&EPA
United States
Environmental Protection
Agency
Environmental Research
Laboratory
Athens GA 30613
EPA-600/3-84-065
June 1984
Research and Development
Application Guide for
Hydrological Simulation
ProgramFORTRAN
(HSPF)
-------
EPA-600/3-84-065
June 1984
APPLICATION GUIDE FOR IIYDROLOG I CAL
SIMULATION PROGRAM - FORTRAN (liSPF)
by
Anthony S. Donigian, Jr.
John C. Imhoff
Brian R. Bicknel1
John L. Kittle, Jr.
Anderson-Nichols and Co.
Resources Technology Division
Palo Alto, CA 9^303
Contract No. 68-01-6207
Project Officer
Thomas 0. Barnwell
Technology Development and Applications Branch
Environmental Research Laboratory
Athens, GA 30613
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
ATHENS, GEORGIA 30613
-------
DISCLAIMER
The information in this document has been funded wholly or in part
by the United States Environmental Protection Agency under Contract No.
68-01-6207 to Anderson-Nichols and Co. It has been subject to the Agency
peer and administrative review, and it has been approved for publication
as an EPA document.
UjS. Environmental Protection
11
-------
FOREWORD
As environmental controls become more costly to implement and the
penalties of judgment errors become more severe, environmental quality
management requires more efficient analytical tools based on greater
knowledge of the environmental phenomena to be managed. As part of
this Laboratory's research on the occurrence, movement, transformation,
impact, and control of environmental contaminants, the Technology Develop-
ment and Applications Branch develops management or engineering tools to
help pollution control officials achieve water quality goals through water-
shed management.
The development and application of mathematical models to simulate the
movement of pollutants through a watershed and thus to anticipate environ-
mental problems has been the subject of intensive EPA research for several
years. The most recent advance in this modeling approach is the Hydrological
Simulation Program - FORTRAN (HSPF), which uses digital computers to simulate
hydrology and water quality in natural and man-made water systems. HSPF is
designed for easy application to most watersheds using existing meteorologic
and hydrologic data. Although data requirements are extensive and running
costs are significant, HSPF is thought to be the most accurate and appropri-
ate management tool presently available for the continuous simulation of
hydrology and water quality in watersheds.
William T. Donaldson
Act i ng Di rector
Environmental Research Laboratory
Athens, Georgia
i i i
-------
ABSTRACT
The nydrological Simulation Program - FORTRAN (HSPF) is a set of
computer codes that can simulate the hydrologic and associated water quality
processes on pervious and impervious land surfaces, in the soil profile, and
in streams and well-mixed impoundments. This document describes the entire
application process of HSPF to demonstrate the decisions, procedures, and
results that are involved in a typical application. The document is intended
as a supplement to the existing HSPF user's manual and programmer's supple-
ment. Together these three documents provide sufficient guidance for the
full and intelligent use of the broad range of capabilities of HSPF.
This report was submitted in partial fulfillment of Contract No.
68-01-6207 by Anderson-Nichols and Co. under the sponsorship of the U.S.
Environmental Protection Agency. This report covers the period from
March 1, 1981 to September 30, 1983. and work was completed as of
September 1983-
IV
-------
CONTENTS
Sec t i on
Abstract
Figures
Tables
Acknowledgments
1. Introduction
2. Study Delinition
2.1 Definition of Study Goals
2.2 Assessment of Data Availability
2.3 Assessment of Time and Resources ....
2.4 Study Definition Process for the Iowa
River Study
2.5 Summary
1 V
V 1 1
V 1 1 1
1 X
1
5
5
7
8
12
. . 15
Development of a Modeling Strategy 17
3.1 Selection of Constituents and Sources
to be Modeled '8
3.2 Preliminary Segmentation of Land Area
Based on Weather Data 27
3.3 Final Segmentation of the Land Area 38
3.1 Segmentation and Characterization of the
Channel and Contributing Areas ^7
3.5 Characterization of Special Actions 55
Operational Aspects of HSPF Use 57
4.1 Steps in Running HSPF 57
4.2 Overview of HSPF Input 58
4.3 Output Options 62
Input and Management of Time Series Data 70
5.1 Creation of Time Series Store (TSS) 70
5.2 Adding Dataset Labels 72
5.3 Input of Data 72
5.4 Management of TSS Datasets 75
Model Parameters and Parameter Evaluation 77
6. 1 Types of Data Needed 78
6.2 Sources of Data 78
6.3 General Considerations 82
-------
7. Calibration and Verification 8k
7.1 General Calibration Procedures 84
7.2 Calibration Guidelines for Major
Constituent Groups 89
7.3 How Much Calibration ? 112
7.4 Verification 114
8. Analysis of Alternate Conditions 116
8.1 Philosophy Underlying Comparison
of Alternatives 116
8.2 Steps in the Analysis Process 118
8.3 Examples of Analyzing Alternatives
with HSPF 119
9. References 138
Appendices 140
A. Sample HSPF Input Sequence 140
B. Use of the NETWORK Block to Connect the Surface
and Instream Application Modules 167
C. Equivalency Table for Selected HSPF and ARM/NPS
Parameter Names 170
v i
-------
FIGURES
Figure Page
2.1 Representative HSPF Project Schedule ........ 13
3.1 Meteorologic and U.S.G.S. Gaging Stations
in and near the Iowa River Basin .......... 31
3.2 Isopleths of Mean Annual Precipitation and
Potential Evapot ranspi rat i on in Iowa ........ 32
3.3 Isopleths of Mean Annual Temperature in Iowa. ... 34
3.4 Preliminary Segmentation of the Iowa River Basin
to Account for Variability in tleteorol og ic
Patterns and Soils Characteristics ......... 41
3.5 Channel Reaches and Contributing Areas
for the Iowa River Basin .............. 42
3.6 Final Segmentation of the Iowa River Basin ..... 43
3.7 Iowa River Low-Water Profile ............ 50
4.1 Sample Short-span Display (first type)
from the DISPLY Module of HSPF ........... 6k
1.2 Sample Short-span Display (second type)
from the DISPLY Module of HSPF ........... 66
4.3 Sample Long-span Display (annual)
from the DISPLY Module of HSPF ........... 6?
5.1 Example of User's Control Input for the
COPY Module .................... k
7.1 Example of Response to the INTFW Parameter ..... 93
8.1 Frequency Curves for Simulated Ammonia and
Nitrate at Marengo, Iowa .............. 128
8.2 Lethality Analysis of Chemical Concentration. . . . 130
8.3 Locations of the 21 Dam Sites for Power
Generation in the Rio Yaque del Norte Watershed,
Dominican Republic ................. 132
8.4 Clinton River Drainage Basin, Michigan ....... 13^
8.5 Dunn-Wilcox Watershed ............... 135
8.6 Hydrograph of Reach 941 for June 26, 1968 Event . . 137
vi i
-------
TABLES
Table Page
2.1 HSPF Release 7.0 Run Costs 11
3.1 Constituent Hierarchy in HSPF for
Instream Modeling 20
3.2 Meteorological Time Series Data Requirements
for HSPF 28
3.3 Summary of Meteorologic Data Used to Represent the
Three Segment Groups of the Iowa River Basin. ... 36
3.4 Definition of Pervious Land Segments for the
Iowa River Basin 45
3.5 Land Use in the 13 Contributing Area
Subdivisions in the Iowa River Basin 46
3.6 Reach Characteristics for the Iowa River 5^
4.1 HSPF Input Blocks and Recommended Sequence 60
4.2 Examples of Input Blocks Required for HSPF Runs . . 61
4.3 Operations Performed by the GENER Module of HSPF. . 69
6.1 Types and Sources of Data Needed to Use the
Various Sections of the HSPF Application Modules. . 80
8.1 Selected Alternatives, Associated HSPF
Assumptions, and Suggested Input Modifications. . . 120
8.2 Selected BMP Scenario for Simulation on the
Iowa River Basin 123
8.3 Comparison of Edge-of-Stream Loadings for Base
Conditions and BMP Simulations in the Iowa
River Basin 124
8.4 Comparison of Loadings in the Iowa River at
Marengo for Base Conditions and BMP Simulations . . 126
8.5 Lethality Analysis of BMP Scenario for Alachlor
in the Iowa River at Marengo, Iowa 131
8.6 Comparison of Maximum Flows for Reaches with
Channel Storage 137
VIII
-------
ACKNOWLEDGMENTS
This uork was sponsored and supported by the U.S.
Environmental Protection Agency. Mr. Thomas Barnuell of the
Environmental Research Laboratory, Athens, GA was Project
Officer; his assistance and guidance has contributed to the
successful completion of this work and is gratefully
acknowledged.
Among the authors, Mr. Anthony Donigian uas the Project
Manager with overall responsibility for technical direction,
supervision, and review. He was also the major author of
the final section of this document, analysis of alternate
scenarios. Among the authors, Mr. John Imhoff was the Task
Leader for this project and initial author for the sections
describing procedures for study definition, development of a
modeling strategy, parameter evaluation, and calibration and
verification. Mr. Brian Bicknell was responsible for the
sections pertaining to the operational aspects of HSPF and
the input and management of time series data. Mr. Jack
Kittle provided significant technical review and guidance,
and was also the key source on all HSPF operational
quest ions.
In addition to the authors, several other individuals at
Anderson-Nichols were active in preparation of this
document. Ms. Kathyrn Lahanas and Ms. Mary Maffei provided
report typing and text editing throughout the project, and
Ms. Virginia Rombach prepared the report charts and figures,
and assisted in final preparation of the document. The
dedication and efforts of these individuals contributed to
the success of the project.
IX
-------
SECTION 1
INTRODUCTION
This document describes the entire application process of
the Hydrologic Simulation Program - Fortran (HSPF) using the
loua River Basin Study CImhoff et al., 1983) to demonstrate
the decisions* procedures, and results which are involved in
a typical HSPF application. The document is intended as a
supplement to the existing User's Manual (Johanson et
al.,1981*) and Programmer's Supplement (Johanson et
al.,1979). Together these three documents provide
sufficient guidance to allow the user to make full and
intelligent use of the broad range of capabilities contained
in HSPF.
The User's Manual provides instructions for building input
sequences and explains the basis for the simulation
algorithms. Included in the User's Manual are an
explanation of basic model concepts, programming standards
and practices, a visual table of contents of program
components, functional descriptions of subprograms, and
format information for the User's Control Input.
The Programmer's Supplement permits the user to follow the
inner workings of the model. Program code, in the form of
IBM pseudocode (IBM, 1974), data structures and file
structures, and sample input sequences and results are
included. The Programmer's Supplement is contained on
magnetic tape.
While the User's Manual and Programmer's Supplement provide
a systematic and comprehensive description of model contents
and operational procedures, many questions which are
critical to the intelligent use of HSPF are left unanswered.
Additional guidance is needed to answer such user questions
as :
(1) How can I develop a modeling strategy which
will address the problems I need to analyze?
(2) What kinds of data do I need for my modeling
effort, and where can I get this data?
-------
(3) What model parameters are most critical to my
application, and how do I develop the most
reasonable values for these parameters?
(4) What is involved in the model calibration and
verification process, and how much calibration
effort is necessary before I can use the model
to analyze my problems?
(5) Once the calibration and verification process
is complete how can I use the model to evaluate
the effects of alternate practices?
(6) How can I use the model's capabilities to
provide me with results which are the most
informative and the most useful for
interpretation and presentation?
The purpose of this document is to answer these and related
questions concerning the application of HSPF to engineering
and planning studies. The discussion of the application
process is divided into the following seven major steps
which are necessary to perform a complete model application:
Study Definition
Development of a Modeling Strategy
Learning the Operational Aspects of HSPF Use
Input and Management of Time Series Data
Parameter Development
Calibration and Verification
Analysis of Alternate Scenarios
The "study definition" process involves (1) identification
of the questions which the model application must answer,
and determination of the level of detail required to answer
these questions; (2) assessment of the availability of
supporting data and its usefulness to the modeling effort;
and (3) comparison of the time and money available to
perform the modeling effort with estimates of resources
required for the intended application.
Successful application of HSPF to a study area requires the
development of a simulation plan or strategy, based on
characterization of the area with regard to meteorologic
conditions (and spatial variability), soils characteristics,
topography, land use, pollutant sources, available historic
-------
data, etc. The purpose of this section is to outline the
general characterization process.
An important step in applying HSPF is familiarizing oneself
with the mechanics of the model so that the input sequences
necessary to build the time series data base (Time Series
Store) and execute simulation runs can be developed. The
goal of this section is to provide an overview of
considerations involved in running HSPF and developing input
sequences, and to direct the user to the proper places in
the User's Manual for additional information.
All HSPF simulation runs involve the use and/or generation
of data in the form of time series. This section describes
the storage, retrieval and management of time series data
using HSPF utility routines, stand-alone programs and a
large random access file known as the Time Series Store
(TSS).
Parameter development focuses on the process-oriented
parameters needed as input to the application modules of
HSPF. Since the model is designed to be applicable to many
different watersheds and water systems, these parameters
provide the mechanism to adjust the simulation for specific
topographic, hydrologic, edaphic, land use, and stream
channel conditions of a particular area. The parameter
development section is designed to familiarize the user with
the types of data which are needed for parameter evaluation
and to direct the user to existing data and documents which
will prove useful in the evaluation process.
Calibration is the process of adjusting selected model
parameters within an expected range until the differences
between model predictions and field observations are within
selected criteria for performance. It is required for
parameters that cannot be deterministically evaluated from
topographic, climatic, edaphic, or physical/chemical
characteristics. Verification is the complement of
calibration; model predictions are compared to field
observations that were not used in calibration. In essence,
verification is an independent test of how well the model
(with its calibrated parameters) represents the important
processes occurring in the natural system. The
calibration/verification section provides recommended
procedures and guidelines for the major sections and
constituents of HSPF.
Because of the comprehensive scope of HSPF, once it has been
applied (i.e., calibrated/verified) to a watershed system it
can be subsequently used to analyze a variety of proposed or
projected alternative conditions. In this process the
calibrated/verified model is used to project changes in
-------
system response resulting from a proposed alternative; this
alternative is represented in HSPF by adjustments (changes)
to model input, parameters, and/or system representation
(e.g., interconnection of PLSs and stream reaches). This
section discusses the basic philosophy underlying the use of
HSPF for analysis of alternatives, enumerates the various
steps involved in this process, provides guidance in
analyzing selected alternatives, and describes related
examples drawn from past experience with HSPF and/or
predecessor models.
In describing the general application process, ue make
numerous references to the Iowa River Basin Study, uhich uas
a preliminary application of HSPF to model water quality and
the effects of agricultural best management practices
(BtlPs). While no one example application can serve to
demonstrate the extensive capabilities and potential diverse
applications of the model, the Iowa River project
illustrates many of the decisions, procedures, and results
involved in using HSPF.
At each step in the application process ue will first
explain what needs to be done; then explain how it was done
in the Iowa River project; and finally discuss additional
considerations and/or actions which may be necessary for
different types of applications. Thus, while the previously
existing documentation instructs the user on HSPF model
contents and operational procedures, this document is
primarily designed to instruct the user on how to use the
model to analyze engineering and planning problems in an
intelligent manner.
The user should note that the Iowa study required the full
range of HSPF capabilities from data management to
pesticide runoff and soil simulation to instream sediment
transport and pesticide fate modeling. Many user problems
and potential applications will require only ssubsets of HSPF
capabilities and significantly less resources.
-------
SECTION 2
STUDY DEFINITION
A realistic assessment of study goals and resources at the
beginning of a modeling project is critical to the
development of an effective modeling strategy and an
appropriate data base. In fact, project goals and resources
will affect every step of the modeling process. A
reasonable division of time and effort between the
individual steps of a complete model application can only be
achieved by careful consideration of the required end-
products of the project and the time and money available to
produce these end-products. The "study definition" process
can be divided into three major tasks:
(1) Identify the questions which the model
application must address, and determine the
level of detail and model accuracy required to
analyze and answer these questions.
(2) Assess the availability of supporting data and
its usefulness to the modeling effort.
(3) Compare the time and money available to perform
the modeling effort to guidelines for the
effort and costs involved in an HSPF
application as outlined in this document.
Each of these three tasks is considered in more detail
belOM.
2.1 Definition of Study Goals
Clearly defined study goals are needed every step of a model
application. Quite often the goals stated at the beginning
of a modeling study are too ambitious or too vague. A study
workplan may call for an "evaluation of watershed water
quality" or a "complete investigation of hydrologic
resources." Without further refinement, such goals do not
provide the model user with a clear understanding of what
information is needed from the model application. As an
example, consider a study which calls for an evaluation of
the effects of tertiary treatment of domestic wastewater on
-------
the quality of downstream receiving waters. Given that HSPF
is capable of modeling nearly 20 individual water quality
constituents, it is essential that the modeling effort be
restricted to critical constituents based on an
understanding of (1) the constituents which are most
affected by the treatment practice and (2) the constituents
which exert the most influence on the overall quality of the
receiving waters. If in this case the study goal can be
refined and stated as "an evaluation of the effects of
tertiary treatment on concentrations of dissolved oxygen,
BOD, ammonia, and nitrate in receiving waters," considerable
effort can be saved in development of the modeling strategy,
data acquisition, parameter evaluation, etc.
While it is wise to acquire and examine all existing data
which could prove useful to a modeling effort, it is
essential to concentrate one's effort from the beginning of
the study on data pertinent to the critical constituents
which will be modeled. Development of data for constituents
which will not be modeled can often squander time and
resources needed at later stages of the model application.
Further detail on selecting appropriate constituents is
provided in Section 3.1.
Many of the issues involved in properly defining a study are
related to requirements for spatial or temporal definitions,
or to the 1evel-of-detai1 needed to answer the study
questions. Early recognition of the spatial and temporal
definition required in the model representation assures the
development of an appropriate modeling strategy. Comparison
of a "wasteload allocation study" to a "watershed water
quality study" serves to illustrate the importance of
recognizing spatial definition requirements.
Hast eload allocation study.
The goal of such a study is to determine an equitable
distribution of chemical loadings to the receiving waters
from existing point sources in a watershed. The resulting
composite loadings must not violate water quality standards
at any point along the channel system. To perform such a
study it is necessary to analyze the effects of each major
point source individually; and thus, detailed data on point
source contributions and channel characteristics are
required. Both factors are pertinent to the development of
the model representation and the model data base.
Watershed water quality study.
The goal of a watershed-oriented study might be to assess
the overall chemical loadings at the downstream terminus of
a stream or river. For such a study, a number of
simplifications can be made in the representation of point
loads. For example, channel reaches can be defined based on
-------
such iactors as hydrogeometric and hydraulic characteristics
and/or reaction rates of critical constituents Following the
definition of the reach system, point source data from all
contributions to a reach can be combined without
significantly affecting model results at the downstream
terminus.
Understanding the requirements for temporal definition in a
study can have an equally significant role in development of
the modeling strategy. For example, the importance of
timing of flow, and hence magnitude of peak flows, to study
results may determine whether or not hydraulic routing is
required as a component of the modeling effort. A study to
determine expected annual runoff at a potential reservoir
site may not require stream routing of runoff, because
determination of the maximum instantaneous flow will not
influence to the study results. On the other hand, accurate
representation of peak flows may be critical to a design
study for a flood control structure.
Precise statement of study goals with careful consideraton
of the spatial and temporal modeling detail necessary to
answer the critical study questions will vastly improve the
likelihood of a successful model application. Additional
issues concerning 1evel-of-detai1 are critical to every step
of the simulation process. For example, the model user must
assess the appropriate detail for representing the
constituent sources and processes which are modeled. Only
those constituent sources and processes which are likely to
have a significant effect on study results should be
included in the modeling effort. The goal is to achieve a
suitable fit between the planned modeling effort and the
data, time, and money available to perform the study. The
role that project resources play in determining realistic
and realizable study goals is discussed in the following
sections (Section 2.2 and 2.3).
2.2 Assessment of Data Availability
Effective use of HSPF requires considerable data to
characterize watershed land use, soils, and meteorology; for
model applications in which channel processes are important,
additional data on streamflow, channel geometry, and
instream chemical concentrations are necessary. Sufficient
knowledge of the physical, chemical, and biological
characteristics of the study area must also be available to
develop numerous parameter values. Subsequent sections of
this document will provide guidelines for the proper
selection and use of all these different kinds of data. The
purpose of this discussion is to emphasize that a model user
must collect and assess available data at the beginning of a
-------
study in order to assure that sufficient data exists to
allow confidence in model results.
Model results can be only as good as the data used to apply
the model. If the data used as input to HSPF is accurate
and comprehensive, the model user can have more confidence
that the model representation is appropriates for the study
area. When simulation results have been produced, they must
be compared to additional data such as obse:rved streamflou
or instream chemical concentrations. A good comparison
between simulated and observed values indicates that the
model algorithms adequately represent the critical processes
in the study area. Unfortunately, a modeler never has all
the data needed to fully represent the study area and verify
simulation results. Filling in missing input data for a
study area based on general knowledge, data from other
watersheds, and previous modeling experience can provide
reasonable simulation results in many cases. The degree of
confidence given to these results should reflect the amount
of missing data, the reasonableness of the assumptions used
in filling data gaps, and the amount of observed data
available to verify the simulation results.
Scarcity of observed data to verify simulation results can
significantly weaken confidence in model results and hence
the achievement of study goals is threatened. At the
initial stage of model application, it is critical that the
user assess whether or not adequate observed data exist to
verify model results. Data must represent the spatial and
temporal variations in flow and/or chemical loadings
resulting from the combined meteorologic, hydrologic,
chemical, and biological processes of the study area. While
an adequate record of meteorologic and hydrologic data
exists for most areas, water quality data are frequently of
poor quality due to infrequent sampling, time-composited
samples, etc. If insufficient data exist to verify the
model results, a supplementary sampling program should be
considered. In many cases a modeling study may not achieve
its goals if simulation results cannot be substantiated by
observed data.
2.3 Assessment of Time and Resources
Data is not the only resource which is important to defining
and analyzing study goals - the time and money available to
perform the study are equally critical. This section
provides preliminary guidelines for the time and costs
involved in modeling studies using HSPF.
HSPF is a new model, with its initial release occurring in
1979. While a number of HSPF applications are in progress,
-------
feu studies are complete; consequently information on time
and costs associated with model application is limited to a
feu pilot studies. The potential model user should use the
guidelines presented belou to make a preliminary assessment
on whether or not the planned model application can be
performed uithin the time and budget available. The
guidelines uere derived primarily from modeling studies
performed by staff who uere heavily involved in the HSPF
model development; lack of familiarity with the model will
increase the time and effort required for model application.
Three topics will be discussed:
(1) Amount of time and effort required for
representative applications (including computer
costs)
(2) Relative effort involved in the seven steps of
model application
(3) Relative timing for performance of the seven
application steps.
The following estimates of level-of-effort, computer costs,
etc., required for representative applications are based
upon two recent pilot applications: the Four Mile Creek
Basin near Traer, Iowa, (Donigian et al., 1983b) and the
Iowa River Basin located in central Iowa. Both studies
involved land surface and instream modeling of runoff,
sediment, and chemicals on agricultural watersheds. In the
Four Mile Creek application, d etai1ed calibration and
verification of the model was performed for three small
field sites each representing a separate land use activity:
corn and soybean cropland and pasture. Simulation periods
were six months for pesticide calibrations and twelve months
for agricultural nutrients. Subsequently, the results were
extrapolated to the entire watershed where the same
constituents were modeled on three land segments and the
results used as loadings to an eight-reach stream system.
Less detailed calibration was performed at the watershed
level where the simulation periods ranged from four months
to thirty months, and two separate agricultural practice
scenarios were simulated.
In the Iowa River study, the methodology developed on Four
Mile Creek was extrapolated to the 7200 sq. km. Iowa River
Basin to demonstrate its applicability on a large river
basin. For modeling purposes, the study area was divided
into nine pervious land segments in order to represent
variability in meteorology, topography, soils, land use, and
agricultural practices and chemical applications (see
Section 3). Runoff and associated loadings of sediment,
inorganic nitrogen, and one pesticide were simulated for a
-------
five year period and were used as input to a thirteen- reach
channel system. Hydraulic routing and instream chemical
reactions were simulated for the 300 kilometers of the loua
River upstream of Marengo, Iowa. The simulation was limited
to approximately six calibration runs for hydrology and
sediment; full scale simulation runs for inorganic nitrogen
and pesticide were performed for two different scenarios
without calibration due to lack of observed data.
As an aid to the user in projecting computer costs, Table
2.1 presents the actual execution time and costs for
representative one-year simulation runs from a number of
applications. It is important to remember that these run
costs are highly dependent on the computer rate structure,
output options such as plots and displays, and other
factors. The user should note carefully what is included in
each of these run descriptions when estimating his own
computer costs. In addition, a significant fraction of the
computer costs incurred by a user (and not considered in
Table 2.1) may be associated with input sequence development
during interactive sessions at a computer terminal.
A major consideration in any application is the division of
the available resources among the tasks to be performed.
Shown below is a representative breakdown of the application
effort into the steps discussed in Section 1, through
calibration and verification; the analysis of alternatives
is excluded because the effort will be highly dependent on
the projected use.
TASK % EFFORT
Problem Definition 5
Modeling Strategy 10
Learn Operational Aspects 10
Development and Input of Time Series 30
Parameter Development 15
Calibration and Verification 30
This table is intended as a guide; the relative effort for
the various steps of an HSPF application will differ from
study to study. For example, application to an area which
has been modeled previously using HSPF will require less
effort for parameter development and calibration due to
knowledge of watershed characteristics. Also, this
distribution may vary considerably depending on the
familiarity of the user with HSPF and experience in its use.
In addition to the division of total effort into the
separate tasks of an application study, the relative timing
for the start and completion of each task should also be
considered at the beginning of the study. Inevitably,
10
-------
T.-.ble 2.1 HSPF Release 7.0 Run Costs
Computers IBM 3081 at Stanford University - Canter for Information Technology
CPU Rate: S23.10/cpu minute (nisht/weeKend)
Ois* I/O Rate' 40.325/1000 (nignt/weekend)
Q.W DESCRIPTION
EXECUTION TIMg DISK I/O ="?IHT
ni nA-r $Afr No./vr ?A-r COST
I. NEWTSS Run - create a new TSS file
0.
0.
121 0.10
0.69
0.71
2. TSSM and COPY Run - create 4 data
labels in the TSS and transfer 4 time
series 13 daily; 1 hourly) into the
datssets. Display tha time series.
0.06 1.39 236* 1.97
5.90
3. PERLND Run - 1 land segment (PWATER),
2 displays, 1 plot. INDELT = 1 hour.
4. PERLHD Run - 3 land sesments (SNOW,
PMATER), 4 displays, 1 plot, 2
duration analyses. INDELT = 2 hr.
0.10 2.31 138GQUAL), 36 displays,
10 plots. INOELT = 1 hour.
7. Watershed Run (Agrlc. nutrients) -
3 land segments (2 Hith 3 BLKS)
(SNOU,PWATER,SEDMNT,PSTEMP,PWTGAS,
MSTLAY,NITR.PHOS.TRAC), 8 stream
reaches (HYDS.ADCALC,CONS,OXRX,
NUTRX), 53 displays, 9 plots.
INDELT = 1 hour.
<».15
(»
95.37
4*. 51 )
-------
delays in completion of one or more tasks will occur, and
the project schedule may be extended; however, many of the
tasks involved in a modeling study may overlap;
consequently, delays in completion of the overall project
can be minimized. Due to differences in goals and modeling
strategy, the schedule for one project may be quite
different from another. For example, defending on the
availability of data. Task t 4, input and management of time
series, may begin very early in the schedule, whereas
calibration must auait some parameter development. By
necessity, production runs related to a specific constituent
or process cannot start until calibration/verification of
that constituent is complete. In order to provide a guide
to the user, a representative project schedule based upon
the Iowa River and Four Mile Creek studies is presented in
Figure 2.1.
In summary, this section is intended to emphasize the
importance of considering the specific budgetary and time
requirements of an HSPF application during the study
definition, and particularly to provide a guide to the user
for estimating the resources required and! the relative
timing of the project tasks. While model applications may
differ greatly in scope and purpose, it is hoped that the
representative data derived from pilot studies and presented
here will be useful in this process.
2.1 Study Definition Process for the Iowa River Study
This discussion illustrates how the guidelines developed in
Sections 2.1-2.3 were used to define a realistic scope of
work for the Iowa River Study. As noted previously, the
Iowa River Study was a demonstration application of HSPF on
a large river basin to evaluate the effects of agricultural
nonpoint pollution and proposed best management practices
(BMPs). Since the study was intended to demonstrate a
methodology, its goals were somewhat different than those of
most engineering applications in that study results were not
intended as a basis for making specific engineering or
planning decisions. Nonetheless, modeling results had to be
reasonable in order to demonstrate the validity of the model
algorithms and the modeling approach. In defining a clear
set of goals for the Iowa River Study the following factors
were significant:
(1) The primary intent of the study was to
extrapolate a methodology developed on nearby
Four Mile Creek (52 km2) to the Iowa River
Basin (7240 km2) to demonstrate its
applicability and functionality on a large
river basin. Consequently, considerable
12
-------
I
I
I
1
c
o
s-
o
in
.*:
OJ
Ol
2
0)
U
o
a
a,
K
8!
-H
4J
fl
4J
C
0)
(0
(U
i-l
a;
CM
0)
J-l
3
Cn
H
t,
CJ
a
JS
§
o
tt
et
O
li
u_
c/)
oo
ca
o
on
a.
a:
l
oo
a
o
M
O£.
UJ
a.
o
CQ
O
13
-------
information on soils, topography, land use, and
meteorology had already been gathered for the
central Iowa area. Model results from Four
Mile Creek were available to give some idea of
the hydrologic response of the regiion. In
addition, useful information on farming
practices (tillage, fertilizer and pesticide
application) had been gathered for the Four
Mile Creek Study, and reasonable reaction rates
for chemicals had been determined. This wealth
of data and experience from the HSPF
application on Four Mile Creek, provided major
benefits for the Iowa Basin Study.
(2) The major nonpoint source pollution problems in
Iowa were identified in the literature as
sediment erosion, and nutrient and pesticide
runoff. All three contaminants were modeled in
the Four Mile Creek Study.
(3) Immediately prior to the Four Mile Creek Study,
we had enhanced the HSPF capabilities with
improved algorithms for sediment transport and
reaction and transport of generalized
nonconservative chemicals, such as pesticides.
Initial demonstration of the improved
capabilities was performed in the Four Mile
Creek Study, and an important aspect of the
Iowa River Study was to expand the
demonstration of these new capabilities.
(t) Data gathering efforts for the Iowa River
yielded adequate streamflow, sediment, and
nutrient data to judge the reasonableness of
subsequent model results.
(5) The best and most abundant data for the Iowa
River was collected at Marengo, Iowa, upstream
from the Coralville Reservoir. This suggested
that Marengo would serve well as the terminus
of the modeled area.
(6) Time and level-of-effort limits for the Iowa
River Study (8 months and 1400 person-hours,
respectively) were sufficient to demonstrate
the methodology, but it was evident that
detailed calibration/verification for all
modeled constituents could not be performed and
that the number of BMP scenarios modeled would
have to be limited. These limitations were
deemed reasonable for a demonstration project.
-------
Based on the above listed considerations, we refined the
study goals to include the following points:
(1) The study area wns restricted to the watershed
above Marengo, Iowa.
(2) The modeling effort was restricted to
hydrology, sediment, nutrients, and one
pesticide.
(3) Data acquisition was limited to material useful
in modeling these four constituents.
(4) The planned calibration/verification effort was
limited. The goal of calibration would be a
general agreement between simulated and
observed values primarily for the flow and
sediment; no further refinements would be made
(5) Simulated BMP scenarios would be limited to one
or two depending on remaining resources in the
later stages of the modeling effort.
The concise scope of work developed above allowed us to
design a modeling strategy which would realize study goals
in an efficient, cost-effective manner.
2.5 Summary
Depending on project goals and resources, the amount of
effort devoted to many aspects of a modeling study can
either be reduced or expanded. Areas of the model
application which exhibit the most flexibility with respect
to required level of effort include the following:
complexity of land and channel segmentation
chemical sources and constituents considered in
the s imulat i on
s impli f i ed
algori thms
versus
d eta i1ed
simulation
level of detail and effort for
calibration/verification process procedures
number and level of detail for analysis of
alternate scenarios
15
-------
All of these topics will be discussed in more detail in
subsequent chapters. It is evident from the above list that
the relative effort devoted to the various steps of a
modeling study can be modified to a certain extent at any
point in the project. Generally speaking,, however, a
modeling study is most likely to be successful if major
changes are not made to the modeling strategy and scope of
work in the later stages of the project unless they are
absolutely necessary. Careful definition oJE study goals,
followed by development of an appropriate and comprehensive
modeling strategy is needed for efficient performance of all
steps of the model application.
16
-------
SECTION 3
DEVELOPMENT OF A MODELING STRATEGY
The second step in applying HSPF to a study area is the
development of a simulation plan or strategy, based on
characterization oi the area with regard to meteorologic
conditions (and spatial variability)> soils characteristics.
topography, land use, pollutant sources, available historic
data, etc. Meteorologic data must be identified which are
representative of the various segments of land to be
modeled. A basin segmentation scheme must be developed
which defines areas of homogeneous hydrologic response based
on soils characteristics and land use, as well as weather
conditions. A representative channel system including both
hydraulic and geometric characteristics is needed.
Streamflou and water quality data which can be used to
calibrate the model must be examined, and a modeling
strategy which makes full use of available data must be
devised.
The relative importance of various pollutant sources must be
ascertained. For those pollutant sources which are deemed
significant to model results, a general characterization of
pollutant behavior (accumulation, removal, influence by, and
response to land use activities) must be defined. The
purpose of this section is to outline the general
characterization process. Frequent references to the Iowa
River Study are made to illustrate the process and decisions
involved in developing a modeling strategy. Important
considerations in developing modeling strategies for other
applications are noted. The discussion is divided into five
subsections:
selection of constituents and sources to be
modeled
preliminary segmentation of land area based on
weather data
final segmentation of the land area
segmentation and characterizaton of channel and
contributing areas
17
-------
characterization of special actions or events
3.1 Selection of Constituents and Sources to be Modeled
An important first step in developing the modeling strategy
for a study is to decide which constituents will be modeled.
Concurrently, the user must assess which sources of
constituents (e.g., point loadings, nonpoint loadings,
chemical transformations, instream sources) are significant
to the water and chemical mass balances for the study area,
and how to characterize these sources for modeling purposes.
This section provides the first-time model user with general
guidelines for accomplishing these tasks.
Select i on of Const i tuents. As discussed in Section 2.1, the
choice of which constituents will be modeled is strongly
influenced by study goals and resources. All constituents
modeled by HSPF are key indicators of one or more different
aspects of water quality. For example dissolved oxygen,
water temperature, and sediment are key constituents which
must be considered if maintaining a suitable environment for
fish is a study concern. On the other hand, nitrates,
phosphates, and pesticides are critical constituents when
evaluating the impacts of nonpoint source pollution from
agriculture. In every case, study goals will necessitate
the modeling of certain constituents, while others will not
be nearly as critical to answering study questions.
Generally speaking, in order to conserve project resources,
one should avoid modeling constituents which are peripheral
to the main concerns of the study.
The resources available to perform a study are an important
factor in the selection process. By consulting others
involved in the application of HSPF and by reviewing the
general cost and effort guidelines for u:;ing the model
(Section 2.3), one should assess whether or not a reasonable
list of constituents has been selected for simulation. At
the same time the user must consider whether or not
existing data is adequate to characterize important
constituent sources and processes and to allow reasonable
calibration and verification of the model. While data
deficiencies do not preclude the modeling o:E a constituent,
one must give careful consideration to validity of results
which are not supported by good data.
An additional factor which must be considered if
constituents other than water are to be modeled is the
hierarchical nature of biochemical interactions. Due to the
interrelationships which exist between various constituents
and processes, the simulation of some constituents cannot be
18
-------
carried out independently of others. In all three modules
(PERLND, IMPLND, RCHRES) water must be simulated (or
available from a previous run or observed data) if any other
constituent is to be simulated. While modeling conventions
and simplifications in the land surface modules (i.e.
PERLND, II1PLND) allow the independent simulation of specific
constituents, a good deal of interdependency is exhibited by
the constituents and instream processes modeled in RCHRES.
For example, while water temperature is not affected by any
other simulated constituent, dissolved oxygen concentrations
are dependent on water temperature and cannot be simulated
independently. Most of the constituents which are modeled
in RCHRES are in some way related to other constituents.
Table 3.1 shows the hierarchy of dependency for RCHRES
const i tuents.
For example, if phytoplankton growth dynamics were the
subject of study, then water temperature, dissolved oxygen,
biochemical oxygen demand, nutrients, and zooplankton (i.e.,
groups 4, 7, 8, and 9) must be modeled in order to fully
model phytoplankton population fluctuations. However, if a
chemically conservative substance such as total dissolved
solids were the only constituent of interest, simulation of
additional constituents is not necessary.
Each constituent within each group does not need to be
simulated. There are allowable variations and minimum
criteria established for each group. The functional
description portions of the User's Manual (Part E, Sections
4.2(1)-H.2(3)) describe the allowable combinations of
constituents within each group and should be reviewed before
the final selection of constituents is made.
While the interdependencies discussed above usually require
that additional constituents be simulated, sometimes these
requirements may be satisfied by a user-input time series.
When available, this option may be preferable in situations
where the required data is easy to estimate or will have
minimal effect on the primary constituents to be simulated.
For example, if the temperature dependence of instream
chemical processes is low, the use of an approximate water
temperature time series is appropriate. Or, if it is known
that suspended sediment concentrations are generally low,
user-estimated time series for use in the instream
photolysis and photosynthesis algorithms are preferable to
the added cost and data requirements of performing a
detailed sediment simulation. The user should note,
however, that this option does not eliminate the
interdependencies specifically within section RQUAL;
simulation of plankton, for example, always requires
simulation of dissolved oxygen, BOD, and instream nutrient
processes.
19
-------
TABLE 3.1 CONSTITUENT HIERARCHY IN HSPF FOR INSTREAM MODELING
GROUP *. CONSTITUENTS GROUP DEPENDENCY
1 hydraulics (water) none
3 conservatives 1
4 water temperature 1
5 inorganic sediment 1,4 **
6 general quality constituent 1,4 ***
7 dissolved oxygen, BOD 1,4
8 inorganic N and P 1,4,7
ammonia
nitrate
nitrite
phosphate
9 plankton 1,4,7,8
phytoplankton
zooplankton
benthic algae
organic N,P,C
10 pH, inorganic carbon 1,3,4,7,8,9
PH
carbon dioxide
total inorganic carbon
alkalinity
* group numbers correspond to module section numbers used
in the Activity Block of RCHRES
** water temperature required if Colby method used for
simulating sand; user may either simulate water
temperature or provide an input time series
*** simulation may be dependent on additional constituents
depending on the algorithm options which are used;
refer to functional descriptions of module section
GQUAL in the User's Manual.
20
-------
Based on the above discussion a reasonable procedure for
selecting the constituents to be modeled is outlined belou:
1. Review project goals and the questions which
must be answered by modeling.
2. Establish which constituents modeled by HSPF
are the best indicators for addressing these
questions* and make a preliminary list of these
const i tuents.
3. Review project resources to make sure that
sufficient time, money, and data are available
to support the simulation of the constituents
contained on this list. If not, review step t2
and reduce the list to an appropriate length.
4. If instream simulation will be included in the
modeling effort, refine the preliminary list to
include constituents which must be modeled or
input due to constituent interdependences. Re-
evaluate available project resources.
Determination of Const ituent Sources to be Modeled. There
are six possible sources of water and/or other constituents
which are modeled by HSPF:
initial storages
nonpoint loadings (including atmospheric
deposition)
point loadings
chemical transformations
releases from the channel bottom
atmospheric gas invasion
Of these, the first three listed are the only sources of
water, while all six are potential sources of other
constituents. Nonpoint source loadings are usually
s imulated with the PERLND and IMPLND sections while point
source contributions are specified as a input t ime series
defined by the user. The chemical transformation, benthal
release, and gas invasion algorithms in HSPF are specific to
certain constituents; consequently only those chemicals
listed below can be introduced by these processes:
21
-------
Chemical Transformations Benthal Releases Gas Invasion
BOD BOD dissolved oxygen
inorganic N (ammonia, ammonia or carbon dioxide
nitrite, nitrate) nitrate
organic N orthophoshorus
orthophosphorus carbon dioxide
organic P
phytoplankt on
zooplankton
benthic algae
carbon dioxide
total inorganic carbon
organic carbon
daughter products from degradation
of generalized constituents
plant nitrogen
plant phosphorus
Specification of initial storages is required for all
constituents to be modeled. Depending on the nature of the
study, one or more additional sources uill be important to
the modeling effort. To a large extent the algorithms which
represent chemical transformations are an integral part of
the model and uill degrade some chemicals and produce others
in a manner which is designed to be consistent with the real
world based on current knowledge. Thus, of the six
potential sources of water and/or constituents, both initial
storages for water and chemicals, and chemical
transformations will be included in almost every study which
is not purely a hydrologic investigation.
The purpose of the remainder of this discussion is to
provide guidelines for assessing whether or not each of the
other four potential sources of constituents (i.e., nonpoint
loadings, point loadings, benthal releases, gas invasion) is
significant to the overall water and/or mass balances for
the study area, and hence must be represented in the
modeling effort.
In making this assessment, one should consider the
foil owing:
1. Nonpoint loadings are commonly associated with
almost any type of human activity within a
watershed. It is unlikely that nonpoint source
pollution can be ignored in most comprehensive
water quality studies of watersheds.
2. It may be possible to model the land surface of
predominantly rural land using only the PERLND
22
-------
module; generally, both the PERLND and IMPLND
modules are required to adequately model urban
areas. Before deciding whether to utilize one
or both modules the model user should review
the differences between the tuo modules in
representing hydrolcgic and water quality
processes which are important to the study
area. Whether or not simulation of both
pervious and impervious surfaces is necessary
is influenced by the constituents which are
being simulated, their relative accumulation on
the two types of surface, and the relative
abundance of each surface type in the
watershed.
3. If instream processes are not simulated,
initial storages, chemical transformations on
the surface and in the soil, and washoff from
the land surface are the only chemical sources
which can be modeled.
4. Simulation of point sources is required under
the following circumstances:
if a significant fraction of the water
volume for the study area is
contributed by point sources, at least
on a seasonal basis. (In some urban
watersheds, all summer streamflow is
from point sources.)
if the chemical loadings associated
with point sources are a significant
source of the constituents being
modeled.
In most areas of the United States, point
loadings from industry and municipalities have
been inventoried in terms of mean flow and type
of effluent, and often some chemical
concentration data is available. A simple,
first-cut technique of assessing the
significance of point sources is to sum the
mean flows of all loadings and compare this
number to mean streamflow and low flow during
the simulation period at various points with
good records in the study area. Comparison of
these values will give a reasonable indication
of the dilution capacity of the stream.
At the same time it is often useful to develop
an estimate of mass contributions of selected
23
-------
constituents from the point sources. This can
be done by developing mean chemical
concentration estimates for each source, then
multiplying mean concentrations by mean flow to
derive mass contributions for each point load,
and finally summing mass contributions from all
point sources. If instream chemical
concentrations are available near the
streamflou gage, a rough estimate can also be
made of total mass loadings to the stream from
all sources. By comparing these estimates, the
modeler can make an intelligent decision on
whether point sources should be modeled.
