example Output of HYPLT Plotting Program.
                                     28


-------
The information presented in Figure 5-4 comes directly from the net transported sediment yield or
deposition section of the ANSWERS output. Several programming steps are required to "recon-
struct" an elemental data format (row and column coordinates) and input the individual element
information to a pmgram called CONTUR. CONTUR allows the user to set the levels of sediment yield
or deposition at which contours are desired. The program produces a map at any desired scale (up to
the maximum size the plotter allows). The shading in Figure 5-4 is for additional visual effect. The
portions of the watershed with closely spaced contqurs show areas with excessive transported soil
loss or deposition. In general, high transported soil loss areas will be found near high deposition
areas. This is due to the fact that steep slopes blend into flat slopes near the channels.
      CONSTRUCTION SETE


          WATERSHED



            Indiana MIP
              Erosion> 20 M.T./ha

              Deposition  10  M.T./ha


                     Figure 5-4. Example Output from CONTUR Program.
                                          29

-------
                           LIST OF REFERENCES

Alonso, C.V., W.H. Neibling and G.R. Foster. 1981.  Estimating sediment transport capacity in
watershed modeling. Transactions of the ASAE. 24(5):1211-1220,1226.

Barfield, B.J. 1968. Studies of turbulence in shallow sediment laden flow with superimposed
rainfall. Technical Report No. 11. Water Resources Institute, Texas A&M University. College Station,
TX.

Beasley, D.B. 1977. ANSWERS: A mathematical model for simulating the effects of land use and
management on water quality. Ph.D. Thesis. Purdue University. West Lafayette, IN. 266 p.

Beasley, D.B., L.F. Huggins and E.J. Monke. 1980. ANSWERS: A model for  watershed planning.
Transactions of the ASAE. 23(4):938-944.

Curtis, D.C. 1976. A deterministic urban storm water and sediment discharge model.  Proceedings
of the National Symposium on Urban Hydrology, Hydraulics, and Sediment  Control. University of
Kentucky.  Lexington, KY. pp. 151-162.

Davis, S.S. 1978. Deposition of nonuniform sediment by overland flow on concave slopes. M.S.
Thesis. Purdue University. West Lafayette, IN. 137 p.

Dillaha, T.A. 1981. Modeling the particle size distribution of eroded sediments during shallow
overland flow. Ph.D. Thesis. Purdue University. West Lafayette, IN. 189 p.

Fair, G.M.,  J.C. Geyer and D.A. Okun. 1968. Sedimentation. In: Water and Wastewater  Engineering:
Volume 2.  Water Purification and Wastewater Treatment and Disposal. John Wiley and Sons, Inc.
New York, NY. Chapter 25.

Foster, G.R. and L.D. Meyer. 1972. Transport of soil particles by shallow flow. Transactions of the
ASAE. 15(1):99-102.

Foster, G.R. 1976. Sedimentation, general.  Proceedings of the National Symposium on Urban
Hydrology, Hydraulics, and Sediment Control. University of Kentucky. Lexington, Ky. pp. 129-138.

Foster, G.R. 1982. Modeling the erosion process.  In: Hydrologic Modeling  of Small  Watersheds.
American  Society of Agricultural Engineers. St. Joseph, Ml.

Holtan, H.N. 1961. A concept for infiltration estimates in watershed engineering. ARS-41-51.
Agricultural Research Service, U.S. Department of Agriculture. 25 p.

Horton, R.E. 1919. Rainfall interception. Monthly Weather Review. 47:603-623.

Horton, R.E. 1939. Analysis of runoff plot experiments with varying infiltration capacity. Transac-
tions of the AGU. 20:693-711.

Huggins, L.F. and E.J.  Monke. 1966. The mathematical simulation of small watersheds. Technical
Report 1. Water Resources Research Center, Purdue University. West Lafayette, IN. 130 p.

Huggins, L.F., J.R.  Burney,  P.S. Kundu  and E.J. Monke. 1973. Simulation of the hydrology of
ungaged watersheds. Technical Report 38. Water Resources Research Center, Purdue University.
West Lafayette, IN. 70 p.

Israelsen,  0. W. and V. E Hansen. 1962. Irrigation principles and practices.  John Wiley and Sons,
Inc. New York, NY.

Mantz, P.A. 1977. Incipient transport of fine  grains and flakes by fluids - extended Shield's
diagram. Journal of Hydraulics Division, Proceedings of the ASCE. 103(HY6):601-615.
                                          30

-------
Meyer, L.D. and W.H. Wischmeier. 1969. Mathematical simulation of the processes of soil erosion
by water.  Transactions of the ASAE. 12(6)754-758.

Overton, D.E. 1965. Mathematical refinement of an infiltration equation for watershed engineer-
ing. ARS-41-99. Agricultural Research Service, U.S. Department of Agriculture. 11 p.

Philip, J.R. 1957. The theory of infiltration: 1. The infiltration equation and its solution. Soil
Science. 83:345-357.

Skaggs, R.W., L.F. Huggins, EJ. Monke and G.R. Foster. 1969. Experimental evaluation of infiltra-
tion equations. Transactions of the ASAE. 12(6): 822-828.

Wischmeier, W. H., C. B. Johnson and B. V. Cross. 1971. A soil erodibility nomograph for farmland
and construction sites. Journal of Soil and Water Conservation. 26(5):189-193.

Wischmeier, W. H. and D. D. Smith. 1978. Predicting rainfall erosion losses - a guide to conserva-
tion planning. Agriculture Handbook 537. Science and Education Administration, U. S. Department
of Agriculture. 58 p.

Yalin, Y.S. 1963. An expression  for bed-load transportation. Journal of Hydraulics Division,
Proceedings of the ASCE. 89(HY3):221-250.

Young, R.A.  and C.K. Mutchler. 1969. So/7 movement on irregular slopes.  Water Resources
Research.  5(5): 1084-1089.
                                          31

-------
                  APPENDIX A - SOIL PARAMETERS
                                    Section A.I
The parameters described in this section are concerned with the physical description of the soil
condition. The parameters do have some seasonal variation, since the bulk density of a 59!! does
vary somewhat throughout the year. The total porosity (TP) is a measure of the bulk density. The
field capacity (FP) describes the upper limit of available water in the soil. It also quantifies that
portion of the pore volume which can contain only gravitational water (above the moisture holding
capacity of the soil). The control zone depth (DF) identifies the volume of soil (depth) that actually
influences the rate of infiltration at the surface. The antecedent soil moisture (ASM) quantifies the
starting point for the soil moisture-based infiltration equation. Methods for determining each of
these important parameters will be presented in this section.