5. Simulation of benthal releases is limited to
inorganic nitrogen, orthophosphorus, carbon
dioxide, and biochemical oxygen demand (BOD).
Generally speaking, benthal releases are only
significant in slow-moving bodies of water
which are subjected to heavy loadings of
nutrients and/or organic material. Settling of
dead organic material and subsequent
decomposition is paralleled by the release of
inorganic materials and soluble BOD. Under
some conditions, particularly periods of scour
from high flows, benthal releases can be an
important source of these constituents.
6. Simulation of atmospheric gas invasion is only
necessary if instream processes are simulated
and either dissolved oxygen or carbon dioxide
are to be modeled. If so, it is useful to use
sections PWTGAS and IWTGAS to estimate the
resulting concentrations of gases in the runoff
entering the channel system from pervious and
impervious areas, respectively. In addition,
gas invasion at the surface of the channel
waters must be simulated using the RQUAL
Sect i on.
Characterization of Sources. Once the modeler has decided
which sources of water will be modeled, the following
suggestions should prove useful in characterizing these
sources:
1. Generally, assigning values to initial storages
is not a major problem. However, one must be
careful not to assign initial values which
exert an unreasonable effect on simulation
results. For example, if an unrealistically
large initial value is specified for the land
surface storage of a particular chemical, it is
2k
-------
possible that simulated washoff for a
significant portion of the simulation period
will be biased. The modeler should always
examine the simulation results in the first
time intervals of initial computer runs to
assess whether problems of this nature are
occurring.
Parameter requirements for characterizing
nonpoint source chemical loadings may be found
in the User's Control Input (Part F, Sections
4.4(1-3)). While considerable data are
available which allow general characterization
of chemical accumulation and removal for
different types of land and different land
usesi the modeler will most often be forced to
make an educated guess at characterizing
nonpoint sources in the study area.
Examination of preliminary simulation results
may convince the modeler to adjust certain
aspects of the characterization. Given the
uncertainties involved in characterizing
nonpoint sources in most watersheds, the
accumulation/removal parameters are often
treated as calibration parameters.
In most caseSf characterization of point
sources is relatively straightforward. For
each point source, a time series of values is
required for flow and for all constituents
which are being simulated. The time series of
data must span the entire period of simulation.
Quite often a constant value for flow and
constant values for chemical concentrations are
used in the absence of better data; daily,
monthly, or seasonal values r. re preferred if
data is available. General guidelines are
available for characterizing municipal and many
industrial effluents (Metcalf and Eddy, 1972;
Dyer 1971). Be aware that if concentration
values for a particular constituent are omitted
for a point source, HSPF will assume a zero
concentration for the volume of water
introduced into the reach by the point source.
The user has a good deal of control over
whether or not particular chemical
transformations or benthal releases are
simulated. If they are, rate coefficients
allow further control on the impact of these
processes on simulation results.
25
-------
Data In p u t Procedures for Characterization o£ Sources.
Because of the large number of constituents and processes
which can be modeled by HSPF, it is not practical to give
detailed instructions on hou to provide the model with the
necessary input to properly characterize each possible
source of each possible constituent. Nonetheless, the
following general statements may be helpful:
1. Initial storages must be specified in the
User's Control Input for each constituent
modeled by PERLND, IMPLND, or RCHRES. The
input tables used to specify initial storages
are usually located after the parameter tables
specified for each module section (see Part F
of User's Manual) and usually have a table name
containing a phrase such as "STOR", "INIT", or
"STATE".
2. The numerous parameters which control the
quantity of nonpoint source loadings simulated
by HSPF are contained in the UCI tables for
modules PERLND and IMPLND.
3. Point loadings data are input to HSPF by using
the External Sources and Network Blocks.
Guidance is provided in Section 4.6 of Part F
of the User's Manual.
4. As already indicated, chemical transformations
are a source of certain constituents in all
three application modules. Numerous tables in
the UCI are used to characterize the
transformations which are modeled.
5. Three tables in the RCHRCS UCI are used to
characterize benthal releases. Table-types OX-
BENPARM, NUT-BENPARM, and PH-PARM2 are used to
provide the necessary input for simulating
bottom releases of BOD, nutrients, and carbon-
dioxide respectively.
6. In HSPF, gas (dissolved oxygen and carbon-
dioxide) concentrations in runoff from both
pervious and impervious surfaces are assumed to
be at saturation; hence user input is not
required. However, for instream gas invasion a
limited amount of information must be supplied
by the user in Table-types OX-CFOREA (for
oxygen) and PH-PARM2 (for carbon-dioxide).
This discussion on characterizing constituent sources is
intended to provide the user with a preliminary
26
-------
understanding for the procedures and effort which will be
necessary to provide the model with the information it needs
to simulate the constituents and sources uhich have been
selected. Additional details for performing the
characterization are provided in the discussion of parameter
development contained in Section 6.
3.2 Preliminary Segmentation of Land Area Based on
Weather Data
This discussion focuses on the development of an appropriate
representation of the meteorologic conditions for an entire
study area based on site-specific weather data from stations
in and near the study area. Topics discussed include
weather data needs for hydrologic and water quality
simulation, importance of different weather data types to
simulation results, interpretation and evaluation of
available data, and criteria for selection of the best
station records and representation scheme for the study
area .
Time series weather data are critical inputs to HSPF for
both hydrologic and water quality simulation. All
hydrologic simulations of runoff require precipitation and
potential evapotranspirat ion data. Hydrologic studies which
simulate snowmelt and water quality studies which simulate
water temperature require additional time series data for
air temperature, wind speed, solar radiation, and dewpoint
temperature. Plankton simulation requires solar radiation
data. Depending on the simulation options selected, time
series data for wind speed and cloud cover may be needed for
simulation of a generalized quality constituent. Wind speed
may be required for simulation of dissolved oxygen.
Table 3.2 summarizes the meteorological data required for
simulating various processes in HSPF. Further details on
time series requirements can be found in Section 4.7 (Time
Series Catalog) of the User's Manual.
A necessary task in the HSPF modeling effort is division of
the study area into land segments such that each segment can
be assumed to produce a homogeneous hydrologic and water
quality response. To determine whether meteorologic
variations should be accounted for in selecting segments,
two factors must be considered. First, the degree of
spatial variability exhibited by the data type must be
examined. For instance, data suggest that in the Iowa River
Basin mean annual air temperature has a much more
significant variability across the watershed than does wind
speed.
27
-------
to
LLJ
a:
I
o
Oi
f
*
0.
in
3:
e
o
in
z
o
O
LU
in
LU
73
01
3
CT
a
at
t_
01
a
E
Ol
o
O
01
a
at
LU
a.
.O
ID
o
o
_t
o
a:
o
LU
* * o
u
3
*
ID
i a c t-
c c_ oo
c ro 3 -*- a
o L. + c -u e
f 0) 01 IS 01 L.
-u o i- E -f 01
IB a c ai 01 73 >
-~ 13 o a > nj -- o
> S O I- C O
Q. LU * Of E "~
IB -H 1. O 73
O L. 73 IB a 3
01-4- L. C~ 3 O
i. o a - o ai ->
E
O 3
1- E
>t- 01
a; a. in ^
uj z: c is
>- LU o «
<£!
3 in -f ID
a. a. ID in
N m
« n c c in
a ai o o >
. .-..-_ m
£- I- *- ^- -M o LU
o 01 o u ID + an
n in or 01 o X
in in o -C o
O 01 > a. ££
E E <- C
^ O O H- «^ H-
T3 -. -O T3 TJ T3 TJ
01 U 01 a 01 01 01
I- C C I. u I. I.
O
333333
tr -M tr cr cr tr cr
ai a. ai oi 01 ai ai
<- o i- L. c. L. L.
a. in
1/1
28
-------
Second, the impact of the data type on simulation results
must be considered. Some data types such as precipitation
and evapotranspiration are direct determinants of water
availability while other data types only affect streamflow
timing by altering the rate of spring snowmelt.
Consequently, if significant variability does exist over the
watershed for a critical data type such as precipitation or
evapotranspirat ion, the use of multiple weather station
records is warranted. Simulation results can be further
improved in those cases where multiple records for the other
meteorological data types are readily available.
It should be noted, however, that there is a limit to the
amount of segmentation which should be performed based
solely on meteorologic considerations; additional
segmentation of the study area, as described in Section 3.3,
will be necessary to represent differences in soil
characteristics and land use. Thus, if three segments are
defined based on meteorologic variability and three land
uses are to be simulated, the total number of land segments
which must be simulated is nine multiplicative. Major
differences in soils characteristics could require an even
greater division of segments and the computer costs for
simulating additional segments are significant (Section
2.3) .
Experience has shown that effective meteorologic
representation of most watersheds greater than approximately
100 square kilometers requires at least three different
rainfall records, perhaps more if rainfall patterns are
highly variable. For watersheds smaller than 100 square
kilometers one rainfall record may be adequate if rainfall
is reasonably uniform and study goals do not require maximum
accuracy. Generally speaking, an effective procedure is to
segment the study area based on three or four sets of data
which include records of somewhat low, average, and somewhat
high rainfall and evapotranspirat ion. Specific conditions
and/or project objectives may require more detailed
representation.
If a range of values for critical weather data is
represented in the records from the different stations, the
model user can maintain a degree of flexibility in
simulation results by adjusting the amount of study area
land which is represented by each of the sets of
meteorologic data. This procedure was used in the Four Mile
Creek Study, and is described in its final report (Donigian
et al., 1983b) .
A number of factors are involved in selecting the most
appropriate weather data for a study area. Among these are:
-------
long term behavior of study area weather
differences between long term area beihavior and
long term record behavior for specific stations
spatial variability in study aresa weather
exhibited in both short and long term records
accuracy and completeness of station records
How these factors affect the selection of weaither data for a
modeling effort is best shown by example. Consequently, the
detailed description of the weather data selection process
for the Iowa River Basin Study has been extracted and
included below. Each data type is considered separately
since the selection procedure varied depending on
availability of data, spatial variability of the data type,
and the impact of the data type on simulation results.
Genera 1 Availability of Data. There are 18 NOAA weather
stations in or near the 7,2*40 square kilometer Iowa River
Basin above Marengo. The location of each station in
relation to the watershed boundary is shown in Figure 3.1.
Additional meteorologic data were available: from the Iowa
State University and Four Mile Creek Weather Station near
Traer. Precipitation, maximum and minimum air temperatures,
humidity, pan evaporation, solar and net radiation have been
recorded at this station. However, the station was closed
during winter months and has experienced numerous equipment
failures; consequently, records are incomplete.
Preci pi tat ion. Mean annual precipitaton for the basin
varies from 762 millimeters in the north to 838 millimeters
in the southeast (Figure 3.2). Given the primary importance
of precipitation data to the simulated water balance, three
records were used. Both long term averages and records for
the selected simulation period (1974-1978) suggest that the
Traer precipitation is representative of the southeastern
third of the basin, which receives 813 to 838 mm of yearly
rainfall .
The central section of the basin has a long term average
annual precipitation in the range of 787 to 813 mm, and can
be well represented by the Iowa Falls record. The Iowa
Falls station recorded an average of 757 mm of annual
rainfall during the 1974-1978 simulation period, somewhat
lower than the long term average. (Lower than average
rainfall was recorded at all stations within the basin for
the 1974-1978 period.) The Iowa Falls record was generally
good. Records were missing for 17 days, and were filled in
using data from the Ames station.
30
-------
The northern section of the basin is characterized by 762 to
787 mm of average rainfall. Inspection of the records for
the two best candidate stations, Forest City and Sheffield,
showed large periods of missing data for both. Sheffield
was selected as the base record and was updated using Forest
City data when available (43 days). Remaining gaps (53
days) were filled using Iowa Falls data.
FOREST CITY
WATERLOO WSO AP
AMES 8 WSW
Meteorologic Station
U.S. Goelogical Survey Recording Gag*
O US. Geological Survey Discontinued Gig*
A Crest-stage Partial-record Station
- -_- Basin Boundary
X 250 Miles upstream from mouth of lowi River
NORTH ENGLISH'
Figure 3.1 Meteorol&gic and U.S.G.S. Gaging Stations in
and near the Iowa River Basin.
31
-------
34 34
Mean Annual Precipitation in Inches
620mm
danowi d«ta nation
uxd lor HSPF
ImuKtlon lion
640 mm
720
720mm
720mm 740mm
Mean Annual Potential Evapotranspiration in Millimeters
Figure 3.2 Isopleths of Mean Annual Precipitation and Potential
Evapotranspiration in Iowa (adapted from Iowa Natural Resources
Council, 1978). Locations of data stations used in simulation
are noted on maps.
32
-------
Potent ia 1 Evapotrnnspi rat i on (£E.T) . Mean annual PET for the
Iowa River Basin varies from 635 mm in the north to 686 mm
in the far south (Figure 3.2). Three sets of PET data are
available: Ames, Iowa City, and Four Mile Creek Weather
Station. The record used for simulation was a composite of
Four Mile Creek Weather Station and Ames data. All data
prior to July 1976 is from Ames, while that occurring after
July 1976 is primarily Four Mile Creek Weather Station data,
with missing values obtained from Ames. The ten year
(1969-1978) average annual PET for the combined record is
630 mm; this suggests that the record may be a little low
for the southern portion of the basin. However, since the
record was used successfully for the Four Mile Creek
simulation, it was considered adequate to represent the PET
for the overall basin.
Air Temperature. Long term records indicate a strong
relationship between station latitude and mean annual air
temperature (Figure 3.3). Short term records show more
variability, but indicate that the 1974-1978 period was
cooler than typical. Given the fact that the stations which
are selected are used to represent temperature
characteristics over large areas of land, stations which
exhibit reasonably close agreement between long and short
term records are more likely to be representative of the
large regions. Selection criteria for air temperature
records are listed below in order of importance:
1. Three stations were needed, one to represent
each of the three basin sections delineated for
the precipitation records.
2. Close agreement between long and short term
records was desirable.
3. The short term record should be somewhat cooler
than long term record.
4. Stations should be within the watershed
boundari es.
Based on these criteria the three air temperature records
chosen for the Iowa River Basin simulation were Iowa Falls,
Marshal 1 town, and Cedar Rapids.
Iowa Falls - The station is located inside the watershed,
and its short term and long term records are similar. The
mean annual temperature is about 8.6 degrees C, and the
record was used to represent the upper third of the basin.
Marshalltown - The station is located inside the watershed,
and its short and long term records are similar, with the
short term record somewhat cooler. The mean anual
temperature is approximately 9.2 degrees C, and the record
was used to represent the middle portion of the basin.
33
-------
(0 O
s "-1
o
2
5
to
0) -H
ft'O
(0 0)
TJ to
CO
rt C
5 O
0-H
H -P
(0
C -P
-H 10
,-. (0
fe
p to
(0 C
M o
(U -H
CX-P
E «
o o
EH O
3 ^
C oo
C
rt
0)
o
o
X! U
-P
(U W
-H 0)
a o
o ^
01 3
HO
co O
M 0)
rH -P
CU ffl O
r) M C
0)
-------
Cedar Rapids - The mean annual temperature of this station
is approximately 9.7 degrees C, which makes the station
representative of the lower porton of the basin. The short
term mean annual temperature is similar but somewhat cooler
than the long term record. The station is located outside
of the watershed, but appears to better represent the lower
third of the basin than any other station.
The quality of all three records was excellent, with a total
of seven records missing for the entire simulation period.
These records were filled in using data from nearby
stations. All records consisted of maximum and minimum air
temperatures. These data were distributed to hourly values
for use by HSPF.
Mind Speed. Wind data for the state of Iowa do not vary
greatly from station to station. Consequently, the wind
data from Four Mile Creek Weather Station (corrected using
Waterloo data), which were used for the Four Mile Creek
simulation, were examined to determine whether the record
would adequately represent the entire Iowa River Basin.
Analysis showed that the mean average hourly wind speed over
any given month of the ten year Four Mile Creek record did
not vary from the long term composite Iowa value for the
same month by more than 1.6 km/hr. Comparison of mean
annual wind speeds showed a composite statewide value of
12.2 km/hr at 0.3 meters above the land surface versus a
value of 12.1 km/hr for the Four Mile Creek data. The Four
Mile Creek record had considerable gaps in it and was
updated for the Four Mile Creek simulation using Waterloo
data. This composite record was used for the entire Iowa
River Basin.
Solar Radi at i on. Comparison was made between the Four Mile
Creek Weather Station solar radiation data and that at Ames
(approximately 80 kilometers away) to assess the variability
of radiation within the basin area. For the 18-month period
from July 1976 to December 1977 the records differed by 3X,
with a maximum monthly variation of 20X for the month of
March 1977. Given the limited variability in these two
records, the radiation record which was used for the Four
Mile Creek simulation was used to represent the entire Iowa
River Basin. This record is a composite of Four Mile Creek
Weather Station and Ames data. All data prior to July 1976
is from Ames, while that occurring after July 1976 is
primarily Four Mile Creek Weather Station data, with missing
values obtained from Ames.
Dewpoint Temperature. Previous studies have shown
similarity between average daily dewpoint temperature and
minimum daily temperature. Comparison of these two values
on a daily basis for a 60-day record at Waterloo verified
35
-------
this relationship.
record (Mason City)
uas decided that
temperature could
Given the fact that only one deupoint
is available near the study basin* it
the best representation of deupoint
be obtained by using daily minimum
temperature records for the three
Falls, Marshal 1 town, Cedar Rapids).
basin segments (Iowa
Based on the above analysis of available weather data from
stations in or near the Iowa River Basin, it uas deemed
necessary to divide the study area into three meteorologic
segments in order to adequately represent observed
variability in precipitaion and air temperature. While the
boundaries between the three segments uere still reasonably
uncertain at this point in the segmentation process, it uas
useful to summarize the planned use of meteorologic data in
Table 3.3 for use in developing input sequences once the
segment boundaries uere finalized.
TABLE 3.3 SUMMARY OF METEOROLOGIC DATA USED TO REPRESENT THE
THREE SEGMENT GROUPS OF THE IOUA RIVER BASIN
source of meteorologic record used to
represent' each segment *
data type
precipitation
potent ial
evapotranspiration
air temperature
wind speed
solar radiation
segment i1 segment t2
Sheffield/ loua Falls
Forest City
FMC**/
Ames
loua Falls
FMC/
Waterloo
FMC/
Ames
FMC/
Ames
Marshal 1 town
FMC/
Waterloo
FMC/
Ames
deupoint temperature loua Falls Marshalltoun
(min. daily
temp.)
(min. daily
temp.)
segment
Traer
FMC/
Ames
Cedar
Rapids
FMC/
Waterloo
FMC/
Ames
Cedar
Rapids
(min. daily
temp)
The second station noted in some entries uas used to fill
in missing data records in the primary station.
**FMC = Four Mile Creek Weather Station
36
-------
It is useful to emphasize several aspects of the data
selection process used in the Iowa River Study. The
following suggestions are general in nature and can be
applied to the data evaluation and selection process for any
HSPF modeling study.
(1) Locate all meteorologic stations in and near
the study area on one map.
(2) Locate long term weather behavior data for the
study area in the form of isopleth maps such as
Figures 3.2 and 3.3. Use these maps to assess
the need for meteorologic segmentation.
(3) For each type of weather data tabulate the
length of record and mean annual value (long
term record) for each station based on NOAA
data summaries.
(1) Locate stations and mean station values on
isopleth maps. Use this information to
determine which stations are most
representative of particular portions of the
study area.
(5) Based on available weather data and an
assessment of the availability of streamflow
and water quality data for calibration and
verification of the model* select the period of
time which will be simulated.
(6) For each type of weather data and for each
station tabulate the mean value for each year
of the simulaton period and assess the quality
of each record in terms of the number of
missing values.
(7) Evaluate these mean annual values to identify
short term weather trends for the simulation
period and possible anomalies in the short term
records which could preclude their use as
representative data for large areas. For
example, 1974 precipitation records at
Sheffield, Iowa, included two very intense
rainfall periods which appeared to be localized
thunderstorms. Use of the Sheffield record to
represent the upper third of the Iowa River
Basin resulted in gross oversimulation of
runoff for 1971.
(8) If snoumelt is to be simulated, compare the
timing of spring warming trends in air
37
-------
(9)
( 10)
temperature data for the various stations to
observed increases in streamflou at gaging
stations. Both the timing and amount of
snoumelt is dependent on air temperature. and
hence, a good simulation of streamflou during
the spring months depends on the use of
appropriate air temperature data.
Select the
each data
segment.
best weather
type for each
station
planned
to represent
meteorologic
Fill in missing
stat i ons.
records using data from nearby
The above discussion assumes that there are a number of
weather stations in or near the study area. Depending on
the size and location of the study area, the model user may
have difficulty obtaining even one set of representative
data for a particular ueather data type. In particular,
data for solar radiation, wind speed* and deupoint
temperature are scarce. As can be seen by the discussion of
the selection process for these data types in the Iowa River
of judgement and approximation is
the best input for the modeling
of minimum air temperature records
records is an example of such an
approximation. It is important that careful consideration
be given to selection of meteorologic data in order to avoid
the necessity of making changes in the data base at a later
point when it is discovered that selected data are not
appropriate.
Study, a certain amount
necessary in developing
effort. The substitution
for dewpoint temperature
3.3 Final Segmentation of the Land Area
The final segmentation scheme for a watershed cannot be
performed until soils characteristics and land uses have
been considered. Guidelines were presented above for
performing preliminary segmentation of a study area based on
meteorologic considerations. This section discusses these
additional factors which must be considered in order to
develop the final segmentation scheme. First, general
definitions for segments and segment groups are provided to
clarify the purpose and process of segmenting the study
area. Following these definitions, the method used to
refine segments in the Iowa River Basin Study is described.
This example, along with supporting discussions* illustrates
how soils characteristics, topography, and boundaries of
contributing areas to river reaches are used to delineate
segment groups and how land-use data is used to determine
the areal breakdown of segment groups into segments.
38
-------
One of the basic concepts of watershed modeling using a
lumped parameter approach (e.g., HSPF and predecessor
models) is the division of the watershed into land segments,
each with relatively uniform meteorolo'gic, soils, and land-
use characteristics. Similarly the channel system is
segmented into 'reaches', with each reach demonstrating
uniform hydraulic properties. The entire watershed is then
represented by specifying the reach network, i.e., the
connectivity of the individual reaches, and the area of each
land segment that drains into each reach. Each land segment
is then modeled to generate runoff and pollutant loads per
unit area to the stream channel. Multiplying the unit area
runoff and pollutant loads by the area of each land segment
tributary to each channel reach determines the runoff and
pollutant loads to each reach; performing these calculations
for each reach in conjunction with modeling the instream
hydraulic and water quality processes results in the
simulation of the entire watershed.
Def inition oi segment and segment g roup. For the purposes
of HSPF, a segment is defined as a parcel of land which
exhibits a homogeneous hydrologic and water quality
response. Hence, one set of hydrologic and water quality
parameters (both calibration and non-calibration parameters)
can be used to characterize all of the land considered as
one segment. For modeling purposes, it is not necessary
that all of the land in a segment be contiguous. The only
requirements are that the segment parameters reasonably
represent the hydrologic and water quality charateristics of
all land considered as part of the segment, and that the
total area of each segment contributing runoff and
pollutants to each hydraulic reach is known.
The hydrologic response of a parcel of land is a function of
meteorologic patterns, soils characteristics, and land uses.
In most cases, meteorologic patterns and soils
characteristics allow for a preliminary division of a basin
into segment groups. A segment group is a parcel of land
which is exposed to meteorologic conditions (rainfall,
evaporation, etc.) which for modeling purposes are
designated by one set of meteorologic time series. In
addition, it is assumed that all of the land in the segment
group would exhibit a homogenenous hydrologic reponse if
there were uniform land use. In order to make this
assumption, soils characteristics must be reasonably
consistent throughout the segment group area. Segment
groups are subsequently divided into segments, with each
segment representing a different land use.
The segmentation process is best shown by example.
Consequently, a detailed description of the segmentation of
the Iowa River Basin has been extracted and included below.
39
-------
Prel imi.n.a rv s egment groups i_o_r- the I oua R i v e r Basin.
Variability in meteorology over the basin indicated that the
loua River Basin should be divided into three segment groups
in order to perform a reasonable hydrologic calibration
(Section 3.2). Based on long-term isopleth information on
rainfall and air temperature, tentative boundaries for the
segment groups were formulated, followed by a slight
adjustment of boundaries based on spatial distribution of
soils.
Most of the Iowa River Basin is covered with prairie soil
formed from glacial drift, an unconsolidated mixture of
gravel and partly weathered rock fragments left by glaciers.
Underlying the drift, at a considerable depth, are
consolidated rocks that outcrop where the river has cut deep
into the drift. The study area has three distinct
topographical areas. The first area is the upper end of the
basin above Alden (Figure 3.4), where topogiraphy is gently
undulating to nearly level. In this area drainage is poorly
developed, and the land is characterized by depressions
which collect water and prevent rapid runoff. Soil
associations are predominantly Storden, Clarion, and
Webster. The second area between Alden and flarshalltoun is
more hilly terrain, but is still predominantly Clarion and
Webster soils. South of Marshalltown the terrain becomes
more level, and the glacial drift soils are covered by
loess, a silty, wind deposited material. The topography and
loess thickness vary in the region, but generally 1.5 to 4.5
meters of gently sloping loess materials are present. This
southern area is in the Tama-Muscatine soil association.
The boundary between the Clarion-Webster and the Tama-
Muscatine areas was compared to the tentative boundary
between the bottom two segment groups, as defined by
meteorologic considerations. It was concluded that the
soils association boundary would serve equally well as a
boundary between the land represented by meteorologic data
sets f2 and t3 (Section 3.3). The preliminary boundary
between the two northern segment groups was drawn based on
long-term precipitation isohyets and the general breakpoint
between the northern flat lands and the central hilly
region. These preliminary segment group boundaries are
delineated in Figure 3.4.
Comparison of preliminarv segment group boundaries to
boundari es for cont ribut ing areas to hvd raulic reaches. A
good deal of time and effort can be saved by defining
segment group boundaries so that they are superimposed on
the boundaries between contributing areas to the individual
reaches. To determine whether or not boundaries can be
superimposed, the model user must first delineate the
contributing area boundaries for reaches as outlined in
40
-------
Section 3.4. For the loua River Basin Study> contributing
area boundaries as delineated in Figure 3.5 were examined
and it was decided that the preliminary segment group
boundaries (Figure 3.4) could be shifted and superimposed
onto contributing area boundaries as shown on Figure 3.5.
Thus, all land contributing runoff to reaches 1-6 uas
contained in segment group i1; all land contributing runoff
to reaches 7-11 was in segment group #2, and runoff to
reaches 12 and 13 was wholly contributed by segment group
#3.
V U.S. Geological Survey Recording Gagt
O U.S. Geological Survey Discontinued Gagt
A Crest-stage Partial-record Station
..v Basin Boundary
\25O Miles upstream from mouth of low* River
^\ Segment Group No.
Figure 3.4
Preliminary Segmentation of the Iowa River
Basin to Account for Variability in Meteorologic
Patterns and Soils Characteristics.
-------
The three segment groups delineated in Figure 3.6 are the
final ones used for the loua River Basin Study. The
boundaries between segment groups are based on
meteorological, edaphic (soils), topographical, and drainage
considerations. Evaluation of land-use practices allows the
model user to further divide each of the segment groups into
segments.
-4
^_..__ Basin Boundary
Reach Boundary
- Local Contributing /tret to Knell
Figure 3.5 Channel Reaches and Contributing Areas for
the Iowa River Basin.
-------
Land use categories. The final subdivision of segment
groups into pervious land segments (PLSs) and/or impervious
land segments (ILSs) is based on land use. Land use types
which uill have the largest impact on runoff or water
quality response in the watershed must be identified. The
user must assess whether or not runoff from impervious urban
areas is a significant contributor of water and/or
pollutants. If so, the amount of impervious area in each
must be determined* and pollutant accumulation
processes on impervious surfaces must be
(Section 3.1).
segment group
and removal
characterized
o
A
\Z50~
'I.S Geological Survey Recording Gige
U S.Geological Survey Discontinued Gigc
Crest-stage Partial-record Station
Basin Boundary
Miles upstream from mouth of lowi River
Segment Group No.
Figure 3.6 Final Segmentation of the Iowa River Basin.
-------
If urban runoff does not contribute significant uater or
pollutants to the study area* it is appropriate to represent
the entire watershed with pervious land segments. For
example, in the loua River Basin between 65X and 85X of each
county which contributes land to the basin is cropland,
while less than 1% is urban. Of all other land use types*
only grassland comprises more than 10X of the area's total
land. As a result, agricultural nonpoint source pollution
in the form of fertilizers and pesticides is the major water
quality concern in the basin. While use of impervious land
segments is not necessary to model this1. study area,
differences in land use and agricultural practices require
the division of each of the three segment groups of the
basin into multiple pervious land segments.
A large majority of the croplands in the basin are planted
in either corn or soybeans. Given the differences in
fertilizer and pesticide application for the two crops, each
crop was considered as a separate land-use type. All lands
not planted in corn or soybeans were considered as a third
composite land-use type. Thus, there were a total of nine
pervious land segments (PLSs) for the Iowa River Basin - one
to represent each of the three land-use types in each of the
three segment groups. The characteristics of the nine
pervious la-nd segments selected for the Iowa River Basin
simulation are summarized in Table 3.4.
Division of s ectment g roup areas into PLS areas . A number of
factors are involved in deciding how many and which land
uses will be modeled as distinct segments. Important
considerations include:
allowable complexity of modeling effort within
time and effort constraints of the study
spatial resolution required to answer study
quest i ons
number of segment groups required to represent
differences in meteorologic, topographic, and
soils conditions
degree of heterogeneity in land use within
segment groups
availability of reliable data which will serve
as the basis for dividing segment groups into
desired segments
When the model user has decided upon an appropriate number
of land use segments based on the above considerations, a
good deal of work is still required to transform and reduce
-------
TABLE 3.1 DEFINITION OF PERVIOUS LAND SEGMENTS FOR THE IOWA
RIVER BASIN
CHARACTERISTICS
PLSt
1
2
3
«»
5
6
7
8
9
meteorology
met. set * 1
met. set t 1
met. set t 1
met. set t 2
met. set * 2
met. set t 2
met. set t 3
met. set t 3
met. set * 3
soils
loess
loess
loess
glacial till
glacial till
glacial till
glacial till
glacial till
glacial till
land use
soybeans
corn
other
soybeans
corn
other
soybeans
corn
other
see Table 3.3 for description of meteorologic data
existing land-use data into the form needed as input to
HSPF. Depending on the size of the study area, local,
county, and/or state statistics and planning maps may be
necessary to properly characterize land use. For large
watershed areas, land-use data is often tabulated on a
county-by-county basis, and this data must be extrapolated
to contributing areas to each reach based on the amount of
various counties contained within each contributing area.
At the same time land-use data in existing documents quite
often is divided into different categories than those
desired for the modeling study. Consequently, some
aggregation or disaggregation of data is almost always
requi red.
For the Iowa River Basin Study, county
year (1976) of the simulation period (
to determine the percentage of land i
to corn, soybeans, and other purposes.
the relative amount of land devoted to
throughout the county. The contri
hydraulic reach was subdivided on
further subdivided into corn, soybeans
on the countywide statistics. Total a
the land uses is summarized in Table
land use data for one
1974-1978) was reduced
n each county devoted
It was assumed that
each use was constant
buting area to each
a county basis, and
, and other land based
rea devoted to each of
3.5 for each of the 13
-------
TABLE 3.5 LAND USE IN THE 13 CONTRIBUTING AREA SUBDIVISIONS IN
THE IOUA RIVER BASIN
Contri but ing
Reach area (sg km)
1 860
2 847
3 355
4 269
S 238
6 723
7 927
8 995
9 122
10 202
1 1 199
12 391
13 1 109
TOTAL 7236
Area planted
in corn
(sa km)
321
332
140
106
98
337
445
495
62
101
98
184
523
3243
Area planted
in soybeans
(so km,)
1 1 1
140
60
47
41
1 17
197
259
28
49
54
129
360
1593
Other land
use (sg km)
427
376
155
1 17
98
269
285
241
31
52
47
78
225
2400
subdivisions of the basin. (Note that all land in the
drainage area for a reach must be classified as belonging to
one of the land use categories.) The information in this
table, combined with parameter values which establish
hydrologic and water quality characteristics for each land-
use type, is needed by HSPF in order to simulate runoff and
chemical washoff from contributing areas (if a reach system
is being modeled) or from segment groups (for studies not
including reach systems).
Transferring Land Segmentat i on Data into a. HSPF Input
Sequence. The following explanations describe how -the
meteorologic, soils, and land-use data used to define land
segments are incorporated into the HSPF input sequence
(User's Control Input):
Meteorologic
Store (TSS)
Section 4 of
data are input to the Time Series
using the procedures outlined in
this guide and detailed in Part F
2.
of the User's Manual.
After the meteorologic data have been input and
cataloged in the TSS, the modeler specifies
which weather data will be used for each land
segment by developing the EXTERNAL SOURCES
BLOCK of the User's Control Input (see Appendix
A for example).
46
-------
3. Soil properties such as particle size and
distribution, bulk density, depth of topsoil,
and others are critical determinants of the
hydrologic and sediment erosion processes on
pervious land segments. Consequently, the
values selected for many of the parameters in
the User's Control Input for module sections
PWATER, MSTLAY, and SEDMNT are determined by
the predominant soils characteristics for each
s egment.
4. Land use activities affect hydrologic,
sediment, and chemical processes on all land
segments, regardless of whether they are
pervious or impervious. Representation of
land use activities is accomplished through the
use of the PWATER, SEDMNT, MSTLAY, PQUAL, PEST
and NUTR module sections of PERLND and the
IWATER, SOLIDS, and IQUAL module sections of
IMPLND. In addition, the 'Special Actions'
option (see User's Manual Section 3.5 and
Sections E 4.03 and F 4.10) is used to
represent chemical applications, tillage
operations, and other abrupt changes to land
surface conditions.
Selecting appropriate parameter values to represent various
soil types and land-use activities is a major aspect of
simulation. Additional discussion on specific PERLND and
IMPLND parameters and their relationships to land surface
and subsurface conditions is provided in Section 6 (Model
Parameters and Parameter Evaluation).
3.4 Segmentation and Characterization of the Channel and
Contributing Areas
The purpose of this section is to outline and discuss the
criteria used for selection and definition of channel
reaches and the areas contributing runoff to the reaches.
Performance of the tasks described in this section is only
necessary if the model user decides that modeling of
hydraulic routing and/or instream processes is essential to
meet the study goals. Situations which often require
modeling of channel processes include:
studies which require the calculation of
accurate instantaneous peak flows and/or
concentrations
studies in which point loadings must be
cons id ered
47
-------
studies in which water quantity and/or quality
results must be determined at locations other
than the downstream terminus of the study area
studies which simulate constituents which
experience significant degradation in the
stream channel during ordinary flow conditions
Basic channel hydrogeometry is a primary consideraton in the
channel segmentation process. Before the segmentation
process begins, the modeler should determine the following
channel characteristics from available maps and supporting
data :
length of channel in study area (from maps or
reports)
average slope of channel (from maps or reports)
velocity at mean flow (from USGS gage; records)
flow-through time for mean flow
The above data gives
actual channel behavio
through time for mean
degradation rate for
assess whether or not
to significantly reduc
the travel time in the
the modeler to more c
to the modeling effort
the model user a rough idea of the
r. For example, by comparing flow-
flow for the study channel to the
a particular contaminant, one can
channel processes should be expected
e quantities of the contaminant during
study area. Such information allows
learly define the processes important
before simulation begins.
General channel characteristics such as average slope are
useful indicators of required segmentation. By comparing
average slope to extremes in slope experienced in localized
portions of the channel, one can ascertain whether or not
hydraulic behavior is likely to vary significantly from one
length of channel to another; if so, additional channel
segmentation may be required to provide a hydraulic
representation which is adequate to satisfy study goals.
The proper use of hydrogeometric considerations in the
segmentation process was demonstrated by the Iowa River
Study in which the three major criteria for definition of
channel reaches were reach length, slope, and entry point of
tributary flow.
(1) Reach length. The hydraulic routing algorithms
used in HSPF are most accurate when flow time
through individual reaches approximates the
simulation time step. Since a 2-hour time step
-------
was used for routing in the loua River* reach
lengths should ideally have been approximately
3.6 km (1.8 km/hr x 2 hours) in order for flow-
through time for mean flow to meet this
condition. If this criterion uere folloued,
more than 80 reaches would have been necessary
for the Iowa River channel. For the purpose of
this demonstration project longer reaches, in
the range of 15 to 30 kilometers, were used.
The use of longer reaches reduced and spread
out short time interval peaks, but effects were
minimal on the mean daily values used for
calibrat ion.
(2) Slope. Individual reaches should have
reasonably homogeneous bottom slope. Major
drops in bottom elevation due to natural falls
or reservoirs should serve as boundaries
between reaches; the change in bottom elevation
at the channel discontinuity should not be
considered in the slope calculation.
A low water profile for the Iowa River was
prepared using U.S.G.S. data (Heinitz, 1973).
The profile (Figure 3.7) indicated a highly
uniform slope for the entire 300 km stretch of
river. A preliminary division of the river
into reaches indicated that slopes range from
0.00026 to 0.00069 m/m. Consequently, slope
was not a major consideration in reach
definition for the Iowa River. U.S.G.S. data
indicated only one significant discontinuity in
the channel bottom: a 24-foot drop below the
Iowa Falls Power Dam (Figure 3.7). The
reservoir site was used as a reach boundary in
definition of the Iowa River reach
configuration.
(3) Entry point of tributary flows. HSPF assumes
that all local flows enter a reach at the
upstream boundary. Consequently, it is
reasonable to define reaches so that downstream
limits are located directly above major
tributary inflows. Hence, inflows enter a
reach at its upstream limit in the same manner
as the routing algorithms assume.
The Iowa River was divided into 13 reaches for
simulation. Of the 12 intermediate reach
boundaries between the study limits, one was
selected at the Iowa Falls Power Dam channel
-------
HVMOa 8
39V9 WV3H1S S9SH
wva SYWJ VMOI
STIVJ »MOI
QNV XiJOJ HinOS N33M13B INIOdQIM
XUOd HinOS
-I- °
3DN3nldNOD
NHOinvHSHVH 6 39*3 MV3U1S SDSn
33N3mdNOO X33HD HVSdS
SDNSnlJNOO X33HO
SMOQ
smvj
_. S
NM0111VHR1IVH
30N3mjNOO X33U3 11VS
3DN3nldNOD JI33UD A3NOH -
SNIVId 3113S
Z'OSI
O'SET) 09N3XVW e 39V9
_. S
8 a a °
C7> OO f^
^ 13A31 V3S 3AOHV '133d NI '(IOUVA313
0)
rH
H
0
M
JJ
-------
discontinuity* 2 were selected at U.S.G.S.
streamflou gage sites (Rowan and Marshal 1 town),
eight corresponded to sites of major tributary
inflow, and one was chosen to subdivide a
section of river which was too long to be
represented as one reach.
While channel segmentation in the Iowa River Study was based
almost solely on hydrogeometric criteria, additional
considerations are important in many other studies. Two
factors which are critical to the development of an
appropriate reach configuration are (1) the location of data
available for model calibration/verification and (2) the
spatial resolution required to answer study questions.
[)ata Availability. As discussed in Section 3.2, the period
of simulation should be selected based not only on
availability of meteorologic data, but also on the
availability of instream quantity and quality data which can
be used for calibration/verification. A good instream
calibration depends on one or more reliable streamflou
records which extend over the entire period selected for
calibration/verification. If water quality is to be
simulated, instream data on chemical concentrations which
characterize both spatial and temporal variation is highly
desirable. In order to compare observed and simulated
values directly, it is useful to define model reaches so
that points where data have been collected correspond to
reach boundaries. When the model user has decided which
quantity/quality data will be used for
calibration/verification, the location of this data should
be considered in the channel segmentation scheme.
Segmenting the channel so that streamflow gages are located
at reach boundaries is a common practice.
Spat ial Resolut i on Reg ui rements. The spatial detail of
simulation results is determined by the number and length of
the reaches defined in the channel representation. If the
modeler wishes to isolate individual point loads for
detailed analysis, no more than one point load can be
contained in a single reach. If the localized effects of an
instream aerator are to be assessed, the reach containing
the aerator should be a short one; otherwise, calculated
increases in dissolved oxygen will be averaged over a longer
stretch of channel than desired. In general, reach
boundaries should be defined at each point where simulation
results need to be examined. For example, if the goal of a
study is to assess a numb'er of potential reservoir sites, a
reach boundary should be defined at each of the sites.
When the model user has developed an appropriate reach
segmentation scheme based on channnel hydrogeometry, data
51
-------
availability, and spatial detail requirements, a number of
supporting calculations and tasks must be performed as
outlined below:
(1) Delineate the study area boundary and the
stream channel on the best available
topographical map.
(2) Locate reach boundaries on the map.
(3) Delineate the watershed area contributing
runoff to each of the reaches.
(4) Using a planimeter or other methods, calculate
the area contained in each of the subdivisions
delineated in step #3.
(5) Determine the average slope of each reach based
on map contours or supporting data.
(6) Concurrent with the final land segmentation
effort described in Section 3.3, determine
whether it is reasonable to superimpose land
segment boundaries on contributing area
boundaries to simplify the modeling effort.
(7) Develop an FTABLE for each reach for use in the
HSPF input sequence. FTABLES specify values
for surface area, reach volume, and discharge
for a series of selected average depths of
water in the reach. In most cases this type of
information is not available for each reach and
some approximations must be performed.
Description of FTABLE development for the Iowa
River Basin Study is provided below as an
examp1e.
FTABLES for reaches 1, 7, and 13 were
developed using U.S.G.S. cross-sections at
gage sites and depth/discharge curves,
combined with specified reach lengths.
FTABLES for reaches 2 through 6 were
developed assuming that the slopes of the
cross-sections were the same as that at
Marengo, but that channel capacity
decreased upstream from reach to reach.
Both the width and depth coordinates of
points on the Marengo cross-section were
multiplied by a factor «1.0) consistent
with relative top width data developed by
Wallace (1971) for each of the reaches
along the Iowa River. The adjusted cross-
52
-------
sections were input to an auxiliary
computer program along with values for
channel slope. Manning's n and flow. The
program generated values for normal depth,
cross-sectional area, and top width for a
series of flow values at each reach cross-
section, providing all the hydraulic data
necessary to generate the FTABLES. The
same procedure uas used to develop FTABLES
for reaches 8 through 12 based on the
cross-section and slope at Rowan. Again,
channel capacity was increased downstream
from reach to reach by applying
progressively larger multipliers (based on
Wallace's data) to the coordinates of the
Rowan cross-section. It should be noted
that the Marshalltown cross-section was not
used to generate other FTABLES, because its
shape was not considered representative of
most stretches of the river (Wallace,
1971 ).
(8) Prepare a summary table including reach
designation numbers, lengths, average channel
slopes, and contributing areas. Table 3.6 from
the Iowa River Basin Study is provided as an
example of such a summary table.
Transferring Channel Characterization and Segmentation Data
into a. HSPF Input Sequence. The three major groups of input
data developed during the channel segmentation and
characterization effort are (1) the hydraulic data contained
in FTABLES, the contributing areas to reaches, and the
configuration of the reach network. The following guidelines
are provided in order to expedite the incorporation of this
data into a HSPF input sequence:
1. The contents and format of the FTABLES are
outlined in Part F (Section 4.5) of the User's
Manual, and typical FTABLES are included as
part of the sample input sequence in Appendix A
of this guide.
2. Contributing area data for each reach is
incorporated into the input sequence in the
MFACTR field of the NETWORK Block. Refer to
Part F (Section 1.6) of the User's Manual for
instructions or to Appendix A of this guide for
an example. (Note that the value of MFACTR is
dependent on the constituent units used in both
the PERLND/IMPLND and the RCHRES application
modules.
53
-------
TABLE 3.6
REACH CHARACTERISTICS FOR THE IOWA RIVER
Description of Reach
Belmond to Gage at
Rowan
Gage at Rouan to Unnamed
Creek Confluence
Unnamed Creek Confluence
to Iowa Falls
Iowa Falls to Midpoint
Midpoint to South Fork
Confluence
South Fork Confluence to
Honey Creek Confluence
Honey Creek Confluence to
Gage at Marshalltoun
Gage at Marshalltoun to
Sugar Creek Confluence
Sugar Creek Confluence to
Deer Creek Confluence
Deer Creek Confluence to
Richland Creek Confluence
Richland Creek Confluence
to Salt Creek Confluence
Salt Creek Confluence to
Honey Creek Confluence
Honey Creek Confluence to
Gage at Marengo
TOTAL
Reach
13
12
1 1
10
9
8
7
6
5
4
3
2
1
Reach
Length
(Km)
15
26
28
28
28
22
16
214
29
26
20
16
17
299
.0
.14
.8
.3
.3
.<4
.7
.3
.8
. 1
. 1
.3
.t4
.8
Channel
Slope
(m/m)
.00026
.00031
.00066
.00069
.00055
...».
.OOOH7
.00033
.00028
.00030
.00032
.00030
.00026
. 000^14
Cont rib
Area (s_
1109
391
199
202
122
995
927
723
238
269
355
8«47
860
7236
3. The configurat
specified in
Flow of water
reaches and c
reach are incl
Further detail
to specify t
segments to
Appendix B of
ion of the reach network is also
the NETWORK Block of the UCI.
and constituents from upstream
ontributing land areas for each
uded in this portion of the UCI.
s on the use of the NETWORK Block
ransfer of materials from land
the channel are included in
this guide.
-------
When the tasks outlined in the previous four sections
(3.1-3.4) have been accomplished, the modeling strategy for
most model applications is complete, and the modeler can
begin to develop the computer data base and input sequences
ior the preliminary simulation runs. However, when HSPF is
used to model certain discrete activities, such as pesticide
or fertilizer applications on farmland, additional effort
must be expended on the modeling strategy to develop an
appropriate model representation. The next section
describes how and when to use the Special Actions routine of
HSPF to model the effects of discrete events occurring on
the study area.
3.5 Characterization of Special Actions
The model user should be aware of the Special Actions
capabilities of HSPF during the development of the modeling
strategy. The Special Actions Block can be used to adjust
the value for any variable in the COMMON BLOCK (operation
Status Vector) of module section PERLND at any point in time
during the simulation period. Among the situations in which
this capability can prove useful are the following:
(1) Representation of natural events which are not
adequately portrayed by model algorithms.
(2) Representation of discrete man-made events or
impacts.
(3) Control of output for critical periods.
Using the Special Actions Block to account for a process in
nature is essentially a corrective action necessitated by
observed deficiencies in the algorithms used to represent
the process. For example, in some model applications the
standard practice of inputting a constant value for
infiltration capacity is not appropriate. Since freeze or
thaw of the ground alters the infiltration and storage
capacity of soil, seasonal adjustment of the infiltration
capacity parameter (INFILT) may be required in order to
adequately model the seasonal differences in runoff
generation due to ground conditions (Donigian et a 1 . ,
1983a).
Many activities related to agriculture, silviculture,
construction, and mining can have significant effects on the
hydrologic and water quality processes considered by HSPF.
The influence of such activities on modeled processes can be
represented by using the Special Actions Block to modify the
values of key parameters and/or variables of the PERLND
module section at appropriate points in time during the
55
-------
simulation period. The Iowa River Basin Study included a
number of situations in which the Special Actions capability
was utilized to represent the effects of agricultural
activities. One example was increasing the value for the
detached sediment storage whenever plowing occurred; this
adjustment was critical to the results for watershed
sediment washoff simulation. Another example was increasing
the values for land surface and soil storages of fertilisers
and/or pesticides to represent chemical application during
the simulation period. Those interested in using the
Special Actions Block to model agricultural activities are
referred to reports on parameter estimation for modeling
agricultural BMPs (Donigian et al., 1983a) and study
descriptions for application of HSPF to Four Mile Creek,
loua (Donigian et al., 1983b) and the Iowa River Basin
(Imhoff et al., 1983).
While the Special Actions Block was originally introduced
into HSPF in order to allow the modeling of agricultural
activities such as plowing, cultivation, fertilizer and
pesticide application, the ability to alter the value of
variables at intermediate points during the simulation
period can be used in a number of creative and effective
ways. In one study the Special Actions Block was used to
increase the values for chemical storage variables
associated with rainfall. Since HSPF (Release No. 7) does
not model the quality of precipitation, chemical storage
values were increased on a monthly basis commensurate with
the quantity of rainfall and associated chemicals occurring
each month. In another case a model user employed the
Special Actions Block at an intermediate point during the
simulation to alter the value of the parameter which
specifies the print interval for output; by doing so it was
possible to generate the detailed output necessary to
understand results from a critical period of simulation
without printing unnecessary information during the
remainder of the simulation period.
It is important for the model user to consider whether or
not the use of the Special Actions Block to alter values of
variables used in the PERLND module section can improve the
model representation of the physical processes which are
being simulated. If so, Section 4.03 of Part E and Section
4.10 of Part F of the User's Manual provide the necessary
details to utilize this option. The proper input format for
Special Actions instructions is further illustrated in the
sample input sequence contained in Appendix A.
56
-------
SECTION <4
OPERATIONAL ASPECTS OF HSPF USE
The third step to applying HSPF is familiarizing oneself
with the mechanics of the model so that the input sequences
necessary to build the timeseries data base (Time Series
Store) and execute simulation runs can be developed.
While creation and modification of input sequences will be a
continuing process throughout the later stages of model
application, it is useful, particularly for the new model
user, to study and understand the general operational
aspects of HSPF prior to attempting to use the model.
Preliminary knowledge of HSPF operations will allow the user
to eliminate much of the cost and frustration involved in a
trial-and-error approach to running the model. * The goal of
this section is to provide an overview of considerations
involved in running HSPF and developing input sequences, and
to direct the user to the proper places in the User's Manual
for additional information.
U.1 Steps in Running HSPF
A necessary first step prior to actually running HSPF is to
obtain the current version of the program. The complete
HSPF system including source code, documentation, and stand-
alone programs is available on tape from the U.S. EPA, and
may be obtained by writing to:
Center for Water Quality Modeling
U.S. Environmental Protection Agency
College Station Road
Athens, GA 30613
The distribution tape includes the following files:
- source code for HSPF
- input sequence to compile HSPF
- object code for HSPF
- input sequence to link HSPF
- HSPF Information File (INFOFL)
- HSPF Error File (ERRFL)
- HSPF Warning File (WARNFL)
57
-------
- HSPF test input
- HSPF test output
- lists of HSPF subroutines (by no. and alphobetical1y)
- HSPF User's Manual (text only)
- HSPF OSV's and data structures
- PERLND variable memory addresses (for use in
SPECIAL ACTIONS)
- source code for NEWTSS
- IBM input sequence to compile and link NEWTSS
- NEWTSS IHFOFL
- NEWTSS ERRFL
- NEWTSS test input
- NEWTSS test output
- FTABLE generation program
Once this tape is obtained and the necessary :£iles have been
transferred to the user's computer system, the following
steps in actually running HSPF are required: (1) compilation
and testing of HSPF and NEWTSS, (2) creation of the Time
Series Store (TSS), (3) development and running of input
sequences, and (4) analysis of the results. The compilation
and testing process will be unnecessary if HSPF is already
operational on the user's computer or on another system.
Otherwise, the HSPF source code (available on the
distribution tape) is required, and must be compiled,
linked, and tested. For installation on computer systems
other than IBM, the user may have to modify the source code
according to system specific instructions available from the
U.S EPA (Athens, GA). HSPF has been sucessfully operated on
a variety of computer systems, such as JLBM., D E C V A){. C_D_C_,
HP3000, and Harris.
Creation of the TSS involves the actual creation of a TSS
file using the stand-alone program NEWTSS, creation of
individual dataset labels in the TSS with the TSSMGR module,
and subsequent input of data (time series) to the TSS using
the COPY and/or MUTSIN modules. This process is described
in detail in Section 5.0 of this document.
Developing and running input sequences, and analysis/display
of the results using the various capabilities and options of
HSPF are the primary operational aspects to be considered in
this section.
t. 2 Overview of HSPF Input
HSPF input sequences consist of the required job control
language (JCL) and one or more HSPF 'input sets'. An input
set is either a TSSMGR input set used to create, modify, or
destroy labels of individual datasets in the TSS, or a RUN
input set, used to perform all other operations of HSPF.
58
-------
The input set is further divided into groups of text lines
(card images). The groups are called "blocks' and may
appear in any sequence in a run; however, a natural or
logical sequence exists, and will be presented here as an
example. Both the new and experienced HSPF user will find
this sequence useful for operational purposes, i.e. ease of
development and modification, and also for understanding and
presentation.
Table 4.1 lists the various blocks of an HSPF input sequence
with reference to the corresponding section(s) in the User's
Manual where additional information and guidance in the
development of that block's input is available. Of course,
a single input sequence or set would not often include every
block; however, a RUN input set must include the GLOBAL and
OPERATION SEQUENCE blocks, at least one OPERATION-type block
(PERLND through MUTSIN in Table U.I), and one of the three
time series transfer blocks (EXTERNAL SOURCES, NETWORK,
EXTERNAL TARGETS). A TSSI1GR input set must include at least
one of the TSSM operational blocks. Several sample HSPF
input sequence outlines are shown in Table 4.2; each set is
a list of the blocks required or typically found in a
different type of input sequence. In addition, a short
description of the run or its function(s) is included. An
example of a complete HSPF input sequence is included as
Appendix A of this document.
The development and manipulation of complex input sequences
for HSPF can be a time-consuming process due to the large
number of user options available. The following list of
recommendations is intended to assist the user in this task
and to facilitate error detection and isolation.
Input Sequence Development
- Consult pertinent sections of the User's Manual
(User's Control Input) for the appropriate format,
create an outline of the run consisting of the
required "blocks", following the sequence given in
Table 1.1.
- Add the required input tables (see Part F, Section
4.4) for each block including all known parameter
values or an easily recognizable dummy value
(e.g., 'xxxx') where the value is to be inserted
later.
- Freely include comment lines (delineated by 3
astericks - *##) in the input sequence to explain
options used, default values assumed, parameter
value units, and to delineate the input format.
Use the comments included in Part F of the User's
59
-------
TABLE 4.1 HSPF INPUT BLOCKS AND RECOMMENDED SEQUENCES
RUN Input Set
JCL *
GLOBAL Block *
OPERATION SEQUENCES Block #
SPECIAL ACTIONS Block
PERLND Block
IMPLND Block
RCHRES Block
FTABLES Block
COPY Block
PLTGEN Block
DISPLY Block
DURANL Block
GENER Block
MUTSIN Block
EXTERNAL SOURCES Block
NETWORK Block
EXTERNAL TARGETS Block
TSSMGR Input Set
JCL *
ADD Block
UPDATE Block
SCRATCH Block
EXTEND Block
SHOWSPACE, SHOWDSL, and
SHOUTSS Blocks
User's Manual Reference(s)
none (use examples on
distribution tape)
F 4. 2
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
3
10, E
4(1),
4(2),
4(3),
5,
4(11)
4( 12)
4(
4(
4(
4(
f
6 .
f
3)
4)
5)
6)
,
9
9
9
9
,
9
f
9
4
4
4
4.
E
E
E
E
E
E
E
E
E
.6
.6
.6
03
14 .
14 .
4 .
4
4
4
4
4
4
.. 2
..3
., 4
2( 1
2(2
2(3
. 2(
.2(
.2(
. 2(
.2(
.2(
1)
2)
3)
4)
5)
6)
, 4.6.5
, 4.6.5
, 4.6.5,
none (use examples on
distribution tape)
2.3
2.4
2.5
2.6
F
F
F
F
F
4.6.6
2.7
* - Always required
Manual or simply modify a sample
included on the distribution tape.
input sequence
When modifying an input sequence, user options for
specific operations may be easily removed by
deleting the corresponding entry in the OPERATIONS
SEQUENCE block or making it a comment line. The
corresponding input tables for that option may be
left intact, or for clarity, they may be deleted
or 'commented out*.
When mod i f ying
input sequence
subsequent run
an input sequence, save the old
file for reference until the
has been sucessfully executed or
60
-------
unt i1 a uel1
concluded .
defined set
of
runs
has been
Maintain a master input sequence file with all
blocks and all tables included. This may be used
as a base sequence from which a new, functional
sequence could be created with minimum effort by
merely deleting all unwanted options and adding
new parameter/variable values.
TABLE 1.2 EXAMPLES OF INPUT BLOCKS REQUIRED FOR HSPF RUNS
RUN T_Y_PE
TSSM Label Run
TSS Data Input
Run
PCRLND Run
RCHRES Run
Watershed Run
BLOCKS REQUIRED
JCL
ADD Block
JCL
GLOBAL Block
OPER. SEQ. Block
COPY Block
EXT. SOURCES Block
EXT. TARGETS Block
JCL
GLOBAL Block
OPN. SEQ. Block
SPECL. ACT. Block
PERLND Block
PLTGEN Block
DISPLY Block
EXT. SOURCES Block
NETWORK Block
JCL
GLOBAL Block
OPN. SEQ. Block
RCHRES Block
FTABLES Block
PLTGEN Block
DISPLY Block
EXT. SOURCES Block
NETWORK Block
JCL
GLOBAL Block
OPN. SEQ. Block
SPEC. ACT. Block
PERLND Block
RCHRES Block
FTABLES Block
PLTGEN Block
DISPLY Block
DURANL Block
EXT. SOURCES Block
NETWORK Block
DESCRIPTION
Add label(s) to the TSS
Input time series data to
the TSS from a sequential
file.
Simulate hydrologic and
water quality processes
on a pervious land seg-
ment and output selected
time series results graph-
ically and as tables.
Simulate hydraulic and
water quality processes
in a stream or mixed res-
ervoir reach and output
selected time series
results graphically and
as tables.
Combination of PERLND
and RCHRES runs including
plots, durational analy-
ses and tabular displays
of selected time series
results.
61
-------
Error Detec t i on
- Interpretation of errors which occur before
execution i.e., Run Interpreter errors, may be
facilitated by changing the Run Interpreter Output
Level to a higher value (maximum = 10) and
executing an "interpret only" run. (See references
to the Global Block in the User's Manual).
- Detection and isolation of more subtle errors
uhich occur only during execution may be aided by
changing the output flags in an operation block to
obtain printout of results at each interval or
timestep of the run. Cost reductions during this
debugging process can be realized by "turning off"
all operations and options uhich are obviously
unrelated to the error, and also by limiting the
time span of the run. Note that printout of
results at each interval will create large volumes
of output, so this option should only be used for
limited time span runs.
- Warning messages due to mass balance differences
in the PERLND module may be caused by operations
performed in the SPECIAL ACTIONS Block. For
example, chemical applications performed through
SPECIAL ACTIONS will generate a mass balance error
for the specific chemical state variable modified.
The user should examine these warnings and verify
their source.
- Error and Warning messages printed with the HSPF
output and some additional pertinent information
may be found in the HSPF Information File, Error
File, and Warning File. The user should have a
listing of these files for reference purposes.
t.3 Output Options
Due to the diversity and flexibility of HSPF output options,
the user should pay particular attention to this subject in
the development of input sequences. Often, analysis of the
results of a run may be greatly facilitated and improved by
judicious choice of output types and format. The following
overview is intended to provide a brief guide to this
subject; however, the user should consult the appropriate
sections of the User's Manual for more detailed descriptions
and for direction in the use of HSPF output options.
The basic output which is available from each of the HSPF
physical process operation module sections (e.g., Section
62
-------
PUATER in module PERLND) may be printed at each time step of
the run or at multiple time step intervals including daily,
monthly, and yearly summaries. This output basically
consists of all state variable values related to the section
in addition to detailed material fluxes over the printout
interval. The printout frequency is user controlled through
the PRIHT-INFO tables in the PERLND, IMPLND, and RCHRES
input blocks. The user may also specify the units system
(English or Metric) used for all printout and output time
series through the GEM-INFO tables of these blocks.
In those cases uhere the user is only interested in specific
variables or time series, the selective printing of these
time series in the form of "displays" (see below) may be
more convenient than the standard output while
simultaneously saving printing costs. In addition, for
those active module sections which are not pertinent to the
run analysis, the user may save printing costs by
selectively specifying that no output be produced.
Display Time Ser i es
While the standard output discussed above usually includes
much of the necessary information regarding a run, the user
may display any time series computed in a run or input to it
in a convenient format by using the DISPLY module. In order
to determine which time series are computed by each module
in a run (and available for output) the user should consult
the Time Series Catalog (Part F, Section 4.7 of the User's
Manual). Sample outputs from the DISPLY module are shown in
Figures 4.1, 4.2, 4.3.
The user can elect to display the data in a "long-span
table" or a "short-span table." The term "span" refers to
the period covered by each table. A short-span table
(Figures 4.1 and 4.2) covers a day or a month at a time and
a long-span table (Figure 4.3) covers a year.
The user selects the time-step for the individual items in a
short-span display (the display interval) by specifying it
as a multiple (PIVL) of INDELT. For example, the data in
Figure 4.1 are displayed at an interval of 5 minutes. This
could have been achieved with:
IHDELT PIVL
5 mi n 1
1 min 5
63
-------
TSS 2 Precip. (in/100)
Summary for DAY I97
-------
month-value, or year-value, one of five "transformation
codes" can be specified:
SUM Sum of the data
AVER Average of the data
MAX Take the max of the values
at the smaller time step
MIN Take the minimum
LAST Take the last of the values
belonging to the shorter
t ime step
SUM is appropriate for displaying data like precipitation;
AVER is useful for displaying data such as temperatures.
The DISPLY module incorporates a feature designed to permit
reduction of the quantity of printout produced when doing
short-span displays. If the "row-value" ("hour-sum" in
Figure U.I; "day- average" in Figure 4.2) is less than or
equal to a "threshold value," printout of the entire row is
suppressed. The default threshold is 0.0. Thus, in Figure
4.1; data for dry hours are not printed.
The user can also specify:
a. The number of fractional digits to use in a display.
b. A title for the display.
c. A linear transformation, to be performed on the
data when they are at the INDELT time interval
(i.e., before module DISPLY performs any
aggregation). By default, no transformation is
per f ormed .
Plot Time Ser i es
One or more time series may also be diplayed graphically
using the PLTGEN module, which prepares the time series for
plotting; and a stand alone plotting program which reads
the prepared plot file and translates its contents into
information used to drive a plotting device. User-
controlled PLTGEN options include coordinate axis scaling
factors, plot and coordinate axis titles, and various curve
drawing options. Alternative uses of an HSPF plot file
(PLOTFL) are:
1. To display one or more time series in printed
form. For example, to examine the contents of a
dataset in the TSS, run it through PLTGEN and list
65
-------
the contents
terminal.
of PLOTFL on
line printer or
2. To feed time series to some other stand-alone
program. For example, one could specify the
contents of PLOTFL as input to a program which
performs statistical analysis or computes cross
correlations between time series.
A stand-alone plot program which will read an HSPF plot file
and drive a plotter is required and must be supplied by the
user .
Other T ime Series
Uti lities
Other analysis capabilities of HSPF involve manipulation and
analysis of time series using the GEHER and DURAML modules.
GENER allows the generation of a new time series by an
operation on one or two existing time series. The operation
is specified by supplying an "operation code" (OPCODE).
TSS 3 Temperature (Deg F)
Summary for MONTH I974/ 8/
Data interval: 120 mins
DAY
AVER
Interval Number.
3*5
10
I I
MONTH AVER: 6.84059E+01
Figure 4.2 Sample short-span display (second type) from the DISPLY module of HSPF
12
1
2
3
4
5
6
7
8
9
10
11
12
13
11
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
63.
68.
68.
64.
64.
66.
66.
70.
68.
69.
72.
70.
70.
65.
67.
70.
66.
66,
70,
73,
74,
73
73
66
64
72
73
60
62
66
67
8
8
6
0
9
7
6
3
7
6
8
8
3
5
, 1
, 1
.8
.2
.3
.8
.7
.3
.4
.2
.0
.9
.8
.3
.7
.9
.0
54.5
61.0
65.5
54.5
58.5
57.5
55.5
64.0
62.5
60 .5
68.5
62.5
65.5
57.5
56.5
62.5
63.5
55.5
59.5
64.5
65.5
65.0
67.5
60.5
51.5
60.5
67.5
55.0
53.5
59.0
62.0
53.5
60.0
65.0
53.5
57.0
56.0
53.5
63.0
61.0
59.0
68.0
61.0
64.0
55.5
55.5
61.0
62.0
54.0
58.0
63.0
64.5
63.5
66.0
58.5
49.5
59.0
66.0
53.0
52.0
58.0
61.0
52.5
59.0
64.0
52.5
56.0
55.0
52.5
62.0
60.0
58.0
67.0
60.0
63.0
54.5
54.5
60.0
61.0
53.0
57.0
62.0
63.5
62.5
65.0
57.5
48.5
58.0
65.0
51 .5
51.0
57.0
60.0
53.0
60.0
64.5
53.5
57.0
56.5
53.5
63.0
61.0
59. 0
67.5
61.0
64.0
55.5
55.5
61.0
61.5
54.5
58.5
63.5
64.5
63.0
66.0
58.0
50.0
59.5
66.0
52.0
52.0
58.0
61.0
59.5
65.0
68.5
60.0
62.0
63.5
61.0
68.0
66.0
65.0
71.5
67.0
69.0
62.0
62.5
67.0
65.5
61.5
65.5
70.0
71.0
69.5
71.5
64.0
58.0
67.5
72.5
57.0
58.5
63.5
66.0
68.0
72.5
73.5
69.0
69.5
73.5
7!. 5
75.0
73.5
74.0
77.5
76.0
76.0
71.0
72.5
75.0
71.5
72.5
76.5
80.0
81.0
78.0
79.5
73.0
70.0
79.5
81.0
64.5
67.0
71.5
73.0
74.5
77.5
77.0
75.5
74.5
80.0
79.0
79.5
78.5
79.5
81.0
81.5
80.5
77.5
79.0
80.5
75.0
79.0
83.0
86.0
87.0
84.5
84.5
78.5
78.0
87.0
87.5
69.5
73.5
76.5
77.5
76.0
79.0
78.0
77.0
76.0
82.0
81.0
81.0
80.0
81.0
82.0
83.0
82.0
79.0
81.0
82.0
76.0
81.0
85.0
88.0
89.0
86.0
86.0
80.0
80.0
89.0
89.0
71.0
75.0
78.0
79.0
74.5
77.5
76.0
75.0
74.0
79.5
79.5
79.0
78.0
79.5
80. 0
81.5
80.0
77.0
79.0
80.0
74.0
79.0
83.0
86.0
87.0
84.0
84.0
77.0
78.0
87.0
85.5
69.5
73.5
76.5
76.5
71.0
75.0
70.5
71.0
70.0
73.5
75.5
75.0
73.5
77.0
75.5
77.5
74.0
72.0
75.0
76.0
69.5
74.5
78.5
81.0
81.5
80.0
78.0
71.0
73.5
82.5
78.5
65.5
70.0
73.0
70.5
66.0
71.0
63.5
65.5
64.5
65.0
70.5
69.5
67.5
73.0
69.5
71.5
66.0
65.0
69.5
70.5
63.5
67.5
72.5
74.0
74.0
74.5
70.0
62.0
67.5
75.5
67.5
59.5
65.0
68.0
62.0
62.5
67.5
57.0
60.5
59.5
53.5
66 .0
64.5
63.0
70.0
65.0
67 . 0
60.0
59.0
64.5
t'j.5
58.0
62.0
67.0
68.0
68.0
69.5
63.0
54.5
63.0
70.0
59.0
55.0
61.0
64.0
55.5
66
-------
Table 4.3 lists the currently available transformations or
operations performed by GEHER. In Table 4.3, A and B are
input time series, and C is the resulting output time
series. A typical application of GENER might be the
calculation of a chemical concentration by dividing the mass
outflow by the water outflow from a reach. The user may
also find it convenient to add new operations to GENER by
modifying the HSPF source code. For example, a recent
application by the Denver Regional Council of Governments,
required the incorporation of an "urban irrigation function"
which was implemented through the development of a neu GENER
operation. Further information related to GENER may be
found in Part E Section 4.2(15) and Part F Section 4.4(15)
of the User's Manual.
DURANL performs duration and excursion analyses on a time
series, computing a variety of statistics relating to its
excursions above and below certain specified "levels."
Day
JAN
TSS 3 Temperature (Deg F)
Annual data display: Summary lor period ending 1974/12
FEB MAR APR MAY JUN JUL AUG SEP
OCT
53. 4
64.6
71.1
68. 4
AVER 25.1 21.6 35.3
-------
Typical applications of DURANL are
3.
Examination of flow modeling results by comparing
simulated and observed flou frequency information
such as the percent of time the flows exceeded
certain specified levels.
Analysis of the frequency and duration of
dissolved oxygen levels to evaluate aquatic
impacts of various uaste-load applications or
water quality management options.
Lethality analysis of chemical concentration time
series. The frequency or percent of time acute,
chronic, and sublethal conditions (pertinent to a
particular aquatic organism) might be determined
for a stream from a simulated time series of
chemical concentrations.
TABLE "4.3 OPERATIONS PERFORMED BY THE GENER MODULE OF HSPF
OPCODE
1
2
3
6
7
8
9
10
11
12
13
It
15
16
17
18
19
20
21
22
Act i orj
C= Abs value (A)
C= Square root (A)
C= Truncation (A)
eg. If A = 4.2, C = «4.0
A=-3.5, C=-3.0
C= Ceiling (A). The "ceiling" is
the integer >= given value.
eg. If A=3.5, C=t.O
A=-2.0, C=-2.0
C= Floor (A). The "floor" is the
integer <= given value.
eg. If A=3.0, C=3.0
A=-2.7, C=-3.0
C= loge (A)
C= loglO (A)
C= K(1)+K(2)*A+K(3)*A**2 (up to 7 terms)
The user supplies the no. of
terms and the values of the
coef ficients (K).
c=
c=
c=
c =
c=
c=
c=
c=
c=
c=
c=
c=
c=
c=
K**A
A**K
A + K
Sin
Cos
Tan
Sum
A + B
A-B
A*B
A/B
MAX
HIM
A#*B
(A)
(A)
(A)
(A)
(A,
(A,
B)
B)
6G
-------
Further information regarding DURANL and its options may be
found in Part E Section 4.2(14) and Part F Section 4.4(14)
of the User's Manual.
Generally, as the user gains experience with HSPF, and
becomes more familiar uith the output and analysis options
available/ he begins to utilize them more fully to improve
the analysis of the results. Examples of much of the common
types of output used in typical hydrologic/water quality
studies of agricultural watersheds may be found by examining
the input sequence included as Appendix A of this document.
Generally, long-span displays of stream flow both in units
of depth over the watershed and flow units (cms) are
included along with concentrations of sediments, pesticides
and agricultural nutrients, and the corresponding areal
loadings of these materials. Typical plots include many of
the same quantities. GENER is typically used to generate
concentrations which are not computed internally by HSPF.
These concentration time series are then displayed using
DISPLY or PLTGEN. Of course many of the results used in the
calibration/verification process may not be required for the
final production runs and may be eliminated to save
computation and printing costs in these runs.
-------
SECTION 5
INPUT AND MANAGEMENT OF TIME SERIES DATA
All HSPF simulation runs involve the use and/or generation
of data in the form of time series. This section describes
the storage, retrieval, and management of time series data
using HSPF utility routines, stand-alone programs, and a
large random access file known as the Time Series Store
(TSS). Topics to be discussed include evaluation of TSS
size requirements, creation of a new TSS file, addition of
TSS dataset label and directory information, input of time
series to the TSS, and general TSS management tools
available within the HSPF system. More specifically, this
section provides a guide to the user in the execution of the
following steps required in any HSPF application where time
series data are manipulated.
Estimate the size of the TSS
Create a TSS with NEWTSS
Create individual dataset
with TSSMGR
labels in the TSS
Input data to the TSS with COPY
Input data to the TSS with MUTSIN
Maintain the TSS with TSSMGR
Where feasible, examples will be presented in order to
clarify the discussion, and relevant sections of the HSPF
User's Manual will be referenced for additional information.
5.1 Creation of a Time Series Store
The Time Series Store (TSS) provides a convenient library
for storage of time series in the HSPF environment. The TSS
consists of a single, large, direct access disc file; HSPF
subdivides this space into many datasets containing time
series, and a directory keeps track of the datasets and
their attributes. Before time series are stored in the TSS,
the file must be initialized and its directory created.
70
-------
This is done by executing the separate program NEWTSS which
is available on the standard HSPF release tape (Section 4)
and documented in Appendix III of the User's Manual. When
running NEWTSS, the user specifies general attributes of the
TSS file including total size, and Fortran unit number.
An estimate of the amount of data to be stored in the TSS is
required to provide a size specification in the NEWTSS
input. At the beginning of an application, this estimate may
be difficult to make due to uncertainty about exactly what
data are both required by the model study and available.
However, if the TSS file is discovered to be too small (or
large) after it has been set up and filled with data, it is
a relatively simple process to open a new TSS file of
different size and copy the contents of the current TSS into
it. This is accomplished with the COPY option contained in
the NEWTSS program.
The first step in estimating the TSS file size is to make an
inventory of all available and expected (i.e. simulated)
time series data including time step, period of record,
source of data, and data format. When a complete inventory
of all required data sets has been completed, a simple
equation may be used to calculate the TSS size
specifications required in the NEWTSS input. Factors
required for the equation include the information compiled
in the inventory and any pertinent compression information
for the various time series. In general, compression of
time series data can significantly reduce the amount of
space required for its storage if many periods of missing or
zero data are present; hence compression options should be
utilized whenever feasible. These options are more
completely described in Part F Section 2 of the User's
Manual, and will be referenced in Section 5.2 of this
document. Convenient guidelines including detailed
worksheets and instructions are available in Appendix III of
the User's Manual to assist the user in both steps of the
TSS file size estimation process.
Creation of the TSS file requires the execution of NEWTSS.
The NEWTSS input sequence used to create the TSS file for
the Iowa River project is shown below as an example, along
with a definition of input parameters.
OPNTSS
TSS FILE LENGTH= 960 (TSS file length in records)
MAX. DSNO= 200 (Maximum number of datasets)
TSS FILE N0= 18 (Fortran unit number of the
TSS file)
71
-------
5.2 Adding Dataset Labels
After the TSS file has been initialized and its directory
created using the NEHTSS program, the individual time series
are put in the store using HSPF utility routines. This
procedure requires two separate steps; first, the HSPF
routine known as TSSMGR is employed to create the specific
datasets or labels in the TSS, and second, the actual data
is copied to the newly created datasets using the COPY
and/or MUTSIN routines contained in HSPF.
Before data can be stored in the TSS, individual labels must
be created and space allocated within the TSS file. Data in
the TSS is stored in "datasets", each of which is identified
by a "label". Labels are created or added to the TSS by
executing the ADD option of the TSSMGR routine; they include
such information as an identifying dataset number, amount of
space in the dataset, the unit system of the data,
compression information, time step, and descriptive
information such as name, location, etc. This information
is very important to the correct and efficient storage of
the corresponding time series data; the user should
carefully review Part F, Section 2 of the User's Manual
where the TSSMGR input and user options are described. Note
that much of the information needed may already have been
compiled during the inventory of time series for evaluating
the total size of the TSS (Section 5.1 of this document).
Shown below is an example TSSMGR ADD input sequence which
was used to create a dataset label for a streamflow record
on the Iowa River. Note that input variables which are not
shown will assume their default values.
ADD
DATASET NO = 45
SPACE= 10
NAME= STFLOW
UNITS= ENGLISH
COMPRESSION= UNCOMP
TII1ESTEP= mMO
NMEMS= 1
LOCATIOH= MARENGO, IOWA
MEMBER NAME= STFLOW
KIND= MEAN
5.3 Input of Data
Input of data to the TSS is accomplished by executing either
a COPY or MUTSIN operation of HSPF depending on the format
of the available data. Normally, time series data is
available as a sequential file in which a number of
72
-------
successive data points or intervals are contained on each
line (card image)* and in a particular format. The HSPF
system is designed to read such a file using either a
default format or a user-specified format. The data is
transferred from the sequential file to the TSS dataset by
employing the COPY utility module of HSPF. Listed in Figure
5.1 and described below is an example of the input required
to transfer two time series into the Time Series Store.
The GLOBAL Block specifies the period for which data are
being input (June 1974), and some other general control
in f ormat ion.
The OPN SEQUENCE Block indicates that there are two COPY
operations in the run, the first having a time step of 1
hour and the second 24 hours.
The COPY Block indicates that, for both COPY operations, a
single mean-valued time series is being handled.
The EXT SOURCES Block specifies that:
1. The file with FORTRAN unit no. 31 contains
.hourly data (format HYDHR), in metric units.
Missing records are assumed to contain zeros
(like NWS hourly precipitation cards). The
multiplication factor field is blank, so it
defaults to 1.0. The time series goes to COPY
operation no. 1 time series group INPUT, member
MEAN 1 .
2. The file with Fortran unit no. 32 contains
daily data (format HYDDAY) in metric units.
This time series goes to COPY operation no. 2.
The EXT TARGETS Block specifies that:
1. The output from COPY operation no. 1 (which
came from sequential file no. 31) goes to
dataset no. 25 in the TSS (member PRECIP 1) and
is stored in metric units. The access mode is
ADD.
2. Similarly, the output from COPY operation no. 2
is to be stored in member PETDAT 1 of TSS
dataset no. 26.
Note that the labels for the TSS datasets must have been
previously created, and that the member identification and
unit system information (METR) supplied by the user must
agree with the corresponding information in the label.
73
-------
Figure 5.1 Example of User's Control Input for tho COPY Module
RUN
GLOBAL
Inputting test data to TSS
START 197<+/06 END 1974/06
RUN INTERP OUTPUT LEVEL 3
RESUME 0 RUN 1
END GLOBAL
OPN SEQUENCE
COPY 1 INDELT 01<00
COPY 2 INDELT 2<»<00
END OPN SEQUENCE
COPY
TIMESERIES
«thru» NPT NMN ***
1 2 1
END TIMESERIES
END COPY
EXT SOURCES
<-Volume-> SsysSaap<--Mult-->Tran <-Target vols> <-Grp> <-Member-> ***
« tern strs<-factor->strB « * » * ***
SEQ 31 HYDHR METR2ERO COPY 1 INPUT MEAN 1
SEQ 32 HYDDAY METRZERO COPY 2 INPUT MEAN 1
END EXT SOURCES
EXT TARGETS
<-Volume-> <-Grp> <-Member-x--Mult-->Tran <-Volume-> Tsys Tgap Amd ***
» * 8<-factor->strg * * tern strg
COPY 1 OUTPUT MEAN 1 TSS 25 PRECIP 1 METR ADD
COPY 2 OUTPUT MEAN 1 TSS 26 PETDAT 1 METR ADD
END EXT TARGETS
END RUN
In addition to storage of various input time series data for
HSPF simulations, the TSS also provides a convenient storage
facility for resulting output time series. Any time series
created during an HSPF run is available to be output and
stored in the TSS; hence making it available as input to a
74
-------
future run. For example, one could store the appropriate
time series results of a completed hydrologic calibration,
and subsequently use them as input to water quality
calibration runs thus saving the costs of resimulating the
hydrology in each run.
Transfer of output time series to TSS files is specified in
the EXT TARGETS Block similarly to the second step of the
COPY operation in the previous example (Figure 5.1) except
that the time series source is a HSPF simulation operation
(e.g. PERLND) rather than a COPY operation. For cases uhere
a number of time series are to be output to a single TSS
dataset (either summed or as individual members of the
dataset), the time series should first be collected by using
a COPY operation (specified in the NETWORK Block) and then
output together in the EXT TARGETS Block. For more detailed
information on time series linkages, the user should refer
to Part F, Section 4.6 in the User's Manual. Also, the time
series created by the various simulation modules of HSPF and
available to be output are listed in Part F, Section 4.7,
the Time Series Catalog.
An alternative method of transferring data to a TSS using
the utility module MUTSIN is sometimes required depending on
the format of the external sequential file containing the
data. MUTSIN is designed to read files which have the same
format as an HSPF plot file. Situations in which MUTSIN is
useful include the following:
(1) To input data with a time interval not included
in the standard HSPF sequential input formats.
(Part F, Section 4.9)
(2) To transfer data from one TSS file to another;
This transfer requires the use of the PLTGEN
utility module to output from the source TSS
and MUTSIN to input to the target TSS.
(3) As an interface between HSPF and other
continuous simulation models; the other model
can output results in the form of an HSPF plot
file and MUTSIN inputs the data to a TSS (or an
HSPF simulation run).
5.4 Management of TSS Datasets
During the course of a model application study, general
maintenance functions associated with the data in the Time
Series Store are handled by the TSSMGR module. In addition
to creation of dataset labels in the TSS file, this module
allows the user to perform general "housekeeping" chores
associated with these datasets.
75
-------
Since the storage of additional data (either model input or
output) is often an ongoing process during the course of a
model application, the creation of neu datasets in the TSS
may occur at any time as long as sufficient space exists in
the TSS file. As discussed above (see example in Section
5.2), new datasets are created by using the ADD command of
the TSSMGR module. This action creates a neu label with
various user-specified and default cha rac t er :L s t ics , which is
then available for transfer of data.
Often, it becomes convenient or necessary to remove datasets
from the TSS. Analysis of data which has previously been
input to the TSS or a redefinition of the study goals or
strategy may result in the conclusion that is dataset is no
longer required. Execution of the SCRATCH command of the
TSSMGR module selectively removes the dataset from the TSS
thus making the space available for use by another dataset.
Two TSSMGR commands which are used to modify the attributes
of a TSS dataset label are the UPDATE and EXTEND options.
Use of these commands is required to increase (or decrease)
the space allocated to a dataset (EXTEND) or to modify other
selected label parameters such as the units, name, location,
security parameter, etc. (UPDATE). For example, UPDATE
would be vised to change the security option of a dataset
from WRITE (unprotected) to READ (protected) in order to
avoid inadvertent replacement or damage to its contents.
In order to examine the overall status of the TSS and its
datasets, one employs the SHOWSPACE and SHOWDSL commands.
SHOWSPACE provides a count of the available space in the TSS
for additional datasets. The SHOWDSL command displays the
attributes of selected or all dataset labels in the TSS and
also provides a summary of the TSS data including the period
of record (years) contained in each dataset and the
available space. The commands used to achieve these and all
other functions of TSSMGR are documented in Part F, Section
2.0 of the User's Manual.
76
-------
SECTION 6
MODEL PARAMETERS AND PARAMETER EVALUATION
For the purposes of HSPF the functional definition of a
parameter is an input datum (not a time series) whose value
is not changed by program computations. Each parameter
supplies the program with information which it needs to
perform its operations. Some parameters are control-
oriented while others are process-oriented. The control-
oriented parameters are used to specify program instructions
such as constituents which will be simulated, .how long the
simulation period will be, or how often program results
will be transferred to the line printer. Selecting the best
values for these parameters is entirely dependent on the
needs of the individual user, and consequently guidelines
for their evaluation are not appropriate. The modeler
should review Section 4 of this document for a general
discussion of user-controlled options in executing HSPF;
Part F of the User's Manual contains formatting information
for input of parameter values as well as a brief discussion
of possible options for each parameter.
This section focuses on the process-oriented parameters
needed as input to the application modules of HSPF. Since
the model is designed to be applicable to many different
watersheds, these parameters provide the mechanism to adjust
the simulation for specific topographical , hydrologic,
edaphic, land use , and stream channel conditions for a
particular area. The large majority of the parameters can
be evaluated from known watershed characteristics.
Parameters that cannot be precisely determined in this
manner must be evaluated through calibration with recorded
data .
At the present time the documentation for HSPF does not
contain the type of detailed information on parameter
evaluation which is available for certain of its predecessor
models, such as the Agricultural Runoff Management Model and
the Nonpoint Source Model. While developing comparable
guidelines for evaluating HSPF parameters on a parameter-by-
parameter basis would be a useful task, it is also a
formidable one since there are over 1000 parameters in the
entire HSPF system. Of course, only a small fraction of
these parameters are part of the User's Control Input for
77
-------
any one type of application. The purpose of this section is
to familiarize the user with the types of data which are
needed for parameter evaluation and to direct the user to
existing data and documents which will prove useful in the
evaluation process. The section concludes with a general
discussion of some of the considerations involved in using
and interpreting existing data in order to develop
reasonable parameter values for a specific study area.
6.1 Types of Data Heeded
Sections 2 and 3 of this document explain how various types
of data are used to develop a realistic set of modeling
goals and an effective modeling strategy. Much of the data
used for these purposes is also useful for evaluating model
parameters. Depending on the type of model application*
additional information from maps, reports, and research
literature may also be needed. While a discussion of data
requirements for evaluating individual parameters is beyond
the scope of this document, it is nonetheless useful to
point out the types of data which are needed to develop the
parameters for each section of the three HSPF application
modules. Table 6.1 provides a preliminary list of data
types needed for each section of PERLND, IMPLND, and RCHRES.
Many of the data types listed in the table are sources of
raw data. In some cases the information required to
evaluate certain parameters may have been developed in
previous reports on the study area and use of raw data may
not be necessary.