Total Porosity (TP).The volume of pore space within the soil is directly related to the bulk density
(weight per unit volume) of the soil. The total porosity (percent pore space) of a soil is defined as:


                                   TP = 100 -  (>)  * 100                          (A-1)
         where:  TP = total  porosity,  percent,
                 BD = bulk density,
                 PD = particle density (assumed  to  be  2.65).


In some cases, comprehensive soil surveys will contain information on the bulk density of the
mapped soil types. However, most soil surveys, even the newer ones, do not contain this kind of
data. One bit of information that is available in all soil surveys is the textural class of the individual
soil types. Without specific information on the bulk density of a particular soil, the modeler can still
make a reasonabjy good estimate of the bulk density, and thus the total porosity, by utilizing the
textural class definition.

Table A-1 lists bulk densities and other soil physical properties for several different soil textural
classes. For those classes not listed, an  interpolation can be performed. Very sandy soils or organic
soils have much larger ranges of values.

Field Capacity (FPJ. As the moisture content of the soil is increased, a point is reached when water
begins to drain due to gravitational forces. Another way of describing this phenomenon is to say that
the moisture holding tension within the  soil becomes less than 1 /3 atmosphere. The point at which
this gravitational drainage begins is called field capacity (FP) and is expressed as a percentage of
saturation. Saturation occurs when the total pore space within the soil is filled with water. Thus, a
soil with a total porosity of 50 percent and a field capacity of 70 percent contains 35 percent water (by
volume) at field capacity.

Although some soil surveys contain information about the actual field capacity of the individual soil
types,  most soil surveys  only state what the available water capacity of the soil is. Using the
information in Table A-1 and the available water capacity of the individual soijs (A horizon), the
modeler can easily estimate the field capacity (percent  saturation) when that information is not
available. By definition, the available water capacity of a soil is that water held within the pores
between field capacity (tension of 1 /3 atmosphere) and the wilting point (tension of 15 atmospheres).
In addition,  the assumption is made that approximately  one half of the water in  the soil is
                                           33

-------
                 Table A-1. Some Representative Physical Properties of Soils.
Soil
Texture
Sandy
Sandy loam
Loam
Clay loam
Silty clay
Clay
Bulk
Density*
(9/cc)
1.65
(1.55-1.80)
1.50
(1.40-1.60)
1.40
(1.35-1.50)
1.35
(1.30-1.40)
1.30
(1.25-1.35)
1.25
(1.20-1.30)
Total
Porosity*
(% volume)
38
(32-42)
43
(40-47)
47
(43-49)
49
(47-51)
51
(49-53)
53
(51-55)
Field
Capacity*
(% saturation)
39
(31-47)
49
(38-57)
66
(59-74)
74
(66-82)
79
(72-86)
83
(76-89)
Wilting
Point*
(% saturation)
17
(10-24
21
(15-26)
30
(26-34)
36
(32-40)
38
(34-42)
40
(37-43)
        *Numbers in parentheses indicate normal range.

         Adapted from: (Israelsen and Hansen, 1962).
unavailable. Thus, if the available water is listed as .15 inches/inch (15 percent by volume), the field
capacity of the soil is twice that amount or .30 inches/inch (30 percent by volume). Further, if the
total porosity has been listed as 50 percent, that means that the field capacity of the soil is 30 percent
divided by 50 percent or 60 percent of saturation.

Infiltration Control Zone Depth (DF). Of all of the parameters used in the ANSWERS, this one is
the least well defined and most arbitrary. Essentially, it describes the volume of soil (depth) that
influences the infiltration rate at the surface of the soil. Experimental data and simulation experience
have lead to the conclusion that the control zone depth (DF) varies with time. However, not enough
information exists to describe a seasonal influence exactly. In general, the control zone depth is
equal to or less than the depth of the A horizon. For most of the soils that have been modeled in the
Midwest, the control  depth (DF) has been assumed to be equal to .25 to .75 of the depth of the A
horizon. In general, a starting value of one half of the A horizon has been used.

Antecedent Soil Moisture (ASM).The infiltration equation in the ANSWERS model is based on the
moisture content of the soil. Since the infiltration rate will be much greater when the soil is "dry" than
when it is "wet", it is very crucial that the correct antecedent moisture  content be used when
simulating actual situations. For hypothetical or "wet weather" simulations, moisture contents at or
above field capacity will generally be used.
                                           34

-------
This section details a simple moisture balance approach for determining the antecedent moisture
content in each soil. The form of the moisture balance equation is:

                              ASM = ASML + RAIN - ET - RO - PERC                    (A'2)
          where:   ASM = antecedent soil  moisture,
                  ASML = last known (initial) soil moisture,
                  RAIN = daily rainfall,
                    ET = evapotranspiration,
                    RO = runoff,
                  PERC = percolation.

In this equation, percolation refers to drainage of gravitational water (water in excess of field
capacity). In order to simplify the moisture balance calculations, several assumptions are made:
    1.  The depth of the soil layer that influences the moisture content is equal to the control
       zone depth (DF),
    2.  The evapotranspiration  rate is one half of normal on days that have rainfall in excess
       of 0.2 inches,
    3.  The soil drains down to field capacity within 1 day at the steady state infiltration rate
       (FC),
    4.  When the soil moisture  content reaches the wilting point, no additional moisture is
       lost due to ET,
    5.  On days when rainfall is less than 0.3 inches, RO (runoff) is zero,
    6.  On days when rainfall is between 0.3 and 0.8 inches, RO  0.10*RAIN,
    7.  On days when rainfall is between 0.8 and 1.5 inches, RO  0.15*RAIN,
    8.  On days when rainfall is greater than 1.5 inches, RO   0.20*RAIN.

The rate of evapotranspiration can be calculated using any of several different equations. Each
method entails certain  assumptions  and the user must determine which equation best serves his
purposes and  utilizes his data. The  average monthly rates shown in the following example are
representative of cropland rates in northern Indiana.

The antecedent moisture calculations should be started approximately one month prior to the time to
be simulated. Field capacity or any other reasonable moisture content can be assumed as a starting
point. During the period of calculation, the soil moisture is not allowed to go below the wilting point.
Once enough rainfall has occurred within the calculation period to equal or exceed field capacity, the
previous history  is wiped out. Table A-2 shows example moisture calculations for a soil with a total
porosity of 50 percent, control depth (DF) of 6 inches, field capacity of 70 percent saturation (wilting
point of 35 percent saturation) and a steady state infiltration rate (see Section A.2) of 0.3 inches per
hour. Using the above information and assuming that an ANSWERS simulation is to be started on
June 14, the ASM value for this soil type would be 67  percent.