6.2 Sources of Data
Generally speaking, there are a greater number of data
sources available for evaluation of physical parameters for
a specific study area than there are for evaluation of
chemical/biological parameters. This disparity in data
availablity is largely due to the fact that physical data
related to topograhy, soils, and/or channel geometry are
collected as a necessary part of agricultural,
si1vicultural, construction, and water supply activities
whereas collection of the data necessary to evaluate the
rates and coefficients involved in chemical/biological
interactions is not nearly as widespread. Numerous federal,
state, and local agencies may be able to provide information
useful in developing values for the physical parameters in
HSPF. Among these are:
U.S. Geological Survey
U.S. Army Corps of Engineers
U.S. Soil Conservation Service
73
-------
state geologic surveys
state departments of water resources
* local universities
Although a substantial body of data has been developed on
water qua1ity-rel ated parameters, the data are scattered
throughout journal articles, government documents and
technical reports. This, of course, makes it difficult for
the modeler to obtain necessary guidance in assigning values
to the various constants and coefficients required by the
model. Fortunately, a number of reports and user's manuals
have been produced which will assist the user in this
process. In particular* the following five sources of
information will prove useful in evaluating water quality
parameters for HSPF:
HSPF User's Manual (Johanson et al., 1981)
ARM Model User's Manual (Donigian and Davis,
1978), NFS Model User's Manual (Donigian and
Crawford, 1979)
Tetra Tech Report: Rates> Constants, and
Kinetics Formulations in Surface Water Quality
Modeling (Zison et al., 1978)
CREAMS User's Manual (Knisel, 1980)
HSPF Iowa Study Reports (Donigian et al.,
1983b; Imhoff et al.,1983; Donigian et al.,
1983a)
HSPF User's M.9JQ-LL9-.1. Parts E and F of the User's Manual are
the most useful. Part E contains the functional
descriptions for the important processes modeled by HSPF.
Included in these descriptions are numerous equations which
illustrate how the input parameters are used to adjust the
model computations in order to represent specific study area
conditions. Part F, the User's Control Input, provides
information on how to input necessary parameter values to
the computer program.
ARM,NPS User's Manuals. Both of these manuals contain
guidelines, on a parameter-by-parameter basis* for
evaluation of all process-oriented parameters needed for
their use. Since these two models are predecessors of the
PERLND and IMPLND modules of HSPF, many of the parameters
are shared in common, and the guidelines set down for
evaluating particular parameters are equally applicable to
HSPF. The names for many of these parameters have been
changed to conform to HSPF naming conventions. In order to
expedite the use of the valuable information contained in
79
-------
TABLE 6.1 TYPES AND SOURCES OF DATA NEEDED TO USE THE VARIOUS
SECTIONS OF THE HSPF APPLICATION MODULES.
PERLND
SECTION ATEMP
SECTION SNOW
SECTION PWATER
SECTION SEDHNT
SECTION PSTEMP
SECTION PUTGAS
SECTION PQUAL
SECTION MSTLAY
SECTION PEST
SECTION NITR
SECTION PHOS
SECTION TRACER
topographical nvaps
topographical maps, vegetation maps or
aerial photos, field observation. ARM
User's Manual
vegetation maps or aerial photos, soils
maps, topographical maps, land use maps,
ARM User's Manual, timing of disturbances
soils maps, data on farming practice*,
ARM User's Manual
air temperature data, field soil tempera-
ture data
none
local stormuater quality data, NPS User's
Manual
ARM User's Manual
ARM User's Manual, pesticide Literature,
field data
ARM User's Manual, field application
rates, kinetic data, crop life cycle
ARM User's Manual, field application
rates, kinetic data, crop life cycle
none
IMPLND
SECTION ATEMP
SECTION SNOW
SECTION IWATER
SECTION SOLIDS
SECTION IWTGAS
SECTION IQUAL
topographical maps
topographical maps, vegetation maps or
aerial photos, field observation, ARM
User's Manual
aerial photos, stormuater management
plans, NPS User's Manual
street cleaning data, land use* data,
local stormuater quality data, NPS
User's Manual.
air temperature data, uater temperature
data
local stormuater quality data, NPS
User's Manual
80
-------
TABLE 6.1 (cont'd) TYPES AND SOURCES OF DATA NEEDED TO USE THE VARIOUS
SECTIONS OF THE HSPF APPLICATION MODULES.
RCHRES
SECTION HYDR
SECTION AOCALC
SECTION CONS
SECTION HTRCH
SECTION SEDTRN
SECTION GQUAL
SECTION OXRX
SECTION NUTRX
SECTION PLANK
SECTION PHCARB
channel geometry data, streamflou gage
records and rating curves, topographical
maps
none
none
topographical maps, aerial photos
bed sediment data, instream sediment
loadings data, particle size analyses
laboratory or field kinetic data, liter-
ature values for partition coefficients,
organic matter content of suspended and
bed sediments, environmental conditions
(e.g. pH, temperature)
literature or field kinetic data, channel
bottom samples, instream oxygen and BOD
data
literature of field kinetic data, instream
nutrient data, channel bottom samples
literature or field kinetic data, instream
biotic data
none
the ARM and NFS User's Manuals, a table which equates former
parameter names with the current HSPF names for selected
parameters is included in Appendix C of this document. The
names for hydrology and sediment related parameters (i.e.,
the first two pages) are shared by both the ARM and NFS
models, while the remainder of the parameter names in the
appendix are specific to the ARM Model.
T e t r a Tech Report: Rates, Co n s t a n t s, and Kinetics
Formulati ons i n Surface Hater Qua 1i ty Model inn. This
document is a comprehensive compilation of data on surface
water quality modeling formulations and values for rate
cons tants
1iterature
chemical,
Cur rent 1y,
evaluat ing
processes
and coefficients. The report contains a
review covering a broad spectrum of physical,
and biological processes and formulations.
it is one of the best sources available for
many of the parameters related to the instream
modeled by RCHRES.
31
-------
CREAMS User' s ManuaJ^. CREAMS is a mathematical model
developed by the U.S. Department of Agriculture to evaluate
nonpoint source pollution from field-sized areas. Volume II
of the CREAMS documentation (all three volumes are bound as
one report) provides the user with guidelines for developing
parameter values associated with hydrology, sediment,
nutrient, and pesticide simulation. Parameters are
organized in tabular form, and for each parameter a short
definition, the best source for evaluating the parameter,
and the expected quality of the derived value are listed.
These tables can provide useful suggestions for evaluating
similar HSPF parameters. However, before using parameter
values from CREAMS or any other model as input to HSPF, one
should compare the model formulations in uhich the
parameters are used. In the case of the CREAMS model,
documentation in Volume I should be reviewed to make certain
that parameter definitions are consistent with those used in
HSPF.
HSPF Study Reports. The Four Mile Creek and Iowa River
Reports contain a number of tables which list the values of
parameters used in each study. While the values for some
parameters may vary greatly from one watershed to another,
these tuo reports will provide a basis for a first guess in
developing values for certain parameters. More detailed
evaluation guidelines and additional values for runoff,
sediment, and chemical parameters are contained in Section <4
of the report entitled "HSPF Parameter Adjustments to
Evaluate the Effects of Agricultural Best Management
Practices (Donigian et al., 1983a)." In many cases the
specific parameter values in this report are pertinent only
to the Iowa-Cedar River Basin, while the guidelines describe
how to estimate parameter values for other areas or
cond i t ions.
The reports and manuals described above give some guidance
in parameter evaluation and in some cases provide a range of
reasonable values for individual parameters: nonetheless
local field data is still the most reliable means of
parameter evaluation. The modeler should carefully review
the reports and documents which have previously been
prepared for the study area to insure that data useful in
parameter evaluation is not overlooked.
6.3 General Considerations
Selecting parameter values almost always requires
considerable interpretation and/or extrapolation of data.
Given the scarcity of definitive guidelines* engineering
judgement and a good understanding of model algorithms are
crucial to the process. The modeler should keep the
82
-------
following points in mind uhile performing this task. A
number of the points discussed are adapted from the Tetra
Tech report cited in Section 6.2.
(1) Selection of reasonable values for physical
parameters is a critical first step to all
model applications. If the physical attributes
of the study area are not represented correctly
and with adequate det'ail, it uill be difficult
to perform a realistic hydrologic simulation,
and without good hydrologic results it is not
possible to obtain reliable sediment or water
quality results.
(2) Values for a number of parameters, in
particular physical parameters, vary on a
seasonal basis. When attempting to develop
values for parameters related to rainfall
interception, upper zone water storage
capacity, land surface roughness,interflow, or
evapotranspi ration from the soil profile,
remember that it may be appropriate to develop
values on a monthly basis. One should also
assess which parameters, if any, are affected
by activities on the land surface which are not
specifically modeled by HSPF. It may be
desirable to modify values for certain
parameters coincident with a particular
activity by using the Special Actions Block
(Section 3.5). If so, the user needs to
develop a value for base conditions and one or
more additional values representative of the
activities which disturb the base condition.
(3) There is rarely concensus on how best to select
a value for a particular water quality rate or
coefficient. Generally, there are a great many
environmental factors influencing a given rate
parameter. The factors can be complex, and
their influence on rate constants inadequately
quantified. In come cases, such as in modeling
stormwater runoff quality, there may be so many
physical and chemical factors involved that
developing a satisfactory mechanistic model may
be impractical or beyond the state of the art.
In such cases a parameter is often relegated to
being a calibration parameter.
(*4) The Tetra Tech report cautions against blind
use of literature values for parameters,
particularly rate parameters, by noting that
some researchers believe that some surface
water quality parameters are highly system-
specific based on the commonly large
differences in observed rates from system to
system. $3
-------
SECTION 7
MODEL CALIBRATION AND VERIFICATION
The calibration and verification process is critical to the
application of HSPF. In this section ths process will be
defined and described, and recommended procedures and
guidelines will be presented. The goal is to provide a
general calibration/verification methodology for users of
the model. As one gains experience, the methodology will
become second-nature and individual methods and guidelines
will evolve.
7.1 General Calibration Procedures
Calibration is an iterative procedure of parameter
evaluation and refinement by comparing simulated and
observed values of interest. It is required for parameters
that cannot be deterministical1y evaluated from topographic,
climatic, edaphic, or physical/chemical characteristics.
Fortunately, the large majority of HSPF parameters do not
fall in this category. Calibration should be based on
several years of simulation (3 to 5 years is optimal) in
order to evaluate parameters under a variety of climatic,
soil moisture, and water quality conditions. The areal
variability of meteorologic data series, especially
precipitation and air temperature, may cause additional
uncertainty in the simulation. Years with heavy
precipitation are often better simulated because of the
relative uniformity of large events over a watershed. In
contrast low annual runoff may be caused by a single or a
series of small events that did not have a uniform areal
coverage. Parameters calibrated on a dry period of record
may not adequately represent the processes occurring during
the wet periods. Also, the effects of initial conditions of
soil moisture and pollutant accumulation can extend for
several months resulting in biased parameter values
calibrated on short simulation periods. Calibration should
result in parameter values that produce the best overall
agreement between simulated and observed alues throughout
.the calibration period.
Calibration includes the comparison of both monthly and
annual values and individual storm events. Both comparisons
-------
should be performed for a proper calibration of hydrology
and water quality parameters. When modeling land surface
processes, hydrologic calibration must precede sediment and
water quality calibration since runoff is the transport
mechanism by which nonpoint pollution occurs. Likewise,
adjustments to the instream hydraulics simulation should be
completed before instream sediment and water quality
transport and processes are calibrated. The overall
calibration scheme for a model application including
hydrology, sediment and water quality simulation is outlined
below. The outline is divided into two parts: land surface
calibration and instream calibration.
Land Surface Cali brat i on ( P E R L N D , IMPLND).
(1) Estimate individual values for all parameters.
(2) Perform hydrologic calibration run, including
snowmelt simulation, if necessary.
(3) Compare simulated monthly and annual runoff
volumes with recorded data.
(4) Adjust hydrologic calibration parameters, and
initial conditions if necessary, to improve
agreement between simulated monthly and annual
runoff and recorded values.
(5) Repeat steps 2, 3, and 4 until satisfactory
agreement is obtained.
(6) Compare simulated and recorded hydrographs for
selected storm events.
(7) Adjust hydrologic calibration parameters to
improve storm hydrograph simulation.
(8) Perform additional calibration runs and repeat
step 7 until satisfactory storm simulation is
obtained while maintaining agreement in the
monthly and annual runoff simulation.
If sediment is simulated:
(9) Perform calibration run for sediment
parameters.
(10) Compare monthly and annual sediment loss with
recorded values, if available.
(11) Compare simulated storm sediment graphs with
recorded values for selected events.
85
-------
(12) Adjust sediment calibration parameters to
improve the simulation of monthly and annual
values and storm sediment graphs.
(13) Repeat steps 9, 10, 11 and 12 until
satisfactory sediment simulation is obtained.
If water quality is simulated:
(14) Perform calibration run for water quality
parameters.
(15) Compare simulated monthly and annual pollutant
loss with recorded values, if available.
(16) Evaluate pollutant state variables (e.g.
surface and soil storages) and compare with
recorded data, if available.
(17) Compare simulated and recorded pollutant graphs
(concentration and/or mass removal) with
recorded data for selected events.
(18) Adjust relevant water quality parameters (i.e.
accumulation/uashoff pollutant potency,
adsorption, decay, leaching and perform
additional pollutant calibration trials until
satisfactory agreement is obtained.
At the completion of the above steps, HSPF is calibrated to
the watershed being simulated under the land use conditions
in effect during the calibration period.
Inst ream Cali brati on (RCHRES)
(1) Estimate initial values for all parameters.
(2) Perform hydraulic simulation run.
(3) Compare simulated and recorded streamflou
hydrographs for calibration period.
(4) If hydraulic routing results do not appear
reasonable adjust FTABLE values, and initial
conditions if necessary, to improve agreement.
(5) Repeat steps 2, 3 and 4 until satisfactory
agreement is obtained.
If water temperature is simulated:
(6) Perform calibration run for temperature
parameters.
86
-------
(7) Compare simulated temperature graphs for
calibration period with recorded values, if
available.
(8) Adjust temperature calibration parameters to
improve agreement between simulated and
observed values.
(9) Repeat steps 6, 7 and 8 until satisfactory
temperature simulation is obtained.
If sediment is simulated:
(10) Perform calibration run for sediment
pa rameters.
(11) Compare simulated monthly and annual sediment
loadings uith recorded values, if available.
(12) Compare simulated storm sediment graphs with
recorded values for selected events,
(13) Analyze behavior of bed sediments compared to
available data.
(11) Adjust sediment calibration parameters to
improve the simulation of monthly and annual
values and for individual storms.
(15) Repeat steps 10 through 14 until satisfactory
sediment simulation is obtained.
If generalized quality constituents are simulated (GQUAL):
(16) Follou the same procedure which was outlined in
steps 10 through 15 for sediment.
If dissolved onygen and BOD are simulated and nutrients and
plankton are not:
(17) Perform dissolved oxygen and BOD calibration
run.
(18) Assess the effects that parameter values are
having on DO and BOD simulations by examining
printed output and constituent graphs.
(19) Compare constituent graphs with observed
values, if available.
(20) Adjust oxygen parameter values to improve the
simulation of both DO and BOD simultaneously.
87
-------
(21) Repeat steps 17, 18, 19 and 20 until the best
agreement between simulated and observed values
is obtained for both constituents.
If nutrients are simulated and plankton are not:
C22) Perform nutrient calibration run.
(23) Assess the effects that nutrient parameters are
having on DO and nutrient simulations by
examining printed output and constituent
graphs.
C2U) Compare constituent graphs with observed
values, if available.
(25) Adjust nutrient calibration parameters to
improve the simulation of DO (if nitrification
is simulated) and nutrients. If adjustments
improve nutrient simulation but harm the DO
simulation, consider whether adjustment of DO
parameters can compensate.
(26) Repeat steps 22, 23, 2U and 25 until the best
agreement between observed and simulated values
is obtained for both DO and nutrients.
If plankton are simulated:
(27) Perform plankton calibration run.
(28) Assess effects that plankton simulation is
having on dissolved oxygen, BOD, nutrient, and
plankton values by examining printed output and
constituent graphs.
(29) Compare constituent graphs with observed values
i f available.
(30) Adjust plankton calibration parameters to
improve the simulation of most or all of the
affected constituents. Consider adjusting
calibration parameters other than plankton
parameters, if necessary (i.e., DO, BOD or
nutrient parameters).
(31) Repeat steps 27, 28, 29 and 30 until the best
agreement between simulated and observed values
is obtained for the majority of affected
cons t i tuents.
88
-------
If pK and the carbon cycle are simulated:
(32) Follou the same procedure which was outlined
for temperature in steps 6, 7, 8 and 9.
At the completion of the above steps, HSPF is calibrated to
the channel system being simulated under the conditions in
effect during the calibration period. Often times,
sufficient data will not be available to complete all steps
in the calibration process. For example, monthly and annual
values of sediment or pollutants will not be available for
comparison with simulated results. In these circumstances.
the user may omit the corresponding steps in calibration;
however, simulated values should be analyzed and evaluated
with respect to data from similar watersheds, personal
experience, and guidelines provided below.
7.2 Calibration Guidelines for Major Constituent Groups
The following discussion provides suggestions and guidelines
for calibrationg the major constituent groups modeled by
PERLND, IMPLND, and RCHRES. In many cases, the guidelines
are presented in terms of parameter categories rather than
using specific parameter names due to the large number of
parameters which must be considered. It should also be
noted that when specific parameter names are mentioned, the
names used are always those corresponding to the input of a
constant parameter value; the user should be aware that in
cases where monthly values are input for a particular
parameter, the variable names of concern for calibration may
be slightly different than those referred to in this
discussion.
Hvd rolog i c Calibration
Hydrologic simulation combines the physical characteristics
of the watershed geometry and the observed meteorologic data
series to produce the simulated hydrologic response. All
watersheds have similar hydrologic components, but they are
generally present in different combinations; thus different
hydrologic responses occur on individual watersheds. HSPF
simulates runoff from four components: surface runoff from
impervious areas directly connected to the channel network,
surface runoff from pervious areas, interflow from pervious
areas, and groundwater flow. Since the historic streamflou
is not divided into these four units, the relative
relationship among these components must be inferred from
the examination of many events over several years of
continuous simulation. Periods of record with a
predominance of one component (e.g., surface runoff during
storm periods, or groundwater flow after extended dry
89
-------
periods) can bo studied to evaluate the simulation of the
individual runoff components.
The first task in hydrologic calibration is to establish a
water balance on an annual basis. The balance specifies the
ultimate destination of incoming precipitation and is
indicated as
Precipitation - Actual Evapotranspiration
- Deep percolation -/Asoil Moisture Storage = Runoff
In addition to the input meteorologic data series> the
parameters that govern this balance are LZSN, INFILT, and
LZETP (evapotranspirat ion index parameter). Thus, if
precipitation is measured on the watershed and if deep
percolation to groundwater is small* actual
evapot ranspi ration must be adjusted to cause a. change in the
long-term runoff component of the water balance. LZSN and
INFILT have a major impact on percolation and are important
in obtaining an annual water balance. In addition, on
extrenely small watersheds (less than 100-200 hectares) that
contribute runoff only during and immediately following
storm events, the UZSN parameter can also affect annual
runoff volumes because of its impact on individual storm
events (described below).
Recommendations for obtaining an annual water balance are as
follows:
(1) Annual precipitation should be greater than or
equal to the sum of annual evaporation plus
annual runoff if groundwater recharge through
deep percolation is not significant in the
watershed. If this does not occur one should
consider using the parameter MFACT in the
NETWORK Block to adjust input precipitation so
that it is more representative of that
occurring on the watershed.
(2) Since the major portion of actual
evapotranspiration occurs from the lower soil
moisture zone, increasing LZSN will increase
actual evapotranspiration and decrease annual
runoff. Thus, LZSN is the major parameter for
deriving an annual water balance.
(3) Actual evapotranspiration is extremely
sensitive to LZETP. Since LZETP is evaluated
as the fraction of the watershed with deep
rooted vegetation, increasing LZETP will
increase actual evapotranspi rat ion and vice
versa. Thus, minor adjustments in LZETP may be
90
-------
used to effect changes in annual runoff if
actual evapotranspiration is a significant
hydrologic component of the watershed.
(4) The INFILT parameter can also assist in
deriving an annual water balance although its
main effect is to adjust the seasonal or
monthly runoff distribution described below.
Since INFILT governs the division of
precipitation into various components,
increasing INFILT will decrease surface runoff
and increase the transfer of water to lower
zone and groundwater. The resulting increase
in water in the lower zone will produce higher
actual evapotranspiration. Decreasing INFILT
will generally reduce actual evapotranspiration
and increase surface runoff. In watersheds
with no base flow component (from groundwater),
INFILT can be used in conjunction with LZSN to
establish the annual water balance.
When an annual water balance is obtained, the seasonal or
monthly distribution of runoff can be adjusted with use of
INFILT, the infiltration parameter. This seasonal
distribution is accomplished by dividing the incoming
moisture among surface runoff, interflow, upper zone soil
moisture storage, percolation to lower zone soil moisture
and groundwater storage. Of the various hydrologic
components, groundwater is often the easiest to identify.
In watersheds with a continuous base flow, or groundwater
component, increasing INFILT will reduce immediate surface
runoff (including interflow) and increase the groundwater
component. In this way, runoff is delayed and occurs later
in the season as an increased groundwater or base flow.
Decreasing IHFILT will produce the opposite result.
Although INFILT and LZSN control the volume of runoff from
groundwater, the AGWRC parameter controls the rate of
outflow from the groundwater storage.
In watersheds with no groundwater component, the DEEPFR
parameter is used to direct the groundwater contributions to
deep inactive grounduater storage that does not contribute
to runoff (DEEPFR = 1.0 in this case). For these
watersheds, runoff cannot be transferred from one season or
month to another, nnd the INFILT parameter is used in
conjunction with LZSN to obtain the annual and individual
monthly water balance.
In watersheds with continuous or intermittent baseflow,
groundwater outflow to the stream is usually the largest
component of the total streamflou. In these watersheds, the
DEEPFR parameter is used to estimate the fraction of total
91
-------
groundwater recharge that reaches deep aquifers that do not
discharge and contribute to baseflow at the watershed
outlet.
Continuous simulation is a prerequisite for correct modeling
of individual events. The initial conditions that influence
the magnitude and character of events are the result of
hydrologic processes occurring between events. Thus, the
choice of initial conditions for the first year of
simulation is an important consideration and can be
misleading if not properly selected. The initial values for
UZS, LZS, and AGWS should be chosen according to the
guidelines in Section 6 of the NFS User's Manual and
readjusted after the first calibration run. UZS, LZS, and
AGWS for the starting day of simulation should be reset
approximately to the values for the corresponding day in
subsequent years of simulation. Thus, if simulation begins
in October, the soil moisture conditions in subsequent
Octobers in the calibration period can usually be used as
likely initial conditions for the simulation. Meteorologic
conditions preceding each October should be examined to
insure that the assumption of similar soil moisture
conditions is realistic.
When annual and monthly runoff volumes are adequately
simulated, hydrographs for selected storm events can be
effectively altered with the UZSN and INTFW parameters to
better agree with observed values. Also, minor adjustments
to the INFILT parameter can be used to improve simulated
hydrographs; however, adjustments to INFILT should be
minimal to prevent disruption of the established annual and
monthly water balance. Parameter adjustment should be
concluded when changes do not produce an overall improvement
in the simulation. One event should not be matched at the
expense of other evenbs in the calibration period.
Recommended guidelines for adjustment of hydrograph shape
are as foilous:
1. The interflow parameter, INTFW, can be used
effectively to alter hydrograph shsipe after
storm runoff volumes have been correctly
adjusted. INTFW has a minimal effect on runoff
volumes. As shown in Figure 7.1 where the
values of INTFW were (a) 1.4, (b) 1.8, and (c)
1.0, increasing INTFW will reduce peak flows
and prolong recession of the hydrograph.
Decreasing INTFW has the opposite effect. On
large watersheds where storm events extend over
a number of days, the IRC parameter can be used
to adjust the recession of the interflow
portion of the hydrograph to further improve
the s imulat i on.
92
-------
2. The UZSN parameter also affects hydrograph
shape. Decreasing UZSN will generally increase
flows especially during the initial portions,
or rising limb, of the hydrograph. Low UZSN
values are indicative of highly responsive
watersheds where the surface runoff component
is dominant. Increasing UZSN will have the
opposite effect, and high UZSN values are
common on watersheds with significant
subsurface flow and interflow components.
Caution should be exercised when adjusting
hydrograph shape with the UZSN parameter to
insure that the overall water balance is not
significantly affected.
3. The INFILT parameter can be used for minor
adjustments to storm runoff
distribution. Its effects have
above. As with UZSN, changes
affect the water balance; thus,
should be minor.
volumes and
been discussed
to INFILT can
modi f ications
Ad justment
hyd rolog ic
of storm hydrographs is
calibration. If the
the final
ef fects of
step of
channel
0)
O)
O
CO
Time
Figure 7.1 Example of Response to the INTFW Parameter,
-------
attenuation on flows are important to the study results,
module section HYDR of RCHRES must be used to periorm
hydraulic routing. If such is the case, the following
guidelines are useful for finalizing hydraulic parameter
values.
Hydrauli c Gali brat ion
The major determinants of the routed flouts simulated by
section HYDR are the hydrology results from PERLND and/or
IMPLND and the physical data contained in the FTABLES
(Section 3.U). The FTABLES specify values for surface area,
reach volume, and discharge for a series of selected average
depths of water in each reach. This information is part of
the required User's Control Input for section HYDR and
consequently must be prepared prior to running the model.
Modification of these FTABLE values is essentially the only
means of calibrating the hydraulic results since the
additional parameters required for section HYDR are not
calibration parameters. If the routed flous simulated by
HYDR do not appear reasonable, the user should revieu the
assumptions and approximations on which the FTABLE values
were based. Particular attention should be given to the
fol1 owing i terns:
the approximations of channel geometry which
were used to develop the depth/volume
rel at i onshi p
the channel roughness coefficients selected for
normal depth calculations (if the reach is free
f1owi ng )
the interpretation and extrapolation of
existing stage/discharge data
For most model applications, calibration of the hydraulics
portion of the model is not a major task. If both the
hydrology results and the physical data provided in the
FTABLES are reasonable, little or no adjustment will be
necessa ry.
Snou Gali brat i on
Snow accumulation and melt can be a significant component of
streamflow from a watershed in many areas of the world.
Over one-half of the continental United States experiences
more than 60 cm of snowfall in an average year. For
mountainous watersheds at high elevations, spring snowmelt
may account for the major portion of annual streamflow.
Thus, accurate simulation of snow accumulation and melt
processes is needed to successfully model many watersheds.
-------
Snow calibration, using module section SNOM> is actually
part of the hydrologic calibration. It can be a major part
of the hydrologic calibration depending on the importance of
snonmelt runoff in the overall hydrologic balance. It is
usually performed during the initial phase of the hydrologic
calibration since the snow simulation can impact not only
winter runoff volumes, but also spring and early summer
st reamf1ou.
Simulation of snow accumulation and melt processes suffers
from two main sources of user-controlled uncertainty:
representative meteorologic input data and parameter
estimation. Uncertainties associated with deficiencies in
model algorithms, such as representation of frozen ground
conditions and effects, are beyond the control of the user
in normal applications. However, we recommend that all HSPF
users interested in snow simulation review the SNOW module
functional descriptions in the HSPF User's Manual and the
Iowa Basin studies (Donigian et al., 1983b; Imhoff et al.,
1983) in order to be aware of algorithm limitations and
assumptions.
The additional meteorologic time series data required for
snow simulation (i.e. air temperature, solar radiation,
wind, and dewpoint temperature) are not often available in
the immediate vicinity of the watershed, and consequently
must be estimated or extrapolated from the nearest available
weather station. Snowmelt simulation is especially
sensitive to the air temperature and solar radiation time
series since these are the major driving forces for the
energy balance melt calculations. Also, traditional
precipitation gages, even when equiped with windshields, can
underestimate snowfall amounts by 50 percent or more
depending on wind conditions (Linsley et al., 1975). This
type of error can have major impacts on the simulation.
Estimation of snow parameters is another possible source of
uncertainty due to less historical experience with snow
simulation than with general hydrologic modeling. Although
the energy-balance approach in module section SNOW is
somewhat more deterministic than the PWATER algorithms, a
degree of empiricism is still needed for many of the complex
processes of snow accumulation and melt. The data and
information sources noted in Section 6.2 should be reviewed
when estimating snow parameters and should be supplemented
with any other relevant information.
In many instances it is difficult to determine if problems
in the snow simulation are due to the non-representative
meteorologic data or inaccurate parameter values.
Consequently the accuracy expectations and general
objectives of snow calibration are not as rigorous as for
95
-------
the overall hydrologic calibration. Comparisons of
simulated weekly and monthly runoff volumes with observed
streamflow during snowmelt periods, and observed snow depth
(and water equivalent) values are the primary procedures
followed for snow calibration. Day-to-day variations and
comparisons on shorter intervals (i.e. 2-hour, 4-hour,
6-hour, etc.) are usually not as important as representing
the overall snowmelt volume and relative timing in the
observed weekly or bi-weekly period. In many applications
the primary goal of the snow simulation will be to
adequately represent the total volume and relative timing of
snowmelt to produce reasonable soil moisture conditions in
the spring and early summer so that subsequent rainfall
events can be accurately simulated. Obviously, if snowmelt
is a key component of the model application, such as
investigating flooding problems from spring snowmelt
conditions, more detailed calibration may be needed.
If observed snow depth (and water equivalent) measurements
are available, comparisons with simulated values should be
made. However, the user should be aware of the possible
tremendous variation in snow depth that can occur in a
watershed, and that the single observed value may not always
be representative of the watershed average.
Guidelines for adjusting snowmelt volumes are as follows:
1. Increasing the SNOWCF parameter should be
considered first if snowmelt volumes are
underestimated. Maximum SNOWCF values in the
range of 1.5 to 1.8 may be appropriate to
account for catch deficiency of the gage.
2. If snowmelt volumes are oversimulated there may
be problems with the precipitation gage
adequately representing the land segment. As
discussed in the hydrologic calibration
(above), the MFACT parameter in the NETWORK
Block can be used to adjust the segment
precipitation.
3. Whether precipitation falls as rain or snow has
a major impact on resulting runoff volumes.
The TSKOW parameter controls this
determination. It can be increased if
observations consistently indicate that snow
occurred and the model assumed the
precipitation occurred as rainfall, and vice
versa. The Special Actions option in HSPF can
be used to adjust TSNOW for specific critical
events if necessary for a reasonable
s imulat i on.
96
-------
4. The MPACK parameter has some impact on runoff
volumes because low values indicate greater
areal coverage of the snoupack. Runoff volumes
increase as a function of the area covered by
snow.
5. The SNOEVP parameter has a relatively minor
effect on snoumelt volumes since snow
evaporation is usually a small component of the
snoupack water balance. However, unusual
conditions may require adjustments to SNOEVP if
snow evaporation is important.
Guidelines for adjusting snoumelt timing are discussed
belou:
1. If significant differences in the timing of
observed and simulated snoumelt runoff occur,
the user should first examine the meteorologic
time series for errors, inconsistencies, and
possible discrepancies betueen the ueather
station and uhat the watershed may have
experienced. Air temperature and solar
radiation are the most critical time series to
examine, although wind and deupoint
temperature, to a lesser extent, also affect
snoumelt timing. Constant adjustments to the
time series are made uith the MFACT parameter
of the NETWORK Block.
2. The rate at uhich melt processes occur directly
impact the snowmelt timing. Increasing the
rate will cause melt to occur earlier in the
season, and vice versa. Radiation melt can be
adjusted only by adjusting the solar radiation
time series as discussed above. Condensation-
convection melt can be adjusted either by
adjusting the air temperature and uind time
series or by the CCFACT parameter, uhich is a
direct multiplier of the condensation-
convection melt equation (see HSPF User's
Hanua1 ) .
3. If observed streamflou or snou def.'th
measurements indicate a relatively constant
melting of the snoupack, the MGMELT parameter
can be used to represent a constant daily melt
component. Usually small but non-zero values
are used for MGMELT unless specific watershed
or meteorologic conditions indicate otherwise.
97
-------
tt. Snowmelt timing in terms of measured runoff can
also be affected by the storage and subsequent
release of melt water from the snowpack.
Increasing the MWATER parameter will increase
the amount of melt water stored within the
snowpack with a subsequent delay in the
snowmelt reaching the watershed outlet or gage.
Unlike predecessor models, HSPF allows the user to run the
SNOW module sections independently of the other PERLND
modules. In this way, the snow calibration runs can be
performed efficiently and cost-effectively on an individual
basis prior to executing complete hydrologic calibration
runs .
Sed iment Eros i on Cali brat i on
As indicated in the description of the general calibration
process, sediment calibration follows the hydrologic
calibration and must precede water quality calibration.
Calibration of the parameters involved in simulation of
watershed sediment erosion is more uncertain than hydrologic
calibration due to less experience with sediment simulation
in different regions of the country. The process is
analogous; the major sediment parameters are modified to
increase agreement between simulated and recorded monthly
sediment loss and storm event sediment removal. However,
observed monthly sediment loss is often not available* and
the sediment calibration parameters are not as distinctly
separated between those that affect monthly sediment and
those that control storm sediment loss.
In general, sediment calibration involves the development of
an approximate equilibrium or balance between the
accumulation and generation of sediment particles on one
hand and the washoff or transport of sediment on the other
hand. Thus, the accumulated sediment on the land surface
should not be continually increasing or decreasing
throughout the calibration period. Extended dry periods
will produce increases in surface pollutants, and extended
wet periods will produce decreases. However, the overall
trend should be relatively stable. This equilibrium must be
developed on both pervious and impervious surfaces, and must
exist in conjunction with the accurate simulation of monthly
and storm event sediment loss. To assist in sediment
calibration, the following guidelines are provided.
1. On pervious areas, KRER, and NVSI are the major
parameters that control the availability of
sediment on the land surface, while KSER and
JSER control the sediment washoff. The daily
98
-------
accumulation or removal of sediments by NVSI
will dominate sediment availability for land
surfaces with high cover factors (COVER). On
exposed land surfaces* sediment generation by
soil splash is important and is controlled
largely by the KRER parameter. To offset the
sediment availability on pervious areas, the
KSER and JSER parameters control sediment
uashoff to prevent continually increasing or
decreasing sediment on the land surface. Thus,
balance must be established between the KRER,
and NVSI parameters and the KSER and JSER
parameters to develop the equilibrium described
above.
2. On impervious areas, soil splash is not
significant. The major sediment accumulation
and removal parameters are ACCSDP and REMSDP
and the sediment uashoff parameters are KEIM
and JEIM. These tuo parameter sets must be
adjusted to maintain a relatively stable amount
of sediment on impervious surfaces throughout
the calibration period.
3. The output for PERLMD and IMPLND indicates the
flow and sediment contributions from pervious
and impervious surfaces in each land use
simulated. In urban areas, the majority of
nonpoint pollutants will emanate from
impervious land surfaces especially during
small storm events and in the early portion of
extended events. Pervious land surfaces in
urban areas will generally contribute a
significant amount of pollutants only during
large storm events and the latter portion of
extended events. The user should note this
behavior from the output provided during
cali brat i on runs.
**. The output also indicates the accumulated
sediment on pervious and impervious surfaces in
each land use. This information is provided to
assist in the development of the sediment
balance.
5. The daily removal factor, REMSDP, is usually
assumed to be relatively constant and fixed.
Also, the exponents of soil splash (JRER) and
sediment washoff (JSER,JEIM) are reasonably
well defined. Thus, the parameters that
99
-------
receive major consideration during sediment
calibration are: the accumulation rates, NVSI
and ACCSDP; the coefficient of soil splash,
KRER (especially for exposed land surfaces);
and the coefficients of sediment uashoff, KSER
and KEIJ1.
6. In general, an increasing sediment storage
throughout the calibration period indicates
that either accumulation and soil fines
generation is too high, or sediment uashoff is
too low. Examination of individual events will
confirm whether or not sediment uashoff is
under-simulated. Also, the relative
contributions of pervious and impervious
surfaces will help to determine whether the
pervious or impervious washoff parameters
should be modified. A continually decreasing
sediment storage can be analyzed in an
analogous manner.
7. The sediment washoff during each simulation
interval is equal to the smaller of two values;
the transport capacity of overland flow or the
sediment available for transport from pervious
or impervious surfaces in ench land use. To
indicate which condition is occurring, the user
should output values for STCAP, the sediment
transport capacity by surface runoff* using the
DISPLY function of HSPF. These values can then
be compared with the washoff values reported in
the output for section SEDHNT of PERLND (DISPLY
cannot currently output transport capacities
for impervious land surfaces.) Generally,
washoff will be at capacity during the
beginning intervals of a significant storm
event; this simulates the "first flush" effect
observed in many nonpoint pollution studies.
As the surface sediment storage is reduced,
washoff will be limited by the sediment storage
during the latter part of storm events.
However, for very small events overland flow
will be quite small and washoff can occur at
capacity throughout. Also, on agricultural and
construction areas washoff will likely occur at
capacity for an extended period of time due to
the large amount of sediment available for
transport.
8. Using the information provided by displaying
the values for STCAP, minor adjustments in
JRER, JSER, and JEIM can be used to alter the
100
-------
shape of the sediment graph for storm events.
For pervious areas when available sediment is
limiting, increasing JRER will tend to increase
peak values and decrease low values in the
sediment graph. Decreasing JRER will have the
opposite effect tending to decrease the
variability of simulated values. When sediment
is not limiting, the JSER parameter Mill
produce the same effect. Increasing JSER will
increase variability 'while decreasing it will
decrease variability.
For impervious areas, the JEIM parameter will
produce the effects described above when
sediment uashoff from impervious areas is
occurring at the transport capacity. All these
parameters will also influence the overall
sediment balance, but if parameter adjustments
are minor, the impact should not be
significant.
9. HSPF includes algorithms to represent scouring
of the soil matrix as an additional component
of the total sediment erosion. Since this
process uas not included in the ARM and NFS
models, there is little experience upon which
to base parameter values. The relevant
parameters are KGER (coefficient) and JGER
(exponent); the mathematical formulation is a
power function of overland flow, identical to
the transport capacity equation, but it is not
limited by available particles since it is
scouring the soil matrix. The parameters are
analogous to those discussed above, and the
scouring algorithm can be employed to increase
sediment erosion on watersheds where scouring
and gully formation is evident.
Sediment calibration should be performed on a single land
use at a time, if possible, in order to correctly evaluate
contributions from individual land uses.
Sed iment Transport Cali brati on
While land surface sediment erosion is simulated in terms of
total sediment, instream sediment transport (using section
SEDTRN of RCHRES) is calculated based on the three component
fractions of sediment (sand, silt, and clay). There are no
calibration parameters involved in simulation of sand
transport by the Colby or Toffaleti methods. If, however,
sand transport is modeled as a power function of stream
velocity (SANDFG=3), the user can control the process to a
101
-------
certain extent by adjusting the values for the coefficient
(KSAND) and exponent (EXPSND) of the transport equation.
The successful simulation of cohesive sediments (silt and
clay) is much more dependent on calibration. The three
parameters used for calibration are the critical shear
stress for deposition (TAUCD); critical shear stress for
erosion (TAUCS); and the rate of erosion* or erodibility
coefficient CM). Successful calibration of the instream
sediment transport processes for cohesive sediments requires
the following five steps:
1. Using the hydraulic calibration, identify a
period of record which contains events which
have a good fit between recorded and simulated
flows. Sediment transport processes., and the
sediment calibration must be based on an
accurate hydraulic representation in order for
the values derived for TAUCD, TAUCS, and M to
be meaningful. The calibration period must
contain significant runoff events in order to
properly define the runoff/sediment washoff
relationship at higher flows.
2. Use the HSPF DISPLAY function to output daily
values for calculated shear stress, TAU.
Identify the range of values for TAU which are
characteristic of periods which exhibit
significant suspension of sediment in the
historical data.
3. Set values for the critical shear stress for
erosion of silt and clay which bracket the
period of increased suspended load. Proper
selection of values for TAUCS should result in
scour and suspension of cohesive materials
during periods of increased flow and shear
stress, but no erosion during periods when the
historical record shows minimal suspended
sed iment.
4. By examining calculated values for TAU during
low flow and less turbulent portions of the
simulation record, select values for TAUCD for
silt and clay which allow deposition only
during appropriate periods.
5. Adjust the erodibility coefficient, M, to
obtain the best overall correspondence between
observed and simulated sediment loads for
events with good hydraulic fit.
102
-------
Sediment transport processes are strongly linked to
hydraulic processes* and a good hydraulic calibration is a
necessity for a good sediment simulation. In order to
perform a meaningful instream sediment calibration, the
erosion must also be reasonably accurate. In essence, the
instream calibration is merely an adjustment of bed
sediment, by deposition or scour, to make up the difference
between edge-of-stream loadings and observed loadings at a
point downstream.
PERLND Mater Qua!i ty Gali brat ion
Pisspived Gases. Calibration of dissolved gases simulation
by PERLND (section PWTGAS) is limited to a feu relatively
simple adjustments.
1. Estimate all dissolved gas parameters and
storages from the literature and all available
information on the study area.
2. Depending on whether or not soil temperature is
simulated, adjust soil temperature simulation
results or input time series data to modify gas
saturation values calculated for surface
runoff .
3. If gas concentrations (or mass loadings) from
the combined outflow from surface runoff,
interflow and groundwater are not reasonable,
adjust user-specified gas concentrations for
interflow and groundwater until acceptable
results are obtained.
General Qua 1 i t y Cons t i tuents. Calibration procedures for
simulation of general quality constituents or pollutants
(using section PQUAL) vary depending on whether constituents
are modeled as sediment-associated or flow-associated.
Calibration of sediment-associated pollutants begins after a
satisfactory calibration of sediment uashoff has been
completed. At this point adjustments in the pollutant
potency factors (POTFW and POTFS) can be performed.
Generally, monthly and annual pollutant loss will not be
available, so the potency factors will be adjusted by
comparing simulated and recorded pollutant concentrations,
or mass removal, for selected storm events. For nonpoint
pollution, mass removal in terms of pollutant mass per unit
time (e.g., gm/min) is often more indicative of the washoff
and scour mechanisms than instantaneous observed pollutant
concentrations. However, the available data will often
govern the type of comparison performed.
103
-------
Storms that are well simulated for both flow and sediment
should be used for calibrating the potency factors. The
initial values of potency factors should be increased if
pollutant graphs are uniformly low and decreased if the
graphs are uniformly high. Monthly variations in potency
factors can be used for finer adjustments of simulation in
different seasons if sufficient evidence and information is
available to indicate variations for the specific pollutant.
However, individual storms should not be closely matched at
the expense of the other storms in the season. Also,
consistency between the sediment and pollutant simulation is
important; if sediment is under-simulated then the pollutant
should be under-simulated, and vice versa. Inconsistent
simulations can indicate that sediment is not a transport
mechanism for the particular pollutant or that the potency
factors have been incorrectly applied. Also, if there is no
similarity between the shapes of the recorded sediment and
pollutant graphs, then pollutant transport is not directly
related to sediment transport and no amount of adjustment
will allow an effective simulation of that pollutant.
Calibration procedures for simulation of pollutants
associated with overland flow are focused on the adjustment
of three parameters: the pollutant accumulation rate
(ACQOP); the maximum pollutant storage on the land surface
(SQOLIM); and the parameter which relates runoff intensity
to pollutant uashoff (WSQOP). As was the case for sediment-
associated constituents, calibration is performed by
comparing simulated and recorded pollutant concentrations,
or mass removal, for selected storm events. In making this
comparison, the following issues should be considered:
1. If too much pollutant washoff is simulated for
all storms, the value used for maximum storage
(SQOLIM) is probably too high. Likewise,
consistently low simulations of pollutant
washoff indicate the value used for SQOLIM is
too 1ow.
2. If too much washoff is simulated for small
storms, but not for large storms, the value
assigned for the washoff rate parameter (WSQOP)
may be too low.