                                     Section A.2
The parameters  which specifically  describe  a soil's infiltration response as described by the
modified form  of Holtan's equation used in ANSWERS are defined in this section. The steady state
infiltration rate (FC) indicates the rate at which the soil will absorb water when the soil is saturated.
The difference between the maximum and steady state infiltration rates (A)  combined with the
infiltration exponent (P) helps to describe the typical exponential "drawdown" of the infiltration rate.

Infiltration Rate Descriptors (FC and AJ. A simple procedure for selecting values for the steady
state infiltration rate (FC) and the difference between  maximum  and steady state rates (A) is
described here. The user is, of course, free to use any information he has concerning these values.
Soil survey information is used in this procedure due to its general  nature and  ready availability.


                                           35

-------
              Table A-2. Antecedent Soil Moisture (ASM) Calculation Example.
Day
1 (5/15)
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18 (6/1)
19
20
21
22
23
24
25
26
27
28
29
30
31 (6/14)
Soil Moisture
(% saturation)
70
69
67
65
64
62
100
94
93
69
67
76
68
67
65
65
73
78
76
77
79
81
85
69
67
65
63
62
92
69
67
Rain
(in.)
.01
0
0
0
0
3.16
.90
.80
.02
0
.29
0
0
0
.05
.27
.27
.21
.25
.35
.39
.53
.03
0
0
0
.02
1.11
.04
0

ET
(in.)
.05
.05
.05
.05
.05
.03
.03
.03
.05
.05
.02
.05
.05
.05
.05
.03
.03
.03
.03
.03
.03
.03
.06
.06
.06
.06
.06
.03
.06
.06

Runoff
(in.)
0
0
0
0
0
.63
.14
.08
0
0
0
0
0
0
0
0
0
0
0
.04
.04
.05
0
0
0
0
0
.17
0
0

Percolation
(in.)
0
0
0
0
0
0
2.26
.73
.69
0
0
.18
0
0
0
0
.09
.24
.18
.22
.28
.32
.45
0
0
0
0
0
.66
0

The range of permeabilities for a given soil type (as listed in the USDA Soil Survey format) are used in
the following manner:
    1. The midpoint of the lower 1/3 of the range is used for FC,
    2. The midpoint of the upper 2/3 of the range is assumed to be the maximum rate,
    3. The value of A is equal to the maximum rate minus FC.

Assuming a permeability range of 0.2 to 1.5 inches  per hour, the following example illustrates the
technique:

                 The total  range =1.3  inches  per hour

                 1/3 of range = 1.3/3 = .43  inches per hour

                 Thus, FC equals the midpoint  of the lower 1/3 of range

                      FC = .2. + (.43/2)  = .42  inches per  hour


                                          36

-------
             The maximum  rate  is  the midpoint of the  upper  2/3 of  the  range

                 maximum rate =  ((.2 +  .43) +1.5)/2  =  1.07  inches per  hour

             The A  value  equals the maximum rate minus FC

                 A^ =  1.07  -  .42  =  .65 inches per hour


Using the entire range, FC would be 0.2 iph and A would be 1.3 iph. The method detailed above
appears to give more realistic numbers.

Infiltration Exponent (P). As stated in the section on component relationships, the infiltration
exponent (P) relates the rate of decrease of infiltration capacity to increasing moisture content. This
property varies among  soil types and is most closely related to the textural  class of the soil. The
heavier the texture (more clay), the larger the value of P. Conversely, sandy soils show little change
in infiltration rate with increasing soil moisture content and, thus, have a much smaller value of P.
Table A-3 lists some starting point values for several textural classes.

                      Table A-3.  "P" Values for Various Soil Testures.
                    Soil Texture       Suggested Values for "P1
Clay
Silty clay
Clay loam
Loam
Sandy loam
Sand
.75 -
.65 -
.60 -
.55 -
.50 -
.35 -
.80
.75
.70
.65
.60
.50
                                     Section A.3
Two different types of information are included in this section. The USLE "K" factor or soil erodibility
of each soil type is described. The general subsurface drainage characteristics of the watershed are
described by the combination of the tile drainage coefficient and the groundwater release fraction.
Both of these drainage terms are defined and methods of parameter value assignment are discussed.

Soil Erodibility - USLE "K" (K). Most of the newest USDA Soil Surveys contain information pn the
"K" factor for each soil type mapped. Other sources of this information include statewide soil loss
handbooks or brochures which are published by most state SCS offices. Wischmeier et al. (1971)
have produced a nomograph technique for determining the USLE "K" factor based on textural class
and other soil characteristics.

Section A.4 details a method for simplifying the soils description file presented in this Appendix.
Essentially, the technique involves the grouping of soils by similarities in drainage response. When
soils are grouped in this manner, the (K) parameter for the various individual soils may not be equal.
If they are not, one of two methods can be used to arrive at an "effective" (K).
      1.  Use an area weighted average of "K" values within a  drainage class, or
      2.  Use the value of "K" for the predominant soil(s).

                                           37

-------
Subsurface Drainage Characteristics. The  two parameters  which describe the rate of
subsurface water movement are the tile drainage coefficient and the groundwater release fraction.
The tile drainage coefficient is simply the design value for tile drainage. Interflow or groundwater
release is described by putting a fraction of the water in the subsurface reservoir into the channel
system at each time step. Experience has shown that the value of the fraction varies from as little as
zero to approximately 0.01. Small values of the fraction may actually cause an increase in the flow
rate on the recession limb of the hydrograph due to the fact that the drainage rate from the control
zone is greater than the groundwater movement rate. Thus, the subsurface reservoir of water
increases, and the interflow  rate rises accordingly. Higher values of the fraction will cause the
hydrograph to "level off" for a period of time and then decrease as the rate of subsurface drainage
becomes less than the interflow rate.

                                     Section A.4
In order to reduce the number of soils that must  be described in the "predata" file, a technique has
been developed for identifying soils with similar drainage characteristics. The similar soils are then
placed in a general group with drainage and erosion characteristics which describe the "average"
response of the soils making up the group.