3. If simulation results for storms following long
periods without rain are good, but too much
washoff is simulated for storms which occur in
close sequence to earlier storms, the value
used for the accumulation rate parameter
(ACQOP) is probably too high and should be
adjusted accordingly. Of course, the opposite
is true if simulated values are low for storms
of this type.
104
-------
In most cases, proper adjustment of SQOLIM, WSQOP, and ACQOP
allows a good representation of the uashoff of flow-
associated constituents. In study areas where pollutant
movement is also associated with subsurface flows, the user
may assign pollutant concentration values for both interflow
and active grounduater. If this option is exercised, one
should pay careful attention to the influence which these
pollutant sources are having on simulated net pollutant
outflows, particularly if observed instream pollutant data
are being considered in the calibration process.
Pesticide Galibration. Ideally pesticide simulation should
require little, if any, calibration since all the pesticide
parameters represent characteristics that can be determined
in laboratory experiments. However, inaccuracies in the
pesticide algorithms, discrepancies between laboratory and
field conditions, variability in measured laboratory values,
or lack of pertinent laboratory values will usually require
some adjustment or calibration of initial parameter values.
Calibration should be done by comparing simulated values
with measured field data. If no field data are available,
data from watersheds under similar conditions and personal
experience should be used to evaluate the simulated values.
The intent of pesticide calibration is to: (1) obtain the
correct time distribution of the amount of pesticide in the
soil following application by adjustment of the degradation
parameters; (2) obtain the correct vertical distribution of
pesticides in the various soil layers by adjusting the
leaching factors; and (3) obtain the correct partitioning
between solution and sediment-associated pesticide by
adjusting the adsorption/desorption parameters. With this
procedure in mind, the following steps and guidelines for
pesticide calibration are recommended.
1. Estimate all pesticide and solute leaching
parameters from the literature and all
available information on the field site.
2. Adjust pesticide decay rates (primarily in
surface and upper soil zones) to better reflect
the observed soil core data.
3. Adjust solute leaching parameters (primarily
surface and upper zone values) to better
reflect the pesticide distribution between the
surface and upper zones, as determined from the
soil core data or calibration with a
nonreactive tracer (e.g., chloride).
4. Adjust adsorption/desorption parameters as
needed to obtain the proper distribution
between solution and adsorbed forms.
105
-------
5. Compare storm event pesticide losses in
solution and adsorbed forms with observed data
and make further parameter adjustments as
discussed above.
Nut ri ent Gali brat ion. Nutrient calibration begins with
analysis and comparison of soil storages with observed soil
nutrient data. Soil nutrient data obtained from sampling
throughout the watershed for the period of calibration
provides valuable information for the calibration of the
nutrient parameters. If no soil nutrient data are
available, calibration consists of merely estimating
reasonable nutrient storages and comparing the recorded and
simulated nutrient runoff results. However, all the
simulation results (storages and runoff) should be evaluated
for reasonableness based on personal experience and data
from similar watersheds.
With or without observed data, the order of calibration is
the same and is analogous to the pesticide calibration
procedures.
Nutrient calibration involves the establishment of
reasonable soil nutrient storages through adjustment of
percolation parameters, plant uptake parameters, and
reaction rates, followed by evaluation of nutrient runoff
and refinement of pertinent parameters. The recommended
order and steps in the procedure are:
1. Evaluate initial soil nutrient parameters from
information available in the literature, and
include fertiliser and rainfall sources of
nutrients as input to the model.
2. Calibrate initial mineralization rates so that
annual amounts of plant-available nutrients
correspond to expected values.
3. Adjust leaching factors based on any data
available for a tracer such as chloride.
4. Adjust plant uptake rates to develop the
expected nutrient uptake distribution during
the growing season and the estimated total
uptake amount expected for the crop.
5. Adjust nutrient partition coefficients based on
available core and runoff data.
6. Refine the leaching, uptake, and partition
parameters based on observed runoff data and
the expected sources of nutrient runoff, i.e.,
surface, interflow, groundwater.
106
-------
As with pesticide calibration, some iteration of the steps
is often required. Parameter values may need to be
readjusted as later steps affect prior adjustments, but the
order designated should help to minimize the number of
iterations in the calibration procedure.
IMPLND Hater Quality Calibration.
Procedures for calibrating the simulation of dissolved gases
and general water quality constituents using the IMPLND
module are the same as those outlined for calibrating land
surface processes in PERLND; however, subsurface processes
are not considered in IMPLND and hence are not a factor in
calibration. (Refer to calibration guidelines for PWTGAS
and PQUAL for assistance.)
RCHRES Hater Quali tv Gali brat ion
Hater Temperature. Given the strong influence that water
temperature has on biological and chemical reaction rates,
it is important to obtain the most reasonable values for
water temperature possible. If available meteorologic data
and observed instream temperature data are adequate to
perform temperature simulation and calibration, the modeler
should use adjustments to four parameters: CFSAEX, KATRAD,
KCOND, and KEVAP as a basis for calibration:
1. CFSAEX is the ratio of shortwave radiation
incident to a reach to radiation incident at
the recording station. If heavy vegetation or
irregular topography shades a reach for all or
part of the day, the value of this parameter
can be lowered accordingly. Since shortwave
radiation is the largest source of heat to the
reach, adjustment of the value for CFSAEX is
the most effective of all four water
temperature calibration parameters.
2. The values for the other three parameters are
physically based, and the default values for
all three should be used for the first
calibration run.
3. An increase in the value of the atmospheric
longwave radiation coefficient (KATRAD) will
tend to increase water temperature.
4. An increase in the value of the conductive -
convective heat transport coefficient (KCOND)
will increase heat transfer between water and
the atmosphere. Consequently, simulated water
temperature may either increase or decrease
107
-------
depending on the relative temperatures of water
and air.
5. Increasing the value of the evaporation
coefficient (KEVAP) uill tend to decrease
simulated water temperature.
The user should note that in situations where point loadings
contribute a significant volume of water to the reach
system, the water temperature values assigned to the point
loading may become the dominant factor in water temperature
simulation. If reasonable adjustments to the four
calibration parameters cannot produce an acceptable
calibration, input data for point loads or meteorology are
most likely unrepresentative of the study reaches and should
be re-examined.
General finality Cons tit u en t s (GQUAL). The specific
procedures used to calibrate the simulation of a generalized
quality constituent, or GQUAL, depend on the relative
adsorption characteristics of the compound and the
availability of laboratory data to characterize the various
decay processes (i.e., hydrolysis, oxidation, photolysis,
volatilization, biodegradation) which can be modeled. The
key parameters are the part i t i on coe_f i ic i enis , the process-
specific or lumped
-------
If field data (i.e., instream and/or bed chemical
concentrations) indicate that the laboratory values are not
appropriate, the estimated partition coefficients can be
adjusted accordingly. In making adjustments, one should
remember that the simulation of sediment-associated
constituents is heavily influenced by sediment simulation
and that adjustment of the partition coefficient values
should not be used as a means of correcting deficiencies
introduced by an inaccurate sediment simulation. Decay
rates are specified separately for the soluble component,
the adsorbed component on suspended sediments, and the
adsorbed component on the bed sediments. The process-
specific rates are available only for the soluble component;
the adsorbed components use a single lumped decay rate for
each size fraction (i.e., sand, silt, clay). General 1y," the
same decay rate is used for all size fractions, unless data
indicates otherwise, but different rates are expected for
the suspended and bed sediments.
For most constituents which are modeled with mcTdule section
GQUAL, detailed laboratory data needed to evaluate
parameters for specific degradation processes are not
generally available. Even if relevant data exists, large
variations in degradation rates can occur in the field. For
this reason, it is a common practice currently to lump the
effects of all forms of degradation into a general decay
parameter (Section 4 . <4 ( 3 ) . 7 . 1 1 , Part F of the User's Manual)
and treat it as a calibration parameter.
Current efforts to develop laboratory protocols for
measuring process rate parameters and prepare data bases for
contemporary compounds should help to provide a better basis
for estimating process-specific rate parameters in the
future. Since environmental conditions such as water
temperature, pH, cloud cover, and others, affect the rate at
which components of the total degradation occur, estimation
of a general degradation rate is always somewhat inaccurate
and adjustment through calibration may be justified, if
possible. In any case, the user should be cognizant of the
primary decay mechanisms of the compound so that the impact
of including or excluding effects of environmental
conditions can be assessed.
The adsorption/desorption transfer rate parameters represent
the rate at which the system approaches equilibrium
adsorption conditions between the soluble and suspended and
bed sediments. This concept was included to allow either an
equilibrium or kinetic approach to adsorption since
equilibrium partitioning conditions are not often achieved
instantaneously in natural aquatic systems. Very little
information is available on which to evaluate these rate
parameters. Sensitivity studies conducted as part of our
109
-------
Four Mile Creek application indicate the following:
a. Large partition coefficients and sediment
concentrations increase the effect of the
transfer rate on the total chemical load
because a greater fraction of the load is
transported with the sediment.
b. If the majority of the chemical load is in
solution, the primary impact of the transfer
rate is to control the amount o:E chemical
adsorbed to the bed and subsequently released
to the water column in the time period
following peak concentrations. This was
observed in Four Mile Creek by measurable
pesticide concentrations for several days
following a storm event.
c. Equilibrium conditions can be approximated
(i.e., ins tanteous1y in each time interval) by
setting the transfer rate equal to three times
the number of simulation time intervals in a
day. Thus, with an hourly time step, a
transfer rate of 72 (3x24) per day would
achieve 95X of equilibrium conditions within
one time interval. Alternatively, a value of
21 per day (assuming an hourly time step) would
achieve 95% of equilibrium within three time
steps, since first-order kinetics are assumed;
this is sufficiently fast as to practically
represent equilibrium adsorption conditions in
most aquatic systems.
d. In our Four Mile Creek study, vie derived
through calibration transfer rates of 8.0 and
0.03 for the suspended and bed sediments,
respectively. Logically, the rate for the
suspended sediments in the water column should
be substantially greater than the bed transfer
rate due to instream mixing and turbulence.
The bed transfer rate also depends, to some
extent, on the assumed bed depth and associated
sediment mass available to adsorb chemicals; a
one-foot depth was assumed in our Four Mile
Creek study.
Detailed Simulat i on pi Selected Constituents Involved in
Bi ochemical Transformations (RQUAL). As the generalized
calibration procedures outlined in Section 7.1 indicate, the
calibration of RQUAL can be quite complicated and time-
110
-------
consuming, depending on the number of constituents and
processes which are simulated. In fact, adjustment of RQUAL
simulation results to more closely duplicate observed values
is not always achieved solely by calibration. In some
cases, simulation of additional constituents and/or
processes may allow improvements to simulation results which
cannot be obtained by adjustment of parameter values. For
example, simulation of plankton may be necessary in order to
duplicate observed seasonal fluctuations in nutrient
concentrations, or volatilization may need to be modeled in
order to reproduce the observed nitrogen mass balance for a
lake. Thus, while the user is allowed to model nutrients
without consideration of plankton and/or volatilization, it
may not be possible to obtain a good fit between simulated
and observed nutrient values in cases where these factors
are important but are not modeled. Module sections GQUAL
and RQUAL contain many user options for simulating or not
simulating various constituents and processes. Simulation
results are equally dependent on the simulation of all
important constituents/processes and on development of
realistic parameter values.
Calibration of RQUAL is complicated by two factors. First,
the interrelationships of the various constituents result in
changes in simulated concentrations for numerous
constituents by adjustment of a parameter value specific to
only one constituent. For example, if one increases the
value for the algal respiration rate parameter in order to
reduce simulated plankton populations, the modification will
also result in increased values for nutrients and inorganic
carbon and a decreased value for dissolved oxygen. Thus,
the final calibration of any one constituent in RQUAL cannot
be completed until all adjustments have been made to
associated constituents. The calibration of RQUAL is
complete when the best overall fit to data is achieved for
all constituents which are simulated.
The second factor which complicates the calibration of RQUAL
is the wide range of values which have been reported for the
model parameters. The variability of literature values for
many of these parameters results from the complexity of the
physical, chemical, and biological factors which influence
the ultimate biochemistry of each individual stream or lake.
Quite often it is difficult for the model user to know
whether or not the values assigned to calibration parameters
are reasonable for the study area, even if the values do
result in a good simulation.
Given the potential complexity of RQUAL simulation, as well
as the flexibility allowed in constituents/processes
simulated, it is not possible to describe a detailed
calibration procedure. Nonetheless, the parameters
111
-------
identified below are generally
useful for calibration of
considered in RQUAL:
cons idered to
the various
be the most
constituents
oxyg en
BOD
nut ri ents
algae
zooplankton
pH/carbon
BENOD
BRBOD
KBOD20
BRCON(I)
KNH320
KN0220
DEBAC
CFSAEX
LITSED
EXTB
MARGR
ALR20
ALDH
ALDL
MZOEAT
ZFIL20
ZRES20
ZD
BRC02
benthal oxygen demand rate
benthal release rate for BOD
decay rate of BOD
benthal release rates for nit-
rate and orthophosphorus
oxidation rate of ammonia
oxidation rate of nitrite
fraction of denitrifying bacterii
correction factor for surface
area exposed to sunlight
light extinction factor to
account for suspended sediment
base extinction coefficient for
light
maximal unit algal growth rate
algal unit respiration rate
high algal death rate
low algal death rate
maximum zooplankton unit
ingestion rate?
fillering rate
unit respiration
zooplankton
zooplankton
rate
zooplankton
unit death rate
benthal release rate for C02
7.3 How Much Calibration?
A common question that is asked by model us
extent of calibration or parameter adjus
before one can say that the model is "cal
test watershed. Obviously this depends to
how well the initial parameter values are
beyond that, the question is really "How
simulated and recorded values be before cal
terminated?" The answer to this questio
number of factors including the extent and
the available data, the problems analyzed
capabilites, and the allowable time
calibrat ion.
ers concerns the
tment necessary
ibrated" to the
some extent on
estimated. But
close should the
i brat i on can be
n depends on a
reliability of
vs . the model
and costs for
Data Problems. The available data are often the most severe
limitation on calibration especially for water quality
variables. A common mistake by model users is to accept the
observed data as being absolutely accurate. In fact, any
measurement obtained under field or natural conditions will
usually contain at least a 5 to 10 percent variation from
112
-------
the actual or true value. Moreover, instantaneous or short
time interval measurements commonly shou variations of 10 to
20 percent and greater for flow or concentration values.
Usually annual volumes and total loss measurements are the
most accurate except when a persistent bias exists in the
measurement technique or calculation method.
The assumption of uniform areal precipitation is a major
source of error with direct effects on the simulation since
precipitation is the driving force of HSPF. Precipitation
is rarely uniform and is highly nonuniform in thunderstorm
prone regions of the country. This nonuniformity makes
simulation of thunderstorms difficult since the actual
rainfall is unknown if the recording gage does not
adequately represent- the rainfall pattern.
The user should be aware of the measurement techniques and
the resulting confidence limits of the observed values for
both the input meteorologic data and the runoff or soil
calibration data. Simulated values uithin the confidence
limits of the observed calibration data cannot be improved
upon; this signals a reasonable end to calibration.
However, this is not an absolute criterion since a good
overall calibration can include simulated individual storm
events or instantaneous values uith larger variations than
the accepted confidence limits. In such cases, analysis of
the discrepancies and personal judgment must be called upon
to decide if calibration is sufficient.
Probi ems Analyzed vs. Mode1 Capabilities. Another source of
frustration in model calibration is the attempt to calibrate
a model for conditions or processes that the model cannot
adequately represent. For example, at present HSPF cannot
fully represent the effects of specific tillage operations
on runoff and soil moisture. While the Special Actions
Block can be used to approximate changes in soil properties
related to tillage, additional research is needed to
determine hou these changes can be simulated
deterministica11y. Runoff for storms occurring soon after a
tillage operation may not be uell simulated/ but this effect
decreases uith the time since tillage. Calibration of
parameters to better simulate such events will produce a
biased set of hydrologic parameters, and subsequent
simulation results will not be realistic.
To avoid such problems, the user should have a basic
understanding of the processes that are occurring on the
watershed, the processes simulated by HSPF, and their method
of representation in the model. Study of HSPF algorithms
provides an additional benefit since the user will acquire a
better understanding of the role of model parameters and the
impact of parameter adjustments. Calibration can be
113
-------
expedited with this knowledge, and with the realization that
certain processes affecting the observed data are not
represented in the model. Parameter adjustments to
circumvent such model limitations are both inappropriate and
futile.
Guidelines. In many applications of HSPF, the time and
costs budgeted to calibration will determine the level of
effort expended. Calibration is a critical step in any
model application and may require 30 to 50 percent of the
total project resources. Its importance cannot be
understated. The arguments provided above should not be
used to justify reducing the time and costs required for a
reasonable calibration. However, our experience has shown
that many diligent users will often spend too much time on
calibration due to insufficient observed data, ignorance of
the accuracy of the data, and misconceptions of model
capabilities and parameter sensitivities.
The agreement between simulated and recorded values required
for an adequate calibration is highly dependent on the
specific watershed, data conditions, and problems analyzed.
Very little quantitative information exists to provide
guidelines for evaluating a calibration. However, from our
experience in applying HSPF and related models and within <
the framework of the considerations discussed above, the
following general guidelines for characterizing a
calibration are provided to assist potential model users:
Difference Between Simulated and Recorded Values (percent)
Calibration Resu11s
Very G o o d Good Fair
Hydrology/Hydraulics <10 10-15 15-25
Sediment <15 15-25 25-35
Mater Quality <20 20-30 30-40
The above percent variations largely apply to annual and
monthly values. Individual events may shou considerably
larger variation for many reasons with little impact on the
overall calibration. These values should be used only as
approximate guidelines. The user should attempt to obtain
the best calibration possible within the limitations of the
available data, the model capabilities, and the allowable
budget.
7 . 4 Veri f icat ion
Model verification is in reality an extension of the
calibration process. Its purpose is to assure that the
-------
calibrated model properly assesses all the variables and
conditions uhich can affect model results. While there are
several approaches to verifying a model, perhaps the most
effective procedure is to use only a portion of the
available record of observed values for calibration; once
the final parameter values are developed through
calibration, simulation is performed for the remaining
period of observed values and goodness-of-fit between
recorded and simulated values is reassessed. This type of
split-sample calibration/verification is highly recommended.
However, in data-poor situations there is a real question as
to whether to calibrate on half the data and verify on the
other half, or obtain the best calibration on all the
observed data. In any case, credibility is based on the
ability of a single set of parameters to represent the
entire range of observed data. Overall model credibility
can be enhanced if the model is applied by independent
users, in a variety of watersheds, and for a range of events
with different magnitudes. If a single parameter set can
reasonably represent a wide range of events, then this is a
form of verification.
Quantitative measures of verification are needed and model
reports should always include comparison of simulated and
observed data. This should be done for runoff volumes,
pollutant loads, hydrographs and pollutographs.
Correlations of point-to-point comparisons may not be valid,
due to time shifts. For nonpoint source pollution, mass
loads are usually more appropriate for comparison than
concentrations .
115
-------
SECTION 8
ANALYSIS OF ALTERNATIVE CONDITIONS
The analysis of proposed or projected alternative conditions
for a watershed or water system will be the most critical
step in many HSPF applications. The results of this step
will often provide direct input to the decision-making
process by supplying the necessary system response
information to evaluate and compare alternatives.
Unfortunately, coming at the end of the model application
process, this analysis step is often plagued by short time
schedules, inadequate resources, and insufficient
data/information for an indepth investigation. The model
user must be aware of these potential pitfalls in order to
preserve sufficient project resources for this final task of
analyzing proposed alternatives. In effect, the ultimate
utility of the HSPF application will often depend on the
successful completion of this analysis, as measured by the
ability of the model to represent alternative conditions and
provide sufficient data for a valid comparison.
Because of the comprehensive scope of HSPF, once it has been
applied (i.e., calibrated/verified) to a watershed system it
can be subsequently used to analyze a variety of
alternatives and associated impacts. Water projects related
to flood control, storm drainage, urban and agricultural
best management practices, water supply, hydropouer,
municipal and industrial waste treatment, etc. can be
analyzed within a comprehensive watershed management
approach. This section discusses the basic philosophy
underlying the use of HSPF for analysis of alternatives,
enumerates the various steps involved in this process,
provides guidance in analyzing selected alternatives, and
describes related examples drawn from past experience with
HSPF and/or predecessor models.
8.1 Philosophy Underlying Comparison of Alternatives
The philosophy underlying the use of HSPF for analyzing
various alternatives is a basic component of the concepts
and assumptions of the continuous simulation approach. The
calibrated/verified model is used as a tool to project
116
-------
changes in system response resulting from a proposed
alternative; this alternative is represented in HSPF by
adjustments (changes) to model input, parameters, and/or
system representation (e.g., interconnection of PLSs and
stream reaches). During the calibration/verification steps,
the model results are compared with observed data for
selected time periods; whereas, in the analysis of
alternatives the model results for a specific alternative
are compared to model results produced by appropriate base
conditions. In this way the relative changes in system
response associated with a proposed alternative can be
identified and analyzed.
Two key aspects of analyzing alternatives involve the
methods and procedures for characterizing both the system
(base condition and alternatives) and the system response.
A common misconception of potential users of continuous
simulation models is that the model is designed to duplicate
observed data on the watershed (i.e., system) for the
extended simulation periods of 10 years or ^ more. In
reality, the observed data reflects dvnami c changes
occurring on the watershed such as land use changes, channel
modifications, water use patterns, etc., whereas the model
describes what would have been observed under static
(constant) watershed conditions. For this reason
calibration and verification time periods are specifically
chosen to be 1o n a enough to cover a range of
hydrometeorologic conditions (to satisfy calibration/
verification needs), but short enough to limit physical
changes that could significantly impact the system response
(to satisfy the static conditions assumption).
In effect, a Monte Carlo type approach is employed where the
input meteorologic data is the driving function used to
generate a corresponding output time series under constant
watershed conditions; the output time series is then
analyzed to characterize the watershed response under the
defined conditions. This characterization can be based on a
variety of numeric measures, such as mean, maximum, and/or
minimum values of flow, reservoir volumes, pollutant
concentrations, and/or loads for monthly, seasonal or annual
periods.
Alternately, a frequency-duration analysis can be performed
for any output time series to determine the 'percent of
t ime' that hourly, multi-hour1y, or daily values exceed (or
are less than) specific target values. Frequency analysis
is generally preferred since it provides a more rigorous
characterization of the system response over the entire
range of dynamic watershed conditions. Moreover, frequency
information provides a means of assessing flood damages,
water quality impacts, fish toxicity conditions, etc.
associated with extreme values of flow and pollutant
concentrations.
117
-------
As described in Section 4, HSPF can provide all the numeric
and statistical measures noted above. A consistent set of
measures must be chosen and generated for both the base
condition and each alternative in order to provide a valid
basis for comparison.
8.2 Steps in the Analysis Process
Prior to analysis of alternatives, the calibration/
verification process must proceed to the state where model
results are sufficient to demonstrate that the model
provides a realistic and credible representation of the
system response. At this point, the proposed alternatives
can be analyzed by the following procedures:
1. Define appropriate bgse conditions to which
alternatives will be compared. This may be the
calibrated condition* or some modification of
it.
2. Define the simulation time period, output time
series, and numeric/statistical measures to be
used to characterize and compare the base
condition with proposed alternatives.
3. Simulate base conditions for the simulation
period, and generate the selected time series
and numeric/statistical measures.
4. Define alternatives to be analyzed. Each
alternative should provide a mean i ng ful and
realistic difference from the base condition.
5. Define and incorporate all effects of the
proposed alternative on model parameters,
inputs, and/or system representation.
6. Perform simulation runs for each proposed
alternative for the identical time period as
the base condition, and generate identical time
series and numeric/statistical measures. Make
sure that t he only d i f ferences between the base
and alternative runs are d ue to the a 11ernative
being ana 1vzed.
7. Compare model output and numeric/statistical
measures of the base and alternative model
runs. The model user should be able to explain
and justify the differences; if the differences
are counter-intuitive, check parameters and
model output for possible errors.
118
-------
Although each of the above steps are important, it is clear
that the critical step in the analysis is Step # 5, defining
the effects of the proposed alternative in terms of speci fie
changes in model inputs, parameters, and/or system
representation (e.g., interconnection of PLSs and stream
reaches). Due to the wide range of alternatives that can be
analyzed with HSPF, the process of determining required
changes is best shown by example.
8.3 Examples of Analyzing Alternatives with HSPF
Table 8.1 presents a summary of how various water project
alternatives can be represented with HSPF and lists the
associated changes in the input sequence. As noted above,
simulation of alternatives will require adjustments to model
input, parameters, and/or system representation. Generally,
changes to model input and system representation will be the
easiest to specify and provide the greatest reliability in
the resulting simulation. For example, model input changes
will include modifications to point load, flow, and/or
rainfall files in the TSS to represent alternatives such as
municipal/industrial waste treatment levels, instream
aeration, flow augmentation, rainfall augmentation,
wasteload allocation, etc. System representation can be
changed to analyze land use changes, reservoir operations,
reservoir site alternatives, stream modifications, etc.
Although it is often stated that modeling should be used
only to analyze d i f f erences between alternatives, a well
calibrated/verified model can provide absolute values uith
an acceptable degree of reliability. This is especially
true if the relatively, straight-forward changes in model
input and system configuration provide a reasonably accurate
representation of alternatives being analyzed.
However, the same degree of absolute accuracy cannot be
attributed to model parameter adjustments used to evaluate
alternatives such as stormwater drainage plans, urban and
agricultural BMPs, and land/soil disruptions from
construction, mining, silviculture, waste disposal, etc.
The impact of these types of activities on certain
parameters, such as infiltration, soil credibility, soil
moisture capacity, etc. is not well defined; model results
should be viewed primarily as describing the relative
d i f f erences between alternatives based on current best
estimates of the relative change in certain parameter
values.
Specific examples of projects where HSPF and/or predecessor
models have been used to analyze alternatives are described
belou:
119
-------
if i c ci t ions
Q
c-
Q.
t~4
73
Q
-+-
LI
01
Ol
Cl
^j
ro
C
t_
o
-J
«*
T3
O
*
U
01
r_«
O
V)
«0
O
^
J]
13
H-
SHOUVJIdI
o
21
1
*-*!
U.
Q_
in
2:
Q
UJ
1
in
C
o
33
Ul
Z
Q
( t
1
<£
t
Z
UJ
tn
UJ
0£
a_
LU
04
I
UJ
0
c
z:
in
UJ
>
* '
** CT>
C
a
o «
3C 3
tfi
a
0
c
a
_c
u
in
.
a
73
Q
0
trf
c
.*-
o
a
c
. =
f5
in a
a <-
a>->-
c m
>3
j: o
0*-
'o
I-
*
tn
3
a
c
-i-
c
-O 01
C E
ro-n
ro
01
13 1-
a-f
,r-
U 01
- -*
C m
3 13
Z. 2
3
a.
-o
c
ra
a
c _*
a u
ai o
XJ3
o
tn
T3LU
130
ca:
3
o
-*- in
^
t
OX
Q.UJ
a-C
o ct
a o
> !_
ai -C
Q-H
s
13
Cl
t_
.tj
in
o
^^
^
a
01
>
x
o
1^
o
n
13
0
c
o
a.
c
0
+1
ro
(_
01
13
S
13
01
L.
+s
in
c
ctors in NETWORK
PERLND, IrlPLND,
c; 01
.^
C *- m
O U C
a o
U o (3
O t-
o
Q. in a
- OJ 0
L. tn
3 *- LJJ
S ca:
01 H
01 0
01 -1C Q£
C (J
13 073
-c-» c
O.Q (u
o
in in
C 01
o -
c_
-t- 0
u en
ro at
(_ -^
*- Hj
(J
at
> 01
tn
-M 3
f3
73
0) C
1- '3
C
01
c
o
O>-t-
c in
ra
-C X
o a*
tn
01
at
c
0
L)
at
tn
3
73
C
ro
_j
t notsi opQrat i ng
vary i ng outflow
ro 1 inkage to mu 1 1 i
, MUTSIN/PLTGEN
o - £.
oi or 3 Ci
~ ~ 3
M O 0
a -M L. t_
L. JC
-V >-»-
O in f5
-** 3 S ^«
^^ d
grt73 73
UJ (0 in o
~-» m =
737; a
* 01 C L. *
O (3 01 3
-D 0 = >73
o c_ oi re o
£ a-c =
in
at
>
t.
3
O
GI
t-
3 0
L_ 4-
O» in in
C73 L.
- c
-»- ra o
f3 S >
t_ 01 L.
073 0
a m
0 2 O
0 U
c
M- en
-M C
03
Ol O-M
c m
[373
J^ C X
o fij at
tn i_ t_
c QI - a
O -t- I- -t-
o a (o
*- 2 2 2
ro o
L. . a *
a en o
a c o
o at 73 c.
-^ 5^-*^
C -C C
in 0"
0 « U >
> tn > ,^
t. >73
ai a o
m rs a o ro
at c 3 3
0£ fO inn- D"
RES, and NETUORK
er vo i r emd add
tit ing procedures;
analys i s
~ (.1 i_ in
o oi a c
a: c Q. o
o
UJ 3 73 fO
O C O L.
z -*- at
UJ -M u Q.
Z3 t- Q 0
o a a
UJ in x c
in c at
o
2: t_ >
a. o o c
o -M «*- at
in
> inuj o
*^ _i£ j (
u so
73 O < 01
O h- Ol
El J3u_ in
_£
u
o
at
1_ C.
= o
r, >
01 '_
t. at
-t-i tn
tn Q
c_
01
C73
.^ a
-M in
in o
^- a.
x o
a i-
a
o
u ru
ro
Q.*-
o
C* 2
O tn
-M C
o
in
t- ra
ai
o
> -*-
t. in
ai ai
in >
at c
Oi
s of f 1 ous or
the NETUORK, RCHRES
o
~ £.
c CUJ
M- rc_i
=CQ
73 01 <
0731
= u.
2
(- O L,
o ^ o
* X
T3*«73
T3 3 C
< o 13
I
^^
^
O
C_ 0
o
(_
t- 01
con
..- (D
o t.
m -»- *v
Cl =
Ol in r3
C 2 01
(j O t_
JI -n
O **- in
X
73
C
ro
f-
o
-H«
F3 in
-*-* C
C O
at
E m
Ol L.
3 at
(3 >
273
O
^^ t_
U- 0
Cki ji
UJ U
Z 0
LU <
O.Q
Oicn
C UJ
o
in Q£
3Z)
Q
m in
'_ h
ox
U UJ
0
!- C
^^
- t
<5O
H- c
.tl .2
73-M
o a
E 0
i
c
, »
in
^^. QJ
73 =
0 3
U 0
a >
a
a> -f
**f ro "
4- m
73 C C
ai at
C fj-M
t_ c
4- "-
GJ C
7J-- -
> in o
-H Q
t_ in -4-1
O fD fO
'ai at L.
t_ 3
O O 73
c
o
1C
c
01
;=
5»
3
ra
^H
^^
ro
**-
c
ro
0£.
o
to
tn ui
^73
t_
c rs
T;
Ul fu
O -M
^-* in
*- >
*->
3 ^^
a 13
C 3
IT
73 C
fy dl
0*-
ro
2
-»j
c "
Q
O 01
a L.
> tn
*- »«
73 Oi
o at
z: s
i
73
0
, «
73
0) Q
-*-* jj
in
tj O
Z Q.
X
ai at
x
J 01
ro c
2 "~
O -M
r-< Ul
^-H -r-
ro x
Gi tn
O t_
-»- ai a
3 c ai
^3 O (_
- = fj
I- ^ -C
-M U
in Ol m
..- c -
O 73
C
O
-»-
ro
u
o
,-«
^^
ro
73
ro
0
^4
01
*
in
ro
3
73
at
15
U
O
in
in ui
fi3 a
ai
c ra
t3 .C
U
^
O -*-
o u
^ Cl
-O »*
l+_
m ai
UJ l_
C£
ZI O
o *-
QC
tn
>UJ
CQ
73<
O h-
Hu.
c
n
u
.
^^
in
- Ul
u c
31 O
u +*
(3 U
(_ a
(g in
JZ
O =
ro
Z at
o u
H- in
73 U
a
u
73 ai
o a.
E in
^
w C
c o
o
f 13
tO N
o
»- a
~o c
0 13
E-C
o^
~ tn
ro ai
01 Ol 01
I- >
f a> a
V)^"-4
D pcirciinet ers and/or
» n ronr-n-sm-1* ill =>.-.
ions and/or 1 ink ci go
eatinent niodGl
"Z. *- '
CX C! '- \
UJ T> 01 C)
o_ ci Q. cn
cf o FO
o> c t-
-^ a: o
to iri LU -^j
t: . z in
_ UJ
a = o c
00 13
i-*. a .a
a t"! i. u
5" in 3
>in a ro
c^
O=: >-«
o o (3
Sac = =
TJ
Q
in -^
o z
a 3
o =
t_ -*-
a r;
,moQ
^ -T "~
T) C u
o o o
Z O 01
0-
£
m
j:
u
13 O
a m
a
~a
a
5
in o
* l_
Ct-
01
c o
0 U
a c
s o
o u
u o c
»- o
-a»-
01 -*-
CTJ
TJ
»-T3 C
IB C O
a 13 u
1
~* ^
3 C
U O
3
L. Ol«
Cn Oi in
10 13 a.
CE
L. taca
o =
X
D -^ tn
c in at
13 01 O
J3
C -f
13 O
.a ro rj
i-i-i.
Z5 3 Q.
D parameters to
changed condit ion;
rate pervious 1 ciii d
rai:ic.'t ers
r: u ~ "J
: O CL C-
C£ G(
1U t>"> "G
Q.-3 0
a -
ca ^- GI *-
fj V "~ ~
^ U 3 O
t_ Jl 3" ~
n. ^ 0)
O " ^-
t_ - -*-
g. c *
Q.-^ 0 3
tu C -**
GJ *- -M
> m o ^
**- Cl GI
-n 5. S
o a GI
E t- 2 n
^
o
,-.
01 a.
C 3
^ t-
»« .1
M
3t!
in
O')-
i- O
m o
C 2.
O >
*" -H
«
- O
T>
C «-
O
u u
9
D a
ai in
c
s
<«- o
Of U
Oi-
tn
c
0-01
c t-
-f 0 3
Q. *
31- "^
1- U 3
in 3 (J 13
.^ t- "- m
TJ*- > 0
m^H a
c m
..- o m -
0 U -T3
in n>
-*. c a
a 01 M
C C in
13 oi ra
_i~ S 3
120
-------
loua River Has in Study
The Iowa River Basin study (discussed throughout this
manual) was designed to evaluate the utility of HSPF as a
planning tool to analyze the runoff, pollutant loading, and
instream water quality changes resulting from proposed
agricultural BMPs. Following the preliminary hydrology and
sediment calibration, and pesticide and nutrient simulation
for the entire basin, the BMP analyses were performed.
Conventional agricultural practices for Iowa provided the
base conditions to which a proposed BMP scenario i.e., a
combination of selected, compatible practices, was compared.
The definition of conventional practices was as follows:
DEFINITION OF CONVENTIONAL AGRICULTURAL
PRACTICES FOR IOWA WATERSHEDS
Conventional agricultural practices for rowa are
assumed to include continuous row-cropping (no
rotation) and moldboard plowing followed by
secondary tillage at least once to smooth and
pulverize the soil for planting, with cultivation
when and where appropriate. Cropping and tillage
operations are assumed to be straight row and
usually parallel to field borders regardless of
slope direction. Fertilizer application and
moldboard plowing are assumed to be done in the
fall, with disking and pesticide application in
the spring prior to planting. In all cases with
conventional tillage, the soil surface is free
from residues for a period of time. One or two
cultivation operations may be performed as needed
during the early growing season (Donigian et al.»
1983a) .
The primary components include moldboard plowing and
secondary tillage for seedbed preparation, one or two
(chosen for simulation) cultivation operations during the
early growing season, and crop residue removal following
harvest in the fall.
The BMP scenario chosen for simulation included conservation
tillage plus the use of contouring; the assumptions used in
representing this scenario are listed in Table 8.2. These
changes are based on studies performed as part of the Iowa
Field Evaluation Program by Donigian, et al., (1983a) to
assess the effects of a variety of candidate BMPs on HSPF
model parameters. Specific parameter values for base and
BMP conditions are included in the Iowa River Study report.
121
-------
As noted in Table 8.2, the primary components of our BMP
scenario were (1) a shift from moldboard plowing to chisel
plowing and field cultivation as primary tillage, (2) one
summer cultivation for weed control in place of two
cultivations under base conditions, and (3) allowing crop
residues to remain on the field following harvest. These
components were modeled by increasing parameter values for
soil moisture retention (UZSN), rainfall interception,
surface roughness (Manning's n), and land cover; and
decreasing the sediment fines produced by tillage. The
infiltration parameter was not changed, under the assumption
that the primary tillage operations have similar effects on
the infiltration process. Also, there was no change in
chemical parameters, soil bulk density, soil temperature, or
chemical application amounts, although fall fertilizer
application was replaced by increasing the spring and summer
applicati ons.
Using these assumptions and associated changes in parameter
values, the resulting comparison of this BMP scenario and
the previously simulated base conditions is shown in Tables
8.3 and 8.4. Table 8.3 presents a detailed comparison of
the edge-of-stream loadings for the BMP and base conditions
while Table 8.4 lists the resulting basin-wide loadings
measured at Marengo, Iowa. The other/pasture land use
category shows no effects since only corn and soybean
cropland was affected under this BMP scenario.
Land Surface Simulati on. Over the five year simulation
period, annual runoff reductions from soybean and corn
cropland were in the range of 4% to 17% with the larger
reductions generally observed for corn. Groundwater
outflow, the largest contributor to streamflow, shows the
smallest effect (average reduction of 4.2%) from the BMP
while surface runoff is decreased significantly (average
reduction of 30% for soybeans and 26% for corn). As a
consequence, sediment losses which come entirely from the
surface were also reduced dramatically with soybean and corn
reductions ranging from 45% to 69% (average 52%) and 33% to
73% (average 47%), respectively. In addition, BMP effects
on erosion were much more pronounced than the resulting
loading at Marengo since most of the sediment loading at
Marengo resulted from channel scour processes rather than
from land surface erosion. Solution alachlor edge-of-stream
loading reductions were in the range of 4% to 42% with
slightly greater reductions occurring on soybeans than corn;
the average decrease for soybeans was 33%, and 19% for corn.
Nutrient simulation for both base conditions and the BMP was
also performed for the entire five year simulation period.
As shown in Table 8.3, total annual nitrate nitrogen
reductions ranged from 3% to 10% for soybeans and 5% to 54%
122
-------
for corn. Ammonia nitrogen was reduced 23X to 35X for
soybeans and 18% to 82X for corn. The nitrate reductions
were lowest in the first year of the simulation period since
we assumed the same initial storages in the soil for both
the base conditions and the BMP. Lower nitrate and higher
ammonia storages in the first year of the BMP simulation
TABLE 8.2 SELECTED BMP SCENARIO FOR SIMULATION ON THE IOWA
RIVER BASIN
CONSERVATION TILLAGE PLUS CONTOURING
1. Chisel plowing replaces fall moldboard plowing on
corn residue
2. Field cultivation replaces spring plowing and disking
on soybean residue
3. No change in infiltration parameter
4. Residues remain after harvest, with the following
reductions by tillage and decay:
Moldboard 90X
Chisel 35X
Light disk 30%
Field cultivation 3OX
Winter decay
Soybeans 30X
Corn 10%
5. Reduction in sediment fines from tillage: 50 - 70X
6. UZSN increases due to contouring and less seedbed
preparation
7. One summer cultivation replaces two cultivations
under base conditions
8. Rainfall interception, surface roughness
(Manning's n), and land cover increase due
to residues and less tillage
9. No change in chemical application amounts, but fall
nitrogen fertilizer application moved to spring and
summer. Incorporation distribution as follows:
Surface Upper
Moldboard OX 100X
Chisel 50% SOX
Disking 20X SOX
NH3 Injection OX 100X
Cultivation 10X 60X
10. No change in chemical (pesticide or nutrient)
parameters, bulk density, or soil temperature.
123
-------
uould have been more consistent with the storages calculated
during the rest of the period. In fact, the initial nitrate
nitrogen storage was high enough to preclude a significant
reduction by the BMP scenario in 1974.
The effects of the BMP scenario upon surface nutrient
processes occurring on the land surface are relatively small
since the primary effect is to reduce surface runoff and not
to affect the plant growth and other biological/chemical
3.3.
Comparison of Edge-of-St ream Loadings for 3ase Conditions and 3,1P
Simulations in the Iowa 3iver Basin
RUNOFF (mm)
Mil
51"?
BASE
CC'Ul
3MP
BA_SE/5f1P.
974
975
976
977
973
age
S *
I *
G *
T *
237
203
124
31
331
21
29
156
207
.3
. ]
. 5
.2
.4
.3
.0
.0
.0
269,
193.
113.
73,
313,
15
24
154.
1 94
. 1
, 7
.4
.3
. 1
.3
.7
.0
.0
-6.
-4 ,
-a ,
- 9
-5 ,
-30,
-15,
-1
-6
. 5
, 5
.9
. 1
. 5
.3
. 3
311.
221 ,
131
37.
342.
26
33
159
220
.9
.7
.8
.9
.3
. 9
t Q
.0
.0
235.
204.
114.
72,
320,
1 9,
29,
150,
200,
,7
,7
.2
.9
. 1
. 9
, 2
, 0
.0
Avar
SEDIMENT (tonnas/ha)
1974
1975
1976
1977
1973
Average
ALACHL03 (kg/ha)
1974 0.0905
1975 0.0518
1976 0.0102
1977 0.00196
197S 0.0764
Average 0.0462
0.616
0.224
0.062
0.018
2.946
0.773
0 .240
0.097
0.019
0.009
1 .331
0.369
-45
-57
-69
-47
-53
-52
-3.3
-7.S
-15.
-17.
-6.6
-26.
-14.
-5.7
-10.
0 .399
0.375
0. 135
0.028
2.665
0.320
0.600
0. 197
0.036
3.015
1 .322
0 .434
-33.
-47 .
-73.
-43.
-50.
-47.
0
0
0
0
0
0
.0630
. 0302
. 0066
.00133
.0524
.0303
-30.
-42.
-35.
-4.5
-31'.
-33.
0 .
0 .
0 ,
0
0.
a
. 185
.0103
.00697
.00132
.0380
.0483
3.
0 .
0.
0.
0.
0.
150
0031
00539
00125
0305
0390
-19.
-21 .
-23.
-5 .
-20.
-19.
36 1 . i
252.0
156 .7
121.0
370 .7
10.7
27.9
214.0
252.0
0 . 032
0 . 323
0.007
0 . 002
0 .423
0 . 107
0 . 0
0 . 0
0 .0
0 .0
0 .0
0 .0
NITRATE (kg/ha)
1974
1975
1976
1977
i978
23.32
9.1S
6.69
4. 12
1 1 .57
22.60
S.66
6.06
3.68
11.18
-3. t
-5.7
-9.4
-10.
-3.4
51 .80
25.27
15.25
6.91
29.46
49.38
13.59
8.36
3. 17
17.93
-4.7
-46.
-45.
-54.
-39.