The procedure requires information about the drainage classification and hydrologic soil group of
each soil type. The technique involves the following:
    1.  Soils are  listed by their drainage classification (i.e., well drained, poorly drained,
       very poorly drained, etc.) and by hydrologic group (i.e., A, B, C or D).
    2.  The soils are first grouped by drainage classification. Then, those soils are examined
       for hydrologic group. Soils that have the same drainage classification and hydrologic
       group are considered to have similar responses. It is possible that within one
       drainage classification there may be soils with two or more hydrologic groups. The
       soils in each hydrologic group should be considered as a separate soil group.
    3.  The soil(s) that predominate the area in each soil group are chosen as representative.
       A more complex method would be to select the descriptive parameters based on area
       weighting.
                                           38

-------
  APPENDIX B - LAND USE AND SURFACE PARAMETERS
                                   Section B.I
The information presented in this section involves the extent of crop cover, the flow retardance of the
surface and the relative erosiveness of the various crops or land uses. The specific land use or crop
(CROP) is simply an identifier that is printed during output. The potential interception (PIT) and
percent cover (PER) are used to describe the interception of rainfall. Manning's n (N) describes the
surface roughness or retardance to flow.  The relative erosiveness parameter (C) is actually a
combinatio_n of the USLE "C" and "P" values with seasonal adjustment.
Interception Parameters (PIT and PER). A certain amount of the precipitation during any event
never reaches the soil surface. Contact with and storage on vegetation accounts for this removal
and is called interception. The potential interception volume (PIT) describes the volume of moisture
that could be removed if the area were completely covered by that crop  or land use. The actual
percentage of cover (PER) assumes the non-covered area has no  interception. Table B-1 lists some
example values for PIT.

                        Table B-1. Potential Interception Values.
                                  Crop                 PIT
                       Oats                           .5 - 1.0

                       Corn                           .3 - 1.3

                       Grass                          .5 - 1.0

                       Pasture and Meadow              .3 - .5

                       Wheat, Rye and Barley           .3 - 1.0

                       Beans, Potatoes and Cabbage     .5 - 1.5

                       Woods                         1.0 - 2.5
Manning's n (N).The measure of surface roughness or flow retardance used in the flow equation in
ANSWERS is Manning's n. This information, when combined with element slopes, rainfall, intercep-
tion, infiltration and  routing considerations, helps yield the solution to the continuity equation,
which is the basis of ANSWERS. There are numerous sources for obtaining reasonable values of n for
channel and overland flow situations.
Relative Erosiveness (C).This parameter is used in determining how much soil could potentially
erode due to a particular crop or land  use, when compared to fallow ground under identical
conditions. It is a direct combination of the USLE "C" and "P" parameters with a seasonal adjust-
ment. Thus, conventionally tilled corn at crop stage 1 will have a higher (more erosive) C value than
the same corn at crop stage 2, when there is more foliage and root structure. Agriculture Handbook
No. 537 (Wischmeier and Smith, 1978) contains information for determining the USLE "C" and "P"
values throughout the year for numerous crops and management systems.
                                         39

-------
                                    Section B.2
Figure B-1 shows a profile of a section of the soil surface. The combination of the peaks and valleys
yields a certain depressional storage volume. In addition, the amount of the surface that area that is
inundated at any time is a direct function of the depth of water on the surface. The infiltration rate
within an element is greatly affected by the amount of pondage within the area.
                             Figure B-1. Soil Surface Profile.

Surface Storage Descriptors (HUand RC). The ANSWERS model uses the maximum roughness
height (HU) and a roughness coefficient (RC) to describe the surface storage characteristics and the
ponded surface area. The roughness coefficient (RC) is, essentially, a shape factor which describes
the frequency and severity of the roughness. The maximum roughness height (HU) is used to
establish the upper limits of the surface roughness and is physically measurable. Table B-2 shows
some typical values for both HU and RC.

                      Table B-2. Typical Surface Storage Coefficients.
                      Surface Condition
              RC
                    Plowed Ground:

                      Spring - smooth
                      Spring - normal
                      Spring - rough

                      Fall - smooth
                      Fall - normal
                      Fall - rough
                                              (mm)
100
130
130

 60
 70
130
.53
.48
.59

.37
.33
.45
                    Disked and Harrowed:

                      Very smooth
                      Rather rough

                    Corn Stubble
 30
 60

110
.42
.43

.59
                                          40

-------
     APPENDIX C - INDIVIDUAL ELEMENT INFORMATION
                                   Section G.I
 One of the strengths of the ANSWERS model is its ability to describe spatially those processes which
 affect the hydrologic and erosion  responses of a watershed. In order to accomplish this, the
 topographic influences of the  various sub-areas within  the watershed must be quantified. By
 applying a square grid to a soils or topographic map, the element pattern is defined. Once done,
 additional information concerning specific soils,  land use,  drainage and BMPs can be added easily.

 In the rather rare situation in which the necessary watershed data is available on machine readable
 media, the only requirement is to develop a simple computer program to reformat the data in a
 manner compatible with the above definitions. Unfortunately, the more common situation is to have
 only maps available from which the necessary information must be manually extracted. Because the
 format outlined in Chapter IV is rather tedious, a  series of programs have been written to facilitate
 the manual  transcription of map information into the file format required by ANSWERS.

 It is recommended that a transparent overlay be prepared with a grid of squares drawn to the
 map-scale size desired for watershed elements. It is further suggested that this grid have its  rows
 and columns numbered beginning in the upper left-hand corner with number 1 for the rows and for
 the columns of elements. This overlay can then be superimposed on topographic and soils maps to
 delineate the elements of which the catchment is composed. The map should be oriented in such a
 manner that the direction of predominate water flow is from the top (row 1) to the bottom of the map,
 i.e., the watershed outlet should be  as near to the highest numbered row of elements as possible
 while still maintaining an orientation which is compatible with most field boundaries. This  often
 means that the rows will be oriented  in one of the four cardinal directions. Although not an absolute
 necessity, this arrangement does facilitate computation.

 Available data formatting programs are written with the expectation that separate files will be
 manualjy prepared for each different physical parameter in the element file (i.e., soil, crop, channel
 type, rain gauge). Data for these programs may be prepared in a free-format style with each numeric
 entry separated by a comma. Any number of data entries may be entered on a single line. For those
 locations which do not have a FORTRAN compiler which accepts free-format input, it is a simple
 matter to modify the input format statements in whatever  manner desired.

 The method for describing the direction of slope  applies to both overland and channel flow direc-
 tions. In those elements which contain channels, the direction is in 45 degree increments which
 contribute all of the flow to one of  the eight surrounding (and touching) elements. The  channel
 direction takes precedence over the  overland direction.

 Simplified Data  File Construction. The basis of the simplified data file construction technique is
 the use of USDA-SCS Soil  Survey  information  to provide a map base for the grid  and slope
 information from  the mapping units themselves.  Although the mapping units used in modern Soil
 Surveys give a range of slopes for each defined area, there is no information provided on slope
 direction. Therefore, directions  must be chosen using a "common sense" approach in conjunction
with topographic  information.

An additional degree of slope descriptiveness can be obtained by using the surrounding elements to
determine which part of the slope range should be used. For example, when a soil with a range of 3-6
 percent is surrounded by soils with steeper slopes, the high end of the range (approximately 6
percent) should be used. On the other hand, a soil with 6-12 percent slopes surrounded by lesser
slopes would use the low end of the range. An element surrounded by soils with similar slopes would
use the midpoint of the range.
                                         41

-------
Data File Construction Using Topographic Maps. ANSWERS requires that the average slope
steepness and direction of steepest slope be specified for each element in the catchment. Rather than
attempt to obtain this information directly from a topo map a program, called ELEVAA, has been
written which requires instead that the elevations of the corners of each element be input. That
program calculates the slope steepness and flow direction of each element and then sets up the
complete format for the elemental data file. That file is subsequently updated to add inf9rmation
concerning the soil type and crop/management practice for each element and to identify those
elements which contain a well defined channel.

ELEVAA is organized so that the identification of the location of the elevation data requires a slight
mental transposition of the element's row and column number to refer to the elevation at the upper
left-hand corner of the element. For example, position 4,6 for an elevation actually refers to the
upper-left corner elevation for the  element in row 4 and column 6 of the watershed. The elevation in
the lower right-hand corner of element 4,6 would be identified by the position numbers5,7 (note that
this elevation is also the elevation of the upper left-hand corner of element 5,7; thus, since a single
point elevation is normally used to help compute the slppe direction and steepness of four different
elements, one never needs to be concerned with specifying the location of anything except the upper
left-hand corner elevation for each element).

Begin preparing the input data file for ELEVAA by starting with the upper corner elevations of the top
row of elements in the catchment. Generally this will be row number 1. The first two entries on a data
line must be, respectively, the corresponding  row number and the  column number of the first
elevation that will appear on that line. The remaining entries on that line will be the elevations  of
sequential elements in that given  row. Any number of elevations, up to a maximum of 28, may be
entered on a single line, but they must be for sequential columns. Thus, if a row is discontinuous
(watershed boundaries are irregular and often exclude  some element positions) or whenever all
elements in a given row have been specified, it is necessary to start a new line of data. The end  of
data input to ELEVAA is signified by including a blank line after all elevation data has been specified.

There are some difficulties  that arise as a result of trying to generate the elemental data file from
corner elevations. It is almost certain that some slope directions calculated on the basis of corner
elevations will be unsatisfactory  and will need to be manually edited. This is especially true for
boundary elements of the catchment. They will often be computed in a direction which would result
in some of their outflow moving out of the watershed. Also, it is mandatory that all of the outflow
from the outlet element move directly out of the catchment, i.e., it should have one of the cardinal (0,
90,180 or270) directions and will usually be specified as a channel element as discussed below. It is
also likely that elements located near to stream channels will not have appropriate slope directions
of slope steepness returned from ELEVAA due to topographic irregularities along a channel. Most of
these difficulties will  be automatically corrected when the  channel data file is processed. In any
case, it is recommended that all manual editing of the elemental data file be postponed until after the
basic soils and channel data have also been processed.

One final warning should be given concerning possible difficulties with the output of ELEVAA for
cases where the watershed of concern has  substantial areas which are relatively flat or even
depressional. For such situations, it is likely that ELEVAA will return element flow directions for a
group of elements which results in a computational ponding of water, i.e., the flow pattern for the
of elements has no outlet for surface flows. These areas, due to the cumulative flow depths that
result from the circulating runoff,  will often show a false very high erosion rate. There are at least
two ways to discover such problem  areas. One is to plot an elemental map with the flow direction
show for each element and to visually inspect it for such regions. A GCS (Graphics Compatibility
System)-based plotting program, called QCKPLT has been devised for this purpose (see Chapter V for
an example). Even if that procedure is followed, the second method is recommended as afinal check.
This second method  involves  running an initial watershed  simulation assuming a  fallow (no
vegetation) condition and then making a critical examination of all regions which show high erosion
rates to make sure they are not the result of circulatory regions caused by erroneous element slope
                                           42

-------
directions. Once such regions are detected it is necessary to manually correct the flow directions to
provide a drainage outlet for the region. An interactive editing program for this purpose has been
written in the "C" language for use on computers utilizing the UNIX operating system. A FORTRAN-
based version of this editor is planned.

                                     Section G.2
In order to fully utilize the ANSWERS model as a planning or evaluation tool, it must be able to
simulate the particular practices that users might apply to a watershed. ANSWERS handles BMPs in
two different ways. Practices which are strictly tillage-oriented are described in both the soils and
land use files. This is due to the fact that tillage-based management changes can affect both the
infiltration response and the surface condition of the soil. On the other hand, those BMPs which are
structural in nature or which require a change in land use (row crop to grass for waterways or field
borders, for example) are described in the individual element data file.

The primary tillage based BMPs considered by ANSWERS are:
    1.  Change from conventional tillage (fall turning plow) to chisel plowing,
    2.  Change from conventional tillage to minimum tillage,
    3.  Change from conventional tillage to no-till.

Each of these changes has similar effects, but to a different degree upon both the soils and land use
(surface) descriptors. This necessitates describing both "new" soils and land  uses.

When conventional tillage is replaced by chisel plowing (#1), the infiltration components (A and FC)
should be increased approximately 20 percent. At the same time, the surface descriptors (HU and N)
should be increased by approximately 15 percent. In changing from conventional tillage to minimum
tillage techniques (#2), the infiltration and surface parameters should be increased by approxi-
mately 25 percent and 20 percent, respectively. The change from conventional to no-till (#3) requires
increases of approximately 25 percent for both infiltration and surface parameters.

Table C-1 lists the BMPs which are based on structures or land use changes. Each BMP, its code
number, additional descriptors required and any pertinent assumptions are listed.


               Table C-1. Structural and Land Use  Change BMP Descriptions.


                                          Additional
                BMP            Code  No.    Information          Assumptions


        Tile Outlet Terraces       1                     Trap efficiency of 90%

                                                        Only lowermost terraces
                                                        are described

        Sedimentation Pond        2                     Trap efficiency of 95%

                                                        Only ponds  in  upland
                                                        areas  should be defined,
                                                        instream structures  are
                                                        treated differently

        Grassed Waterways         3          Width       Decreases credibility

        Field  Borders             4       Total  width    Decreases credibility
                                           43

-------
If a terrace or pond exists only in a portion of the element, the assumption is made that all incoming
flow is influenced by that BMP. Thus, elements which have only a small portion of a practice within
their boundaries should not be given credit for that practice. The most effective BMP found in an
element should be the one described. As an  example, in an  element with a pond and a grassed
waterway, the pond, with its very effective sediment trapping characteristics, should be chosen for
description.
                                          44

-------
          APPENDIX D - ENHANCED SEDIMENT MODEL
                                     general
In order to better understand the complex processes of sediment and nutrient movement on a
watershed scale, it is necessary to have a transport model capable of determining not only the
magnitude of the material in movement but also the makeup. Since phosphorus transport is highly
influenced by the amount of fine material in the sediment load, a model which can predict the actual
"enrichment" of fines in the sediment load should be a very positive addition to a total "water
quality" model.