7.16
4.10
2.92
2.53
7.16
Average S * 0.0129 0.0091 -29. 0.193 0.1SO -6..7
I * 1.14 0.975 -14. 7.69 6.10 -21 .
G * 9.82 9.45 -5.3 15.5 12.21 -21.
T * 11.0 10.44 -5.1 23.4 13.49 -21.
a.ooso
0 .349
3 . S2
4.77
-------
Table 8.3. continued
SOYBEANS
BASE BMP
AMMONIA (kg/ha)
1974
1975
1976
1977
1973
Average S
I
G
T
MINERALIZATION
1974
1975
1976
1977
1973
Average
DENITRIFICATION
1974
1975
1976
1977
1978
Average
0.719
0.420
0.634
0.217
0.807
0. 0861
0.422
0. 0500
0.559
tkg/ha)
51.3
56.0
56 .3
56.4
56.3
55.3
Ckg/hai
8.23
4.91
5.26
6.14
5.30
6. 1
0
0
0
0
0
0
0
0
0
51
55
56
56
56
55
8
4
5
5
5
6
.551
.296
.409
.141
.587
.0637
.233
.0488
.397
.0
.3
. 1
.3
.9
.2
.23
.36
. 16
.93
.78
.01
2 DIFF
-23.
-29.
-35.
-35 .
-27,
-26.
-36.
-2,
-29,
-0 ,
-0 .
-0.
-0.
-0,
-0,
+ 0,
-0
- 1
-2
-0,
-1 ,
.4
.5
.3
.3
.2
,7
.2
.7
. 9
. 9
.6
. 4
.6
SASE
3
2
*>
1
5
0
2
0
3
47
51
51
51
51
50
17
14
13
13
' 4
14
. 12
.95
.45
.29
.28
.297
.65
.0578
.02
. 1
.0
.0
. 1
.9
.4
.3
.2
.7
.6
.3
.6
CORN
BMP 2 DIF
2
1
0
0
4
0
1
0
1
46
50
50
50
51
49
13
12
12
12
12
13
.23
.90
.641
.233
.34
.773
.03
.0479
.87
.4
.5
.5
.7
.3
. 9
.5
.4
1
.2
.6
.6
-28
-35
-74
-82
-18
+ 160
-61
-17
-38
-1
- 1
-1
-0
-1
-1
+ 6
-13
- 1 1
-10
- 1 1
-6
.3
. 1
.0
. 9
,3
.0
.9
.8
PLANT UPTAKE (kg/ha)
1974
1975
1976
1977
1978
Average
48. 1
34.4
37.7
55.0
39.7
43.0
49
35
33
54
40
43
.6
.8
.4
.4
.7
.8
+ 3
+ 3
+ 1 ,
-1 ,
+ 2,
+ 1 ,
. 1
.8
.8
. 1
.3
.9
147
138
148
186
141
152
.8
.6
.5
.5
.2
.0
177
153
159
190
156
167
.6
.0
.2
.4
.9
.4
+ 2
+ 10
+ 7
+ 2
+ 1 1
+ 10
.0
.
OTHER/PASTURE
BASE/BMP
0.772
0.516
0.521
0.282
0.976
0.0528
0 .446
0.114
0.613
38
33
37.6
33.
38.
33.3
2.46
2.26
2.17
2 .32
3.11
2.6
25.9
25.5
25.S
35.7
32. 1
29.0
S = Surface OutfloN
I = Interflow Outflow
G - Groundwator Outflow
T. = Total Outflow
processes occurring in the soil. In addition*
storage of nutrients generally present on the land
any significant change in the nutrient processes
relatively small change in runoff. Consequently,
uptake, denitrification, and mineralization
significantly changed under the BMP scenario.
the large
precludes
due to a
the plant
are not
Instream Simulation. Table 8.4 shows the effects of the BMP
scenario on the runoff and water quality of the loua River
125
-------
TABLE 8.1* COMPARISON OF LOADINGS IN THE IOWA RIVER AT MARENGO
FOR BASE CONDITIONS AND BMP SIMULATIONS
BASE
RUNOFF (mm)
SEDIMENT
(tonnes/ha)
SOLN. ALACHLOR
(kg/ha)
SED. ALACHLOR
(kg/ha)
NITRATE N
(kg/ha)
AMMONIA N
(kg/ha)
1974
1975
1976
1977
1978
Average
1974
1975
1976
1977
1978
Average
1974
1975
1976
1977
1978
Average
1974
1975
1976
1977
1978
Average
1974
1975
1976
1977
1978
Average
1974
1975
1976
1977
1978
Average
183.0
124. 0
80. 0
47.8
299.0
147. 0
3.91
0.88
0.56
0.019
5.69
2.21
0.0278
0. 0023
0. 0003
0.00
0 . 0068
0.0076
0.0032
0.0002
0.00
0 . 00
0. 0007
0.0008
31 .0
14.9
9.5
4.9
18.5
15.8
0.48
0.57
0.53
0.37
0.91
0.57
BMP
170.0
116.0
73.9
42.4
280.0
136.0
2.62
0.147
0.12
0.012
5.49
1 .74
0.0219
0.0017
0.0004
0 . 00
0.0048
0.0058
0.0020
0.0001
0.00
0 . 00
0.0004
0.0005
29.8
9.5
6.2
3.0
13. 1
12.3
0.41
0. 30
0. 20
0.09
0.46
0.29
DIFFERENCE
-7. I
-6. 4
-7 .6
-11.3
-6.4
-7.5
-33.0
-47.0
-79 . 0
-37. 0
-3.5
-21 .0
-21 .0
-35.0
-50.0
-29 . 0
-24. 0
-38.0
-50. 0
-43. 0
-38.0
-3.9
-36.0
-35. 0
-39. 0
-29.0
-22.0
-15.0
-47.0
-62. 0
-76.0
-49. 0
-49. 0
measured at Marengo, loua. Over the five year simulation
period, total annual runoff reductions at Marengo were in
the range of 7% to 1IX with an overall average of 7.5%
reduction. Annual sediment loss reductions uere generally
higher varying from 4% to 79X reduction with an overall
average of 21% reduction over the simulation period. These
sediment loss reductions are somewhat less than what might
be expected; however, as discussed above, a significant
portion of the total sediment loss is derived from the
channel system itself which would not be significantly
affected by the BMPs. Also, the 4% reduction in 1978 biased
the average; the average reduction in 1974 to 1977 was 49%.
Solution alachlor at Marengo was reduced from 0% to 50% with
126
-------
an average of 24% reduction over the simulation period;
sediment alachlor was also reduced from 0% to 5054, averaging
37.5% over the period. The 0% reduction in alachlor
occurred in 1976 and 1977, the years of extreme drought in
central Iowa.
The instream nutrient results are also presented in Table
8.4 as annual nitrate and ammonia loadings at Marengo. The
nitrate nitrogen reductions ranged from 4% to 39% over the
simulation period with an average reduction of 22%. Ammonia
nitrogen was reduced by 15% to 76% with an average reduction
of 49%; this reduction was considerably higher than the
nitrate reduction due to reduced sediment loadings that
transport the adsorbed ammonia nitrogen. As discussed above
for the edge-of-strearn loadings, the reductions for nitrate
and ammonia were lowest in the first year of the simulation
period due to the same initial nutrient storages in the soil
for both the base conditions and the BMP.
Figure 8.1 compares the frequency curves for nutrient
concentrations at Marengo resulting from simulation of base
and BMP conditions for the 1974-1978 period. Both the
nitrate and ammonia curves indicate a general decrease of
instream nutrient concentrations for the BMP scenario;
extreme and median values are reduced for both constituents.
Generally speaking, reductions in ammonia concentration
resulting from the modeled BMP scenario were relatively more
pronounced than reductions in nitrate, particularly for
extreme values. For example, the 10% level for ammonia
(i.e., the concentration which was exceeded 10% of the time)
was reduced by 60% from the base conditions to the BMP
scenario, while only a 13% reduction in nitrate occurred.
The best management practices are more effective in reducing
peak concentrations of ammonia than nitrate because improved
sediment erosion control prevents adsorbed ammonia from
reaching the channel, while erosion control has a limited
effect on the highly mobile nitrate species. Reductions for
median concentrations resulting from the BMP scenario were
18% and 34% for ammonia and nitrate, respectively. The
relatively large reduction in nitrate concentration for mid-
range events can be attributed to two phenomena resulting
from best management practices: (1) increased nitrate uptake
by plants and (2) decreased groundwater flow. Large
quantities of nitrate are carried to the river by
groundwater flow, and reduction of instream nitrate
concentrations is a natural consequence of decreasing
groundwater flow and associated concentrations. On the
other hand, ammonia loadings from groundwater are relatively
small, and instream concentrations of ammonia are not nearly
as sensitive to reductions in groundwater flow.
127
-------
100.0
50.0
20.0
10.0
5.0
2.0
o
°-5
0.2
0.1
0.05
0.02
0.01
^
"w
1
I
\
\
\
X
^\- I
X ".
V
\>
N
*
.\
\
\
\
t
\
X
\
\
X
\
\
\
\
\
\
\
\
\
\
i
i
\
\
\
\
I
^
\
\
\
\
%
\^
\
\
\
\
\
\j
NO - N
\\
\ N
\ V
\
\
NH
\
\
\
S, i
\ *
\ ^
\ \
\\
\v
\\
\
\
N N
X
-N
\
X
N\
BMP
BASE
0.1 1.0 5.0 10.0 20.0 50.0 75.0 90.095.098.0 99.8
Percent time concentration exceeded
Figure 8.1 Frequency Curves for Simulated Ammonia and Nitrate at
Marengo, Iowa. Base condition and nutrient BMP results
are illustrated.
128
-------
Risk Assessment. One of the possible uses of continuous
modeling of chemical fate and transport is to evaluate the
risk or exposure of aquatic organisms to various magnitudes
and duration of chemical concentrations. Figure 8.2
demonstrates how the frequency, or percent of time, of
acute, chronic, and sublethal conditions might be determined
for a particular organism and stream given a time series of
chemical concentrations. This methodology was developed in
work by Onishi et al., (1979) in providing a procedure to
assess the risk of chemical exposure to aquatic organisms.
Using these procedures the simulated chemical concentrations
under both base conditions and the BMP scenario were
analyzed to determine the percent of time conditions within
each region shown in Figure 8.2 would exist. The results of
this analysis are shown in Table 8.5; the table title
indicates a hypothetical organism because all the values
observed for alachlor concentrations were considerably lower
than any of the MATC (maximum acceptable toxicant
concentration) values for common species of fish found in
the Iowa River.
Table 8.5 also shows the reductions in the fraction of time
when acute and lethal conditions exist under the simulated
BMP scenario. The specific choice of MATC and lethality
data chosen for this analysis resulted in no change in the
percent of time when acute conditions existed, primarily
because the maximum simulated value was still sufficiently
large to exceed the values that define the acute region
under both conditions (base conditions and BMP scenario).
A concentration of 30 ppb solution alachlor defined the
single day (24-hour) acute concentration threshold for our
hypothetical organism. The maximum observed solution
alachlor concentrations in each year for both the base
conditions and BMP scenario are listed below:
Annual maximum daily concentrations of
solution alachlor (ppb)
Year Date Base BMP £ Change.
1974 5/16 286. 262. - 8.4
1975 6/15 27. 16. -41.
1976 5/29 17. 12. -29.
1977 5/22 2.0 1.6 -20.
1978 5/18 105. 90. -14.
Thus, although the BMP scenario provided substantial
reductions in the peak concentrations ranging from 9X to
41%, the absolute reductions in 1974 and 1978 were not
sufficient to reduce the concentrations below the 30 ppb
threshold used in our risk analysis.
129
-------
QC
^
Q
flj
-p
(0
Q
c
o
H
4J
(0
M
4J
c
H)
O
C
o
o
(0
o
H
0)
A
U
M-l
O
CO
H
to
rt
c
(0
fi
4J
(1)
CN
QO
(U
S-l
ri
tji
H
NOIlVdlNBONOO
130
-------
TABLE 8.5 LETHALITY ANALYSIS OF BMP SCENARIO FOR ALACHLOR
IN THE IOWA RIVER AT MARENGO, IQUA
Global Exceedance
(X of time)
Base Conditions BMP Scenario X Chance
Acute Region 0.49 0.49 0
Above NATO Value 3.50 2.68 -23.4
Sublethal Region 96.50 97.32 + 0.8
(below HATC)
MATC - Maximum Acceptable Toxicant Concentration
(0.003 mg/1 used above)
The fraction oi time when lethal conditions exist, both
acute and chronic, is represented by the values listed on
the line entitled "Above The NATO Value" in Table 8.5. The
reductions indicate a 23% reduction in the percent of time
when lethal conditions occurred in the watershed.
Obviously, reductions in the percent -of time for lethal
conditions will correspond to an increase in time for sub-
lethal conditions. Although the values listed here 'are
specific to the conditions under which this BMP scenario was
simulated, the overall methodology and analysis indicates
how the procedures described here can be used to evaluate
the effects of BMP scenarios on the resulting risk of
exposure of aquatic organisms to chemicals.
Dominican Republic HvdTOPower Study
One of the early applications of HSPF was in a hydropower
study of the Rio Yaque del Norte Basin for the Dominican
Republic (Hydrocomp, Inc., 1980). Hydropower is a major
source of electricity in this developing country, which is
experiencing an 1IX annual increase in demand. Twenty
potential hydropower sites were identified and 10 potential
network configurations were hypothesized. The analysis
procedure consisted of the generation of 99 years of
synthetic precipitation, calculation of land surface runoff,
and calculation of natural streamflow at 21 sites (shown in
Figure 8.3). Power generation was simulated by running the
streamflow through the 10 different hydropower
configurations. The time series for depth of flow (head)
131
-------
and flow rate uere
estimate the most
Johanson, 1981).
then analyzed using the GENER module to
efficient configuration (Barnuell and
Generation of hydroelectric power involved operation of HSPF
to first simulate a hypothetical 99-year streamflow period
and then route the streamflou through diversion works or
storage reservoirs to a penstock and turbine facility. The
flow was then returned to the river for reuse further
downstream.
The operation of the diversion dams was simulated using the
RCHRES module of HSPF.
streamflow and diverted
FTABLE to specify the
output was multiplied
compute simulated power
The input to the system was natural
flow, and spill was output using
diversion demand. The diversion
by a power conversion factor to
generated. Duration analyses of
power and spill were performed using the DURANL module.
Figure 8.3 Location of the 21 Dam Sites for Power Generation
in the Rio Yaque del Norte watershed, Dominican Republic
(Hydrocomp Inc., 1980)
132
-------
The storage dams were operated in a similar manner using
RCHRES. The reservoir depth-storage relationship was
incorporated in the FTABLE and the variable stage (head)
calculated. Power generated and duration analyses were
treated in the same manner as for the diversion dam
analysis.
In a single HSPF run, an entire multi-diversion and storage
configuration was completely analyzed. Operations proceeded
in sequence from upstream to downstream, with each result
routed to further operations as required by the particular
configuration. Complex configurations, such as interbasin
water transfers and streamflow alteration by upstream
generation facilities were handled without problems.
Clinton River Stormwater Management Study
The Macomb County (Michigan) Public Works Department has
used an early version of HSPF to evaluate stormwater
management alternatives for small study areas within the
Clinton River basin (Minn and Barnes, 1982). The objective
of the study was to determine the effect of stormwater
retention from upstream areas on downstream flows. The
study area selected as a sample case was the Dunn-Wilcox
watershed in southeast Shelby Township (Figure 8.4). This
watershed within the Clinton River basin has an area of 1942
hectares (1800 acres), of which 32 percent is developed with
most development occurring in the upper part of the
watershed. Drainage is provided by nine county drains. In
addition, five state-owned borrow pits and seven man-made
lakes are available to store stormwater runoff. Future
development in the watershed is expected to increase the
severity of flood problems.
Prior to the investigation of possible stormwater management
alternatives, the model was calibrated for the entire
Clinton River basin and all sub-basins containing streamflow
records. Following the calibration of the model, 48 years
of simulated streamflow data were created by the model using
historical precipitation data and present land use
conditions. This was done to generate a consistently long
period of streamflow data for the entire basin without
having to consider the effect of land use changes on the
recorded flow data in the past 48 years. The simulated
streamflow record is more representative of the runoff that
would occur under current land conditions in response to
historical meteorological data, than the observed historical
streamflow record.
After the completion of calibration, the June 1968 flood was
selected as the study flood for evaluating stormwater
133
-------
management alternatives on the Dunn-Wilcox watershed (Figure
8.5). This flood is the largest of record. Two sets of
flows were simulated for the June 1968 event. The first was
using existing conditions (present land use patterns); the
second assumed full development of the watershed in
accordance with the Township's Master Land Use and Zoning
Plan.
These flows were then routed through a combination of
different drain facilities with and without retention. The
drain facilities consisted of (1) drains in their present
condition, (2) enlarged drain channels, and (3) enlarged
drain channels with extra stormwater storage (wide channel
tops and lake storage).
Results were analyzed for three sets of land use and drain
channel combinations. These combinations are: (1) present
land use and present channels, (2) future land use and
improved (enlarged) channels, and (3) future land use and
Lake St. Clair
Location of Dunn-Wilcox Watershed
OAKLAND CO. \ |MACOMB Co|
WAYNE CO.
Figure 8.4 Clinton River Drainage Basin, Micnigan
-------
<0
3 O> "O
^} OQ C
00 of
s §
£ .2
£ » S S
o c c o
*r dj C8 **
^ 0 6 CC
QC
C1J
x:
to
s_
OJ
+J
X
O
o
I
c
c
LO
co
CD
135
-------
improved channels with ext
simulation results show that
(combination #2) is increas
present conditions (combinati
channel tops and lake storage
system for the future land us
peak flood flows to the 1
conditions (see Table 8.6).
in the peak flow at the downs
extra storage is used (comb
improving the channels (combi
ra stormwater storage. The
the future condition peak flow
ed two to three times over the
on #1). The addition of wider
to the improved drain channel
e case (combination #3) reduces
evels seen with the present
Figure 8.6 shows the reduction
tream end of the watershed when
ination t3) compared to just
nation 12) .
As urbanization of the Dunn-Wilcox watershed increases to
its planned maximum concentration it will not be sufficient
to just enlarge the present drain channels. In addition
enlarging the channels, extra channel and lake storage will
be required to contain major floods.
As shown by this study, watershed simulation makes possible
the analysis of different land use conditions and potential
solutions to flooding and other problems. The authors of
this study noted in their report (Uinn and Barnes, 1982)
some of the advantages this watershed simulation approach
offers to public works engineers and planners. These
advantages are:
1. Consolidation of detention facilities, thus
minimizing the number of small private and
troublesome basins.
2. Large basins offer multiple use potential, thus
minimizing maintenance problems and
expend i tures.
3. Channel storage is an extension of county
drains which now exist and no additional
maintenance would be required.
4. Considerable savings in drain construction by
comparison of open drain versus (inclosed drain
construction costs.
5. A reduction in culvert and bridge sizes for all
road crossings.
I 36
-------
TABLE 8.6 COMPARISON OF MAXIMUM FLOWS (CFS) FOR REACHES
WITH CHANNEL STORAGE
Reach
Number
908
912
935
936
937
939
941
Combination tl
258
321
533
Combination #2 Combination #3
635
66.3
723
895
271
1458
1456
1755
1815
300
300
150
450
375
625
600
2000 -
CO
u.
O
S^
Ul
O
O
CO
Q
1500 -
1000 -
500 -
1815
Combination *2
I 1 i i i i
0400 0800 1200 1600 2000 2400
A.M. P-M.
TIME (Hours)
Figure 8.6 Hydrograph of Reach 9^' for June 26,
1968, Event
137
-------
SECTION 9
REFERENCES
Barnuell, T. 0., Jr., and R. C. Johanson. 1981. HSPF: A
Comprehensive Package for Simulation of Watershed Hydrology
and Water Quality. In: Nonpoint Pollution Control: Tools
and Techniques for the Future. Interstate Commission on the
Potomac River Basin. Rockville, MD. pp. 135-153.
Donigian, A.S., Jr. and H.H. Davis, Jr. 1978. User's
Manual for Agricultural Runoff Management (ARM) Model. U.S.
Environmental Protection Agency. Athens, GA.
EPA-600/3-78-080.
Donigian, A.S., Jr. and N.H. Crawford. 1979.
for the Nonpoint Source (NPS) Model. U.S.
Protection Agency. Athens, GA.
User's Manual
Envi ronmental
Donigian, A.S., Jr., J.L. Baker, D.A. Haith, and M.F.
Walter. 1983a. HSPF Parameter Adjustments to Evaluate the
Effects of Agricultural Best Management Practices. Draft
Report. U.S. Environmental Protection Agency. Athens. GA.
Donigian, A.S., Jr., J.C. Imhoff, and B.R. Bicknell. 1983b.
Modeling Water Quality and the Effects of Agricultural Best
Management Practices in Four Mile Creek, Iowa. U.S.
Environmental Protection Agency. Athens, GA.
Dyer, H.L. 1971. Liquid Waste Emissions Factors.
National Laboratory. Argonne, IL.
Argonne
Heinitz, A.J. 1973. Floods in the Iowa Basin Upstream from
Coralville Lake, loua. U.S.G.S. Open File Report.
Hydrocomp, Inc. 1980. Analysis of Power Generating
Configurations in the Rio Yaque del Norte Watershed.
Mountain Vieu, CA.
Imhoff, J.C., B.R. Bicknell, and A.S. Donigian, Jr. 1983.
Preliminary Application of HSPF to the Iowa River Basin to
Model Water Quality and the Effects of Agricultural Best
Management Practices. U.S. Environmental Protection Agency.
Athens, GA.
138
-------
International Business
Programming Textbook
Prog ram.
Machines Inc
£ Workbook
Johanson, R . C . , J . C
Imhoff
1974. Structured
Independent Study
Programmer's Supplement
Program - FORTRAN. This
and H.H. Davis,
for the Hydrological
material is on magnetic
Jr. 1979.
Simulation
tape.
Johanson, R.C., J.C. Imhoff, H.H. Davis, Jr., J.L. Kittle,
Jr., and A.S. Donigian, Jr. 1984. User's Manual for the
Hydrological Simulation Program - FORTRAN (HSPF): Release
8.0. U.S. Environmental Protection Agency. Athens, GA.
Knisel, W.G., editor. 1980. CREAMS: A Field-Scale Model
for Chemicals, Runoff, and Erosion from Agricultural
Management Systems. U.S. Dept. of Agriculture. Conservation
Research Report No. 26.
Linsley, R.K., Jr., M.A. Kohler, and J.L.H. Paulhus. 1975.
Hydrology for Engineers. 2nd Edition. McGraw-Hill Book
Company. New York, NY. 482 p.
Metcalf and Eddy, Inc.
Collection, Treatment,
Company. New York, NY.
1972. Wastewater Engineering:
and Disposal. McGraw-Hill Book
Onishi, Y
Wise, and
Instream
Battelle
, S.M. Brown, A.R. Olsen, M.A.
W.H. Walters. 1979. Methodology
Migration and Risk Assessment
Pacific Northwest Laboratories.
Prepared for
GA. 204p.
U.S
Environmental Protection
Parkhurst, S.E.
for Overland and
of Pesticides.
Richland, WA.
Agency. Athens,
Wallace, D.A. 1971. Determination of Land Runoff Effects
on Dissolved Oxygen by Mathematical Modeling. Ph.D. Thesis.
University of Iowa. Iowa City, IA.
Wallace, D.A. and R.R. Dague. 1973. Modeling of Land
Runoff Effects on Dissolved Oxygen. Journal of Water
Pollution Control Federation. Vol. 45, No. 8. Washington,
DC.
Winn, G.E. and F. Barnes. 1982. Alternatives for Storm
Water Management in the Dunn-Wilcox Watershed, Shelby
Township, Macomb County, Michigan. Office of the Macomb
County Public Works. Mt. Clemens, MI.
Zison, S.W., W.B. Mills, D. Deimer, and C.W. Chen. 1978.
Rates, Constants, and Kinetics Formulations In Surface Water
Quality Modeling. U.S. Environmental Protection Agency.
Athens, GA. EPA-600/3-78-105.
139
-------
APPENDIX A
Sample HSPF Input Sequence
This input sequence was developed and
used in the loua River Study- The
sequence provides the input instructions
and parameters necessary to simulate
hydrology* hydraulics, sediment and
pesticide processes on the basin land
surface and within the loua River.
-------
/VHSPF7 JOB (R72SXA, 185,30. , 99) , 'IOWA-6' , REGION=5 1 2K
//HSPF7 EXEC PGM-HSPF,REGION=512K
//STEPLIB DD DSN=WYL . XA . Q 1 1 . HSPF7 . LM, DISP=SHR ,
// UNIT=DISK,VOL=SER=PUBOI2
// DD DSN=SYS2.F03.PROD.LINKLIB,DISP=SHR
//FT01F001 DD DSN=WYL.XA.Q1t . HSPF7 . INFOFL , DISP = ( OLD, KEEP ),
//FT02F001 DD DSN=WYL .XA . R72 . HSPF . TEMP . UCI FL , DISP = ( OLD, KEEP) .
84, (2000,5)) ,
//FT03FOO) DD DSN^WYL .XA . Q 11 . HSPF7 . ERRFL , DISP = (OLD, KEEP ) ,
//FTO,
(RECFM=F,BLKSIZE=2000,BUFNO=1 )
DD DSN=WYL ,XA, R72. HSPF. TEMP . I<* . TSGETF, DISP = ( OLD, KEEP) ,
=DISK, VOL =SER=PUBO 10, SPACED (800, (500,5)),
(RECFM=F,BLKSIZE-800,BUFNO=1 )
DD DSN=WYL.XA.R72.HSPF.TEMP.I<».TSPUTF,DISP=(OLD.KEEP).
=DISK,VOL=SER=PUB010,SPACE::(800,(500,5)>,
(RECFM=F,BLKSIZE=SOO,BUFNO=1 )
DD DSN=CSPACFL.DISP=(NEW, DELETE),
=SYSDA,VOL=SER=SCR001 ,SPACE=(36,( 100,5)),
(RECFM=F,BLKSIZE=36,BUFNO=1)
DD DSN=WYL .XA . R72 . TSSFL . 14 , DISP = ( OLD, KEEP ),
=DISK,VOL=SER=PUB010,DCB=(BUFNO=1 )
DD DSN=UIYL .XA . R72 . PLOTFL 1 . IOWA . PEST . C2 , UNIT = DISK,
DISP = ( NEW, KEEP) , VOL = SER = PUBO 1 0 , DCB = ( LRECL=80 , BLKSIZE = 2000 ,
RECFM=FB,BUFNO=1),SPACE=(TRK, (5, 5), RISE)
DD DSN=WYL .XA . R72 . PLOTFL2 . IOWA . PEST . C2 , UNIT = DISK,
DISP=(NEW,KEEP),VOL=SER=PUB010,DCB=
-------
//FT76F001
//FT77F001
//FT78F001
//FT79FOOI
/VFT80F001
//FT81F001
//FT82F001
//FT83F001
//FT05F001
RUN
DD
DD
00
DD
DO
00
DD
DD
00
SYSOUT=A
SYSOUT=A
SYSOUT=A
SYSOUT=A
SYSOUT=A
SYSOUT=A
5YSOUT=A
SYSOUT=A
*
,DCB=(RECFM=FBA
.DCB=
,BLKSIZE=133)
,BLKSIZE=133)
.BtKSIZE=133>
.BLKSIZE=133)
,BLKSIZE=133)
,BLKSIZE=133)
,BLKSIZE=133)
GLOBAL
IOWA RIVER: PERLND/RCHRES (RUNOFF, SEDIMENT. C PESTICIDE) CALIB *2
START 197
-------
INCREASE INPUT DUE TO THAWED GROUND *»*
PERLND 1 1974/03/31 3
PLOWING ***
(UNITS ARE IN/IVU
334 1 0.14
PERLND 1 1974/04/15
INCREASE INFILT FOR TILLAGE
PERLND 1 1974/04/15
DISKING
PERLND 1 1974/05/15
ALACHLOR APPLICATION OF 2.5
(TOTAL RATE DISTRIBUTED OVER
PERLND 1 1974/05/24
PERLND 1 1974/06/03
PERLND 1 1974/06/11
RESET ALACHLOR SURFACE DECAY
PERLND 1 1974/06/20
CULTIVATION
PERLND 1 1974/06/21
CULTIVATION
PERLND 1 1974/07/14
12
12
12
LB/AC
THREE
12
12
12
RATE
12
12
3
3
3
***
###
3
3
3
**#
3
3
3
RESET INFILT TO NOMINAL VALUE
PERLND 1974/08/15
3
REDUCE INFILT FOR FROZEN GROUND
PERLND 1 1974/12/15
PERUND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLNO
PERLND
PERLND
PERLNO
PERLND
PERLNO
PERLND
PERIND
1975/03/31
1975/04/15
1975/04/15
h 1975/05/15
1975/05/24
1975/06/03
1975/06/10
1975/06/20
1975/06/2 1
1975/07/14
1975/08/15
1975/12/15
1976/03/31
976/04/16
976/04/16
976/05/14
976/05/25
976/06/03
976/06/1 1
1976/06/20
1976/06/21
1976/07/14
1976/08/15
1976/12/15
PERLNO 1 1977/03/31
PERLND 1 1977/04/15
PERLND 1 1977/04/15
PERLND 1 1977/05/16
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
977/05/25
1977/06/03
1977/06/10
1977/06/20
1977/06/21
1977/07/14
1977/08/15
1977/12/15
1978/03/31
1978/04/15
1978/04/15
1978/05/15
1978/05/25
1978/06X04
1978/06/10
1978/06/20
1978/06/21
1978/07/14
1978/08/15
1978/12/15
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
918
***
334
#«*
918
APPLIED TO
SEPARATE
2716
2716
2716
2576
**»
918
#-**
918
***
334
*»*
334
334
918
334
918
2716
2716
2716
2576
918
918
334
334
334
918
334
918
2716
2716
2716
2576
918
918
334
334
334
918
334
918
2716
2716
2716
2576
918
918
334
334
334
918
334
918
2716
2716
2716
2576
918
918
334
334
1
1
1.2
0. 18
2.0
SURFACE ADSORBED STORAGE
APPLICATIONS:
2
2
2
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
1
2
2
2
2
2
2
25* 505{ 25%)
0.625
1.25
0.625
0.06
1 .5
1.5
0. 14
0.08
0.14
1.2
0. 18
2.0
0.625
1 .25
0.625
0.06
1.5
1.5
0. 14
0.08
0. 14
1.2
0. 18
2.0
0.625
1.25
0.625
0.06
1.5
1.5
0. 14
0.08
0. 14
1 .2
0. 18
2.0
0.625
1.25
0.625
0.06
1.5
1 .5
0. 14
0.08
0. 14
1 .2
0. 18
2.0
0.625
1.25
0.625
0.06
1.5
1.5
0. 14
0.08
INCREASE INFILT DUE TO THAWED GROUND ***
PERLND 2 1974/03/31 3
(UNITS ARE IN/IVL)
334 1 0.14
-------
DISKING »*«
PERLND 2 1974/04/25 12 3 334 1 0.18
DISKING *»*
PERLND 2 1974/04/25 12 3 918 1 2.0
ALACHLOR APPLICATION OF 2.5 LB/AC »** APPLIED TO SURFACE ADSORBED STORAGE
(TOTAL RATE DISTRIBUTED OVER THREE **» SEPARATE APPLICATIONS: 25% 50% 25%)
PERLND 2 1974/05/01 12
PERLND 2 1974/05/15 12
PERLND 2 1974/05/20 12
RESET ALACHLOR SURFACE DECAY RATE
PERLND 2 1974/05/30
CULTIVATION
PERLND 2
CULTIVATION
PERLND 2 1974/07/01 12
RESET INFILT TO NOMINAL VALUE
PERLND 2 1974/08/15
REDUCE IHFILT FOR FROZEN GROUND
PERLND 2 1974/12/16
1974/06/11 12
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1975/03/31
1975/04/25
1975/04/25
1975/05/01
1975/05/10
1975/05/20
1975/05/30
1975/06/10
1975/07/01
1975/08/15
1975/12/15
12
12
12
12
12
12
12
1976/03/31
1976/04/26 12
1976/04/26 12
1976/05/01 12
1976/05/10 12
1976/05/20 12
1976/05/30
1976/06/11 12
1976/07/01 12
1976/08/15
1976/12/15
1977/03/31
1977/04/25 12
1977/04/25 12
1977/05/01 12
1977/05/10 12
1977/05/19 12
1977/05/29
1977/06/10 12
1977/07/01 12
1977/08/14
1977/12/15
1978/03/31
1978/04/25 12
1978/04/25 12
1978/05/01 12
1978/05/10 12
1978/05/20 12
1978/05/30
1978/06/10 12
1978/07/01 12
1978/08/15
1978/12/15
INCREASE INFILT DUE TO THAUIEP GROUND
PERLND 3 1974/03/31
REDUCE INFILT FOR FROZEN GROUND
PERLND 3
PERLND
PERLND
PERLND
PERLND
1974/12/16
1975/03/31
1975/12/15
1976/03/31
1976/12/15
3
3
3
***
3
3
*»*
3
*ft*
3
*»*
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
ID ***
3
***
3
3
3
3
3
2716 2 0.625
2716 2 1.25
2716 2 0.625
2576 1 0.06
918 1 1.5
918 1 1.5
334 1 0.14
334 1 0.08
334 1 0.14
334 1 0.18
918 1 2.0
2716 ;2 0.625
2716 ,2 1.25
2716 ,2 0.625
2576 1 0.06
918
918
334
334
334
334
918
2716
2716
2716
2576
918
918
334
334
334
334
918
2716
2716
2716
2576
918
918
334
334
334
334
918
2716
2716
2716
2576
918
1.5
1 .5
0. 14
0.08
0. 14
0. 18
2.0
0.625
1.25
0.625
0.06
1.5
1.5
0. 14
0.08
0.14
0.18
2.0
0.625
1 .25
0.625
0.06
1.5
1 .5
0. 14
0.08
0.14
0. 18
2.0
0.625
1 .25
0.625
0. 06
1 .5
918 1 1.5
334 ' 0.14
334 11 0.08
(UNITS ARE IH/IVL)
334 t 0.22
334 t 0.12
334 11 0.22
334 'I 0.12
334 II 0.22
334 11 0.12
PERLND 3
1977/03/31
334
0.22
-------
PERLND 3
1977/12/15
334
0. 12
PERIHD 3
PERLND 3
INCP.EASE INPUT
PERLND 4
PLOWING
PERLND 4
INCREASE INPUT
PERLND 4
DISKING
PERLND 4
1978/03/31
1978/12/15
3
3
DUE TO THAWED GROUND
1974/04/07
1974/04/22 12
FOR TILLAGE
1974/04/22 12
1974/05/22 12
ALACHLOR APPLICATION OF 2.5 LB/AC
(TOTAL RATE DISTRIBUTED OVER THREE
PERLND 4
PERLND 4
PERLND 4
RESET ALACHLOR
PERLND 4
CULTIVATION
PERLND 4
CULTIVATION
PERLND 4
RESET INFILT TO
PERLND 4
1974/05/30 12
1974/06/05 12
1974/06/15 12
SURFACE DECAY RATE
1974/06/25
1974/06/26 12
1974/07/21 12
NOMINAL VALUE
1974/08/20
3
3
3
3
**»
***
3
3
3
*««
3
3
3
3
REDUCE INFILT FOR FROZEN GROUND
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
PERLND 4
1974/12/08
1975/04/07
1975/04/20 12
1975/04/20 12
1975/05/22 12
1975/05/30 12
1975/06/07 12
1975/06/15 12
1975/06/25
1975/06/26 12
1975/07/21 12
1975/08/20
1975/12/07
1976/04/07
1976/04/22 12
1976/04/22 12
1976/05/21 12
1976/05/30 12
1976/06/08 12
1976/06/16 12
1976/06/25
1976/06/26 12
1976/07/21 12
1976/08/20
1976/12/07
1977/04/07
1977/04/22 12
1977/04/22 12
1977/05/22 12
1977/05/30 12
1977/06/07 12
1977/06/15 12
1977/06/25
1977/06/26 12
1977/07/21 12
1977/08/20
1977/12/06
1978/04/08
1978/04/21 12
1978/04/21 12
1978/05/22 12
1978/05/30 12
1978/06/08 12
1978/06/13 12
1978/06/23
1978/06/26 12
1978/07/21 12
1978/08/20
1978/12/06
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
334
334
1
1
0.22
0.12
*** (UNITS ARE IN/IVL)
334
*»*
918
***
334
«**
918
APPLIED TO
SEPARATE
2716
2716
2716
2576
***
918
»*»
918
tttf*
334
***
334
334
918
334
918
2716
2716
2716
2576
918
918
334
334
334
918
334
918
2716
2716
2716
2576
918
918
334
334
334
918
334
918
2716
2716
2716
2576
918
918
334
334
33 >
918
334
918
2716
2716
2716
2576
918
918
334
334
1
1
1
1
0. 16
1 .2
0.22
2.0
SURFACE ADSORBED STORAGE
APPLICATIONS:
2
2
2
1
1
1
1
1
1
1
1
1
2
2
2
1
1
1
1
1
1
1
1
1
2
2
2
1
1
1
1
1
1
1
1
1
2
2
2
1
1
1
1
1
1
1
1
1
2
2
2
1
1
1
1
1
25% 50* 25X)
0.625
1.25
0.625
0.06
1.5
1.5
0.16
0. 10
0. 16
1 .2
0.22
2.0
0.625
1.25
0.625
0.06
1.5
1 .5
0. 16
0. 10
0.16
1.2
0.22
2.0
0.625
1.25
0.625
0.06
1 .5
1 .5
0. 16
0. 10
0. 16
1.2
0.22
2.0
0.625
1 .25
0.625
0.06
1 .5
1 .5
0. 16
0.10
0. 16
1 .2
0.22
2.0
0.625
1.25
0.625
0.06
1 .5
1.5
0. 16
0. 10
-------
INCREASE INPUT DUE TO THAWED GROUND *»* (UNITS-ARE SN/IVL)
PERLND 5
DISKING
PERLND 5
DISKING
PERLND 5
1974/04/07
197
-------
PERLND 6
PERLHD 6
PERLND 6
PERLND 6
PERLND 6
PERLND 6
INCREASE INPUT
PERLND 7
PLOWING
PERLND 7
INCREASE INPUT
PERLND 7
DISKING
PERLND 7
1976/04/05
1976/12/10
1977/04/05
1977/12/10
1978/04/04
1978/12/10
3
3
3
3
3
3
DUE TO THAWED GROUND
1974/04/1-5
1974/04/29 12
FOR TILLAGE
1974/04/29 12
1974/05/29 12
ALACHLOR APPLICATION OF 2.5 LB/AC
(TOTAL RATE DISTRIBUTED OVER THREE
PERLND 7
PERLND 7
PERLND 7
RESET ALACHLOR
PERLND 7
CULTIVATION
PERLND 7
CULTIVATION
PERLND 7
RESET INPUT TO
PERLND 7
1974/06/05 12
1974/06/13 12
1974/06/20 12
SURFACE DECAY RATE
1974/06/30
1974/07/01 12
1974/07/25 12
NOMINAL VALUE
1974/08/25
3
3
3
3
**»
***
3
3
3
«**
3
3
3
3
REDUCE INPUT FOR FROZEN GROUND
PERLND 7
PERLHD 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLHD 7
PERLHD 7
PERLND 7
PERLND 7
PERLHD 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLND 7
PERLHD 7
PERLHD 7
PERLND 7
PERLND 7
1974/12/01
1975/04/16
1975/04/29 12
1975/04/29 12
1975/05/30 12
1975/06/05 12
1975/06/13 12
1975/06/20 12
1975/06/30
1975/07/01 12
1975/07/25 12
1975/08/25
1975/12/02
1976/04/15
1976/04/29 12
1976/04/29 12
1976/05/27 12
1976/06/05 12
1976/06/13 12
1976/06/20 12
1976/06/30
1976/07/01 12
1976/07/25 12
1976/08/25
1976/12/02
1977/04/15
1977/04/29 12
1977/04/29 12
1977/05/30 12
1977/06/05 12
1977/06/13 12
1977/06/20 12
1977/06/30
1977/07/01 12
1977/07/25 12
1977/08/25
1977/12/01
1978/04/15
1978/04/29 12
1978/04/29 12
1978/05/29 12
1978/06/05 12
1978/06/13 12
1978/06/21 12
1978/06/30
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
334
334
334
334
334
334
**» (UNITS ARE
334
***
918
**»
334
***
918
APPLIED TO SURFACE
1
1
1
1
1
1
0.26
0. 14
0.26
0.14
0.26
0. 14
IN/IVL)
1
1
1
0.20
1 .2
0.26
2.0
ADSORBED STORAGE
SEPARATE APPLICATIONS:
2716
2716
2716
2576
***
918
***
918
*»«
334
***
334
334
918
334
918
2716
2716
2716
2576
918
918
334
334
334
918
334
918
2716
2716
2716
2576
918
918
334
334
334
918
334
918
2716
2716
2716
2576
918
918
334
334
334
918
334
918
2716
2716
2716
2576
2
2
2
1
1
1
1
1
1
1
1
1
2
2
2
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
2
2
2
1
25% 50% 25%)
0.625
1.25
0.625
0.06
1.5
1 .5
0.20
0. 12
0.20
1 .2
0.26
2.0
0.625
1.25
0.625
0.06
1 .5
1 .5
0.20
0.12
0.20
1 .2
0.26
2.0
0.625
1.25
0.625
0.06
1.5
1 .5
0.20
0.12
0.20
1.2
0.26
2.0
0.625
1.25
0.625
0.06
1.5
1.5
0.20
0.12
0.20
1.2
0.26
2.0
0.625
1.25
0.625
0.06
-------
PERLND 7
PERLND 7
PERLND 7
PERLND 7
1978/07/0! 12
1978/07/25 12
1978/08/25
1978/12/01
918
918
334
334
1 .5
1 .5
0.20
0. 12
(UNITS ARE IN/IVL)
334 1 0.20
INCREASE INPUT DUE TO THAWED GROUND .***
PERLND 8 1974/04/15 3
DISKING »**
PERLND 8 1974/05/06 12 3 334 I 0.26
DISKING ***
PERLND 8 1974/05/06 12 3 918 1 2.0
ALACHLOR APPLICATION OF 2.5 LB/AC *** APPLIED TO SURFACE ADSORBED STORAGE
(TOTAL RATE DISTRIBUTED OVER THREE *** SEPARATE APPLICATIONS: 25* 50* 252)
PERLND 8 1974/05/11 12
PERLND 8 1974/05/20 12
PERLND 8 1974/05/30 12
RESET ALACHLOR SURFACE DECAY RATE
PERLND 8 1974/06/10
CULTIVATION
PERLND 8 1974/06/20 12
CULTIVATION
PERIND 8 1974/07/15 12
RESET INFILT TO NOMINAL VALUE
PERLND 8 1974/08/25
REDUCE INFILT FOR FROZEN GROUND
PERLND 8 1974/12/01
PERLND 8
PERLND 8
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLNO 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND 8
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
1975/04/13
1975/05/07
1975/05/07
1975/05/09
1975/05/20
1975/05/30
1975/06/10
1975/06/20
1975/07/15
1975/08/25
1975/12/02
12
12
12
12
12
12
12
1976/04/15
1976/05/07 12
1976/05/07 12
1976/05/10 12
1976/05/20 12
1976/05/30 12
1976/06/10
1976/06/20 12
1976/07/15 12
1976/08/25
1976/12/02
1977/04/15
1977/05/07 12
1977/05/07 12
1977/05/10 12
1977/05/19 12
1977/05/30 12
1977/06/10
1977/06/20 12
1977/07/13 12
1977/08/24
1977/12/01
1978/04/15
1978/05/05 12
1978/05/05 12
1978/05/10 12
1978/05/20 12
1978/05/30 12
1978/06/10
1978/06/18 12
1978/07/15 12
1978/08/23
1978/12/01
3
3
3
*«»
3
*«*
3
***
3
*»*
3
**«
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2716
2716
2716
2576
918
918
334
334
334
334
918
2716
2716
2716
2576
918
918
334
334
334
334
918
2716
2716
2716
2576
918
918
334
334
334
334
918
2716
2716
2716
2576
918
918
334
334
334
334
918
2716
2716
2716
2576
918
918
334
334
0.625
1.25
0.625
0.06
1.5
1.5
0.20
0.12
1
1
1
2
2
2
1
1
1
1
1
1
1
1
2
2
2
1
1
1
1
1
1
1
1
2
2
2
1
1
1
1
1
1
1
1
2
2
2
1
1
1
1
1
0.20
0.26
2.0
0.625
1 .25
0.625
0.06
1 .5
1 .5
0.20
0. 12
0.20
0.26
2.0
0.625
1.25
0.625
0.06
1.5
1 .5
0.20
0. 12
0.20
0.26
2.0
0.625
1 .25
0.625
0.06
1 .5
1 .5
0.20
0.12
0.20
0.26
2.0
0.625
1.25
0.625
0.06
1.5
1.5
0.20
0. 12
INCREASE INPUT DUE TO THAWED GROUND »**
PERLND 9 1974/04/10 3
REDUCE INFILT FOR FROZEN GROUND ***
PERLND 9 1974/12/05 3
(UNITS ARE IN/IVL)
334 1 0.30
334
0.16
148
-------
PERLND 9
PERLND 9
PERIND 9
PERLND 9
PERLND 9
PERLND 9
PERLND 9
PERLND 9
END SPEC-ACTIONS
PERLND
1975/04/10
1975/12/05
1976/04/10
1976/12/06
1977/04/10
1977/12/05
1978/04/10
1978/12/05
334
334
334
334
334
334
334
334
0.30
0. 16
0.30
0.16
0.30
0.16
0.30
0. 16
ACTIVITY
ATMP
0
0
0
0
0
0
ACTIVE SECTIONS <1=ACTIVE; 0=INACTIVE) ***
SNOW PWAT SED PST PWG PQAL MSTL PEST NITR PHOS TRAC **»
END ACTIVITY
PRINT-INFO
PRINT FLAGS
I - I ATMP SNOW PWAT SED PST
19 444
END PRINT-INFO
*** v PIVL
PWG PQAL MSTL PEST NITR PHOS TRAC
44
PYR
***
12
GEN-INFO
1
2
3
4
5
6
7
8
9
BEANS
CORN
OTHER
BEANS
CORN
OTHER
BEANS
CORN
OTHER
UNIT SYSTEM
NAME NBLK USER IN OUT ENGL METR
51 0
**«
##*
52
53
54
55
56
57
58
59
END GEN-INFO
SECTION SNOW ***
ICE-FLAG
0= ICE FORMATION NOT SIMULATED;
* - »ICEFG
1 9 1
END ICE-FLAG
1= SIMULATED
***
***
SNOW-PARMt
SNOW INPUT INFO: PART 1
t - * LAT MELEV SHADE
1 42. 925. 0.0
2 42. 925. 0.0
3 42. 925. 0.0
4 42.5 1110. 0.0
5 42.5 1110. 0.0
6 42.5 1110. 0.0
7 43. 1225. 0.0
8 43. 1225. 0.0
9 43. 1225. 0.0
END SNOW-PARM1
SNOW-PARM2
SNOW INPUT INFO: PART 2
I
2
3
4
5
6
7
8
9
- *
RDCSN
0. 12
0. 12
0.12
0. 12
0.12
0. 12
0.12
0. 12
0.12
TSNOW
32.