An expanded capability sediment transport model is available as part of the ANSWERS package
which can produce information on the differential detachment and transport of the various particles
which make up the soil surface and sediment load. This model utilizes many of the assumptions that
the original detachment/transport model within ANSWERS uses. It is based  on the Yalin (1963)
model with modifications for multiple particle classes and for varying densities.

The component relationships, additional input information  required, and output formats are des-
cribed in this section. The general release of the ANSWERS model does not contain this version of
the sediment model. It is significantly slower and  requires more computer  resources  than the
original sediment algorithms. A complete discussion of the development and testing of this section
of ANSWERS can be found in Dillaha (1981). A modified version of ANSWERS, incorporating this
model, can be obtained by contacting the Agricultural Engineering Department, Purdue University,
West Lafayette, IN 47907.

                              Model Development
Sediment transported by overland flow occurs in two related forms: bedload and suspended load.
Bedload is that portion of the load which moves along the bottom of the flow by saltation, rolling and
sliding. Bedload is generally composed of the larger soil particles (sand, gravel and aggregates).
Bedload movement  is highly transport dependent. A decrease in transport capacity  causes the
immediate deposition of the excess bedload.

Suspended load is much more uniformly distributed throughout the flow depth as contrasted with
bedload.  A decrease in the transport capacity  will not result in the immediate deposition of
suspended sediment. This delay is the result of the small fall velocities of most suspended particles
which prevents them from depositing immediately. Part of the suspended  load has such small fall
velocities (see Table D-1) that they are effectively permanently suspended, when compared with the
larger sediment particles. In this research the smallest particles, primary clay and silt less than 10
microns in diameter, were considered to be washload and were not allowed to deposit unless the
overland flow rate was zero. This means that their yield was determined entirely by their weight
fraction in the original soil  masses and the total amount of detachment.


                        Table D-1. Typical Sediment Characteristics.
               Particle Type   Size  Specific Fall Velocity  Time to Settle
                              (mm)  Gravity     (mm/s)        1.0 mm
Primary Clay 0.002
Primary Silt 0.010
Small Aggregate 0.030
Primary Sand 0.200
Large Aggregate 0.500
2.65
2.65
1.80
2.65
1.60
0.003
0.080
0.349
24.
40.
5. min
13. s
3. s
.04 s
.03 s
                                         45

-------
The fraction of particles larger than 10 microns which were deposited when there was a transport
capacity deficit was defined as:
                       RE. = FVi *     .     (if REi >  1, then  RE.  =  1)               (D"1}
         where:   RE  =  fraction of particles of partiqle  size class  i
                          depositing,
                  FV. =  fall velocity of particle  i, m/s,              ?
                 AREA =  area of an overland  flow or channel element,  m ,
                    Q =  flow rate, m /sec.
This is the classic relation describing the settling efficiency of particles in water (Weber, 1972). This
equation is not strictly true for the case of overland flow since it was developed assuming that:
    1. The settling is occurring under quiescent conditions.
    2. Flow is steady and the concentration of suspended particles of each size is initially
      distributed uniformly throughout the flow.
    3. Once the particles are deposited they are not resuspended.

Although overland flow conditions violate some of these assumptions (resulting in a possible
overestimation of the removal efficiency), it was decided to use the equation anyway. It was chosen
because it is an improvement over the normal procedure which assumes that all excess sediment is
deposited immediately and because the overestimation of removal efficiency due to turbulence is
potentially offset by the high concentration of the large particles in the lower flow region.

The fall  velocity, FVi, in Equation (D-1) was calculated as (Fair, et al., 1968):


                                   4 *  AGRAV *  (SG-l) * DIA. *5                    (D-2)
                            FVi =  (	r^TD	L)


                                                         o
         where:  AGRAV = acceleration due to gravity, m/s ,
                   SG. = particle density, g/cm  ,
                  DIA. = diameter of particle class i, mg,
                    CD = Newton's drag coefficient.


Newton's drag coefficient, CD, is defined as:

                             CD  =  18 * VISCOS       for REYN < .1                   (D-3a)
                          CD =  lL +    3    + .34    for REYN > .1                (D-3b)
         where:  VISCOS  =  kinematic viscosity of water, m2/s,
                    REYN  =  Reynolds number = FVi * DIA1 * VISCOS.




                                           46

-------
For Reynold's numbers less than or equal to 0.1 the fall velocity calculation was direct; but for values
greater than 0.1, an iterative solution was required. If experimentally determined fall velocities are
available for the actual particle classes being modeled, they should be input directly and the model
will not have to estimate the fall velocities.

The rate at which sediment is transported by flow is a function of both the sediment supply rate and
the transport capacity of the flow. The net rate of movement is proportional to the lesser of the
transport or the rate at which the particles are available for transport.

The selection of a transport equation for the overland flow regime is probably  the most critical
component of a sediment yield model. It is very difficult because all  accepted transport models,
developed to date, were developed for streamflow conditions. The differences between deeper
channel flows and shallow overland flows are  significant. Shallow  flow undulates constantly,
resulting in flow regime changes; rainfall drastically increases the turbulence of the flow (Barfield,
1968); the ratio of particle diameter to flow may be high; surface tension may affect flow conditions;
(Young and Mutchler, 1969); and the slope of the energy gradeline is difficult to estimate (Foster,
1982). In spite of these differences, streamflow transport relations must be used because of the lack
of any relations developed from overland flow data.

Foster and Meyer (1972), Davis (1978), Alonso, et al. (1981) and Foster (1982) have done considerable
work in studying the use of streamflow equations for modeling the overland flow situation. Their
general concensus is that the Yalin (1963) equation best fits experimental overland flow data,
especially when one is modeling the particle size distribution of the eroded sediment. Based upon
their success with the equation it was decided  to use the Yalin equation in this work.

Yalin (1963) developed an expression for the bedload transport of uniform, cohesionless grains over
a movable bed for steady, uniform flow of a viscous fluid. His derivation is based upon dimensional
analysis and upon the mechanics of the average motion of a grain. The flow was assumed to be
turbulent with a laminar sublayer not exceeding the thickness of the grains. Grain motion was
assumed to be by saltation and a critical tractive force was assumed. The critical tractive force as a
function of the laminar sublayer thickness was assumed to be demonstrated by the Shields' curve.

The Yalin equation is:


                           TF = PS * SG *    * DIA * VSTAR * AGRAV                 (D'4a)
          where:
                          PS  =  YALCON  * DELTA * (1  - 1n (1  * S.IGMA))
                                                         SIGMA
                              SIGMA =  2.45  * SG~'4 * Y™  * DELTA                   (D-4c)
                                                      l/K
                        DELTA  = JL -  I     (when  Y  <  YrD,  DELTA =  0)               (D-4d)
                                 ICR                   CK
                                     	VSTAR2                                (D-4e)
                                     (SG - 1) * AGRAV * DIA
                                           47

-------
                                 VSTAR = (AGRAV * r * SI)
                                                         .5                       (D-4f)
                TF = transport capacity,
                e  = mass density of fluid,
             VSTAR = shear velocity,
               YpR = critical shear stress from Shield's diagram,
                5C = slope of the energy grade! ine,
                 r = hydraulic radius,
            YALCON = an empirically derived factor = 0.635.


In the new sediment model, the hydraulic radius,  r, was assumed to be equal to the flow depth
(actually depth of stored water less the depth of retention storage).

Foster and Meyer (1972) developed a method by which Yalin's equation could be used to predict the
transport capacity of each particle size class in a mixed sediment. Yalin assumed that the number of
particles in transpprt was equal to DELTA. Foster and Meyer, therefore, assumed that for a mixture,
the number of particles of a size i is proportional to DELTAj. Values of DELTAj for each particle class
were summed to give the total transportability, SDEL:

                                           NPART
                                     SDEL = £  DELTA.                            (°-5)
         where:   NPART = number of particle size classes,
                 DELTA. = DELTA as defined by Equation (D-4d)  for
                      1     particle class.



The number of transported particles of class i  in a mixture, (Ne)j was taken to be:


                                           _ N1  * DELTA.                            (D.6)
                                     v Vi      S~D"ET
          where:   N. = the number of particles transported in a uniform
                        sediment for DELTA..
In a similar manner, PS was assumed to be proportional to DELTA, so:
                                            PS,  * DELTAi                            (D.7)

                                    1  e'1       5DEC
         where:  (P ). = the effective PS for particle class i  in  a
                           mixture,
                   PS  = the PS calculated for a uniform sediment  of  class  i.
                                           48

-------
The actual transport rate TFj of each particle class in a mixture can then be expressed as:
                       TFi = (lVi * SGi * PW * AGRAV * DIAi * VSTAR
                                                                                   (D-8)
Yalin's method, as originally derived, was developed from laboratory flume studies of the transport
of sand and gravel with flows much deeper than those encountered in overland flow. The constant
0.635 was determined from these flume studies. Davis (1978) found that 0.88 was a better value for
sand with a diameter of 342 microns and that 0.47 was better for coal particles with diameters of 156
microns and 342 microns and specific gravities of 1.67 and 1.60 respectively. No other information is
available concerning the value to use for YALCON unless one determines it by calibration. It was
decided to leave it fixed at 0.635 in this study.

The original Shields' diagram used by Yalin was not developed for sediment particles of low specific
gravity and small size. Fortunately, the Shields' diagram was extended for small particles by Mantz
(1977) and Figure D-1 can be used to estimate YCR. In the model, Shields' diagram, as extended by
Mantz, was broken down into the 5  regions listed in Table D-2.
                   0.1             1.0             tO              100

                         PARTICLE REYNOLDS NO. , REYN = VSTAR •  DIA/VISCOS


                  Figure D-1. Shield's Diagram as Extended by Mantz (1977).



              Table D-2. Critical Shear Stress Relations Used in Transport Model.
        Region
Reynolds
 number
Critical  shear stress relation
           I      0.03 j 450.
               YCR  = 0.1  *  REYN
                                                      -0.3
               YCR  =  EXP  (-2.3026  -  .5546  *  log   (REYN))

               YCR  =  0.033

               YCR  =  EXP  (-3.9793  +  .19212 * logg  (REYN))

               YCR  =  0.06
                                           49

-------
A  flow diagram  of the  model is presented in Figure D-2 which  gives an overall picture of the
interactions between the various parts of the model.
HYDRAULIC AND SEDIMENT INFLOW
DATA FROM MAIN MODEL
1

                                      DOES FLOW RATE EQUAL ZERO?
                              WASHLOAD
               \
               /
                                 YES
                                                          LARGER
                                                         PARTICLES
                                            CALCULATE TRANSPORT CAPACITY
                                               OF SUSPENDED SEDIMENT
                                              GROSS TRANSPORT CAPACITY
                                             CALCULATE DETACHMENT RATES
                                           AND POTENTIAL SEDIMENT OUTFLOWS
                                             DISTRIBUTE TRANSPORT CAPACITY
                                              EXCESSES TO PARTICLES WITH
                                              TRANSPORT CAPACITY DEFICITS
IS THE TRANSPORT CAPACITY
GREATER THAN THE MAXIMUM
POTENTIAL SEDIMENT LOAD?
                                                         NO
                                              IS TRANSPORT CAPACITY LESS
                                           THAN THE POTENTIAL SEDIMENT LOAD J
                                              WITH NO FLOW DETACHMENT?
                                                         NO
                                            NO DEPOSITION, MAXIMUM RAINFALL
                                               DETACHMENT WITH PARTIAL
                                                  FLOW DETACHMENT
                                             APPLY CONTINUNITY EQUATION TO
                                              ROUTE SEDIMENT AND COMPUTE
                                               CUMULATIVE AGGRADATION
                                                                           YES
                            YES
                                              PARTITION SEDIMENT OUTFLOW
                                                TO ADJACENT ELEMENT
                                                    NEXT ELEMENT
                                        Figure D-2. Flow Diagram.
                                                    50

-------
A brief summary of the basic assumptions on which the model is based follows:
    1. The particle size distribution of detached sediment is the same as the weight fractions
      of the soil particles in the original soil mass (no enrichment during detachment).
    2. Rainfall detachment is not limited by the transport capacity of the flow.
    3. Flow detachment occurs only if there is excess transport capacity and can never
      exceed the transport capacity excess.
    4. Deposition and flow detachment never occur at the same time for the same particle.
    5. Washload transport is independent of the transport capacity of the flow and does not
      influence the transport of the larger particles.
    6. Deposited sediment requires the same amount of energy  as  in the  original
      detachment to become redetached.
    7. Enrichment is controlled by the  deposition process.
    8. The rate at which a particle will deposit is proportional to its fall velocity.
    9. Channel erosion does not occur.
   10. Subsurface or tile drainage produces no sediment.