32.
32.
32.
32.
32.
32.
32.
32.
SNOEVP
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
SNOWCF
1 .45
1 .45
1 .45
1 .45
1 .45
1 .45
1 .45
1 .45
1 .45
CCFACT
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
COVIND
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
MWATER
0.08
0.08
0.08
0.08
***
**»
0.08
0.08
0.08
0.08
0.08
MGMELT
0.0001
0.0001
O.OOOt
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
««*
**«
-------
END SNOW-PARM2
SNOU-INITI
INITIAL SNOW CONDITIONS: PART 1
» - » PACKSNOW PACKICE PACKWATER RDENPF DULL
1 9 4.0 0.0 0.0 0.2 0.0
END SNOW-INIT1
SNOW-INIT2
INITIAL SNOW CONDITIONS: PART 2 ***
t - * COVINX XLNMLT SKYCLR *#*
t 9 0.01 0.0 1.0
END SNOW-INIT2
SECTION PWATER ***
PWAT-PARM1
PWATER VARIABLE MONTHLY PARAMETER VALUE FLAGS **»
» - * CSNO RTOP UZFG VCS VUZ VNN VIFW VIRC VLE ***
1 9 1 0 0 t t 1 0 0 t
END PWAT-PARM1
PAKTMP
32.
***
**#
PWAT-PARM2
#** PWATER INPUT INFO: PART 2 (PART 1 ONLY FLAGS)
**** INPUT INPUT VALUES ARE FOR FROZEN GROUND ***
t - t ***FORE5T LZSN INPUT LSUR
t 0.000 7.0 0.040 300.
2 0.000 7.0 0.040 300.
3 O.OtO 8.0 0.060 300.
4 0.000 7.0 0.050 320.
5 0.000 7.0 0.050 320.
6 0.010 8.0 0.070 320.
7 0.000 8.0 0.060 350.
8 0.000 8.0 0.060 350.
9 0.010 9.0 0.080 350.
END PWAT-PARM2
PWAT-PARM3
*** PWATER INPUT INFO: PART 3
| - * ***PETMAX PETMIN INFEXP INFILD
1 40. 35. 2.0 2.0
2 40. 35. 2.0 2.0
3 40. 35. 2.0 2.0
4 40. 35. 2.0 2.0
5 40. 35. 2.0 2.0
6 40. 35. 2.0 2.0
7 40. 35. 2.0 2.0
8 40. 35. 2.0 2.0
9 40. 35. 2.0 2.0
END PWAT-PARM3
PWAT-PARM4
*** PWATER INPUT INFO: PART 4
« - ft *** CEPSC UZSN NSUR INTFW
t 0.01 0. .0
2 0.01 0. .0
3 O.Ot 0. .2
4 0.01 0. .0
5 0.01 0. .0
6 0.01 0. .2
7 0.01 0. .0
8 O.Ot Q. .0
9 0.01 0. .2
END PWAT-PARM4
MON-INTERCEP
ONLY REQUIRED IF VCSFG=1 IN PWAT-PARM1
* - * INTERCEPTION STORAGE CAPACITY AT START
JAN FEB MAR APR MAY JUN JUL AUG
t 0.03 0.03 0.03 0.03 0.01 0.01 0.08 0.16
2 0.03 0.03 0.03 0.03 O.Ot 0.03 0.10 0.16
3 0.06 0.06 0.06 0.07 0.07 0.08 O.tO O.tO
4 0.03 0.03 0.03 0.03 0.01 0.01 0.08 O.t6
5 0.03 0.03 0.03 0.03 0.01 0.03 0.10 0.16
6 0.06 0.06 0.06 0.07 0.07 0.08 0.10 0.10
7 0.03 0.03 0.03 0.03 O.Ot 0.01 0.08 0.16
8 0.03 0.03 0.03 0.03 O.Ot 0.03 0.10 O.t6
9 0.06 0.06 0.06 0.07 0.07 0.08 O.tO 0.10
SLSUR
0. 050
0.050
0.050
0.020
0.020
0.020
0.010
o.o to
0.010
DEEPFR
0.0
0.0
0.0
0.0
0.0
0.0
. 10
. 10
. 10
IRC
0.60
0.60
0.80
0.60
0.60
0.80
0.60
0.60
0.80
KVARY
0.3
0.3
0.3
0.3
0.3
0.3
0.5
0.5
0.5
BASETP
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
LZETP
OF EACH MONTH
SEP OCT
0.18 0.14
0. 18 0. 14
O.tO O.tO
0. 18 0. 14
0. 18 0. 14
0.10 0.10
0.18 0.14
0. 18 0. 14
0.10 0.10
NOV DEC
0.03 0.03
0.03 0.03
0.07 0.06
0 .03 0.03
0.03 0.03
0.07 0.06
0.03 0.03
0.03 0.03
0.07 0.06
AGWRC
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
AGWETP
0.03
0.03
0.03
0.03
0.03
0.03
0.08
0.08
0.08
**»
***
***
END MON-INTERCEP
150
-------
MON-UZSN
ONLY REQUIRED IF VUZFG=1 IN PUAT-PARM1
1 - * UPPER ZONE STORAGE AT START OF EACH MONTH
1
2
3
4
5
6
7
8
9
JAN FEE MAR APR MAY JUN JUL
0.3 0.3 0.3 0.3 0.9 0.5 0.5
0.3 0.3 0.3 0.3 0.8 0. ONLY REQUIRED IF VNNFG=1 IN PWAT-PARM1
1 MANNING'S N FOR OVERLAND FLOW AT
JAN FEB MAR APR MAY JUN JUL
0.25 0.25 0.25 0.25 0.25 0.15 0.15
START
AUG
0.20
0.25 0.25 0.25 0.25 0.25 0.15 0.15,0.20
0.30 0.30 0.30 0.30 0.30 0.30 0.30
0.25 0.25 0.25 0.25 0.25 0.15 0.15
0.25 0.25 0.25 0.25 0.25 0.15 0.15
0.30 0.30 0.30 0.30 0.30 0.30 0.30
0.25 0.25 0.25 0.25 0.25 0.15 0.15
0.25 0.25 0.25 0.25 0.25 0.15 0.15
0.30 0.30 0.30 0.30 0.30 0.30 0.30
0.30
0.20
0.20
0.30
0.20
0.20
0.30
OF EACH MONTH
0
0
0
0
0
0
0
0
0
SEP
.22
.22
.30
.22
.22
.30
.22
.22
.30
0
0
0
0
0
0
0
0
0
OCT
.25
.25
.30
.25
.25
.30
.25
.25
.30
0
0
0
0
0
0
0
0
0
NOV
.25
.25
.30
.25
.25
.30
.25
.25
.30
0
0
0
0
0
0
0
0
0
DEC
.25
.25
.30
.25
.25
.30
.25
.25
.30
##*
#«*
***
END NON-MANNING
MON-LZETPARM
ONLY REQUIRED IF VLEFG=1 IN PUAT-PARM1
t LOWER ZONE ET PARAMETER AT START
JAN FEB MAR APR MAY JUN JUL
0.20 0.20 0.20 0.23 0.23 0.25 0.60
0.20 0.20 0.20 0.23 0.23 0.25 0.60
0.25 0.25 0.25 0.25 0.30 0.35 0.40
0.20 0.20 0.20 0.23 0.23 0.25 0.60
0.20 0.20 0.20 0.23 0.23 0.25 0.60
0.25 0.25 0.25 0.25 0.30 0.35 0.40
0.20 0.20 0.20 0.23 0.23 0.25 0.60
0.20 0.20 0.20 0.23 0.23 0.25 0.60
0.25 0.25 0.25 0.25 0.30 0.35 0.40
OF EACH
AUG
0.80
0.80
0.40
0.80
0.80
0.40
0.80
0.80
0.40
0
0
0
0
0
0
0
0
0
MONTH
SEP
.75
.75
.45
.75
.75
.45
.75
.75
.45
0
0
0
0
0
0
0
0
0
OCT
.50
.50
.35
.50
.50
.35
.50
.50
.35
0
0
0
0
0
0
0
0
0
NOV
.30
.30
.30
.30
.30
.30
.30
.30
.30
0
0
0
0
0
0
0
0
0
DEC
.20
.20
.25
.20
.20
.25
.20
.20
.25
***
*##
***
END MON-LZETPARM
PUAT-5TATEI
*#* INITIAL CONDITIONS AT START OF
« #** CEPS SURS UZS
0.0 0.0 0.8
0.0 0.0 0.8
0.0 0.0 2.0
0.0 0.0 0.5
0.0 0.0 0.5
0.0 0.0 1.0
0.0 0.0 0.5
0.0 0.0 0.5
0.0 0.0 1.0
SIMULATION
IFUS
0.0
0.0
0.0
0 . 0
0.0
0.0
0.0
0.0
0.0
LZS
8.0
8.0
9.0
8.0
8.0
9.0
7.5
7.5
8.5
AGWS
0
0
0
0
0
0
0
.45
.45
.50
.30
.30
0.4
.20
.20
0.4
GWVS
0.9
0.9
t .0
0.6
0.6
0.8
0.4
0.4
0.8
END PWAT-STATE1
SECTION SEDMNT
***
SED-PARMI
***
* - * CRV VSIV SDOP **»
19101
END SED-PARMt
SED-PARM2
***
I - t
1
2
3
4
5
6
7
8
9
END SED-PARM2
SMPF
KRER
.45
.45
.40
.45
.45
.40
.40
.40
.35
JRER
2.2
2.2
2.2
2.2
2.2
2.2
2.2
2.2
2.2
AFFIX
.030
.030
.003
.030
.030
.003
.030
.030
.003
COVER
1 .0
1 .0
1.0
1.0
1 .0
1 .0
t .0
1.0
1.0
NVSI ***
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
151
-------
SED-PARM3
***
» - * KSER JSER KGER JGER
1 3.0 2.2 0.0 1.0
2 2.0 2.0 0.0 1.0
3 1.0 2.0 0.0 1.0
4 3.0 2.2 0.0 1.0
5 2.0 2.0 0.0 1.0
6 1.0 2.0 0.0 1.0
7 2.5 2.2 0.0 1.0
8 1.8 2.0 0.0 1.0
9 0.5 2.0 0.0 1.0
END SED-PARM3
(ION-COVER
MONTHLY VALUES FOR EROSION-RELATED LAND COVER **#
SEP OCT NOV
77 .61 .26
62 .51 .38
90 .90 .90
77 .61 .26
62 .51 .38
90 .90 .90
77 .61 .26
62 .51 .38
.90 .90 .90
BLK5 ***
* - It
1
2
3
4
5
6
7
8
9
JAN
. 17
.25
.90
. 17
.25
.90
. 17
.25
.90
FEB
. 13
.22
.90
.13
.22
.90
.13
.22
.90
MAR
.09
.20
.90
.09
.20
.90
.09
.20
.90
APR
.06
. 18
.90
.06
. 18
.90
.06
. 18
.90
MAY
.01
.03
.90
.01
.03
.90
.01
.03
.90
JUN
.03
.08
.90
.03
.08
.90
.03
.08
.90
JUL
.43
.40
.90
.43
.40
.90
.43
.40
.90
AUG :
.67
.70
.90
.67
.70
.90
.67
.70
.90
END MON-COVER
SED-STOR
* - *
1
2
3
4
5
6
7
8
9
DETACHED
SEDIMENT
BLOCK!
0.2
0.2
0.1
0.2
0.2
0.1
0.2
0.2
0.1
BLK2
STORAGE
TONS/ACRE ***
BLK3
BLK4
DEC **«
.21
.29
.90
.21
.29
.90
.21
.29
.90
END SED-STOR
SECTION MSTLfcY ***
MST-PARM
* - #
1
2
3
4
5
6
7
8
9
END MST-PARM
MST-TOPSTOR
MST-TOPFLX
SLMPF
0.7
0.7
0.5
0.7
0.7
0.5
0.7
0.7
0.5
INI
INI
ULPF
5.0
5.0
.0
.0
.0
.0
.0
. 0
5.0
LLPF
.5
.5
.5
.5
.5
.5
.5
.5
.5
#**
***
INITIAL MOISTURE STORAGES DEFAULTED TO ZERO ***
INITIAL MOISTURE FLUXES DEFAULTED TO ZERO ***
SECTION PEST ***
PEST-FLAGS
OPTIONS FOR SIMULATION OF UP TO 3 DIFFERENT PESTICIDES ***
* - * NPST MAX ITERATIONS ADSORP OPTION ***
PST1 PST2 PST3 PST1 PST2 P5T3 **»
1 2 1 20 2
4 5 1 20 2
7 8 1 20 2
EHD PEST-FLAGS
SOIL-DATA
SOIL LAYER DEPTHS AND BULK DENSITIES **»
- * DEPTHS (IN) BULK DENSITY (LB/FT3) ***
SURFACE UPPER LOWER GROUNDW SURFACE UPPER LOWER GROUNDU *««
1 2 0.25 5.71 41.30 60. 62.4 79.2 81.7 85.5
4 5 0.25 5.71 41.30 60. 62.4 79.2 81.7 85.5
7 8 0.25 5.71 41.30 60. 62.4 79.2 81.7 85.5
152
-------
END SOIL-DATA
**# PESTICIDE NO. I - ALACHLOR *»*
PEST-ID
**«
» - * PESTICIDE NAME ***
1 2 ALACHLOR
4 5 ALACHLOR
7 8 ALACHLOR
END PEST-ID
PEST-CMAX
ONLY USED IF ADOPFG=2 OR 3 IN PEST-FLAGS ***
» - » CMAX ***
(PPM) ***
I 2 242.
4 5 242.
7 8 242.
END PEST-CMAX
PEST-SVALPM SURFACE LAYER
ONLY USED IF ADOPFG=2 (SINGLE VALUE FREUNDLICH) IN PEST-FLAGS ***
I - * XFIX K1 Nl ***
(PPM) ***
1 2 0.0 4. 1.4
4 5 0.0 4. 1.4
78 0.0 4. 1.4
END PEST-SVALPM
PEST-SVALPM UPPER LAYER
ONLY USED IF ADOPFG=2 (SINGLE VALUE FREUNDLICH) IN PEST-FLAGS *»*
t - * XFIX K1 Nl ***
(PPM) ***
12 0.0 4. 1.4
4 5 0.0 4. 1.4
78 0.0 4. 1.4
END PEST-SVALPM
PEST-SVALPM LOWER LAYER
ONLY USED IF ADOPFG=2 (SINGLE VALUE FREUNDLICH) IN PEST-FLAGS ***
» - » XFIX K1 Nl ***
(PPM) ***
1 2 0.0 3. 1.4
45 0.0 3. 1.4
7 8 0.0 3. 1.4
END PEST-SVALPM
PEST-SVALPM GROUNDWATER LAYER
ONLY USED IF ADOPFG=2 (SINGLE VALUE FREUNDLICH) IN PEST-FLAGS **»
I - » XFIX Kl N1 ***
(PPM) ***
I 2 0.0 3. 1.4
45 0.0 3. 1.4
7 8 0.0 3. 1.4
END PEST-SVALPM
PEST-DEGRAD
PESTICIDE DEGRADATION RATES (PER DAY) «#*
i - t SURFACE UPPER LOWER GROUNDW ***
t 2 0.120 0.045 0.04 0.04
4 5 0.120 0.045 0.04 0.04
7 8 0.120 0.045 0.04 0.04
END PEST-DEGRAD
PEST-STOR1
INITIAL PESTICIDE STORAGE IN SURFACE LAYER (LB/AC) ***
* - » CRYSTAL ADSORBED SOLUTION ***
1 2 0.0 0.0 0.0
4 5 0.0 0.0 0.0
7 8 0.0 0.0 0.0
END PEST-STOR1
PEST-STOR1
INITIAL PESTICIDE STORAGE IN UPPER LAYER (LB/AC) **»
t - t CRYSTAL ADSORBED SOLUTION «**
1 2 0.0 0.0 0.0
4 5 0.0 0.0 0.0
7 8 0.0 0.0 0.0
END PEST-STOR1
-------
PEST-STORI
LOWER LAYER STORAGE
» - » CRYSTAL ADSORBED SOLUTION
1 2 0.0 0.0 0.0
4 5 0.0 0.0 0.0
7 8 0.0 0.0 0.0
END PEST-STORI
PEST-STORt
* - It
t 2
4 s
7 8
END PEST-STORt
GROUNDWATER STORAGE OF PESTICIDE
CRYSTAL ADSORBED SOLUTION
0.0 0.0 0.0
0.0 0.0 0.0
0.0 0.0 0.0
*«»
***
*«*
***
END PERLND
RCHRES
ACTIVITY
RCHRES ACTIVE SECTIONS (1=ACTIVE; 0=INACTIVE) *»*
» - I HYFG ADFG CNFG HTFG SDFG GQFG OXFG NUFG PKFG PHFG
113)100110000
END ACTIVITY
PRINT-INFO
RCHRES PRINTOUT LEVEL FLAGS
t - * HYDR ADCA CONS HEAT
1
2
7
8
13
12
END PRINT-INFO
GEN-INFO
RCHRESNEXIT< UNIT SYSTEM XPRINTER > ***
UCI IN OUT ENGL METR IKFG ***
LEN
(MI)
10.8
10.1
12.5
16.2
18.5
15. 1
10.4
13.9
17.6
17.6
17.9
16.4
DELTA H
14.8
16.0
21 . 1
25.7
27.3
26.3
25.8
32.3
51.1
64. I
62.4
26.8
71
72
73
74
75
76
77
78
79
80
81
82
83
***
00000
DATUM H
KS «** DB50
0.5 0.014
0.5 0.014
0.5 0.014
0.5 0.014
0.5 0.014
0.5 0.014
0.5 0.014
0.5 0.014
0.5 0.014
0.5 0.014
0.5 0.014
0.5 0.014
-------
13
END HYDR-PARM2
13
9.3
12.8
0.5
0.014
HYDR-INIT
RCHRES INITIAL CONDITIONS FOR HYDR *#»
*#* 1/1/74 FLOW: MARENGO 1000 CFS
* - »VOL(AC-FT) PAIR OF COLS FOR F(VOL)
EX I EX2 EX3 EX4 EX5
1
2
3
4
5
6
7
8
9
10
1 1
12
13
END
1016.
716.
776.
974.
1066.
783.
397.
472.
432.
376.
347.
370.
206.
HYDR-INIT
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
INITIAL GCT) COMPONENT
EX1 EX2 EX3 EX* EX5
««*
***
SECTION SEDTRN ***
SANDFG
RCHRES ***
< » SDFG #**
1 13 1
END SANDFG
SED-GENPARM
RCHRES
* ft
1
2
3
4
5
6
7
8
9
10
11
12
13
BEDWID
(ft)
150.
140.
130.
125.
110.
110.
100.
95.
95.
90.
84.
85.
85.
BEDUIRN
(ft)
15.
15.
15.
15.
15.
15.
15.
)5 .
10.
10.
10.
10.
to.
END SED-GENPARM
SAND-PM
RCHRES
tt ft
1 13
END SAND-PM
D
( in)
.014
W
( in/sec)
2.5
SILT PARAMETERS ***
SILT-CLAY-PM
RCHRES D Ul
* * (in) (in/sec)
1 13 .00063 .0066
END SILT-CLAY-PM
CLAY PARAMETERS ***
SILT-CLAY-PM
RCHRES D W
« * (in) (in/sec)
1 13 .000055 .000034
END SILT-CLAY-PM
POR ***
***
RHO
2.65
KSAND
RHO TAUCD
(Ib/ft2)
2.2 0.05
RHO TAUCD
(Ib/ft2)
2.0 0.04
EXPSND #**
«**
TAUCS M ***
Ub/ft2) (Ib/ft2d) ***
0.15 3.0
TAUCS M **»
Ub/ft2) Ub/ft2d) ***
0.12 6.5
SSED-INIT
RCHRES Suspended sed cones (mg/1) «*»
» t SAND SILT CLAY «**
1 13 0.0 16. 24.
END SSED-INIT
BED-INIT
RCHRES
» I
1
2
BEDDEP Initial bed composition ««»
(ft) Sand Silt Clay ***
10. 0.60 0.20 0.20
9. 0.60 0.20 0.20
155
-------
5
6
7
8
9
10
1 1
12
13
END
BED-INIT
0.60
0.50
0.50
0.50
0.50
0.50
0.50
0 .50
0.50
0.50
0.50
0.20
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.20
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
SECTION GQUAL ***
GQ-GENDATA
RCHRES GQUAL General Info
* » NQL TPFG PHFG ROFG CDFG 5DFG PYFG
I 13 1 1
END GQ-GENDATA
***
LAT ***
9UAL
- ALACHLOR »**
GQ-QALDATA
RCHRES ***
* t *#*
1 13
END GQ-QALDATA
GQID DQAL(mg)
ALACHLOR 0.0
CONCID CONV
mg I.6017E+*
QTYID
Ib
GQ-QALFG
RCHRES First set of flags for a qual ***
* » HDRL OXID PHOT VOLT BIOD GEN SDAS ***
1 13 11
END GQ-QALFG
GQ-GENDECAY
RCHRES FSTDEC
I * (/day)
1 '13 0.080
END GQ-GENDECAY
GQ-SEDDECAY
RCHRES KSUSP
* * (/day)
1 13 0.100
END GQ-SEDDECAY
THFST ***
**#
1 .07
THSUSP
1 .07
KBED
(/day)
0. 120
THBED »«*
***
1 .07
GQ-KD
RCHRES
« «
1 13
END GQ-KD
Partition coefficients (1/mg)
ADPM1 ADPM2 ADPM3 ADPM4
2.0E-6 1.0E-5 5.0E-5 1.0E-5
GQ-ADRATE
RCHRES Ads/Des rate parameters (/day)
f « ADPM1 ADPM2 ADPM3 ADPM4
1 13 8.0 8.0 8.0 .03
END GQ-ADRATE
ADPM5
5.0E-5
ADPM5
.03
6Q-ADTHETA
RCHRES Ads/Des temperature correction parameters!
* * ADPM1 ADPM2 ADPM3 ADPM«i ADPM5
1 13 1.0 1.0 1.0 1.0 1.0
END GQ-ADTHETA
GQ-SEDCONC
RCHRES Initial concentrations on sediments (mg/mg)
t * SQAL1 SQAL2 SQAL3 SQAL4 SQAL5
1 13 0.0 0.0 0.0 0.0 0.0
END GQ-SEDCONC
**«
ADPM6 ***
1 .OE-4
#*#
ADPM6 ***
.03
***
ADPM6 ***
1 .0
**«
SQAL6 ***
0.0
END RCHRES
FTABLES
FTABLE
156
-------
ROWS COLS
17 4
DEPTH
(FT)
0.0
4.0
5.0
6.0
7.0
8.0
9.0
to.o
11.0
12.0
13.0
14.0
15.0
16.0
17.0
18.0
19.0
END FTABLE
FTABLE
AREA VOLUME DISCH
(ACRES) (ACRE-FT) (CFS)
0.0 0.0 0.0
250.0 579.9 270.0
281.5 848.3 710.0
294.5 1 137.6 1210.0
302.4 1437 .4 1721.0
307.6 1741.1 2258.0
314.2 2048.7 2840.0
322.0 2366.8 3490.0
360.0 2700.7 4249.0
435. 9 3101.2 5120.0
455.6 3543.7 6300.0
484.4 4296.4 8206.0
517.1 5364.6 10900 .0
549.8 6837.4 14730.0
589.1 8960.7 20330.0
628.4 11497.7 27490.0
667.6 12188.9 36500.0
1
2
**tt
FLO-THRU***
OIRS>**»
0 . 0
26 . 0
14.4
11.4
10.1
9.3
8.8
8.2
7 .7
7.3
6.8
6.3
6.0
5.6
5.3
5. 1
4.0
ROWS COLS ***
13 4
DEPTH
(FT)
.000
1 .517
3.033
4.550
6.067
7.583
9.100
12.133
15. 167
18.200
24.267
30.333
36.400
END FTABLE
FTABLE
AREA VOLUME DISCH
(ACRES) (AC-FT) (CFS)
.000 .0000 .000
184.248 267.8384 233.5109
199.551 558.8862 753.5811
214.854 873.1438 1507.3840
230.158 1210.6110 2479.9080
245.461 1571.2870 3666.0350
260.763 1955. 1730 5065. 1 170
291.370 2792.5750 8510.6010
321.975 3722.815012844.6200
352.581 4745.890018105.8600
847.719 8386.800034735.1900
1342. 857 15031. 54005 96 49. 6300
1837.99524680 .120095395. 1800
2
3
FLO-THRU ***
(MIN) #**
0.0
832.7
538.4
420.5
354.4
311.2
280.2
238.2
210.4
190.3
175.3
182.9
187.8
ROWS COLS «**
13 4
DEPTH
(FT)
.000
1 .433
2.867
4.300
5.733
7. 167
8.600
1 1 .467
14.333
17.200
22.933
28.667
34.400
END FTABLE
FTABLE
AREA VOLUME DISCH
(ACRES) (AC-FT) (CFS)
.000 .0000 .000
210.732 288.9285 198.0271
229.040 604.0977 639.6987
247.348 945.5090 1280.9470
265.656 1313. 1630 2109.7010
283.964 1707.0570 3122.2200
302.272 2127. 1930 4318.51 10
338.889 3046.1910 7271.6710
375.505 4070.156010996.9500
412.120 5199.078015530.5400
991.242 9222.035029932.8100
1570.36616565.300051548.8900
2149.48927228.850082592.3700
3
4
FLO-THRU ***
(MIN) *#*
0.0
1059.3
685.6
535.9
451 .9
396.9
357.6
304. t
268.7
243.0
223.7
233.3
239.3
ROWS COLS #*»
13 4
DEPTH
(FT)
.000
1 .383
2.767
4. 150
5.533
6.917
8.300
1 1 .067
13.833
16.600
22.133
27.667
AREA VOLUME DISCH
(ACRES) (AC-FT) (CFS)
.000 .0000 .000
262.472 347.2415 191.2105
285.382 726.1736 617.7407
308.291 1136.7960 1237.1120
331 .200 1579. 1 1 10 2037.7360
354.109 2053.1160 3016.0610
377.018 2558.8100 4172.1520
422.836 3665.2770 7026.7810
468.654 4898.503010628.8100
514.472 6258.488015013.5400
1238.83211 109.270028952.0300
1963. 19519968.200049882.5400
FLO-THRU «**
(MIN) ***
0.0
1318.4
853.4
667. 1
562.6
494.2
445.3
378.7
334.6
302.6
278.6
290.6
157
-------
33.200 2687.55732835.260079952.6200
END FTABLE 4
298.2
FTABIE
ROWS COLS
15 4
DEPTH
(FT)
0.0
1.0
1.8
2.3
3.5
4.2
5.5
7.3
10.0
12.2
13.9
15.*
18.1
20.5
23.0
END FTABLE
FTABLE
ROMS COLS
16 4
DEPTH
(FT)
0.0
1.0
1.7
2.3
3.3
4. 1
5.3
6.3
7.1
9.9
12.0
13.7
15.2
16.6
18.0
19.4
END FTABLE
FTABLE
ROWS COLS
11 4
DEPTH
(FT)
0.0
1.0
2.0
4.0
6.0
8.0
10.0
12.0
13.0
14.0
33.3
END FTABLE
FTABLE
ROMS COLS
19 4
DEPTH
(FT)
0.0
0.6
1.4
2.3
2.9
4.2
5.2
6. 1
6.9
5
AREA
(ACRES)
0.0
103.2
188.4
246.7
347.6
374.5
405.9
430.5
468.7
491.1
515.8
531.5
531.5
531.5
531.5
5
6
AREA
(ACRES)
0.0
82.4
150. 1
195.8
265.4
289.2
305.7
316.6
327.6
353.2
375.2
395.3
395.3
395.3
395.3
395.3
6
7
AREA
(ACRES)
0.0
181 .5
277.3
332.8
385.7
534.5
557.2
584.9
611.4
642.9
642.9
7
8
AREA
(ACRES)
0.0
160. 1
203.9
217.3
227.5
246.0
259.5
271.3
283. 1
VOLUME
(ACRE-FT)
0.0
51.6
170.4
287.0
634.6
888.0
1390.3
2159.5
3374.8
4408.6
5274.2
6070.2
7491.9
8902.4
10090.0
VOLUME
(ACRE-FT)
0.0
38.4
130.0
219.6
479.5
675.4
1046.9
1352.6
1625.3
2549.6
3316.5
3966.3
4564.8
51 17.5
5673.9
6223.0
VOLUME
(ACRE-FT)
0.0
110.9
344. 1
934. 1
1651 .4
2622.0
3706.2
4853.3
5798.8
7563.6
20000.
VOLUME
(ACRE-FT)
0.0
45.5
195.4
385.8
534. 1
837.4
1091 .8
1320.9
1536.6
DISCH
(CFS)
0.0
10.0
50.0
100.0
300.0
500.0
1000.0
2000.0
4000.0
6000.0
8000.0
10000.0
14000.0
18000.0
22000.0
DISCH
(CFS)
0.0
10.0
50.0
100.0
300.0
500.0
1000.0
1500.0
2000.0
4000.0
6000.0
8000.0
10000.0
14000.0
18000.0
22000.0
DISCH
(CFS)
0.0
163.8
460.7
1389.0
2660.0
4197.0
67 18.0
10880.0
14860.0
20460.0
65000.0
DISCH
(CFS)
0.0
10.0
100.0
300.0
500.0
1000.0
1500.0
2000.0
2500.0
***
FLO-THRU***
(HRS)»»*
0.0
62.5
41.3
34.8
25.7
21 .5
16.9
13. 1
10.2
8.9
8.0
7.4
6.5
6.0
5.6
t*»
FLO-THRU***
(HRS)***
0.0
46.6
31 .5
26.6
19.4
16.4
12.7
10.9
9.9
7.7
6.7
6.0
5.5
4.4
3.8
3.4
«**
FLO-THRU***
(HRS)***
0.0
8.2
9.0
8.2
7.5
7.6
6.7
5.4
4.7
4.5
3.7
#**
FLO-THRU***
(HRS)***
0.0
55.2
23.7
15.6
13.0
10.2
8.8
8.0
7.5
158
-------
7.6
8.8
10.8
12.6
13.9
15.2
16.3
17. 4
18.5
42.8
END FTABLE
FTABLE
ROWS COLS
19 4
DEPTH
(FT)
0.0
0.5
1 .3
2.2
2.8
4.0
4.9
5.7
7. t
8.3
9.3
10.3
11.1
11.9
12.6
13.2
13.8
14.4
34.5
END FTABLE
FTABLE
293.2
310.0
350 .4
382.5
384. 1
384. 1
384 . 1
384. 1
384. 1
384. 1
8
9
AREA
(ACRES)
0.0
192.0
258.
273.
283.7
305.
322.
337.
362.7
384.0
405.3
428.8
448.0
467.2
486.4
486.4
486.4
486.4
486.4
9
10
1737.1
2114.5
2783,4
3410.0
3930.8
4407 . 6
4852.4
5305.7
5676.3
15000.
VOLUME
(ACRE-FT)
0.0
51.2
230 .4
456.5
631.5
983.5
1284.3
1555.2
2039.5
2483.2
2888.5
3266. 1
3630.9
3989.3
4339.2
4531 .2
4925.9
5203.2
15000.
3000.0
4000.0
6000.0
8000.0
10000.0
12000.0
14000.0
16000 .0
18000.0
52000.0
DISCH
(CFS)
0.0
10.0
100.0
300 .0
500.0
1000.0
1500.0
2000 .0
3000.0
4000.0
5000 .0
6000.0
7000.0
8000 .0
9000.0
10000 .0
1 1000.0
12000.0
45000.0
7.0
6.4
5.6
5.2
4.8
4.5
4.2
4.0
3.8
3.5
**»
FLO-THRU***
(HRS)**»
0.0
62.1
27.9
18.5
15.3
11.9
10.4
9.4
8.2
7.2
7.0
6.6
6.3
6.0
5.8
5.5
5.4
5.3
4.0
ROWS COLS ***
13 4
DEPTH
(FT)
. 000
0.967
1 .933
2.900
3.867
4.833
5.800
7.733
9.667
1 1 .600
15.467
19.333
23.200
END FTABLE
FTABLE
AREA
(ACRES)
.000
221 . 156
241 .778
262.400
283.022
303.644
324.267
365.51 1
406.755
448.000
997.925
1547.8501
VOLUME
(AC-FT)
.0000
203.8163
427.5671
671 .2532
934.8733
1218.4290
1521 .9190
2188.7030
2935.2270
376 1 .4890
DISCH
(CFS)
.000
105.3339
340.9438
684. 1250
1 129.0880
1674.4200
2320.6460
3922.5970
5953.2500
8435. 0070
6556.945016461 .7100
1478.770028863.3700
20 97. 774 18526. 940047091. 1000
10
1 1
FLO-THRU
(MIN)
0.0
1404.8
910.5
712.3
601.1
523.3
476, 1
405.1
358.0
323.8
289.2
288.7
285.6
***
**»
ROWS COLS **#
13 4
DEPTH
(FT)
.000
0.850
1 .700
2.550
3.400
4.250
5. 100
6.800
8.500
10.200
13.600
17 .000
20.400
AREA
(ACRES)
.000
196.357
214.800
233.242
251 .685
270. 127
288.569
325.454
362.339
399.224
891 .021
1382.817
VOLUME
(AC-FT)
. 0000
159.0658
333.8074
524.2251
730.3184
952.0884
1 189.5340
171 1 .4540
2296.0790
2943.4050
5136.8200
DISCH
(CFS)
.000
60. 7026
196 .5158
394.3945
651 . 0386
965.6660
1338.6120
2263.4980
3436.4480
4870.5540
9513.9370
9002.332016694.9700
1874.61314539.940027256.7500
FLO-THRU
(MIN)
0.0
1902.4
1233.2
965.0
814.4
715.8
645. 1
548.9
485. 1
438.7
392.0
391 .5
387.3
***
*«*
END FTABLE 11
FTABLE 12
ROWS COLS ***
159
-------
13 4
DEPTH AREA VOLUME DISCH
(FT) (ACRES) (AC-FT) (CFS)
.000 .000 .0000 .000
0.875 186.364 155.4583 43.8048
1.750 203.757 326.1360 141.7904
2.625 221.151 512.0332 284.5178
3.500 238.545 713.1506 469.5835
4.375 255.939 929.4878 696.4016
5.250 273.333 1161.0440 965.1946
7.000 308.121 1669.8160 1631.5490
8.750 342.909 2239.4680 2476.2790
10.500 377.697 2869.9980 3508.7200
14.000 841.534 5003.6480 6848.3780
17.500 1305.372 8760.738012008.9400
21. 000 1769. 21014141. 2500 195 94. 5400
END FTABLE 12
FTABLE 13
ROWS COLS
14 4
DEPTH AREA VOLUME DISCH
(FT) (ACRES) (ACRE-FT) (CFS)
0.0 0.0 0.0 0.0
0.5 94.7 36.4 11.8
1.0 112.7 89.4 48.8
2.0 120.6 205.2 152.0
3.0 130.8 331.4 308.0
4.0 142.0 466.7 504.0
5.0 151.0 613.2 744.0
6.0 160.1 768.8 1020.0
6.5 164.6 850.0 1220.0
7.5 174.7 1090.1 1750.0
8.5 186.0 1454.2 2580.0
9.5 197.3 2014.4 3860.0
10.5 214.2 2660.4 5500.0
68.1 214.2 15000. 45000.
END FTABLE 13
END FTABLES
DISPLY
DISPLY-INF01
» - « *#* TITLE
1 FLOW (IN) MARENGO (SIM)
3 SED LD (LB/AC)MARENGO(SIM)
5 FLOW (CFS) ROWAN (SIM)
7 FLOW (CFS) MARSHLTWN (SIM)
9 FLOW (CFS) MARENGO (SIM)
11 SED WSHFF PLS1-BEANS(LB/AC)
12 SED WSHFF PLS2-CORN( LB/AC)
13 SED WSHFF PLS3-PAST. (LB/AC)
14 SED WSHFF PLS4-BEANS(LB/AC)
15 SED WSHFF PLS5-CORN( LB/AC)
16 SED WSHFF PLS6-PAST .( LB/AC)
17 SED WSHFF PLS7-BEANS( LB/AC)
18 SED WSHFF PLS8-CORN( LB/AC)
19 SED WSHFF PLS9-PAST .( LB/AC)
20 SOL ALAC CONC(MG/L) MARSHLT
21 SOL ALAC LOAD(LB/AC)MARSHLT
22 SED ALAC CONC(PPM) MARSHLT
23 SED ALAC LOAD( LB/AOMARSHLT
24 SOL ALAC CONC(MG/L) MARENGO
25 SOL ALAC LOAD(LB/AC)MARENGO
26 SED ALAC CONC(PPM) MARENGO
27 SED ALAC LOAD( LB/AOMARENGO
END DISPLY-INF01
END DISPLY
PLTGEN
PLOTINFO
Hthrut FILE NPT NMN Labi PYR PIVL
1 31 2 12
2 32 2 12
3 33 2 12
4 34 2 12
5 35 2 12
6 36 2 12
FLO-THRU *»*
(MIN) ***
0.0
2576.5
1669.9
1306.5
1 102.6
969.0
873.3
743.0
656.6
593.8
530.4
529.6
523.9
«**
FLO-THRU***
(HRS)**»
0.0
36.9
22.0
16.2
13.0
11.2
10.0
9. 1
8.4
7.5
6.8
6.3
5.9
4.0
TRAN PIVL DIG1 FIL1 PYR DIG2 FIL2 YEND
SUM
SUM
AVER
AVER
AVER
SUM
SUM
SUM
SUM
SUM
SUM
SUM
SUM
SUM
AVER
SUM
AVER
SUM
AVER
SUM
AVER
SUM
*«*
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
4
2
1
1
1
4
4
4
4
4
4
4
4
4
5
4
5
4
5
4
5
4
61
62
63
64
65
66
66
66
66
66
66
66
66
66
67
68
67
68
69
70
69
70
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
160
-------
7 37 2
8 38 2
END PtOTINFO
GEN-LABELS
MAL m j> Y Y T 1 C
1 FLOW: ROWAN
2 FLOW: MARSHALLTOWN
3 FLOW: MARENGO
4 SED LOAD: MARENGO
5 SOLN ALAC CONC
6 SED ALAC CONC
7 SOLN ALAC LOAD
8 SED ALAC LOAD
END GEN-LABELS
SCALING
dthru* YMIN YMAX
1 0. 1500.
2 0. 5000.
3 0. 15000.
4 0. 200.
5 0.0.1
6 0. "0.5
7 0. 0.0005
8 0. 0.00005
END SCALING
CURV-DATA (first curve)
*thrul < Curve label > ***
1 SIMULATED 10
2 SIMULATED 10
3 SIMULATED 10
4 SIMULATED 10
5 MARSHALLTOWN 10
6 MARSHALLTOWN 10
7 MARSHALLTOWN 10
8 MARSHALLTOWN 10
END CURV-DATA
CURV-DATA (second curve)
Mhru# < Curve label > **#
1 OBSERVED 10
2 OBSERVED 10
3 OBSERVED 10
4 OBSERVED 10
5 MARENGO 10
6 MARENGO 10
7 MARENGO 10
8 MARENGO 10
END CURV-DATA
END PLTGEN
GENER
OPCODE
* TO » OP- ***
1 2 19
END OPCODE
END GENER
EXT SOURCES
12
12
N^
CFS
CFS
CFS
LB/AC
MG/L
PPM
LB/AC
LB/AC
IVLIN ***
20.
20.
20.
20.
20.
20.
20.
20.
8
8
8
8
8
8
8
8
<-VOLUME->
* »
TSS 39 PRECIP ENGLZERO 1.03
TSS 131 PRECIP ENGLZERO 0.98
TSS 132 PRECIP ENGLZERO 0.95
TSS 121 ARTEMP ENGL 0.98
TSS 123 ARTEMP ENGL 0.92
TSS 122 ARTEMP ENGL 0.88
TSS 41 EVAPOR ENGL 0.7
TSS 42 WINDXX ENGL
TSS 46 SOLRAD ENGL
TSS 124 DEWPNT ENGL
TSS 126 DEWPNT ENGL
TSS 125 DEWPNT ENGL
TSS 171 ARTEMP ENGL
TSS 123 ARTEMP ENGL
SAME
SAME
SAME
SAME
SAME
SAME
SAME
SAME
AVER
AVER
AVER
SUM
AVER
AVER
SUM
SUM
AVER
AVER
AVER
SUM
AVER
AVER
1 SUM
1 SUM
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
RCHRES
VOLS-> ***
* * * * ***
1 3 EXTNL PREC
4 6 EXTNL PREC
7 9 EXTNL PREC
1 3 ATEMP AIRTMP
4 6 ATEMP AIRTMP
7 9 ATEMP AIRTMP
1 9 EXTNL PETINP
1 9 EXTNL WINMOV
1 9 EXTNL SOLRAD
1 3 EXTNL DTMPG
4 6 EXTNL DTMPG
7 9 EXTNL DTMPG
1 6 EXTNL GATMP
7 1 1 EXTNL GATMP
161
-------
TSS
TS5
TSS
TSS
TSS
TSS
TSS
TSS
TSS
TSS
TSS
TSS
END EXT
NETWORK
122
41
42
46
124
126
125
134
127
136
1 13
1 19
ARTEMP
EVAPOR
WINDXX
SOLRAD
DEWPNT
DEMPNT
DEUPNT
UATEMP
SEDMNT
STFLOW
STFLOW
STFLOW
ENGL
ENGL
ENGL
ENGL
ENGL
ENGL
ENGL
METR
ENGL
ENGL
ENGL
ENGL
0.7
1 .0
1 .0
1 .0
1 .0
SAME
SAME
SAME
SAME
DIV
SAME
SAME
SAME
RCHRES 12 13 EXTNL
RCHRES
RCHRES
RCHRES
RCHRES
1 13 EXTNL
1 13 EXTNL
1 13 EXTNL
1 6 EXTNL
RCHRES 7 1 1 EXTNL
RCHRES 12 13 EXTNL
RCHRES
1 13 HTRCH
PLTGEN 4 INPUT
PLTGEN
1 INPUT
PLTGEN 2 INPUT
PLTGEN 3 INPUT
GATMP
POTEV
WIND
SOLRAD
DEWTMP
DEWTMP
DEWTMP
TW
MEAN
MEAN
MEAN
MEAN
2
2
2
2
SOURCES
<-VOLUME->
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLHD
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
*
7
8
9
7
7
7
8
8
8
9
9
7
7
7
8
8
8
13
7
8
9
7
7
7
8
8
8
9
9
7
7
7
8
8
8
12
4
5
6
4
4
4
5
5
5
6
6
4
4
4
5
5
5
11
4
5
6
4
4
4
5
5
5
PUATER
PWATER
PUATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
ROFLOW
PUATER
PWATER
PWATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
ROFLOW
PUATER
PUATER
PUATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
ROFLOW
PWATER
PWATER
PUATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
<-MEMBER-X-MFACT >
*
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
POPST
SOSDPS
SOSDPS
POPST
SOSDPS
SOSDPS
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
POPST 1
SOSDPS 1
SOSDPS
POPST 1
SOSDPS 1
SOSDPS 1
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
POPST 1
SOSDPS 1
SOSDPS 1
POPST 1
SOSDPS 1
SOSDPS t
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
7413.
10770.
4640.
8896.
44480.
35584.
12928.
64640.
51712.
33408.
22272.
26688.
3203.
23485.
64640.
7757.
56883.
1 .0
2667.
3787.
1600.
3200.
16000.
12800.
4544.
22720.
18176.
11520.
7680.
9600.
1 152.
8448.
22720.
2726.
19994.
1 .0
1120.
2027.
960.
1344.
6720.
5376.
2432.
12160.
9728.
6912.
4608.
4032.
484.
3548.
12160.
1459.
10701 .
1 .0
1013.
2080.
1067.
1216.
6080.
4864.
2496.
12480.
9984.
« «
RCHRES 13 INFLOW
RCHRES 13 INFLOW
RCHRES 13 INFLOW
RCHRES 13 INFLOW
RCHRES 13 INFLOW
RCHRES 13 INFLOW
RCHRES 13 INFLOW
RCHRES 13 INFLOW
RCHRES 13 INFLOW
RCHRES 13 INFLOW
RCHRES 13 INFLOW
RCHRES 13 INFLOW
RCHRES 13 INFLOW
RCHRES
RCHRES
3 INFLOW
3 INFLOW
RCHRES 13 INFLOW
RCHRES 13 INFLOW
»*#
IVOL
IVOL
IVOL
ISED
ISED
ISED
ISED
ISED
ISED
ISED
ISED
IDQAL
ISQAL
ISQAL
IDQAL
ISQAL
ISQAL
* « #**
1
2
3
1
2
3
Z
3
1
2 1
3 1
1
2 1
3 1
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES 12 INFLOW
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
IVOL
IVOL
IVOL
ISED
ISED
ISED
ISED
ISED
ISED
ISED
ISED
IDQAL
ISQAL
ISQAL
IDQAL
ISQAL
ISQAL
IVOL
IVOL
IVOL
ISED
ISED
ISED
ISED
ISED
ISED
ISED
ISED
IDQAL
ISQAL
ISQAL
IDQAL
ISQAL
ISQAL
1
2
3
1
2
3
2
3
1
2 1
3 1
1
2 1
3 1
1
2
3
1
2
3
2
3
1
2 1
3 1
1
2 \
3 1
RCHRES 10 INFLOW
RCHRES 10 INFLOW
RCHRES 10 INFLOW
RCHRES 10 INFLOW
RCHRES 10 INFLOW
RCHRES 10 INFLOW
RCHRES 10 INFLOW
RCHRES 10 INFLOW
RCHRES 10 INFLOW
RCHRES 10 INFLOW
IVOl
IVOL
IVOL
ISED
ISED
ISED
ISED
ISED
ISED
1
2
3
1
2
3
162
-------
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
6
6
4
4
it
5
5
5
10
4
5
6
4
4
4
5
5
5
6
6
'
H
4
5
5
5
9
4
5
6
4
4
4
5
5
5
6
6
4
4
4
5
5
5
8
6,
5
6
4
4
4
5
5
5
6
6
4
4
4
5
5
5
7
1
2
3
1
1
1
2
2
2
3
3
1
1
1
2
2
2
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
ROFLOW
PUATER
PUATER
PWATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
ROFLOUI
PWATER
PWATER
PWATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
ROFLOW
PUATER
PWATER
PWATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
ROFLOW
PWATER
PWATER
PWATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
SOSED
SOSED
POPST
SOSDPS
SOSDPS
POPST
SOSDPS
SOSDPS
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
POPST
SOSDPS
SOSDPS
POPST
SOSDPS
SOSDPS
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
POPST
SOSDPS
SOSDPS
POPST
SOSDPS
SOSDPS
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
POPST 1
SOSDPS 1
SOSDPS 1
POPST 1
SOSDPS 1
SOSDPS t
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
POPST
SOSDPS
SOSDPS
POPST
SOSDPS
SOSDPS
7680.
5120.
3648.
438.
32)0.
12480.
1498.
10982.
1 .0
587.
IZSfl.
640.
704.
3520.
2816.
1536.
7680.
6144.
4608.
3072.
2112.
253.
1859.
7680.
922.
6758.
1 .0
5333.
10190.
4960.
6400.
32000.
25600.
12224.
61120.
48896.
35712.
23808.
19200.
2304.
16896.
61 120.
7334.
53786.
1 .0
4053.
9173.
5867.
4864.
24320.
19456.
1 1008.
55040.
44032.
42240.
28160.
14592.
1751.
12841 .
55040.
6605.
48435.
1 .0
2400.
6933.
5547.
1440.
15840.
11520.
4160.
45760.
33280.
39936.
26624.
8640.
1037.
7603.
41600.
4992.
36608.
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
to
10
10
10
10
10
10
10
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
ISED
ISED
IDQAL
ISOAL
ISQAL
IDQAL
ISQAL
ISQAL
IVOL
IVOL
IVOL
ISED
ISED
ISED
ISED
ISED
ISED
ISED
ISED
IDQAL
ISQAL
ISQAL
IDQAL
ISQAL
ISQAL
IVOL
IVOL
IVOL
ISED
ISED
ISED
ISED
ISED
ISED
ISED
ISED
IDQAL
ISQAL
ISQAL
IDQAL
ISQAL
ISQAL
IVOL
IVOL
IVOL
ISED
ISED
ISED
ISED
ISED
ISED
ISED
ISED
IDQAL
ISQAL
ISQAL
IDQAL
ISQAL
ISQAL
IVOL
IVOL
IVOL
ISED
ISED
ISED
ISED
ISED
ISED
ISED
ISED
IDQAL
ISQAL
ISQAL
IDQAL
ISQAL
ISQAL
2
3
1
2 1
3 1
1
2 1
3 1
1
2
3
1
2
3
2
3
1
2 1
3 1
1
2 1
3 1
1
2
3
1
2
3
2
3
1
2 1
3 1
1
2 1
3 1
1
2
3
1
2
3
2
3
1
2 1
3 1
1
2 1
3 1
1
2
3
1
2
3
2
3
1
2 1
3 1
1
2 1
3 1
163
-------
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
6
1
2
3
t
1
1
2
2
2
3
3
1
1
1
2
2
2
5
1
2
3
1
1
1
2
2
2
3
3
1
1
1
2
2
2
4
1
2
3
1
1
1
2
2
2
3
3
1
1
1
2
2
2
3
1
2
3
1
1
1
2
2
2
3
3
1
1
1
2
2
2
2
1
2
3
1
t
1
2
ROFLOW
PWATER
PWATER
PWATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
ROFLOW
PWATER
PWATER
PWATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
ROFLOW
PWATER
PWATER
PWATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
ROFLOW
PWATER
PWATER
PUATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
ROFLOW
PWATER
PWATER
PWATER
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
POPST 1
SOSDPS 1
SOSDPS 1
POPST 1
SOSDPS 1
SOSDPS 1
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
POPST
SOSDPS
SOSDPS
POPST
SOSDPS
SOSDPS
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
POPST 1
SOSDPS
SOSDPS
POPST
SOSDPS
SOSDPS
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
POPST
SOSDPS
SOSDPS
POPST
SOSDPS
SOSDPS
PERO
PERO
PERO
SOSED
SOSED
SOSED
SOSED
1 .0
853.
2027.
2027.
512.
5632.
4096.
1216.
13376.
9728.
14592.
9728.
3072.
369.
2703.
12160.
1459.
10701 .
1 .0
960.
2187.
2400.
576.
6336.
4608.
1312.
14432.
10496.
17280.
1 1520.
3456.
415.
3041 .
13120.
1574.
1 1546.
1 .0
1227.
2880.
3200.
736.
8096.
5888.
1728.
19008.
13824.
23040.
15360.
4416.
530.
3886.
17280.
2074.
15206.
1.0
2880.
6827.
7733.
1728.
19008.
13824.
4096.
45056.
32768.
55680.
37120.
10368.
1244.
9124.
40960.
4915.
36045.
1.0
2293.
6613.
8800.
1376.
15136.
1 1008.
3968.
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
IHFLOW
INFLOW
IHFLOW
IVOL
IVOL
IVOL
ISED
ISED
ISED
ISED
ISED
ISED
ISED
ISED
IDQAL
ISQAL
ISQAL
IDQAL
ISQAL
ISQAL
IVOL
IVOL
IVOL
ISED
ISED
ISED
ISED
ISED
ISED
ISED
ISED
IDQAL
ISQAL
ISQAL
IDQAL
ISQAL
ISQAL
IVOL
IVOL
IVOL
ISED
ISED
ISED
ISED
ISED
ISED
ISED
ISED
IDQAL
ISQAL
ISQAL
IDQAL
ISQAL
ISQAL
IVOL
IVOL
IVOL
ISED
ISED
ISED
ISED
ISED
ISED
ISED
ISED
IDQAL
ISQAL
ISQAL
IDQAL
ISQAL
ISQAL
IVOL
IVOL
IVOL
ISED
ISED
ISED
ISED
1
2
3
1
2
3
2
3
1
2
3
1
2
3
1
2
3
1
2
3
2
3
1
2
3
1
2
3
1
2
3
1
2
3
2
3
1
2
3
\
2
3
1
2
3
1
2
3
2
3
1
2
3
1
2
3
1
2
3
1
1
1
1
1
t
1
1
1
1
1
1
1
1
1
1
t
164
-------
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
PERLND
RCHRES
RCHRES
RCHRES
RCHRES
GENER
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
GENER
RCHRES
RCHRES
RCHRES
GENER
GENER
RCHRES
RCHRES
RCHRES
RCHRES
2
2
3
3
1
1
1
2
2
2
13
7
1
1
1
1
13
7
1
1
2
3
4
5
6
7
8
9
7
7
7
7
1
7
1
1
1
1
2
1
7
1
1
2
7
1
7
1
SEDMNT
SEDMNT
SEDMNT
SEDMNT
PEST
PEST
PEST
PEST
PEST
PEST
HYDR
HYDR
HYDR
SEDTRN
HYDR
SEDTRN
HYDR
HYDR
HYDR
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
SEDMNT
GQUAL
GQUAL
GQUAL
SEDTRN
OUTPUT
GOUAL
GQUAL
GQUAL
GQUAL
SEDTRN
OUTPUT
GQUAL
GQUAL
GQUAL
OUTPUT
OUTPUT
GQUAL
GQUAL
GQUAL
GQUAL
SOSED
SOSED
SOSED
SOSED
POPST
SOSDPS
SOSDPS
POPST
SOSDPS
SOSDPS
ROVOL
ROVOL
ROVOL
ROSED
ROVOL
ROSED
ROVOL
ROVOL
ROVOL
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
SOSED
DQAL
RODQAL
ROSQAL
ROSED
TIMSER
ROSQAL
DQAL
RODQAL
ROSQAL
ROSED
TIMSER
ROSQAL
DQAL
DQAL
TIMSER
TIMSER
RODQAL
RODQAL
ROSQAL
ROSQAL
4 1
4 t
1 1
4 t
4 1
4 1
t 1
4 1
4 1
4 1
1 1
1 1
4 1
4 1
43648.
31744.
63360.
42240.
8256.
991 .
7265.
39680.
4762.
34918.
6.05
6.05
6.05
1 . 1 18E-3
6.711E-6
1 . 1 I8E-3
6.05
6.05
6.05
2000.
2000.
2000.
2000.
2000.
2000.
2000.
2000.
2000.
1 .026E-6
500.
1 .026E-6
5.592E-7
500.
5.592E-7
500.
500.
1 .026E-6
5.592E-7
1 .026E-6
5.592E-7
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
RCHRES
PLTGEN
PLTGEN
PLTGEN
PLTGEN
DISPLY
DISPLY
DISPLY
DISPLY
DISPLY
DISPLY
DISPLY
DISPLY
DISPLY
DISPLY
DISPLY
DISPLY
DISPLY
DISPLY
DISPLY
DISPLY
GENER
GENER
DISPLY
DISPLY
DISPLY
DISPLY
GENER
GENER
DISPLY
DISPLY
PLTGEN
PLTGEN
PLTGEN
PLTGEN
PLTGEN
PLTGEN
PLTGEN
PLTGEN
1
1
2
3
4
1
3
5
7
9
1 1
12
13
14
15
16
17
18
19
20
21
1
1
22
23
24
25
2
2
26
27
5
5
6
6
7
7
8
8
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INFLOW
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
INPUT
ISED
ISED
ISED
ISED
IDQAL
ISQAL
ISQAL
IDQAL
ISQAL
ISQAL
MEAN
MEAN
MEAN
MEAN
TIMSER
TIMSER
TIMSER
TIMSER
TIMSER
TIMSER
TIMSER
TIMSER
TIMSER
TIMSER
TIMSER
TIMSER
TIMSER
TIMSER
TIMSER
TIMSER
ONE
TWO
TIMSER
TIMSER
TIMSER
TIMSER
ONE
TWO
TIMSER
TIMSER
POINT
POINT
MEAN
MEAN
MEAN
MEAN
MEAN
MEAN
2
3
2
3
1
2
3
1
2
3
1
1
1
1
1
2
1
2
1
2
1
2
1
1
1
1
END NETWORK
END RUN
/»
// EXEC FORTGO.PROG=PLOT,LIB='WYL.XA.R72.PLOT',VOL=PUB010,
// REGION.GO=512K
//GO.FT05F001 DD DSN=WYL.XA.R72.PLOTFL1.IOWA.PEST.C2,
// DISP=
-------
// DISP=(OLD.KEEP).UNIT=DISK,VOL=SER=PUB010
/V EXEC FORTGO,PROG=PIOT,IIB='WYL.XA.R72.PLOT1,VOL=PUB010,
// REGION.GO=512K
//GO.FT05FOOI DD DSN=UYl.XA.R72.PLOTFLS.IOWA.PEST.C2,
// DISP=(OLD,KEEP),UNIT=DISK,VOL=SER=PUBO10
166
-------
APPENDIX B
Use of the NETWORK Block to Connect the
Surface and Instream Application Modules
In HSPF, the operational connection between the land surface
and instream simulation modules is accomplished through the
NETWORK Block. Time series of runoff, sediment, and
pollutant loadings generated on the land surface are passed
to the receiving stream for subsequent transport and
transformation simulation. This connection of the IMPLND
and/or PERLND modules with the RCHRES module requires
explicit definition of corresponding time series in the
linked modules. A one-to-one correspondence exists between
several land segment outflow time series and corresponding
stream reach inflow time series (e.g. runoff, sediment,
dissolved oxygen, etc.); however in order to maintain
flexibility, some of the time series are more general, and
no unique correspondence exists. Also, in some cases, a
process or material simulated in the stream will have no
corresponding land surface quantity. For example, the
inflow of plankton to a stream occurs only from upstream
reaches and not from a land segment.
The following table is a list of the more common or likely
time series correspondences between the IMPLND/PERLND
modules and RCHRES. The table is structured such that the
right hand section consists of a list of all possible
materials or quantities simulated in the RCHRES module.
Information included for each is the HSPF section in which
the material is simulated, the variable name, and its units.
The left hand column indicates the corresponding time series
from the land segment module (or a possible one) and
includes the same information as the right side. In
addition, a conversion (CONV FACTOR) factor between the two
corresponding time series is specified. The actual
multiplication factor (MFACT) to be used in the NETWORK
Block is calculated as: MFACT = area * CONV. FACTOR. The
user should note that the module sections PQUAL, IQUAL, and
GQUAL involve the simulation of one or more genera I quality
constituents; consequently, their inclusion in these tables
reflects only possible or recommended correspondence. Other
combinations are possible depending on the particular
application. The user should consult the individual time
series catalogs (Part F, Section 4.7 of the User's Manual)
for more detailed information about particular time series.
167
-------
CO
LU
H-*
Of. "
LU h-
COO
<£
LU LL
t-Oi
O
r-LL.
^
Q_ ^
zco
1-fLU
a:
co o
LUJ-CMrO--l O 01
i o o z z z z z z z a. Q_ a_ a. a. a. a. LU LU -<- in
in > a: oi 3 "a c
LU xrc-naioai
&Z 1 o O -4-» tn T3
o| a * c c. 3 QI
Oi utnincaiu- ^3^3in
Z Ot-t-OCO3 01
o zzz cQcat-OiaiooincrL-Ti-
DC o; Qi QS -i _i -i i XXXXXXX-X^^^^CZIX tn -K +s a moi^ . t_
co I oxco co mo o ooooz zz zzz z a_ a. 0.0. a. 0.0. en ^ ^ 01 T><: in a
ro a OP rouJuooTi
T3-H> UI_ :C -i-> 73 -M 10
O C '13 co.OO73OmcoCn
in in ca: oiroro -c
O T3-i->,inino ft: o 73 73 - a L.
»- 73
roro < ^x Cki O Irt H-» C In X!
'Oouuoooooooooooooo uoooo eat & vr- LU co i +
^fQfQfyfyf^fyrorOfurOfO^DnjfOrjt^fDf^ (0 lu H3 fy fD ^-^ 3Z ^o CD sJ" ft (_ *t *^- o L
Z XXXXXXXXXXXXXXXXXX XXXXX i^ Z3I 0-0 O O QIC
X >3 cccininininin>inm>inininin min>>in o;c z a. C1*- <-' *- in-
C^--r-OOOjQX!^J3J3-t-'^lXH-'ZlJ3J3il Q.CI-t--t-_a OO « < CCO D
o" CQ -»-»+ ^j<<« H cr«» 0"*-^ -H t *-* o" cr^^ 13 rocroc oooi m
I^j ^->o -i-'O mz EC roro
LU LU ro o o ai _i -M o
a: 3 "a in ai
CQ 33L.LUJ3(D^ E
-lOrOCM - QIM- O ' Q < *^Q_ mro ro
oto .C ZozocLgrin
O 'in _iin QIE *£ >»
111 QjQ£a^oin-+^+*-^ai
Q_i i cototoj:_i _i _i_iz:caLUinLUinoL. in tTJC
a_
oi t-i
LU Oi LU O OO OO O OOO OO O OO CO CO CO LU LU LU O O O CM to vT IT1 ^
tOCOQ_Q_a_^^ ^ s^ ^^ ^^ ^>
z
o
o
LU
CO
za.Q-CL.a-
168
-------
Ul
n
tu ^«.
in i
O
LU >>>>>>>>>>>>>>>
_IZ-
C G
"D -4J > in in o
to c ~H (C
to ID - C C C
oo in in u o O O
000 0 0 0 0 0 0 0
1- i. L. (_ u t-
>>>>>>>>>>>>>>>
iOOOOOOOOOUOOOOO
xxxxxxxxxxxxxxx
CTCQ -^ -<- -^ cr a cr cr cr cr cr cr cr
Odd x
001 _i_ i loinininoooooo
Qicyzininino'o'GrcyciC'O'cyo'O'
oi in in in in in
>>>>>>
]! J-! J= J3 .3 .S
U. U_ U. LL LL U-
i«; vs ^ ^ HI T:
OO CO OO «O O O
CQ CO
Q_D- Q- CLD_ Q-
O O CD CD O
> > > > >
O O O O U
X X X X X
>* ^x !lx >* in
cr cr D- cr
CM
o o o o o
OO O'O'O
o o o o o
in
O'O'C'OTS
5 1 group
cn
L
in
Qi o
O U-
HH
.^
o tx
O C£
CQ 1
^ C
H- -t-*
LU ro
*-> o
01
c a
*- in
in
(U
0
z
o
no contriL
:omponent .
+* in
15 LL)
o o:
T3 0
C fV
in c
E ~o
-* *-
0 d
01
Z L
_J O
E:
l-H L
O
C M-
Q
>n z
t-2I
C
0) E
O
O L
CM-
CSJ
-«-' C
in
U Oi -*-
01-3
in _Q
in 3
(5 C" L
01 -^
*> l- u\
c
and contain i nc
SILT and CLA^
za fraction c
C Q
£<* c
"D -)-
Q) O in
C O
s- - O
0
-+- *+-
O Gl
EC
*- o
J3 QI +-
I- E
in -+- -4-»
^- -^ in
T3 3 01
to
T3
QI
C
13
01
Si in
Oi
-. t.
3 C'
o in
r~
in Qi
E
C -^
15
cn
Ti
c c
D 0
0.
' i-
- O
GJ
Q. GJ
tn *~
1 -»-
l_
0) C
in -
in
(0 ro
E
r in
-H- 0)
cr-C
^ -*-*
169
-------
APPENDIX C
Equivalency Table for Selected HSPF
and ARM/NPS Model Parameter Names
HSPF and Corresponding ARM Model Parameters*
PROCESS
Runoff-related
Interception
Depression/Surface Storage
Soil Moisture Storage
Overland Flow
Infiltration
Subsurface Flow
HSPF
PARAMETER
CEPSC(M)
UZSN(M)
LZSN
LSUR
SLSUR
NSUR(M)
INFILT
INFEXP
INFILD
INTFW(M)
IRC(M)
DEEPFR
AGWRC
KVARY
CORRESPONDING
ARM PARAMETER
EPXM
A
UZSN
LZSN
L
SS
NN
INFIL
none
none
INTER
IRC
K24L
KK24
KV
Interception storage capacity.
Values in ARM vary with monthly
crop cover.
Impervious areas are handled as
a separate segment in HSPF.
Upper Zone Nominal Moisture
Capacity.
Lower Zone Nominal Moisture
Capacity.
Length of overland flow path.
Slope of overland flow path.
Manning's n of overland flow path.
Index to infiltration capacity
of soil.
Exponent in infiltration equation.
Value of 2.0 is used in ARM.
Ratio of max to mean infiltration
capacities of the soil. Value of
2.0 is used in ARM.
Interflow inflow parameter.
Interflow recession parameter.
Fraction of groundwater inflow
to deep aquifers.
Groundwater recession parameter.
Variable groundwater
recession parameter.
Parameters followed by '(M)' indicate that 12 monthly values can be specified.
70
-------
01
19
i. in
41 +
P C
01 41
E E
IB -H
L. in
IB D
n.'i-i
H^
Ul
in
01 41
O L
N Oi
in
t-
4) 41
3 E
O
-1 t-
41
£
+
C
1-
3*
u. u
z: o
i
X 03
H
ii
3 D--
0
oi s
"D H
19 Ul
E Z
> 01
.Q 3
c
-O
a f
i. C
01 O
> a
0
O -1
H
to -
5!s
£
H- 0
°x
C 3
O
+
H in
U 41
IB i-
i. 0
U. 4-
L
01
H
01
E
IB
L 1.
41 IB
-f^ Q_
C
c
3 O
H
C 41 -H
"O IS
O 1.
l» Q.
E in
CLD; c
in < 19
C L
IB C -H
i. O
H a
+ IB
o o >
H Z Ul
01 E
Ul O
a t-
Q
H C
in o
i. *
0) IB
P L
"^ £L
12
3 IB
0 1.
u +*
OI O
t&
0 >
4- Ul
>H
01
O
z:
y*
fy
^
C
H
O
t
3
O
H
4-
01
in
19
account
O
H
^
01
01
E
IB
19
a
6
8
o
to
01
tn
IB
Dl
" ?
> E
§ §
..- 1-
0 0 *
4-0
4- 4)
4- I. >
01 O C
+ O
x o o
O 19 1
H 4- C
19 O
° "c '+
L 41 IB
O E Ul
O + C
a in 4i
t.3,"g
O-DO
4- < O
melt
series
SOI
E
p 'f
IB
"O O
IB
IB
+ TJ
IB
in L.
C <-
C (U 19
.2^-3
* in in
D T-ITJ
CTT3 C
01 < 19
C
O
u. u.
z: a.
x x
H
3 O
41 O
E O
i. i-H
0 03
4-
i. ^
41 CK
o O
01 t-
L Ul
19 Z
.HJ5
t. 01
o yi
o a:
IB 01 <
IB 4-
* o
-i in
01 1. C
E 01 O
H i^
TJ 41 -H
CEO
5 19 41
o t. in
1. 19
oi arc
o
> v z
rH Ul tO
41
19 £ C
OH
§
O
41
in
D£
Ul
t-
£
c
o> u!
in a.
3 to
o,1
t- 4-
II O
i!
K1 Ul
^ a.
Iri
L)
U
U
§
cc
o
01
I
in
Jg
19
U
i
»H
19
>
X
/S
Parameters
HSPF
PARAMETER
LZETP(M)
FOREST
a.
i-
Ul
to
OJ
+>
a
OL
a
in
01
!_
L
O
O
?
19
U.
a.
to
X
lp.
+>
a
t.
'a
8
to Z
to
UJ
o
O
tt
a.
,^>
a
19
>
Ul
o S
5: 8
in
oi
a
L.
!_
D.
ft S
19
to
171
-------
.£
3 01
4- U
°L
> 41
*>§
H -
O +*
fB (9
4- C
sl
.- oi
H
re
t.
C - TJ
01 41
sss
'IS
oi J in
^j uj .-
u.
01 T) D-
o c OP 4-
fc 41 OI
41 -4
H UJ
UJ
«.£!«
TJ
01
gs
'i* 01
O
re _
t. *
L. re
41 1
>
o o o.
41 4- 10
Vc
TO 41 -r-
41 -H
J= 41 -H
in E O
i. re c
H re oi
re a <-
3 re
_ 0"
£ in re
41 x re
x: K -a
C X
'i- -H
in c
4) I-
£ U
Su. o re
y o 5* '
"O .£ CO 3 '
re -H cr
^ 01
in 3 o:
4) o i.
41 O UJ ra
in -a z. 3
o
01
> 3 U.
<:
> i
UJ M
UJ
z:
re
a.
> :
ii
<
3
§
M
O
u
'i
+
o
I
172
-------
a
co
x
1-
Ul
O
U
to cr
5£
is
o <
0. a:
CO <
ui QL
DL
fv y
O C£
a;
t-
Ul
u. <
a. ce
10 <
X CL-
IO
CO
Ul
8
-f
t. c
01 01
H E
i-5
re re
L. -H
re oi
O.T)
a^
£ °
re
0 £
a""
O -M
U 01
^H .r-
X 4-
f 4-
goi
o
s: u
^
JE
Q
a. a
> Ul
8£
i
ui ct
> Ul
o o:
u *:
rH
i-H
re
4-
c
o 're
oi ce
f
re >
rH J2
41
L. H->
1 £
*H 01
£ E
41 X
E O
.- re
01 41
co a
^
g
g
O
^*
01
H
'5
in
C
£ + £
0 C 0
f» C -H
a o a
oi ui oi
ce
Ul
a:
rv
Ul
CC
41
U
H
U
re
L.
a
E
01
ai
re
re
E
.?
H
o
a
CO
u.
a.
u.
a.
i:
CO
i
u
in
rH
'o
in
X
t.
re
E
£
44
»C
O O
O 4- -H
»^ 4- re
O 4) 3
re o &
4- O 41
01
g
a.
Ul
CO
J3
L
in o
N M
+* 4-
1"?
x re
O i-H
re L.
H 41
t.
§
u
in
i-H
'o
in
X
i.
re
C
+"
£
O
z
o
cc
Ul
CO
->
1
in
£
re
L
H
fj
S
E
5
01
in
£
4J
o 'o
f 4-
ra 4-
3 01
tr o
01 U
cc
Ul
CO
a
Ul
to
§
i~f
g
re
41
O
n
4*
O
a
re
t.
c
0
a
re
L.
f
B
'v
01
in
£
g^g
+ £ -H
re o re
B-X-&
01 ui 01
a
Ul
CO
1
a
Ul
CO
F-l
'o
in
41
£
£
o
01
TJ
3
|
in
OI
>,
H
H
CO
£
g
t.
01
41
e
re
i_
o
?
1
in a:
Ul
O CO
X ^
i.
stream
on/scou
o
;*
L. UJ
^ ^
i^- i-H
H iH
O CO
i
rH
H
Jf
"8
3
.i!
H
S8
& £
01
*^"
L
3 -H
80
in co
_]
01 DC
£ Ul
O C£
£ CO
ce co
Ul I
co ui
-> Q
4-
o
+
1
+
si
re 4-
H
nj
> O
< Ul
T,
O
re
*-»
41
T3
C
o
H
O
T3
01
i.
^
(_l
re
a
u
a.
5
V)
X
M
U.
U.
g
H
O
n
C
-
01
41
L
OI
01
* II
.r- re in
+ .-< o
o 3 a
a § ~
E o in
o u oi
u re c
o o £
t. rH
01 oi
3 X O
t» a in
in 41
in o 4- >
41 e o -
£ OI-H +
- E re -H re
4* *r- re ot
rH > > 01
r-H -f rH O C
'o 41 're 41 i.
in in a t. o
S
g
£
M
^
173
-------
TD
1
4*
R
2
i
g
£
£
H
r
?
r respond
5
g
a
in
IU
i
ID Q£
£ iU
Q UJ
IS
&S
IU IL
ii
e
r
&;*
CE
££
tO
(0
IU
B
g
?1
1- *
c *
1 0
« 41
U +
*!
*
tn
o
in
X-H
c&
- a
H *
in Ji
41
M
41
L. C
" 1
Hg
- in 5
* 41 .X
i» 5 f
O -H (I
» a o
Is,!
II 0
H t.
n o H-
o -H o
" in
lli
o * *
«
I
8
n
i 1}
* M j
J5 ^2
2 .r-
1 fH
-1 41
n J) o
O i» r»
to **
w L
si
a -*
*
i. «
c "
orages i i
nterflow
+ .1-
in
*l
'o **
- «.
+ 41
21
KJ M
n
M A
n !?
M 3
J
o
8 in
C * 41
N c 'B
~'-H "in
n C
L + O
0 «- 3
5f||
I?J
S^i-o^
O -) 41
t. r» J3 ^H
Ol 1. «H
u. O n
"O D. in
C WT3 C
n x n "-
8(0 (0
0. Q.
~ ^ ^^
12 S 12
M -I <
(O
*
i
* '£* a
IS aM
-wo
2V"8
41 m .
M Z > ft K1
Cl (. *>
41 « 41 3
> K> «. +-!
- o o c o
U. Z H- - M
L8
W -H
1. >*-
Ii o c a «-
i. * -t ~4 41
n *
TJ fi-t C
at V u -
«- O O
&H- S- IA +
it .r- a
«. O ^ X 41
O 41 U U U
H O. fB 13 X
W Ul 41 41 41
ui i/t CM ui in
§5523
,
ra
4<
Z
(0
i|
< (.
1 O
S-o"
p'o o «
^s.0-?
"1*1
t <- 1.
1. Z 41 41
D < >*
O -J B C
u_ Q. -H -
O
*
If) ***
5 S
-4 1.
a -H
g'E
»<*-<**
a a a. o. a.
10 O M -J <
?
C
I
i!
HOC
gSR-f!
4« * «H C It
a a o o
i. u j i
O O 0 I H-
HJ:*
iii^l
«<5r
55z M
to Q £ C z
o < w < o
1
*
! i
H- PI
i!
-------
g
H
re
u
4-
tr-
i.
*
Z
g
ir-
H
re
N
rH
O
e
g
z
z
Kl
1.
4-
41
.!*?
ni
H
O
5
c
re
rH
a
4-
o
c
0
f
u
re
i.
u.
o
rH
O
^-*
t
m
g-
41
in
g
in
^
7j*
y
TJ
re
UJ
I
t-
.g
u
01
t.
1.
o
o
01
I.
3
H
re
i.
41
a
E
01
1-
designatei
ction
C
o
o
ra
u
L.
U
re
01
t.
0
4-
in
41
3
rH
re
>
re
41
t.
41
H
4-
0
+«
O
t.
4-
c
i
^
JQ
i-l
01
1.
U.
01
3
o
0
H
o.
0
§
0
rH
re
01 U.
E 0-
re 10
C I
i.
o
4-
in
c
o
^>
re
3
U
rH
re
u
e
'L
JQ
*H
IP_
&
41
order
in
^
o
-o
a
01
-H
in
c
c
o
H
a
i.
o
in
"O
re
01
t.
re
in
F
41
ffi
t-
ra
a
4)
in
01
h-
in
o
^J
41
C
^
01
H in
-O 01
- 3
in i
f. re
01 >
(y 2
- c
re
^ *.
.t^ ^
-H
C
|.2
o a
in i-
o
01 in
jc -u
H re
< 01
H +
uj a
cl
O -H
c - o
o + a
u ai
01 t.
a o.-- 41 i.
W J3 re
re 0.
o o o
in in E
41 t! E
~o re
C 3 O «l
£2
re
t. in
r- a 41 ai
~ °- 3
>* 01 E rH
o (- 41 re
D. O I- >
c
. 4)
o cn
c o
01 -H
01 ,r-
10 C
r
rH M 0
OT
01 <
C * O
O I -H
C I- 01
.< « o> 01 01
<0
-------
O 3
UJ
o
u
in o
ZS
C L
01 O
D14-
O
i. in
H 01
'E re
o >
«)X
ro c
+ O
in «H
3
O -I
H 5
re in
§
o
f-H
U.
D.
to
X
*
^
O
Q.
H-
0
0)
\s
re
H
o
H->
C
re
H
Q.
L
01
>
re
-I
(_i
'o
in
u
re
01
c.
0
H-
in
0)
-H
t.
H
X
f-»
c
E
>
H
-I
^Q
3
,_|
O
in
E
'x
re
z:
H
'o
in
o
T3
01
x
H-
>^
rH
H
gm
3
C -H
ra re
E >
OJ ^
a
oi "6
a --
'o ~a
-H 3
in 01
01 L.
a. u.
01
3
fH
re
z
o
H
?
D
01
1.
u.
01
H
in
i
o
c
o
H-
01
3
re
Z
X
u
1 1
T3
3
01
t-
U-
on.
rameters
.- re
H p
a
0 01
m
> 01
ax
o
in in
OJ 3
"° -3
r-H
a re
i.
O U.
in a.
~a to
re x
T5 .
3 01
re o
> z
i.
ot
re
!-(
'o
in
X
u
ro
01
X
+^
'3
>
1.
re
>
o
H
(.
O
M-
in
L.
OI
01
E
re
L
re
a
01
-H
re
i.
t.
01
T3
L
O
H
in
S.S
'H C
a o
L. ir-
o +
in Q-
01 O
\ r-1
c re
0 C
~ O
+i ii-
a*-
O "O
in TJ
a re
re
01 re
-a
p- in
o .-
'? in
in '
01 X
a»-
?
UJ
P
C
O
+>
o
01
1.
1.
o
o
01
3
f
re
0)
u- a
a. a
to oi
x H
1
o
44
in
01
D
'o
H
in
01
a
L
o in
4- 0)
H
in re
01 t-
3
H O
re x
ID CD ID CD
a o o o
o:
ui
t-
UJ
i:
u. <
x a.
i O U
_ i CD UJ
a a a a
tn 5 -J <
u
ID
x x
< M
~ U.
X Sri
511. -! -H CJ
X Sri Z Z
IO (0
a. a.
to o
o <
in
in
UJ
8
E
re
TJ
re
L,
cn in
01 0)
0)
01
o
t.
a
o
oi o
T3 -
F- -H
o a
- i.
H O
Sin
01
176
-------
O or
Z UJ
O <
a ce
10 <
uj a.
o
41
in
i.
41
I
a
i.
re
D.
ct
UI
UI
u. <
a. a
in <
x a.
§
^_
_i
in
<
E
H
(0
CO
in o
to -H
** 41
I- "O
41 3 01
a-f -H
o re re
rH L rH
in 4i 4i
Q. i.
TO E
C SI £
rj -H O
"» t. +
t- oi re
-J > 3
10 re a
< rH 01
rcept
urface
ess i on
oi in i.
H 01
KJ NJ
in in
< CO
~ ~
c n
5 5
S 2
g
+>
oi rH re
o o &
-H in oi
ature.
(ULTPl) and s
upper layer
e regression
1. 4- L.
41 -H O 3
D. D. -H
E 41 " a
01 U OJ 1.
* 1- Q. 41
41 1- Q.
i. * -1 E
C 3 01
re IH ~ -H
Is
E E
^^ ^^
H C\J
a a.
1- h-
3 3
modeled
or ui th a
d as follows:
41 01
M 41 C
C 0 £
ro .a oi
o re -o
41
41 TJ L.
L. 01
3 C in
H - t
re H- OP
1-5)+-'
01 "D OP
E in re
oi re L.
H re
c a
-H O
.r-.r-X
O -H +
in re .r-
3 3
1- V
5,4-1
re c -H
rH O -
r- (.
i. in o
41 in oi
a. 01 rH
J E*
U. "-
o- re x
(n +
X X 0
_ * g
C "- E
M 3 in
41
4-*
O
rH
'5
in
factor
rence between
41
CT4-
C 4-
X 73
-H
0 C
O re
E 01
in z:
none
none
E E
w «M<
rH OJ
D. 0.
1- t-
_J 1
L.
01
H
re
3
"D
mperatures.
wer and group
i_
re
4^
eratures at s
nth.
oi o a o
H .H E E
41
i. >»- X
r- rH O
re x i- re
o "c
C 0
re i:
.s
r
LZTEMP
E
^-*
^4
a
CD
41 41
ro H-
rH O
I.
O
rH
able for the
rameters:
rH re
- a
re
> en
re c
o '3
in o
rH rH
TO rH
O
in 4-
ir-
41
O +
,fm
+ X
° 'I
E
X in
H I-
is;
O r&
ClrH
f9 t.
V
|-i
"o §
o o
UI CD
Ql ^J
C C
**
41
M
O
N
01
3
0
rH
I- H-
O
+ 01
O i-
re 3
D1 re
c a
a
x -a
H
0 C
O re
E a
in r
none
none
E E
** *t
~i OJ
a. a.
t- t-
ts u
41
I-
3
re
r soil temper
mperature.
41 01
*-» -H
re
3 1.
c re
h
01 re
t.
rH Cl
"- >
O re
in rH
T3 i.
41 AI
§§
UI rH
in
re 01
4.f.
X -
D:
0**
in M
in
Is
U 41
"- >.
jC re
SZDPTH
UZDPTH
I_
01
> t.
re ai
j ^
re
41 -J
u
re (.
t £
3£
O
^**
1_
41 41 01
X 0 >.
+ re re
4** rH
in i-
r- 3 L
in oi
x a
t- 4i a
n. x 3
Q -H
N 01
3 E X
0 +-
C 4- 4-
re o
K i< c
H a o
01 O -H
41 TJ +
4- O
H O
-o ro
M 41
in o x
41 41
C C
82
L
01
ro L
-1 41
>,
I- re
01 -J
0 3
_l o
re
r- C t.
o ai o
in
ties for each
: does not as
» bulk dens it
parameters.
r- 0. X X
m m -H -H
ex a
01 O 01
T3 -H T)
Ji 41 01 rH
rH > 6 .r-
3 re ro o
CQ -H c m
N N M 41
(O 3 -J C
a o a o
CO CO CO C
.
AI
*f- i.
re ai oi
-J >> >>
re re i.
41 -1 -J 41
o >
re L l. ro
4- 01 OP _l
t- 0.3
41 -!
£ '5
ol
a
a
o
in
c
41
O
177
. U S GOVERNMENT PRINTING OFFICE 1984- 759-102/0971
------- |