                      Input and  Output Format Changes
Figure D-3 depicts the additional information required to run the expanded version of the ANSWERS
sediment model. The data block is inserted just after the soils information and before the drainage
and groundwater release information. It includes a description of each particle class (size, specific
gravity, and fall velocity—if known) and the make-up (in terms of the defined size classes) of each
of the listed soil types. In this example, there were five total size classes with two (the first two)
defined as washload classes. Additionally, there were ten soil types. Soil number three was
composed of the following percentages of the five size classes:
    1. 10 percent of class  1 (.002 mm),
    2. 18 percent of class  2 (.010 mm),
    3. 50 percent of class  3 (.030 mm),
    4. 17 percent of class  4 (.500 mm),
    5. 5 percent of class 5 (.200 mm).
                   PARTICLE SIZE AND TRANSPORT DATA FOLLOWS
                  NUMBER OF PARTICLE SIZE CLASSES   = 5
                  NUMBEF< OF WASH LOAD CLASSES       = 2
                    SIZE  SPECIFIC GRAVITY  FALL VELOCITY
                  0.002
                  0.010
                  0.010
                  0.50
                  0.20
                   .JOO
                   .100
                   .160
                   .080
                   .160
                   .100
                   .160
                   .024
                   .200
                   .200
         2.65
         2.65
         i.ro
         1.60
         2.65
                  0.0
                  0.0
                  0.0
                  0.0
                  0.0
.180
.180
,160
.120
.) 60
.180
.160
.031
.200
.20u
.500
.500
.530
.350
. 530
.500
.530
.240
.200
.200
.170
.170
.110
.150
.110
.170
.110
.239
.200
.200
.050
.050
.040
.300
.040
.050
.040
.466
.200
.200
                            3a
                  DRAINAGE COEFFICIENT FOR TILE DRAINS  =6.35 MM/24HR
                  GROUNDWATER RELEASE FRACTION =    .001

                         Figure D-3. Additional Input Information.
                                          51

-------
Classes 1,2, and 5 are defined as primary clay, silt, and sand, respectively. Classes 3 and4, as noted
by their lower specific gravity and larger sizes, are small and large aggregates, respectively.

The input data "echo" for the expanded  model shows the newly calculated fall velocity (unless
previously defined) for each size class and the "sand equivalent" diameter of particles with specific
gravities less than 2.65. Figure D-4 shows the data "echo" which is displayed immediately following
the soils information.
         q
        10
55.0

 1.0
 po.n
100.0
38.1
   0
127.0
 25.4
 .60
1.00
203.0
   .1
 70.0

100.0
.20

  0
PARTICLE SIZE DISTRIBUTION DATA
NUMBER OF PARTICLE SIZE CLASSES = 5
NUMBER OF WASHLOAD CLASSES = 2
CLASS
1
2
->
A
5
PARTICLE
CLASS
SOIL 1
SOIL 2
SOIL 3
SOIL 4
SOIL 5
SOIL 6
SOIL 7
SOIL 8
SOIL 9
SOIL10
DIA,MM
.002
.010
.•030
.500
.200
EQ3AND,MM





SIZE DISTRIBUTION
1 2
.100 .IPO
. 100 . 1RO
.160 .160
.080 .120
. 160 . 160
.100 .180
. IbO .160
.024 .031
.200 .200
. 200 . 200
3
.500
.500
.530
.350
.530
.500
.530
.240
.200
.200
.002
.010
.021
.278
.200





OF SOILS .AS
4
.170
.170
.110
.150
.110
.170
.110
.239
.200
.200
5
.050
.050
.040
.300
.040
.050
.040
.466
.200
.2.00
SG FALL VELOCITY, M /S
2.650
2.650
1.800
1.600
2. 650
DETACFIED
6 7










.0000036
.0000896
.0003910
.0432082
.0263375

P










       TILE DRAINAGE COEFF. = 6.35  MM/24H
       GROUNDWATER RELEASE FRACTION =  l.OOOE-03
                             Figure D-4. Data "Echo" in Output.


Immediately following the statistics on total rainfall and runoff and average sediment yield (at the
end of the printed hydrograph), statistics on the composition of the sediment leaving the watershed
are presented. Figure D-5 indicates that both of the washload classes (clay and silt) were
significantly enriched during this particular event.
                                            52

-------

3487.0
RUNOFF VOL
0 .1633 532016. 0.
JME PREDICTED FROM 63.40 MM OF RAINFALL
AVERAGE SOIL LOSS = 745. KG/HA
PARTICLE SIZE DISTRIBUTION
OF ERODED SEDIMENT
PARTICLE CLASS 1 = 24.4fi PERCENT
PARTICLE CLASS 2 = 38.85 PERCENT
PARTICLE CLASS 3 = 28.13 PERCENT
PARTICLE CIASS 4 = 3.45 PERCENT
PARTICLE CLASS 5 = 5.12 PERCENT

                                                       14.708  MM
                INDIVIDUAL ELEMENT NET SEDIMENTATION
ELEMENT SEDIMENT  ELEMENT SEDIMENT  ELEMENT SEDIMENT  ELEMENT SEDIMENT


              Figure D-5. Sediment Sizes Exiting Watershed.
                                 53

-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA-905/9-82-001
4. TITLE AND SUBTITLE
ANSWERS Users Manual
7. AUTHOR(S)
David B. Beasley and Larry F. Huggins
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Agricultural Engineering Department
Purdue University
West Lafayette, IN 47907
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Great Lakes National Program Office
536 South Clark Street, Room 932
Chicago, IL 60605
3 RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
December, 1981
6. PERFORMING ORGANIZATION
8, PERFORMING ORGANIZATION
CODE
REPORT NO
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
EPA-G005335
13. TYPE OF REPORT AND PERIOD COVERED
Computer Model Users Manual
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
      This report  is  an expanded and edited version of the Users Manual for
 the ANSWERS  (Areal Nonpoint Source Watershed Environment Response  Simulation)  model,
 first published in September,  1980.  ANSWERS is a distributed parameter model
 capable of predicting  the hydrologic and erosion response of primarily
 agricultural watersheds.   Particle-size distributions of the eroded sediment
 (at all points in the  watershed, as well as the outlet) are available.  The
 manual provides insights  into  model concepts, input requirements,  output
 interpretation, and  planning applications.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Water quality model
Hydrology
Sediment yield
Particle-size distributions
Best Management Practices
Runoff model
tS. DISTRIBUTION STATEMENT
Document is available from U.S. EPA,
Chicago and Agr. Engr. Dept. , Purdue Univ.,
West Lafayette, IN.
b. IDENTIFIERS/OPEN ENDED TERMS

19. SECURITY CLASS (This Report}
20. SECURITY CLASS f This page)
c. COSATI Field/Group

21. NO. OF PAGES
54
22. PRICE
EPA Form 2220-1 (9-73)
                                           54
                                                             U  S. Uovernmont Printing Office  1981  750-79S

-------